Produced by Chuck Greif and the Online Distributed
Proofreading Team at http://www.pgdp.net (This file was
produced from images available at The Internet Archive)






  [Every attempt has been made to replicate the original as printed.
Archaic usages have been retained (i.e. shew, shews, divers, Friulan Alps, surprized.)
Some typographical errors have been corrected; a list follows the text.
Letters preceded by and equal sign [=Q] between brackets indicate that this letter
   was over-lined in the original. (note of the etext transcriber.)]




                           BRICK AND MARBLE
                                  IN
                           THE MIDDLE AGES.


               _Uniform with the present Volume._

     THE GOTHIC ARCHITECTURE OF SPAIN: FROM PERSONAL OBSERVATIONS MADE
     DURING SEVERAL JOURNEYS. By G. E. STREET, R.A. Second Edition, with
     100 Illustrations. Medium 8vo, 30_s_

     “Mr. Street has opened a new vein of architectural interest. Every
     part of this work presents evidence of the labour and deep interest
     with which he pursued his investigations, and the result is one of
     the most curious and valuable architectural works which we have
     received for some time.”--_Guardian._

     “A valuable contribution to the history of Gothic architecture. It
     will form a useful addition to the few books with which a traveller
     may profitably equip himself for the Peninsula. With the exception
     of the great work of Villa-Amil and Escosura, we have no
     publication which throws so much light on the architectural
     monuments of Spain--especially on those of the earlier Christian
     period.”--_Edinburgh Review._

[Illustration: 1.--NORTH PORCH, STA. MARIA MAGGIORE, BERGAMO.

_Frontispiece._]




                           BRICK AND MARBLE

                                  IN

                           THE MIDDLE AGES:

                 Notes of Tours in the North of Italy.

                     BY GEORGE EDMUND STREET, RA.,

  MEMBER OF THE IMPERIAL AND ROYAL ACADEMY OF THE FINE ARTS, VIENNA,

                               ETC. ETC.

                           _SECOND EDITION._

                     WITH NUMEROUS ILLUSTRATIONS.

                                LONDON:

                    JOHN MURRAY, ALBEMARLE STREET.

                                 1874.

               [_The right of Translation is reserved._]

                                LONDON:

                  PRINTED BY WILLIAM CLOWES AND SONS,
                  STAMFORD STREET AND CHARING CROSS.


                             To the Memory
                                  OF
                  THE RIGHT REV. SAMUEL WILBERFORCE,
                      LORD BISHOP OF WINCHESTER,
                              ETC. ETC.,
           IN TOKEN OF THE AUTHOR’S MOST SINCERE AFFECTION,
                                  AND
        IN GRATITUDE FOR NUMBERLESS BENEFITS RECEIVED FROM HIM,
                             THIS VOLUME,
                 ORIGINALLY DEDICATED TO HIM IN 1855,
                           IS NOW INSCRIBED.




PREFACE TO THE SECOND EDITION.


The First Edition of this volume has been so long out of print that I
had almost ceased to regard myself as responsible for all that it
contains. It was the rapid and fresh summary of a happy journey
undertaken in the early years of my artistic career. There were, I knew,
many details in which it could easily be improved, and many journeys
taken over the same ground might have enabled me to go far more into
detail than I was able to do when I published it. I find, however, on
reading again what I wrote so long ago, that age and greater knowledge
of the subject have generally confirmed my old ideas, and that, as far
as regards the principles of my book, I still believe them to be true
and just. In revising what I wrote, however, I have found myself obliged
to make many alterations and additions, sometimes in relation to towns
not visited on my first journey, sometimes in reference to buildings
either not described at all or at best insufficiently described before.
In doing this I have endeavoured not to increase too much the bulk of
the volume, and as far as possible not to interfere with the general
character or tone of its contents, though in the process of revision the
larger portion of the book has had to be re-written.

I hope, if other occupations admit of it, before long to add to this
volume a second, containing notes of tours in the centre and south of
Italy, undertaken with the same object of studying and describing the
too little appreciated art of Italy in the Middle Ages, which seems to
me to be almost equally full of interest in all parts of the Peninsula.

The materials which I have accumulated for this purpose are only too
considerable, and the very richness of the subject has made me shy of
approaching it; but the necessity of publishing another edition of this
volume has revived my resolution to complete as soon as possible the
work which I originally proposed to myself, and of which I have never
lost sight. But whether I accomplish this or not, the volume which I now
republish may, I hope, give a tolerably complete view of Italian Gothic
architecture north of the Apennines.

Those who wish for further archæological details as to the age and
history of buildings, will not find great difficulty in supplementing
what I have written. For myself I confess at once that I have not had
time or opportunity for examining the documentary history of the
buildings I have described, and that the dates where I have given them
are generally obtained at second-hand, though never given save where
they accord with the architectural character of the work. It was never
for merely archæological purposes that I made my many journeys in Italy.
Before I first travelled there I had made myself well acquainted with
all the best remains of the Middle Ages in England: I had travelled,
sketch-book in hand, through France and Germany, and I knew, therefore,
something of the art of our fathers in most districts north of the Alps.
But so far I had found no time or opportunity for the study of those
early Italian buildings which give the key to the history and style of
ours, or of those later works in which, with more or less distinctness,
the architects north of the Alps repaid the debt they had previously
incurred to the South.

I felt then, as I do now, that no study of architecture was complete
which did not proceed exhaustively with the study of all European
varieties, and above all of that of Italy. Moreover, there was something
in the practice and tendencies of our own day which gave special
interest and fascination to such a study. We had gone on in our old
paths, studying the works of our own country, which in some respects
were deficient in their teaching. We wished to combine the best
architecture, the best painting, and the best sculpture in our works.
The world seemed to respond to our aspirations, and it is south and not
north of the Alps that examples of such a combination have to be looked
for. So again it was desirable at any rate to meet the demand which was
naturally arising for colour in construction, and here again it was in
Italy only that numerous ancient examples of such a combination were to
be found. These were the special inducements to me in my earliest
journey to Italy, and their influence is as strong as ever upon me. I
feel, indeed, even more now than then, the importance of such study to
the English architect of to-day. The more men educate themselves by the
study of ancient examples, the more is their work likely to become
refined and scholarly, whilst at the same time there is no real risk of
its becoming less original.

It is quite possible, and one wishes above everything to see it usual,
for architects to design all their work without special reference to, or
really copying from, any old work. But before doing this they ought at
least to put themselves in the same position as to knowledge of what had
been done before as that in which their forefathers were. Unless they do
so, the desire for originality will only be satisfied by the production
of excesses and monstrosities, whose only claim is that they are
new--one which, in spite of those who demand a new style, I venture to
declare to be their sufficient condemnation.

It remains only to say, that since I first visited Italy two works have
been published which add infinitely to our knowledge of the sister arts
of painting and sculpture, and enable the artist to travel with a full
certainty that he will not lose anything by reason of the ignorance or
carelessness of his guide. I need hardly say that I refer to Messrs.
Crowe and Cavalcaselle’s ‘History of Painting in Northern Italy,’ and to
Mr. C. C. Perkins’s admirable volumes on ‘Italian’ and ‘Tuscan
Sculptors.’ To the latter, indeed, I owe I know not how many
acknowledgments for the information I have derived from him. He has done
that for the History of Mediæval and post-Mediæval sculpture in Italy
which had before hardly even so much as been attempted, and his facts
and conclusions are almost always so stated as to command the assent of
his readers.




PREFACE TO THE FIRST EDITION.


In these days of railways and rapid travelling there is scarcely any
excuse for ignorance of Continental art. The most busy man finds some
short holiday in the course of the year, and, if wise as well as busy,
spends it not in quiet sojourn at home, but in active search of the
picturesque, the beautiful, or the old, in nature or in art, either in
his own country or abroad.

And as the holidays of busy men are short, and therefore to be made as
much of as possible, I conceive that I shall be rendering some service,
and providing myself with a fair excuse for my presumption, if I venture
to shew, by a simple narrative of a tour undertaken in the course of the
year before last, how much it is possible to accomplish with pleasure,
and, when one has some definite object in view, with profit of no common
kind, even in a short holiday.

There are many classes of travellers, and each doubtless flatters itself
that its own is the very best of all modes of travelling; and sorry
should I be to attempt to disabuse any one of so pleasant a self-deceit.
But the more I think of it, the more certain it appears that the reasons
and objects which always take me away from home are precisely such as
make up the sum of happiness and pleasure to a traveller.

Indeed, without some definite object before him, beyond the mere desire
of relaxation and pleasure, few travellers know that thorough joy of
heart which an architect feels as he begins the journey which bears him
away from home on some ecclesiological or architectural ramble.

Such an one, hard-worked for more than five-sixths of the year, may, if
he will, press into the short remainder left to him for a holiday as
much both of profit and of pleasure as it is possible to conceive. He
goes, sketch-book in hand, with some ancient town or thrice noble
cathedral set before him as his goal; and, passing along smiling
valleys, or over noble mountains, drinks in all that he sees, not the
less gratefully or delightedly in that he views it as the preface only
to his more intense enjoyment in the study and pursuit of his own
well-beloved art.

If such be my case--and such it is--wonder not, gentle reader, that I
desire to shew how much enjoyment may be snatched from time in little
more than one short month, nor that I am anxious to put on paper the
thoughts that have been uppermost in my mind as I travelled, and looked
at and drew the old builders’ works in the north of Italy, the more as
they seem to bear with much force upon questions debated with more and
more eagerness and anxiety every day, by very many of those who take the
most lively interest in the progress of Christian art.

In past years I had travelled--rapidly, it is true, but not without
learning much, very much, of what was useful--by the noble cities of
Belgium, up the church-besprinkled banks of the fair Rhine, over the
plains of Bavaria, and through much that was most noble and interesting
in different parts of France and Germany; I had dreamt of old times and
old men in the antique streets of Bruges and Nuremberg, and under the
shade of the still more ancient walls of Regensburg, in the solemn naves
of Amiens, of Köln, of Freiburg, of Strasburg and Chartres, and of many
more most noble piles; I had paced the ruins of old abbeys, and
studied, so far as I could, in all of them the science and the art of my
forefathers; but so far all my time had been devoted to the study of
Northern art, and I had found no time and no opportunity for the study
of that modification of the pointed style which distinguishes the cities
and the churches of the north of Italy. No wonder then that, with a
prospect at last of a first sight of Italy and Italian architecture
before me, I looked forward long and anxiously for the end of summer,
for that happy autumn which brings ease and relaxation to so many a
wearied heart; and that when at last, at the latter end of August, I
found myself absolutely on my way, I was in no common degree disposed
for the thorough enjoyment of all that I met with.

It is well here to observe, by the way, that there is much in the
present position of architects and the world which may give to these few
remarks upon the pointed architecture of the north of Italy--slight and
sketchy though they may be--a degree of value beyond what they would
have had only a short time since.

It is impossible not to feel that the great and general interest in art,
created by the revival of true principles within the last few years, is
a subject of the greatest congratulation to all true artists. It is not
only in architecture, but happily in painting also, that first
principles are now studied with some determination by men who command
the respect of a world educated hitherto to admire and believe in the
falsest and weakest schools of art. It was, therefore, with the desire
to see how far these first principles were worked out by the architects
of the Middle Ages in Italy, how far moreover they were developed in
directions unattempted by their brethren in the North, and how far they
have succeeded in leaving us really noble works for our study and
admiration, that I undertook my journey.

Let me say, too, at the same time, that I started without either the
intention or the desire to examine at all carefully the works of the
Renaissance architects. For this there were many reasons--among others
my own unfitness by predilection and education for the task, the
shortness of my time, and the fact that, as it appears to me, their
works have already received as much both in the way of illustration and
of description as they deserve.

I should wish also, I must confess, in all my studies of foreign
architecture, to confine myself to those buildings in which there appear
to me to be the germs at least of an art true and beautiful in itself,
and of service to us in our attempts to improve our own work. It does
not appear to me that the works of the Italian Renaissance architects
really contain this. I see no reason whatever for doubting that if we
wish for a purer school of art we must either entirely forget their
works, or remember them so far only as to take warning by their faults
and failures. I see no reason for allowing that they have succeeded in
carrying out true principles, either of construction or ornamentation,
to any greater extent than their imitators in England. The same
falseness of construction, and heaviness, coarseness, and bad
grotesqueness of ornamentation, seem ever to attend their works,
together with the same contempt of simplicity, repose, and delicacy
which we are so accustomed to connect with them. In short, I see but
little reason to differ from the estimate which Mr. Ruskin has given of
their merits in the ‘Stones of Venice,’ and what he has so well said I
need not attempt to enlarge upon.

My own feeling is, that as in the pointed arch we have not only the most
beautiful, but at the same time incomparably the most convenient feature
in construction which has ever been, or which, I firmly believe, ever
can be invented, we should not be true artists if we neglected to use
it.

I hold firmly the doctrine that no architect can properly neglect to
avail himself of every improvement in construction which the growing
intelligence of this mechanical age can afford him; but this doctrine in
no way hinders the constant employment of the pointed arch; on the
contrary, it makes it necessary, because it is at once the most
beautiful and the most economical way of doing the work that has to be
done.

There are, I well know, advocates for the round arch, whose theory
appears to be that we ought to go back for some ages, to throw ourselves
as it were into the position of men who knew only the round arch, and
from this to attempt to develope in some new direction: this is Mr.
Petit’s theory, and it is, as appears to me, one which it is not
difficult to meet.

Its supporters assert that pointed architecture is so essentially the
effort of a particular age, and marked by certain peculiarities so
decided, as to be filled, even in its most noble works, with a kind of
spirit which in this age it is vain to attempt again to evoke. The old
Gothic spirit is, they say, dead; and, glorious as it was, its flight
was but meteor-like, and, having passed across the horizon of the world
in its rapid course, it has sunk beyond all possibility of revival.

It appears to me that those who so argue confound the accidents with the
elements of the true Gothic architecture of the Middle Ages, and mistake
altogether the object which, I trust, most architects would propose to
themselves in striving for its revival. The elements are the adoption of
the best principles of construction, and the ornamentation naturally and
properly, and without concealment, of the construction; the accidents
are, as it appears to me, the particular character which individual
minds may have given to their work, the savageness, or the grotesqueness
as it has been called, which is mainly to be discovered in the
elaboration of particular features by some particular sculptor or
architect, and which in the noblest works--and, indeed, I might say, in
most works--one sees no trace of. The true Gothic architects of the
Middle Ages had, in short, an intense love of nature grafted on an
equally intense love of reality and truth, and to this it is that we owe
the true nobility and abiding beauty of their works; nor need we in this
age despond, for if we be really earnest in our work, there is nothing
in this which we need fear to miss, nothing which we may not ourselves
possess if we will, and nothing therefore to prevent our working in the
same spirit, and with the same results, as our forefathers.

The mediæval architecture of Italy presents, however, one further
practical argument against this theory of the lovers of the round arch
which they cannot, I think, meet.

It will be found in the following pages that in Italy there did not
exist that distinction between the use of round and pointed arches which
did exist for three centuries north of the Alps. They were content there
to use whichever was most convenient, and whichever appeared to them to
be most effective in its intended position. We therefore find, in most
Italian mediæval buildings, round and pointed arches used in the same
work, the former generally for ornament, the latter for construction;
and the effect of this is in some degree to make us lessen the rigidity
with which a study of Northern art might otherwise affect our views on
this point. But I think no argument can be used by the lovers of the
round arch which would ever go farther than to leave us open to the
choice of both round and pointed arches, just as in these old Italian
buildings: they have no right to say, “You may not use the pointed arch
at all,” but they perhaps may be allowed to ask, “Why exclude for ever
the round arch?” and then I should refer them to Italy for a proof that
as a rule the mixture of the two is neither harmonious nor satisfactory;
whilst at the same time I should go on to shew them that, when they talk
of the virtues of Roman and Romanesque architecture, of the repose and
the simplicity which distinguish them, of their grandeur and their
general breadth and nobility of effect--in all these things they do but
sing the praises of the best Italian architecture of the thirteenth and
fourteenth centuries, and that in studying the style we may well be
guided by it in what we do, not to the forgetfulness of the glories of
our own land, but to the development in a forward direction of what we
inherit from our forefathers of that architecture which, after a lapse
of three centuries, we now see on all sides reviving with fresh vigour
from its temporary grave, and which requires only prudence and skill on
the part of its professors to make even more perfect than before.

My object therefore in the following pages will be mainly to shew the
peculiarities of the development of pointed architecture in Italy, and
specially to shew in what way the materials so commonly used
there--brick and marble--were introduced both in decoration and in
construction. All these points are of the very greatest importance to
us, for I am persuaded that not only will some reference to Italian
models do somewhat towards the improvement of our art, but that in no
matter is information more needed, and improvement more easy, than in
the use of brick in architecture; whilst working in marble has been as
yet so little practised among us, that we may almost regard it as at
present unattempted, though, as I hope to shew, there is no longer any
reason why this should be the case.

It is impossible to conclude this Preface without mention of the
obligations which not only all who travel in Italy, but all who are
interested in good architecture, owe to Mr. Ruskin. No man need or can
profess his acquiescence in every one of the opinions which he has
propounded, but as an architect I feel strongly that a great debt of
gratitude is owing to him for his brilliant advocacy of many laws and
truths in which every honest architect ought gladly to acquiesce. He may
be well content to bear the opposition which he has evoked, satisfied
that all that he has written is in the main most certainly for the
benefit and exaltation of art of all kinds.

Nor less is a debt of gratitude to be acknowledged by every traveller to
my friend Mr. Webb for his most excellent and trusty work on
‘Continental Ecclesiology:’ it is certainly the most absolutely correct
guide-book ever drawn up for ecclesiologists anywhere; and in travelling
over the same ground, as I have done in this tour, my excuse for giving
what I have in the way of descriptions of the same buildings is, that
what I have written has been all with a view, beyond that of merely
describing the churches, of shewing the principles upon which their
builders worked, and giving, so far as the limits of such a work will
allow, drawings of the buildings I have described.

It will depend on circumstances whether I am able at some future day to
continue my inquiries among the churches and domestic buildings of
Central Italy, a tract at least as rich as that over which the tour
described in the following pages took me.

It remains only to say that all the illustrations which I have given are
engraved from my own drawings on the wood from my sketches made on the
spot, and that I have endeavoured as much as possible to avoid giving
subjects which have been before published. It would have been easy to
add largely to them, especially from my sketches in Venice, but it
seemed to me that, as this could only be accomplished by adding also to
the cost of the book, it was much better to omit them. I have avoided
therefore giving drawings of any buildings already drawn by Mr. Gally
Knight, to whose work I must refer my readers for representations of
several of the buildings described, and for illustrations of Venice I
must refer to Mr. Ruskin’s engravings and to the photographs which have
rendered her features so well known to almost all students of
architecture.

In conclusion, I cannot speak too highly of the assistance afforded to
the architectural student by Murray’s Handbook of Northern Italy: it is
almost invariably correct, and gives just what one wants to know of
nearly all buildings of any interest or importance.

_Oxford, 1855._




CONTENTS.


 CHAPTER I.

 .....PAGE

 Routes to Italy--Paris--Strasburg--Rouffach--Basel.....1

 CHAPTER II.

 Churches of Basel--Storks--Rheinfelden--Frick--Baden--Zürich:
 the Cathedral--Fondness of the Swiss for bright colours--Lake of
 Zürich--Rapperswyl--Linth Canal--A Wayside Inn--Wesen.....14

 CHAPTER III.

 Wallenstadt--Sargans--Gorge of the
 Tamina--Ragatz--Chur--Ems--Reichenau--Thusis--Zillis--Andeer--Splügen--The
 Splügen Pass--The Custom-house--Cascade of the Medessimo--Campo
 Dolcino.....28

 CHAPTER IV.

 Chiavenna--Lake of Riva--Colico--Gravidona--Lake
 of Como--Varenna--Stelvio Pass--Lecco--Bergamo:
 Broletto--Churches--Castle of Malpaga.....44

 CHAPTER V.

 Palazzuolo--Coccaglio--Brescia: new and old
 Cathedrals--Broletto--Churches--Donato--Desenzano--Lago di
 Garda--Riva--Trent--Verona.....63

 CHAPTER VI.

 Verona: Campanile of the Palazzo dei Signori--Sta.
 Anastasia--Monuments--Piazza dell’ Erbe--The Duomo--The
 Baptistery--Sta. Maria l’Antica--Cemetery and Palace of the
 Scaligers--Domestic Architecture--Piazza di Brà--The Austrians--Ponte
 di Castel-Vecchio--San Zenone--San Fermo Maggiore--Chapel near the
 Duomo--Romeo and Juliet--Dwarfs--Wells.....84

 CHAPTER VII.

 Neighbourhood of Verona--Vicenza: Cathedral--San Lorenzo--Santa
 Corona--Palazzo della Ragione--Gothic Palaces--Palladio’s
 Works--Teatro Olympico--Padua: Giotto’s Chapel--The
 Eremitani--Sant’Antonio--The Duomo.....126

 CHAPTER VIII.

 Padua and Venice Railway--Venice: Piazza and Church of
 S. Mark--Torcello--The Lagoon--Murano--Sta. Maria dei
 Frari--SS. Giovanni e Paolo--Sta. Maria dell’ Orto--Other
 Churches--Domestic Architecture--Fondaco de’ Turchi--Other
 Byzantine Palaces--The Ducal Palace--Foscari Palace--Ca’
 d’Oro--Other Gothic Palaces--Balconies--Venetian Architecture--A
 Festival--Paintings.....149

 CHAPTER IX.

 New Roads to Venice--The Pusterthal--Innichen--Dolomite
 Mountains--Heiligenblut--Kötschach--Kirchbach--Gail
 Thal--Hermagor--Ober Tarvis--Predil
 Pass--Gorizia--Aquileja--Grado--Udine--Pordenone.....238

 Chapter X.

 Venice to Verona--Verona to Mantua--Villa
 Franca--Mantua: its Churches and Palaces--The
 Theatre--Montenara--Campitello--Casalmaggiore--Longadore--Cremona:
 the Cathedral--Churches and Public Buildings--Lodi--Pavia: its
 Churches--Castle of the Visconti--The Certosa--Drive to Milan.....253

 CHAPTER XI.

 Drive from Padua to Ferrara--Monselice--Rovigo--Ferrara:
 Cathedral--Castle--Gallery--Road to Bologna--Altedo--Bologna:
 Cathedral--San Petronio--San Domenico--San Giacomo--Sta. Maria
 Maggiore--San Francesco--San Stefano--Leaning Towers--Casa dei
 Mercanti--Domestic Remains--Academy--Modena: Cathedral--Parma:
 Cathedral--Correggio--The Baptistery--Piacenza: The Palazzo
 Publico--Cathedral--San Francesco--Sant’ Antonino--San Giovanni in
 Canale--Asti: Cathedral--San Secondo--Campanili.....286

 CHAPTER XII.

 Milan: the Cathedral--Sant’Ambrogio--Sant’Eustorgio--Sta. Maria
 delle Grazie--Certosa of Chiaravalle--Novara--Vercelli--Monza: the
 Cathedral--The Broletto--Sta. Maria in Strada--Como: the Broletto--The
 Cathedral.....315

 CHAPTER XIII.

 Departure from Como--Varese--Lake of Varese--Italian
 Boatmen--Intra--Laveno--Lago Maggiore--Magadino--Road to
 Hospenthal--The Dazio Grande--Airolo--Hospenthal--Ascent of the
 Furca--Valley of the Reuss--Lake of Luzern--Luzern--The Unter
 Hauenstein--Strasburg.....344

 CHAPTER XIV.

 Concluding Summary--Classic and Gothic Architecture--Italian
 Gothic--Shafts--Cornices--Monuments--Cloisters--Windows--Brickwork--Colour
 in Construction--Truth in Architectural Design and Construction.....361

 APPENDIX.

 Catalogue of the Subjects of the Sculptured Capitals in the Lower
 Stage of the Doge’s Palace, Venice.....409





LIST OF ILLUSTRATIONS.

(THE FULL-PAGE ENGRAVINGS ARE NUMBERED IN ORDER.)


                                            PAGE

 1. _Frontispiece._ N. Porch, Sta.
    Maria Maggiore, Bergamo.
    Shop-window, Rheinfelden.....16

 2. Cloister of Zürich Cathedral....._to face_19*

    Church on the Lake of Zürich.....22

    Wooden spire, Ragatz.....32

 3. Interior of Cathedral, Chur....._to face_33

    Gravidona, ground-plan of Baptistery.....48

 4. Gravidona, the Baptistery    ....._to face_.....48

 5. The Broletto, Bergamo.....“.....54

    Campanile, Bergamo.....59

 6. Malpaga Castle....._to face_62

    Window, Coccaglio.....64

 7. House at Coccaglio....._to face_65

    Detail of windows and corbelling
    for chimneys, Coccaglio.....65

    Wooden balcony, Coccaglio.....66

 8. Broletto, Brescia....._to face_69

    Cloister, Broletto, Brescia.....69

    Detail of circular window, Broletto,
    Brescia.....70

    Doorway, Broletto, Brescia.....71

    Brick cornice, Broletto, Brescia.....71

    San Francesco, Brescia.....72

 9. Details of brickwork in Broletto....._to face_72

    Cloister of the Carmine, Brescia.....74

10. Duomo, Trent, Eastern doorway....._to face_81

11. Campanile of Scaliger Palace,
    Verona....._to face_84

    Plan of Sta. Anastasia, Verona.....87*

12. Pavements, Sta. Anastasia, Verona....._to face_89

    Sta. Anastasia, Verona, aisle
    window.....90

    Door-frame, Sta. Anastasia.....92

    Door-frame, San Pietro Martire.....94

    Crest of metal railing, Verona.....102

    Metal railing, Verona.....103

13. Courtyard of the Scaliger Palace,
    Verona....._to face_104

    Doorway, old house, Verona.....105

    Windows, old house, Verona.....106

    Brick battlement, Viccolo Cavaletto,
    Verona.....106

14. Courtyard of old house, Verona_to face_106

15. Cloister of San Zenone, Verona....._to face_111

16. Interior of San Zenone.....112

17. San Fermo Maggiore, Verona,
    west front....._to face_121

18. Italian brickwork.....122

    Domestic window, Verona.....123

19. Porta di Castello, Vicenza ....._to face_128

20. Houses in the Contrada Porto,
    Vicenza....._to face_131

    Arena Chapel, Padua, west end.....137

    Arena Chapel, side window.....139

    Sant’Antonio, Padua, ground-plan.....145*

21. Sant’Antonio, view from the
    east....._to face_147

    S. Mark’s, Venice, ground-plan.....156*

22. Duomo, Torcello, ambon and
    screen....._to face_168

    Sta. Fosca, Torcello, east end.....171

23. Interior of Sta. Maria dei Frari,
    Venice....._to face_177

24. Exterior of Sta. Maria dei Frari,
    Venice....._to face_180

    Window, San Stefano, Venice.....186

25. Campanile, San Giacomo del
    Rialto....._to face_187

26. Byzantine well, Venice.....“ 193

27. Corte del Remer.....“ 196

28. View of Venice, from the Romance
    of Alexander, Bodleian
    Library....._to face_204

29. Archway, Ponte del Paradiso....._to face_210

30. Doorway on the Ponte San Tomà....._to face_212

    Brick battlement, Venice.....214

31. Angle window....._to face_216

    Balcony, Venice.....217

    Capital of window-shaft, Venice.....220

32. Palazzo Segredo....._to face_221

33. Window, Ponte del Fornaro.....“.....222

34. Staircase, Casa Goldoni.....“.....223

    Venetian chimneys.....224

    Aquileja, Patriarch’s throne.....243

35. Aquileja, interior of Duomo ....._to face_244

36. Udine, Palazzo Publico.....“.....248

37. Udine, steps to Palazzo Publico....._to face_249

    Udine Cathedral, aisle windows.....250

38. Udine, tower and cathedral doorways....._to face_250

39. Ducal Palace, Mantua.....“.....255

40. Windows in ditto.....“.....256

41. Castello di Corte, Mantua.....“.....257

    Brick window, Sant’Andrea.....258

42. Gateway, Palazzo della Ragione....._to face_258

43. Campanile of Sant’Andrea....._to face_259

    Brick window, Sant’Andrea.....259

    Brick window, Campitello.....262

    Brick window, near Casalmaggiore.....262

44. North transept, Duomo, Cremona....._to face_266

    Brick window, Duomo, Cremona.....267

    Rose window, ditto.....268

45. Palace of the Jurisconsults,
    Cremona....._to face_269

    Window-jamb, ditto.....269

    Chimney and battlement, ditto.....270

46. San Michele, Pavia, interior....._to face_276

47. Castle of the Visconti, Pavia.....“.....277

48. Bay of courtyard of ditto.....“.....278

49. San Pantaleone, Pavia.....“.....279

50. West end of San Francesco, Pavia....._to face_281

51. Certosa of Pavia.....“.....284*

52. Ferrara, the Duomo....._to face_288

53. San Petronio, Bologna, south
    aisle....._to face_291

    San Petronio, Bologna, plan.....292*

    San Petronio, Bologna, section.....293*

54. San Petronio, Bologna, interior....._to face_294

55. Monument near San Domenico,
    Bologna....._to face_296

    Cloister, San Stefano, Bologna.....298

    Brick window, Palazzo Publico,
    Bologna.....301

56. South front of Duomo, Modena....._to face_304

    Baptistery, Parma, plan.....307*

    Baptistery, Parma, section.....307*

57. Piacenza, view of Palazzo Publico....._to face_308

    Brickwork, Palazzo Publico, Piacenza.....310

    Sant’Antonino, Piacenza, plan.....312*

    Sant’Antonino, Piacenza, section.....313*

    Duomo, Milan, ground-plan.....317*

58. Sant’Ambrogio, Milan, baldachin....._to face_326

59. Piazza dei Mercanti, Milan.....“.....329

60. Certosa of Chiaravalle.....“.....330

61. Sant’Andrea, Vercelli, interior....._to face_332

    Window in Duomo, Monza.....335

62. The Broletto, Monza....._to face_337

    Window in Broletto.....337

63. The Broletto, Como....._to face_340

64. Sta. Maria, Como.....“.....342

    Cornice, San Francesco, Brescia.....379

    String-course, Palace of Jurisconsults,
    Cremona.....391

    Window in north transept, Cremona.....392

    Detail of window-jamb, Cremona.....393

    Brick archivolt, Vescovato, Mantua.....394

    Arch-mould, Cremona.....394

    Window, Verona.....396

    Brick window, Sant’Andrea,
    Mantua.....397

65. Key-plan of Capitals in Doge’s
    Palace, Venice....._to face_409

⁂ The engravings marked * are taken from Mr. Fergusson’s ‘History
of Architecture.’




BRICK AND MARBLE

IN THE

MIDDLE AGES.




CHAPTER I.

    “Yet waft me from the harbour-mouth,
       Wild wind! I seek a warmer sky,
       And I will see before I die
     The palms and temples of the South.”
             _Tennyson._

Routes to Italy--Paris--Strasburg--Rouffach--Basel.


An architectural tour in Italy seems to afford about as much prospect of
pleasure and information combined as any which it is possible for an
English student to take. He may see, if time allows, so much on his
road, that whether one thinks of the journey or the end of it all is, at
any rate in the perspective, charming. And in these days when, what with
railways, through-tickets, and Cook’s and other guides for timid
tourists, the journey from one end of Europe to the other is made so
quickly and so cheaply as to be within most educated men’s reach, it is
no wonder if most of us in our turn make the venture.

Many are the ways by which one may reach the North of Italy, but one or
two only of them seem now to be commonly used, to the exclusion of all
others, and with great loss of pleasure to all travellers who make the
journey more than once. The natural, because the quickest, road is now
by the Mont Cenis tunnel to Turin, and for the country described in
these pages nothing can be more convenient. But when my first journey
was made, it was more easy to take one of the passes leading to Milan,
and so I went by the Splügen. Since those days I have found my way to
and from Italy by other roads which I recommend strongly to others. I
pass by such a well-known road as that by railway over the Brenner, in
order to suggest three other roads, either of which brings the traveller
down upon Venetia in the happiest possible frame of mind if he is at all
capable of being moved to pleasure by the sight of exquisite scenery,
pleasant and religious people, and roads and country not too much
crowded with tourists.

The first of these is by the Lake of Constance, the Vorarlberg, and the
Vintschgau to Botzen; the next by the Brenner pass as far as
Franzensfeste, and thence by the Ampezzo pass through Cortina and
Cadore--Titian’s country--to Conegliano, and so by railway to Venice, a
road lighted up by the wild beauty of the Dolomite mountains and now
unaccountably neglected by English tourists; the third, and perhaps the
most charming, though somewhat indirect, and requiring more time, again
by the Brenner as far as Franzensfeste, thence by railway to
Lienz--stopping to see the fine church and Dolomite mountains at
Innichen on the way--and then by country carriages from the Pusterthal
into the Gailthal where there is the most charming combination I know of
pastoral and picturesque scenery, seasoned by interesting old churches;
and thence to Ober Tarvis and by the stern and magnificent Predil pass
to the head of the Adriatic at Gorizia, whence--after seeing Aquileja
and Grado--the traveller may, with halts at Udine and Pordenone, reach
his goal at Venice by railway.

But in this my first journey to Italy, I was sufficiently happy in
finding the Splügen prescribed for me as on the whole the most
convenient mode of reaching in succession all the spots which had most
special interest for me. My scheme was to make myself fairly well
acquainted with some of the most interesting Italian cities north of the
Apennines, and for this purpose to descend from the Splügen on Bergamo,
and from thence to go on to Venice, halting as often as necessary by the
way, and then to return by Mantua, Cremona, and Pavia, or by Ferrara,
Bologna, Parma, and Piacenza to Milan, and so home. And railways, if
they have made the journey somewhat more easy than it was, and have
deprived it now and then of the charm which always attends the
recollection of impediments and difficulties on the road, have not in
any way altered the advantages of such a route for those whose tastes
are at all akin to those which I carried abroad with me in those days,
and carry still with undiminished strength. Whether, however, one enters
Italy by one pass or another, the first part of one’s journey is by the
well-worn road to Paris, which, by reason, I suppose, of its being the
prelude to nearly every holiday tour that I make, never seems to be
stale, old, or too well-known.

There is something very novel, and it strikes me more every time it is
seen, in the aspect of everything directly you have crossed the Channel;
indeed, there is no country in Europe so much as France, and no city,
perhaps, so much as Paris, which strikes an Englishman as being foreign
in its aspect, and new in all its customs and proceedings. The dress of
every one, the arrangements of the railways, the harnessing and
character of the horses, the mode of life in hotels, and the ordinary
habits and pleasant traits of the middle classes are all quite fresh to
the English eye. Nor is the aspect of the country less so: fields cut up
into small strips of a dozen kinds of crops; unprosperous-looking cows,
each feeding discontentedly and drearily, tethered to a man or woman on
a small patch of grass; corn cut and then stacked in small cocks for a
month or two of exposure to the pleasant changes of the atmosphere; and
the entire absence of hedge-rows and other trees than poplars, all go to
make up a thoroughly un-English picture.

After skirting the coast and its dreary expanse of sandhills, reminding
one very much of those singular sands on the north coast of Cornwall,
which are so often shifting about, covering up new churches, or
uncovering the old oratory of some early British saint, we reach the
banks of the Somme, and then travel along a poor peaty tract of country
until the famous west front and short but lofty nave of Abbeville come
in view. Thence by a valley (rather more rich than is common in good
churches) we continue our race for Amiens. Among these churches I may
instance the hipped saddle-back roofed steeples of Picquigny, Hangest,
and Pont Rémy, as very valuable examples of their order; that of
Picquigny, indeed, surmounting a central steeple, and finished at the
top with some delicate open ironwork, is about as graceful a specimen as
I know.

At Longpré is another church with a steeple of some pretension, but not
satisfactory. It has a perforated spire of stone much too small for the
size of the tower, and ungraceful in the extreme.

At Amiens one always longs to stop again and again to feast one’s eyes
upon its glorious cathedral, perhaps after Chartres and the Parthenon
the noblest and most masculine piece of architecture in the world. But
with us this was impossible; our destiny was--come what might--to
endeavour at any rate to discharge ourselves in Paris within the
shortest possible number of hours from London; and the dusk of the early
autumn evening prevented our having more than the very slightest glimpse
of the Minster.

The refreshment-room at Amiens is one of the best I have ever been
in--reasonable, clean, and good--and placed just at that happy distance
from the sea at which the poor wretches who have been in the depths of
woe on the passage begin to recover their presence of mind, and with it,
of course--as good Englishmen--their appetites; what wonder then if the
Buffet at Amiens prospers!

The rest of our journey to Paris was all performed in the dark, relieved
only by the sight of the then long-expected comet, and it was almost
midnight ere we found ourselves settled at our hotel.

I am never sorry to have a day in Paris. In spite of alterations and
reconstructions which have converted an interesting old city into the
most spick-and-span place in the world, there are even to the present
day parts which are untouched by the improver, and full of a pleasant
national character which seems to be little to the liking of the rulers
of the French. There is, too, in spite of the changes which a great and
rich city must always undergo, a great deal which is interesting to the
architect. We may look at old engravings, and wish ourselves back in
those old times when the walls surrounded the city where now the
Boulevards run round its heart, when the Temple and a number of other
important buildings, now wholly destroyed, adorned the country just
outside the walls; but the city which has still, among other
architectural treasures, such churches as Notre Dame, S. Germain des
Près, the Sainte Chapelle, S. Martin des Champs, and a host of lesser
lights, and the Chapel of Vincennes, and S. Denis within a short drive,
is in quite a different category from such a city as London, and is
indeed hardly second to any other in Europe in architectural interest.

To come to much later times and very different work, it is always
pleasant to be able to walk down the Boulevard des Italiens to the
Madeleine, and for a few minutes to gaze at a church which certainly
presents one very grand idea--that of space--clothed in very gorgeous
dress. One always feels a certain sympathy for a church in which so many
people are ever praying; and I have never yet been into this church
without being able to count them by scores. The last time I was at Paris
I remember being struck by seeing for the first time a peripteral
building made really useful. The walls within the columns were hung with
rich draperies, and a long procession coming out marched round the
circuit of the church between the columns and the walls, and in again at
the west door; the effect was, as may be imagined, very striking.

From the Madeleine we found our way to the new church of S. Clothilde, a
large cruciform church, and the last erected in Paris in the Gothic
style. Its design is intended to be of early character, but in reality
is quite late in its effect; nor do I know when I have seen anything
much less successful than the two western steeples rising but a short
distance above the nave roof, and looking mean and weak to a degree. In
plan the church is not badly arranged; there is just such a choir as
might easily be properly used, and a large space for congregational
purposes.

How much we want churches, in this respect at least, somewhat like S.
Clothilde, in our large cities in England!

There are here a great many windows filled with stained glass, executed,
I believe, by Mons. Marischal. His windows are illustrations of a truth
which men are very slow to receive and act upon, viz. that in decorating
a transparent material, one whose transparency moreover is the sole
cause of its use, we have no right to shade it with dark colours so far
as to destroy its brilliancy. These windows were elaborately shaded,
and, as a necessary consequence, were heavy and dismal in their effect;
besides which, most unpleasant mixtures of green, yellow, and ruby, and
of ruby and blue--very glaring and very bad--abounded.

The carving of the capitals is, as is usually the case in recent foreign
works, all derived from natural types of foliage, and is fairly well
done: but the carving of rather elaborate sculptures of the “Stations”
did not please me, having none of the severity of ancient examples. When
shall we see a school of sculptors rise able really to satisfy the
requirements of the times? I confess I despair more on this point than
on any other; for I have as yet seen no fair attempt made to recover the
style, or work upon the principles, of the best mediæval sculptors. The
work of our modern sculptors is nearly all foreign and unreal, and
almost always involves the assumption that they are representing the
proceedings of the Greeks or Romans, and not of the English: it is
impossible therefore that such a school can be healthy, strong, or
successful. We lack men who will give us (clothed with as much
anatomical correctness as they like, so that they do not leave them
lifeless and academical) representations of subjects from English
history and national life, illustrations of the Scriptures which we
still believe, of the faith which we still profess, conceived in
something of the architectonic and yet really dramatic and romantic
spirit which marks the best sculpture of the middle ages. The strange
thing is that with works near at hand which few living men could rival,
they absolutely refuse to study them at all, and I believe if we were to
summon all the most eminent sculptors to a conclave and put them to the
question, not one in four of them would confess to having ever been to
Chartres or Bourges, and four out of five would assert that it would
have done them no good if they had. If they would give us anything at
all comparable to the great works of the best Greeks the case would be
altogether different, but to be served with a réchauffé of the antique
when one is crying out for something suitable to the present, is cause
enough for the apathy of the English public about sculptors’ work. We
ask for English history or Bible story, and are treated to nymphs
combing their hair; and for figures of our Lord and St. Peter, and get
nothing but Musidoras and Clyties. No sculptor would lose much by the
study of the best mediæval examples of drapery--and there are among the
gothic statues which deck the doors and porches of the churches I have
named, some of the most admirable description, such as warrant any one,
who is at all troubled with feeling for his art, in using strong
language about those who neglect them. In Italy we shall find the same
careful shutting of men’s eyes to what is good, simply because it
belongs to the thirteenth or fourteenth centuries. Orvieto is left on
one side in order to spend time over work not possessed of a tithe of
the beauty of that on its cathedral façade; and, indeed, just as the
French examples, they appear only too often never even to have been so
much as heard of!

The study of ancient sculpture in England is not quite so easy, because
our old buildings are not so rich in it as are the French; but if one is
told--as one is too often--that the art of sculpture in the middle ages
was unknown or rude in comparison with its state now, one may fairly
refer to some of the modern attempts at its imitation for a proof that
this was not the case, as e.g. to the recumbent effigy of Archbishop
Howley at Canterbury, or to another, of some more humble individual, in
the south transept of Chichester Cathedral; a glance only at which, and
a comparison with some of the noble mediæval effigies lying in all the
stateliness of their repose by their sides, will at once show any one
that it is not merely necessary to put an effigy upon its back with its
hands in prayer in order to vie with the effigies of the thirteenth and
fourteenth centuries. The position is something, but not all, and
requires very much more skill in its treatment than of late years we
have had to bestow.

From S. Clothilde we went first to the pleasant gardens of the
Luxembourg--gardens which always make one envious for London--and thence
to Notre Dame. Here I always feel no slight pride in the success which
its architect has achieved. Six hundred years have passed over Paris,
one effort after another has been made, vast sums of money have been
spent, and still this great work stands supreme and separated by a vast
distance from all competition, and greatest beyond comparison of all
Parisian buildings, not only in its general scheme, but equally in the
admirable design and execution of every detail. There is much to be seen
and learnt here in every way. The west doors are superb. The planning
and construction are very fine, and the series of sculptures behind the
stalls full of interest and well worthy of study.

From Notre Dame one goes, of course, to the Sainte Chapelle. When this
journey was undertaken everything about this chef-d’œuvre was
gradually growing to perfection: the flèche was being put up on the
roof, the painting on the walls was nearly finished, and the altar was
in progress. Since then it has escaped, as it were by a special
providence (and why not?), from the incendiary fire which destroyed
almost the whole of the surrounding Palais de Justice, and it still
rises uninjured among the ruins. Of all the chapels of the same kind it
is certainly the most beautiful--and whether one names our own St.
Stephen’s, or thinks of others, such as the Chapel at S. Germer and the
other at Riom, the Paris chapel is certainly by far the finest--being in
truth a real work of inspired genius.

Altogether, I cannot help thinking that the effect upon the mind of what
one sees in Paris is very unsatisfactory; the revival of Christian art
seems, as it were, to be only skin-deep; there seems to be no enthusiasm
for it. What is done is done in the same way as other public works, as
the business of the state, not by the will of the people. The
scaffolding, which was just being removed from the avenue leading from
the Tuileries to the Barrière de l’Etoile, after having assisted at the
fête of Napoleon, was an illustration sufficiently apt of the work which
seems to engage too many of the artists of Paris; Parisian fête
composers and decorators really appear to be the architects of the day,
and of course this fact must mitigate very much against real art in
every branch, as its tendency is to make people accustomed to temporary
exhibitions, the shortcomings of which are pardoned on the score of
their temporary character, and so the artist is lowered in his tone by
assisting in the production of works which are not intended--as all
great works ought to be intended--to last for ages.

A day in Paris is generally a long and tiring one; and so we found it;
but nevertheless we pushed on without delay, and leaving our hotel
before the table-d’hôte was much more than half over, we drove to the
station of the Strasburg Railway, and in a few minutes we were _en
route_. If any one doubts the possibility of really resting one’s body
in a railway carriage, let him take the same precaution that we took,
and he need not despair: a day of sight-seeing in Paris is certainly the
best possible recipe for sound sleep in a railway carriage, and I
believe that when we arrived at Strasburg, at about eight the next
morning, we were very fairly rested. I confess, however, that I did feel
a twinge of horror when I found that the train by which we were anxious
to reach Basel left again in about half an hour--too long to wait, but
not long enough for either breakfast or dressing. There seemed, however,
to be no alternative, and so on we went, comforting ourselves as best we
might with some sour grapes and bad dry bread--the sole edibles
procurable at the Strasburg buffet!

The Railway from Strasburg to Basel is much more enjoyable than
iron-ways generally are. There is scarce a cutting during the whole
extent of the journey, and the views of the chain of the Vosges
are--before one has gazed on real mountains in Switzerland--very
delightful.

The railway runs up the broad valley of the Rhine, and within a few
miles of Strasburg approaches very near to the mountainous district. The
outlines of the hills are bold, picturesque, and well varied; and, as
they rise rather precipitously from the valley, are often crowned with
ruined castles, and have on their lower slopes large and
populous-looking villages, they are at any rate very pleasing neighbours
for a railway journey.

A few architectural notes of such churches as are passed on this route
(which I travelled not for the first time) will not be out of place,
though, with one exception, there is not anything of great value.

At Schlestadt there is a large tower of late date to the principal
church, which is rather fine in its effect. It has its two upper stages
nearly similar, which is rarer at home than abroad. Another church has
an early spire; and there is a smaller church with a good open turret.
Opposite Schlestadt the chain of the Vosges is very striking, and some
of the picturesque outlines of hills capped with ruined castles remind
one of the more famous banks of the lower portion of the Rhine. Beyond
Schlestadt we reach Colmar, the cathedral of which is large, and has a
late tower capped with an ugly bulbous roof. Another church in Colmar
has a good open-work and very light turret rising from the middle of the
length of its roof. The effect of this kind of turret, of which we in
England have no examples, is always very satisfactory.

But the best church in the whole extent of this journey is that of
Rouffach, one on whose merits ‘Murray’--whose services all travellers
must gratefully acknowledge--is silent. It is of early date, cruciform
in its plan, and the crossing surmounted by a good early tower and spire
of octangular form. Each side of the tower has a good window, above
which a string-course forms the base to a gable on each side. The angles
of the spire spring from the bases of these eight gables, and the whole
design reminded me somewhat of the only example of the same type in
England--the beautiful steeple of Lostwithiel. Rouffach has a good choir
terminating in an apse, and a south-western steeple, surmounted by a
slender spire too small for the tower. Altogether, the general effect of
the church is very fine. Beyond this point there are no features of
interest; the Vosges retreat into the distance, and nothing is to be
seen but a dead flat of field and wood, relieved occasionally by a
village or town, remarkable mainly for the ugliness of its church. The
busy manufacturing town of Mulhausen is passed, the number of stations
is carefully reckoned, and long before you catch the first view of Basel
you are heartily sick of the slow pace at which the Strasburg and Basel
Railway Company always arrange to carry their passengers.

Those who know the Hotel of the Three Kings at Basel will understand how
grateful was the information given to us, as we mounted its steps, that
the table-d’hôte was to be ready in half an hour. Refreshing enough at
any time, such an announcement was doubly so to travellers just arrived
from a journey from Paris without a stoppage; and in no bad spirit did
we enter the salle à manger, whose windows, opening into balconies which
absolutely overhang the great and glorious Rhine, flowing strong and
quick for ever in the same unceasing current, make it about the
pleasantest room of the kind that I know.

There are few things in the world so fine as a mighty river, few rivers
so fine as the Rhine, and few spots so favourable for its contemplation
as the balcony at Basel. As you look at the deep colour of the water,
you think of all the wonders which on its way it has seen. You remember
your own exploits and pleasant walks in past times along the lovely
valley of the Aar, and over the barren and stony waste of the Grimsel,
to the source of this beautiful feeder of the Rhine; or you think of
Lake Constance and Schaffhausen, and of the beautiful valley of the
Upper Rhine, and of the lakes of Wallenstadt, Lucerne, Brienz, and
Thun--every one of which seems to the mind’s eye to be represented and
brought near by each wave that dashes madly along before your gaze. And
then, whither do they all so swiftly wend their way? Down by minsters
and by castles, along broad plains, through narrow water-worn chasms,
and again through great, dreary, but many-peopled flats, into the sea,
there to mix themselves and all their recollections in the great,
glorious, but tradition-despising depth of Old Ocean.




CHAPTER II.

    “For pallid autumn once again
     Hath swell’d each torrent of the hill;
       Her clouds collect, her shadows sail,
       And watery winds that sweep the vale
     Grow loud and louder still.”
                _Campbell._

     Churches of Basel--Storks--Rheinfelden--Frick--Baden--Zurich: the
     Cathedral--Fondness of the Swiss for Bright Colours--Lake of
     Zurich--Rapperswyl--Linth Canal--A Wayside Inn--Wesen.


At Basel we engaged a voiturier to take us to Baden, whence the only
Swiss railway was to have the privilege of conveying us to Zurich. Our
scheme for reaching Italy was to pass by the lakes of Zurich and
Wallenstadt, and then, following the valley of the Rhine, to cross over
the pass of the Splügen to Chiavenna, and so to reach Lake Como.

We left Basel at two o’clock in the afternoon, hoping to reach Baden by
about nine; the weather looked threatening, but we took a cheerful view
of this, as of everything else, as all good travellers should, and
comforted ourselves with the thought that at any rate we could better
afford to have a wet day between Basel and Baden than between Zurich and
the Splügen.

The view of the city as you leave it is certainly very striking; the
cathedral spires are picturesque in their outline, and the number of
churches with turrets and steep roofs combine with them to produce a
most ecclesiastical-looking town. Nor need any one interested in
architecture despair of finding much pleasure in a more careful
inspection of its buildings. They are full of interest, though generally
passed too rapidly by people in a hurry to get on to enjoy the pleasures
which await them beyond.

The roofing of the cathedral is worthy of notice as being composed of
variously coloured tiles, arranged in diamond patterns over the surface
of the roof, and giving a degree of richness to the colouring of this
generally heavy part of the building which is very admirable.

In another fine church of the early part of the fourteenth century here,
I remember being amused to see how quietly the storks possess themselves
of all kinds of places for their nests, and think even the ridge of the
steep roof of a church a proper place for their abode. The good people
at Basel build their chimneys with flat tops for the express benefit of
their long-legged friends; who, from their elevated and well-warmed
abodes, look down sedately, and with a well-satisfied air upon their
unfledged brethren below.

Why the people here love storks, the people of Venice pigeons, and the
people of Berne bears, I leave to more industrious inquirers to decide,
satisfied only to notice the fact that it is so, as each of these
fancies adds one to the list of local peculiarities so valuable in the
recollections of a journey.

The road from Basel to Baden is for the first half of the way very
pretty; we came in, unfortunately, for rather drenching rain, and so
lost all beyond the suggestion of some striking views. The towns through
which we passed were not of much interest, though there were many
picturesque and pleasant-looking subjects for the pencil. The most
striking place on the road was Rheinfelden, a largish village (or
perhaps I ought to say small town, as it rejoices in a Rath-haus of
some pretension), surrounded by very high walls, and entered by tall
stone gate-towers, pierced with pointed arches, and surmounted by upper
stages of timber, with tiled roofs of quaint and effective character;
and here and at Stein and Baden I noticed that almost all the houses
were old and very little altered. I observed particularly the old
shop-windows of very simple design, closed with folding shutters, and
taking one back to old times most decidedly in their design.

[Illustration: SHOP-WINDOW, RHEINFELDEN.]

Beyond Rheinfelden the road, which so far has skirted the Rhine rather
closely, leaves it again for a few miles until it touches it for the
last time at the small town of Stein.

From Stein we saw an imposing-looking church on the other side of the
river at Sekingen. It has a great western front with two bulbous-topped
steeples, and is of very considerable length. The division between choir
and nave is marked by a delicate turret, and the whole church, as far as
one can judge by a distant view, looks as though it would well repay a
visit. There are six bays in the nave, five and an apse in the choir.
The former has very simple windows, whilst in the latter they are rather
elaborate. There is no aisle to the choir and no transept.

The rain continued incessantly until we reached the long straggling
village of Frick, a quaint and antique-looking place, where our
voiturier stopped for an hour to bait his horses, who, however, at
Rheinfelden had enjoyed a treat in the shape of a loaf of very brown
bread, a kind of food second only, in the estimation of foreign steeds,
to the precious _morceaux_ of lump sugar with which Swiss voituriers are
so fond of encouraging and petting them.

We were nothing loth to stretch our legs; and finding that the church
was worthless--one of those unhappy bulbous-spired and bulbous-roofed
erections so common in some parts of the Continent, and the roof of even
the eastern apse of which was twisted into a most ingenious and ugly
compound curve--we took up our quarters in the respectable hostelry and
“Bierbrauerei” of the Angel, and devoted ourselves to the consumption of
coffee and beer of no bad quality. Our host wished sadly to see us
located under his roof for the night, but we were resolute in our
determination to reach Baden that night, and so persisted in going,
though to our subsequent regret.

It was soon dark, and the new moon, which shone cheerfully upon us, gave
us just a glimpse occasionally of the scenery, which about Brugg, where
we crossed the Aar, and again at Königsfelden, seemed to be remarkably
good.

At last, at about half-past ten o’clock, we reached what we fondly hoped
was to be our resting-place. But Baden chose not to take us in, and to
our horror, as we drove up to the chief and only available inn, we were
met with the dismal announcement from the mouth of the civil landlord,
that all the rooms were full.

However, we dismounted, and found that there was no other inn in Baden
proper, but that at the Baths there were several; at them our landlord
assured us that he knew we should find no room, and so we thought it
useless to return and try. Our only course seemed to be to feed our
horses again and then go on to Zurich; and as Swiss drivers and Swiss
horses never seem to tire of trotting on slowly and drowsily along the
road, there was no difficulty in at once coming to an arrangement with
our coachman.

Accordingly, at midnight we started again, hoping at some early hour in
the morning to reach Zurich. It was sufficiently provoking to be toiling
on slowly and sleepily for nearly four hours almost alongside of a
railroad which would have taken us early the next morning in
three-quarters of an hour; but there was no help for it, and so we did
the best we could, by sleeping whenever we were able, to pass the weary
hours away.

At last, just as the day began to dawn, we came in sight of Zurich and
its lake, and last, not least, we reached the great hotel. Here we
pulled up, knocked desperately, awoke the slumbering porter--but, alas!
only to hear again the unwelcome sounds which had greeted our ears at
Baden! He suggested, however, that at the Hôtel Belle Vue we should
probably find beds, and so on we drove, rather in despair at our
prospects, though, happily, unnecessarily so, for the Belle Vue gladly
opened its arms for our reception, and ere long we were, oblivious of
all our toil, comfortably ensconced in bed. From our windows we had a
pleasant view of our quarters; it was broad daylight, and the prospect
was--as from such a position, looking up a lake, it always is--very fair
and charming.

We were up again soon after eight, and were glad to find the morning
fine, though the clouds were low, and we saw, consequently, nothing of
the distant view of mountains which lends its greatest charms to Zurich.
The town is, however, pretty and striking. The picturesque houses, with
wooded hills on all sides beyond them, and very charming views of the
lake, if they do not make its attractions first-rate, at any rate make
them very considerable.

The main feature of interest for me was the cathedral, a fine Romanesque
church, very fairly perfect, but mutilated in its interior arrangements
by the Calvinists, in whose hands it now is. In plan, it has a nave with
aisles of six bays, a

[Illustration: 2.--CLOISTER, ZURICH CATHEDRAL. Page 19]

short choir, and east of this a square-ended sanctuary, the aisles
having apses, roofed with semi-domes. In the nave two of the
aisle-arches make one groining bay. The transverse groining-ribs are of
a simple square section, the diagonal ribs having in addition a large
round member. The triforium is very large and fine, and is made use of
for congregational purposes, being fitted up with seats, which,
curiously enough, are all made to turn up as misereres. There are no
transepts. The sanctuary arch is loftier than the choir arch, and seems
to have been intended to be very distinctly marked. In the clerestory
there are two simple round-headed lights in each bay; the choir is
arcaded all round internally, and for frigidity of effect cannot be
surpassed; the internal fittings comprise an immense pulpit, but, so far
as I could see, not even an apology for an altar.

The exterior has two western steeples,[1] and a north doorway, each jamb
of which has three detached shafts, standing considerably in advance of
the wall, which is entirely covered with diapers. The arch itself is
semi-circular, and very simple in its moulding; but this simplicity
rather adds to than detracts from its general grandeur of effect. The
whole is inserted in an additional thickness of wall, set on, as it
were, against the original wall, and the extreme width of the doorway
itself is no less than eighteen feet nine inches. The cloisters were
remarkable, and very good of their kind; the arches rested on detached
shafts, the capitals of which were elaborately carved in a very peculiar
manner, but very effectively. The whole design was unlike any Northern
Romanesque, and bore much more similarity to the best Lombard work.
Unfortunately, the whole of this cloister was rebuilt in 1851, the
carving having been re-worked, or renewed throughout in imitation of the
original. It will be seen, however, that, in spite of alterations, this
is a very fine church, of a very early type, and peculiarly valuable in
a country which, like Switzerland, has comparatively little left that is
really good in the way of architectural examples.

There are other churches in Zurich, but I believe not old, and at any
rate I had no time to examine them. One of them is appropriated to the
use of the Roman Catholics; and there is one desecrated, rising from the
edge of the lake, and forming a prominent object in the general view of
the town as you leave by the steamer; this is of good outline, but has
no details remaining of any value. The point chiefly to be noticed in
the churches of Zurich appears to be the way in which their spires are
all painted red, looking in the full sunshine very bright and
picturesque.

The Swiss have a great feeling for bright colour, and on our way from
Basel to Baden we noticed one of the many instances of this in several
turrets covered with brightly-coloured glazed tiles. A light green seems
to be the favourite colour, and is commonly used without mixture with
any other. They look best with their lower side rounded, and when of
small size; and are constantly used in turrets rising out of roofs which
are entirely covered with plain tiles. I remember, two years before,
noticing with extreme pleasure the beauty of some dark green tiles used
at Schaffhausen; and I have already had occasion to mention those on the
cathedral at Basel with equal commendation. Unhappily, we have to lament
that English people, in their insane hatred of bright colours, if they
saw such tiles used in England, would be horrified at such a violation
of the correct simplicity and uniformity of colour to which the
cheapness of slate has made them accustomed. Some modern attempts,
however, at introducing coloured tiles have not been so successful as
could be wished; and of all, perhaps the least so is the roof of the new
Maria Hilf church at Munich, on which tiles of light blue colour are
used in such large masses, that at first sight it seems that half the
roof is stripped, and that the pale blue sky is seen instead of roof.

At ten o’clock we left our hotel by the steamer for Schmerikon at the
head of the Lake of Zurich. The weather still looked doubtful, though
much better than on the previous day, and our host of the Belle Vue,
taking a good view of this, as is a landlord’s duty, conducted us to the
boat with smiling anticipations of fine days to come.

The shores of the lake are, for the greater part of its length,
literally fringed with houses all painted white, and contrasting
violently with the trees, vineyards, and green hills by which they are
backed. On the north the shore is low and gradually shelving down to the
water; on the south it is rather more precipitous, but after all not
very striking. At the head of the lake heavy dark round clouds hung upon
the hills, and left us in pleasant doubt as to whether or no we had fine
mountains to discover when they cleared away; a doubt, as it happened,
not settled, as far as we were concerned, save by certain lively and not
too trustworthy representations which we afterwards met with, in the
shape of advertisements of the Zurich hotels, and which showed a line of
snow mountains as the ordinary horizon of their visitors.

The churches on the lake are very numerous and very similar. The
steeples are almost always gabled, and from these gables rise spires
painted red, and very thin and taper in their form. The gabled sides of
the towers are generally made useful rather than ornamental by the
introduction of enormous clock-dials. The only decidedly mediæval church
which I saw between Zurich and Rapperswyl was at one of the villages on
the north shore of the lake, I think at Meilen, but I am rather
uncertain as to the name. Its design is both novel and very good; the
pinnacles on the gable being unusual in saddle-backed steeples, and
giving considerable picturesqueness of outline. The accompanying woodcut
will show the general character of the design, and it will be seen that
the tower is on the north side of the choir. The steeple roof is covered
with greyish-red tiles, with a pattern marked on them with yellow tiles.

[Illustration: CHURCH ON THE LAKE OF ZURICH.]

The steamers on this, as on most Swiss lakes, are somewhat tedious in
their journeys, as they take a most zigzag course, first calling on one
side of the lake and then on the other, until one doubts whether one
will ever reach the journey’s end. At Horgen of course we discharged a
large proportion of our English passengers, who were all bound for the
Rigi, but their places were soon occupied by the umbrella-loving
natives, who flocked in and out of the boat in great numbers at every
station, and by the time we reached Rapperswyl we had no more
fellow-countrymen in the boat, and perhaps, like many Englishmen, to say
the truth, we then first thoroughly realised that we were abroad. Much
as one loves one’s country, certainly one source of pleasure when abroad
is the not hearing too much English spoken or seeing too many English
faces.

At Rapperswyl, famous for having the longest bridge in the world, there
is a most conspicuous group of buildings on rising ground above the
lake, very picturesquely thrown together; it consists of a church and a
castle; the latter has several towers capped with pyramidal and
saddle-backed roofs, and the former has two towers in the position of
transepts, with saddle-back roofs gabled north and south, the southern
tower being considerably the larger of the two. Altogether, the group is
one of uncommon variety and picturesqueness of outline. Below, in the
town, is a small church, with a most happily-conceived though very
simple bell-turret rising out of the roof, square in its plan, but
capped with an octagonal spirelet. This is a not uncommon plan in this
part of Switzerland, and is always most agreeable in its effect. The
views from the terrace by the side of this castle are of singular
beauty. It is high enough above the lake to command a good view of its
whole expanse, and to secure a not too distant view of some of the
mountain peaks of Glarus. Rapperswyl is a good point to stop at, for the
sake of a visit to the famous pilgrimage church at Einsiedeln, certainly
one of the spots in Switzerland most curious and interesting, though its
buildings have no claims to our regard on the score of architectural
beauty.

Passing under, or rather through, the bridge, we found that it was very
narrow and had no side railing of any kind, so that it appears to be far
from a pleasant contrivance for crossing the mile or two of shallow
water which here scarce serves to keep up the appearance even of a lake;
and perhaps it is upon the score of the absence of real danger of
drowning if one fell over that they dispense with any protection. At
Schmerikon, which we reached in four hours from Zurich, we left our
steamer, and immediately embarked on a barge in order to go by the Linth
canal to Wesen; but we found that, however expeditious this might be in
descending, it was a kind of conveyance not to be recommended highly to
any one wishing to ascend the canal, inasmuch as--unlike ordinary
canals--this is neither more nor less than the glacier-torrent of the
Linth bringing down the melting snow from the Glärnitsch and Todi
glaciers, and rushing along at a really tremendous pace; to those,
however, who have time, it may be commended as affording magnificent
views of the mountains of Glarus and of those which rise so grandly
above the Lake of Wallenstadt.

As we entered the canal from the lake we were amused by the unsuccessful
attempts of our crew to secure some wild-fowl, two of which they
succeeded in shooting, and then, without any kind of regard for the
feelings of passengers panting to arrive at Wesen in the promised two
hours and a half, they deliberately proceeded--of course in vain--to
chase the unhappy birds, which, though wounded, were quite able to dive
much deeper than their enemies could reach, and so the only consequence
of the chase was a hearty laugh at the expense of the baffled sportsmen,
half an hour’s delay, and much lost ground to be made up.

The entrance to the canal was very striking; a low hill covered with
larch and birch rose from the water’s edge, and above this, the
mountains, gradually shelving upwards, were terminated in a line of
rocky ridges of very grand and rugged character. Whilst we were admiring
the view a slight shower passed over us, and the sun suddenly breaking
out, produced one of those lovely effects of colour so peculiar to
mountain scenery; a rainbow seemed exactly to fill up one of the great
basins formed by the undulations of the mountains, and, after bathing a
great sweep of mountain-side in the richest and most distinctly marked
colours, gradually died away.

The canal, which at first looks more like a river, soon takes a bend to
the S.W., and then, passing under a quaint wooden bridge, over which
passes the road to Uznach, we found ourselves in what certainly looked
sufficiently canal-like. The stream is so rapid that the walls built up
on either side are preserved from being washed away by stone groins
running out into the stream, and acting as so many breakwaters to keep
the water in the centre. Slowly and steadily our horses pulled us up,
whilst we, mounted on the top of the cabin, were able to see over the
walled sides of the canal, and to enjoy the glorious prospect before us.

Before long our captain blandly informed us that he was going to stop
for dinner at a wayside house, so we, anxious to make the same good use
of our time, attempted to follow his example. Unfortunately the
landlord, though very jolly-looking, had a very badly stocked larder,
and we had to satisfy ourselves with bread, honey, and wine. It is true,
indeed, that our host did produce some cold meat--portion, as I
imagined, of a goat dressed some ten days back--but this was not
eatable, and was valuable only as furnishing an opportunity to him of
showing his perfect power of making the best of a bad thing. To season
the goat he brought in vinegar and oil, and, putting them upon the
table, exclaimed with some _empressement_, “Voilà, monsieur; mais le
vinaigre n’est pas bon!” just as if this was the strongest
recommendation he could give us! We laughed heartily, avoided the
vinegar, and parted good friends with our host, thanking him from our
hearts for having saved us the painful operation of making the discovery
about its quality for ourselves!

Our not very satisfying repast finished, we embarked again upon our
barge, and in the occasional intervals, when sudden and heavy storms of
rain obliged us to seek shelter in the cabin, we were much amused in
watching the proceedings of some men belonging to the boat, who spent
the whole of the five hours consumed in the journey in an unceasing game
of cards; I must do them the justice to say that they played very
good-humouredly, and laughed without ceasing. Under no circumstances
could we have seen the scenery more gloriously; occasional bright gleams
of sunshine broke in upon and followed clouds of the most inky hue, and
then came pelting down heavy showers, accompanied by howling wind and
darkness; and as we reached the opening of the valley, looking up beyond
Glarus to the great mountains which close in its upper end, I think the
effect was really more grand and terrific than anything I have ever
seen. The mountains are of very fine outline, and of great height, as we
saw by the more than occasional glimpses which we had of snow about
their summits. By the time we reached Wesen the wind was so violent that
we found it difficult to keep our places upon the top of the cabin; and
we disembarked just before dark, in time to see the fine mountains on
each side of the Lake of Wallenstadt here and there through the
storm-clouds, and its waters beaten by the wind into not insignificant
waves. We had to walk through the entire length of the village--a
picturesque, quaint little place, sheltered under the almost overhanging
rocks at the side of the water--and arrived at last at the capital and
thoroughly Swiss inn, the Hôtel de l’Epée, where we were to sleep.

Travellers now speed very differently along this country, and, I fear,
see less than they ought of its beauties. Steamboats no longer attempt
to pass beyond Rapperswyl, and the railway hurries one along by the
beautiful Lake of Wallenstadt to the valley of the Rhine, only earning
one’s gratitude when one is in violent haste, and because by a branch
line it makes a détour to Glarus and Stachelberg much more possible than
it was when first I made the journey. On the whole I fear, where
railways pass through beautiful scenery, the tourist loses more than he
can possibly gain, not only in the views of the country, but equally in
the incidents of travel, which are becoming only too monotonous and
similar everywhere.




CHAPTER III.

                                  “Where the mountains
    Lift, through perpetual snows, their lofty and luminous summits.”
                      _Evangeline._

  Wallenstadt--Sargans--Gorge of the
  Tamina--Ragatz--Chur--Ems--Reichenau--Thusis--Zillis--Andeer--Splügen--The
  Splügen Pass--The Custom-house--Cascade of the Medessimo--Campo
  Dolcino.


The storm of the evening gave no kind augury of sunshine on the morrow,
and with rather anxious thoughts we listened as it roared among the
mountains which overhung our hostelry. But it seemed that we had
suffered enough, and when we woke we found that, though the clouds had
not yet cleared off from the sides of the mountains, there was
nevertheless every prospect of a fine day.

We were obliged to leave by an inexorably early steamer at half-past
five for Wallenstadt, and so lost all but the suggestion only of the
magnificence of the mountains which tower up so grandly over the north
shore of the lake. Like Goethe on his way into Italy, we might exclaim,
“What do we not pass over, both on the right hand and on the left, in
order to carry out the one thought which has become almost too old for
the soul!” But our time was limited, and our chief anxiety to spend as
much of our short holiday as we could in Italy; and so, sad though we
were to miss what was doubtless so well worthy of being seen, on we were
bound to go without delay.

Before we started I had secured a voiturier whose carriage was at
Wallenstadt to take us on to Chur, so that on this score I had no
trouble before me. Our voyage was only too soon made. Unlike the Lake of
Zurich, where the traveller rather hopes that each place at which he
stops may be the last, on this lake, as the tiny steamer ploughs its way
rapidly over its surface, with its goal always in view, and with not a
place to stop at on its road, he ceases not to long that his pleasure
may be prolonged!

By seven o’clock we were in our carriage, and _en route_. The sun began
to shine, and every minute the clouds rose higher and higher; so that,
before we finally lost--by turning into the valley of the Rhine--the
last view of the valley of the lake, we could see the peaks of the
mountains which we so wished to have seen before, the Sieben-Churfürsten,
which tower so grandly over the lake.

Wallenstadt is but a poor place, its situation being unwholesome, and
its inns not much to be commended. It has a church of modern character,
with an old-looking tower in the position of a transept, with a
saddle-back roof, gabled north and south. On the lower part of the south
side of this tower are paintings of the Crucifixion and some other
subjects, apparently of some antiquity. Just above the town, on the
right, we passed the ruins of an old castle; and at a slight rise in the
road had a beautiful view of the calm waters of the lake, looking blue,
but very much smaller than it really is. This, no doubt, is owing to the
great height of the precipitous rocks on its north side, which we now
saw for the first time, the clouds having at last risen and disclosed
some of the beauties which they had been concealing from us.

The valley from Wallenstadt to Sargans, just beyond which our route,
after crossing the very low watershed, joined the valley of the Rhine,
was strikingly beautiful. Its ecclesiological features were not,
however, remarkable, if I except the constant repetition of what I have
often noticed in the Catholic cantons of Switzerland and in Tyrol--the
occurrence, namely, of grated openings on either side of the western
door-way, commanding the interior and protected by an open porch,
through which passers-by, though not able to enter, might still see the
altar. On our journey from Basel to Zurich we passed a church the altar
of which was lighted up, and the doors behind these gratings left open
very late at night. It was in a lonely place, and when I passed there
was no one in or near the church. I never see this arrangement without
wishing to introduce it in England. There are so many of our Churches
which cannot conveniently be left always open, and where such a
provision might suggest to passers-by, as it does here, the propriety of
using a church at other times than those of public service.

The cultivation of this valley is not so uninteresting as its
ecclesiology. Here we first found the vines trained about in the
horizontal Italian fashion, whilst under them great gourds and pumpkins
developed themselves to a prodigious size.

Sargans is a very picturesque old town, and has some capital examples of
good Swiss carpentry in its houses; in addition to which there is a
picturesque and antique-looking castle, rising high above the houses on
a rock, guarding the eastern entrance to the town, and commanding the
junction of our road with that of the valley of the Rhine leading to the
Lake of Constance.

Our coachman was under a bond to travel as fast as a diligence which
lumbered on slowly in advance of us, and as far as Ragatz was quite true
to his word; there, however, we determined to pause for a few hours, not
willing to pass anything so famous as the baths of Pfeffers without a
visit. Leaving our carriage, we mounted a light car, and were soon
ascending the beautiful gorge of the Tamina to the baths. The road is
capitally made, and follows the windings of the mountain torrent so
closely as to require some nerve in those who drive rapidly along on
their road to or from the baths. The ascent is steep, but in rather less
than an hour we found ourselves at the baths. The rocks rise nearly
perpendicularly behind the ledge of rock on which they stand, and the
only mode of access to the upper and more wonderful part of the chasm
was by passing through the long corridors, which betokened the once
religious object of the building. These passed, and in charge of a
guide, we crossed the torrent by a rude bridge, and then by a rather
precarious path made our way, as it seemed, almost into the bowels of
the mountain. The gorge is so very narrow that in many parts the light
of the sky is no longer visible, the rocks overhanging each other above
the head. All the while the torrent is roaring by our sides, and we feel
that we are indeed enjoying an excursion into the very heart of the
rocky earth. At last we reach the end of the path, are compelled by our
guide to ensconce ourselves, one by one, in a small kind of box formed
round the source of the spring--to pronounce it very hot and very nasty
(its two most eminent qualities)--and then, still admiring the matchless
grandeur of the rocky way, we regain our car, and are soon again whirled
down the hill to Ragatz.

Our driver is a cheerful, pleasant fellow, talks German much better than
the man we brought from Wesen, is communicative, moreover, and seems to
enjoy a laugh and a joke uncommonly. Of course we become friends, and
with no trouble on our parts, though with some little on his, it is
arranged that our old driver shall remain where he is, and that our new
friend, proud in the possession of the then very necessary Austrian
passport, shall take us on as far at any rate as Chiavenna. A hurried
Swiss luncheon--wine, honey, bread and butter--is soon despatched, and
again we are on our way under the auspices of our new voiturier.

But we must not leave Ragatz without noticing its church, remarkable for
its exceedingly good octagonal wooden spire springing in an unusual
manner out of a square wooden belfry stage, and another church at (I
think) Vilters, close to Ragatz, which has a lofty tower finished on
each side with a sharp gable, and a thin octagonal spire rising from the
intersection of the cross-gabled roof; both these steeples are in a
position which for some reason is very popular in this district--the
south side of the chancel.

[Illustration: WOODEN SPIRE--RAGATZ.]

From Ragatz to Chur the churches are all very similar; they have tall
towers generally in the same position as those near Ragatz, and capped
with bulbous roofs, or sharp spires covered with metal. The road is not
quite the most agreeable we have travelled; some of the views, it is
true, are most lovely, and the mountains--among which towers pre-eminent
the grand outline of the Falkniss--are very noble; but, despite all
this, the valley is too wide, and the Rhine, by periodical inundations,
manages to secure so nearly its whole extent to itself, that there is a
waste, desolate, and pestilential look in the foreground which is not
prepossessing. We arrived at Chur at about half-past

[Illustration: 3.--CATHEDRAL. CHUR. Page 33.]

one, and, not sorry that our horses required rest, betook ourselves to
the inspection of this very curious town.

It is entered by old gateways, and many of the streets are still full of
ancient houses. The curious feature of the place is however its complete
division into two quarters--the Protestant and the Catholic--the latter
walled off, and entered by its own gates.[2] It occupies the upper part
of the town, and contains in the cathedral church of S. Lucius an
attraction for architects which has unusual merit and interest. Its plan
consists of a nave of three bays, a choir of one bay raised by twelve
steps above the nave, and a sanctuary much narrower than the nave and
choir, and also of one bay. The steps from the nave to the choir are
narrow and on each side, and between them is a very flat wide arch,
under which access is obtained to the crypt, the floor of which is a few
steps below the nave, and extends under the choir and sanctuary. The
plan is, it will be seen, not unlike that of the Cathedral at Zurich,
save that here there are no apsidal terminations at all.

A sketch of the interior of so singular a church cannot be
uninteresting, and it will be seen from this that the whole is of the
very earliest pointed work, and good of its kind; the crypt is supported
in the centre by a column resting upon a grotesque animal. Two of the
altars have fine shrines of metal of the thirteenth century, and two
other altars have ancient pricket candlesticks, and there are some fine
brass standard candlesticks also; the choir stalls are old, and there is
a late triptych behind the high altar, and a very fine Sakramentshaus
with metal doors just below the northern flight of steps to the choir,
which reminded me of the very fine example in a similar position in the
cathedral at Ulm. The altar is of stone of the thirteenth century, with
five detached shafts in front, supporting the slab or mensa. The whole
church is groined. It is worthy of notice that the choir makes a great
bend out of the straight line towards the north--so much, indeed, that
it is impossible to avoid noticing it as one enters the church. The
steps from the nave to the choir lack dignity. But it is true that if
they had been in the centre, and the entrances to the crypt on each
side, the crypt would not have been seen, as it now is, from the nave,
and a striking effect would have been lost. The west end has a fine
round-arched doorway with several shafts in each jamb, above this a
large window of the same character, and in the gable a small
middle-pointed window. About ten feet in advance of the west doorway is
a curious remnant of a gateway with piers and shafts resting upon
monsters, looking, however, very much as though it had been removed from
elsewhere.

Service commenced just as I was obliged to think of leaving the church;
the priests wore red cassocks and tippets, and very short surplices
edged with lace, and looked unclean and untidy; there was no one in the
body of the church, and the sacristan, after the service had commenced,
walked backwards and forwards about the choir, down the steps into the
nave, and then--after a little attention bestowed on some matter
there--out of the church. On a subsequent visit to this church (in 1872)
I found repairs in progress, which bade fair to destroy some of its
great archæological interest.

Descending from the melancholy and squalid-looking Catholic quarter, we
soon came upon the Protestant church, dedicated in honour of S. Martin,
which is now somewhat remarkable. It is old, but it has been plastered,
whitewashed, and then painted by some original artist over its whole
exterior, in an extraordinary imitation of all kinds of inconsistent
architectural devices; pilasters, cornices, mouldings, tracery, and the
like, are all boldly represented with black paint, and in such style
that we all stopped the moment we saw it, struck by the conviction that
it must be a scene from some play, so utterly absurd, flat, and out of
all perspective did the whole look.

The situation of Chur is very lovely, placed as it is just at the point
where the Schalfiker Thal joins the valley of the Rhine, and upon the
steep and rugged bottom slope of the mountains.

The weather was every moment becoming more glorious, and just as we left
Chur, along the road which leads to Reichenau, we had one of the most
lovely views we had enjoyed. It is not always the case, however
beautiful may be the scenery, or however lovely the weather, that one
finds everything group together perfectly; here, however, it did, and I
commend the subject to the pencils of those who follow me on this route.

We soon reached Ems, whose church, situated upon a green knoll above the
village, has the peculiarity of a small apsidal building east of the
chancel apse. The key was not to be found, so that I could not go in to
examine what this building was. This church had an octangular steeple,
whilst another church in the same village had one of the bulbous
coverings of which I have before complained. At Reichenau it is proper
to go to see the house in which Louis Philippe acted in 1793 as
schoolmaster under a Monsieur Jost, and I fear we fell rather in the
good opinion of our driver when we neglected so proper and regular a
custom; but so it was. The garden of the inn is charming, and from its
edge you obtain the best view of the junction of the Vorder and Hinter
Rhine, and having enjoyed this thoroughly, we passed rapidly through
Reichenau, across its two quaint covered wooden bridges, and by the
beautiful meeting of the waters, until we found ourselves following the
course of the Hinter Rhine and fairly on the Splügen road.

We only wished to reach Thusis by sunset, and so our time was ample for
enjoyment; we walked much of the way, detecting eagerly every here and
there patches of snow on the mountains in the distance, each of which is
hailed as a discovery by every fresh traveller, who feels himself
transported with delight by the distant view of the pure white against
the sky.

Castles are here as numerous as ever upon the Rhine, and at least a
dozen, I should think, might be reckoned perched on every favourable
spot between Reichenau and Thusis. As the road advances the valley
widens out into a kind of basin, into which flow two streams, the one
through the as yet unperceived gorge of the Via Mala rather to the
right, the other through an opening in the mountains directly in front
of us, which allows us a charming view of the snowy heights above the
Julier Pass, drinking in the last red rays of the setting sun, long
since passed away from the ground on which we stand; then there is a
long ascent, and, passing peasants coming in from hay-making, merrily
laughing and singing, we drive up the straight ugly street of Thusis to
the Via Mala Hotel. But the evening is too glorious to lose, and in five
minutes we are out again on foot to explore the commencement of the
black defile; and until we are absolutely turning into it, so narrow is
the gorge that it is not seen, but when seen, and by such a light, how
grand and beautiful it is! We ascended some distance and then stood and
admired. Above us tremendous rocks towered high into the air, riven in
two for the narrow chasm in which we stood, at whose bottom we heard the
distant roar of the Rhine, and down below and beyond, framed as it were
between the grand outline of rocky crag and pine-covered mountain, lay
the valley of Domleschg, still retaining, by contrast with the gloom
around us, some light upon its fields, and castles, and villages. Rest
was well earned after such a pleasant and actively spent day, and, if we
were late in starting in the morning, it was as much the fault of our
coachman as of ourselves. However, though not so early as we intended,
we left soon after six, and in a few minutes were again in the Via Mala.
And now by daylight I doubt whether we were not all disappointed; there
is so much in a name that one expects something _very_ terrific from
such a name, and this it scarcely is. It is seldom fair to compare one
piece of scenery with another, but still I feel that this was certainly
not the most savage I had ever seen, and therefore not justly _my_ Via
Mala. But beautiful in the extreme it was, and I believe we all
regretted that we so soon found ourselves again in the more open valley
on the road to Zillis. Here we found a church with a lofty tower, in the
same position, and with a spire of the same design, as that at Ragatz;
the nave low and ugly, the chancel lofty, with a steep pitched roof and
apse; the windows pointed but modernized; the belfry windows of the
steeple of three lights, with circular arches, and divided by shafts,
which were continued on in blank panels on each side of the windows, so
as to form an arcade of five arches on each side. And this I believe was
the last noticeable church we saw before we reached Chiavenna, and in
its arcaded belfry I fancied that I saw something of an Italian
influence at work, which might well have been the fact.

We soon reached Andeer, where we waited but a short time, and then
commenced a steep ascent. The lovely scenery, the mountains closing in
round us, and the roar of the falls of the Rofla making music in our
ears, made our way very enjoyable. There was but little chance, however,
of rapid progress, as from Andeer to Splügen the road is almost always
on the ascent, sometimes gradually, at others in steep zigzags up the
shoulder of some obstructive hill, and constantly overhanging or
crossing the rapid, white, foaming mountain-stream, sole representative
here of the noble river whose broad waters have been admired at Basel.
The air of desolation becomes more decided as one reaches Splügen. Trees
and shrubs more scarce, and often blasted by the fierce rush of the
wintry wind, or the keen sharp blow of the fallen rock, or the swift
sweep of the avalanche, aid in making up the desolate picture.
Vegetation has well-nigh ceased, and the eye, though deceived at first
by the intensely red colour perceived every here and there on the
hill-sides and on the rocks, discovers presently that not to flowers or
plants, but to lichen or other such desolate vegetation, is it owing.

By the time we caught the first sight of Splügen the sky was
overclouded, the wind rose, and a sudden heavy storm of rain gave us a
lesson in the customs of the weather in these regions, to which our
driver’s quiet assurance that we should probably have a snowstorm on the
pass added the few remaining drops required to make up the draught which
we saw ourselves doomed to swallow.

Splügen, however, was reputed to have an inn which would give us
enviable shelter for a couple of hours, and we entered at once, hoping,
if we waited, again to see the blue sky before we crossed the boundary
between the North and the South--between Switzerland and Italy.

The table-d’hôte was just about to commence, and in came a diligence
from Milan, and out came the passengers: another carriage, which had
pursued us relentlessly all the way from Andeer, came in at the same
moment, and down we sat, about fifteen English people, not one of whom
had been in the house ten minutes before, not one of them stopping for
more than their own and their horses’ dinners, and all proceeding in
different directions, either on their way home satiated with travel, or
just about to dive like ourselves in full quest of pleasure and
excitement, into a new country. These meetings are always curious,
generally amusing, and to the quiet and attentive observer of character
not a little edifying. On this occasion there was subject-matter enough,
and we found an old gentleman, travelling sorely against his will, under
the care of an active and thoroughly vulgar wife, some literary old
maids of another party, and the enthusiastic damsel of a third, each in
their way amusing, and not the less so in that it was necessary to
inspect them and part with them so rapidly.

Splügen, in a soaking rain, is not a pleasant place; and as I employed
myself in sketching from the inn window the very picturesque old bridge,
which gives[3] all its architectural character to the village, I
conceive that I accomplished all that was necessary; and when we got
into our carriage again, and, crossing by the bridge, left the Bernardin
road to the right, and finally plunged really into the Splügen route, it
seemed like a reward for my industry to find the rain cease and the sun
again occasionally shine out.

The ascent begins with a series of zigzags, which rapidly carry the road
high above the valley of the Rhine, and then, passing through one of the
long covered galleries for which this route is famous, it emerges in an
upland valley or dip between two mountains, up which it takes a steady
course along a road macadamized, by-the-by, mainly with the white marble
which abounds here, until, just below the summit, it comes again upon a
steep mountain-side, to be surmounted only by a patient unravelling as
it were of the intricacies of an endless zigzaging, which at last brings
us to the Swiss guard-house and the entrance to the great gallery. The
clouds are low and gathering; but still as we see below us white patches
of snow every here and there, and above us the blue edge of a great
glacier marked with lines of crevasses and fringed with a white edge of
snow, we feel that we have really at last achieved the summit. Noisily
we trot through the arched gallery, and then, after another slight
ascent for a few minutes, we stop and put on the drag, and then down we
go rapidly and cheerily, backwards and forwards, occasionally giving a
merry tap to some corner post at the turns of the road, in order to let
it be known that we, our driver, and our horses, are all of us heartily
glad that we are at last on the south side of the pass--no longer the
German Splüg_en_, but, as we learn from divers notices along the road,
the Italian Splug_a_. A short drive takes us to the custom-house--not
looked forward to cheerfully by those who have met, as we had at
Splügen, a man turned back by mistake, and after two days’ delay again
retracing his steps--but happily, in our case, passed easily enough, and
with an exhibition of the greatest courtesy and civility from the
Austrian officer, the mention of whom reminds me of the great change
which has taken place in the political status of this country since
first I made acquaintance with it. It is a change of no little
importance to the traveller, who now goes without let or hindrance
almost everywhere, instead of being worried out of his life by troubles
about passports which even Austrian courtesy could not make tolerable.

We are soon off again across a drear and peaty-looking plain, with no
view of the neighbouring mountains, and accompanied along the road by a
troop of wild smuggler-like fellows, in broad-brimmed steeple-crowned
hats, loose jackets, knee breeches, and coarse stockings, riding wildly
along on rough horses, without saddles or bridles, but every one of them
handsome grand-looking fellows, showing, as they smiled, teeth of the
purest white, and more nearly coming up to one’s idea of real Italians
than any with whom, later in our journey and more in Italy, we happened
to meet. Before long, however, we again commenced the descent, and then,
after passing through two or three galleries of prodigious length, at
last came out upon one of those spots, the view from which, as much
perhaps by reason of its associations as for its intrinsic beauty, rests
on the mind for ever after, as one of the most lovely ever seen. On our
right a steep mountain track slopes rapidly and almost perpendicularly
down to a narrow valley, whose opposite and no less precipitous side we
are about to descend; below us, far down, we see the village roofs of
Isola, with its church and Italian campanile; beyond--and this is indeed
the great charm of the prospect--down the valley, where the atmosphere
seems redolent of the South, we see a grandly formed mountain, and again
to its right another but more distant; between these two dim and distant
shades lies the lake of Como--beyond them the broad rich plain of
Lombardy; the sun shines forth, and we dream henceforward of that
valley, looked down upon from the gallery on the Splügen, as one of the
brightest prospects of our lives!

We had not gone far beyond the last gallery before our voiturier made
good a boast which he had often repeated, of showing us a real waterfall
on a grand scale before we parted company, and, pulling up his horses,
made us--not unwilling--dismount to look _down_ the cascade of the
Medessimo. A passage has been formed from the road to a point which just
overhangs the fall, and here, securely parapeted round, you look down
over a grand sheer fall of some eight hundred feet, in the course of
which the torrent which goes to feed the threadlike Lira down below us
in the valley, and just now roaring in bold volume underneath our road,
loses itself in soft, delicate, and fairy-like spray, and ere it reaches
the rock below, seems like some delicate mist falling from the sky for
ever in endless and exquisite change of form. Just beyond the cascade
the most wonderful part of the descent in an engineering point of view
commences, and the road seems really to descend the perpendicular face
of the rock, surpassing in boldness most other roads that I know, and
affording very fine and varied views of the cascade on the descent. We
soon reached Campo Dolcino, a miserable and most dirty-looking village;
and were, sorely against our will, obliged to wait for our horses to
bait; and then on we went, the sun some time set, and the night dark and
cloudy. Presently a storm arose; and without lights, and travelling
along a road turning sharp angles every minute, and never losing the
music in our ears of the roaring Lira, our lot seemed more wild than
enviable; at last we came to a house and tried unsuccessfully to borrow
a light, but presently at another house we succeeded, and then guided by
a lantern we pursued our way safely enough. I have seldom been out in so
grand a storm; the lightning was vivid beyond all that I could conceive;
and as at one minute it played about on the foaming water beneath us,
and at another lighted up the whole mountain-side beyond with pale and
intensely lovely light, flickering, playing, and dancing about in the
wildest fashion, I believe we felt half sad when house after house
appeared, and at last we entered the long, narrow, and thoroughly
Italian streets of Chiavenna.

Another journey took me to Chiavenna at the same time in the evening, on
my way north from Como. It was the night of the 8th September, the
Nativity of the Blessed Virgin, and every peasant in his solitary châlet
on the mountain-side was burning a bonfire in her honour. There seemed
to me to be something very touching in this flaming burst of distant
greeting from mountain to mountain, and few circumstances have ever
brought home more vividly to me the isolation of these mountaineers,
than the compensating power of a sympathetic faith which made them thus
bid each other welcome by their flaring fires.




CHAPTER IV.

                        “But now ‘tis pass’d,
    That turbulent chaos; and the promised land
    Lies at my feet in all its loveliness!
    To him who starts up from a terrible dream,
    And lo, the sun is shining and the lark
    Singing aloud for joy, to him is not
    Such sudden ravishment as now I feel,
    At the first glimpses of fair Italy.”
                       _Rogers._

     Chiavenna--Lake of Riva--Colico--Gravidona--Lake of
     Como--Varenna--Stelvio Pass--Lecco--Bergamo:
     Broletto--Churches--Castle of Malpaga.


The situation of Chiavenna is eminently beautiful: in a deep valley
surrounded on all sides by mountains whose slopes are covered with soft
and luxuriant foliage of oak and chestnut, and where every available
open space is devoted to trellised vineyards, it contrasts strongly with
the pine-covered hills so lately passed on the northern slopes of the
Alps; placed, too, at the confluence of two streams--the Meira and the
Lira--it rejoices in the constant rushing sound of many waters.

It was only necessary to move out of the shade of our hotel into the
melancholy piazza in which it stands, to discover that an Italian sun
lighted up the deep blue sky; and a walk to the principal church,
dedicated in honour of S. Lawrence, a stroll through the narrow streets,
and a rather toilsome ascent through a vineyard formed upon a rock
which towers up behind a kind of ruined castle, and from which a capital
view is obtained of the singular and beautiful cul-de-sac in which the
town is planted, sufficiently convinced us of its power.

The church of S. Lawrence is entered from a large oblong cloister, in
one angle of the space enclosed by which rises a tall campanile, its
simple form, and its arcaded belfry full of musical bells, contrasting
well with the outline of the hills which overhang and hem it in. On the
east side of the cloister are the church, an octagonal baptistery, and a
bone-house, all ranged side by side and opening into it, and the latter
curious as an example of the extent to which the people of Chiavenna
amuse themselves by arranging sculls and arm-bones into all kinds of
religious and heraldic devices, and with labels to mark the names of
their former owners. The _tout ensemble_ is picturesque in its effect,
and the cool pleasant shade of the cloister, with the view of the church
and its tall campanile, and irregularly grouped buildings looking
brilliantly white in the clear sunshine, was very pleasing.[4]

Italian beggars, persevering, and, at any rate in appearance, very
devout, did their best to annoy us here and everywhere when we ventured
to stop to examine or admire anything; and Italian beggars are certainly
both in pertinacity and in filth about the most unpleasant of their
class.

My voiturier gave me a lesson worth learning, and not perhaps unworthy
of note for other unsuspicious travellers. We had a written contract to
Chiavenna, and thence to Colico he had agreed verbally to take us for a
certain sum; before we started I found, however, that he intended to
charge us three times as much as we had agreed upon, and, as very
luckily we found a diligence on the point of starting, we secured places
in the cabriolet at its back, from which we had the best possible
position for seeing the views, and so left him in the lurch, with divers
admonitions to behave himself more honestly for the future.

At ten we left, and had a very enjoyable ride to Colico. The valley,
however, bore sad traces of the havoc made by the inundations of the
Meira, and of the storm of the previous night. We soon reached the
shores of the little Lake of Riva, along whose banks our road took us
sometimes in tunnels, sometimes on causeways built out into the water,
until at last we reached the valley up which runs the Stelvio road, and
then, after passing along the whole length of a straight road lined on
each side with a wearisome and endless row of poplars, we were at
Colico. Here we prudently availed ourselves of the opportunity of an
hour’s delay in the departure of the boat for an early dinner, and, then
embarking, waited patiently the pleasure of our captain.

The scenery of Lake Como has been so often extravagantly praised that I
was quite prepared to be disappointed; but for the whole distance from
Colico to Lecco it is certainly on the whole more striking than any lake
scenery I have seen. The mountains at its head are extremely irregular
and picturesque, and throughout its whole length there is great change
and variety. In this respect it contrasts favourably with most other
lakes, and I certainly think that not even in the Lake of Lucerne is
there any one view so grand as that which one has looking up from within
a short distance of the head of Como over the Lake of Riva to the
mountains closing in the Stelvio, and rising nobly above the sources of
the Meira and the Lira.

Somewhat, too, may be said of the innumerable villages and white villas
with which the banks of the lake are studded; they give a sunny,
inhabited, and cheerful feeling to the whole scene; and, reflected in
the deep blue lake in those long-drawn lines of flakey white which are
seen in no other water to such perfection, add certainly some beauty to
the general view.

One of these villages--Gravidona--within half an hour’s sail of Colico,
ought not to be left unvisited by any one who cares about architecture.

Close to its little harbour stand two churches side by side, one an
oblong basilica, the other a baptistery of, as it seems to me, such
great interest that I give illustrations both of its plan and of its
exterior. It will be seen that the dimensions are small, the total
internal width being less than forty feet, whilst the design of the east
end is most ingeniously contrived so as to give no less than five
eastern apsidal recesses. There are two stair-turrets in the wall on
each side of the western tower which lead up to a sort of
triforium-passage which is formed behind an arcade in the side wall of
the church, and one of them leads also to the first floor of the tower.
The triforium consists of an arcade of seven arches in each side wall.
The three small apses at the east have each their own semi-dome, and the
chancel as well as all the other apsidal recesses are similarly roofed.
All the walls retain more or less traces of old paintings, the
Coronation of the Blessed Virgin occupying the principal apse, and the
Last Judgment the west wall. The whole church is built in white marble
and black lime-stone used in courses, or stripes, with extremely good
effect.

The roof of this Baptistery is of wood. The exterior is best explained
by reference to my drawing of the west front. It stands on a charming
site with a background of lake and mountain, such as one seldom enjoys.
There is a contrast here, which strikes one very much, between the
ingenious skill of the planning of such a building as this and the
rudeness of the execution of the details. I know nothing as to the
history of Gravidona; but it looks as though the plan came from the
hands of men who knew something of the Church of San Vitale at Ravenna,
whilst its execution was left to the rustic skill of the masons of the
country.

[Illustration: PLAN--BAPTISTERY, GRAVIDONA.]

[Illustration: 4. BAPTISTERY. GRAVIDONA. p. 48.]

The Baptistery is dedicated to S. John the Baptist. Close to it, as I
have said, stands the Church of San Vincenzo, which though Romanesque in
its foundation has been much modernized, and is now mainly interesting
on account of the exquisite examples of late fifteenth century
silversmiths’ work which still enrich its sacristy. Conspicuous among
these is a silver processional cross. This cross is nearly two feet
across the arms by three feet in height from the top of the staff. There
is a crucifix on one side and a sitting figure of Our Lord on the other;
figures of SS. George, Vincent, Sebastian, Christopher, and Victor, and
Our Lord on the base or knop; and half-figures of the Evangelists on the
arms of the cross. The ornaments consist of crockets bent and twisted,
of blue enamels, filigree-work, nielli, and turquoises set in the centre
of dark-blue enamels. It is, in short, a piece of metal work which might
well make a modern silversmith run down swiftly into the lake and drown
himself in despair at the apparent impossibility in these days of
rivalling such a piece of artistic and cunning workmanship, in spite of
all our boasted progress!

Not much less splendid is a chalice of about the same age. It is ten and
three-quarter inches high, has a plain bowl, but knop, stem, and foot
all most richly wrought with figures, niches and canopies, and the flat
surfaces filled in with fine blue and white Limoges enamels. The paten
belonging to this chalice is very large--nearly ten inches across, and
quite plain.

Half the passengers on the steamboat were, of course, Austrian soldiers
and officers, the other half English or Americans, either resident at or
going to Como. We, however, stopped on the way, and, leaving the
steamboat in the middle of the lake, after a row of about twenty minutes
found ourselves at Varenna, a village exquisitely placed just where the
three arms of the lake--the Como, the Lecco, and the Colico
branches--separate, affording, whether seen from here, from Bellagio, or
from Cadenabbia, the most lovely lake views it has ever been my good
fortune to see.

Here we had what seemed likely to be an endless discussion upon the
relative merits of a four-oared boat and a carriage as a means of
conveyance to Lecco. We inclined to the latter; but, leaving the matter
in the hands of an active waiter, we busied ourselves with eating
delicious fruit, admiring the tall cypresses growing everywhere about
the shores of the lake, and watching the exquisite beauty of the
reflections of Bellagio and the opposite mountains on the smooth bosom
of the water.

We were soon off again, and well satisfied to find ourselves trotting
rapidly along the well kept Stelvio Road, instead of dragging heavily
and slowly along as one always does with a Swiss voiturier; soon,
however, we were to find that our driver was an exception to the Italian
rule, and that he who wishes to travel fast must not expect to do so
with vetturini.

The churches which we passed were in no way remarkable; they all had
campanili, with the bells hung in the Italian fashion in the belfry
windows, with their wheels projecting far beyond the line of the wall;
but they all seemed alike uninteresting in their architecture, so that
we were in no way sorry to pass them rapidly on our way to Lecco. This
eastern arm of the lake, though of course much less travelled than the
rest of its course, is very beautiful, and its uninhabited and less
cultivated looking shores, with bold cliffs here and there rising
precipitously from the water, were seen to great advantage, with the
calm unrippled surface of the lake below, and the sky just tinged with
the bright light of the sun before it set above.

Lecco contains nothing to interest a traveller; we had an hour to spend
there before we could get fresh horses to take us on to Bergamo, and
wandered about the quaint-looking streets, which were full of
people--some idly enjoying themselves, others selling luscious-looking
fruit. We went into a large church not yet quite completed; it was
Renaissance in style, almost of course, and on the old plan, with
aisles, but very ugly notwithstanding. In the nave was a coffin covered
with a pall of black and gold; six large candles stood by it, three on
either side, and two larger than the others on each side of a crucifix
at the west end. The whole church revelled in compo inside and out and
there was external access to a wretched bone-house in a crypt.

Leaving Lecco, we had a long drive in the dark to Bergamo; the night was
very dark, but the air was absolutely teeming with life and sounds of
life; myriads of _cicale_ seemed to surround us, each giving vent to its
pleasure in its own particular note and voice with the greatest possible
determination; and had I not heard them, I could scarcely have believed
it possible that such sounds could be made by insects, however numerous
they might be. We changed horses at a village on the road, and went on
rapidly. The old town of Ponte San Pietro was passed, having been taken
at first to be Bergamo, and remembered by the sound of a troop of men
singing well together as they passed us in the dark in one of its narrow
streets, awakening with their voices all the echoes of the place, which
till then had seemed to us to be supernaturally silent. It was eleven
o’clock before we reached Bergamo, and tired with our long day’s work,
we were soon in bed.

A prodigious noise in the streets before five o’clock the next morning
gave us the first warning that the great fair of Bergamo was in full
swing; sleep was impossible, and so we were soon out, enjoying the busy
throng which crowded the streets of the Borgo, in a before-breakfast
walk; the crowd of women selling fruit, the bright colours of their
dresses, the rich tints of stuff hung out for sale, the display of
hair-pins and other ornaments in the innumerable silversmiths’ shops,
and the noisy, laughing, talking people who animated the whole scene,
made the narrow arcaded streets of the busy place most amusing.

After breakfast we started at once for the Città, as the old city of
Bergamo is called. It stands on a lofty hill overlooking the Borgo San
Leonardo, within whose precincts we had slept, quite distinct from it
and enclosed within its own walls. The ascent was both steep and hot,
but the view at the entrance gateway of the Città over the flat Lombard
country was very striking, and well repaid the labour of the ascent.
This vast plain of bluish-green colour, intersected in all directions by
rows of mulberry-trees and poplars, diversified only by the tall white
lines of the campanili which mark every village in this part of
Lombardy, and stretching away in the same endless level as far as the
eye could reach, was grand if only on account of its simplicity, and had
for us all the charm of novelty.

Through narrow and rather dirty streets, which do little credit to the
cleanly habits of the Bergamask nobility, to whom it seems that the
Città is sacred, and whose palaces are, many of them, large and
important buildings, we reached at last the Piazza Vecchia, around which
is gathered almost all that in my eyes gives interest to Bergamo.

Across the upper end of the Piazza stretches the Broletto, or town-hall,
supported on open arches, through which pleasant glimpses are obtained
of the cathedral and church of Sta. Maria Maggiore, which last is the
great architectural feature of the city.

But we must examine the Broletto before we go farther. And first of all,
its very position teaches a lesson. Forming on one side the boundary of
a spacious Piazza, on the other it faces, within a few feet only, the
church of Sta. Maria Maggiore, and abuts at one end upon the west front
of the Duomo. It is to this singularly close--even huddled--grouping
that much of the exquisite beauty of the whole is owing. No doubt Sta.
Maria and the original cathedral were built first, and then the
architect of the Broletto, not fearing--as one would fear now--to damage
what has been done before, boldly throws his work across in front of
them, but upon lofty open arches, through which glimpses just obtained
of the beauties in store beyond make the gazer even more delighted with
the churches when he reaches them, than he would have been had they been
all seen from the first. It is, in fact, a notable example of the
difference between ancient grouping and modern, and one instance only
out of hundreds that might be adduced from our own country and from the
Continent of the principle upon which old architects worked; and yet
people, ignorant of real principles in art, talk as though somewhat
would be gained if we could pull down S. Margaret’s in order to let
Westminster Abbey be seen; whereas, in truth, the certain result would
be, in the first place, a great loss of scale in the Abbey seen without
another building to compare it with and measure it by; and in the next,
the loss of that kind of intricacy and mystery which is one of the chief
evidences of the Gothic spirit. Let us learn from such examples as this
at Bergamo that buildings do not always require a large open space in
front of them, so that they may be all seen and taken in at one view, in
order to give them real dignity.

The whole design of the Broletto is so very simple as to be almost
chargeable with rudeness of character. The ground on which it stands is
divided by columns and piers, the spaces between them being all arched
and groined. Towards the Piazza three of these arches, springing from
rather wide piers, support the main building, and another supports an
additional building to the west of it. Above the three main arches are
three windows, of which that in the centre, though very much altered,
still retains a partially old balcony in front, and was evidently the
Ringhiera, from which the people standing in the Piazza were wont to be
addressed by their magistrates. The windows on either side are very
similar in their design and detail; their tracery is of fair
middle-pointed character; and the main points in which they strike one
as being different from English work are the marble shafts with square
capitals in place of monials, a certain degree of squareness and
flatness in the mouldings, and the very pronounced effect of the sills,
which have a course of foliage and moulding, and below this of trefoiled
arcaded ornament, which in one shape or another is to meet the traveller
everywhere in Northern Italy; either, as here, hanging on under the
sills of windows, or else running up the sides of gables, forming
string-courses and cornices, but always unsatisfactory, because
unmeaning and unconstructional. The origin of this sort of detail is to
be found in the numerous brick buildings not far distant, where the
facility of repeating the patterns of moulded bricks led (as it did in
other countries also) to this rather unsatisfactory kind of enrichment.
The detail of the arcades supporting the upper part of the building is
throughout bold and simple, and I should say of the thirteenth century;
the bases are quite northern in their section, the caps rather less deep
in their cutting, but still in their general design, and in the grouping
of tufts of drooping foliage regularly one above the other, reminding
one much of Early French work, though they are certainly not nearly so
good as that generally is. There is a flatness about the carving, too,
which gives the impression of a

[Illustration: 5.--BROLETTO, BERGAMO. Page 54.]

struggle, in the hand of the carver, between the Classic and Gothic
principles, in which the latter never quite asserted the mastery. The
lesson to be learnt from such a building as this Broletto appears to me
to be the excessive value of simplicity and regularity of parts
carefully and constructionally treated; for there are no breaks or
buttresses in the design, and all its elements are most simple, yet
nevertheless the result is beautiful.

To the west of the Broletto is a good open staircase (much like that in
the Piazza dei Signori at Verona),[5] forming a portion of one side of
the Piazza, and leading to the upper part of the buildings, and, I
think, to the great clock-tower, which, gaunt and severe in its outline,
undecorated and apparently uncared for, rears its great height of rough
stone wall boldly against the sky, and groups picturesquely with the
irregular buildings around it. I have omitted to notice that the whole
of the Broletto, with the exception of the window-shafts, is executed in
stone, and without any introduction of coloured material, so that it in
no way competes with the exquisite piece of coloured construction which
we have next to examine, immediately behind it.

A few steps will take us under the open-arched and cool space beneath
the Broletto, to the face of the north porch and baptistery of Sta.
Maria Maggiore. This is a very fine early Romanesque[6] church, but with
many additions and alterations on the outside, and so much modernized
inside as to be quite uninteresting to any one who thinks good forms and
good details necessary to good effect. The plan is cruciform, with apses
to the choir, on the east and west sides of the south transept, on the
east of the north transept, and at the west end of an additional north
aisle; in all no less than five apsidal ends. The nave is of three bays
with aisles, and to each transept have been added, in the fourteenth
century, porches, thoroughly Italian in their whole idea, and novel to a
degree in their effect upon an English eye.

A domed chapel, erected as a sepulchral chapel by Bartolomeo Colleoni in
the Renaissance style, on the north side of the nave, is most
elaborately constructed of coloured marbles. The effect is too bizarre
to be good; there is an entire absence of any true style in its design,
and there is nothing which makes it necessary to criticize it with much
minuteness.

The best and most striking feature in the whole church is the north
porch,[7] a most elaborate structure of red, grey, and white marble, to
which a drawing without colour can hardly do justice. It is supported
upon detached marble shafts, whose bases rest upon the backs of rather
grand-looking lions, curiously grouped with children and cubs. Above the
arches which rest upon these shafts, and which, though circular, are
elaborately cusped, is another stage divided by columns and trefoiled
arches into three spaces, the centre of which is occupied by a noble
figure of a certain Duke Lupus on horseback, with a saint on either side
in the other divisions. All the shafts except those in the upper
division are of red marble; the highest stage of all is entirely of grey
marble; in the middle stage all the moulded parts are of red, and the
trefoiled arches and their spandrels of grey marble; the space at the
back of the open divisions and the wall over the main arches of the
porch are built in courses of red and white marble. All the groining is
divided into diamond-shaped panels, composed alternately of black, red,
and white marble, all carved in the same kind of pattern. In the great
arch of the porch the outer moulding is of red marble, and all the
cusping of grey. The construction of the whole is obviously very weak,
and depends altogether for its stability upon iron ties in every
direction.

The approach to the porch, by seven steps formed alternately of black
and white marble, increases the impressiveness of the grand doorway, in
front of which it is built, the whole of which is of white marble, whose
carved surfaces and richly moulded and traceried work have obtained a
soft yellow colour by their exposure to the changing atmosphere, and are
relieved by one--the central--shaft being executed in the purest red
marble. There are three shafts in each jamb, carved, twisted, or moulded
very beautifully. These shafts are set in square recesses, ornamented,
not with mouldings, but with elaborate flat carvings, in one place of
saints, in another of animals, and with foliage very flat in its
character, and mainly founded on the acanthus.

To an English eye these columns in the doorways are some of the most
charming features of Italian architecture; but they must be always
looked at as simply ornamental, and not as constructional features; and
perhaps in all doorways the shafts, being really incapable of supporting
any considerable weight, would be better if, by their twisting and
moulding, it were clearly shown that their architect meant them to be
simply ornamental. In the Bergamo doorway the spaces between the shafts
are so strong in their effect, though carved all over their surface,
that any lightness in the columns themselves is amply atoned for. Such a
work as this northern porch at Bergamo is indeed a great treat to an
English architect, teeming as it does with fresh and new ideas, and in a
small compass showing so many of the radical points of difference
between northern and southern Gothic, and at the same time offering so
beautiful a study of constructional colouring, that it is impossible to
tire of gazing at it.

The porch to the south transept is of a simpler but somewhat similar
design. Both are placed against the western half of the gable against
which they are built, with a pleasant ignorance of those new-fangled
views of regularity of plan which are the curse of modern architects.
This southern porch is round-arched, and fitted exactly to the doorway
which it shields. Its outer arch is carried on detached shafts resting
on the backs of monsters, and it is mainly constructed of black and
white marble. It is of only one stage in height, and has a deep cornice
enriched with a series of niches with figures. An inscription below the
cornice gives the date as 1360.[8] Above the porch, but independent of
it, is a lofty monumental pinnacle corbelled out from the wall, and
richly sculptured with crocketed pinnacles and gablets. When the church
is entered, the reason for the apparently eccentric position of the
porches is seen. They were so placed to give more space for the altars
to the east of the transepts, and their successful effect is good
evidence that no artist need ever distress himself about a want of
regularity, if it is the result of a little common sense attention to
convenience in the arrangement of his plan.

The southern side of the church gives a very fair idea of what the
general character of the original building of 1134 was. The windows were
very plain, the walls lofty, the roof flat, and ornamented with
corbel-tables up the gables and under the eaves, and pilasters were used
at intervals instead of buttresses. There is a central octagonal
lantern which may be old, but which is entirely modernized. The most
interesting remains are the various apses already mentioned. They are of
two divisions in height, the lower adorned with very lofty,
boldly-moulded arcades, above which is an elaborate cornice, and above
this again a low arcade on detached shafts, behind which the walls are
considerably recessed to form galleries which produce a very deep
shadow. The capitals are elaborately carved, and the upper cornice is
again very rich. Altogether, little as remains unaltered of the old
fabric, it is enough to give an idea of a very noble and interesting
phase of art. Near a doorway into the north chancel-aisle the external
walls have traces, faint and rapidly decaying, of some very exquisite
frescoes or, more probably, tempera paintings.

[Illustration: CAMPANILE--BERGAMO.]

The steeple is in a most unusual position--east, namely, of the south
transept--not less, I believe, than some three hundred feet in height,
of good and very simple pointed character, without any approach to
buttressing, and remarkable as having an elaborately arcaded
string-course a few feet below the belfry windows, which have
geometrical traceries enclosed within semi-circular arches, affording,
like the south transept porch, a curious illustration of the
indifference of Italian architects to the use of the pointed arch where
strength was not of consequence.

Italian campanili have quite a character of their own, so distinct from
and utterly unlike the steeples of Northern Europe, that this, the first
Gothic example I had seen, interested me exceedingly. Perhaps its detail
was almost too little peculiar, if I may venture to say so; for
certainly it has left no such impression of individuality on my mind as
has the beautiful campanile to whose grace so much of the charm of
Verona is due.

The cathedral at Bergamo, which is close to the Broletto and Sta. Maria,
may be dismissed in a word. It has been rebuilt within the last two
hundred years, and appeared to be in no way deserving of notice. In a
courtyard on its north side is a small detached polygonal baptistery,
founded in 1275, which must have been very interesting. It is all built
of marble, and richly adorned with shafts; but so far as I could see
every portion of it has been renewed within a few years. Beside Sta.
Maria Maggiore and the Broletto we found little to see. Two
churches--one in the Città, and another, desecrated, in the Borgo--have
very good simple pointed doorways, with square-headed openings and
carved tympana; but beyond these we saw scarcely any trace of pointed
work. We had a luxuriously hot day in Bergamo, and, as we sat and
sketched the Broletto, a crowd, thoroughly Italian in its composition
and proceedings, gathered round us and gave us a first lesson in the
penance which all sketchers must be content to undergo in Italy. Before
long I found that my only plan was to start an umbrella as a defence
both against the sun and the crowd, and this, though not entirely
successful, still effected a great improvement.

The walk down the hill to the Borgo was more pleasant than the climb up,
and we were soon at our inn again; and then, after a most delicious
luncheon of exquisite fruit and coolest lemonade, concluded by a very
necessary dispute with our landlord about the amount of his bill,
ending, as such disputes generally do in Italy, with a considerable
reduction in the charge and the strongest expressions of regard and good
wishes for our welfare on our way, we mounted our carriage, and were
soon on the road towards Brescia.

Not far from this road and within about eight miles of Bergamo lies one
of the most interesting of the many castles of which one so frequently
sees remains in the North of Italy. This is the Castle of Malpaga, which
was inhabited by the famous Condottiere Bartolomeo Colleoni, of whom we
have already heard at Bergamo, and of whom we shall see something again
at Venice. It belongs now to a nobleman who lives in the Città of
Bergamo, and leaves this old and stately pile to the keeping of his
hinds, who tend his silk-worms, gather his grapes, make his wines, look
after his corn and cattle, and do as much as in them lies to gather the
fruits which mother earth yields in these parts with such ungrudging
profusion, but trouble themselves little about the preservation of the
old castle or its belongings, seeing that they seem to give scant
pleasure to their lord.

The castle as originally built was a square building enclosing a
courtyard built of brick externally, and adorned with a forked
battlement, which is common everywhere in old buildings between this and
Vicenza, and with four square corner towers, of which one larger than
the others has a very bold and fine overhanging machicolated parapet. In
the centre of the south front the drawbridge still remains in use, and
was lowered for our exit from the castle. Outside the square castle was
a space, and then a low wall again furnished with the forked battlement.
This must have been a very picturesque arrangement; but unfortunately
its real character is now only intelligible to the skilled eye. For the
great Colleoni, finding himself in possession of a castle which gave him
insufficient space for his magnificence, built up walls on the top of
the old battlemented outer wall, and created his state rooms in the
space between this new wall and the old external wall of the castle.[9]
These rooms of his have much damaged the effect of the outside of the
castle; but internally they are still interesting, owing to the
sumptuous character of the painted decorations with which he had them
adorned. These were executed at about the time of the visit of Christian
II. of Denmark to Colleoni, and are interesting if not great works of
art. The old courtyard though small is very fine in its effect. The
upper walls are carried on pointed arches and are covered with fresco or
distemper paintings, said to have been executed by Giovanni Cariani of
Bergamo, or by Girolamo Romanino of Brescia, extremely striking and
attractive in their general style of colour and drawing. The most
picturesque incidents are illustrations of Colleoni’s career--the Doge
of Venice giving Colleoni his bâton in the presence of the Pope, and a
fine battle subject.

A squalid area for rubbish, children, pigs, cats, and what not, is left
all round the moat, and beyond this are all the farm buildings and
labourers’ residences, which go to make up the _tout ensemble_ of a
great Lombard farmyard. The surroundings are not clean nor very
picturesque, but the castle itself has so great an interest, that no one
who visits Bergamo should pass it by unseen.

[Illustration: 6. CASTLE OF MALPAGA. p. 62.]




CHAPTER V.

    “Am I in Italy? Is this the Mincius?
     Are those the distant turrets of Verona?
     And shall I sup where Juliet at the masque
     Saw her loved Montague, and now sleeps by him?”
                     _Rogers._

     Palazzuolo--Coccaglio--Brescia: new and old Cathedrals,
     Broletto--Churches--Donato--Desenzano--Lago di
     Garda--Riva--Trent--Verona.


Our drive from Bergamo to Brescia was strikingly unlike what we had
hitherto been so much enjoying. Mile after mile of straight roads,
between fields so closely planted with fruit-trees that one never sees
more than the merest glimpse of anything beyond them, are certainly not
pleasant; and the hot sun above us, and the thirsty and dry beds of
rivers which we crossed on our way, made us feel glad when evening drew
on, and we found ourselves rapidly nearing Brescia.

I made notes at two or three places on the way. At Palazzuolo is a great
circular belfry, ornamented with a large figure at the top and divers
others about its base, built of brick rusticated to look like stone, and
altogether about as base a piece of architecture as could well be found,
but pardoned here because of the pure blue of the sky I saw behind it,
and partly on account of the view which it commands, reaching, it is
said, as far as Milan, and including the great plain out of which, upon
a slight hill, it rises. Palazzuolo is nicely situated, and upon the
first of the many rivers which we had passed from Bergamo which had any
water in its bed. The houses, too, were almost all supported on arcades,
giving pleasant shelter from the sun.

Beyond this we came to Coccaglio, a small village with a wretchedly bad
modern church, glorying in a glaringly sham front, and faced on the
opposite side of the street by the remains of a mediæval church--whose
place it has taken--and which is now shut up and rapidly going to ruin.
The new church is built north and south--the old one orientating
properly; but then the west front was the great feature of the new
church, and therefore it was necessary, of course, to place it towards
the road!

[Illustration: WINDOW--COCCAGLIO.]

Coccaglio still has, however, some very valuable remains of mediæval
domestic work in its houses, of which I was able

[Illustration: 7.--HOUSE AT COCCAGLIO. Page 65.]

to obtain some sketches. They were entirely executed in brick and
terra-cotta, except, of course, the capitals and shafts of the windows,
and appeared to be of the fourteenth century.

[Illustration: DETAIL OF WINDOWS AND CORBELLING FOR
CHIMNEYS--COCCAGLIO.]

The upper portion of the house of which I give a sketch remains very
fairly perfect, though its lower story has been entirely modernized. It
will be seen that it is very uniform in its design, the large and small
windows alternating regularly; and that semi-circular arches are used in
the windows in connection with ogee trefoils. This is one of the
apparent inconsistencies which occur in almost all Italian Gothic work;
and might seem to give us ancient authority for any amount of licence in
our combination of the elements of what we ordinarily consider to be
thoroughly different styles. The windows are marked by the same
elaboration of their sills which we noticed in the Broletto at Bergamo,
and the detail of these, as also of the corbelling out from the wall of
several chimney-breasts, is exceedingly good.

In a back street in the village I found a house the balconies around
which were corbelled forward on finely moulded beams, which, judging by
the moulding, could hardly be of later date than the commencement of the
fourteenth century.

[Illustration: WOODEN BALCONY--COCCAGLIO.]

Wooden mouldings of this kind are much rarer in Italy than they are in
the North, and I particularly notice this little relic, therefore, which
still remains to show how well the science of moulding was sometimes
understood even there.

Such a village as Coccaglio is, as I found afterwards, a place to be
made much of; for generally, except in public or important buildings in
large towns, one sees very little trace of any mediæval domestic work
beyond the perpetually recurring arcading under the houses which is so
general a feature in all the towns in the North of Italy.

There is nothing further of any interest on the road, and just after
sunset we reached Brescia, too late to see anything of the general
effect of the city.

Brescia is mainly famous, I believe, first for its connection with a
story of the generosity of Bayard, the “chevalier sans peur et sans
reproche,” and next for the large discoveries of Roman remains which
have from time to time been made there. It is one of those towns,
moreover, of which guide-books, with an immense list of churches and the
pictures they contain, give perhaps too grand an idea before they have
been seen. It is, however, undoubtedly a place of much interest, not
only for the antiquary, but also for the student of mediæval art, since,
though its churches are generally uninteresting, it has in the Broletto,
sadly mutilated and modernized as it is, the remains of one of the most
extensive and grand of these buildings, and to a considerable extent
executed in very excellent brickwork.

Our first visit in the morning was to the Piazza, in which stand the two
cathedrals--the old and the new--side by side, and just beyond them the
front of the Broletto, stretching its great length up a slight hill and
along a narrow lane beyond the Piazza, whilst at its angle, towering up
between it and the cathedrals, stands a tall and rugged stone campanile,
without break or window until at the top, where, just as in the
corresponding tower at Bergamo, great rudely-arched openings are left,
through which appear the wheels and works of the bells.

The new cathedral, approached by a flight of steps from the Piazza, has
a great sham front. It has, moreover, a large dome, said to be inferior
only in size to those of S. Peter’s and the cathedral at Florence, but
not prepossessing in its effect; nor did the church seem to contain any
pictures of value. By a descent of some twenty steps from the south
transept the old cathedral is reached. This is of very early date, and
constructed partly in stone and partly in brick. The most remarkable
feature is the nave, which is circular in plan, with an aisle round it;
the central portion, divided by eight arches from the aisles, being
carried up into a dome. The choir and transepts are projected on the
east side of the dome, and the former is groined and has a five-sided
apse. The walls retain some fair mediæval monuments, and beneath the
church is a large crypt. The old stone altar in the choir is a fine
example of the thirteenth century. The mensa is, as so often is the
case, carried on shafts, no less than sixteen in front and six at the
ends, with carved capitals. The stalls are in the apse, and a fine
lectern stands behind the altar.

The whole air of the Duomo Vecchio is chill and dismal to a degree; it
is neglected and dirty, and apparently shut up except for occasional
services, and left no pleasant impressions on our minds of our first
Lombard cathedral; and yet undoubtedly there is both here and at
Aachen--where the plan of the cathedral is so very similar--much to
admire in the idea of the plan, and I can quite imagine that a very
noble and useful church might in any age have been founded upon this old
Lombard type.[10]

Those who know anything of Spanish churches will be reminded here of the
two cathedrals at Salamanca, the relative positions of which are just
the same, a steep flight of steps in either case, leading down from the
south aisle of the new cathedral into the old and deserted one.

From the cathedral we went at once to the Broletto, which

[Illustration: 8.--BROLETTO. BRESCIA. Page 69.]

stands in the same Piazza. The main portion of this immense building
appears to have been built rather early in the thirteenth century. The
arches throughout are both round and pointed, used indifferently; but
this mixture does not betoken any diversity of date, as it would in
England.

[Illustration: CLOISTER--BROLETTO, BRESCIA.]

A large quadrangle is formed by the buildings, which has a cloister on
two sides, and traces of another cloister on a third side now built up.
The cloister still remaining on the cast side is ancient and on a large
scale; it opens to the quadrangle with simple pointed arches resting
upon heavy piers, and a row of piers running down the centre divides it
into two portions, so that it may be judged that its size is very
considerable. The groining has transverse and diagonal ribs, the former
being very remarkable, and, as not unfrequently seen in good Italian
work, slightly ogeed; not, that is to say, regular ogee arches, but
ordinary arches with the slightest suggestion only of an ogee curve in
the centre. Of the external portion of the building the west front is
the most perfect, and must always have been the finest; it consists of a
building containing in the upper story five windows, the centre being
the largest, and possibly once the Ringhiera, to the south of which
rises the great belfry of rough stone, and beyond that a wide building
with traces--but no more--of many of the original windows; north of the
building with the five windows is a very beautiful composition executed
almost entirely in finely-moulded bricks; it has an exquisite door with
some traces of fresco in its tympanum, executed mainly in stone, of
which I give a drawing, and a magnificent brick rose window, above which
is a brick cornice, which continues over the remainder of the west front
and along the whole of the north side.

[Illustration: DETAIL OF CIRCULAR WINDOW--BROLETTO, BRESCIA.]

[Illustration: DOORWAY--BROLETTO, BRESCIA.]

[Illustration: BRICK CORNICE--BROLETTO, BRESCIA.]

The size of the building is prodigious, and certainly the detail of all
the parts (excepting perhaps the cornice, which is of the common arcaded
kind) is most beautiful and valuable. The brickwork is so good and
characteristic that I have given several sketches of it. All the arches
have occasional voussoirs of stone, and the centre of the arch is always
marked by a key-stone, and these are sometimes slightly carved to
distinguish them from the other stone voussoirs. The abaci are of brick,
moulded and very varied. The doorway given in the woodcut on the
preceding page has stone jambs, caps and bases, lintel and outer arch,
the label and cusps being of terra-cotta; above this the whole of this
portion of the front is of brick, and very admirably built.

[Illustration: SAN FRANCESCO--BRESCIA.]

Of the churches of Brescia there seem to be but few of any interest;
that of San Francesco, of whose west front I give a sketch, is the best,
and, though not of uncommon design, is worth notice; the mixture of
white and black

[Illustration: 9. BROLETTO--BRESCIA.

Details of Archivolts]

marble and brick is very judicious; but I must protest once more against
the arcaded eaves-cornices, which are very elaborate and heavy; nor can
I bring myself to like the great flat gable, covering both nave and
aisles, and divided only by pilaster strips, which characterizes so many
mediæval Italian churches. In this west front of San Francesco the
cornices and the mouldings of the small circular windows are all of
brick, and the rest of the front of stone, the rose window having
voussoirs of black and white marble. The only other part of the church
which appeared to be of any interest was a campanile on the south side
of the choir: this had stone belfry windows, well treated with simple
plate tracery, and there is a singular and lofty lantern over one of the
chapels on the north side of the church, all of rich brickwork dating
from about 1480.

The sun was at its hottest as we wandered about the streets of Brescia;
but there was so much pleasure in the examination of the busy people who
thronged its narrow tortuous streets, that we enjoyed it very much. In
Italian towns, too, there is not much difficulty in finding the way; we
ask the road to some church, and forthwith, in place of a long and not
very intelligible direction, in which we are sure entirely to confuse
our right hand with our left, the person we ask turns round with us,
walks by our side, shews us our object, and, politely taking off his hat
and bowing, takes leave of us. It was by such aid as this that we found
the church of the Carmine, which is another very late Gothic church. The
west front is most fantastic and unpleasing, and the pinnacles composed
of round bricks, disposed alternately over each other, and common in
most Italian brick buildings, are very ugly; there is, however, a good
simple cloister attached to the church on the north; it is of the same
design as almost all in this part of the world, having simple round
shafts with carved caps and circular arches. An inner

[Illustration: CLOISTER OF THE CARMINE CONVENT--BRESCIA.]

cloister which I remember of old as occupied by the ever present
Austrian soldiers, is now (1872) open to all the world, and neither
cared for nor used. Here the south side of the cloister is of two
stories in height, the lower similar to the our just mentioned, the
upper having two arches to one arch of that below, and the arches
picturesquely shaped, being cinquefoils, with the central division of
ogee form, and with moulded terminations to the cusps. There is a fair
campanile here, with brick traceries and strings, but with a modern
belfry-stage.

A little bit of cloister, or gallery, on the north side of Sta. Afra,
has arcading of similar character in its upper gallery, but the arches
are trefoiled.

In the Contrada della Pace there remains a very bold fragment of a
castle tower. It is built of very roughly-jointed stone, and is
perfectly plain till near the top, where it has a bold machicolation
with tall square angle-turrets, the whole battlemented with a forked
battlement. Out of the centre of the tower a tall thin tower rises to
some height above the battlements.

One of the most picturesque spots in the city is the Piazza, at the end
of which stands the Palazzo della Loggia; the effect on coming into it
from the narrow streets in which we had been wandering was very
pleasant, the large open space being surrounded with rather elaborate
Renaissance work with rich coloured sun-blinds projecting from the
windows over the sunny pavement, which in its turn was thronged with
people in picturesque attire selling fruit and vegetables. The streets
are all arcaded, and some of them have very considerable remains of
frescoes on the exterior, giving much interest to the otherwise ugly
walls; they have, however, suffered very greatly from exposure, and are
only in places intelligible; still they give traces of brilliant
external colour, and are therefore much valued in my recollections of
Brescia.

Compared with Bergamo, Brescia has the air of a smart and busy place;
its streets are wider and better paved, and the smells which still
greeted us were not quite so bad as there. The staple manufacture of the
city seems to be that of copper vessels; shop after shop, indeed street
after street, is full of coppersmiths’ shops; the men all sitting at
work, and keeping up a ceaseless din of hammering, in open shops, so
that all the world may see them. Nor is the coppersmiths’ the only trade
that loves publicity, for here as elsewhere the barbers’ shops are very
amusing, quite open in front, with perhaps a yellow curtain hanging down
half way, affecting only to conceal the inviting interior, which however
is always sufficiently visible, occupied in the centre by a chair, on
which sits the customer gravely holding a soapdish to his chin whilst
the barber operates; and this going on all day makes one think that
shaving is, after all, one of the great works of an Italian’s time!

When we left Brescia the heat was intense; the road, too, was deep in
dust to an extent not to be understood in England. There had been a
drought of some weeks’ duration, and the much-travelled road from Milan
to Verona, along which our way now lay, plainly told the tale which the
dry, parched, cracked-looking earth on each side of our way, and the sad
faces of every one as they talked about the failure of the vintage,
amply confirmed. Unluckily for ourselves we had not taken the advice of
our driver to have a close vehicle, but had insisted upon having an open
carriage, the consequence of which piece of self-will was that we had
hard work, even with the aid of umbrellas, to protect ourselves from
_coups de soleil_. We now learnt that in hot weather in Italy it is not
always the best plan to have as much of the sun as one can get. In
England it always is, but he who acts on his English experience in Italy
will surely repent his mistake.

We managed, however, to exist through the clouds of dust, relieved
perhaps by the sight of a regiment of swarthy and unpleasant looking
Austrian soldiers marching through the sun and dust, many of them with
their knapsacks and arms, but all with great-coats on to preserve their
white uniforms. When we saw them we could not help contrasting our
relative lots, and then, feeling how much worse off they were than
ourselves, we went on a little more contentedly than before. The road,
too, became slightly more interesting; instead of miles upon miles of
straight lines, we had a more winding way, and after a time occasional
beautiful glimpses of the mountains which marked to us the situation of
the Lago di Garda.

We drove without stopping through Donato, a place of no interest
apparently save for the huge dome of its church, and then passing under
a very fine viaduct resting upon a long range of pointed arches, (which
carries the railway which soon after our return was opened, and now
whisks only too many travellers from Milan to Venice and back without a
halt on the way,) we commenced the descent towards the town of
Desenzano, beyond and above whose roofs stretched the beautiful expanse
of fair Lago di Garda, with its great calm surface, and fine group of
distant mountains hemming in with picturesque and irregular outline, its
upper end.

We soon reached the poor and desolate streets of the town, and diving
into the dark court of the not over-clean looking hotel, gave ourselves
up for a time to the contemplation of the quiet loveliness of the scene.
The contrast between the flat shores of the lower part of the lake and
the mountains which crowd around its head is very striking, and to this
it is that Desenzano owes all that it has of interest. We strolled out
for a short time, looked at washerwomen kneeling in small tubs on the
edge of the lake, and washing their linen upon the smooth face of the
stones which pave its shore, and then went on, as in duty bound, to look
into the church. This we found to be neither very old nor very
interesting, but curious as illustrating the extent to which, in Italy,
the practice is sometimes carried of putting altars in every direction
without reference to their orientation. Here the high altar and some
others faced due south, whilst most of the remainder faced east, and I
think scarcely one turned to the west.

The remains of an old castle rise picturesquely above the little
harbour, from which steamboats sail for the tour of the lake, and
bidding farewell for a time to dusty roads, let us embark on one of
these for Riva at the head of the lake on our way to Trent, which is,
artistically speaking, the northernmost really national city and
cathedral in Italy.

Our steamboat kept to the west side of the lake, touching at a few
villages and towns, and for the most part ploughing its way along
beneath some of the highest and most precipitous rocks that I know. We
took a band on board on the way, and discharged them into a big barge
under a cliff, on the top of which was being held a village festa, at
which they were to perform; and we all looked with no little compassion
upon the heavily weighted performers as we saw the people who had
preceded them climbing the steep mountain sides above us.

The Lago di Garda seems to me to be in its upper reach one of the most
beautiful of Italian lakes. But it should always be taken in the way we
went, for the contrast between the sublimity of the upper end and the
tameness of the lower end is so great that nothing but disappointment
would be felt by those who saw the head of the lake first.

One or two of the towns on the western shore have churches of some
interest. At Salo there is a Gothic church with windows which have a
wide external splay and an enriched brick moulding or label all round
them. The windows are of one light, and have ogee-cusped heads. Another
church, at (I think) Gargnano, is of much more importance. It is
cruciform, with a domical lantern at the crossing. The nave has a simple
clerestory and aisles, and the west end, built in black and white
courses, has one great arch which encloses the doorway, above this a
lancet window, and above this again a circular window without tracery.

From Riva--one of the most pleasant resting-places on the Italian
lakes--a good road through fine scenery leads to Trent, a journey of
some six or eight hours. The descent upon Trent is very fine. The town
standing by itself well away from the fine mountains which form the
background to the view, the old walls and towers around it, and the
interest of the cathedral and other buildings behind them, combine to
give a sense of the importance and grandeur of the city which the facts
of the case hardly justify. Perhaps, too, there is something in the
historical importance of the place which, without one’s knowledge, sways
the judgment.

There is in fact only one building of great architectural
importance--the cathedral; but, as I shall shew, it has the greatest
interest, not only on account of its real merit, but also because it is
a startling example of the way in which, in the thirteenth century,[11]
the Lombard architects adhered to their old lessons and habits in spite
of all the developments which were then universally accepted on the
northern side of the Alps.

The church is a round-arched building throughout, but the mouldings and
details everywhere show a knowledge of thirteenth-century work, and have
none of the character of true Romanesque or Lombard art. Yet at the same
time there is in many respects a most close imitation of Lombard
features. There are arcades under the cornices of the aisles, arcades
under the eaves of the apses, open porches supported on shafts whose
bases rest on monsters, and other features which, looked at apart from
the sections of mouldings and details of sculpture, might well warrant a
much earlier date being fixed on for the execution of the work than I
have named. An inscription which fixes the date of some works here in
1212[12] may fairly, I think, be assumed to give the date of the greater
part of the fabric, though some portions, as e.g. the western wheel
window and the northern porch, are probably not so early by at least a
hundred years. But I am not concerned to deal with the question only
from an archæological point of view, and will at once therefore go on to
give some description of the building.

The ground-plan is in the shape of a Latin cross--with an eastern apse,
two small apses to the east of the transepts, and a nave and aisles of
seven bays. There is an octagonal lantern over the crossing, and the
whole church is groined. The doors are, two in the east walls of the
transepts, one at the west end (of marble), and one with a projecting
open porch over it near the east end of the north aisle. Two western
towers were intended to be built, and the staircases to them are carried
up in the western and southern aisle walls in a very unusual and
picturesque fashion. They commence in the third bay east of the towers,
and are carried up in a continuous rise, opening to the church with a
series of arches stepping up to suit the level of the staircases. The
western bay to which these stairs lead is groined at a lower level, as
well as at the nave level, so as to form a very lofty gallery open to
the church.

The clerestory consists of very small windows, and there is no
triforium; the main portion of the columns goes up to

[Illustration: 10. DUOMO. TRENT. p. 81.]

and carries the groining, and though the main arches are all
semi-circular, there is nevertheless an evident attempt--and it is
successful--to give an impression of height to the interior. The
continuous arcades under the eaves are carried also across the front of
the transepts, and give a great effect of richness to the external
architecture. Of the two towers only one is complete, and this was built
in the sixteenth century. The northern porch is the only place in which
the pointed arch appears, and it seemed to me to be of the fourteenth
century, though the doorway is of Lombard character, with very quaint
but poor carving in its tympanum, of Our Lord with the four Evangelists.
The whole church is built of stone, and has a Classical want of life and
vigour which one notices only too often in the best Lombard work. Were
it not for the building attached to its north-east angle, I suspect the
general impression would be much less agreeable than it is. This is a
lofty erection with two square turrets and a small apse at the east,
parts of which seem to be earlier than the date I have given to the
cathedral. It is connected with the north-east angle, but its axis is
not parallel with that of the cathedral, and there is consequently a
good deal of picturesqueness in the perspective, besides which it
prevents the otherwise insipid outline of the whole building being
perceived.

The porch on the east side of the south transept, of which I give an
illustration, is one of the most interesting portions of detail in the
building. Its front is supported on two shafts, one of which is an
octagon, resting on the back of a lion, the other four shafts cut out of
a single block and ingeniously knotted together in the centre, and
resting on the shoulders of four sitting figures--altogether about as
strong an illustration of mediæval love of change and variety as could
be found. It must have been the work of a sculptor who was just a little
savage with the somewhat tame uniformity of the whole of the
architectural scheme of the cathedral.

I found little else to see in Trent. Sta. Maria Maggiore has a
Romanesque steeple, quite plain below, but with its two upper stages
arcaded, the arcades resting in the lower stage on shafts coupled one
behind the other, and in the upper tripled in the same way. I do not
remember before to have seen this last arrangement. A tower in the walls
between this church and the cathedral is a rhomboid in plan, and was, I
suppose, built of this strange shape to suit some necessity arising out
of the position of streets and walls. Considerable portions of the walls
remain; they are of stone, finished on the top with the forked Italian
battlement, and having square projecting towers at short intervals.

Trent Cathedral and the fine church at Innichen in the Pusterthal, are
quoted frequently as the two finest churches in Tyrol. That at Innichen
has not the same entirely Italian character which marks that at Trent.
In the latter I always feel that climate, people, and town are all in
concert to make one suppose oneself in Italy, which certainly is not the
case at Innichen. North of Trent the architecture of the Tyrol (as at
Botzen and Meran) is entirely German, whilst in Trent itself, were it
not for a steep roof here and there covered with bright glazed tiles,
and a few such slight indications, no one would suspect the presence of
any German influence whatever.

I have travelled so frequently from Trent to Verona by the railway that
I always regard it as one of the most natural and obvious roads of
approach to Italy for Englishmen. It takes its course through so fine a
country that one does not easily tire of the journey, and finally, it
sets travellers down in the city which, perhaps more than any other in
northern Italy, charms the cultivated traveller by the beauty, interest,
and grandeur of its buildings. Who that has taken this way to Verona
does not remember with pleasure the last quarter of an hour of his
journey, as the railway, making a circuit round two-thirds of the city,
reveals first a mixed group of lofty steeples, presently the great
church of San Zenone, then Sta. Anastasia, anon San Fermo, then crosses
the swift-flowing Adige, and at last lands one at the station, full of
anxiety to make the nearer acquaintance of the buildings of “Verona la
degna,” which from afar look so wondrous brave and fine?




CHAPTER VI.

    “Come, go with me. Go, sirrah, trudge about
     Through fair Verona.”
              _Romeo and Juliet_, act i. scene 2.

     Verona: Campanile of the Palazzo dei Signori--Sta.
     Anastasia--Monuments--Piazza dell’ Erbe--The Duomo--The
     Baptistery--Sta. Maria l’Antica--Cemetery and Palace of the
     Scaligers--Domestic Architecture--Piazza di Brà--The
     Austrians--Ponte di Castel-Vecchio--San Zenone--San Fermo
     Maggiore--Chapel near the Duomo--Romeo and Juliet--Dwarfs--Wells.


We reached Verona in the evening, and were up early on the next morning,
anxious to get a general idea of the city. But I was no sooner out of my
bed than I saw from my window, over the roofs of the opposite buildings,
the campanile of the Palazzo dei Signori, a lofty, simple, and almost
unbroken piece of brickwork, rising, I suppose, at least three hundred
feet into the air, and pierced with innumerable scaffold-holes, in and
out of which, as I looked, flew countless beautiful doves, whose choice
of a home in the walls of this tall Veronese tower will make me think
kindly of putlog-holes for the future. Certainly, if the Italian and
English principles of tower-building are to be compared with one
another, the Italian need give no fairer example of its power than this
simple and grand erection.

It rises, as we found afterwards, out of a large pile of buildings, and
for a short distance above their roofs is built in alternate courses of
brick and a very warm-coloured stone,

[Illustration: 11.--CAMPANILE, PALAZZO SCALIGERI, VERONA. Page 84.]

and then entirely with brick, pierced with only one or two small
openings, and terminating with a simple belfry-stage; the belfry
windows, with their arches formed without mouldings, and with the sharp
edges only of brick and stone used alternately, are divided into three
lights by shafts of shining marble; the shafts, being coupled one behind
the other, give strength with great lightness, and are very striking in
their effect. These windows have, too, remarkably large balconies, but
without balustrading of any kind. The upper and octangular stage of the
campanile is comparatively modern, but rather improves the whole effect
than otherwise.

I could hardly tear myself away from this noble work; but much more was
to be seen, so I dallied not long before I set forth on a journey of
discovery, giving myself up gladly to sketching and ecclesiology.

The hotels in Verona are both of them near Sta. Anastasia, and at the
eastern end of the long and at first narrow and picturesque Corso. The
Adige separates the city from its eastern suburb, and from the hills
crowned by the Castel San Felice and the picturesquely stepped city
walls. Its yellow waves wash with an angry rush the foundations of the
houses which overhang it all along its course, but the only views of it
are to be obtained from the bridges, and from the open space near the
Castel Vecchio. At the extreme north-western angle of the town stands
the church of San Zenone. One soon finds one’s self constantly on the
Corso, and to the north of this lie the cathedral, Sta. Eufemia, the
Castel Vecchio, and San Zenone, whilst to the south of it are the tombs
of the Scaligers, San Fermo Maggiore, the Roman Amphitheatre, and the
Palazzo Publico. Without further attempt to describe the map let us
visit the buildings, of which the list I have given, though by no means
exhaustive, includes the finest.

The Veronese architects in the Middle Ages were certainly some of the
best in Italy. San Zenone is by very much the finest church of its kind
that I know; Sta. Anastasia is on the whole one of the best churches of
a later date; and San Fermo Maggiore affords some of the best detail of
brickwork, and the tombs of the Scaligers the best examples of monuments
in all Italy.

The first thing seen on turning out of the hotel is the west front of
the church of Sta. Anastasia, looking so beautiful at the end of the
narrow street, whose dark shade contrasts with the bright sunshine which
plays upon its lofty arched marble doorway and frescoed tympanum, and
lights up by some kind of magic the rough brickwork with which the
unfinished church has been left so brightly, that, as you gaze, thoughts
pass across your mind of portions of some lovely painting or some
sweeter dream; you feel as though Fra Angelico might have painted such a
door in a Paradise, and as though it were too fair to be real. There,
however, it is, rich and delicate in colour, shining with all the
delicate tints of the marbles of Verona, pure and simple in its
softly-shadowed mouldings, beautiful in its proportions, and on a nearer
approach revealing through the dark shade of its opening, and over and
beyond the people who early and late throng in and out, the vague and
misty forms of the solemn interior.

Sta. Anastasia is one of the most complete and representative pointed
churches in the North of Italy, and deserves, therefore, a rather
detailed description. Its date is about 1260 to 1290. The ground-plan is
very simple--a nave of six bays, then one which is the crossing of the
transepts, a very short choir of one bay finished with an apse and two
chapels on the east side of each transept. The nave aisles fire narrow,
and the whole design is characterized by intense simplicity of detail
and arrangement. The width of the nave, and the height of the columns
and arches, give, on entering, an idea of vast space and size. The
columns are very simple, cylindrical in section, and support arches
built of brick, and only chamfered at the edge; from the caps of the
columns flat pilasters run up to the commencement of the groining, and
above the nave arcades there are two small circular openings, one in the
place of a triforium, opening into the roof of the aisle, the other
above it and larger, filled in with plate tracery in stone, and forming
the clerestory.

[Illustration: STA. ANASTASIA--VERONA.]

The arrangement of the plan is in several respects unlike that of
northern Gothic buildings, and as most complete Gothic churches in Italy
are very much of the same type, it is as well here to point out the
peculiarity. The most marked features are first the shortness of the
choir, and next, the fact that whereas in northern Europe it is usually
the aisle vaulting bay which is square, whilst the nave bay is oblong
from north to south, here the nave groining bay is square, and the bays
of the aisle oblong in the direction of east and west. The difference of
effect is great, simple as the statement seems. In these Italian
churches there is much greater space between the columns of the nave
arcades, the groining is consequently divided into much larger bays, and
the whole interior has a largeness of treatment which is not common in
the North. On the other hand there is much less of the complexity and
intricacy which are so charming in our interiors, and you see at a
glance the whole of the church.

In Sta. Anastasia the apses on each side of the choir have an even
number of sides, as has also the sacristy, which is a room on a
magnificent scale, to the north of the church. This is a peculiar
feature, producing as it does an angle in the centre of the apse, which
we shall see again at Venice and Vicenza, and which is seldom seen out
of Italy.[13]

The whole is so simple in design and construction that it depends for
its rich effect on the painting which covers almost every part of it,
and which harmonizes well with the architectural lines. The decorations
appear to have been done, or at any rate commenced, within a short
period of the completion of the church, and are therefore very valuable.
The ground of the painting is white, many of the patterns of borders
being very elaborate compositions of flowers and foliage. The main
arches are painted to represent voussoirs of red brick and stone, but I
am inclined to think that they are really entirely of brick; their
soffeits, which are very broad and flat, are all painted with large
scroll patterns of foliage. In the groining the diagonal ribs are
painted at the intersection with stripes of colour alternating with
white, and on each side of all the ribs a wide border of foliage is
painted, whilst in the centre of each groining-cell some large device is
painted in a medallion, some of these being merely ornamental, others
having figures. The detail of much of the painting is cinquecento in its
character, and not valuable as an example to be literally copied,[14]
but its general effect is certainly very beautiful, and it is worthy of
all praise in respect of the strictness with which it is kept
subservient to the architecture, and in some respects, indeed,

[Illustration: 12. SANTA ANASTASIA. VERONA.

Pavements. Page 89.]

even serves to atone for its deficiencies, as, e.g., in the broad
painted horizontal borders, which take the place, very successfully, of
brick or stone string-courses, which in the construction are entirely
omitted.

It surprised me, I confess, very much, to find a church painted
throughout without any use of gold, and yet with good result; so it is,
however; the effect is most solemn and religious, and there is a very
rich effect of colour; the fact is, that the white ground answers the
same purpose in a degree, though of course not to the full extent, that
gold would.

But if the walls are beautiful in their colour, not less so is the
pavement, which, from one end of the church to the other, remains to
this day to all appearance just as it was on the day that the church was
finished. The nave and transepts are all in one pattern; the spaces
between the columns in a variety of beautiful designs, and divided from
the nave and aisle pavements by a strip of white marble on each side;
and the aisles again are on the same scheme throughout. The colours of
the marble used are white, red, and bluish grey, and the patterns very
simple and generally geometrical in outline, and there is a quiet
richness of effect in their arrangement which is exceedingly beautiful.
Such a pavement must unhappily be for ever Italian, and we in England
can scarce hope ever to attain to anything so exquisite; but we do not
well to forget that by the mixture of a small quantity only of marble
with our encaustic tiles we should attain to much greater beauty of
effect than we can by the use of tiles alone, and there are many
towns--as, e.g., Plymouth--the very pavement of whose streets is of a
material which might most advantageously be introduced, more often than
it has yet been, inside the walls of our sanctuaries, as well as under
the feet of every passer along the streets.

There are some monuments and paintings here quite worth looking at. The
Pellegrini Chapel, next to the choir, has two fine trefoil-headed
monuments in red marble with the background painted with subjects of
about the same age (circa 1392); and in the Cavalli Chapel there is an
admirably painted wall, against which has been put a monument which,
though somewhat rude and coarse in its sculpture, nevertheless produces
a very fine effect of colour and architecture combined.[15]

[Illustration: AISLE WINDOW--STA. ANASTASIA, VERONA.]

The window tracery of Sta. Anastasia is rather singular plate tracery,
consisting of mere piercings through the stone with very little
moulding; most of the windows are of two very lofty trefoiled lights
with circles and trefoils pierced above, very simple and severe, and
remarkable for the quaint way in which the cusping is arranged, not with
some reference to vertical lines, as is ordinarily the case with us, but
just as fancy or chance seems to have dictated. The clerestory windows
are circular openings cusped.

There is a curious but not very happily treated arrangement inside--a
step all round the inside walls, projecting some three or four feet from
them, and panelled all round against the riser with a small trefoil
arcading--the whole in red marble.

Externally the church is almost entirely built in red brick with rich
cornices and rather ungainly buttresses and pinnacles of brick; the
windows have brick jambs with stone tracery, and on the north side of
the choir is a fine lofty campanile, finished at the top with a low,
very plain, and octangular capping, and unpierced with openings, except
in the belfry-stage. Of the west front only the doorway has been
completed; this is in courses of red, grey, and white marble, and most
effective; the rest of the front is left in brick, finished exceedingly
roughly, with a view to leaving a key for the marbles with which, no
doubt, it was intended to veneer the entire front. The wooden framework
of this door, of which I give a detail, is very curious; it is of deal,
coëval with the doorway, and the framework is external, not internal.

The west front of the church stands in a small piazza, on the north side
of which is the little church of San Pietro Martire, between the east
end of which and Sta. Anastasia is a wall dividing a small burial-ground
from the Piazzetta; on the top of this wall, and supported upon corbels,
is one of those monuments peculiarly associated with Verona, because so
numerous there, though they are often met with elsewhere in Italy; they
are either large pyramidal

[Illustration: DOOR-FRAME--STA. ANASTASIA.]

canopies supported upon trefoiled arches resting on four marble shafts,
with a kind of sarcophagus or an effigy beneath; or else, when attached
to a wall, they have two detached shafts supporting the same kind of
trefoiled arch and surmounted by a flattish pediment. Their effect is
almost invariably beautiful in the extreme, and their only defect is,
that they all require to be held together by rods of iron connecting the
capitals of the columns. This, however, is soon forgotten when one feels
that there is no pretence ever or anywhere at its concealment; and
notwithstanding this slight defect, one cannot help loving and admiring
them; for there is a grace and beauty about the form and proportion of
the Veronese trefoiled arch, such as is never seen, I think, elsewhere,
and the very flatness of the carving and the absence of deep moulding
seem all adopted in order that nothing may interfere with the simple
beauty of the outline of the arch. In this case the monument is
supported on a large slab of stone corbelled forward and balanced upon
the top of a thin wall over the archway which leads into the churchyard
of San Pietro. Four shafts with sculptured capitals, resting on the
angles of this slab, support four trefoiled arches, (those at the ends
narrower than the others), which are almost destitute of moulding save
that the outer line of the arch has a broad band of delicate sculpture
all round it. The arch terminates in a small cross, and above on each
side is a very flat pediment, moulded and finished on the under side
with one of the favourite Italian arcaded corbel-tables; the finish is a
heavy pyramidal mass of stone rising from behind the pediments. The four
bearing-shafts are of white marble, all the rest of the monument of red.
Within the four supporting shafts stands a sarcophagus, supported on the
backs of couchant lions, very plain, but ornamented at the angles in
very Classic fashion, and bearing a recumbent effigy.

The church of San Pietro has three or four smaller monuments enclosed
within its small courtyard; two of them are on the ground and have round
arches on detached shafts with an Agnus Dei carved in the centre. The
third is corbelled out from the wall, and the face of the monument is
covered with sculpture, that of the figures being very inferior in style
to that of the foliage enrichments. This little group of monuments,
varying in date from 1283 to 1392, is well worth study, and I found it
more than usually interesting owing to its entire unlikeness to any
English work. The church against which these monuments are built is
small but deserves notice; it is of brick with a stone canopy on shafts
corbelled out above the west door; the buttresses are mere pilasters,
and run up without any weathering till they finish in an arcaded
corbel-table at the eaves; the windows have wide brick splays outside,
and trefoil heads of stone without any chamfer or moulding; on the south
side, the church--in point of size a mere chapel--is divided into four
bays, one of which has a monument corbelled out by the side of the
window. It is built entirely of red brick, not relieved in any way,
except that the window arches are in alternate voussoirs of brick and
stone. The wooden framework of the west door deserves notice as being
very unlike the English mode of door-framing, and very good in its
effect. The accompanying sketch will best explain it: and it must be
understood that, instead of being internal, as such framing would
usually be in England, in this example it is external, just as in the
somewhat similar door of Sta. Anastasia, of which I have already made
mention.

[Illustration: DOOR-FRAME--SAN PIETRO MARTIRE.]

Turning back from the Piazzetta of Sta. Anastasia, and traversing again
the narrow and gratefully shady street by which we reached it, we soon
found ourselves at the end of the Piazza dell’ Erbe, the most
picturesque square in the city, and at an early hour in the morning
quite a sight to be seen. The whole open space was as full as it could
be of dealers in vegetables and fruits, all of them protecting
themselves and their stores from the intense glare of the sun under the
shade of prodigious umbrellas, at least five times as large as any of
ordinary size, and certainly five times as bright in their colours, the
prevailing colour being a very bright red. Altogether it was a
thoroughly foreign scene. The houses, some ancient, but all picturesque
and irregular, surrounding the irregularly-shaped Piazza, the
magnificent campanile of the Scaligeri Palace rising proudly behind the
houses on the left, the fountain in the centre, and the great column of
red Veronese marble rising close to us, which at one time bore the
winged lion--mark of the dominion of the Venetians--all combined to
produce a very striking picture. An hour or two later in the day when we
passed, the people, the umbrellas, and the fruit were all gone, and
somewhat of the charm of the place was gone with them.

From the Piazza dell’ Erbe we went to the cathedral, anxious to see and
hear somewhat of a service which we found was to take place. There was a
great throng of people, and we had some difficulty in finding even
standing room among them; we were not at all sorry, however, to have
gone, for we came in for a sermon most energetically preached, and
enforcing in very powerful language the necessity of repentance. The
pulpit was very large, and as the preacher delivered his sermon he
walked from side to side, and often repeated again to those on his left
the substance of what he had already said to those on his right. The
people, who crowded every available place within hearing, were
exceedingly silent and attentive; and at intervals the preacher stopped
for a minute to cough and use his handkerchief, which was a signal for
an immediate general blowing of noses and coughing all over the church.
The “_Ebben infelici!_” with which he commenced his sermon, was a good
index to its whole tone, and makes me remember with pleasure the vast
crowd listening to Christian doctrine in the grand nave of the Duomo.
Would that we could see any prospect of the day when in England our
larger churches may be used in this way, when, with pews and all their
concomitant evils swept away, we may see a vast crowd standing and
sitting, leaving no passage-way and no waste room, anxious only that
they may, by pressing near, join in the services at the altar, and hear
every word of warning and of advice! The nave of Westminster, so
thronged, would soon show how great has been our mistake in leaving our
large churches so long unused.[16]

The Duomo is a really fine church, Romanesque in its shell, but altered
completely internally in the fourteenth century. It has only five bays
in length, but the dimensions are so large that the nave and aisles
alone measure about 225 feet × 97 feet inside. There are slightly
recessed Romanesque apses in the side walls, and a great tomb or shrine
of S. Agatha, which is worth looking at. The columns in the interior are
very lofty, of red marble, and, instead of being plain and cylindrical,
as at Sta. Anastasia, are moulded with very good effect. Their capitals
are, however, heavy, and the carving of the foliage on them not at all
satisfactory. The whole interior is very solemn, and specially beautiful
on this bright September day, with almost all the light excluded by
thick curtains--here and there a stray gleam of the most intensely
bright light finding its way through some chink, lighting up with sudden
brilliancy the deep cool shade, and--reflected from the bright surface
of some great marble shaft--suggesting the grandeur which can hardly
really be seen.

The choir is divided from the nave by a screen of marble, consisting of
detached columns with Ionic capitals and a continuous cornice, the whole
screen semi-circular in plan, and coming forward into the nave. This,
the work of Sanmichele, is certainly about the most effective work of
this kind and age that I have seen.

The west front of the Duomo is still in the main in its original state.
It has a rich porch supported on shafts (whose bases rest on the backs
of lions), and of two stages in height, a flat pediment surmounting the
upper stage. Another flat pedimental cornice is carried over the whole
front at the same level; but out of this a wall rises above the central
portion the width of the nave and height of the clerestory, which is
again surmounted by a third flat pediment--a confused and not very
graceful arrangement, which found an imitator in Palladio, when he built
that ugliest of fronts, the west end of San Giorgio Maggiore at Venice.

The windows at the ends of the aisles are very Italian in their
character. They are insertions in the wall of two narrow lights within
an enclosing arch, which is again surrounded by a square line of
moulding. The effect of this not uncommon Italian arrangement is
exceedingly unsatisfactory, as it appears to make the window with its
arch and tracery quite independent, constructively, of the wall in which
it is placed; it appears in fact to be merely veneered on to the face of
the wall.

On the north side of the cathedral is a cloister of good detail, which
was originally of two stages in height, but is now considerably altered
in parts. It is exceedingly similar to the cloister of San Zenone, which
we shall see presently, consisting of a long arcade of pointed arches
carried on marble shafts; it is now in a very ruinous condition.

The Baptistery--San Giovanni in Fonte--is a detached church near the
east end of the cathedral, which does not look as if it had been at
first intended for its present use. It is a building of the twelfth
century, with nave and aisles ending in three apses, lighted by a small
clerestory, and with some columns which are probably Roman. The font,
which stands in the centre, is of enormous size, and is sculptured with
eight subjects beginning with the Annunciation and finishing with the
Baptism of Our Lord. These sculptures are very rude. The bason of the
font is almost seven feet in diameter, and in its centre is another
smaller bason of graceful shape. I do not happen to have seen an
explanation of these inner basons in the fonts in Italian baptisteries.
They look as if they were meant for the priest to stand in whilst he
immersed the catechumens around him.

And now that we have visited two of the great churches, we must no
longer delay our visit to the church of Sta. Maria l’Antica, whose small
burial-ground is fenced from the busy thoroughfares, which on two sides
bound it, by an iron railing of most exquisite design, divided at
intervals by piers of stone on whose summits stand gazing upward as in
prayer, or downwards as in warning to those who pass below, a beautiful
series of saintly figures. Within, a glorious assemblage of monuments
meets the eye--one over the entrance doorway, the others either towering
up in picturesque confusion above the railing which has been their
guardian from all damage for so many centuries, or meekly hiding their
humility behind the larger masses of their companions.

The monuments are all to the members of one family--the Scaligeri--who
rose to power in the thirteenth century, and held sway in Verona until
almost the end of the fourteenth. In this space of time it was,
therefore, that these monuments were erected, and they are consequently
of singular interest, not only for the excessive beauty of the group of
marble and stone which, in the busiest highway of the city, among tall
houses and crowds of people, has made this churchyard, for some five
hundred years, the central point of architectural interest, but because
they give us dated examples of the best pointed work during nearly the
whole time of its prevalence in Verona. In the monument of the first
Duke we see the elements of that beauty which, after ascending to
perfection in that of another, again descends surely and certainly in
the monument of Can Signorio, the largest and most elaborate of all,
and, therefore, I am afraid, the most commonly admired, but the one
which shows most evidence of the rise of the Renaissance spirit, and the
fall of true art. Nor is it, I think, to be forgotten, as an evidence of
the kind of moral turpitude which so often precedes or accompanies the
fall of art, that this Can Signorio first murdered his own brother
Cangrande II. that he might obtain his inheritance, and then, before he
died, erected his own monument, and adorned it with effigies of SS.
Quirinus, Valentine, Martin, George, Sigismund, and Louis, together with
allegorical figures of the Virtues with whom he of all men had least
right to associate himself in death, when in life he had ever despised
them. The inscription, which records the name of the architect on this
monument, does but record the vanity of him who was content thus to
pander to the wretched Can Signorio’s desire to excuse the memory of his
atrocious life by the sight of an immense cenotaph.[17]

The tomb of Cangrande I. forms the portal of the church as well as the
monument of the first and greatest of the family. It is perhaps
altogether the finest of all; the shafts which bear the pyramidal canopy
are supported on corbels; between them is a simple sarcophagus
sculptured with a bas-relief, and upon it lies Cangrande with his arms
crossed in token of his resignation and faith. At the top of the
pyramidal covering is the figure of the brave knight riding forth to war
on his gaily caparisoned steed.

Next to this monument in date, as in merit, is that of Mastino II.,
wanting perhaps in some of the severe simplicity of the other, but even
more striking, as it stands at the angle of the cemetery. It is a
thoroughly grand and noble erection of two stages in height, the lower
unimportant, and only serving as a means of raising the monument
sufficiently high to be well seen from the exterior; upon this stand
four shafts, between which, and supported upon four much smaller shafts,
is the sarcophagus on which lies the recumbent effigy, at whose head
stand angels with expanded wings[18] guarding the deceased. The
sarcophagus is adorned with bas-reliefs--that on the west side being the
Crucifixion--and has engaged angle-shafts. The four main bearing-shafts
at the angles of the monument have finely carved caps with square abaci
from which rise simple trefoiled arches with steep pediments on each
side filled with sculpture in relief, and between these are exquisitely
simple niches, each a miniature reproduction of the entire monument, and
containing between their delicate detached shafts figures of saints. The
whole is finished with a heavy pyramidal capping, crocketed at the
angles with crockets so abominable in their shape and carving that they
go far to spoil the entire work, and surmounted by the figure of the
Capitano del Popolo, spear in hand, riding on his war-horse; the horse
and horseman riding with their faces towards the setting sun, as all in
life must ever ride; the effigy below lying so that at the last day the
beams of the day-star in the East may first meet its view, and awaken
him that sleepeth here in peace.

This contrary position of the figure in life and in death, observed also
in others of these monuments, is an evidence of the care and
thoughtfulness with which every detail of these noble works was wrought
out.

The monument of Can Signorio is not worthy of so long a description; it
is octagonal in its plan, and in many respects below the idea shadowed
out so beautifully in the others; the reduplication of niches and
gables, far from improving, only perplexes the design: and when to this
is added that the carving throughout, as well as the other details, show
strong signs of a leaning towards Renaissance, one may see some reason
why this, the most elaborate and complicated of all the monuments, is
after all far from being the most successful.

Mastino II. died in the year 1351, and we may therefore, I think, look
upon his monument as a fair enough example of Italian architecture just
at the period at which in England it had reached its culminating point,
and a careful examination of it cannot, therefore, be thrown away. In
the first place, I must notice that the sculpture, which has the air of
being rather sparingly used as too sacred a thing to be idly or
profusely employed, is exceedingly good. The foliage is almost always
very closely copied from natural forms, is very thin and delicate in its
texture, and thus really present to the gazer that idealized
petrifaction of nature which it ought always to be the sculptor’s effort
to give, and not, as is, I fear, sometimes the case, even in good
English work, so profusely scattered over the whole surface as to give
one a sense of its lack of great value. The worst part of the carving
is, as I have before said, that of the crockets, which are as bad as the
worst modern Gothic could be. The sculpture of the human figure is
throughout very good; remarkable for simple, bold, deep folds in the
draperies, quite Gothic in spirit, and much more akin to our best
fourteenth-century work than to any Classic examples.

As an example of the science of moulding this work is however valueless;
there is absolutely no moulding upon it; and why should there be? Would
it have been well that the lovely marble, whose brilliant white gloss
was sure ere long to be stained with dark streaks of black by the
beating of rain and the staining of age, whilst here and there the white
would stand out more brilliantly than ever,--would it have been well, I
say, that this should have been still further streaked with deep lines
of many mouldings? Most assuredly not: the architect had to deal with a
material which best takes its polish and exhibits its beauty and purity
when used in flat surfaces and in shallow carving, and he did right
therefore in not moulding it as he would have moulded stone.

There is a sharpness and hardness about the lines of the arches,
however, which perhaps almost verges upon rudeness, and, though I can
see that it may be fairly defended, I could yet wish that it might have
been softened.

[Illustration: CREST OF METAL RAILING--VERONA.]

But the points in which such work is a grand example to us are, first,
the value which it shows that we ought to place upon the simple detached
circular shaft, and, next, the beauty and strength of effect which the
cusping of a large arch in a proper manner gives. In these two points
this monument and most of its class teach us lessons which we ought not
to be unwilling to learn, and which, if we at all wish to develope
beyond the point at which our own ancestors ever arrived, we must not
fail to attend to in our own work.

[Illustration: METAL RAILING--VERONA.]

And now I must bid farewell to this lovely spot, the most attractive
certainly, to me, in Verona. The situation of the monuments, rather
huddled together, with the old church behind them, the archway into the
Piazza dei Signori on the other side, and the beautiful iron grille[19]
which surrounds them, the number of saintly and warlike figures and the
confused mass of pinnacle and shaft, half obscured by the railing, do, I
verily believe, make the cemetery of Sta. Maria l’Antica one of the best
spots in the world for the study of Christian art in perfection. What
either Köln or Regensburg Cathedral, or the Wiesen-Kirche at Söest is to
Germany, the Choir of Westminster Abbey or the Chapter-House at
Southwell to England, Amiens Cathedral or the Sainte Chapelle of Paris
to France, such is the Cemetery of the Scaligeri in Verona to Italy--the
spot where at a glance the whole essence of the system of a school of
artists may be comprehended, lavished on a small but most stately effort
of their genius.

Close to their burial-place stands also the Palace of the Scaligers. The
old portion of this fronts towards the Piazza dei Signori, a small
square used only by foot-passengers, and surrounded by elaborate
Renaissance work. The buildings surround a quadrangle--the Mercato
Vecchio--out of one angle of which rises the immense campanile which I
have already noticed, and which is said to have been erected by Can
Signorio about A.D. 1368, though I should have thought that a rather
earlier date would have tallied better with its style. Besides this the
most striking feature is the external staircase in the courtyard, whose
treatment, of a kind not uncommon in ancient Italian architecture, is
very beautiful, though I fear very weak and unstable, if I may judge by
the number of the iron bars by which it is held together. There are many
windows here of very good detail, and an arcaded cornice all round the
courtyard, and close by and also facing the Piazza dei Signori there is
another fine lofty battlemented tower.

In a street close to the monuments of the Scaligers, whose name I have
forgotten, but in a line with the Viccolo Cavaletto, I found a most
valuable example of domestic work in a very fairly perfect state. As far
as I could make it out, it consisted originally of three sides of a
quadrangle, the fourth side towards the street being enclosed by a wall
and arched gateway. The buildings all had arcades on the ground-level,
forming a kind of cloister, and the staircase to the first floor was
external, and built against the wall on the road side. A great many
alterations have been made in the

[Illustration: 13.--COURT-YARD OF THE PALAZZO SCALIGERI, VERONA. Page
104.]

[Illustration: DOORWAY--OLD HOUSE, VERONA.]

house at various times, but in the sketch which I give I have shewn so
much only of it as appeared to belong to the original foundation. In its
construction pointed and round arches have been used quite
indiscriminately, and in some of the arches the depth of the voussoirs
increases towards the centre of the arch. This is a rather favourite
Italian device, and I was always as much pleased as at the first with
the effect of strength and good proportion which it produces. Most of
the arches are built with alternate voussoirs of brick and stone, but
beyond the outside line of the brick and stone arch there is invariably
a line of very thin bricks laid all round the arch, delicately defining
without pretending to strengthen the main arch, just as a label does
with us. I noticed too, generally, that this thin brick was of a
deeper, better colour than the other bricks, which are seldom any better
than the common English bricks, and are always built with very coarse
joints. This house is finished at the top with the quaint forked or
swallow-tailed Ghibelline battlement, so characteristic of Verona, and
which, as we found afterwards, was in use at Mantua, Cremona, and for
some distance south of Verona, but which I first met with in Verona.

[Illustration: WINDOWS--OLD HOUSE, VERONA.]

[Illustration: BRICK BATTLEMENT--VICCOLO CAVALETTO, VERONA.]

I am not pretending to journalize regularly, but rather

[Illustration: 14.--COURT-YARD OF OLD HOUSE, VERONA. Page 106.]

to note down the remarkable points of the various buildings as they
occur to me, and, before I forget that it was Sunday when I was first
looking at the Veronese churches, I must mention that in the evening we
found our way to the great Piazza di Brà, surrounded by barracks and
public buildings, and containing the vast Roman amphitheatre for which
Verona is so celebrated. Its size is prodigious, and, except in the
outer circuit of wall, it is nearly perfect; indeed, it is impossible to
look upon such a vast structure without a great admiration for the men
who ventured to conceive and carry it into execution.

It is difficult now to conceive how the audience could be found who
would fill so vast a space; and certainly the modern efforts in this
direction are mainly serviceable as shewing the immensity of the
theatre. When I was last in Verona a theatre had been erected in the
arena, and a performance was in progress. The audience might have been
tolerably large in an enclosed theatre, but here it seemed to be the
merest handful; and when we stood on the highest attainable part of the
walls, we found ourselves so far from the stage as to be unable to hear
a single word that was said. There is no need to describe here so
well-known a building as this; suffice it to say, that though the detail
of the architecture is poor, the general design and execution of the
structural arrangements, and the magnificence of the whole scheme cannot
fail to strike one with the same wonder that one feels in the presence
of many of these great Roman works; and it is striking indeed, to see
one of them so perfect as to be still capable of use, and really used.

All the Austrian portion of the inhabitants of Verona crowded the Piazza
di Brà on Sunday evening to hear an Austrian military band, and we
enjoyed not a little a stroll among a crowd of uniforms of all shapes,
kinds, and colours. Verona more than most towns, even in Austrian
Lombardy, seems to be sacrificed entirely to Austrian soldiery. It is
quite melancholy to walk along a street of palaces, some of them
converted into old-furniture stores, others going to ruin; and when
suddenly you do come upon a flourishing and smart palace, if you look in
you are sure to see an Austrian sentinel, and find that it is an
officer’s quarters; and equally when you meet a conveyance, if it is
smart and dashing, with good horses and a stylish coachman, it is quite
certain to be occupied by some dignified-looking military man. So, too,
on the Monday evening, when we went to the French opera (a very pretty,
tastefully got-up theatre by the way) there were absolutely none but
Austrians in the house--in the boxes, officers and their wives, in the
pit, subordinate officers and privates. Who can see this immense staff
of foreigners in occupation of a city like Verona without feeling sadly
for the people who live under such a rule, and for the ruler who is
compelled to maintain such a force to keep his subjects in order?[20]

This, however, is a digression, and I must go on to describe the
remaining architectural features of the old city.

On the way to San Zenone Maggiore, which is quite on the extreme western
verge of the city, one passes the Castello Vecchio, a very grand pile of
simple mediæval fortifications erected in the fourteenth century by
Cangrande II. There are several towers and lofty walls, all topped with
the forked Veronese battlement; and connected with it is the magnificent
Ponte di Castel-Vecchio, a great bridge across the rushing Adige, built
entirely of brick, the parapet of the regular Veronese type, and the
piers between the arches rather large and angular, and finishing with
battlements rather above those of the bridge. The main arch is of great
size--it is said to be not less than one hundred and sixty feet--and one
of the most remarkable points in its appearance is, that, instead of
being in the centre, it is on the side of the river next the castle,
while the other two arches, descending rapidly to the north bank of the
river, give the bridge an odd, irregular, and down-hill kind of look.
The architectural features of this bridge are, however, not the only
objects of interest on this spot; for just after passing the castle the
road bends down to the side of the river, and presents an admirable view
of the campanili, steeples, and spires, with the steep hills on the
opposite bank of the stream, and the mountains in the distance, with the
rapid, turgid, white-looking Adige flowing strongly at one’s feet.

A longish walk through squalid suburbs leads us to the open space in
front of the noble basilica of San Zenone; it is a desolate
waste-looking space, and the poor, old, uncared-for church looks now as
though its day was well-nigh past; as if neglect and apathy were all
that men could give now where once they were wont to lavish so much of
their treasure, and love, and art.

The church, as it now stands, seems to have been entirely rebuilt in the
course of the eleventh or twelfth century, and its proportions are so
very grand, and its detail generally so perfect, that I think it may
certainly be regarded as on the whole the noblest example of its class;
indeed, except the very best Gothic work of the best period, I doubt
whether any work of the Middle Ages so much commands respect and
admiration as this Lombard work. There is a breadth and simplicity about
it, and an expression of such deep thought in the arrangement of
materials and in the delicate sculpture, which with a sparing hand is
introduced, that one cannot sufficiently admire the men who planned and
executed it. Beyond this, the constructive science was so excellent and
so careful, that with ordinary care such a church as San Zenone would
seem still likely to last for ages.

The view of the west front is certainly very striking. The whole church
has been singularly little modernized. By its side to the north is a
fine simple red brick tower, I suppose originally belonging to the city
walls; behind and near the east end of the church, but visible here, the
tall and much-arcaded campanile; and on the other side the little church
of San Procolo with a fourteenth-century painting of Our Lord under a
gabled canopy overhanging the doorway. The west front of San Zenone is
simple but dignified. The nave and aisles are finished with cornices
following the flat pitch of the roofs. The walls are divided by many
vertical lines of pilasters which rise from the plinth to the
eaves-cornice. The main part of this front is stone, but a good deal of
marble is used, e.g., the rose window has tracery of red marble enclosed
within an order of white marble; the doorway and sculpture are in white
marble, once much enriched with colour, and the arcaded band all across
the front is of red marble. Add to this that the stone used is all of an
extremely warm yellow colour, and an idea may be formed of the effort
that was made here in the principal front (as often in Italian churches)
to shew that God’s house was the noblest that could be built.

The doorway well deserves a chapter to itself. Its lintel has
illustrations of the labours of the twelve months, its jambs subjects on
the right from the Old Testament, on the left from the New. In front of
this door there are detached shafts standing on monsters and supporting
a low canopy. An inscription on the façade “Salvet in eternum qui
sculpsit ista Guglielmus,” gives the name of the sculptor, the same man
probably who about the middle of the twelfth century sculptured the
western doorway of the cathedral at Modena.

[Illustration: 15. SAN ZENONE. VERONA.]

No less worthy of study are the bronze doors of this doorway. Here, as
is so often the case in mediæval Italian works, we have the names of the
artists employed--Guglielmo and Nicola da Figarola--with the pious
expression of hope that he who sculptured the work might be saved
eternally. The subjects are very rude in their detail; they illustrate
subjects from the Old Testament, and are executed in thin plates of
bronze nailed on to the wooden doors. A row of small windows, one in
each of the many divisions of the front, extends all across near the top
of the porch; whilst above is a large circular window, filled in with
wheel tracery, and treated as was not uncommon as an illustration of the
Wheel of Fortune. Round this window is an inscription explaining its
symbolism.

We went first into the cloister on the north side of the nave. The
arches are very small, and of brick, supported on coupled shafts of red
Veronese marble, which have marble caps and bases, and rest on a dwarf
wall of stone capped with a thin course of marble. The arch bricks are
of a rich red colour, and contrast well with the brickwork of the
ordinary kind above them; they are used without any kind of moulding or
ornament--and yet I doubt whether I have ever seen a more lovely
cloister than this. The arcades on the north and south sides have round
arches; those on the east and west are pointed; and on the north is a
projecting arcade of the same detail, which once formed the lavatory.
The whole of this cloister is in a very sad state of filth, neglected
and unused, and will, I fear, ere long become ruinous.[21]

From the cloister you enter by a side door into the north aisle of the
choir; much better, however, would it always be to enter from the west,
for it is there, when standing at the top of the flight of ten or
twelve steps which leads down from the door to the floor of the nave,
looking down the great length of the church, scanning its singular
perspective of timber roofing, the great height and simplicity of its
walls, and the mysterious view down into the crypt under the choir
through the recently opened arches, that one feels most deeply the great
and religious effect of the church. To an eye used to northern Gothic
there is something very new in such a building. Its shape, its material,
its arrangement, are all unlike what an English eye is used to, but I
cannot say that I paused for an instant in doubt as to whether I might
really admire or not; for I felt at once how very good the work was, not
only in its general effect, but as much in the treatment of the details,
in its colour, and in its arrangement.

The general plan is very simple--a great parallelogram, divided into a
nave of vast width with northern and southern aisles; the aisles
terminated with square east ends, the choir with an apse of five bays,
which is, however, of later date than the rest of the church. The chief
singularity in the design is the division of the piers of the main
arcades into primary and secondary--the first being large heavy piers
supporting great arches spanning the nave and aisles, which are finished
in a line with the top of the walls; and the latter more delicate
circular marble columns of very Classical character, with finely carved
capitals, and looking almost too slight to support the vast height of
clerestory wall which towers up above the arcade which they carry. The
timber roof, or ceiling, is curious; the framing is all concealed, with
the exception of the collar-beams, which connect the points of the
trefoil which forms the internal line of ceiling. This trefoil outline
is all boarded, divided into panels, and painted. The effect of this
great length of panelled roofing, partly concealed by the great arches
which

[Illustration: 16. SAN ZENONE. VERONA. p. 112.]

cross the nave, is certainly fine. The wooden roofs of the aisles, too,
are original, and their beams are painted very much like the Austrian
sentry-boxes, in zigzag lines of black and buff. Much of this painting,
however, did not appear to me to be old.

At about two-thirds of the length of the church between the west door
and the apse it is cut in two, so to speak, by that which perhaps is now
the greatest charm of the interior--the crypt. When I first visited San
Zenone this crypt existed, but its existence was not realized from the
nave. The only access to it was from the aisles, and even here the
arches were partially blocked up. A flight of steps across the whole
east end of the nave led up to the choir and concealed the old entrance
to it. This, it has lately been found, was originally formed by three
open arches from the nave with a flight of steps descending under them
to the level of the crypt, whilst two arches on either side of these
gave access to it from the aisles. This is the old scheme, and the only
possible approach to the raised choir must originally have been that
which has now been restored, viz., two narrow flights of steps against
the side walls, so contrived as not to interfere with or conceal any
part of the sculpture or other decorations on the western face of the
arches to the crypt. The church as now restored yields to few with which
I am acquainted in the solemn effect which is the result of mysterious
light and shade, multiplied vistas of columns and arches, and
picturesque originality of design. It is evidently rather a result of
growth than of first intentions. The crypt--like our own remarkable
example at Wimborne--seems to be an insertion. The columns and piers of
the choir pass on into the crypt, whose piers and vaults are built
against them. There is an obvious difference too in the style of the
church and of the crypt, showing that at least a hundred years must have
elapsed between the erection of the former and the insertion of the
latter. The choir now occupies the three eastern arches of the old
constructional nave, and beyond this has a square bay and a five sided
apse, the two last divisions being groined and decorated with a good
deal of colour. The apse is of the fourteenth century, but all its
windows have been modernized. The crypt follows exactly the dimensions
of the choir, but is divided into no less than nine bays in width, and
six bays in length, exclusive of the apse. The red marble columns which
support its groined roof are all monoliths, delicate in their
proportions, and many of them probably antique. The vaulting is
quadripartite, and there are considerable remains of wall paintings,
which appear to be nearly coëval with the crypt. In the centre of this
crypt is the shrine of San Zeno, half concealed by the gloomy but
effective lighting, and surrounded by a metal railing made of
quatrefoils, buckled or tied together, very much in the same style as
the railings round the Scaligeri monuments. The altar in this crypt is
worth notice on account of its sculptured front. This has in the centre
a small crucifix with SS. Mary and John, and on either side, under
arches, figures of the Evangelists. The face towards the nave of the
seven arches which form the western part of the crypt is decorated with
extreme care and finish. They are admirable examples of the really
polished work of the Italian artists of the end of the twelfth century.
The arches from the north aisle deserve special notice. They are carried
on coupled shafts which, as well as their capital, are of red Veronese
marble, the archivolt being of stone. The section of both shafts is
circular, but one of them is delicately twisted to a spiral curve on its
upward course. The refinement of this treatment of shafts is very
characteristic of the best Italian work, and this coupled shaft, simple
as its treatment is, and common as are the elements of its design, is so
beautiful that it makes a real sunshine in a shady place. Not less are
the arches above it worthy of admiration. The sculpture here, of a
trailing branch of foliage, is very slightly relieved, but its outline
is so graceful, its imitation of nature so close without being merely
realistic, and its fitness for its position so complete, that I think I
have never seen anything in its way more satisfactory, and certainly
never anything really ornamental in the best sense, the elements of
which were more severely simple.[22]

The colour of the whole interior is, to my mind, charming. It was first
of all built in alternate and very irregularly divided courses of brick
and stone. On this warm-coloured ground, one pious man after another
came and painted what seemed to him best--a Madonna, a crucifixion, a
saint, or a group of figures--with not much thought beyond that of
making the particular work in which he was interested tell its own story
well and produce its own effect. So little did he think of other men’s
previous work that the same subjects are not unfrequently repeated. The
result is that the walls were here sober and there gorgeous, but
everywhere coloured, and everywhere more or less interesting. Yet the
materials of which they are built are just those which we see every day
of our lives, and it was the skill of the workman, not the richness of
his materials, which made his work so worthy of our admiration.

Only one portion of the church is decorated upon a regular system; this
is the eastern part of the choir and the apse, which has part of its
walls and its groining very elaborately painted, though with but little
gold; the groining ribs are richly coloured, and on each side of them is
a wide border, generally subdivided into regular geometrical figures,
and the spaces between these borders are painted blue and powdered with
gold stars. In the south aisle is an altar under a baldachin, supported
at the angles by four clustered shafts knotted together in mid height--a
capricious custom of which Italians seem to have been especially fond,
and the only excuse for which, so far as I can see, is that it proves
that all the shafts were cut out of one block, and therefore of more
value than four plain detached shafts cut out of separate blocks could
be. There is perhaps, also, a relief to the mind, after looking at a
long series of similar shafts, to come at last upon some one or two
marked by capricious singularity such as this. Be this, however, as it
may, the eye certainly always feels inclined to admire them, though the
reason is never quite satisfied; and perhaps some better excuse does
exist for their use than I have as yet been able to discover. It is
probable that these columns belonged originally to the baldachin over
the high altar. The canopy which they now support is not old.

The vestry on the north side of the choir is worth a visit, if only for
the sake of its prettily panelled and painted ceiling. Here the panels
are very small, and left in the natural colour of the pine, and the ribs
are decorated with white, red, and black. There is also in this room a
very good fourteenth-century marble cistern and lavatory under an arch
in the wall; and lastly, in the wardrobes, among many things not worth
seeing, a finely embroidered bishop’s mitre of the twelfth century,
wrought in gold on linen. In front is a figure of Our Lord, at the back
one of the Blessed Virgin, with emblems of two Evangelists on each, and
a band below with figures of the Apostles, the stoles also being adorned
with figures.

The arrangement of the choir shows the common Italian plan of stalls
round the apse behind the altar. They are of early Renaissance
character, with some relic of Gothic feeling in the traceries of the
backs and elbows. There is also a good example of a square choir
lectern, with a large base to contain books, and a revolving gabled
desk.

Until the beginning of this century, there were only two altars in this
church, one in the choir the other in the crypt. The verger told me that
his father remembered the other altars being brought from a suppressed
church, and erected here.

The campanile is well seen from the cloister, where it composes finely
with the coursed walls of the church, and the many-shafted arcades of
the cloister. It is a lofty square tower of several stages, with small
pinnacles, and a low circular brick spire crowning it.

I must not leave San Zenone without mentioning the construction of the
exterior, which--with the exception of the west end, which is of stone
and marble--is entirely of red brick and very warm-coloured stone. The
courses of stone are, as a general rule, of about the same height,
whilst those of brick are very varied, some only of one course, others
of four or five. The cornices at the tops of the walls, too, are very
good, supported upon corbel-tables with round arches resting upon
corbels, and much improved in their effect by the judicious introduction
of thin deep-red bricks between the courses of carved stone, which are
thus thrown out forcibly. It is in this use of red brick, and in the
bold and successful way in which brick and stone are shown in the
interior, that this church is so full of instruction to an English eye;
and I could not see such a work without regretting bitterly the insane
prejudice which some people indulge against anything but the cold,
walls, chilling respectability of our English plastered walls, which to
me seems to be fit only for occupation by savages.

Every time I visit this noble church, I leave it with greater regret;
its exceeding grandeur appears to deserve a better fate than the spare
use to which it seems now to be abandoned. To see all these painted or
coloured walls, all these marble piers, and all this vast expanse of
wall and roof waste and desolate, apparently not half used and never
filled with a throng of worshippers, reminds me too strongly of the sad
and similar fate of some of our own English churches not to awaken a
sigh as I look at it. To some men it is a comfort to find that their
neighbours are no better than themselves in these matters, but I confess
that to my mind a great church disused is a subject only for mournful
recollection, just as a noble church much used and filled with crowds of
worshippers is an object for emulation and admiration. Here, in good
truth, I know not where the worshippers are to come from, so decayed and
forlorn is the neighbourhood.

On the way back from San Zenone into the city a small church is
passed--the oratory of the same Saint--where his body is said to have
rested for a time before it was taken to the Basilica. The only
architectural features are of a long subsequent period, a very good
circular window in the west gable, and a doorway with a pointed canopy
supported on shafts above it, under which of old no doubt there was a
painting.

Beside San Zenone I think the only very grand church as yet unmentioned
is that of San Fermo Maggiore--a vast Romanesque basilica without
aisles, but with small transepts, and a chancel and north and south
chancel-aisles opening into the nave by three arches, which exactly
correspond with its vast width; a not very beautiful arrangement, which
we shall meet with again in the church of the Eremitani at Padua, and in
others of the great churches of the preaching orders of monks.

The fabric of the east end, and the eastern half of the nave appears to
be of very early date--I should be disposed to say the end of the
tenth, or beginning of the eleventh century. A lofty crypt is
constructed under the whole of this part, all the columns of which are
square; some of them mere masses of masonry, others slender monoliths.
The mouldings here are rather Roman in character than Lombard. The
groining is all of brick, and very extensive remains of paintings are
still to be seen throughout, the large columns having single figures
painted on them, one on each face. Access to the crypt is obtained (by
the clergy) from the cloister, south of the church, and by the people
through a very spacious staircase entered from the outside by a door
just west of the north porch. So good indeed are the means of access,
that no doubt the crypt was once extensively used by the laity, for whom
these stairs were specially intended. It is now not used at all--just as
is the case with old crypts all over Europe--but then it is fair to say
that the church is no longer served by the Regular Clergy by whom it was
built, and that their conventual buildings were when first I saw them
occupied by Austrian soldiers, and are now still turned, I believe, to
some equally secular use. I think we may fairly assume that in 1313,
when the church was restored and in part reconstructed, the old crypt
was retained partly on account of its associations, and partly because
of its convenience for those who might at first not quite sympathize
with the novel arrangements of the church above, a great unbroken area
built and contrived for the use and convenience of an order of
preachers, and not for receiving a number of altars.

A monument close to the entrance of the crypt is worth notice. It
represents a professor with his pupils sitting each at a desk. The books
have inscriptions on them. The professor’s has “Vita brevis;” a pupil,
“Ars longā;” another, “judicium difficile;” and another, “tēpus
fugit.”

In the interior of the nave we have, as has been observed, a work which
has been so much altered that it is in fact, as we now see it, a work of
the fourteenth century. It is about fifty feet in width, and covered
with a timber roof, constructed so as to form a ceiling of a number of
cusps boarded and panelled on the under side, and tied with iron ties in
place of collar-beams. Two vertical divisions of the panelling are
arcaded, and filled with paintings of saints, and the whole roof is
darkly stained, and richly painted. Beyond this the only very striking
feature is the pulpit, which is corbelled out from the south wall about
midway in its length. It is old and picturesque, and is surmounted by a
delicately carved and lofty canopy--the whole in marble. It is
surrounded by wall paintings of about the same age, and presents a
fairly unchanged example of what such combined works of the painter and
the architect were in the palmy days of the fourteenth century. These
paintings are of much interest. Behind the pulpit are the four doctors
and the four Evangelists seated, the ascent of Elisha in a chariot of
fire, and twelve prophets with scrolls. Under the canopy is the
Crucifixion.

Going to the exterior one finds on the north side two transepts, and
north of the chancel a tower. East of the eastern transept and of the
tower are small apsidal projections of Romanesque character, both of
them in ruins and unused. The masonry here is different from that of the
later work, being of alternate courses of single brick, and of stone.

The exterior of the principal apse is very remarkable, and belongs to
the fourteenth century. Each side has a steep gable with elaborate
cornices, mouldings, and pinnacles, partly of stone and partly of brick.
The gables are built with circular bricks, and there is a cusped
circular window in each gable. Seen from the bridge which crosses the
Adige close to the church, this picturesque east end is one

[Illustration: 17.--S. FERMO MAGGIORE, VERONA. Page 121]

of the most picturesque things in Verona. Unfortunately the campanile
does not equal in importance the church to which it belongs.

The west end will be best understood by the accompanying sketch. It is
constructed entirely in red brick and warm-coloured stone, and I confess
that it impressed me most pleasantly, as having in its four delicate
lancet windows some sort of affinity to our own English work. The north
porch is very fine of its kind, and the jambs of its doorway are
constructed of black, white, and red marble, used alternately. The
arcading against the walls is noticeable as shewing the use of thin
courses of red brick for the purpose of defining the lines of the
stonework.

The monuments on each side of the west door are good simple examples of
a favourite Italian type. They are, as we shall see, of all dates, and
even when developed to a great size, are still corbelled in the same way
out of the walls. In the North of Europe, we have no analogous treatment
of monumental memorials, and this rather enhances their value. The arch
to the monument on the left of the west doorway is painted in a
charmingly simple Giottesque style, and there is a painting also behind
a modern statue in the door-arch, of Our Lord surrounded by angels. In
the arcades above the monuments there were figures of saints painted,
with imitations of mosaic. All these details are worth mentioning in
order to give some idea to those who have not seen Verona, of the extent
to which everywhere the eye is feasted with remains of early art.
Indeed, one feasts uninterruptedly there on all that can delight the eye
in form and in colour!

With the mention of one more church I believe I may bring my notes in
Verona to an end, and this is a small chapel which stands just opposite
the south side of the Duomo, and whose name I could not learn. It is
much like San Pietro Martire in its general arrangement, but remarkable
for the exquisite beauty of its windows, the arrangement of the bricks
and stonework in which is beyond all praise. These windows are
constructed with trefoiled heads of stone, enclosed within an arch of
mixed stone and brick, round whose outer edge runs a band of delicate
terra-cotta ornament. The spandrels of the trefoils are filled in with
refined sculpture, instead of being pierced with the dark eye usually
found in northern Gothic. There is, too, an entire absence of mouldings,
yet, notwithstanding this, the general effect is one of combined
delicacy and richness of no common kind, so much does carefully-arranged
and contrasted colour do for architecture. A third window is entirely of
brick, save the trefoil head of the opening. The side elevation of this
little chapel is very singular in its whole arrangement; there are three
bays divided by pilasters, which finish at the top in an arcaded cornice
of brick; in each of the two western bays is a lancet window, and the
centre bay has in addition a doorway, and a corbelled out monument above
it; in the eastern bay the window looks just like one of those curious
English low side windows, as to the use of which we have had so many
ingenious theories; the east end has no trace of any window, and is
finished with a flat-pitched roof, and a brick corbel-table running up
the pediment. There is no stone used except in the window-heads and
arches.

There are many other churches in Verona on both sides of the river, and
into several of them we went, but without finding any of equal merit to
those which I have already noticed. Santa Eufemia has a fair west front,
of late pointed, and we found one or two good cloisters just like those
mentioned at Brescia. Other churches have fronts built, and interiors
remodelled, by Sanmichele and his successors, in a style which by no
means approved itself to me; others

[Illustration: 18. ITALIAN BRICKWORK:

  1. 2. Windows at Verona.
  3. Cornice S. Ambrogio, Milan.
  4. Cornice Broletto Brescia.
  5. Windows in Broletto, Monza.
  6. Wall Arcade S. Fermo Maggiore.]

there which I did not succeed in reaching,[23] and there is one
dedicated in honour of S. Thomas of Canterbury, which is not, however,
otherwise of any interest; it has a very late Gothic west front, of poor
character.

It is impossible to walk about Verona without meeting at every turn with
windows whose design is similar to those so often seen in Venice, but
the execution and arrangement are generally so inferior here to what
they are there, that I shall defer saying much about them until I am
describing the palaces and ancient buildings of Venice. They are almost
always finished with ogeed trefoils at the top, and are arranged singly,
or in couples or more together, and one above the other, the same in
each story of the house; their mouldings are thin and reedy, and the
carving of their finials, when they have any, is very poor. Examples of
these windows will be seen by most travellers in the rooms of the
Albergo delle Due Torre.

[Illustration: DOMESTIC WINDOW--VERONA.]

The views from the bridges across the Adige are very striking. The main
part of the city is on the right bank, and the river describes nearly a
semi-circle round it. The opposite bank is only partially built over,
and has a largish suburb, upon rather rapidly rising ground; beyond this
the walls of the city are seen with occasional towers, and marked all
the way by their serrated battlements climbing the irregular outline of
the hills in the boldest fashion. Then crossing over to the other side
and turning round, you see the thickly-built city full of towers and
churches rising far above the turmoil of the crowd below into the pure
sky, and, by their number and size, making Verona one of the most
striking old cities I know.

Of course no one goes to Verona without thinking of Romeo and Juliet. I
fear, however, that when I was shown the Casa de’ Cappelletti, a small
inn in a narrow street, and asked to connect it in any way for the
future with the creation of Shakespeare’s brain, my fancy refused to be
sufficiently lively to perform the required feat. The simple fact is
that, real relics not existing, the good people of Verona have wisely
met the demand which Shakespeare has created, and have discovered a tomb
for Juliet, and other reminiscences of the fair Veronese, which I dare
say satisfy very well the majority of travellers.

At Verona, as in the other towns through which we passed in Italy, we
were quite astonished at the number of misshapen dwarfs that we saw; we
could not account for this at first, but I suppose it is because
children, until they can walk, are tied up in rolls of linen so stiffly
as to deprive them of all power of motion. The only wonder is how any of
these unfortunate children ever manage to walk at all.

In the courts of the houses at Verona there are generally wells, with
ingeniously contrived arrangements for enabling the occupants of the
various surrounding houses and balconies to let down their buckets for
water without themselves going down to the wells. There are guide-ropes
to the well from each angle of the courts round which the houses are
usually built, along which the buckets run, suspended by rings and held
by ropes from the balconies, until they reach the iron-work over the
well, and then fall perpendicularly down to the water.

I have visited Verona many times, and each visit seems to me to give
greater pleasure than the last. I fear I have given but a faint idea of
the indescribable charm which it has to all who are fond of early art.
There is little which one can compare with the situation and
surroundings of such cities as Venice and Florence, and yet I suppose
most travellers would agree with me in reckoning the interest of these
three towns as not far from equal, and greater in very many ways than
that of any other Italian cities.

On this first journey we were driven away by bad weather which, when it
sets in, generally continues for several days, and we left, inwardly
resolving that no long time should elapse before we returned--a
resolution which has been abundantly and often fulfilled--and as the
waiter at our hotel honestly told us that we should be very likely to
find fine weather at Padua, whither we were next to journey, we took his
advice, and then, getting into an omnibus contrived to hold thirty
persons, and I should say at least twenty feet long, with four horses
harnessed with long drawn-out traces to increase the already prodigious
length, we were soon at the terminus of the Verona and Venice railway.




CHAPTER VII.

    “‘And whither journeying?--‘To the holy shrine
     Of Saint Antonio in the city of Padua.’”
                    _Rogers._

     Neighbourhood of Verona--Vicenza: Cathedral--San Lorenzo--Santa
     Corona--Pallazzo della Ragione--Gothic Palaces--Palladio’s
     works--Teatro Olympico--Padua: Giotto’s Chapel--The
     Eremitani--Sant’Antonio--The Duomo.


Many of the villages near Verona are remarkable for the remains of
castles of the middle ages. I have never, however, been able to find
time for the examination of any of them, and, judging from hurried views
which I have had of three or four castles south of Verona, I suspect
that they would scarcely repay a long _détour_; they seem generally to
be more remarkable for their general contour, and their quaint forked
battlements, than for any of that delicate detail and appliance to
ordinary wants which it was especially my object to see and study.

The railroad from Verona to Vicenza and Padua is not interesting; the
country is beautiful and luxuriant in detail, but rather tame, flat, and
over-green in the general view. The Veronese mountains, however, are in
view on the north, and as one approaches Vicenza the hills throw out
their spurs into the flat country, covered with vineyards, orchards, and
fruit trees; village follows close upon village, each with its white
church and white campanile, contrasting strangely with the rich colour
above and around, and at last the towers and roofs of Vicenza are
descried on our left.

And is it possible, my readers will exclaim, that you, an architect, can
have dared to pass within sight of Vicenza without making long sojourn
there to drink in the lessons which the works of your great master
Palladio are there to instil! Even so, reader; for in this world there
are unhappily two views of art, two schools of artists--armies of men
fighting against each other; the one numerous, working with the
traditions and rules of their masters in the art, exclusive in their
views, narrow in their practice, and conventional in all their
proceedings, to the most painful forgetfulness of reality either in
construction or in ornament; the other young and earnest, fighting for
truth, small in numbers, disciples of nature, revivers of an art to all
appearance but now all but defunct, yet already rising gloriously above
the traditional rules of three centuries. The one class representing no
new idea, breathing no new thought, faithful to no religious rule; the
other rapidly endeavouring to strike out paths for themselves as yet
untrodden, gathering thoughts from nature, life from an intense desire
for reality and practical character, faithful moreover to a religious
belief, whose propagation will be for ever the great touchstone of their
work. The one class, the disciples of Palladio, journeying towards
Vicenza with a shew of reverence to learn how he built palaces of compo
with cornices of lath and plaster, already in two short centuries
falling to decay, wretched and ruinous! the other stopping long at
Verona, dreaming over the everlasting art of the monuments of the
Scaligers, and of the nave of Sta. Anastasia, still, though five
centuries have passed with all their storms about their heads, fresh and
beautiful as ever, fit objects of veneration for the artist in all ages!

A disciple, therefore, of the last of these two schools, I stayed not
longer at Vicenza than was necessary to satisfy myself of the truth of
the charges against Palladio’s work there, and to note the few, but
interesting, mediæval remains. The situation of the city is beautiful.
Near it to the north are mountain ranges, and where these descend into
the plains there are smaller hills covered everywhere with luxurious
vegetation. The first view of it is also very fine. In front is the old
brick tower at the Porta di Castello, with its deep brick machicoulis
sloping boldly outwards, and finished with a square battlement under a
flat roof. Beyond this are seen the steeples of the city, and highest
among them the Torre dell’ Orologio, a tall slender brick tower in the
Piazza dei Signori, in the centre of the town. The most important Gothic
churches are the Cathedral, San Lorenzo, and Santa Corona. They seem all
to be in very much the same style--one derived no doubt from Venice, as
we shall see when we arrive there. The plans usually are of this kind.
The bays of the nave are square in plan, those of the aisles oblong in
the direction of the length of the church, the choirs apsidal, and the
chapels on each side of them also apsidal, but with an equal number of
sides so that there is an angle in the centre. The transepts are
square-ended, the sacristy at the end of one of them, and the tower at
the side of the chancel aisles. The Cathedral departs from the type of
plan just mentioned, but is, I think, the only exception to the rule. It
has a very wide nave without aisles some 55 or 60 feet in the clear, but
it has been so much repaired and altered that it is now very
uninteresting. Good effect is obtained by the very great elevation of
the altar which is raised above a crypt, the entrance to which is by
flights of steps on each side of the steps which lead up to the choir.
The exterior is mainly of brick save at the west end, which has an
arcade of stone with a doorway in the centre division.

[Illustration: 19. PORTA DI CASTELLO. VICENZA. p. 128.]

This kind of west front is repeated in the more interesting church of
San Lorenzo, which is finished with one vast gable in front of nave and
aisles, and has for its lower stage an arcade of seven divisions, with a
fine pointed doorway occupying the three central arches. This part of
the front is mainly of stone, with enriched members round the arches in
brick, and it is divided from the upper stage by a corbel-table. The
gable is of brick with a large stone circular window in the centre, and
five smaller circular windows following the line of the flat gable which
crowns the whole. I was not much impressed by this design, the only
virtue of which is a certain amount of simplicity and breadth.

The interior of San Lorenzo is lofty and spacious. The nave and aisles
alone measure about 150 feet by 90, and there are spacious transepts,
choir, and chapels. The columns are circular, and have capitals so badly
carved that it is somewhat difficult to say whether they have shells or
tufts of foliage for enrichment. Very small circular windows in the
clerestory, and a groined roof, complete the design. The best portion of
the exterior is, I think, the elevation of the bays of the nave aisles.
Here there are two simple trefoiled lancets with a shallow buttress
between them, and a circular window above in each bay. This design is
refreshingly pure and simple. The steeple is on the east side of the
north transept. It is of six stages in height, the stages being marked
by slightly sunk panels, and corbel-tables under string-courses, which
are formed by a pattern in the bricks, not by any projecting moulding,
so that the straight outline of the steeple is not broken. The result is
not particularly good.

Santa Corona has, like the other churches, a single gable in front of
the nave and aisles, with a western doorway and a large circular window
above. I have a recollection of it as of one of the most ungainly of
fronts. The campanile here is much like that of San Lorenzo, but has a
low octagonal belfry-stage finished with a small circular brick spire.

Santa Corona may well be more visited, for a picture by John Bellini
than for its own merits. This is a picture over one of the altars in the
north aisle representing the Baptism of Our Lord--one so quiet and
beautiful in colour, so dignified and solemn in its design, that it is
impossible to admire it too much. Behind the group of angels who hold
Our Lord’s garments, is a mountain landscape such as one sees from
Vicenza. It is a sublime work. The marble reredos is coeval with the
picture: it is furnished with two low screen walls right and left of the
altar, a common and proper arrangement for its protection when, as here,
the altar stands in the aisle itself and not in a chapel.

I did not find any other church worth noticing, and soon made my way to
the Palazzo della Ragione. The great feature of this building is the
enormous hall, no less than 72 feet wide inside, and covered with a
great arched timber roof boarded on the under side, and divided into
vertical panels by bold ribs painted black and white. This roof is held
together by two tiers of iron ties, and being arched and boarded at the
end as well as at the sides, has somewhat the look of the inverted hull
of a great ship. The effect is imposing, though at the same time it is
somewhat gloomy, owing to the absence of all high light.

This great hall was Gothic inside and out until Palladio cased the front
with open arcades, standing out from the walls and entirely concealing
from below the old windows. These were large single lights with moulded
jambs, but not of very good style. The original staircase remains with
good marble shafted balustrades. The old work here is said to have been
done before the year 1444, the hall having been burnt in 1389. This date
is of some importance, as the walls of the upper portion are faced like
the upper stage

[Illustration: 20. CONTRADA PORTO. VICENZA. p. 131]

of the Ducal Palace at Venice, with marble arranged in a diaper.

The slender and lofty tower of brick which rises at one end of the
building, the two Venetian columns (Vicenza became subject to Venice in
1404), and the Palazzo itself, in spite of its small architectural
merit, combine to make a charming picture, rendered more beautiful when
I saw it, by the animated crowd of peasants who filled the Piazza. The
streets here are very picturesque, rather in spite of Palladio and
Scamozzi than in consequence of what they did. Some of them are arcaded,
and the Gothic houses are still very numerous. They are all, however, of
late date--at least I saw none earlier than about 1350. They are of the
same design as some of the well-known Venetian palaces, only here they
rise out of narrow streets, instead of as they do there from the water.
The usual arrangement is to have on the ground floor a single doorway,
not necessarily central, and on the _piano nobile_ a fine traceried
window with balconies in the centre, single windows also with balconies
near the angles, and intermediate windows of the same design but without
balconies. These are always treated in the same way with small shafts,
and with animals seated on their angles, and are supported on bold
corbels. All the carving that I saw was weak and confused in outline,
and poor in detail, and the capitals are generally too large for the
arch mouldings which rest on them--a common fault in Italian Gothic
work.

I give a view in the Contrada Porto which illustrates two of the best of
these houses. Almost the whole of this street happens to consist of
houses of the same age, and one of them has on one side of its internal
courtyard open arcades on each story, the upper one having its
balustrades remaining between the columns, similar in design to those in
window-balconies.

In their original state most of these houses seem to have been left in
red brick, the windows being of stone, with thin white marble slabs
fitted into the spandrels above the arches. Projecting balls of marble
are often fixed in front of this marble lining. Some houses seem,
however, to have been plastered almost from the first with a view to
painting, and I can hardly say a word against such a plan, with the
recollection of the glowing--in spite of their being faded--tints which
one still sees at Brescia, Genoa, and elsewhere in Italy. But where the
house has architectural features, which are at all good of their kind,
the painter is very apt to ignore them entirely in his work, so that
what was meant to be a good piece of architectural work, becomes in the
end a badly cut-up ground for a painting. Where there is no
architectural detail to be spoilt, any amount of painting may be
lavished on an external wall; and I know few examples which better show
with how much good effect it may be done than the great house of the
Fugger family in Augsburg, which many Italian tourists now-a-days may
see and admire on their way to or from Italy. The other objection to
external painting is its evanescent character; but good colour is
beautiful even in its decay, and I suppose the best artist will paint
what will do most good to his own generation, and trust to his
successors for doing as much for their own times!

One of the most fanciful houses in Vicenza is the Casa Rigafetta, below
the Palazzo Pubblico. The ornaments are not pure, and there is too much
straining to make the most of an opportunity by putting everything
possible into a small space, but still the whole is decidedly pretty.
The balconies here are in plan half a quatrefoil. Near this house is one
with carved angle-shafts, a feature which I do not remember to have seen
in Venice.

Palladio’s works are supposed now to be the glory of Vicenza. I cannot
forgive the artist who did not care to give solidity to his work, and
the power of executing a vast amount of enrichment in the cheapest way,
and with the commonest materials, is about the greatest snare into which
an architect can allow himself to fall. I am well aware that Palladio
was not the inventor of trumpery modes of construction. His admirers
might quote the architects of Pompeii fifteen hundred years before him,
as offenders in the same way, and the curious preservation of their
works as the justification of their offence; but Palladio followed after
men of his own kind and craft who for centuries had studiously
endeavoured to do their work honestly, and he deserves, therefore, all
the hostile criticism of those who object to a revival of bad practices
which in our own day and country have done more real damage to
architecture than anything else that can be named.

One only of Palladio’s works interested me, and this rather as a curious
experiment than as a work of art. This is the Teatro Olimpico, a famous
open-air theatre. There is first of all a semi-circular auditorium open
to the sky, and only remarkable for a mean arrangement of pilasters at
the back. The great object of interest is the stage, on which a
permanent scene has been constructed by Palladio. In order to make this
look much larger than it really is, the streets, palaces, and temples
which are represented are built in perspective. To accomplish this the
stage rises very rapidly, the buildings are squeezed up, and built in
sharp perspective, so that in the end a triumphal arch, which is really
forty feet from the front, looks as if it were four hundred. Should an
actor by any chance so far forget himself as to walk into what looks
like a practicable street, in a minute he would find himself able to
shake hands with the statue on the top of the arch, the illusion would
be entirely destroyed, and the scenery would all look like a collection
of dolls’ houses. As an ingenious deception from one point of view, and
under certain conditions, the scheme is successful, and probably this is
as much as Palladio himself would have claimed for it.

We had now seen all that we cared to see in Vicenza, and gladly found
ourselves again _en route_. There was nothing to see on the road, and we
were not sorry when our engine gave token by its whistle of our approach
to Padua. The omnibus discharged us in a few minutes at the hospitable
doors of the Stella d’Oro, and we were soon out again with the view of
making the most of our time.

Padua, when I first saw it, seemed to me to be a most melancholy city;
grass grew in the streets, the footways were all formed under dark and
dismal arcades, and not only the externals of the half-occupied palaces,
but those even of all the houses, looked squalid, dirty, and miserable;
nor was there any relief when one got into the more open spaces, for the
large piazze on either side of the Palazzo della Ragione, or townhall,
looked as squalid and uncared-for, as dirty and unprepossessing, as they
well could; nor was this universal squalor rendered at all less
remarkable by the fact that Padua rejoices in a caffè, which is said to
surpass any other even in Italy, for its smartness; and the array of
well-dressed gentlemen who frequented it, certainly made the
neighbourhood look more wretched by contrast than it otherwise would.

The Caffè Pedrocchi was however soon passed, and our first object was
the Palazzo della Ragione, whose vast and singular hall, about two
hundred and fifty feet long by ninety feet wide, is one of the greatest
architectural curiosities in the city. Its exterior has been modernized,
so that now it is only remarkable for its long expanse of roof, but the
interior is still in its original state. The access to the Hall, which
in this and other respects much resembles that at Vicenza, is from
external arcades on the first floor, to which four staircases lead from
below. The walls are low, and covered with paintings arranged in arcaded
panels; some of these are said to be by Giotto, and the whole of them, I
believe, were at any rate painted in his time, but have probably been
repaired and retouched extensively long since. The windows are small,
low down in the walls, and admit scarcely more than sufficient light for
the lower part of the hall. The roof shews in section a vast pointed
arch of timber, boarded and divided into panels by a succession of heavy
vertical ribs scarcely at all moulded. The construction is obviously so
weak as, from the very first, to have made the iron ties which hold it
all together absolutely necessary. A curious feature in the design is,
that instead of having gable walls the roof is hipped, and shews
therefore at the end just the same section as at the sides. What little
light finds its way into the dark obscurity of the roof is admitted
through some small dormers high up in its framework. The effect of the
hall is gloomy, and, compared to our own great halls, certainly shows
some lack of knowledge of construction on the part of its architect, and
its bald heaviness makes it absurd to compare it to our own noble
Westminster hall, though their very similar dimensions might naturally
tempt us to do so. It dates from about the beginning of the fourteenth
century, and the story runs that it was designed by a certain Frate
Giovanni, who, travelling in India, saw the roof of a great palace the
construction of which so pleased him that he brought back drawings of it
with him, and erected its fellow here in Padua. How much truth there is
in this tradition I cannot say, but this much seems clear, that in some
way Padua has, if not a very beautiful, at any rate a very remarkable
Sala, and one which is quite unlike any other room in Europe, with the
single exception of the corresponding room at Vicenza, which no doubt
copied from it.

It has been burnt and damaged in one way and another repeatedly since it
was first built, and in the course of the restorations the paintings on
the walls have been excessively damaged, in many parts repainted, and in
some obliterated altogether. The work was commenced at any rate, if not
completed, just at the time that Dante and Giotto were together in this
part of Italy. The walls are all divided into four panels in height by
borders, with painted pilasters for vertical divisions, and the panels
are generally arched and cusped. The paintings include the apostles, the
signs of the zodiac, representations of the months, the planets, and the
constellations. The whole scheme is far too complex to be intelligible
without a key. This fortunately is accessible in the very careful and
complete account of all the subjects of the paintings on the walls which
was prepared in 1858 by Mr. W. Burges, and printed in the ‘Annales
Archéologiques.’ With infinite pains he made out the meaning of the
whole of the figures--no light task, as the walls being divided by
painted borders and arcades into several stages in height and an almost
interminable number in length, spaces are provided for some four hundred
subjects.[24]

At one end of the hall was the chapel of San Prodoscimo, formed I
presume by screens. The judges sat round the hall, forming so many
courts in one room. At the opposite end was a cage or prison, so that
here, under one roof, with walls covered with illustrations, sat all the
courts of Padua, without any of those ingenious divisions and
subdivisions which are now necessary for the administration of the very
smallest sort of justice, and it may be hoped with as much honesty as
there certainly was simplicity.

Now-a-days the hall is quite unused save as a receptacle for lumber, of
which the most remarkable example is the remnant of a gigantic horse
made by Donatello to travel on rollers in some old Paduan pageant.

From the Palazzo della Ragione, we found our way to what must, so long
as it lasts, be the great glory, as it is the chief charm, of
Padua--Giotto’s Chapel, founded in 1303.

[Illustration: ARENA CHAPEL--PADUA.]

This stands in private grounds and on one side of a desolate green walk
which leads up to a private house to which it now forms an appendage.
From the first it was a little private chapel, and in no respect
remarkable for size or costliness of material or design. The plan is a
simple oblong nave with an apsidal chancel, and a sacristy on the north
side, and nothing can be simpler than the exterior. The walls are of
brick, divided into bays by narrow pilasters. The west door is round
arched, as are also the windows. The interior is even more simple; the
whole nave has not a moulding, the walls are continued on into the
semicircular ceiling without any cornice, and all the ornament is added
in colour.

The windows all have a deep splay outside, very simple stone traceries,
and glass fitted to wooden frames placed inside against the stonework.
There seem also to have been shutters outside, for which the hooks still
remain. A sort of penthouse, or perhaps a cloister-roof was carried
along in front of the chapel, but of this nought remains but the corbels
which carried it. There is no more to be said about the exterior than
that it is simple and good of its kind--the kind being very humble.

Let us go inside, and we shall pass a very different verdict. Giotto is
to many of us not only a person singularly gifted in a great age, but in
some sort the embodiment of an idea. The idea is that of an artist,
pure, simple, and direct in his work, who should excel equally in all
the arts, and show--even though his work be an exception to all
rules--the consummate success of such a course. The man was fortunate in
his day and in his friends. Here, where we stand to admire, he painted,
whilst probably Dante looked on. For merely human and artistic interest
there is, therefore, no room which more rightly deserves to be the
object of endless pilgrimages; and there is none in which one will find
the artist more faithful to his calling, more full of recollection and
self-restraint than in this.

I know, therefore, no one building, of such very small size and cost,
which can claim the same degree of interest as this small Chapel of the
Arena. It is, indeed, one of the glories of art that the works of its
great masters cannot diminish in value, or even be competed with by
subsequent masters: when once done, they are done for ever; and so the
Pietà of Giotto, in this little chapel at Padua, is now--as it was when
first painted in the commencement of the fourteenth century, and as it
will continue to be so long as the neglect with which it is now treated
allows it to exist--one of the great paintings of the world, one of
those fountains from which school after school and age after age of
artists may drink instruction and knowledge, and never fail to gain
more, the more they study its many excellences, and its intensity of
feeling and conception.

[Illustration: SIDE WINDOW--ARENA CHAPEL.]

The architectural portion of the interior may be first of all described.
The apse is simply a sanctuary, and the chancel is formed by marble
screens on each side of the nave, leaving a broad entrance-way between
them, and enclosing about one-third of its length. Against the west side
of these screens are altars, each with a small carved marble reredos;
whilst on the east of them are steps leading to the two ambons; that on
the north being a book-rest, carved in marble, and fixed with its face
to the east; that on the south of iron, and turning upon a pivot.
Between these screens and the sanctuary arch are modern stalls on each
side. The sanctuary has seats all round the apse (except in the eastern
bay), each with a delicate white marble canopy. The sacristy is groined,
and has a thirteenth-century press of wood of a design rather curious
than beautiful, but very rich in its detail. In the nave, as I have
already said, walls have neither cornices nor string-courses to break
their even surfaces, and their face is continued on in semicircular
waggon vault. There are six lancet windows on the south, none at all on
the north, and a three-light window very high up in the gable at the
west end above the doorway.

The architectural merit of the building is simply, I think, that it
performs satisfactorily the office of giving ample unbroken surfaces of
wall for paintings.

The arrangement of these is very regular. The vault is divided into two
parts by wide coloured borders, the space between which is painted blue,
powdered with gilt stars, and in each bay there are five small
medallions with figures on a gold ground. The side walls are divided by
borders into three divisions in height; the upper division containing
subjects from the life of the Blessed Virgin; the central, those
illustrative of the life of Our Blessed Lord; whilst those nearest to
the ground are representations of the Virtues and Vices opposed to each
other; the last division tinted only in one colour, the others richly
painted in bright colours upon a field of blue.

The borders which divide the paintings are very well designed, their
patterns being always very clearly defined with white leading lines, and
a line of red on either side always accompanying each line of white. The
paintings themselves are very wonderful; there is an earnestness of
purpose and expression about them such as one rarely meets with; each
subject is treated with a severe conscientiousness, not always
conventionally where a departure from strict rule is for any reason
necessary, but still, generally speaking, in accordance no doubt with
the ancient traditional treatment. This, illuminated as it is by the
thought and love and earnest intensity of feeling which Giotto lavished
on all that he did, makes his work here the most perfect example of a
series of religious paintings that I have ever seen. Of course in such a
large series of subjects there must be great variety of excellence, and
I am content to agree with the rest of the world in awarding the palm
of excellence to the Pietà, in which the expression of intense feeling
in the face of the mourners over the body of Our Lord is certainly
beyond anything of the kind that I know.

The series is very complete, and, beginning with the history of Joachim
before the birth of the Blessed Virgin (the seventh subject), is
continued down through the leading acts of Our Lord’s life to the
descent of the Holy Ghost on the day of Pentecost;[25] whilst the west
wall is occupied by a Last Judgment; and throughout the subjects Our
Lord, the Blessed Virgin, and the apostles are always represented in
vestments of the same colour.

Most of these paintings are in very perfect condition, and the _tout
ensemble_ is nearly as charming as it was when first painted. I was
sorry, however, to notice some of the paintings lined all over by a
recent copyist, and much damage has been done by damp, especially in the
Last Judgment on the west wall. I do not care very much for the painting
on the lower part of the walls. The figures of Virtues and Vices are
very finely designed, but the imitations of marbles and mouldings
painted in perspective were, I hope--being the last work to be
finished--done after Giotto had completed his work.

Close to Giotto’s Chapel stands the great, and to an English eye,
singular church of the Eremitani. It has a very broad nave of immense
length, unbroken by aisles, and roofed with one of the cusped roofs
already noticed at Verona, in which the real construction is (with the
exception of the tie-beams) entirely concealed by boarding on the under
side; this boarding being generally arranged in a succession of large
cusps or curves; the effect here is, I think, very heavy and
unsatisfactory, but we must bear in mind that the span is prodigious,
and the pitch of the roof very flat. The chancel and an aisle on either
side of it open into the east end of the nave with three arches, and
look so small as to be more like mere recesses than important integral
parts of the plan. There are in this church a great many frescoes and
paintings of much interest--among which are some by Mantegna in a chapel
on the south side of the nave which are worthy of careful study as
being, probably beyond almost any other wall paintings which exist, an
evidence of the fact that interest in treatment of subject, drawing and
design of consummate excellence, perspective and decorative colouring of
the walls, may all be included in a fresco without interference with the
wall surface, or indulgence in tricks of chiaroscuro, without which no
painter now seems willing to do his work. Yet if Mantegna cheerfully
accepted such rules for his wall-painting, would it be beneath modern
painters to do the same for theirs, and could they ask for a better
teacher or guide than this consummate artist?

Less interesting to the artist than these works of Mantegna are the
paintings in the apse executed by Guarienti in the middle of the
fourteenth century. Here are figures of the planets: the Sun, Moon,
Venus, Mercury, Mars, Saturn, and Jupiter, each with allegorical
figures, male and female, of the Seven Ages of Man influenced by the
planets.[26]

At the west end of the nave is a great painted Rood, one of those
curiously shaped crosses meant to receive a painting of the Crucifixion
instead of a carved figure, and cut round with quaintly carved and
crisped indentations, of which so many examples remain, though they are
generally found now in picture galleries. This no doubt stood originally
on a Rood beam, where one sees such a Rood represented in one of the
wall paintings in the upper church of S. Francis at Assisi.

When I was last at Padua the west front of this church was being
repaired--a very dangerous and terrible operation in Italy, where, so
far as I have seen, there is less feeling for, or knowledge of, Gothic
architecture than in any other part of Europe. The interior, too, has
been ruined by the way in which the old ceiling has been painted, in
blue and shaded white. In a building whose characteristic feature is a
certain grand simplicity and austerity, it is especially disgusting to
see light and tawdry colouring introduced, seeing how completely out of
harmony with the whole idea of the church it is.

At the west end are some fine monuments of stone and marble boldly
corbelled out from the wall, adorned with good carving of foliage, and
angels looking out from circles.

The east end is less altered than any part of the exterior of the
building. The immense gable of the nave is divided into four parts, the
outer of which have lancet windows, whilst in the centre is an
insignificant apse, which is almost exactly repeated on the east side of
the south transept. The large sacristy on the north side, and a
campanile of not much interest on the same side redeem to some extent
what would otherwise have been a most uninteresting elevation. The
windows here have a wide external splay, semi-circular arches, and stone
trefoil heads inserted rather clumsily under the arches.

What a grand idea it was on the part of the preaching orders to build
these enormous naves for their congregations! Here, when enthusiasm for
preaching was new born and general, a congregation, numbered rather by
thousands than by hundreds, may have gathered round the preacher, all
within sight of him, and, with the aid, probably, of an awning stretched
across the church over the pulpit, all within sound of his voice.

From the Eremitani we found our way with some difficulty through
miserable streets to the church of Sant’ Antonio, probably the most
remarkable architectural work in many respects in this part of Italy.

It seems that about A.D. 1231 it was determined to erect a great church
in honour of S. Antony, the patron of Padua, and Nicola Pisano, then one
of the most eminent men of his day, was sent for to undertake the
work.[27] The view of the exterior which I give, will best serve to shew
in how singular and original a manner he accomplished his work. S.
Mark’s at Venice must have been in his eye when he designed his church,
and the crowd of cupolas which form its roof remind one forcibly of its
most distinguishing feature.

On first sight Sant’Antonio certainly does not prepossess the beholder
in favour of such a bold departure from every-day rules of art. It is
built almost entirely of a light red brick, not much better than the
common London brick in colour--a poor material wherewith to attempt the
construction of a noble church. Stone is used very sparingly in the
voussoirs of the arches and elsewhere. The cupolas are heavy in their
effect, but relieved by that over the intersection of the nave and
transepts, which rises higher than the others, and is certainly striking
in its design and outline, which is, for the main portion of its height,
that of a simple cone.

[Illustration: GROUND PLAN--SANT’ANTONIO, PADUA.]

Round and pointed arches are used indiscriminately, and the walls are
divided everywhere by pilasters, and surmounted by arcaded
corbel-tables, in all these respects giving the building the appearance
of being much earlier than it is. The west front is very peculiar, and
recalls the fronts of the churches in Vicenza which I have described,
and is entirely unlike any of the churches of Pisa, which would hardly
have been the case had it really been designed by Nicola Pisano. One
great flat gable with an arcaded eaves-cornice finishes the whole, and
out of its apex rises a tall polygonal turret, almost as high as the
dome in front of which it stands. The lower part of the west front has a
central entrance of mean character, and on either side two unequal
arches of construction, the walls within which are pierced with
windows. Above these, and just beneath the great pediment-like gable, is
a long arcade of simple pointed arches, behind which are a passage and
three windows opening into the church. This front is a sham front, and
not excusable on account of its grandeur or its beauty. Indeed, had it
followed the outline of the fabric, it would have been neither ungainly
nor heavy, both of which it most assuredly is now.

The interior is striking from its height, but cold in the extreme in
effect; the domes are all whitewashed in the brightest and freshest
manner. The plan gives three domes to the nave, one and an apse of seven
bays to the choir, and one to the transept. The aisles open into the
nave with pointed arches--two to each dome. The choir aisle is continued
all round the choir, and a chapel is thrown out to the east of this,
which is again crowned with a dome.

The north transept contains the chapel of the patron saint, full of
gorgeous ornaments of all kinds, but not very ancient. Opposite it, in
the south transept, is a curious groined chapel, divided from the church
by five pointed and trefoiled arches of yellow marble, resting upon
Classical-looking columns, and all very richly painted and inlaid. Above
the arches are five statues in niches, and the intermediate wall-surface
is inlaid with white and red marble in a regular pattern, such as we
have seen in the pavement of Sta. Anastasia at Verona, with very good
effect.

The cloister on the south side of the church is very large and good, and
some fine arches occur in it, composed of black and yellow marble with
bricks of varied colour introduced. On its east side three open arches,
filled in with a double iron grille, open into what was, no doubt, the
chapter-house. Going from this into a second cloister to the east of it,
no one can fail to be struck by the extreme picturesqueness and novelty
of the view. In the foreground is the simple

[Illustration: 21. SANT’ANTONIO, PADUA. p. 147.]

pointed-arched and open arcade of the cloister; above this rise the
gable of the south transept and the eastern apse with its surrounding
aisle, and two lofty octagonal brick turrets, on each side of the apse,
which look like minarets from Cairo, and combined with the collections
of domes on the roof give a completely Eastern effect to the whole view.
If, however, the detail of this striking building is examined, one can
hardly be satisfied. There is throughout, as there so often is in Italy,
a sad want of skill and neatness in the adjustment of details as
compared to what is common in northern Gothic buildings. This indeed is
a feature of all the works of the Pisani, and gives them the
character--so common and so fatal in modern works--of being to a great
extent the work of assistants and not of the master. Nothing can be much
more clumsy than the provision for the steps leading to the turrets, nor
weaker than the rectangular tracery inserted in the circular window of
the transept. This, however, is a work of the middle of the fourteenth
century, and Nicola Pisano, even if he were the first architect of the
church, would not be responsible for it. I cannot say that I was at all
satisfied either with the internal or external effect of the
church--though it must be confessed that, when seen from a distance,
there is excessive grandeur in the grouping of the multitude of domes,
with the steep cone rising in the centre, and giving point and emphasis
to the whole.[28] The arrangement of the windows and arches round the
apse, for instance, is confused and weak to a degree; and I do not feel
that Nicola Pisano has fairly settled the question of the adaptation of
the dome to pointed buildings by his treatment of the domes here. The
question is still, I think, an open one; and though it may be doubted
whether with our present opportunities it will soon be satisfactorily
answered, I still feel that it would not be difficult to answer it far
more successfully than has been done here.

In the evening we heard some very fine music of ancient character in
Sant’Antonio, after which there was a sermon; and though it was a
week-day, there was a large congregation, very attentive and quiet.

The Duomo is a cold, unattractive church--said, however, to have been
designed by Michael Angelo--and rather bold in the treatment of the
pendentives under its dome. By its side stands a Lombard baptistery, the
interior of which I did not succeed in getting a sight of; and I believe
that I missed some valuable examples of fresco-painting with which its
walls and domed roof are covered.

We wandered about the melancholy streets of Padua, searching in vain for
objects of any interest to our antiquarian eyes. It is true that the
columns and arcades which support the houses are, many of them, ancient;
but they are of a character very common throughout the north of Italy,
and were not sufficiently novel or striking to draw off our attention
from the melancholy and dilapidated look of the houses and shops which
they half concealed and half supported. We saw also one or two old
monuments at the corners of streets--one of them called the tomb of
Antenor--similar in their idea to those which are so frequent in Verona.

The next morning, therefore, saw us making our way to the railway
station for Venice, sad only in leaving Padua that we could not spend
more time there for the study, more quietly and carefully, of the lovely
little Arena Chapel and the paintings of Giotto.




CHAPTER VIII.

    “Break, break, break,
       On thy cold gray stones, O sea!
     But the tender grace of a day that is dead
       Will never come back to me.”
                   _Tennyson._

     Padua and Venice Railway--Venice: Piazza and Church of S.
     Mark--Torcello--The Lagoon--Murano--Sta. Maria dei Frari--SS.
     Giovanni e Paolo--Sta. Maria dell’ Orto--Other Churches--Domestic
     Architecture--Fondaco de’ Turchi--Other Byzantine Palaces--The
     Ducal Palace--Foscari Palace--Ca’ d’Oro--Other Gothic
     Palaces--Balconies--Venetian Architecture--A Festival--Paintings.


Little is to be seen as one leaves Padua to distinguish the city at all
vividly, save the oriental-looking cupolas of Sant’ Antonio, and the
great roof of the Sala della Ragione, looking like the inverted hull of
some great ship, with its convex sides towering up above the otherwise
not remarkable-looking city.

And now our destination was Venice, and anxiously we looked out ever and
anon, impatient, long before the time, to see the tall outline of S.
Mark’s campanile against the horizon. The view was too contracted to be
really beautiful; on each side of the railway rank and luxuriant hedges
of acacia sprouted their tall branches above the carriages, and beyond
them might be seen plantations of maize, flax, and other crops, all
remarkable for their prodigious growth and size, and watered by
countless small canals, and here and there with turgid, muddy-looking
streams rushing on from the mountains to discharge themselves into the
Adriatic, and gliding between great artificial banks which gave them the
appearance rather of large canals than rivers. Nothing breaks the dead
monotony of the scene, for it is misty, and the Tyrolese Alps, which
ought to show their jagged peaks to the north, are invisible. The
stations follow each other in quick succession, and at each stands some
diligence or carriage whose ingrained panoply of white dust, added to
that which rises here and there white and cloud-like in the dry glare of
the Italian sun which beams overhead, makes one feel grateful
that--however unpoetical such a mode of approach may be--a railroad
carries us to Venice. Nor is such a country as this along which one
passes from Padua to Venice without its moral. I suppose there is
nothing more certain than that, ordinarily, the appreciation of high
art, and success in its practice, have never been so marked in countries
whose natural features are lovely as in those in which the devoutest
student of nature’s beauties can see nothing to admire or love. So
Venice, surrounded by waters, and then by a country which would be
entirely tame and uninteresting, were it not for the exquisite distant
views of the Friulan Alps, fell back on her own resources, and provided
herself with that substitute for the loveliness of nature which--alone
of man’s works--the loveliness of beautiful art can be. And perhaps, it
is better for the traveller, that no interest should be felt save in the
end of the journey, where the journey has so brave an end!

At last, however, the broad watery level of the Lagoon is reached;
Venice rises out of the water at a distance of some two miles; and then
across an almost endless bridge the railway takes us into the outskirts
of the city; a confused idea of steeples and domes is all that is
obtained, and one finds oneself going through that most painful of
processes to an excitable man, the usual examination of luggage; at
last, however, we are outside the station, the Grand Canal lies before
us, and we are vehemently urged to get at once into an omnibus-gondola.
But no, this is too absurd a bathos for our first act on entering
Venice, and we step therefore into a private gondola, and, propelled
rapidly and lightly over the still, unruffled water, sink at once into
that dreamy kind of happiness which of all conveyances in the world the
gondola is best calculated to encourage. A short reach of the Grand
Canal is soon passed, and then, with the picturesque cry of warning,
“Ah, stalì!” we turn sharply into a narrow canal to the right, and,
shooting down the tortuous street of water, presently cross the Grand
Canal again midway between the Rialto and the Foscari palace, dive into
another canal still narrower than the former, and at last, after
frequent glimpses of mediæval houses and palaces, find ourselves safely
housed at our hotel close to the Grand Canal, and within two or three
minutes’ walk of the Piazza San Marco.

We stopped but a short time here, so impatient were we to obtain our
first glimpse of the church and palace of S. Mark, to which Venice owes
so much of her fame. We passed along some narrow winding alleys, lined
on either side with open shops, on the counters of which lie exposed for
sale not over dainty-looking edibles, and in whose dim interiors little
light of day seems ever to enter to help the busy workmen who may always
be seen there plying their trade; and then, going under an archway
crowded with busy folk in rather noisy consultation, we found ourselves,
in four or five minutes, standing under the arcade at the upper end of
the great Piazza San Marco. Long lines of regular architecture, arcaded
below, heavy with cornices and elaborated windows above, carry the eye
not unpleasantly down to the lower end of the piazza, at the right-hand
angle of which towers up into the air a vast campanile, simple and
unbroken in its outline, without visible window or any buttress-like
projection, until its upper stage, where it has a very simple open
arcaded belfry, capped with a pyramidal roof; and then across nearly the
whole width of the piazza, and partly concealed by the curiously
irregular position of its campanile, stretches the low singular and
Eastern-looking church of S. Mark. Before it rise the masts from which
of yore hung waving in the wind the banners of the old Venetian state,
and crowds of pigeons, fed here by civic liberality, cover the pavement
with their pleasant fluttering presence. The charm of the west front is
certainly most indescribable; and I confess to feeling a doubt, as I
looked at it, whether it was not more akin to some fairy-like vision,
such as one might see in dreams, than to any real and substantial
erection of stone and mortar; for, indeed, to a mind educated in and
accustomed to the traditions of northern Gothic, there is something so
very _outré_ in the whole idea, something so startling in its novelty,
that it is hard at first to know whether to admire or not. It is far
from imposing in size, but yet, as it is looked at more and more
carefully, it grows much and rapidly on one’s love, and at last imprints
itself on the mind as a real work of art of a very beautiful and unusual
kind, wrought out with an abundance of beautiful detail, but in such a
way as to prevent its being possible that it could ever be absolutely
reproduced or taken as a model.

As you pace down the broad level space of the Piazza, the feeling of the
strangeness of the whole scene increases. There are of course no horses,
and no vehicles of any kind; it is a large square in which all the space
is footpath, and on which, in addition to the many men who pace it
rapidly with busy brow, or idly lounge whilst enjoying the weather and
the place, hundreds of pigeons are constantly fluttering or walking
about in quiet confidence, sure that they will not be molested by any
one. And then, as one draws near to the church, it is easy to understand
rather better than at first in what its real charm lies; this is no
doubt before anything else in its beautiful colour; the whole front is
shafted to a greater extent than almost any building I know, the shafts
all rather heavy, but of marble of the richest kind; the groining of the
seven entrance arches is filled with mosaics, and the walls are
encrusted everywhere with marble. Instead of ordinary gables masking the
roof, the front is finished with great ogee gables most extravagantly
crocketed, and obviously a modern alteration of the original Romanesque
finish; behind these a cluster of fourteenth-century cupolas completes
the view. Of the seven arches which compose the façade, four have
doorways; the outer arches on either side are very narrow, and answer in
width to the kind of cloister which masks the church on the south, west,
and north sides; and the central arch is much wider and loftier than the
others, rising indeed so high as to break through the line of
balustrading which runs across the front just above the other arches.

Within this central arch is a grand doorway, the stilted semicircular
arch of which is of three orders, the central plain, the others covered
with carvings. The piers supporting the main arches have tiers of shafts
in two heights; the lower tier corresponding in height with those of the
doors pierced within the arches, and the others, which are smaller in
diameter, and more numerous than those which support them, rising to the
springing-line of the main arches. The side doorways have very
Eastern-looking arches, the semi-circular line being carried on nearly
to the centre, and then turned up into an ogee. These and some other
portions as, e.g., the windows over two of the doors, and the pinnacles
between the gables, belong to the fourteenth century, whilst the finish
of the gables themselves--great ogee crocketed gables with figures at
the apex of each--is probably of the fifteenth century. All these
archways open into the cloister, or narthex already mentioned. Here,
where the roofs still glow with the most precious early mosaics, and the
walls with marble, either used as inlays or for shafts, one gets a first
and not unworthy hint of the beauty which awaits one in the interior.

And then, on entering the nave, the deep tones of an organ are heard
reverberating through the old building; many people kneel devoutly at
their prayers around us; the hot glare of the sun has gone, and in its
place a cool, quiet, dim light reveals the whole magnificence of the
design. It is quite in vain to describe this in formal architectural
terms. The colour is so magnificent that one troubles oneself but little
about the architecture, and thinks only of gazing upon the expanse of
gold and deep richly-tinted mosaic all harmonized together into one
glorious whole. The mosaics commence throughout the church at the level
of the crown of the main arches dividing the nave from the aisles, and
are continued up the remainder of the wall and into the domes; even the
angles or arrises of the walls and arches are covered with gold mosaic;
so that all architectural lines of moulding are entirely lost, and
nothing but a soft swelling and undulating sea of colour is perceived.
There is nothing violent or garish in all this profuse decoration. The
gold mosaic, used as it always is in early works, set irregularly in,
and surrounded by a white line or joint of plaster, is never
conspicuously bright; the drawing of the figures and subjects may be
criticized, but at any rate it is always direct, simple, and
intelligible, and the colours of the draperies bright and harmonious,
whilst the arcading and wall-lining which fills the lower part of the
walls is all of marble, which has now a rich, warm, but quiet tint,
singularly suitable as a base to the more gorgeous colouring of the
upper part of the walls and domes.

Altogether this is a church beyond most others suggestive of worship.
Other churches sometimes suggest the same feeling either by enormous
size or vast height. At Köln or Milan man feels so small and so
contemptible in comparison with the vastness of his own work that he is
subdued in spite of himself. In countless other great Gothic minsters
the same feeling is produced. But at S. Mark’s it is produced in an
intensified degree, and by a building the scale of which is in every way
small, if not almost insignificant. There is no long vista of arches, no
complicated perspective, and no vast height to awe the beholder, yet the
mystery of colour does for it even more than the mystery of size does
for Köln or Beauvais, Milan, Toledo, or Bourges. It is, therefore,
emphatically a church for worship, one in which even the most careless
treads with hushed footstep and bated breath, and where, in spite of
crowds, an aweful silence seems always to reign supreme save when it is
broken by the religious sound of the services of the church.

The ground plan is no doubt known to most of my readers. It is a typical
example of the Greek or Eastern church as distinguished from the
Romanesque, a cross whose arms are not far from equal, covered by a
series of cupolas, one in the centre, one to each of the three eastern
arms, and two to the western--with aisles to the nave and choir, and a
cloister round the north, west, and south sides of the nave, of which
the two former are the porches, and the latter the baptistery of the
church.[29]

Under the eastern limb of the cross is a crypt, which has in recent
years been opened and drained, and is now always open to inspection.
This is divided under the choir into five aisles in width by a
multitude of small shafts carrying quadripartite vaults, and in the
centre of which just under the choir altar, is the shrine of S. Mark.
Another apse is formed under the south aisle of the choir, and under the
north aisle is a corresponding crypt save that there is no apse to it.
Much modernized as this has been in the course of repair, and entirely
devoid of all colour or decoration as it is, it is still full of
character, and adds largely to the interest of the church.

[Illustration: GROUND PLAN--S. MARK, VENICE.]

If we return to the nave, we shall find that it is not only in general
effect it is so very worthy of admiration; it still retains much of its
old furniture, and in spite of a few modern mosaics, and one or two more
modern altars, is less altered in its general effect since the
fourteenth century than any great church that I have ever seen. The
screen between the nave and choir with the ambons on either side of it
first deserve notice. The screen is mainly a work of A. D. 1394.[30] It
consists of a series of columns carrying a flat lintel or cornice on the
top of which is a row of extremely good statues of the apostles. They
have that grand sweep of the figure which one knows so well in early
fourteenth century work in France, and are free from the somewhat heavy
and clumsy treatment which marks so much of the work of the Pisani. The
screen has been raised on the base of the older Byzantine screen, which
consisted of a simple continuous arcade now nearly hidden by the more
modern steps to the choir. The ambons are probably of the same age as
this older screen; the gospel ambon being of two stages in height, with
a good staircase to it from the choir aisle, that for the epistle being
comparatively low and simple, but still large enough to contain two or
three modern pulpits. The screens to the choir aisles are of the same
sort as the main screen, but are placed one bay to the east of it. They
are all three interesting as showing that a Gothic architect could use
with good effect a common Classic arrangement, and indeed lend fresh
grace to it by the detail of the sculpture and inlaying with which he
adorned it.

Dimly seen from the nave through the Rood-screen, but far more
interesting than even it, is the great baldacchin or canopy over the
altar in the choir. Here we have the simplest form--four columns
carrying round arches and the wall above them finished with a plain
horizontal capping. The arches may be modern; though if they are so,
they are copied from the old, as is evidenced by the painting at the
back of the Pala d’Oro, which shows the placing of the shrine of S. Mark
under a similar baldacchin; but the groining is old, and the alabaster
columns are of extreme interest, being covered all over with most
elaborate sculptures of Scripture subjects. The subjects in the
north-east column give the history of Joachim and Anna, and the birth of
the Blessed Virgin Mary; the north-west has the nativity of Our Lord,
the marriage in Cana, &c.; the south-west subjects from the Passion; and
the south-east the miracles of Our Lord. Few modes of decorating an
altar are altogether so fitting and beautiful as this, and I hope the
day is not far distant when we shall see many of our English altars
standing under canopies of the same sort. St. Paul’s cathedral may well
prepare the way for us in this, by reviving what was usually accepted as
the best kind of reredos by our English church-builders in the
eighteenth century.

Here, too, is a brass eagle so like one of our own, that one might
almost give it credit for coming from an English smith or founder.

Returning to the nave, one finds nothing more worthy of admiration than
another smaller baldachin over an altar between it and the north aisle.
This is hexagonal, carried on shafts with stilted arches and roofed with
a steep roof. Its dimensions render a small altar a necessity--a matter
of common occurrence in old examples. Another reredos and altar in a
chapel at the north end of the north transept, dating from 1430, may
also be noticed. Here the altar is panelled in front and carved with two
angels censing a cross, and low open screens with arcades carried on
shafts are placed a few inches from the ends of the altar. The footpace
is not carried round the altar, so that it can only be approached from
the front.

Of another sort of furniture--monuments of the dead--S. Mark’s has, as
might be expected, a good many examples. The earliest are the probably
Roman sarcophagi,[31] which lie in the outer aisle or cloister right
and left of the entrance; the next, near them,[32] where the sarcophagus
is still retained, but adorned with Christian emblems and sculpture; and
of considerably later date, and much more artistic interest, are the
tombs of the Doge Andrea Dandolo, and of Sant’Isidoro. Here the
sarcophagus is surmounted by a canopy, reverent angels stand on either
side drawing back partially the curtains from the front of the effigy,
and in the centre of the tomb is a bas-relief of the Madonna, and at the
ends the Annunciation, S. Gabriel on one side, the B. V. Mary at the
other. This is the type of monumental memorial on which so much of the
time of Venetian sculptors seems to have been spent. Here, indeed, and
on the very similar figures of the Virgin on so many of the tympana of
doorways throughout the city, we have to study the sculptor’s art from
the time of the Byzantine carvers who wrought the still numerous early
capitals, until the artist of the Ducal Palace came to revive the art
with his original and splendid series of capitals.

But of all the features of this grand church, that which next to the
gorgeous colour of the walls most attracted me was the wild beauty of
the pavement. I know not what other word to use which quite describes
the effect it produces. It is throughout arranged in the patterns common
in most Opus Alexandrinum, but instead of being laid level and even, it
swells up and down as though its surface were the petrified waves of the
sea, on which those who embark in the ship of the church may kneel in
prayer with safety, the undulating surface serving only to remind them
of the stormy sea of life, and of the sea actually washing the walls of
the streets and houses throughout their city. It cannot be supposed that
this undulation is accidental, for had it been the consequence of a
settlement of the ground we should see some marks of it in the crypt and
in the walls, and some tokens of disruption in the pavement itself. And
the corresponding example of Sta. Sofia, Constantinople, where we have
it on record that there was an intentional symbolism in just such a
floor, is conclusive as to the intention of its imitators here.

Of the mosaics with which the church is richly adorned I cannot pretend
to give a complete account. They deserve a volume to themselves. As
regards choice of subjects, it is noticeable that the most prominent
figure is that of Our Lord, who is seated and surrounded by prophets.
Below are the emblems of the four Evangelists, and the four rivers of
Paradise. Whilst again in the west dome He is surrounded by the apostles
and the Evangelists, and everywhere the general scheme is a lesson to
those who now-a-days too often forget the relative importance or the
proper order and arrangement of the divine story in the schemes they
adopt for stained glass and mural decoration. As regards colour, I need
not repeat what I have already said; but it may be observed that
wherever modern mosaics have taken the place of old ones there at once
we see a complete collapse, and a loss of all good effect. This is
mainly owing, beyond doubt, to the attempt which their designers made to
produce the effect of pictures, instead of thinking first and mainly of
the decorative effect of their work on the building. But at the same
time it is obvious that their eyes had lost all feeling for good colour,
and that in attempting to draw with a certain amount of academical
accuracy, they had equally lost all sense of the prime necessity in such
works of simplicity of arrangement, and directness in the telling of
their story. There is no part of the church in which some of the best of
this sort of decoration can be studied with more ease and advantage than
in the cloister on the north side of the nave. Here the mosaics are so
near the eye, and the details of design and colour so fine that one is
never tired of admiring them.

I never leave S. Mark’s without taking one look at least at the four
bronze horses, which, placed as they are on columns high above the
ground, add so much to the strange character of the west front, and are
in themselves such exquisite examples of their kind. Strange ornaments
these for the façade of the chief church of a city where horses’ feet
have hardly ever trod! Equally strange, if you are to have horses in
such a position at all, is the way in which these are supported. They
stand balancing themselves nicely on the caps of small columns. Extremes
meet; and I am not so sure but that this extraordinary arrangement is
not better than that which is usually adopted. If horses are to be
supported above the ground, they may almost as well be so in this way as
on the ordinary pedestal, which looks equally unsafe if the bronze is
instinct with life. These horses were brought from Constantinople after
the fourth Crusade, circa 1203. They are of admirable character, and are
probably of Greek workmanship. With every other moveable thing worth
moving, they were taken to Paris, and returned after the Peace in 1815.

There is a picture in the Accademia by Gentile Bellini, which ought to
be looked at after a visit to S. Mark’s. In it we see the church much as
it is at present; but an enormous procession which winds its tortuous
way about the piazza, defiles before houses every one of which seems to
be ancient, and I never look at the now uninteresting lines of houses
which surround it without wishing for the resuscitation of the buildings
which G. Bellini saw and drew.

We went into the treasury to see the treasures and plate belonging to
the church, but I was much disappointed to find that, in an artistic
point of view, there was really very little to admire, or else what was
admirable was not shewn. The treasury is a dark room lighted up by a few
wax candles, but so badly that it was difficult to see at all
satisfactorily.

I was unable to obtain a sight of the Pala d’Oro, as the altar-piece
behind the high altar is called; it is only uncovered on feast days, and
I have never happened to be in Venice when it was visible. I was very
anxious to have seen it, as it is a most magnificent piece of
workmanship in gold and enamel. It was executed in Constantinople, and
brought to Venice in 1102. Some Italian writers have claimed it for
their forefathers as an Italian work; but the documentary evidence of
its Eastern origin is supported by the details of the design and
execution of the earliest portions of the work. M. Durand has published
a very careful description of it in the ‘Annales Archéologiques,’ vol.
xx. He gives a list of no less than one hundred and sixty-nine panels or
figures, in a considerable number of which the accompanying inscriptions
are in Greek characters. The Pala was “restored” in the thirteenth
century and again in the fourteenth, when no doubt considerable
additions were made to it. The painting at the back has fourteen
subjects on a gold ground, and is dated 1345.

Over and over again when at Venice must one go into S. Mark’s, not to
criticize but to admire; and if ever in any building in which the main
object is the study of art, assuredly here one must go for worship also.
I think I never saw an interior so thoroughly religious and
religion-inspiring as this, and it is well, therefore, not lightly to
pass it by as useless for our general purposes. It seems to shew, as
strongly as any one example can, how much awefulness and grandeur of
character even a small building may attain to by the lavish expenditure
of art and precious materials throughout its fabric; for it is to this
that S. Mark’s owes its grandeur, and to this only. There is nothing
imposing either in its size or in its architecture; on the contrary,
they appear to me to be both moderate, and the former rather mean; and
yet this grand display of mosaics upon a gold ground makes the building
appear to be both larger and better than it is, and fully atones for all
other defects. Could we but place one of our cold, bare places of
worship by the side of S. Mark’s, and let the development of Christian
art in the construction of the fabric be ten times as great in our
Northern church as in the Venetian, we may yet rest assured that every
religious mind would turn at once to the latter, and scarce deign to
think of the former as a place of worship at all. If this is so, does it
not point most forcibly to the absolute necessity for the introduction
of more colour in the interior of our buildings, either in their
construction, or afterwards by the hand of the painter? And architects
must remember that this ought all to be within their province as
directors or designers, and therefore that they must not, as now,
venture to design cold shells which may or may not afterwards receive
these necessary and indispensable decorations, but from the very first
must view them as part and parcel of the work in which they are
personally concerned; and then, but not till then, shall we see a
satisfactory school of architects in England.

The interest of S. Mark’s is not, however, only religious and artistic;
on other grounds it is certainly one of the buildings most worthy of
study in all Europe. Its architecture is purely Byzantine; and whether
its design was derived from Constantinople or from Alexandria, it
presents us with an almost unique example of the architecture of the
Eastern church transplanted almost without alteration to the domains of
the Western. Nor is this all. It played no small part in modifying the
distinctly Roman influence by which otherwise the whole of Northern
Europe would have been affected. When we see a church so far from S.
Mark’s as that of S. Front at Périgueux modelled after it, and in its
turn influencing a vast number of churches in that and the neighbouring
districts, we may realize what S. Mark’s did towards the development of
Romanesque into new forms and combinations, and may then value properly
every portion of its fabric. Byzantine architecture was the development
of Greek art in the hands of the then vigorous and active Eastern
Church. It is not a direct reproduction therefore of Classic art which
is to be seen in S. Mark’s, but one stage of a development the influence
of which--partly owing to the effect of commerce, partly to her
isolation--was largely felt, down to the very last days of active
Venetian artistic life. This has been well condensed in a short sentence
by Mr. Ruskin. “All European architecture,” he says, “good and bad, old
and new, is derived from Greece through Rome, and coloured and perfected
from the East. The Doric and Corinthian orders are the roots, the one of
all Romanesque buildings--Norman, Lombard, Byzantine; the other of all
Gothic--Early English, French, German, and Tuscan. The old Greeks gave
the shaft, Rome gave the arch. The Arabs pointed and foliated the arch.”
But in the colouring and perfecting the church of S. Mark had the lion’s
share, just as in the ground-plan it is to Venice and the East that we
owe the cruciform arrangement of so many of our buildings, instead of
the basilican form to which we might otherwise have been condemned.

There is another respect in which S. Mark’s is extremely Eastern. This
is in the almost entire absence of figure-sculpture in its original
construction. The subjects and figures on the columns of the baldachin
are too delicate to be noticed from a distance, and it was not until
A.D. 1394 that the choir-screen was introduced, with figures of the
Apostles on either side of the rood, erected no doubt to supply a want
which had been long felt before it was gratified. At the same time
figures were added in niches between the gables of the exterior, but
even now they form a small and inconsiderable part of the decoration of
the church.

I have lingered on paper as I did in reality about S. Mark’s; but if we
wish to see Venice we must tear ourselves away from it. We will go out
by the baptistery, and here we are at once on the Piazzetta, the noble
façade of the Ducal Palace on one side, and a great work of
Sansovino’s--the library of S. Mark--on the other; at the end of the
Piazzetta are two monolithic granite columns, one of which bears the
lion of S. Mark, the other the figure of the ancient patron saint of
Venice, S. Theodore; between them is seen the dark-blue line of the sea
rippled into a thousand twinkling waves, and beyond this the Isola San
Giorgio, remarkable for one of Palladio’s churches--a building, as I
think, irredeemably ugly, but, nevertheless, much admired by many. If
you walk down to the strand, where a hundred gondolas wait for
hire--some black and funereal-like, others dressed up with gay awnings,
and all of them proud and swan-like with their bright steel prows rising
lightly and high out of the water--and then, turning round, look first
down the Riva dei Schiavoni, towards the sea, taking in the long
sea-front of the Ducal Palace, then the narrow gap bridged by the famous
Bridge of Sighs, and on again, noting bridge after bridge, and the
Gothic palace now turned into the Hôtel Danieli, and then on to the
promontory running out towards the Adriatic, occupied by the Public
Gardens and planted with the only trees that Venice boasts--how lovely
is the scene! or if, looking back up the Piazzetta to S. Mark’s, noting
the tall campanile and the quaint clock and clock-tower beyond, and the
domes and turrets, niches and figures, which crown the church, how much
more vividly does it not impress the mind!

Venice is full to excess of striking pictures, and it would be endless
to say in how very many respects it has a character of its own which can
never be forgotten. The strange silence of its watery streets, broken
only by the cry of the gondolier or the delicate plash of his oar in the
water, is not the least impressive thing to the stranger; and when,
after trying in vain to thread on foot the labyrinth of passages which
confuse him irrecoverably in a few minutes, he commits himself to the
dark recesses of a gondola, how delightful is the quiet, smooth, and yet
rapid way in which, without more labour than is necessary in looking
about, he finds himself now following the narrow winding of some small
canal, awakening the echoes between the high walls of palaces or
warehouses on either side of the way, or anon, upon turning with a
graceful sweep into the smooth broad reach of the Grand Canal, making
his gondolier move gently and slowly, as one by one the great palaces
which grace its banks and form its retaining walls are carefully
scanned, whilst the various and ever-changing perspective of the whole
is dwelt upon, to be remembered afterwards with such intense pleasure!

From S. Mark’s I remember trying to find my way to San Stefano; and,
taking a map of Venice, and calculating upon the orientation of the
churches being fairly correct, I flattered myself that I might without
difficulty make my way: the result was simply that for half an hour I
was threading the mazes of all the passages around the church, and at
last reached it only by chance; and I found afterwards that it would not
at all do to take the churches as marking the cardinal points of the
compass, for, as may be seen from the campanile of S. Mark, there are
scarcely two churches in the city exactly alike in their orientation.

But pleasant as it is to recall one’s recollections of highways and
byways in Venice, I think, if we wish to understand her architecture
thoroughly, we shall do well first of all to make a pilgrimage to
Torcello, that sad and weird cathedral, standing forlorn and deserted on
a wretched island in the lagoon, wherein we see the handiwork of the
earliest Venetians, and the prototype of much of the later work in
Venice itself. The story of Torcello has been told often--by no one with
more feeling or more pathos than by Mr. Ruskin, and I need not attempt
to repeat it. Suffice it to say that about A.D. 641 the church was first
built, whilst in 864 it was restored, and being nearly decayed by age,
was again most studiously repaired in 1008;[33] whilst in 1361,
according to the testimony of Cornaro,[34] there were still standing on
the same island divers churches (forty-two in number) adorned with
columns of pietra dura, and with mosaics. Strange indeed is the
difference here between then and now!

My first visit to Torcello was sad and sombre enough to begin with. We
started early from Venice in a thick mist, making our way first to the
cemetery, where there is a good tall campanile, then to Murano, and
through its shabby water streets, passing the end of San Donato (which
shall be described later), and then on through long canals edged on
either side by miles of green mud, and thronged with market boats and
noisy boatmen. Here and there across the lagoon we saw a tall campanile
marking the position of each settlement or island in this waste of sea
and mud, and, last of all, that of the cathedral of Torcello, some five
minutes’ walk among decaying walls and unmown grass from the half-ruined
landing-place to which our gondolier tied his boat. All that remains of
the city is before us in small compass. On the left a fourteenth-century
building, said to have been the Palazzo Publico; in front a stone seat
or throne in the centre of what was once the market-place; on the right,
the Byzantine church of Sta. Fosca; and beyond, and connected with Sta.
Fosca by a cloister, the modernized-looking cathedral, plain, bare, and
uninteresting on the outside, with a detached campanile near the east
end. This is all Torcello has to shew; but, forlorn and decayed as
everything in the place is, it is precious in the highest degree to the
architect who cares about the growth of his art. The cathedral is full
of interest, though much damaged by extensive repairs, carried on in a
reckless mood by the Austrians, not long before they lost Venetia, when
new roofs were put on, and the mosaics were so much damaged that I
remember collecting a handful of fragments from a barrow of rubbish
before it was shot into the canal; at the same time a scaffold was
erected for a proposed restoration of the great western mosaic which,
though threatened and indeed commenced in 1857, had not in 1872 been
proceeded with. The exterior has been completely modernized. In plan the
cathedral consists of three parallel naves of ten bays, all finished
with apses. The columns dividing the nave from the aisles are of veined
marble, with capitals of exquisite workmanship, founded, indeed, on
Corinthian examples, but modified by Byzantine influence and by study of
nature. The arches are stilted and high; above them is a small
clerestory of very simple windows. The central apse retains its raised
rows of seats, though their brickwork alone is left; the throne for the
bishop is placed in the centre, and retains some of its marble inlay.
Under this raised east end a descending passage is formed, connecting
the two smaller apses, and with a small apse formed in the thickness of
the east wall opening into it. Three bays of the church are given to the
choir, which is fenced round with richly sculptured marble screens. On
the west the screen has marble columns carrying a flat entablature, and
below them is a solid portion some four feet high covered with panels of
flat sculpture, one having two peacocks drinking out of the same vase,
another two lions at the foot of a branching tree, and another a
complicated interlacing pattern of foliage. Such a screen is the obvious
prototype of that in S. Mark’s, and the sculptures with which it is
adorned are evidently the work of some early Byzantine workman--whether
brought from Aquileja from

[Illustration: 22. DUOMO. TORCELLO. p. 168.]

the ruins created by Attila’s invasion, or wrought on the spot, I cannot
say, but of a character which we still see in some of the nearly
corresponding screens in the existing cathedral of Aquileja. North-west
of this rood-screen stands the marble ambon--a pulpit of two divisions,
one (circular) facing south, the other (square) facing west. This and
the staircase leading to it are full of delicate and good carved work.
The arrangement has an absurd likeness to many a modern English scheme
of pulpit and reading pew, and there is certainly force in an
observation which Mr. Webb makes,[35] that such an arrangement would
never have been thought of, unless the Gospel was to be understood by
the people. Now they do not understand it, it is no longer said from an
ambon, and ambons seem to be much less useful to Romans than
rood-screens are to us!

The screens north and south of the choir do not seem to be so old as the
other, and are simple low screens.

In the mosaics of the apse and western wall we have, perhaps, the finest
examples of Venetian mosaics. The apse is lined with slabs of veined
marble below, and has above a mosaic of the Blessed Virgin Mary with Our
Lord and the twelve Apostles, and the patron of the church, S.
Heliodorus. The whole west wall is covered with a grand mosaic of the
Crucifixion at the top, the Descent into Hell under it, and a Last
Judgment at the base, which is carried down on each side of the west
door, in the tympanum of which is a half figure of Our Lady. A mosaic at
the end of the south aisle has Our Lord with SS. Michael and Gabriel,
and below them S. Gregory, S. Martin, S. Ambrose, and S. Austin.[36]
Save where these mosaics occur, the walls have been persistently
whitewashed, so that the appearance of the church is now far from
attractive. It is not the less of great interest, as an example of a
very early church founded on a Roman basilica, but with Byzantine
influence most conspicuous in the sculpture of its ornaments. Here, as
in S. Mark’s, the floor is paved with Opus Alexandrinum, of which, in
spite of damage done during the late repairs, it is a fine example.
There is nothing to admire on the exterior, though the large stone
shutters to the windows--single slabs of stone about four inches thick,
working on stone pivots--have a most primitive air. The campanile has
not much to distinguish it from others, but the top affords an
interesting view of the lagoon and the sea with Venice in the distance,
and the Alps of Friuli far away to the north.

A few yards through the cloister bring us to the church of Sta. Fosca.
Here by the side of the Romanesque we have a capital example of a
Byzantine plan, which seems to me to be of the greatest value in
connection with the whole of the round-arched palaces of Venice of which
so many remains still exist. Sta. Fosca is a square church with small
projections on the north, south, and west sides, and a deeper projection
for the altar on the east. There are three eastern apses, and the
western side is screened by an open cloister, which is octagonal in
plan. The square centre is domed on very simple pendentives, and the
capitals are similar in character to those in the cathedral. The best
detail is to be seen outside the east end, where there is some good
arcading and an enriched band of chevron ornament, formed by recessing
the brickwork, and a mixture of red and buff brickwork, which is very
effective.

The last time I was here, I found myself, in the middle of making a
sketch of the west front of Sta. Fosca, suddenly struck by the strange
likeness of its octagonal cloister to the most typical elevations of
the Byzantine palaces in Venice. These always have a centre and wings
divided by piers; and whilst the arches in the centre are of ordinary
proportions, those in the wings are narrow and considerably stilted. In
Sta. Fosca precisely the same effect is produced by the elevation of the
three sides of the octagonal cloister, two of them being reduced in
width and seeming to have narrow stilted arches, owing to their being
canted and not seen in true elevation. I confess I could hardly help
thinking that here I saw the accidental germ of an arrangement which,
commenced in Romanesque or Byzantine buildings, was imitated in many of
the finest of the Gothic palaces, and was revived with invariable
persistency in the Renaissance.

[Illustration: EAST END--STA. FOSCA, TORCELLO.]

The return to Venice was more pleasant than the journey out had been.
The water had risen enough to cover the mud everywhere, and now a vast
expanse of apparent sea was lighted up by the hot sun, and in the
far-off distance the horizon was lined with the long picturesque range
of the Alps, tender and transparent in hue, and sweet reminders to the
dwellers on this monotonous lagoon of the world which lay outside their
boundaries in the far north. On the road we stopped at Mazzorbo, where
there is a dated example of a Gothic doorway. This has a square-headed
opening, and above this an ogee canopy or label over a figure of Our
Lord, and some kneeling figures. The date inscribed on it is A.D. 1368.

Farther on Murano is passed, and a halt made for a visit to the church
of San Donato--once a building of the highest interest and well known to
all readers of Mr. Ruskin’s books. Unfortunately my first visit to this
church was after it had been in part “restored,” in the largest and
worst sense of the word. The old brickwork was being renewed, plastered,
and painted up, till most of its interest had vanished; and now, I fear,
only those who saw San Donato some ten years ago can have any idea of
its architectural value and interest. This was chiefly centred in the
east front, where there is a central apse with a lean-to end to the
aisle on either side. The wall is divided into two stages, by a bold
string-course and double line of chevrons formed by recessing the
brickwork and inserting panels of coloured and carved white marble. The
lower stage is arcaded mainly in red brick, whilst the upper has a wall
deeply recessed behind arcades under the eaves, with delicate
balustrades between the columns which carry the arcades. This upper part
of the building is mainly of buff-coloured bricks, with thin lines of
red to mark the pattern of arches, and it is curious that the light
bricks are much larger than the red.[37] The pavements here are very
fine examples of Opus Alexandrinum, with a more than usual proportion of
black marble, and there is a grand mosaic in the apse, of the B. V. Mary
and Our Lord on a gold ground.

One or two Gothic houses in semi-ruinous condition, and a very fine
fragment of late Byzantine work quite in ruins--the Palazzo da
Mula--remain in Murano, but of these there is such good store in Venice
itself that we may pass them by.

Most visitors, I suppose, go to Murano in order to visit Dr. Salviati’s
glass and mosaic manufactory. He has succeeded in reproducing a material
quite equal to that used in the old mosaics. Still more difficult
feat--he has succeeded also in making glass so like the old Venetian
glass in colour, texture, and design, as to puzzle all ordinary judges.
I cannot sufficiently admire or praise the singular power which Dr.
Salviati has shewn in the education of his men. A party of six or eight
made for me before my eyes, in a few minutes, a tall, delicate, and
richly adorned goblet, in which every part was done by eye and fancy; no
modern accuracy was attempted, and the result was a thoroughly beautiful
and artistic work. All artists know how difficult it is to get a workman
nowadays out of the hard mechanical groove of dull uniformity, and Dr.
Salviati’s success is an encouragement to all of us when we are tempted
to despair of making the attempt.

From Murano a few minutes take us again into the watery streets of
Venice. We have now seen the two buildings which ought first of all to
be studied--Torcello and S. Mark’s, and in them we have the key to
everything that follows. The Venetians commenced in their earliest
buildings with works which shewed but little original invention or
power. It was their fortune to have, by reason of their situation and
their commerce, a great connection with the East. They received,
therefore, a great impetus at the first from Byzantine art. Nowhere in
Europe was so great an influence of the kind exerted; and to us, whose
early architecture was almost entirely Romanesque in its origin, it has
a special interest and novelty. But if the early Venetians copied
Byzantine models, employed Byzantine workmen, and thought rather more of
the beautiful colours for which their Eastern acquaintances gave them a
taste than did their neighbours on the mainland, it must be frankly
conceded that in later times they developed a very original form of
Gothic out of these very materials, and owed comparatively little to any
external aid in their great works of the fourteenth and fifteenth
centuries.

The first business of a tourist in Venice is to secure a gondolier with
intelligence enough to understand his proclivities, and patience enough
to humour them. I have more than once or twice had to thank my gondolier
for shewing me old work which otherwise I should never have seen, and I
am grateful accordingly. I am grateful, too, whenever I think of a
gondola, for the most luxurious machine for sketching from which has
ever been constructed; and the more so, when I recollect how for an hour
at a time I have been persecuted in most Italian towns by all the
idlest, dirtiest, and worst-behaved people of the place whilst I have
made my sketches of their buildings. In a gondola in Venice one knows no
such troubles, and the sketcher’s life, as long as he can work in it, is
as happy and undisturbedly serene as is possible. But no artist must
suppose that everything worth seeing can be seen from the water: a few
walks will convince him that, in the narrow _calli_ as well as on the
water-side, much that is interesting is to be found; and when he has
studied Venice both by boat and by pavement, he will find, as I do, that
the subject is too large for a chapter, and requires rather a volume for
its thorough elucidation.

The buildings of Venice divide themselves into two great classes--the
churches and the civil buildings; of these the former is the smaller
and the less interesting class. But as we have already seen at S. Mark’s
and Torcello the earliest examples of the churches, it will be best to
say all that has to be said about them here, and to take the palaces and
houses by themselves afterwards.

We have seen that S. Mark’s was built in the eleventh and twelfth, and
largely altered in the fourteenth century. Between these two periods
little if anything was done in church-building in Venice; or if it was,
it has disappeared. Just as in Germany, the thirteenth century seems
hardly to have existed for Venice, and we go at a bound from the simple
nervous round-arched work of S. Mark’s to the here somewhat poor and
tasteless churches of the fourteenth century. One or two small
campanili--San Polo, San Samuele, and San Barnaba are the best--remain
to show what the size and character of the earlier work were. They have
plain arcades in the walls, rising from the ground to the belfry, and
this has generally windows of two or three lights carried on shafts. At
San Barnaba[38] a spire with parapets and pinnacles was added to such a
steeple in the fourteenth century, the spire being circular in plan and
built of round-ended bricks. San Paterniano has an hexagonal brick tower
with two light belfry windows, also of Romanesque character. The whole
of these works are of brick, and usually the walls batter outward
towards the base. In this respect, as in the general design, the great
tower of S. Mark’s follows those early examples, as also in its means of
access to the top, which is a continuous slope in the thickness of the
wall in place of the newel staircase in use all over the North of
Europe. Finally, all these older works are very small and modest in
scale and design.

Let us now give up all thought of early works, and see what the
fourteenth and fifteenth centuries did for Venice in the way of
churches. Taking them in their order of merit, we will go first to the
Frari, the church of the Franciscans, thence to SS. Giovanni e Paolo,
that of the Dominicans, to the Madonna dell’ Orto, San Stefano, the
desecrated church of the Convent of La Carità, now forming part of the
Accademia, San Gregorio, Santa Zaccaria, San Giacomo del Rialto, and
some smaller fragments.

I must confess that on the whole, in spite of the grand size of some of
them, I was rather disappointed on first seeing these buildings. One
cannot but be impressed with the magnificent size of such a church as
the Frari, with its many interesting details, and its monuments and
woodwork. But in spite of all this, there is something wanting. I had
not expected larger churches, but I had imagined that their style would
be more pure, and at the same time more unlike what I was accustomed to
elsewhere. The impression they left on my mind was decidedly that they
were very inferior in almost every respect to churches of the same size
and degree of ornament in the North of Europe, whilst in scarcely any
point did they seem to me to have features which could with any
advantage be imitated by us. I had allowed myself to expect a very
different result, and was proportionately disappointed. There is no
church in Venice--(in what I am now saying I mean always to except S.
Mark’s)--comparable either to Sta. Anastasia or to the cathedral at
Verona in the interior; and the exteriors, though fine as examples of
the bold use of brick, are nevertheless not first-rate, nor at all
superior to what one sees elsewhere.

Sta. Maria Gloriosa dei Frari ought first to be described, as being
certainly the finest of its class. The first stone is said to have been
laid on April 3rd, 1250, Nicola Pisano being the architect. The
campanile was begun in 1361

[Illustration: 23.--INTERIOR OF STA. MARIA GLORIOSA DEI FRARI. VENICE.
Page 177.]

under Jacopo Collega, and completed in 1396 by Pietro Paolo his son.

The first impression of the church on landing from the gondola on the
desolate-looking piece of pavement which here, as in many of the
Venetian churches, forms a court between the canal and the west front,
is not pleasing. The design of the west front is nothing short of being
positively ugly; it is finished with a great sham gable, with a curved
outline, somewhat akin to the degraded taste of our worst Jacobæan art,
and entirely without any beauty or even picturesqueness of appearance;
the doorways, too, are particularly poor, consisting of a succession of
twisted and reedy mouldings, thin and shadowless, like so many cords
stretched from cap to base and round the arch, without any proper
distinction of jamb and archivolt.

The internal effect of the church is much finer than its west front
would lead one to expect. The plan is simple; a nave and aisles of six
bays, transepts with three eastern chapels to each, and a choir of one
bay with an apse of four bays projecting beyond the others. The tower is
in the angle between the north transept and the nave, and a large
sacristy with an eastern apse is built against the south transept. The
nave and aisles measure about 230 feet by 104, and the transept 160 feet
by 48,--magnificent dimensions undoubtedly. The columns are simple,
cylindrical, and very lofty, their capitals carved with foliage, which
looks late and poor in its execution, though grouped in the old way in
regular tufts or balls of foliage. The arrangement of the wall above the
main arcade is very similar to that of the Veronese, and, indeed, to
that of most Italian Gothic churches; a plain wall being carried up to
the groining, relieved only by a small clerestory window at the highest
point. One is apt to compare this arrangement with the artistic
arrangement of clerestory and triforium in our own churches; but herein
we do not act quite fairly to Nicola Pisano, who is said to have
designed the Frari, and his brethren. They had to work in a country
where light must be admitted very sparingly, and where therefore it is
impossible for architects to revel in the rich traceries which fill the
bays of the churches of the North; they lived among a nation of
painters, and deemed, perhaps, that these plain surfaces of wall would
one day glow with colour and with Scripture story. For these reasons,
then, I defend them for the bareness and over-great plainness which are
certainly at first felt to be so remarkable in their work. The real
beauty of these interiors is owing, more than to anything else, I
believe, to the simplicity and purity of the quadripartite groining
which covers them in, and which, even where other features would seem to
tell of debasement and absence of pure feeling, invariably recalls us to
a proper recollection of the infinite value of simplicity in this
important feature--a point lost sight of in England after the thirteenth
century, to the incalculable detriment of the beauty of some of our
greatest churches. It is not difficult to prove that this must be the
case, for I take it for granted that we all feel that ornament for its
own sake is valueless; and equally, that doing in a troublesome, and
therefore costly way, that which may be done as well and as strongly in
a simpler manner, is unpleasant and distasteful as an exhibition of the
wasteful expenditure of human skill and energy; and therefore, as simple
quadripartite groining with diagonal and transverse ribs, and no lierne
or intermediate ribs, is quite sufficient for the construction, and as
the vaults are in no degree whatever strengthened by the multiplication
and ramification of perplexing ribs, such as we see in later days in fan
tracery and other contemporary modes of vaulting, that it is the truest
and most agreeable system of roofing in stone.

The simple groining of the Frari is entirely executed in brick, and
springs in the aisles from pilasters corbelled out of the walls midway
in height, just as in Sta. Anastasia at Verona, and in the nave and
choir from clusters of shafts rising from the caps of the columns.

The apse is the noblest feature of the whole church; its windows, with
their singular and not quite pleasing transome of tracery, are
refreshing because they have tracery, though indeed it is of a rude and
heavy kind.

There is something impressive about the arrangement of the church. The
choir is prolonged by the length of about one bay and a half into the
nave, and fenced off to the west by a great screen, surmounted by
figures of the Apostles, with a crucifix rising in the centre. The nave
is, of course, quite free from any fixed seats; and this, with the great
area of the transept and the fine perspective of the long range of seven
apsidal chapels on its east side, gives a grand air of spaciousness to
the whole interior. There are some fine monuments here, quite worth
notice as very characteristic of Italian art. They are generally high
tombs corbelled out from the walls, with arched canopies over them,
inclosing paintings. Here the south transept wall over the door to the
vestry contains a group of such monuments, which is extremely
picturesque. The monument of “Beatus Pacificus” (A.D. 1437) has a
graceful painting of the Annunciation over its arch, and sculptures
under it of the Baptism, and, on the tomb, of the Resurrection and the
Descent into Hell.[39] Another monument has a life-size figure on
horseback, and all have so much freshness to an English eye, and yet so
much identity in principle with our own old monuments, that they are
well worthy of study. Last, but not least, are two immense monuments
facing each other, near the west end of the nave, to Canova and Titian,
preposterous in size, heavy, ugly, and cold in character, quite
unsuitable to a church, and, so far at least as I could judge, entirely
devoid of merit as works of religious art. There is, too, a painting by
Giovanni Bellini of the Madonna and Saints, which ought to be visited,
in the grand and well-used sacristy--a room such as one never seems to
see save in Italy. It is still in its old frame over the sacristy altar.
Both in artistic interest and in religious effect it is perfectly fine;
the subject--a Madonna and Child, such as Gian Bellini alone could
paint. Angels playing instruments, sweet and pretty in character, and
saints full of reverence and awe for Our Lord, all treated with a colour
of exquisite depth and richness throughout, make this as worshipful a
picture as I know. There is also in the north transept a most
elaborately framed Gothic triptych, with figures well drawn and rich in
colour.

The stalls in the Frari are all placed in the nave west of the transept,
as in Westminster Abbey. They are of very rich Renaissance character,
but with some late Gothic features. In the north transept is some
elaborate Gothic panelling--very German in character--which looks as if
it had come from the back of the old choir stalls. Here, too, is a
crucifix, probably the original rood. Some fragments of stained glass
are still visible; they are coarse and rude in detail, but extremely
fine in colour; and one must picture the church full of rich glass in
order to do justice to the scheme of the mediæval architect.

To the south of the nave are large uninteresting cloisters, and it is
only at the east end that the exterior at all repays the ecclesiologist
for the pains he must take to get all round it. The view which I give
will best illustrate its general character. The windows are all
transomed, the tracery and portions of the arches being executed in
stone, the rest of the wall being entirely of brick or terra-cotta with
some

[Illustration: 24.--STA. MARIA GLORIOSA DEI FRARI. VENICE. Page 180.]

red marble in the eaves-arcading; the bricks are not particularly good,
and the terra-cotta borders, cornices, and ornaments are poor and meagre
in their design. The most observable point about the detail is the great
and ugly splay on the exterior of the windows, and the facts that the
window mouldings are returned round the sills and that all the apsidal
terminations in the church finish with an angle in the centre--a
peculiarity which is very seldom met with, but very much to be commended
as a variety.

There is a degree of clumsiness about the way in which the arches of the
windows are set upon the jambs which is very characteristic of Italian
Gothic; but this, and other points open to criticism, do not prevent the
east end of this church from being a very noble conception, broad and
grand, unbroken with the lines of buttresses which generally too much
confuse apsidal terminations, and yet very vertical in its effect. There
is no petty attempt at relieving or ornamenting plain wall where it
occurs, but it is left in the native rudeness of the rather
rough-looking red brick, which is in no respect better than the bricks
one may get anywhere in England. The cornices are very marked, and those
in the clerestory have the common and ungraceful corbelled arcading in
brick, to which I have a special antipathy. The clerestory windows of
the transepts and choir are, I need hardly say, quite modern, and of a
kind, unfortunately, most popular throughout the north of Italy. North
of the choir is a tall brick campanile, leaning rather dangerously to
the North, finished with an octagonal upper stage, and, though not very
remarkable, making a conspicuous feature in most of the views of this
part of Venice, and at any rate to be admired for its simplicity and the
absence of effort in its design.

Next in order of merit to this church are those of SS. Giovanni e Paolo
and of the Madonna dell’ Orto, both of them savouring most strongly of
the influence of the Pisani, and in very many points remarkably like the
church of the Frari.

We will take SS. Giovanni e Paolo first. The plan is of the same sort as
that of the Frari--nave with aisles, and transepts with two chapels
opening on each side of them. These are all apsidal, but planned in the
usual way and not as at the Frari. The east end is a fine composition,
having an apse of seven sides, and is the only part of the exterior to
which much praise can be given. It is divided into two stages by an
elaborate brick cornice and a good balustraded passage in front of the
upper windows. The traceries are all unskilfully designed, and set back
from the face of the wall with a bald plain splay of brickwork round
them; the lower windows here have two transoms, and the upper a single
band of heavy tracery which performs the part of a transom in an
ungainly fashion, though not so badly as in the great south-transept
window in the same church. Here, just as at the Frari, it is obvious
that the absence of buttresses to these many-sided apses is the secret
of the largeness and breadth which mark them; and, to say the truth, not
only are large buttresses to an apse often detrimental to its effect,
but at the same time they are very often not wanted for strength. The
interior is remarkable on account of the fine scale on which it is
built, and for the large number of interesting monuments corbelled out
from its walls. Many of them are mediæval and rich in sculpture of
figures, not only on the tombs themselves, but again in the face of the
wall, around their canopies. The effigy of the deceased is almost always
placed on the top of the high tomb or sarcophagus, which, in order that
it may be visible from below, is made with a slope towards the
spectator, the effect of which is most distressing. Much more beautiful
generally is the curtained tester often put above the figure, on either
side of which guardian angels, holding back the folds of the draperies,
allow us to join them in looking at the figure on the tomb. There is
here a very fine lectern--a double-headed eagle standing on a
scorpion--with a rich mediæval stand and base.

There are small two-light windows just over the arches in the nave which
take the place of a triforium, and which look almost as if they were the
clerestory windows of an earlier church whose arches were much less
lofty than those which now exist.

In the small piazza in front of the church stands one of the glories of
Venice--the monument of Bartolomeo Colleoni. As is the case with too
many equestrian statues the base seems dangerously small for the steed,
slow and stately as his movement is. What a grand air of valiant
determination this old warrior wears! what a serious purpose the artist
had in his work, and how carefully he has rendered every detail of
trapping and armour on both man and horse! We have already heard of this
famous condottiere in his chapel at Bergamo and in his castle of
Malpaga. His statue was the work of Andrea Verocchio, but was completed
by Alessandro Leopardi, between 1479 and 1488. Colleoni had left his
whole fortune to the republic of Venice on condition that his statue
should be placed in the Piazza of S. Mark. This being contrary to the
laws, an ingenious loophole for escape was discovered; the bequest was
secured by the erection of the statue in front of the Scuola di San
Marco, whose strange Renaissance front (built with coloured marble in a
horrible sort of perspective, which is the lowest depth to which
architecture ever reached) stands at right angles to the front of SS.
Giovanni e Paolo. For any one who wished to be remembered near S. Mark’s
church, the catastrophe would be as great, if he cared about art, to
find himself connected, instead, with such an abortion as the Scuola of
that ilk, as it would be to Bartolomeo Colleoni to find himself here in
the suburbs when he stipulated so carefully for a place in the very
centre of the city!

Next in order of merit to SS. Giovanni e Paolo I should place the church
of Sta. Maria dell’ Orto. This church is in a very bad state, and so far
ruinous as to require to be supported in its interior by a forest of
shores and scaffold-poles, which makes it quite impossible to get a good
idea of the general effect. It has fair pointed arcades resting upon
very Classic-looking columns, with capitals of poorly grouped and
executed foliage. It is decidedly inferior to the two churches just
described, in every respect save the treatment of its west front, which,
poor as it is, sins less against all acknowledged rules than do theirs;
its character is of a kind of pseudo-pointed, very flat, hard, and
awkward. The cornice, with the open Italian pinnacles above it, over the
central portion, is better in its effect than the singular row of niches
which stands in lieu of cornice for the ends of the aisles; but it is
worth while, nevertheless, to observe how simple is the design of these
niches, taken separately, and how far this simplicity and the genuine
beauty of their cusping and arching go towards redeeming the want of
taste which is shown in the choice of their location. The doorway and
rose window in the west front are of red and white marble, and in the
side windows the tracery and monials are of white marble, and the jambs
alternately red and white. The rest of the wall is brick, but has been
plastered and washed with pink. The windows at the end of the aisles are
remarkable for transoms of tracery supported upon two heights of
delicate marble shafts, and entirely independent of the glazing which is
fixed in frames within them. This kind of arrangement, incongruous and
unsatisfactory as it is here, is worth recollecting, as being suggestive
of an obvious opening for the use of traceried windows in domestic work;
and it is a plan of most frequent occurrence in the best Italian
ecclesiastical architecture. Many of the windows of Sta. Anastasia at
Verona are constructed in this way, showing on the outside elaborately
cusped and pierced plates of stone, against which on the inside the
glazing is fixed, surrounded only with a plain circle of stone.

San Stefano is another really striking Gothic church. Its interior,
notwithstanding the gaudy red damask with which the Venetians here and
elsewhere delight to clothe the columns of their churches, is very fine
and unlike what is common in the North of Europe. The dimensions are
very large. The nave is about forty-eight feet wide, and the whole
length about one hundred and seventy feet. There are a cloister and a
chapter-house north of the nave, and a campanile detached at some
distance to the east. The arcades of six pointed arches dividing the
nave from either aisle are very light, and supported on delicate marble
columns, whose capitals, with square abaci and foliage of Classical
character, hardly look like Gothic work. The masonry and mouldings of
the arches are not arranged in a succession of orders, as is the case in
almost all good pointed work, but have a broad, plain soffeit, with a
small and shallow moulding at the edge, finished with a dentil or billet
ornament, which, originally used by the architect of S. Mark’s in order
to form the lines of constructional stonework within which his encrusted
marbles were held, was afterwards, down to the very decline of pointed
architecture, used everywhere in Venice--not only in its original
position, but, as at San Stefano, in place of a label round the arch.
Its effect is much like that of the English dog-tooth ornament--a
succession of sharp hard lights and shades, useful as giving value and
force to a very small piece of stonework, and therefore exceedingly
valuable when used as it is at S. Mark’s, and equally contemptible, I am
bound to say, when used, as it is in later work at Venice, simply as an
ornament; for this it is not and cannot be, as it is the result of no
skill or taste on the part of the workman, but just such an enrichment
as might be rather better done by machine than by hand. The roof of the
nave is a painted timber roof, boarded in a series of cusped lines on
the under side of the constructional framework, so as to hide it. I must
not forget to add that the interior of San Stefano requires to be held
together by iron ties in every direction--a sin to which, in Italy, the
eye soon has to become accustomed.

The whole of the exterior is very carefully executed in brick, the
moulded work being well done, though very late in date and not good in
effect. The western doorway is of a favourite Venetian type. It is
square-headed, enriched with mouldings and carving, and above it is an
arched canopy with pinnacles on each side and with an ogee arched label
carrying enormous crockets. The finial is a three-quarter figure, and an
angel occupies the spandrel between the arch and the label. Above the
door is a large circular window, unadorned with tracery or filling-in of
any kind. The window from the east end of the church, of which I give an
engraving, is a very characteristic example, of great width, and utterly
unlike any example out of Venice.

[Illustration: WINDOW--SAN STEFANO.]

Perhaps the very worst traceries in Venice--which is saying a good
deal--are in the windows of the apse here, where the traceried arches of
the head are repeated over the transom, but inverted and standing on
their points. More worthy of admiration is a fine tomb corbelled out
from the cloister wall to Andreas Contarina, “MCCCLVII. Dux creatus

[Illustration: 25. SAN GIACOMO DEL RIALTO. VENICE. p. 187.]

MCCCLXXXII. in cœlumn sublatus;” the arched bridge under the choir
(which is carried over a canal) should also be noted, as well as the
very fine campanile, which, though not boasting of any Gothic detail, is
full of the spirit which made the earlier campanili so effective. But if
we wish to see the best campanile in Venice, I think we must go back to
the Rialto, and there, not far from the Grand Canal, we shall see in
that of San Giacomo a perfectly fine example.[40] It is almost entirely
of brick, and the fine long lines of its arcades give a great effect of
height, whilst the details are all good and quite Gothic in their
character.

The other churches in Venice are of less importance than those which I
have described, but the number of remains, of which only too many are
desecrated, is very large. The Accademia has attached to it the
desecrated church of the convent of La Carità. This has three parallel
aisles ended with apses, the usual traceries and cornices, and the
unusual (I am glad to say) feature of three western gables with arched
outlines[41] filled in with much small tracery in brick and terra-cotta.
Another desecrated church near this--that of San Gregorio--is more
interesting. It is of the same general design as La Carità and, like it,
is built of yellowish bricks. The window traceries are of white marble.
The most interesting feature here is the cloister, entered by a
remarkable doorway from the Grand Canal. The doorway is square-headed,
with an ogee trefoiled archway or window on either side, and a sitting
figure of a bishop under a slight canopy over the doorway. The cloister
has five bays on each side, divided by columns which rest on a marble
and brick base, and carry a wooden framework enriched with very good
mouldings.

Another desecrated church is that “dei Servi,” which has a fine lofty
brick front with a large rose window.

In the Campo Sta. Zaccaria is a portal much like that of San Stefano,
save that it has in the tympanum a good figure of the Blessed Virgin
with Our Lord with a saint on each side--the two Saints John, I think.
The Virgin is seated on a Gothic throne carved in very low relief, and
the whole composition is decidedly fine; comparing it with the doorway
at Mazzorbo, I should say this must be a work of A.D. 1380. The church
of Sta. Zaccaria is an early Renaissance building, with many of its
arches pointed. It has an aisle and chapels round the choir, an unusual
plan in Venice, but otherwise it has no interest.

The church of l’Abbazia has some fair detail in its cornices, with
pinnacles at its west end of the same type as those in the Madonna dell’
Orto, and has poor ogee-headed pointed windows; near it is another of
the canopied doorways--the gate of the Corte Vecchia--with an outer
arched canopy, within which under an ogee-shaped label stands the
Blessed Virgin with Our Lord in an aureole on her breast. Two saints
stand at her side, and groups of little figures kneel at her feet,
whilst from the upper finial Our Lord gives His blessing. This bears the
date of 1505.

I think I have now said enough about these late Gothic churches. I have
never been able to interest myself much about them. The work of which
they are specimens is so exceedingly poor, cold, and distasteful to me,
that I feel much inclined when I attempt to sketch them to give up
ecclesiology in despair. The truth is that, S. Mark’s excepted--and of
course it is a very wonderful exception--the churches in Venice do not
come up to the expectations of any one who has ever experienced the
delight of visiting the churches of much smaller cities in France,
Germany, and England. True, indeed, there are much interest and a great
breadth and dignity about the general effect of such a church as that
of the Frari; but for all those lovely points of detail which in every
direction amaze us by the art they display and the rich array of beauty
with which they clothe the walls of Northern cathedrals, there is here
no kind of equivalent.

When I had thoroughly come to this conclusion, and settled in my own
mind by repeated inspection that my judgment was not harsh or unfair, I
confess I felt a weight off my mind. I was now free to indulge myself to
the full in the search for what Venice really has in greater abundance,
perhaps, than any other city in Christendom--remains, namely, of
mediæval domestic work. Nothing can be conceived more delightful than
such a search. You seldom go a hundred yards--often it is much
less--without coming upon some remains, or perhaps some nearly perfect
example, of an old Venetian palace; and then, with the gondola fastened
to one of the great posts which line all the canals, the well-satisfied
gondolier lying stretched on his back behind the awning, your friends
laughing and talking within its dark recess, you sit most luxuriously,
and make your notes and sketches with a degree of quiet comfort which is
not a little conducive to accurate and careful sketching, and to
diligence in its pursuit.

Venetian palaces divide themselves naturally into two great classes--the
Byzantine and the Gothic; and it surprized me very much to find remains
so perfect and so extensive of the former class even on the banks of the
Grand Canal itself, where change has been ever so frequent and so rife.
Indeed, it is singular that nearly all the Byzantine palaces are
situated on its banks.

Of these palaces, certainly the most striking by far are the Ca’
Loredan, the Ca’ Farsetti, and the Fondaco de’ Turchi. They all agree
singularly in the general idea of their design, and consist of a grand
scheme of arcading over the entire front. Divided generally into two
stories in height, they are again divided in a marked manner in width
into a centre and wings. This division is effected solely by a great
difference in the spans of the arches forming the arcades, which in the
wings are much narrower than in the central division. In the upper
arcade the spaces between the columns, and indeed the whole arrangement,
are often studiously unlike those in the lower range; but, at the same
time, there is so very much similarity in the detail of the whole, that
this variety, far from being perceived as an irregularity or a fault,
does in truth just suffice to give force and vitality to what might
otherwise appear to be monotonous and too often repeated, and recalls to
mind not a little the very similar kind of difference between the upper
and lower order of shafts already described in the west front of S.
Mark’s.

We cannot do better than take, as an example of the finest type of a
Byzantine palace, the magnificent, though now desolate, decaying, and
ruined façade of the Fondaco de’ Turchi, once the palace of the Dukes of
Ferrara. The whole front of this was originally cased with a thin facing
of marble, like the coeval works at S. Mark’s--a kind of decoration
which, neglected as this fine relic has been for years, we cannot be
surprized to find almost altogether destroyed; small fragments do,
however, still here and there remain to tell of the original
magnificence of the work. The lower stage of the Fondaco consists of a
continuous arcade of ten open arches, with three narrower arches at
either end, forming the wings, so to speak; the upper stage has eighteen
arches in the centre, and four in each wing. In the wings the piers
supporting the arches are, I think, all moulded pilasters; in the centre
all the arches rest upon columns; and throughout the whole building the
arches, which are all semicircular, are considerably stilted. The entire
building is constructed in brick, which was originally, as I have
before said, covered all over with a thin veneer of marble; in the
spandrels of all the arches this is relieved by small circular
medallions delicately carved, and over the upper stage is a
string-course, above which there would seem to have been a long series
of slightly sunk panels with round-arched heads, filled in with
delicately arranged and beautifully sculptured patterns in marble. These
panels are immediately below the eaves of the roof. Many of the abaci
and string-courses, and all the thin pieces of marble which form the
soffeits of the arches, have their projections finished either with a
nail-head or dentil moulding, and between the shafts of the upper stage
there are traces of balconies.[42]

A very noticeable point in the general effect of the façade of the
Fondaco de’ Turchi is that, from the peculiar shape and great projection
of the capitals of the shafts and the narrow span of the arches, the
whole of the arcading has, at a small distance, almost the effect of a
series of trefoils, and so seems to pave the way for the continuous
traceries of the Ducal Palace and other later buildings.

There is a ruined fragment of a house of the same age as the Fondaco de’
Turchi in a canal behind the Foscari Palace. Here the centre arch is
very wide, has four stilted arches on the sides, the archivolts are all
delicately carved, and small sculptured medallions are introduced in the
spandrels. Here, I think, the red brick of the walls was always intended
to be seen. Of very similar character is a much larger fragment on the
right of the Grand Canal after passing under the Rialto. Here there are
two stories still remaining. The round-arched doorway has two open and
stilted arches on each side, and then a space of blank wall; and the
upper stage has a group of seven arches in the centre, and a single
arch at each end over the blank wall below. The labels of these upper
arches are turned up at the point into an ogee shape, which, strange as
it may seem, must be original, as the detail is early, and they are
surmounted by a collection of carved medallions and a carved
string-course of early style.[43]

The Ca’ Loredan has two stages in height, above which all is modern--but
all the Byzantine arrangements of these two stages are perfect. There
are five open arches in the centre, and an arcaded pier on each side;
and in the next stage, though the division of centre and wings is
preserved, the arches are increased in number, and consequently the
columns of this stage do not come above those of the lower stage. The
string-courses are formed with a billet mould; the capitals are some of
them genuine Byzantine, and some copies of Corinthian. The wall-faces
were all inlaid, but they were in part altered in the fourteenth
century, when some coats-of-arms and figures were added. Those at the
extreme angles are of David and Goliath, and on each side of the centre
sitting figures of Justice and Force.

Next to the Ca’ Loredan is the Ca’ Farsetti (now the Municipio), which
is, I think, slightly the older building of the two. Here there are
three arches resting on shafts in the centre, and an equal number
resting on piers on each side, and a continuous arcade of fifteen arches
on the upper stage resting on coupled shafts.

The Palazzo Businetto on the Grand Canal opposite the Ca’ Grimani, has
remains of Byzantine work in its two lower stages. Here the caps are
Byzantine in character, the archivolts flat inlays, with a billet mould
on each side, and a carved string-course of running foliage inclosed
between two lines of notched or billet mould.

[Illustration: 26. BYZANTINE WELL. VENICE. p. 193.]

This short notice of some of the more important Romanesque and Byzantine
remains enables me to make a few general deductions: (1.) These
buildings were always of two stages in height. (2.) They had the
entrance in the centre, and had generally a distinction between the
centre and the wings. (3.) The capitals were generally Byzantine in
character, but often copied from Corinthian. (4.) They were of brick,
but generally veneered with thin slabs of marble. (5.) They were
enriched with circular, square, and arched medallions inclosing carving
of foliage and animals, and frequently of coupled birds or animals
regarding each other,--a device always indicative of an early date and
an Eastern origin; and (6.) The string-courses were generally carved
either with continuous running foliage, or with leaves arranged in
threes; the centre turning over, the side leaves extended flatwise, and
upward. This last string-course is exactly copied from Sta. Fosca,
Torcello, and is carved all round S. Mark’s inside; whilst the former,
though it is Byzantine in origin, is carried round the wall of the Ducal
Palace between the south-east angle and the Bridge of Sighs. The
illustration of a Byzantine cistern from the centre of a courtyard which
I give, is useful as shewing very clearly the character of the carved
foliage which adorns the string-courses and panels of these Byzantine
buildings. This is always effectively carved with deep cuttings, which
produce bright and sparkling effects of light and shade.

One especial fault of the Venetians seems to have been their proneness
to repeat the same architectural idea an infinite number of times; and
there is something in this so characteristic of the place and the
people, that the reason for it is worthy of some consideration. Venice,
surrounded by water, and cut off from that kind of emulation which in
other places always has the effect of producing life and change very
rapidly in the phases of art, seems to have contented herself, when once
she had well done, with the conviction that improvement was either
impossible or unnecessary, and so, whilst changes were going on in the
mainland, to have rested satisfied with a slight alteration only, and
that one of detail always, for centuries; and it is thus that I account
for the singular sameness which characterized all the efforts of her
Gothic artists. The façade of the Ducal Palace is really precisely the
same in its idea as that of the Fondaco de’ Turchi or the Ca’ Loredan,
altered only in detail--its very beautiful traceries taking the place
of, but doing the same work as, the simple encrusted arcades of its
predecessors. And again, in the fronts of other and much smaller
palaces--indeed, in all the fronts of the Gothic period--it is singular
how exactly the same idea in the general arrangement is always
preserved. Let me describe an ordinary palace. It is divided into three
or four stories in height, the several stages being generally separated
by string-courses. The lower story opens, by an arched doorway in the
centre, to the water; and on either side of this doorway a few small
windows serve to light the basement. The second stage has a grand window
of some five or six lights, divided by shafts of marble, and rich with
tracery, in the centre; and on either side, one or two single lights,
with tracery corresponding with--and often, as it were, cut out in a
slice from--the traceries of the central window. The third stage is
nearly a reproduction of the second, though sometimes slightly less
important; and the upper stage is either again a repetition of the
others, or else consists of a few small windows placed over the others,
and very unimportant and unpretending. The whole is crowned by a
slightly projecting eaves-cornice, generally very meagre in its
character, and with a line of genuine dog-tooth ornament on its lower
edge. Above this, probably--for only one or two examples remain at all
in their original state--was a parapet like those which still in part
remain on the Ca’ d’Oro, at the back of the Ca’ Foscari, and on the
Ducal Palace, light and fantastic to a degree, and almost masking the
flat roof behind.

Such, as will be seen by the views with which, I doubt not, almost all
my readers must be familiar, is the general idea of the Gothic palaces
in Venice, and it admits of very slight modification. Occasionally, as
in the Ca’ d’Oro, the windows are inclosed within a square line of
delicate moulding, the space within which is encrusted with marble, and
entirely distinct from the string-courses, so as to give very much the
impression of a plain wall veneered here and there with a window; or,
again, sometimes the whole central division of the first and second
stories is veneered on to a façade in which the other windows are
treated constructionally but in all cases from first to last (except, as
we shall see, in the Ducal Palace, and for this exception there is some
explanation in its vast size and other reasons), the distinction between
the centre and the wings was never lost sight of, and never forgotten.
This was the great idea of all these buildings, and most perseveringly
was it reproduced down to the last, when, gradually losing even the life
which beautiful detail had once lent, it sank through successive stages,
until at last, easily and well-nigh imperceptibly, it succumbed, without
a struggle, to the rise of the Renaissance feeling, giving only in
revenge to its successor, the curse of an obligation still to go on
building to the last, for whatever want or on whatever occasion, with
the conviction that a centre and two wings must ever be necessary to a
grand façade. It so happens that, in addition to the large and purely
Byzantine palaces in which this arrangement is preserved,--in a delicate
manner, it is true,--there still remains one remarkable example of the
period of transition from Byzantine to Gothic, in a house which forms
one side of the Corte del Remer (facing the Grand Canal just above the
spot where it is spanned by the Rialto), which serves to shew clearly
the first attempt at translation of this Byzantine idea into Gothic.

In the principal story of this house the central feature is the entrance
doorway, whose finely ornamented arch of markedly horseshoe outline is
very conspicuous. On either side of this, and connected with it in one
group, are two windows divided by shafts and with arches of very
singular shape; it is as though a stilted semicircular arch had been
suddenly turned up in the centre, not with the graceful ogee curve of
later days, but with simple, hard, straight lines. Beyond these windows,
one of later date, but probably inserted in the place of the original
window, completes the similarity which the arrangement of the openings
in this house bears to that common in all the later Gothic palaces. The
arches which support the staircase in front of this house are entirely
executed in brick, and are probably later in date than the house itself,
though it is noticeable that they are of a very early and pure type, and
that here, as generally throughout the North of Italy, the pointed arch
was first used in construction, and then, some time after its first
introduction, and very generally in some modified form, for
ornamentation also.

And now, having so far cleared the way, let me ask my readers to go with
me to the Ducal Palace, and there undertake a somewhat careful
examination of its very famous design.

I shall not enter into a general description of the entire building,
because, as this has undergone prodigious alterations since its first
erection, it is unnecessary to do much more than refer to the two
fronts, which still retain, nearly without alteration, their mediæval
design, and to those portions only of the interior and courtyard which
have not been altered.

[Illustration: 27.--CORTE DEL REMER, VENICE. Page 196.]

The whole building forms three sides of a hollow square: one side rises
out of the deep recesses of the Rio del Palazzo, spanned near its outlet
by the famous Bridge of Sighs, and is entirely of Renaissance work; the
next side, rising from the Riva dei Schiavoni, faces the Giudecca, and
is of the purest Venetian Gothic; and the third, facing the Piazzetta di
San Marco--the small square which connects S. Mark’s with the water--is
also Gothic, and of the same type. The back or north side of the palace
abuts upon S. Mark’s.

I cannot pretend to decide at all absolutely upon the vexed question of
the dates of the mediæval portions, because, as the reader will find in
an interesting discussion on the subject in the second volume of Mr.
Ruskin’s ‘Stones of Venice,’ it is a source of hot disputes. But the
following appear to me to be the main points.

The Ducal Palace was burnt in 1105, restored in 1116, and rebuilt in
1173-1177. The two columns on the Piazzetta were brought to Venice in
1172, about which time the two Piazze were formed. Between this date and
1301, nothing is recorded to have been done to the palace. But even at
this date it was a grand building, and is described in 1275 by Maestro
Martino da Canale as “Grande e bellissimo a maraviglia.”[44] He was
equally enthusiastic about the church of “Monsignore San Marco,” and of
his campanile, “so great and so high that one cannot find its equal.”
Sivos in his chronicle (A.D. 1621) says that the Sala Grande was
commenced in 1301, and completed in 1310, at which date the Grand
Council consisted of nine hundred members; and one Pietro Baseggio is
said to have been the architect between 1309 and 1361; he was succeeded
or assisted by Filippo Calandario, and both of them, according to
Zanotto, were described as being architects, sculptors, and
navigators![45] Calandario had raised himself from the humble post of
shipbuilder at Murano, to that of Capo Maestro of the Ducal Palace, a
man of great weight in the city, but finally finished his career only
too much in accordance with custom, being convicted as one of Marino
Falieri’s fellow-conspirators, and hung from the balcony of the Ducal
Palace in A.D. 1355.

On the 28th of December, 1340, a decree was issued ordering the
construction of a staircase on the east side of the palace to lead to
the new rooms, which seems to establish the fact that at this date a
considerable portion at any rate of the second stage was built. The
plague visited Venice in 1359 and 1361, and stopped all work. In 1362,
because the unfinished work was going to ruin, the Council determined to
complete the new hall; and in 1365, this being done, Guariento of Padua
began to paint it, in the time of the Doge Mario Cornaro.[46]

The capital next the south-west angle of the lower stage bears a date
which appears to some[47] to be 1344 (as to which I have never been able
to satisfy myself), and long afterwards we find the date, 1404, on the
large window of the highest story of the sea-front. Finally, in 1419,
there was a great fire which damaged the old portion of the building, so
much that a decree was passed to rebuild it in conformity with the rest,
and this work was completed in 1423, when the council sat in their great
council chamber for the first time; and in 1439-41 the last Gothic work
was added to the palace by the Doge Foscari, viz. the Porta della Carta,
built (as appears by their contract) by Giovanni and Bartolomeo Bons,
between the years 1438 and 1443, in the small space which intervenes
between the north-west angle of the Ducal Palace and the south side of
S. Mark’s. All these dates are important, and I believe undisputed, the
only question being as to which parts of the building they refer to.

And now, before I say more about dates, let me describe these two Gothic
fronts--the sea-front and the Piazzetta-front--and then we may perhaps
see our way to some sort of comprehension of the relative ages of the
various portions of the fabric.

The whole design is divided into three stages in height, the upper
nearly equal to the united height of the two lower stages, and faced
entirely with a delicate diaper of marble cut in small oblong pieces,
which look, save in their texture and colour, only too much like bricks.
In this marble-faced wall are pierced a number of windows with pointed
arches--the tracery of which has been taken out--and in or near the
centre of each façade a much larger window and balcony, which look as
though they had been subsequently inserted. The lowest stage consists of
a long and uniform arcade of very simple pointed arches resting upon
circular columns with elaborately carved caps; these have been shortened
by some twenty inches of their old height, by the rise of the water, and
the consequent elevation of the pavement of the Riva, to the great
damage of their effect. The intermediate stage is a magnificent arcade,
supporting very vigorous tracery, too well known to everybody to require
much description, and divided from the stages above and below it by
large and pronounced lines of carved and moulded string-courses.

It is important to observe that up to the top of the second
string-course the whole of the architecture is of the very best kind of
Venetian pointed; the arches of the lowest stage are well proportioned,
and, though very simply, still well moulded; and the detail of the
whole of the second stage is, to say the least, not at all inferior.
They form together, without exception, I believe, from all I have either
seen myself or heard, the very best and truest specimen of Gothic
architecture south of the Alps.

Above this noble work the third stage comes, and I confess, to my eye,
with patent marks in every stone of which it is composed, that it was
designed by some other hand than that which had been so successful
below. There is something quite chilling in the great waste of plain
unbroken wall coming above the extreme richness of the arcades which
support it; and moreover, this placing of the richer work below and the
plainer above is so contrary not only to all ordinary canons of
architecture, but just as much to the ordinary practice of the
Venetians, that I feel sure that the impression which I have had from my
first acquaintance with drawings of it is substantially correct, viz.
that the line at which alterations and additions have been made is to be
looked for rather in a horizontal than in a vertical direction; that in
all probability, consequently, the builder of A.D. 1301 commenced with
some portion of the sea-façade and gradually carried on the greater part
of the building to the height of the two stages as we now see them,
leaving his building finished in precisely the same way as the
corresponding halls at Padua and Vicenza--two stories in height, with
arcades covering the outer walls of the upper as well as of the lower
stage; and that when the Council Chamber was found to be too small, and
larger rooms were required, another architect suggested the advantage of
obtaining them by raising an immense story above the others, and,
without destroying much of his predecessor’s work, providing rooms on
the most magnificent scale for the Doge and his Council.

The assumption that the Piazzetta-front has been copied from the
sea-front involves a belief in a veneration for and exact imitation of
older work which is (to say the least) extraordinarily rare, if not
unique, in mediæval works. It involves a belief also in the possibility
of a spirited and successful copy being made of an old capital by a
mediæval sculptor without fresh thought or any fresh invention of any
kind. This will be seen if we examine the capitals of the lower stage of
the palace. Here at first sight one is struck by what appears to be the
astonishing variety of the capitals. They are nearly all adorned with
figures or subjects as well as with foliage, and are certainly in both
fronts of various degrees of merit; but on closer acquaintance it is
perceived that the variety of capitals is not so great as it seems, for
that several of those in each front are merely replicas of those in the
other. If any portion of the two lower stages had been built before the
rest, it would have been the whole of the sea-front and six arches of
the Piazzetta-front, for at the end of these there is a column equal in
size to those at the angles, and which might therefore by possibility
itself have been an angle column for a time. But its larger size may
also fairly be accounted for by the fact that it comes under the side
wall of the large building above, and was in any case therefore a
convenient arrangement if not quite a necessity; but the real difficulty
seems to me to be, that if there were any considerable difference in the
date of these works, all experience would lead one to expect that the
earlier works would be the most uniform and the best, whereas in point
of fact this is far from being the case. For instance, in the sea-front
there are various capitals which are of poor execution. These are,
counting from the south-east angle, the third (large and coarse heads),
the eighth (also coarse heads), the thirteenth (lions’ heads), the
fourteenth (beasts), and the fifteenth, which is certainly not so fine
as the replica in the Piazzetta-front (the twenty-sixth capital,
counting as before).[48]

The case for the contemporaneous erection of these two fronts becomes
even stronger if we ascend to the open gallery on the first floor and
examine the capitals there. They are all similar in general character;
and though they become gradually better as one goes from the south-east
to the north-west, they give the impression of all being of nearly one
date, and moreover they all appear to be later in date than the whole of
those in the lower stage.

On the other hand, there seem to be at least two points which make
strongly in favour of the later date which has been given to the twelve
northern arches of the Piazzetta-front. These are, first, that all the
eight capitals which are replicas are in this northern part of the
front, and none of them in the first six arches from the south-west
angle; and, secondly, that the plate armour in the sculpture above the
capital of the north-east angle (Judgment of Solomon) is later in date
than the period which I assume for this work, and later than the chain
mail shown in the third and eighth capitals of the sea-front. To the
first objection a sufficient answer is that some of the best of the
capitals to these twelve arches are of original design--not replicas;
and in reply to the last objection (which is of much force) the only,
but at the same time obvious reply is, that the Trajan capital, just
under this late armour, is one of the most beautiful of the entire
series, and that on the whole it is much more likely that some of the
sculpture was left in block and finished later, than that no difference
should be made in any of the mouldings or details of the work of two
periods. There is indeed some, though not very strong, evidence that the
sculpture of this capital is not earlier than about 1423. There is,
according to Zanotto, an inscription on it, “╋ Duo Soci
Florentini incise,” and he argues from the use of this last word, that
the two men were the same who made the monument to the Doge Tommaso
Mocenigo in 1423, in which, according to Sansovino, they used the same
unusual term “inciserunt;” this interpretation of the inscription is the
more allowable in that, as I have said, the details of armour in
themselves suggest a date not much earlier than this for the sculpture
connected with it on the angle just above the cap. This date of course
brings this work to very nearly the same period as that of the execution
of the Porta della Carta, and it has been assumed by some that these
capitals were the work of the Bons who built it. Such an assertion is as
wild as that of M. Didron, who says something as to their belonging to
the thirteenth century, for the very shortest inspection of the Porta
della Carta would convince the most sceptical that no part of the
capitals could have been executed by any of the men who wrought at it.
In comparing the merits of the carvers of the earliest and the latest
capitals, it is due to the latter to say that one of the finest of the
whole series is this Trajan capital at the north-west angle, and there
is no internal evidence in it which could lead one to suppose it to be
the work of a man who would ever condescend to copy another’s work.

No one can examine the building without seeing that there is not only in
the detail, but equally in the general design, a marked difference
between the two lower stages and the upper stage. In place of the
extreme boldness which marks every part of the former, we see mouldings
reduced in the latter to the smallest and meanest section possible; the
windows of the upper stage are badly designed, whilst the traceries of
the second stage are as fine as they can possibly be; the angle-shafts
of the upper stage are of the latest type, elaborately twisted and
violently defined, instead of being merely delicate roundings off of the
hard line of wall, as all the early Venetian angle-shafts are; the
parapet, too, is not equal in its design to any of the lower work, and
crowns with an insignificant grotesqueness the noble symmetry of the two
lower arcades; and finally, the chequer-work of marble which forms the
whole of the upper wall is a mode of construction which I have not seen
in any early work, though it is seen in the Porta della Carta (A.D.
1429), and in one other late work, the Palazzo in the Campo Sta. Maria
Mater Domini.[49]

Looking at all the circumstances of the case, I think the fairest
explanation of them is, that the whole sea-front and the six arches of
the Piazzetta (columns 1 to 24 on the plan) were first built, that the
extension to the north (columns 25 to 36 on plan) was then immediately
undertaken by the same artists, and finally that the whole upper story
was built and the sculpture of its capitals completed before 1423. The
capitals of the lower arcade were probably sculptured by degrees, and
certainly not by one hand, between the years 1310 and 1361.

There is a confirmation to some extent of this view in a MS. in the
Bodleian Library of the fourteenth century (the Romance of Alexander),
which contains a curious contemporary view of Venice. This drawing has
been engraved at p. 26 of the second volume of the ‘Domestic
Architecture of the Middle Ages;’[50] here it will be seen, with the
usual amount of licence which characterizes most mediæval

[Illustration: 28--VENICE IN THE FOURTEENTH CENTURY.

From the Romance of Alexander.

Page 204.]

representations of places or towns, that it has nevertheless been
intended as an absolute representation of what its draughtsman had seen.
The columns with S. Theodore and the lion of S. Mark on their capitals,
the bronze horses and the domes of S. Mark’s, the position by the
waterside, and the representation of the Ponte di Paglia, are all proofs
of this; but the important point for my present purpose is, that he drew
the Ducal Palace as _a building of two stories in height_--the first a
simple arcade, the second an arcade with tracery. In the distance behind
this, his drawing shews a picturesque assemblage of buildings, whilst
figures are represented behind the upper arcade as though it were only a
kind of immense balcony. There can be no doubt whatever that this old
drawing tells in favour of the view that the upper stage was not built
until after a considerable interval; for it is almost impossible--looking
at the way in which the rest of the drawing is made--to believe that all
reference to it would have been omitted, had it been in existence at the
time the artist saw it.

It will be seen that my supposition that the original design of the
Ducal Palace was of considerably less elevation than the present
building, would tend to make it very much more like the Byzantine type
than it is; but even now no one can dispute the family likeness. The
amount of constructive art is as nearly as possible the same. The weight
is supported by a succession of shafts placed at very short intervals
from each other, and in neither is there any approach to the system of
pier, arch, and buttress, so distinctive of Gothic art in the North of
Europe. The pointed arch is used, it is true, in the palace; but, after
all, the mere use of the pointed arch does not make thorough pointed
architecture, and therefore, interesting as it is as a variety of the
style, the Ducal Palace is, I think, not properly to be placed in the
first class of Gothic buildings. Indeed, the second stage, whose
exquisite beauty is the charm of the whole building, does not exhibit
the pointed arch at all in a properly developed form, and is strong
enough to support the great weight of wall above, only by reason of the
massy character of its tracery, and not by the proper application of
constructional arches. I have already said that there is no approach to
buttressing; but the angles require some help, and this is given partly
by increasing considerably the size of the shafts, and partly by iron
ties at the springing of the arches running for some distance in each
direction from the angles.

All the mouldings are very simple; they are generally composed of
three-quarter beads, small fillets, and large flat hollows, constantly
arranged in the same order. The label of the main arcade is a plain
bead. In the string-courses boldly-carved flowers are repeated with a
slight interval between each, and the upper string-course has a row of
nail-heads in one of its members. The cusping of the tracery is quite
square in its section, and the cusps finish with a square end, to which
is attached--and with good effect--a small circular ball of red marble.
The parapet is of the somewhat peculiar kind I have already mentioned,
and I confess I have never been long enough in Venice to accustom myself
to, or to admire, its extreme peculiarity of both outline and design.

And now before we leave this subject let me offer a remark, as every one
who writes on it must, on the admirable story of these sculptures. I
have never sat in front of one of them for any space of time without
seeing some wayfarer stop to study the story of some one of the
capitals. They are a book at which more thousands have looked with
pleasure for some five hundred years than at any other single book in
the world, with the one exception of the Bible. And the lesson to
architects is obvious. Concentrate your labour and your story on some
one part of your building where all men may read it; tell some simple
story and you will interest your readers, if you will but tell it so
simply that by good chance they may be able to read it. Lay out a scheme
so well that if you die your successor may carry it on. Here, as I
believe, the architect completed his two stages of arcades, whilst the
sculptor was changed, but kept generally to the scheme of subjects first
of all laid down. At the three exposed angles are the three archangels,
below them the moral lesson--as much wanted now as then--of the
Drunkenness of Noah, and the Story of Tobit (with S. Raphael); at
another angle The Fall (under S. Michael); and at the last the Judgment
of Solomon (under S. Gabriel). In the lower range of capitals the
stories and catalogues of virtues and vices; the illustrations of
fruits, animals, every-day life; the labours of the months, trades,
sciences, and arts--are all illustrated, and complete a cycle of
subjects which, ill-treated, would always have a certain value, and
which, well treated as most of them are here, have the very highest
charm.

For a building which owes its general impressiveness entirely to the
uniform character of its architecture, it is especially fortunate that
there should be so much also in the detail to attract and reward
constant and minute examination. It is for this reason that the range of
great capitals to the columns of the lower arcade is of so much
importance. They are so large, so close to the eye, so interesting in
their story, and on the whole so carefully and artistically executed, as
to afford the greater pleasure the more the building is known. The
key-plan to these capitals which I give[51] will be useful to shew what
the general arrangement of the subjects is. I have already shewn that
there are repetitions of many of the subjects, but it is equally worth
notice that the foliage which forms the framework for the subjects is
also repeated. There are, I think, only four varieties in its
arrangement. In the first the capitals are arranged very simply--in some
cases rudely--with tufts of foliage or heads. The capitals numbered 2,
3, 6, 13, 16, 20, 23, 27, and 34, are examples of this. In the next the
foliage of the lower part grows up vertically, bending slightly out to
support the sculptured subjects. These are generally the most graceful
of all, and infinitely richer in effect than the first class. The
capitals numbered 1, 7, 9, 12, 18, 24, 26, 28, 33, and 36, are examples
of these. In the third class the foliage is generally marked by the same
feeling, but it rises vertically to the angles, and curls over under the
subject; the 19th and 25th capitals are examples of this class. In the
fourth class the foliage curves over downwards, both at the angles and
under the subject. The neckings below the capitals are wrought on the
shaft itself. They are sometimes moulded, sometimes corded, and
sometimes delicately carved with foliage; these last are by very much
the more beautiful, and generally accompany the best wrought of the
capitals, whilst the inferior capitals have, in all cases, the plainer
necking.

The capitals in the upper arcade have not so much story as those below.
They have generally a head on each side in the midst of foliage, and are
square in plan, though the lower caps are octagonal--a few only have
their names written over them; but on the ground-story most of the
capitals have, or have had, explanatory inscriptions. Some of the upper
capitals close to the north-west angle are among the best. The curves of
the foliage in the angles of the capitals are admirably wrought, and may
be compared, to the damage of the latter, with some of the lower
capitals in the sea-front. The upper range of capitals gradually
deteriorates from the north-west angle as you go to the south-east.
These last are really very bad, having rude gross carving of the human
figure, and foliage feebly massed and treated; but the upper capital of
the south-west angle with the figures of the four winds, and the two or
three capitals near it, must be excepted from this remark, being superb
in design and execution.

The remains of original work in the quadrangle are much less important.
The arcade on the first floor remains, but none of its details are good,
and on the east side it is a poor Renaissance copy of the other sides.
The whole of the lower arcade has been destroyed or altered. But in the
upper walls, which are faced with brick, some of the original windows
remain; they are small, but of the same sort of detail and character as
the larger windows in the outer walls.

The building has lost much by the gradual raising of the pavement. This
is now about twenty inches above the old base of the columns, and their
proportions are so far altered for the worse. And it has lost immensely
also by the destruction of the inlaid marble which once filled all the
spandrels of the main arcade. Two panels only of these remain, and both
in the sea-front. They are charmingly designed, inclosing circles which
exactly touch the labels and strings.

Of the modern additions to this grand building I shall not say anything.
They are not beautiful in themselves nor interesting by reason of their
decorations, if I except those walls on which Tintoretto has lavished so
much of his skill. The architects of the fourteenth and fifteenth
centuries were artists in very deed, and it is with their work only that
I can feel any real sympathy.

Such, then, is the Ducal Palace: a building certainly in some respects
of almost unequalled beauty, but at the same time of unequal merit; its
first and second stages quite perfect in their bold nervous character,
and in the almost interminable succession of the same beautiful features
in shaft and arch and tracery, forming perhaps one of the grandest
proofs in the world of the exceeding value of perfect regularity and of
a repetition of good features in architecture, when it is possible to
obtain it on a very large scale.

Leaving the Piazzetta, and stepping into the gondola which has been
waiting for us hard by, let us now go in search of other palaces; but
let us not imagine that we are to see anything equal to the Ducal
Palace. There is, it appears to me, a great gap between it and all other
Venetian buildings; and yet all others seem to have been founded on it,
or on the buildings out of which it grew. Their traceries, seldom
absolutely alike, have still so much general similarity that at first
one may well fancy that there is no variety at all; and, as I have
before said, the general arrangement of their windows and doors is so
nearly identical, that this impression is the more likely to grow upon
the mind.

We will not attempt to take the buildings as they come; but rather as we
think of them, and to some extent in the order of their merit, let us
note down a few of the glories of the domestic work of Venice. And first
let us stop in this narrow canal, for we have by our side one of the
most exquisite little pieces of detail in the whole city. It is an
archway, simple and delicate in its proportions, lovely as it is simple,
and appropriately placed hard by the bridge called “del Paradiso.” I
trust that my sketch is clear enough to shew how pure and good the work
is. The main points to be noted are the characteristic flatness of the
details, and the line of dentil-moulding, which defines all the leading
architectural features, originally invented for borders of incrustations
at S. Mark’s, and here, as everywhere in Venice, used for decoration
afterwards. The incrusted circles of marble on each side of the figure
give great life to the spandrel beneath the arch, and the windows seen
behind shew us a late example of the not unfrequent

[Illustration: 29--ARCHWAY, PONTE DEL PARADISO, VENICE. Page 210]

use of the semicircular and ogee arches together in the same window.

Another precious fragment--the Palazzo San Giorgio, I believe--is
reached from the land side by passing under an arch somewhat similar to
that on the Ponte del Paradiso. This arch is turned between the upper
stories of two houses at the end of a _calle_ properly yclept “dell’
arco detto bon,” and is finished with a steep gable. Beyond it is seen a
fragment of wall veneered with marble, with the upper part of an early
two-light window, and two circular medallions; and above this a piece of
wall veneered in diamonds of red and white marble--so far as I know, a
unique example of such a treatment. The window-head is of that earliest
form of ogee, a circle just turned up to a point in the centre, which
has so manifestly an Eastern origin, and which must not be confounded in
date with our English ogee arches.

In another, and rather desolate, canal in the outskirts of the city,
wider than usual, and with a footpath at the side of the water, instead
of having the walls of the houses running down into it, and forming its
boundary, is the Palazzo Cicogna, which I remember gratefully because it
is one of the few exceptions to the general rule of regularity. The
whole design of this building is very irregular: a detached shaft at one
angle supports a portion of the house which overhangs and forms a sort
of open passage-way; to the right of this opening is a four-light
shafted window, and then a plain wall pierced with two windows, each of
a single ogee trefoiled light. The upper story has two single windows
over the others, whilst over the larger windows and the passage-way is a
large window conspicuous from its size and the peculiarity of its
tracery. It is of six lights divided by very good shafts, and properly
arched with pure and good trefoiled arches; above these, and inclosed
within the perpetual indented or billeted string-course, is a
complicated system of intersecting circles pierced at regular intervals
with quatrefoils. The section of this upper part is very much thinner
than that of the arches beneath. This window is in a most shaken and
decayed state, and not likely, I fear, to be long preserved. The whole
elevation is finished with a shallow cornice supported on corbels.

A doorway on the Ponte San Tomà is quite worthy of a visit. It has the
usual square opening of reddish marble, and above this is a pointed arch
of moulded brick; the tympanum is filled in with a square carved centre
panel, and the ground beyond this with quatrefoils of brick or tile very
prettily disposed, and quite deserving of illustration.

And now let us go back to the Grand Canal; we shall enter it by the side
of the Palazzo Foscari, which, with two other contiguous palaces,
occupies quite the post of honour at the bottom of the principal reach
of the canal, and commands the whole view of its noble and ever-busy way
to where the arch of the Rialto and another bend in the canal close in
the view. We will go a few strokes only towards the Rialto, and then
turn round to look at the palaces we have just passed. They certainly
form a most magnificent group, and are in every way worthy of their
conspicuous position. The palace at the junction of the two waters is
that of the Foscari; the others belonged, I believe, to two of the
Giustiniani family; and but a few yards up the canal, which runs by the
side of the former, is one of the smaller remnants of Byzantine work
already referred to. This group is so well known as scarcely to need any
description--suffice it to say, therefore, that throughout these palaces
the windows are shafted, and the glass is fixed in wooden frames behind
the stonework. This is beyond all doubt what we ought to do; it is the
only sensible and rational mode of adapting the system of traceried and
shafted windows for domestic purposes, and

[Illustration: 30. DOORWAY. PONTE S.TOMA. VENICE. p. 212.]

has here, as elsewhere, the prestige of ancient authority to recommend
it to the consideration of those amongst us who will do nothing without
it. I have enlarged on this point elsewhere, and will, therefore, say no
more upon it now, save that in Venice such a thing as an English monial
ordinarily is, was never known. Windows were invariably shafted from the
earliest period to the latest, and so far invariably of the highest
order, inasmuch as they admitted of the definite expression of the point
at which the monial terminated and the arch commenced, and inasmuch,
too, as the coloured surface of the detached marble shaft must ever be
far more lovely than the lines of tracery mouldings carried down even to
the sill.

The angle-shafts of the Palazzo Foscari have caps and bases in each
stage of the building; those of the other palaces continue up without
interruption.

The date of the smaller palaces, and probably of the large one also, is
very early in the fifteenth century; and the latter had, in 1574, the
honour of being the grandest palace that the Venetians could find in
which to lodge Henry III. of France. They are all three very similar in
their design. Their water-gates are pointed, and the windows in the
water-stage small and unimportant. The second stage is more important,
and has cusped ogee window-heads and balconies. The third stage is,
however, the _piano nobile_, all the windows having deep traceried heads
and large balconies. The fourth stage is very nearly like the first,
save that instead of balconies there is a delicate balustrading between
the shafts of the windows, which is very frequent in good Venetian work,
and always very pretty in its effect. All the windows in these three
palaces have ogee-heads generally finished with carved finials, and
inclosed within a square outline formed by the small dentilled moulding,
and giving what I have before had to refer to--to some extent the
effect of a panel with a window pierced in it, veneered on the front.
The Foscari Palace is the only one of these three that has any
string-courses. The arrangement of the windows--large in the centre and
smaller at the sides--is so nearly regular and of a sort of two-and-two
kind of uniformity, that one scarcely notices that nevertheless, when
internal arrangements make it necessary, a departure from this strict
rule is allowed.

The back entrance to the Foscari Palace is on the side canal. It is of
some interest as retaining, in a very perfect state, an example of a
very picturesque treatment in brick of the Venetian battlement. This
consists of a series of piers finished with a steep gabled outline, and
pierced with trefoiled openings. A good example of this sort of
battlement remained near the Fondaco de’ Turchi, and deserves
illustration. It is quite a Venetian invention, and errs on the side of
quaintness.

[Illustration: BRICK BATTLEMENT--VENICE.]

In a small courtyard, desolate and dreary, reached after crossing the
Ponte di Paglia and one or two other bridges on the Riva dei Schiavoni,
is the Palazzo Badoer, a fourteenth-century palace, the ogeed arches of
the windows in which are more than usually good; whilst the beauty of
the central window, inclosed within a square line of moulding, within
which the wall is incrusted with marble relieved by medallions, is very
great. The structure of this, as of most Venetian palaces, is brick
which has been frescoed; but it is now in a very lamentable state of
decay. The balconies of the lower windows are clearly modern, but there
is a trace of the original balustrade between the shafts of the windows
in the second stage; and in front of the side-lights to the upper window
is a grille of iron-work taking the place of a balcony, and composed of
a combination of quatrefoils. The arrangement of the windows in this
front is not absolutely regular, but still the centre is very marked;
and though it is of early date, the true use of the arch nowhere
appears. The usual[52] dog-tooth cornice finishes the walls under the
eaves. In the courtyard of this house are two of the wells which give so
much character to all the courts in Venice. They appear generally to be
of early date, and look, frequently, like the capitals of large columns,
taken down and placed upon the ground. Those in front of the Palazzo
Badoer are perhaps more like fonts.

Another palace, also said to belong to one of the Badoer family, placed
at the junction of two canals very near the Scuola di San Giorgio and
the Greek Church, is remarkable as being now one of the very few houses
in which the red brick walls are still in their original state, and not
defiled by compo, paint, or whitewash. This house has three fronts, with
one old doorway on the canal, and two on the land. The back facing the
Campo San Severo has an unusually fine and lofty entrance, a square
doorway with a pointed arch above (which I suppose once held tracery),
and a group of five windows with circles or disks of marble in their
spandrels above. Amongst other things it is remarkable as an instance
of the way in which windows were sometimes placed absolutely at the very
angles of the building. Judging by the similarity of its tracery to that
above the Porta della Carta in the Ducal Palace, this angle window must
date from about 1400 to 1430. It has a very bold shaft at the angle,
whilst the jambs have pilasters ornamented at their angles by a twisted
cord-like moulding, which is frequently met with in the later work.
There is a small angle-shaft elaborately twisted just above this window,
and very much like the angle-shafts of the Ducal Palace.

The composition of the main window in the front of this house is, I
think, very striking. The lower window of four lights (one of which is
larger and loftier than the others--a curious instance of the junction
of regularity with irregularity), and whose arches are ogee trefoils, is
surmounted by another window of four lights, with delicate balustrading
between the shafts; and on each side of this upper window, and forming
part of the composition, is a single light, with projecting balcony. The
effect of the whole arrangement is pleasing, and is frequently repeated
in other palaces. The marble incrustation over this window is very much
like that in the other Palazzo Badoer; and from the centre of the
medallions of marble small balls of marble project, fixed with metal,
and giving great life and beauty to the medallions, and I think without
any sacrifice of truth. The main fabric of this building must be of the
latter part of the fourteenth century--the arches of the principal
window being of a very excellent though simple type. Venetian balconies,
of which this palace affords such good examples, are very beautiful and
very characteristic. Nowhere else are they seen in such perfection;
nowhere else, perhaps, were they ever so absolutely necessary. The
palaces rose out of the dark water which washed against their
foundations, and no ground

[Illustration: 31.--ANGLE WINDOW, VENICE. Page 216.]

[Illustration: BALCONY--VENICE.]

could be given up for shady arcades as in other Italian cities, nor were
there any paths to be strolled along; the only resource was, therefore,
to gain from the air that which the land could not afford, and by
projections in front of the windows to obtain that power of enjoying the
delicious evening atmosphere, so cool and pleasant after the fatigues of
the too sultry day. These balconies are almost always very similar,
consisting of a number of delicate shafts with carved capitals,
supporting a piece of stone whose under side is notched up in a series
of trefoils (generally ogee), resting upon the capitals of the shafts.
These are divided occasionally by pilasters, under which are corbels
jutting out boldly to support their weight; and above which sit,
generally, quietly and placidly eyeing the gondolas as they shoot
silently by, small lions, dogs, or other animals--a quaint finish which
one soon learns to like; their angles are often marked by corded
mouldings, and the edges of their floors and copings are almost always
moulded and specked with the perpetual notchings of the nail-head, and
their under sides or soffeits are frequently carved or panelled.

There was great variety in the planning of these balconies. In the
Palazzo Persico, for instance, in which the central windows of the
second and third stages form one great panel, the lower balcony is
continuous across all four lights of the window, whilst the outer lights
only of the upper window have balconies, the two middle lights having
instead a balustrade between their shafts. In other cases the balconies
extend to four lights only of a six-light window, whilst in most they
are confined to the central windows, to which they give much additional
dignity. The Ca’ Fasan affords an almost solitary example[53] of tracery
in a balcony; and the effect of this is so vastly inferior to the usual
shafted balconies, that it seems scarcely necessary to pause to consider
why it should be so. Obviously, however, it is not very convenient to
have the fretful points of cusps and traceries set, as it were, to catch
every projection or point of your dress whenever you lean over the edge
of the balcony to inhale the fresh air or scan the busy scene below.

In the Casa Persico, to which reference has already been made, the
central window is an elaborate composition of the same kind; but the
lower one is of more importance, and has a continuous balcony; and here
I may notice the finials with which the ogee arches of Venetian windows
are so often finished. They appeared to me to be invariably tasteless
and poor in execution, and very mean in their outline. I did not see one
finial in Venice which was satisfying, even when found in conjunction
with otherwise fine work; and I used to wish heartily, when I reached
some palace not before seen, that I might find its arches finished
without them. There was some reason for the wish, too, in the fact that
it is in the later work that these tasteless ornaments are commonest. I
saw them first at Verona, and lamented over them there, but at Venice I
was positively annoyed by the persevering and endless thrusting of their
poverty and badness upon my wearied eyes.

And now let us go again into the Grand Canal, and we shall not have gone
very far up the broad water above the Rialto before we shall find, on
our right hand, one of the most striking groups of mediæval palaces and
houses which can be seen anywhere, even in Venice; this is where the
famous Ca’ d’Oro unites with some three or four other houses, of rather
earlier date, and gives a very fair idea of what the water-scenery of
the ancient city once was. There is some difficulty in criticizing the
Ca’ d’Oro, because, in the first place, it has been restored to render
it fit for the occupation of Mdlle. Taglioni; and, in the next place,
much of the elaborate decoration from which it derived its name, has
perished or been destroyed. As it is, however, it is still a very
sumptuous example of the later fourteenth-century Gothic. Its whole face
is inlaid with squares of red and white marble, and a great amount of
carving is spread over the entire surface, round and between the
windows. This is very flat, but good in its effect. The arcade on the
water-story, and the traceried arcades above, all open into recessed
courts--an arrangement peculiar, I think, among Gothic houses, and
similar in its purpose to the arcades in the Byzantine palaces. Some of
the balconies are good, and the carving of the capitals and moulding of
the window-traceries are very characteristic of Venetian pointed. The
whole design is one-sided, and gives the impression of a house to which
an additional wing has been added. The water-stage consists of an open
arcade of five arches, the central arch round, the remainder pointed,
and on one side of these are two windows with a continuous balcony. The
second and third stages have, above the five open arches, elaborately
traceried windows, of no less than eight lights in width, filling almost
the entire front, the outside lights having balconies, whilst the others
have balustrading. Over the two windows of the water-stage are
single-light windows in each stage. There are throughout this front many
medallions of dark marble, which, let into a field of light marble, are
most brilliant in their effect.

The most remarkable features in the Ca’ d’Oro are, however, the triple
and elaborately carved and chevroned angle-shafts, which I have nowhere
else seen,[54] and the very singular parapet. The height of this is
greater about the centre and at the two ends than elsewhere; but this
appears to have been done rather with the intention of carrying up to
the very top the noticeable division in the building itself than for any
other reason. A very small portion only of the parapet is perfect, and
this it is rather difficult to get at. The small balls of marble affixed
to the outer edge of the trefoils are like those in the tracery of the
Ducal Palace, and in the centre of the medallions of marble everywhere
throughout the city. Their effect is certainly very piquant.

[Illustration: CAPITAL OF WINDOW-SHAFT--VENICE.]

By the side of the Ca’ d’Oro there are three ancient houses of
considerable interest, and the second from the Ca’ d’Oro,

[Illustration: 32. PALAZZZO SEGREDO. VENICE. p. 221.]

the Palazzo Segredo, was a very good example indeed; it has unhappily, I
believe, all been restored and painted, so that now few would believe
that it could ever have been (as it was) one of the very best works in
Venice of its age. It quite deserves illustration, on account of the
extreme vigour and beauty of its great window, which has more of the
flavour of the arcade in the Doge’s Palace, than anything else in
Venice. These three houses are all more than usually irregular in the
arrangement of their windows.

Lower down the Grand Canal, and nearly opposite the Post-office, is the
Palazzo Pisani-Moretta--a very late building, in which all the balconies
are Renaissance, with ordinary balustrading; but this occurs so often in
connection with the latest examples of Gothic work, that I am disposed
to believe that they were possibly, after all, contemporary in their
erection. This palace, too, is remarkable for its double entrance-doors,
with ogee arches, and for the manner in which the central window is
carried up in an uninterrupted way to the very cornice; the lower
traceries being very fair, those in the upper story very weak and bad.

The Palazzo Falcanon (alla Riva Tonda) is another fine house. It has two
water-gates; is four stories in height, the third being the principal
floor; the angle shafts are all spiral, and the string-courses all
ornamented with cable mouldings, which, as is usual, are twisted in
reverse ways from the centre of the front.

The Palazzo Celsi, near the Frari, is, like the Badoer Palace, an
example of a fine regularly designed house with its brickwork left in
its natural state; and the Palazzo Orfei is an instance of the finest
(and a very fine) front being turned towards a _campo_ and not towards a
canal. The long group of mediæval houses which formed one side of the
Campo Sta. Maria Formosa was equally worthy of admiration, but has
lately been modernized--a fate which is only too rapidly overtaking
most of what one used to admire in this once fortunately neglected city!

The window of which I give an illustration, on the Ponte del Fornaro, is
a rare but extremely good example of the combination of sculpture and
tracery. Here the carvings are good examples of the emblems of the four
Evangelists very ingeniously treated, and the whole window has more
force than most of the traceried windows.

With notices of two more buildings, the Palazzi Cavalli and Barbaro, I
shall conclude my remarks upon the existing examples of Venetian
domestic work. Neither of them calls for much remark. The traceries of
the Cavalli Palace are heavy and unsatisfactory, and contrast
unfavourably with the greater simplicity of the windows in the Palazzo
Barbaro. The two palaces stand, however, in a very fine position on the
Grand Canal, commanding the view from the Foscari Palace in one
direction to the church of the Salute and the mouth of the canal in the
other. Nearly opposite them is a very striking house, the Ca’ Dario,
built, I imagine, about the commencement of the sixteenth century,
before the revived Classic feeling had fully possessed the Venetians,
and displaying some effective and beautiful arrangements of
constructional decoration with coloured marbles. It is, in fact, an
attempt to revive, to some extent, the art of incrustation, as practised
at S. Mark’s; and so successful is it, that I wonder much that more
examples are not met with.

In the Grand Canal, and near this spot, are many other buildings, all
worthy of illustration, but adding, I think, nothing to what we already
know. The Ca’ Fasan is the most unlike the other mediæval houses of any;
but it pleased me so little that I could not bring myself to waste time
by sketching it. It is only fair to say that in its traceried balconies
it approaches more nearly to the latest

[Illustration: 33. WINDOW. PONTE DEL FORNARO. VENICE. p. 222.]

[Illustration: 34. CASA GOLDONI. VENICE. p. 223.]

northern pointed than any other building in Venice, and that it has
perhaps at the same time less breadth and dignity than any.

Two fine palaces[55] are now turned into hotels, and that at which I
stopped was full of remains of pointed windows; indeed traces of pointed
work are singularly plentiful, and I might go on to an interminable
length were I to attempt to describe them all. The Arsenal is old, I
believe, but has been modernized. It may be visited now for the sake of
the grand and quaint old lions which sit before its entrance.

Of the interiors of these houses I cannot say very much. They usually
have a great hall in the centre of each floor, into which the various
rooms open; and the windows of these halls are generally the most
important in the elevation. The frames of the windows were of wood,
placed behind the traceries, and the original ceilings were the moulded
beams of the floors. I have only seen one good Gothic staircase in
Venice. This is in the Casa Goldoni, and has for its balustrade a series
of shafts with piers at intervals. Its detail in short is that of the
balcony, but sloped up to suit the rise of the steps. Pointed arches of
brick carry the steps. This house may well be visited by other than
architectural pilgrims, and will be found near the Ponte San Tomà. A
fine early Renaissance staircase remains in the Palazzo Minelli, near
San Paterniano. This is circular, with continuous open arcades following
the rise of the steps, the usual shafted balustrade filling the lower
part of the openings between the columns. The chimneys of these palaces
are very singular. Not many old examples remain, but they are still
copied, and that some of them are really old is proved by the extent to
which they are shown in early Venetian paintings, as e.g. in the works
of Gentile Bellini and Carpaccio. In my illustration the examples Figs.
2 and 3 are copied from paintings, and Fig. 1 is from a palace near the
Ponte Bernardo.

[Illustration: CHIMNEYS--VENICE.

_fig. 1_      _fig. 2_      _fig. 3_]

And now that we have so far passed in review a series of the finest
remains of mediæval architecture in Venice, it is time to inquire how
much is to be learnt by what we have seen, and in what degree it differs
from the developments of pointed architecture with which we are familiar
in Northern Europe.

I think the very first point to be observed is that in Venice
architecture was never essentially constructional in the sense in which
it was in our own land. The pointed arch is rarely used except in
churches, and in its place traceries, increased in size and scale to do
their work, are made to carry the entire weight of walling above them,
as is the case, to take the foremost example, in the second stage of the
Ducal Palace. And it is remarkable that, when the arch was used, from a
very early date it was the ogee arch, and not the arch formed by two
simple curves; indeed it may almost be said that the pure pointed arch
was never used, save where it would have been quite impossible with any
other contrivance to bridge the necessary gap, or provide sufficiently
for the weight to be supported. How striking a contrast this is to the
way in which in England men worked with and exhibited the pointed arch,
evidently as if, and because, they loved it!--using it not only as a
sturdy servant to do heavy work, but as the friend of whose friendship
they were ever the most anxious to boast. I do not complain of the
flatness and lack of breaks or recesses in the masses of the great
Venetian buildings, because this no doubt arose in part from the value
of every foot of ground so hardly gathered from the sea, and the
difficulty of throwing out buttresses into the narrow depths of the
canals out of which they rise. And the same conditions which enforced
this flatness are grateful because they involved the charming balconies
which are so peculiarly Venetian, and gave a breadth and simplicity to
the outline which has its own artistic charm.

In the science of moulding I cannot but think that it is quite useless
to compare works executed for the sunny skies of Venice with those
fitted for the gloomy sunlessness of a northern climate. The one kind
are as properly soft, gentle in their alternations of light and shade,
and delicate, as the other are piquant and sharp, rejoicing in the dark
shade of deep hollows and endless intricacy of outline and arrangement.
But I feel no doubt whatever that, unfair as it may be to compare one
school with the other, seeing that each worked for its own wants, it is
yet most clear that the Northern architects were developing a much
deeper art, and working with much more consummate skill, than were the
Venetian. The endless variety of the arrangement of; capitals, and the
necessary grouping of mouldings to fit, their varying outlines, was
carried to the extreme point of perfection by the one school, whilst in
the other not only was there much less depth and relief, but also very
much less variety. The abacus of the Italian capital was almost always
square in plan, and, as an almost necessary consequence of this,
mouldings retained very much the same arrangement and shape for the
whole period of the prevalence of the pointed style, and generally
rather leaned to the side of heaviness than of delicacy. Venetian
mouldings are composed of the constant combination of a three-quarter
bead and a shallow hollow, divided by small fillets, and so invariably
arranged in almost exactly the same order, that it requires very great
care to decide upon the date of buildings by their mouldings with any
sort of approach to certainty.

In addition to simple mouldings, there are also the ever-recurring
ornamented mouldings which are so peculiarly characteristic of Venetian
works of all dates. These consist generally of sections which in England
we should consider Romanesque, but which in Venice appear to be much
more common in the latest works than in the earliest; chevrons,
cable-mouldings, billets, and the like, are seen everywhere, and suggest
the question whether this class of ornamented mouldings, so largely used
in the early days of architecture in England and so little afterwards,
might not with some advantage be rescued from the contempt into which it
has fallen with modern builders. They have the advantage of being within
the power of any ordinary workman to execute, and do not, therefore,
require the handiwork, which is so rare and so precious, of thoroughly
good carvers. Add to this, that some features originally invented for
use in the way of holding together marble incrustations, were afterwards
used universally for their own sakes as ornamental mouldings, for which
office they were in no way fitted, and I think nearly as much has been
said as can be of Venetian mouldings in stone. Those in brick are even
less satisfactory; but they occur mainly about the churches, and, as I
do not recognize anything at all distinctively Venetian in their design
or arrangement, it will be better to say more about them after we have
seen the brickwork in other cities in the North of Italy, compared with
which that at Venice is not of the first order.

In the practice of carving, as in that of moulding, I see no reason for
yielding the palm to the Venetian. It is true indeed, that the Byzantine
capitals--of which such magnificent examples exist at S. Mark’s--are
some of the most exquisite I have ever seen, true and precise in their
sculpture, revelling in the utmost delicacy of intricate work, and
always refined and elaborated with great evidence of care and
thoughtfulness; but after the earliest school, and those later examples
in which they were copied and regarded as models, there appears to me to
be much less to admire. There is a confusion and want of fixed purpose
about many of those which are commonly referred to as the best types of
Gothic sculpture, which is at best not satisfactory; and I confess that
I came away much more pleased with some of the Byzantine capitals than
with any others. They have some notable points of difference from those
to which we are used. They are generally much larger in proportion to
the shaft than ours; and instead of having a regular neck-moulding, they
rise out of the shaft with a kind of swell, which, as being less
definite, is to me less satisfactory than our neck-moulding. The
capitals of all dates are very generally similar in their outlines--this
in part arising from the constant occurrence of circular columns with
capitals whose abaci are square, and in part from the imitation, more or
less closely, of Byzantine models. Indeed, it is impossible not to see
how great an influence the earliest remaining work--that of the eleventh
century--had in Venice until the end of the fourteenth and far into the
fifteenth century; the most beautiful and striking arrangements of the
former age are reproduced and only slightly modified in the finest work
of the latter to a very remarkable extent: and so much more decidedly
and frequently than are the traces, in northern pointed, of any
hankering after the features of Romanesque buildings, that I think but
one conclusion can fairly be drawn from the sculpture of Venice as well
as from its architecture, viz. that pointed architecture was never
developed as purely and thoroughly in Venice as in the North of Europe;
and that, though it retained its sway there nearly as long as it did
elsewhere, it never thoroughly understood or felt its own strength, and
worked and toiled tied down and encumbered by Byzantine fetters and
Classic sympathies. There is much, notwithstanding this, to admire--and,
above all else, the greatest beauty of the style, wherein it so far left
us behind, the thorough appreciation and unsparing use of the shaft. It
is quite astonishing how very little this was ever used in England.
Occasionally, indeed, it was freely used in grand buildings, and in some
individual features it was frequently seen in thirteenth-century
buildings; but at the very period when, if ever, architecture was in its
perfection--in the early part of the fourteenth century--it was almost
entirely forgotten and thrown aside. All honour, therefore, to the men
who so perseveringly and determinedly used it as did the builders at
Venice for three centuries! And all shame to us if we do not attempt for
the future so far at any rate to follow in their steps! So rare are any
but shafted windows in Venice, that at present I hardly remember a
single instance of a window with monials formed by the continuous
mouldings of the tracery; and it is obvious that this gave occasion, not
only to the use of beautiful marbles--never so well used as in
shafts--but also to the constant use of carved capitals. In domestic
buildings, as I have before remarked, this arrangement of shafted
windows is very valuable, because it suggests one obvious way in which
we may unite traceried windows with the very newest arrangement of
window-frames or sashes in the most comfortable nineteenth-century
houses; for in these Venetian palaces the glass was always contained in
a separate wooden frame set within the marble shafts and tracery.[56]

Besides the use of the shaft in the ordinary way, I must not forget to
say that parapets frequently (or perhaps it were better to say
balustrades)--as, e.g., at S. Mark’s--and balconies everywhere, are
composed of a vast number of very delicate shafts, set very close to
each other, and surmounted by long pieces of stone cut out in imitation
of arching, and not really to be regarded as a succession of
arch-stones, but rather as coping-stones to hold the shafts together.
And, again, they are used very beautifully for the support of open
pinnacles, one at each angle, inclosing a figure, just as in the
monuments of the Scaligers at Verona. Examples of this are to be seen in
the pinnacles which have been added between the gables of S. Mark’s,
which are exceedingly good in their effect; and again in the pinnacles
which terminate the church of the Madonna dell’ Orto.

One more point is worthy of remark--the treatment, namely, of the angles
of buildings. These were almost always marked either by a roll-moulding
or by a succession of nook-shafts, sometimes extravagantly chevroned or
otherwise ornamented. This, when done simply, was always satisfactory,
but, in its later and more elaborate form, was, I think, as
unsatisfactory. The delicate rounding off of the angles of walls was a
point not unthought of in England. In the thirteenth century a
nook-shaft was the common contrivance; in the fourteenth, a chamfer; and
in the fifteenth men reverted entirely to the square form. Here,
however, there is a great and very interesting variety in this
apparently simple feature. The most satisfactory plan of all is where a
quarter-circle forms the angle, and is finished with a small incision in
the form of a V on either side, as it unites simplicity with strength of
construction and softness of contour, and does not force itself too
prominently upon our observation; and, next to this, the most
satisfactory form is where, instead of the moulding being round, it is
pointed at the angle. The twisted shafts of the upper stage of the Ducal
Palace, and the triple and chevroned shafts of the Ca’ d’Oro, are not
improvements upon the refinement of the earlier mode.

I have already spoken of the exquisite beauty of the inlaid marbles in
S. Mark’s; nothing can be better than their effect, and nothing seems
more wonderful than that they should not have been used more frequently
in later buildings. I was, perhaps, a little disappointed in not
finding, as I had expected, the marble arranged generally in geometrical
patterns; but this is quite the exception; and one sees only, in a
medallion here and there, the exquisite beauty which their arrangement
in this way may produce. As a rule the walls are faced with thin slabs
of marble, each of the size in which it came to hand, sawn into as many
slices as its substance would allow, and then riveted to the walls and
held in place securely by projecting thin lines of stonework built into
the wall, and cut with indented or billet ornaments along their edges.
There is, however, a degree of real as well as apparent weakness which
is not at all satisfactory in this system of incrustation, and I thought
how much more noble such work might well become, were it to be inlaid
only where no strong work was required to be done--as, e.g., in
spandrels of arches,[57] or within arches--and not as here to the
concealment of every one of the necessary constructional features. It is
to be observed, however, that the slabs of marble are generally higher
than they are wide, so as at once to destroy any thought of their being
really constructional.

The south side of S. Mark’s is, perhaps, the place above all others in
Venice where this inlaid work may be seen to the greatest advantage.
Some of the great arches which stand in place of gables are divided into
four or five square-headed lights by shafts supporting semicircular
arches, the tympana of which are filled in with delicate and perpetually
varied filigree-work in marble, whilst above them a succession of panels
or medallions shews all the resources of the rich materials which were
to be exhibited. In another case, just over the entrance from the
Piazzetta to the church, the tympanum of the arch is filled in with
large medallions, one exquisitely carved, the others plain; whilst the
arch of the window below the tympanum has its beautiful marble spandrels
adorned on either side with medallions which, for exquisite arrangement
of vari-coloured marbles in geometrical patterns, are perfectly
admirable. There is enough, therefore, in the Venetian system of
incrustation, though much unhappily be lost, to give ample food for our
study and admiration; and its only weak point is, as I have said, its
too frequent neglect or concealment of the constructional features of
the buildings it adorns.

It is easy, however, to cavil at particular details, and scan with a
critical eye the architectural beauties of Venice; but let it not be
thought for an instant that all the wonderful pictures which every new
turn or new point of view brings before the eyes are unappreciated. A
few days spent there suffice almost to fill a lifetime with
reminiscences of all that is novel, beautiful, and strange; and days
such as I have spent, year after year, rejoicing in the daytime in the
full brilliancy of a September sun, and at night in the calm loveliness
of a Venetian night, have been just the most delightful in every way
that could be passed.

We were at Venice on the festival of the Nativity of the Blessed
Virgin--a great feast-day, which it had been my fortune to spend some
two or three times before in Roman Catholic countries. I confess that
here we were not edified. We came in, as we went from church to church,
for rather more than the usual number of the _désagrémens_ which always
seem to attend the decoration of the churches, and especially the
altars, for such festivities abroad. The strongest impression left on my
mind was one of wonder at the paltry character of the long array of what
by courtesy are called, I suppose, wreaths of flowers, manufactured of
pink gauze, or some equally unnatural material. These, with vulgar
draperies hung outside the church doors, and in additional quantity
about the altars, with the most noisy and gladsome ringing of bells,
completed the external demonstrations; all the shops were most
studiously closed, and the churches and open places were thronged with
people. At S. Mark’s we heard[58] some abominably light opera-music,
which sounded, as may be imagined, very discordant within its solemn
walls.

One morning we devoted partly to the ascent of the campanile in the
Piazza. The ascent is entirely by inclined planes; the outer walls of
the tower are in fact double, and in the space between them these
inclined planes are formed; and it is worth notice that to this day, in
all buildings which we have seen in progress in this part of the world,
inclined boards are used instead of ladders for obtaining access to
scaffolding; and in one of the mosaics in the entrance-porch of S.
Mark’s, where the building of the Tower of Babel is depicted, precisely
the same kind of arrangement is shewn. This is interesting, as shewing
the tenacity with which old customs are adhered to. The view, when the
top is reached, quite repays the labour of the ascent, as it gives the
best possible idea of what Venice really is. We get an impression of a
very densely populated town, hemmed in on all sides by water, and
looking very flat and low; in the distance small islands pave the way to
the mainland, or shelter us from the sea; these, where they are more
distant, look like mere black spots on the smooth, unrippled expanse of
water: and in the far horizon we see to the west the purple outline of
the mountains about Vicenza; and to the north of these, and rising
grandly into the sky, the snowy peaks of the southern range of the
Friulan Alps. Below and around are countless churches, all placed
confusedly without respect to orientation--a neglect, if anywhere
excusable, surely so here, where land is the exception and water the
rule.

The last day we spent in Venice was most enjoyable. We had been all day
in our gondola, now stopping to sketch some Gothic palace, anon shooting
into some narrow canal to escape the bright heat of the sun, winding our
way now here, now there, just as the fancy of the moment seized us, and
realizing more than ever that “the longest summer’s day was all too
short” for a last day in so fair a place. In the evening, just before
sunset, we went out into the Lagoon, and, rowing round the small island
of Giudecca, watched the gradually waning light reflected on the smooth,
calm water, which seemed too silent and too soft to be disturbed by a
word from any of us; and then at last, turning back and coming suddenly
through a short canal into the main stream just opposite the Dogana, we
moved on gently till we came abreast of the Ducal Palace. It was just
dark; the moon was rising behind us in all her beauty, and in front,
lamp after lamp was suddenly lit along the Piazzetta, then along the
palace-front, all along the Riva dei Schiavoni, until at last, before we
landed, as far as we could see, the bright lights, reflected in a
hundred gleaming, flashing lines, were fitfully dancing in long streams
of light upon the bosom of the waters.

We stepped on shore to find ourselves led on by the sound of military
music, and to be tempted by the luxury of ices eaten _al fresco_ in the
Piazza; and then, when the crowd gradually dispersed, we too, among the
last, found our way to our hotel, charmed so much with our last night in
Venice that it is impossible not to recollect that evening with the
deepest pleasure.

It is not without purpose that I have held silence with regard to the
churches and buildings generally of the Renaissance school in Venice.
These have had in their time many more admirers than have the examples
of architecture which it was alike my business and my delight
particularly to examine; and to the present day I doubt not that nine
people out of ten, led by their valets-de-place, go to see what is worst
in point of taste, and so reap the reward of allowing themselves to be
made to see with another’s eyes, instead of enjoying the intense
pleasure of working out and exploring for themselves all the treasures
of this mine and storehouse of ancient art. It is partly because I feel
the greatest repugnance to the buildings themselves, and partly because
I fear to make my notes, already lengthy, far too long for the patience
of my readers, that I do not venture upon this additional field of
study; but not in the least degree because I doubt the result, for I
believe firmly that, tried by the fair rules which must regulate merit
in a constructive art, the Renaissance buildings of Venice would be no
nearer perfection than those of any other city. Something perhaps there
is in the gloomy grandeur of their vast masses rearing their rusticated
walls and deeply recessed windows darkly above the comparatively
cheerful and bright-looking walls of the neighbouring Gothic palaces,
which may impress the minds of some, but they must be of a sombre
temperament who really love them. Still more must they be of a tasteless
temperament who can endure with patience the succession of
eccentricities with which Palladio and his disciples have loaded their
churches. I pretend not, however, to discuss the point. I had not time
for everything, and preferred giving up the attempt to like what from my
heart I have ever disliked, and what nothing that I saw in Venice would
make me dislike at all less heartily.

Neither do I pretend to say anything about Venetian pictures; guides
without number may be found of more service and more knowledge, and to
their hands I leave their proper charge. A word only upon one
point--their adaptation, namely, to the sacred edifices of which they
are the most notable ornaments.

Now I must at once say that there is no church, so far as I saw, in
Venice, with the single exception of S. Mark’s, which is to be compared
in this respect (in its effect, that is, as heightened by colour) with
such buildings as the Arena Chapel at Padua or the church of Sta.
Anastasia at Verona--the one an example of the very noblest art working
under strict architectural limitations; the other, of simple decorative
painting. The fact is, that the Venetian pictures give the impression
that they might do elsewhere as well as in a church, and therefore
entirely fail in identifying themselves with the walls on which they
hang; whilst no one can ever think of the noble works of Giotto at
Padua, without recalling to mind the religious order of his works and
their identification with the building which contains them; and at
Verona the result of the system adopted in the painting is marvellously
to enhance the effect of the architecture without in any way concealing
or damaging it. In Venice the case is quite different. The church of San
Sebastiano, in which Paul Veronese is buried, and which internally is
almost entirely covered with his paintings, is an example of what I
suppose I must call the best Venetian treatment. This consists, however,
of immense oil-paintings covering entire walls, and absolutely
requiring, in order that they may be at all properly appreciated, that
the spectator should stand in a particular spot--in some cases by the
side of the altar--and that the windows should first have blinds drawn
down, and then, when he goes to look at another painting, have them
drawn up again. This is all very unpleasant. But besides this, there is
no very sensible advantage to the colour of the buildings from these
decorations; certainly they are far behind mere decorative paintings as
vehicles for bringing out the architectural features; and so they are
visited very much as pictures in a gallery, and without in any case
being identified with the churches in which they are preserved. The
mosaics at S. Mark’s are, on the other hand, some of the very grandest
examples of the proper mode of decorating interiors with representations
of religious subjects, all conceived and arranged with some order and
relation to each other. But of the other Venetian churches there does
not seem to me to be any one whose artists at all succeeded in equalling
the example so early set them.

I do not pretend in these pages to speak at all of paintings
irrespective of architecture, or I might find much to say upon the store
of works, of a very noble school, in which this great city is so rich.
The immense rooms of the Ducal Palace, covered as their walls and
ceilings are with the works of Tintoretto, Titian, and Paul Veronese,
cannot be forgotten; still less can the many works of Giovanni Bellini,
and of other painters in the churches, and in the collection in the
Accademia--rich among others in the works of that great and interesting
painter Carpaccio--be passed over; whilst the decorated walls of the
various Scuole are in many cases of hardly inferior interest. I am sorry
that I was obliged to take the great merits of some of the grandest
works somewhat on faith; it was in vain to think of actually studying
them in a short time, and, educated as I have been to love the works of
an earlier date and another school more heartily than these, I must
confess, barbarous as the confession may appear to be, that I was not
thoroughly pleased with what I saw. The magnificence of the chiaroscuro
and colouring of these great pictures scarcely atoned to me for the
degree to which--owing generally to the immense array of figures and
confusion of subject--I failed to carry away distinct conceptions of the
story intended to be told. It may be said that this is the result of
want of taste or education, but still the feeling is so different when
for the first time pictures by Fra Angelico, Giotto, Raffaelle,
Perugino, or Francia are looked at, that it is hard to avoid believing
that, though their power over colour may have been somewhat less, their
power of attaining to the highest point of the true painter’s art--that
of leaving indelible impressions on the minds of all beholders--was
immeasurably higher. Thus much only by way of excuse for not saying more
about what the world in general rightly conceives to be one of the great
glories of Venice.

And now I must say farewell, and, doubtful though I may be as to the
claims of Venetian art in the Middle Ages to be considered as at all
equal to that of the same period in Northern Europe, I am very grateful
for many new ideas gathered and much intense pleasure enjoyed in the
examination of its treasures; and so, rather sadly laying myself down to
sleep for the last time in Venice, I began to deem that my journey
henceforward must be rather less interesting than it had been; with
Venice a thing of the past, instead of, as it was on my outward course,
full of all the beauties with which the liveliest fancy could crowd its
walls and palaces by anticipation.




CHAPTER IX.

                              “A sea
    Of glory streams along the Alpine height
    Of blue Friuli’s mountains.”
              _Childe Harold._

     New Roads to Venice--The Pusterthal--Innichen--Dolomite
     Mountains--Heiligenblut Kötschach--Kirchbach--Gail
     Thal--Hermagor--Ober Tarvis--Predil
     Pass--Gorizia--Aquileja--Grado--Udine--Pordenone.


To those who wish to find new roads to old haunts let me recommend the
road to Venice described in this chapter. A more interesting way for any
one who has already travelled through Lombardy to Venice cannot be
desired. It affords a sight not only of charming scenery, primitive
people, and churches of some interest, but gives an opportunity for a
visit to Aquileja, Grado, and Udine, all of them places well worthy to
be known by all lovers of architecture. Leaving the Brenner railway at
Franzensfeste, we made our way first of all to Innichen. Here I found a
very fine Romanesque church which, placed as it is not very far to the
north of the distant mountains which one sees from Venice, and full as
it is of Italian influence in its general design, may well be included
in my notes. It is a cruciform church with a central raised lantern,
three eastern apses, a lofty south-western tower, and a
fifteenth-century narthex in front of the rest of the west end. The nave
is divided from the aisles by columns which are, (1) ten-sided, (2) four
half columns attached to a square, and (3) octagonal. The first and
third are massive columns decreasing rapidly in size from the base to
the capital. The central lantern has an octagonal vault upon very simple
pendentives, and the apses have semi-dome roofs. A fine south doorway
has the emblems of the four Evangelists, sculptured around Our Lord in
the tympanum. Innichen is a small and unimportant village, but boasts, I
think, of no less than five churches; and fine as is the mother-church,
I suppose most travellers would agree with me in thinking the background
of mountains to the south of it, the most delightful feature of the
place. Truly I know few things more lovely than the evening view of the
church and village, with the tall fantastic peaks of the Dolomite Drei
Schuster behind, lighted up with the glowing brilliancy which is so
characteristic a result of the Dolomite formation, by the last rays of
the setting sun. Below all was gloomy, dark, and shaded; above the whole
series of towering peaks seemed to be on fire, and most unearthly did
they look. The attraction of such sights as I had seen before compelled
me to give a day to an excursion southwards to the Kreuzberg pass, to
have a glimpse, at any rate, of the Auronzo Dolomites, and I had no
reason to repent the day so spent.

Leaving Innichen and going eastward, we went first to Lienz; then, after
a _détour_ to Heiligenblut, we crossed from the Pusterthal to the Gail
Thal, and from thence across the Predil pass to the Adriatic at Gorizia.
From Innichen till we reached the Italian sea-board, we saw and were
much interested in a series of churches, generally of the fifteenth
century, and all built apparently by the same school of German
architects. They are small mountain churches, and are mainly remarkable
for the complicated and ingenious character of their groined roofs. They
have usually aisles, columns without capitals, and no distinct arches
between them, but only vaulting-ribs. The panels between the ribs are
often ornamented with slightly sunk quatrefoils, or in some cases
regularly filled with tracery.

One of the best of these churches is that at Heiligenblut, in Carinthia.
Here, where the main object of every one is the exploration of the
mountains grouped around the beautiful snow-peak of the Gross-Glockner,
it is not a little pleasant to find again, as at Innichen, a remarkable
church just opposite the inn-door. This was built as a pilgrimage church
to contain a phial of the sacred blood, and is extremely interesting
architecturally as a church, built with a regular system of stone
constructional galleries round the north, south, and west sides of the
nave. The aisles are narrow and divided into two stages in height--both
groined--and the upper no doubt intended for a throng of people to stand
in, and see the functions below. Now, however, just as in most modern
galleries, raised tiers of seats are formed in them, and their effect is
destroyed. A pretty Retable at the end of the north gallery suggests
that originally perhaps they were built in part to make room for side
altars, but this was clearly not the primary object. The fronts of the
galleries are covered with paintings of no merit, which illustrate the
beautiful legend of S. Briccius, who is said to have brought the phial
of blood from the East, and to have perished with it in the snow just
above Heiligenblut. There is a crypt under the choir, entered by a
flight of steps descending from the nave; a grand Sakramentshaus north
of the chancel where the holy blood is kept (not over the altar); and
there is a lofty gabled tower and spire on the north side of the
chancel, whose pretty outline adds not a little to the picturesqueness
of the village.

From Heiligenblut, looking at churches by the same hands on the way at
S. Martin Pockhorn and Winklern, we made our way back to Lienz, and
thence, crossing the mountains, descended on Kötschach in the Gail
Thal, passing a good church on the road at S. Daniel.

Kötschach is in one of the most charming situations for any one who can
enjoy mountains of extreme beauty of outline, even though they are not
covered with snow to their base, nor are more than some nine thousand
feet in height. To me this pastoral Gail Thal, with its green fields,
green mountain sides, wholesome air, and occasional grand views of
Dolomite crags, among which the Polinik and Kollin Kofel are the finest
peaks, is one of the most delightful bits of country I have ever seen.
At Kötschach the architectural feature is a fine lofty gabled steeple
with an octagonal spire. It is very remarkable how German these Germans
are! Here, close to the Italian Alps, we have a design identical with
those of the fine steeples of Lübeck, and as vigorously Teutonic and
unlike Italian work as anything can possibly be.

From Kötschach a pleasant road runs down the valley to Hermagor, another
charming little town beautifully placed, and with--no small
attraction--a capital hostelry. On the road, at Kirchbach, the drivers
of the country waggons in which we were travelling pulled up their
horses, to my no small delight, in front of a most interesting mediæval
churchyard-gate; this is a simple archway overshadowed by a shingled
pent-house roof, to whose kindly guardianship we owe it that a
fifteenth-century painting of S. Martin dividing his cloak with the
beggar, and several saints under craftily-painted canopies, are still in
fair preservation on the wayside gate, making one of the most lovely
pictures possible on the road.

At Hermagor, where the grand and massive mountain range of the Dobratsch
to the east, and the Gartner Kogel to the west, give never-failing
pleasure to the eyes whichever way they turn, there is another fine
church, very much of the same character as that at Heiligenblut, but
without galleries.

Between Hermagor and Ober Tarvis the churches are not important, but one
in the village of S. Paul has the unusual feature of a cornice under the
external eaves effectively painted in the fifteenth century, with
elaborate and very German traceries in red and buff, which are still
fairly perfect.

At Ober Tarvis the Predil Pass is reached; and starting from thence in
the morning, passing on the ascent the pretty Raibl See, and on the
descent some of the most stupendous and aweful rocky precipices I have
ever seen, we reached Flitsch to sleep, and on the following afternoon
emerged from the mountains at Gorizia, not far from the head of the
Adriatic, after a long and beautiful drive down the valley of the
Isonzo.

I found absolutely nothing old to see here. It is a smart town, in which
the hand of the improver has been particularly busy in the work of
destruction; but it is the most convenient starting-place for a visit to
Aquileja and Grado, and provides good horses and vehicles.

It is a drive of about a couple of hours from Gorizia to Aquileja. The
country is perfectly flat, but teeming with vegetation, and it is not
until the end of the journey is reached that one realizes under what
baleful conditions life or existence is endured here. A Roman capital
and a fragment or two of Roman columns or mouldings are all that one
sees at first to show that one is driving into one of the greatest of
the old Roman seaports. Here, where before its destruction by Attila in
A.D. 452 the population is said to have been about a hundred thousand in
number, there are now only a few poor houses, and a sparse population,
pauperized and invalided by fever and swamps on every side, whilst the
sea has retreated some three miles, and left the place to its misery
without any of the compensating gains of commerce. Certainly Torcello is
a degree more wretched and deserted, but these two old cities have few
compeers in misery, and I advise no one but an antiquary to make the
pilgrimage to Aquileja, who is not quite prepared to tolerate dirt,
misery, and wretchedness with nothing to redeem them.

[Illustration: PATRIARCH’S THRONE.--AQUILEJA.]

The one great interest in the city now is the cathedral. This is a great
cruciform basilica, with a central and two small apses east of the
transept, and eleven arches between the nave and aisles. The
arrangements of the apse are interesting; two flights of steps lead up
to it from the nave, and in the centre of the east wall is the
patriarch’s throne of white marble, well raised on a platform above the
seat which goes round the apse. The whole arrangement is singularly well
preserved, and looks very well in spite of the destruction of most of
the mosaic pavement with which originally no doubt the floor was laid,
of which only a few tesseræ now remain, and in spite also of the
modernization of the rest of the apse. This throne appeared to me to be
not earlier than circa 1150, though the church is said to have been
built between 1019 and 1042. These dates must, I think, be taken with
large allowance for alterations. With the exception of the apse and the
crypt under it, I believe the greater part of the church was rebuilt in
the fourteenth century; for though the Roman capitals (which were
everywhere ready to the hand) were used on the ancient columns, the
arches carried by them are pointed, and the clerestory is evidently of
the same age. This combination of Classic columns and sculpture with
pointed arches is so very unusual, that it is quite worth while to give
an illustration of the interior. The columns, capitals, and bases are of
varied shapes and sizes, and evidently a mere collection of old
materials which happened to be handy for the builder’s use; the arches
are rudely moulded, and the clerestory of cinquefoiled windows, each of
a single light, is as insignificant as possible, and yet withal there is
so grand an area inclosed that the effect is good and impressive. The
nave is divided from its aisles by eleven arches on each side, and
measures about one hundred and fifty feet in length, by one hundred and
five in width. The aisle roofs are modern, but the nave still retains
its old roof, a fine example of a cusped ceiling, boarded and panelled
in small square panels. The whole of this ceiling is painted, and with
extremely good effect, though the only colours used are black, white,
and brownish yellow. Each panel is filled with a small painted hexagon
filled with tracery painted in black and white, and all the ribs and
leading lines are yellow and black. The purlines, which are arranged so
as to form the points of the cusps, are very decidedly marked with
black. Simple as the treatment is, the effect is admirable, and it
appeared to me

[Illustration: 35. DUOMO. AQUILEJA. p. 244.]

to be owing to the large amount of white in the panels. Near the west
end of the north aisle is a singular circular erection, which is said by
the cicerone to be the receptacle for the holy oil, but which without
this information I should have taken for the baptistery. It is a
perfectly plain circular mass of stonework about fifteen feet across,
with a doorway on the west side, a moulded base and cornice, and above
the latter a series of detached shafts carrying a second cornice of
marble. A square projection on the north side abuts against the aisle
wall, and seems to have been the special receptacle for the vessel which
held the oil. At present it seems to be as little used and understood by
the people of Aquileja as it would probably be if it were in some
country beyond the Roman pale; a remark by the way on old church
arrangements which one finds oneself making almost everywhere, when one
contrasts the intentions of the old builders with the uses to which more
modern ideas--reformed or deformed, whichever they may be--are in the
habit of applying them.

At Aquileja the appropriation of pagan fragments was carried so far that
we found Classic capitals doing service as holy-water stoups.

The interior of the eastern part of the church is more interesting than
that of the nave. It is all probably of the original foundation, and
retains most of the old arrangements. The floor of the choir is raised
some ten or twelve steps, with two flights of steps on each side of the
centre. At the top of these steps, projecting sideways into the
transepts, are tribunes with open balustrades which seem to have served
as ambons. The apse has two rows of seats, with the patriarch’s seat
raised in the centre, and the altar stands in front of this on the chord
of the apse. It is curious that this, which is an apse internally, is a
square projection from the transept externally.

A descent on each side under the tribunes leads to the crypt under the
raised choir. This is very small, but is divided into three aisles in
width, and four bays in length. The central space is screened round
jealously with close grilles reaching from floor to vault, so as to
protect the shrine of S. Hermacora, which occupies the centre. But
little light steals into this crypt, and that little has to find its way
between rank weeds which grow up round the windows; but there is quite
enough to reveal vaults covered with paintings of subjects, and to show
as picturesque and beautiful an _ensemble_ as one need wish to see.
Kneeling desks were placed round the shrine, but the cultus of S.
Hermacora seems to be no longer popular, and the only pilgrims are
curious visitors like ourselves. The paintings on the groining appeared
to me to be of not earlier date than the fourteenth century, and are
very cleverly contrived to suit the early vaults.

The transepts remain to be mentioned. Each has a small eastern apse near
the extreme end, and a tomb or shrine between this apse and the
choir-tribune. These are of the thirteenth century, and are enormous
blocks of stone, panelled and carved in front, and supported on four
detached shafts. In the south transept there are fragments of a
Byzantine screen round the altar in the small apse, which are of rare
beauty and intricacy. The screen consisted of a solid base, breast-high,
covered with carving, and upon which columns stood originally at
intervals of six feet, just as in the screen at Torcello, of which I
have given a view.

There is an early painting of Our Lord, seated on a throne in the semi
dome of this apse, and there are remains of an early wall-painting in
the choir-apse, partly covered by a fifteenth-century picture in a good
frame. The choir stalls are of elaborate intarsiatura work, and date
from the end of the sixteenth century.

A little way to the north of the church stands its campanile, a tall
plain mass of masonry, with the date MDXLVIIII. on the upper stage, and
the inscription “_Tadeus Luranus hoc o. fecit_.” It is worth the climb
to the top to get the view over the flat surrounding country, which
reveals what one fails to see from the dead level of the road, that the
Adriatic is not far off--far enough, it is true, to have ruined the port
of Aquileja--but so near as to be a very important element in the fine
prospect. From here we saw through the haze the island of Grado, on
which I cast longing eyes in vain. My information as to the distance had
been all at fault, and I thought that in a long day from Gorizia, I
might see both Aquileja and Grado. This is, however, quite impossible,
as the boatmen required, they said, three hours for the _trajet_ each
way. It was a misfortune to miss the church at Grado, which contains
much that is worth seeing, and has considerable historical interest, as
the seat of a patriarch, whose jurisdiction included Malamocco, Venice,
Torcello, and Chioggia--and whose importance is vouched for by its old
titles, “Venetæ oræ Istriæque Ecclesiarum caput et mater,” and “Aquileja
nova.”

The patriarch’s throne and the ambon or pulpit, which still remain in
the church at Grado, are evidently extremely fine examples of Byzantine
furniture. The former corresponds with that of Aquileja, but has the
rare addition of a flat canopy or tester supported in front by two
columns, which rest on the side walls of the steps leading up to the
seat. Probably there was a similar canopy at Aquileja. The dignity of
the patriarchal throne is not a little increased by the addition, simple
as it is in its decorative features. The pulpit is even more striking;
it is six-sided, all the sides being arranged in a series of bold
circular projections, with sculptures of the Evangelistic emblems on
their face. The pulpit is supported by a central shaft, and six smaller
columns alternately plain and spiral, and above the pulpit a series of
octagonal shafts are provided to support a canopy or dome over the head
of the preacher. These columns carry arches which are of the common
Venetian ogee trefoil outline, and, there can be little doubt, are later
than the pulpit. The combination is, however, very picturesque, and not
the less interesting in that it has a most strangely Eastern look.[59]

The rest of my party went, whilst I was sketching in the cathedral at
Aquileja, to look at the baptistery. They reported it to be as
completely modernized inside as it certainly is outside, and so I failed
to enter it. I believe I lost nothing, though at one time it was well
worthy of a visit.

A rapid drive back to Gorizia was made with the advantage of a view of
the mountains before us all the way; and we arrived in time to avail
ourselves of the last train to Udine, which we did not reach until after
dark.

I arrived here in entire ignorance of what might be in store for me in
the way of my art. I had seen no drawings of any of its buildings, and I
suspect that most of my readers are in the same state of ignorance. It
was with no little pleasure, therefore, that my earliest stroll in the
morning brought me to a Palazzo Publico, which if not exactly
magnificent in scale is at least very important, and has the special
merit in my eyes of being all Gothic, and almost unaltered on the
outside since its erection. It stands in a piazza which some sixteenth
or seventeenth century scenic architect has treated with considerable
skill. One or two public buildings and a steep hill behind them have
been dealt with in such a way as to call to mind such a disposition of
buildings as one sees, e.g., on the Capitol at Rome, and no doubt so as
to increase very much the apparent importance

[Illustration: 36. PALAZZO PUBLICO. UDINE. p. 248]

[Illustration: 37. PALAZZO PUBLICO. UDINE. p. 249.]

of this little city. The Palazzo Publico is a building of two stages in
height, the lower entirely open with pointed arches resting on columns,
and the upper presenting on its principal front a large balconied
window, or Ringhiera, in the centre, and smaller windows on each side of
it, and at the ends. The cornice and roof are modern, otherwise the
whole design is intact, and exactly in the state in which its architect
left it. The character of the design is clearly Venetian, and the date
about the beginning of the fifteenth century, but still it is not
slavishly Venetian as the houses of Vicenza are, but on the face of it
the work of a local architect who knew enough of what was being done in
Venice to profit by it without absolutely copying.

The lower or ground story is open on three sides, and has ten arches in
front, and five at each end. The space inclosed is irregularly divided
by a longitudinal line of columns, carrying semicircular arches, which
support the walls of the rooms above, the access to which is by a
modernized staircase in the rear. The materials of the walls are
generally red and white marble. The balustrades between the columns and
the staircases leading to them are so good and complete as well to
deserve illustration. The upper part of these, including the cusped
heads to the openings, is of white marble, whilst the shafts are
alternately of the same material and of serpentine. The upper story is
modernized within; but one learns to be grateful for small mercies, and
it was certainly with every feeling of gratitude to later architects
that I sketched this really beautiful building, which they have been
good enough to leave so nearly unaltered on the outside.

The state of the cathedral is less a subject for thankfulness! The whole
building has been completely modernized within and without, with the
exception of the west front and the lower. The former was the façade to
a nave with two aisles on either side, or perhaps with one aisle and
chapels beyond. All the roofs are of the same flat pitch, and stepped
regularly so as to give a broken and bad outline to the mass. The work
is mainly of brick, with some good detail in the windows of the outer
aisles, of which I give an illustration. The west doorway is of the
fourteenth century, with a very steep crocketed gable between pinnacles,
and a badly sculptured tympanum with a curious assortment of subjects;
in the centre the Crucifixion, right and left of this the Resurrection
and an Agnus Dei, and above it the Nativity. Three circular windows
light the three centre divisions of the front, and the two lower are
connected by a broad band of brick arches which crosses the entire front
just below the central circular window. There is not a word to be said
in favour of such a design. It _is_ old, and that is its only virtue!

[Illustration: AISLE WINDOWS--DUOMO, UDINE.]

[Illustration: 38. DUOMO. UDINE. p. 250.]

The tower is more interesting, though it is only an incomplete fragment.
The lower stage is of stone built in dark and light courses, with a
large sunk recess on each side. On the west side is a fine doorway built
of alternate courses of white marble and serpentine, and there are small
circular windows in the cardinal sides just above the lowest stage.
Above these the whole is a plain mass of brickwork, of which a very
small portion only seems to be original. This tower is no less than
fifty-two feet in outside diameter, and its lowest stage is finely
groined, with no provision for the passage of bells. It might almost as
well have been intended for a baptistery as for a tower! It stands close
to the north side of the choir, and by its side is a rather fine doorway
leading into the transept, with a good deal of late Gothic sculpture and
architectural detail. There are niches and figures in the jambs and
round the arch, the Coronation of the Virgin under the latter, and
figures of the Annunciation stuck against the wall on either side in a
very haphazard fashion. The strange contrast in style between these two
doorways will be seen in the illustration which I give. Here we have,
side by side, examples of the most pronounced kind of two national
styles of Gothic; the door into the tower being as clearly Italian in
its beautiful colour and refined simplicity, as that into the church is
German in its cleverness, want of repose, and hard angularity of detail.

The only other old churches I could find in Udine were San Giacomo and
that of the Ospidale. The former is modernized, but retains an early
square brick belfry, arcaded below, and with simple pointed windows of
two lights above. The church of the Ospidale is also modernized. The
façade has a gable with an old brick eaves-arcade, and the only too
common feature of a large circular window inclosed within a square
border.

A picturesque Renaissance well-canopy (dated 1487) over the Fonte di
San Giovanni was the only other feature I could find worth sketching or
making a note of; and having seen everything, I took the railway on
again to Venice.

The views of the Friulan Alps, under which one travels for some
distance, are very exquisite. We passed Conegliano, where I once left
the railway for a journey through the heart of the Dolomite country to
Cortina d’Ampezzo, and, to my regret, hurried past Pordenone, having
forgotten that at any rate a tall brick campanile was there, which
seemed to promise some reward to the visitor. It is of plain arcaded
brickwork below, and the upper stage is slightly battered out with very
tall machicoulis, from within the parapet of which a smaller octagonal
stage rises, covered with a low spire. The whole composition as one sees
it from the railway is unusual and very good, and recalls just a little
the campanile of the Palazzo Publico at Siena.




CHAPTER X.

    “With all its sinful doings, I must say
       That Italy’s a pleasant place to me,
     Who love to see the sun shine every day,
       And vines (not nail’d to walls) from tree to tree
     Festoon’d.”
                    _Beppo._

       *       *       *       *       *

     Venice to Verona--Verona to Mantua--Villa Franca--Mantua: its
     Churches and Palaces--The
     Theatre--Montenara--Campitello--Casalmaggiore--Longadore--Cremona:
     the Cathedral--Churches and Public Buildings--Lodi--Pavia: its
     Churches--Castle of the Visconti--The Certosa--Drive to Milan.


Our gondolier, anxious not to be too late for us in the morning, slept
in his gondola beneath our windows, and did his best, when the sun rose,
to rouse the sleepy porter of our hotel, but in vain; and at last, when
I awoke, I found we should have a very narrow escape, if indeed we did
not absolutely lose our train. The thing was, however, to be done, and
was done. We shot rapidly--only too rapidly for the last time--along the
smooth waters on which we had been so pleasantly loitering before, and
soon found ourselves at the railway station. Our journey was much like
what such journeys usually are: as far as Verona we were only retracing
our steps, but now the hot sun had quite cleared away the clouds which,
when we passed before, hid the Tyrolese Alps from our sight, and these,
whenever the high acacia hedges which line the railway allowed us a
sight of them, made the journey so far beautiful.

The names of the engines on this railway are very unlike the kind of
nomenclature indulged in at home; we were drawn to Verona, I believe, by
the Titian, and saw, as we rushed along, engines named after Dante,
Sansovino, and other artistic and literary celebrities.

We reached Verona at ten o’clock; the station, however, is so much out
of the town, and the day was so intensely hot, that we gave up the idea
of again going into it, and, contenting ourselves with the general view
of its quaint and picturesque walls rising over the rugged hills which
girt the city on its northern side, we sat down to a breakfast of iced
lemonade and some of those deliciously light cakes which are never had
in such perfection as in Italy, and amused ourselves by watching the way
in which the guards and drivers of the train by which we had travelled
proceeded to solace themselves with a game at billiards, upon a table
provided, I suppose, by the very considerate directors of the railway
company.

The railway from Verona to Mantua crosses a country which is thoroughly
uninteresting in point of scenery; it carried us on well into the great
plain of Lombardy, rich, teemingly rich, in its produce, but flat, arid,
and sultry to a degree. This was altogether one of our hottest days, and
took us fairly into a kind of district in which the heat is most
oppressively felt.

On the road we passed Villa Franca, a small town which has a rather
striking castle, with battlemented walls and a good many square towers,
still very fairly perfect; the whole built in brick, and with
battlements finished square at the top, and not forked like those at
Verona.

We reached the station at Mantua by twelve o’clock, but, as this was
very far from the city, it was nearly an hour later before we were
fairly landed at one--I forget which--of the abominably dirty and bad
inns to which sojourners

[Illustration: 39.--DUCAL PALACE, MANTUA. Page 255.]

within its walls have to submit with the best grace that they can.

Mantua is nearly surrounded by water; two large shallow and
unwholesome-looking lakes giving it this far from pleasant kind of
isolation. Over a long mediæval bridge between these waters the way into
the city from the terminus lies. One of the lakes is higher than the
other, and accordingly twelve mills, each adorned with a statue of an
apostle, are formed upon the bridge, and give it its name of Ponte
Mulina.

The general aspect of Mantua is very dreary and unpleasing, not less
forlorn in its appearance than Padua, and possessing but little
attraction for an architect. The chief architectural feature of the city
is the Ducal Palace, which contains, in the midst of a mass of
Renaissance work of the poorest and most unsatisfactory kind, some very
good remains of pointed architecture.

The finest portion is a long building of vast height, and retaining more
or less of Gothic work throughout, but especially remarkable for the
range of windows in its upper stage. Its front faces on one side towards
the Piazza di San Pietro, and on the other with a very nearly similar
elevation towards the Piazza del Pallone, one of the courts in the vast
palace of the Gonzagas, of which it forms a part. This building is said
to have been commenced about A.D. 1302 by Guido Buonacolsi, surnamed
Bottigella, third sovereign of Mantua, and this date quite agrees with
the character of all the detail. The interior has been completely
modernized, mainly by Ginlio Romano, who carried out very extensive
works in other parts of the palace. The windows in the upper stage of
this portion of the palace deserve notice as being about the most
exquisite examples of their class that I anywhere met with, though those
in the campanile of Sant’Andrea, hard by, are only second to them. The
main arch is of pure pointed form, and executed in brick with
occasional voussoirs of stone--one of which forms a key-stone--and over
it there is a label of brick effectively notched into a kind of
nail-head. The same kind of label is carried round the arches of the
window-openings, and down the jamb as a portion of the jamb-mould, and
again round a pierced and cusped circle of brick in the tympanum. In the
sub-arches the key-stones and cusps are formed of stone. The whole of
the jambs are of brick, but instead of a monial there is a circular
stone shaft, with square capital and band and base. The whole is so
exceedingly simple as to be constructed with ease of ordinary materials,
and it is quite equal in effect to any stone window of the same size
that I have ever seen.

The accompanying drawings will, I trust, sufficiently explain the merit
of this magnificent piece of brickwork. The arcading upon which it
rests, and the perfectly unbroken face of the whole, are very
characteristic of Italian work.

On the opposite side of the Piazza di San Pietro is the cathedral, the
only ancient portion of which is a small part of the south aisle. It is
of very elaborate character, entirely built in brick, and so far as it
remains appears to have been part of an aisle finished with a succession
of gables, one to each bay, a common arrangement in German and French
churches, where additional aisles are so frequently met with, but
uncommon in Italy, where, as in England, churches have seldom more than
one aisle on either side of the nave.[60] The brickwork in this small
fragment of the cathedral, though elaborate, was not pleasing, being of
rather late date.

On the same side of the Piazza as the cathedral is the Vescovato, a
large pile of ancient building, but very much modernized. There still
remain, however, some good three-light windows in the upper stage,
inclosed within a circular arch, without tracery, and divided by marble
shafts. Some

[Illustration: 40. MANTUA

Window in Ducal Palace.]

[Illustration: 41.--CASTELLO DI CORTE, MANTUA. Page 257]

old arches remain also in the lowest stage, which, though now built up,
are still valuable as examples of the best mode of treating brickwork.
They consist of three orders--the two inner formed of alternate
voussoirs of brick and stone, carefully and regularly counterchanged,
and the outer of a moulded terra-cotta ornament. Between each of these
lines a brick of deep red colour is set edgeways, shewing a dark line of
little more than an inch and a half in width, and valuable as very
clearly defining the lines of the arch. All these courses are on the
same plane; and probably another rim of the arch is concealed by the
walling which has been filled in underneath.[61]

Going on from the Piazza San Pietro, and passing under an archway, we
came upon the Castello di Corte, also a part of the ancient palace of
the Gonzaga family, who were for a long time lords of Mantua. It is
certainly a very remarkable piece of mediæval fortification, but its
effect is much damaged by the erection of walls between the battlements,
which in my view I have thought it much better to shew in their original
state, which is evident enough upon careful inspection. The heavy
machicolations which run round the main building have a peculiar and
rather grand effect, particularly in the flanking towers. This portion
of the palace is said to have been erected just at the close of the
fourteenth century.

Close to the Castello di Corte is the Ponte di San Giorgio, one of the
entrances to the city, and built between the Lago di Mezzo and the Lago
Inferiore.

Retracing our steps, we soon found ourselves at the great Palazzo della
Ragione, or town-hall. It has been very much altered, but one gateway
remains in a very perfect state, and is quite worthy of illustration.
The marble shafts in the upper stage of the building are coupled one
behind the other with very beautiful effect. Brick and stone are used
alternately in the main arch of this gateway, with thin dividing lines
of brick, as in the Vescovato. In a wall close to the gate is a sitting
figure, intended, it is said, to represent Virgil, of whom the Mantuans
are still, as in duty bound, very proud. I cannot say much for the
figure or its canopy, both of which are, however, mediæval.

[Illustration: BRICK WINDOW--SANT’ANDREA, MANTUA.]

We found nothing else worthy of notice in this building; but close to it
stands the church of Sant’Andrea, a hideous Renaissance edifice tacked
on to a most beautiful brick campanile.

The detail of this is throughout very fine. The tracery

[Illustration: 42.--GATEWAY, PALAZZO DELLA RAGIONE, MANTUA. Page
258.]

[Illustration: 43.--CAMPANILE, S. ANDREA, MANTUA. Page 259.]

is all of a kind of plate-tracery, consisting, that is to say, of cusped
circles pierced in a tympanum within an inclosing arch; the shafts
between the lights are of polished marble, and coupled one behind the
other. The relative proportion of the cusps in this and in most other
Italian buildings is very good. In trefoils, for instance, the upper
cusp is usually smaller than the lower; and in all good cusping it must
be so. Modern men generally reverse the order, and, at the present day,
so little is the subject really understood that at least ninety-nine out
of every hundred cusped window-openings are designed without feeling,
and quite unlike the best old examples; and this, though apparently a
point of very small importance, is really of great consequence to the
perfection of any pointed work.

The faulty portions of this campanile are the elaborate arcadings in
brick beneath the string-courses, and the awkward and abrupt manner in
which the octagonal stage and the round tile spire are set upon the
square tower. The present appreciation of the building by the good
people of Mantua is shewn by the opening pierced in its lower stage, in
front of which the modestly withdrawn folds of a green curtain disclose
the interior devoted to a barber’s shop, and in which the patient,
seated in the middle of the shop, and looking into the Piazza, submits
to the painful operation of shaving--a common picture in almost every
street of an Italian town, but not pleasant when the place is a portion
of a church.

[Illustration: BRICK WINDOW--SANT’ANDREA, MANTUA.]

The guide-books speak of the church of Sant’Andrea as “among the finest
existing specimens of an interior in the revived Roman style.” If it
really is so, I advise all architects interested in the failure of the
said style to venture, notwithstanding the forbidding west front, into
the nave, when they will perhaps find comfort in seeing how miserable a
building “one of the finest” of its class may nevertheless be!

The people at Mantua seemed to be excessively disturbed by my attempts
at sketching, and at Sant’Andrea they mobbed me so thoroughly that I was
really beginning to think of giving up the attempt in despair, when a
kindly-disposed hatmaker, seeing my distress, came down to the rescue,
and gave me and my party seats in a balcony on the first floor of his
house, in which, sitting at my ease above my persecutors and listening
to the good man’s wife and daughters, I finished my sketch with great
comfort.

In Mantua there are two or three other churches with brick campanili,
but they are very inferior in their character to that of Sant’Andrea,
and hardly worth special notice. We owe it to the French that there are
not more interesting churches, for, having succeeded in capturing the
city after a very prolonged siege, they sacked it, and are said to have
destroyed no less than about fifty of them.

Here, as elsewhere in this part of Italy, most of the streets are
arcaded on either side, affording pleasant shelter from the hot sun, but
every twenty yards we come upon one of an unpleasant class of shops, in
which cheese, oil, and the like comestibles are sold, with most
objectionable effects on all people blessed with noses.

In the evening we found an Italian performance going on at the theatre,
and so thither we went, anxious to see how far Italian comedy might be
amusing. I fear our inquiry was not much to our edification, for the
favourite performers were mainly remarkable for the prodigious rapidity
with which they uttered their facetious sayings, and so we lost more
than half the dialogue. The theatre was almost entirely filled with
Austrians, but still there was a sprinkling of Italians among them,
which did away with the absurdly martial appearance of the only other
theatre we had been into--that at Verona.

The next day was Sunday, but we were obliged to push on; and so,
resigning ourselves to the diligence which left Mantua at about nine, we
booked ourselves for Cremona, under the promise that we should be
delivered there punctually by five o’clock.

We lost sight of Mantua almost immediately, for, travelling along a dead
flat and by roads whose sides are lined with high hedges of acacia or
orchards thickly planted, you never see any place or building until you
have absolutely arrived at it. There was not much to interest me on the
road, and the weather, at first cloudy only and sultry, gradually became
worse, and, before we had gone far, settled into a steady pouring rain;
so we read, wrote, and occupied the many hours in the rumbling diligence
as best we might.

At Montenara, which we passed on our road, the church has a brick
campanile, with pilasters at the angles, and in the belfry two-light
windows, with marble central shafts and round arches. It has one of the
usual brick conical spires, with small angle-pinnacles--a finish to
these campanili which certainly does not improve upon acquaintance. They
are constructed of bricks with semicircular ends laid side by side, the
joints being broken in each course, and so making a very jagged kind of
cone.

The only noticeable point about the church at Montenara is that it has
been lately rebuilt in the very worst taste, and at an angle of
forty-five degrees with the old steeple!

At Campitello there are several remains of interest.

There is a small domestic building, with four pointed windows of two
lights at the side; the windows have central shafts of stone, but are
otherwise entirely of rough brickwork. The church has a kind of double
belfry-stage, arcaded similarly in each stage with round arches. There
are also here the remains of a castle by the river, with a fine tower of
the same type as the angle-towers of the Castello di Corte at Mantua,
and covered with a very flat-pitched roof.

[Illustration: BRICK WINDOW--CAMPITELLO.]

[Illustration: BRICK WINDOW--NEAR CASALMAGGIORE.]

At Casalmaggiore, a town of some importance on the Po, we stopped for
dinner; but it was too wet to attempt to look at the river, and the only
note I made was of a large new church now in course of erection,
Renaissance in style, and with a large dome, and a choir and transepts,
all terminated with circular ends. The redeeming feature about it was
that it was entirely constructed in brick with considerable care, though
probably ere long this will be covered with a coat of plaster, of which
modern Italians are not one whit less enamoured than are modern
Englishmen.

At a village, the name of which I did not learn, between Casalmaggiore
and Cremona, the church had a remarkably good simple brick campanile.
The belfry windows were pointed, of two lights, with a small pierced
circle in the head, the shafts being of stone of course. Beneath the
string-courses there was arcading, and the tower was finished with three
forked battlements of the Veronese type on each face, and behind these
rose a circular brick spire. This tower was to the south-east of the
church.

At Longadore we saw another church with a good early campanile, of which
I made a sketch. This was Romanesque, with angle pilasters, and a
central pilaster carried up as high as the belfry-stage. The belfry
windows were of three lights and shafted. The battlement was most
peculiar--a quarter circle at each angle and a half circle in the centre
of each side, with a narrow space between them; the whole executed in
brick and covered in with a flat modern roof. The angle pilasters
finished under arcaded string-courses. Generally speaking, in these
churches the only ancient features seem to be the campanili, and these
are always of brick and nearly similar in their general design, with
pilasters at the angles, a succession of string-courses--generally
arcaded underneath--and windows in the belfry-stage only.

It was quite six by the time we reached Cremona, and, depositing our
passports at the gate, we trotted on along the smooth granite (which in
these towns is always laid in strips between the rough ordinary paving
for the wheels to travel on), and after traversing a long tortuous
street, and getting a glimpse only of the cathedral as we passed near
its east end, we were soon deposited at the Albergo del Capello. a
comfortable hostelry, which we enjoyed the more by contrast with the
miserable quarters with which we had to put up at Mantua.

Cremona is a city full of interest. The piazza in front of the cathedral
is equal in effect to almost any small piazza I know of. On one side is
the great marble west front of the Duomo, backed by its immense brick
campanile, whose wide fame is proved by the old rhyme, of which the
Cremonese are still so proud--

    “Unus Petrus est in Roma,
     Una turris in Cremona.”

On another side is the Lombard baptistery, a grand polygonal building;
on the third, a most interesting domestic building--the palace of the
Jurisconsults--and the Gothic Palazzo Publico; whilst on the fourth, a
narrow, busy street makes up, by the diversity of colour and costume of
the crowd which is always passing along it, for what it wants in
architectural beauty.

The cathedral must be first described, and it is rather difficult to do
this clearly; but so far as can now be made out it seems much as though
it had at first been built upon a simple plan, with nave, north and
south aisles, and three semicircular eastern apses; and that then to
this, in the fourteenth century, had been added, with hardly any
disturbance of the original fabric, immense transepts, loftier even than
the nave, and so long and large as to give the impression now that two
naves have been placed by some mistake across each other. The groining
of the nave is original in its outline, but barbarously painted in sham
panelling so as entirely to spoil its effect, but otherwise there is
little to notice in the interior, the whole of the church having been
converted with the plasterers’ help into Renaissance in the most
approved manner. The walls are covered with painting, and round the
columns, when we were there, were hung great tapestries, all of which
gave the building a rich though rather gloomy colour.

The west front (if you can forget that it is a great mask only to the
real structure) is rather grand from its large plain surface of arcaded
wall; it has been grievously damaged by alterations, but the old design
is still not difficult to trace. The doorway is very noble, and the
open porch in front of it is carried up with a second stage, in which,
under open arches, stand a very fine figure of the Blessed Virgin, and
figures of other saints of more modern character on either side of her;
above this is a great circular window, whilst the wall on either side of
the porch and window is nearly covered with small arcading. The marbles
in the wall, where the arcading does not occur, are arranged very
regularly in horizontal lines alternately of red and white, each course
being about ten inches or a foot high, and divided from the next by a
strip of white marble about two or three inches in height. The great
rose window is all of red marble, with the exception of one line of
moulding which looks like green serpentine. There are some round windows
in the lower stage on each side of the entrance, but they are quite
modern.

On the north side of the nave rises the Torrazzo, as the campanile is
called here--the “una turris in Cremona”--rising about four hundred feet
from the pavement of the piazza. Its design is much like that of all the
other brick campanili in this district--a succession of stages of nearly
equal height, divided by arcaded string-courses, and marked with
perpendicular lines by small pilasters, and almost without windows until
near the summit. The dark red outline of this magnificent tower tells
well against the deep blue Italian sky, which shone brightly behind it
when we saw it; and the effect of its immense and almost unbroken
outline, rising to such an extraordinary height, is so utterly unlike
that of any of our Northern steeples that we need not trouble ourselves
to compare them. Both are fine in their way; but the Italian campanili
are made up of the reiteration of features so simple and so generally
similar that we cannot fairly class their builders with the men who
raised in England such a multitude of steeples, all varying one from
another, and yet all so lovely.

A door in the east wall of the north transept leads into a small
courtyard, sacred now to the cathedral clergy, from which the original
scheme of the eastern part of the church may be fairly well seen. It
appears to have been a stone building treated in the common fashion of
Lombard churches, but with buttresses and a passage through them round
the apse in front of the windows. There is a modernized crypt under the
choir. The side walls of the north transept are seen very well from the
same courtyard; they are well arcaded in brick, and entirely concealed
from sight elsewhere by the enormous false transept-fronts, the backs of
which as seen from here are certainly among the most ungainly works ever
erected for the mere sake of being beautiful.

The rest of the exterior of the Duomo is almost all of brick. The most
remarkable features are the two transept fronts, which are certainly
magnificent in their detail, though most unreal and preposterous as
wholes; they are, both of them, vast sham fronts, like the west front,
in that they entirely conceal the structure of the church behind them,
and pierced with numbers of windows which from the very first must have
been built but to be blocked up. They have in fact absolutely nothing to
do with the building against which they are placed, and in themselves,
irrespective of this very grave fault, are, I think, positively ugly in
their outline and mass. And yet there is a breadth and grandeur of scale
about them which does somewhat to redeem their faults, and a beauty
about much of their detail which I cannot but admire extremely. Both
transepts are almost entirely built of brick and very similar in their
general idea; but, whilst only the round arch is used in the south
transept, nothing but the pointed arch is used in the northern, and it
is quite curious to notice how very much more beautiful the latter looks
than does the

[Illustration: 44.--NORTH TRANSEPT, CATHEDRAL, CREMONA. Page 266.]

former. The filling-in of stilted round-arched windows with ogee pointed
tracery and much delicate cusping gives the south transept a singularly
Eastern look, and it is impossible not to feel that some such influence
has been exercized throughout its design. It would indeed be most
interesting to find out what this was, but I am not aware that there is
likely to be any clue to it. The date of the work is in all probability
somewhere about the latter part of the fourteenth century. The detail
and management of the whole of the brickwork are exceedingly delicate
and effective, surpassing in their way anything I have yet seen.

[Illustration: BRICK WINDOW--CREMONA CATHEDRAL.]

The putlog-holes are left unfilled, as they almost always are in Italy.
The only stone used is in the doorway and the window-shafts, and these
last are almost always coupled in depth. The windows are elaborately
moulded, and courses of chevrons, quatrefoils, and other ornaments are
introduced occasionally as a relief to what might otherwise be the
tedious succession of mouldings which are necessarily rather similar.
The cusping of brick arches is always managed in the same way; the
bricks all radiate with the arch (not from the centre of the cusp), and
look as though they might have been built, allowing plenty of length of
brick for the cusps, and then cut to the proper outline, the edges of
the cusps being almost invariably left square. Some of the terra-cotta
arch ornaments and diapers are exceedingly good of their kind. The most
remarkable feature, however, about these transepts is the prodigiously
heavy open arcade which runs up the gables under the eaves-cornice--so
heavy and so rude-looking, that, taken by itself, it would probably be
put down as being of much earlier date than it really is. The façade
finishes with three heavy pinnacles arcaded all round, and finished with
conical caps.

To the north transept very nearly the same description would apply, save
that the doorway is much finer, and entirely of marble.[62] It is part
of the original Lombard church, and has no doubt been taken down and
rebuilt where we now see it. The tracery of the rose windows is all
finished in brick, and the detail generally is better and more delicate
in its character than that of the south transept. In both the bricks are
all of a pale red colour, and no dark bricks are anywhere used.

[Illustration: ROSE WINDOW--CREMONA CATHEDRAL.]

The baptistery--which, as has been said, stands south-west of the
Duomo--is entered by a doorway with a projecting porch, whose shafts
rest on the backs of animals. It is

[Illustration: 45.--PALACE OF THE JURISCONSULTS, CREMONA. Page 269.]

octagonal in its plan, built of brick with the exception of the side in
which the door is placed, this being of marble, and is very simple in
all its detail. There are three altars in it, and an immense erection of
masonry in the centre, which, though not open, is evidently a font,
amply large for immersion. Each side has three recessed arches on marble
columns, above which the whole is of red brick with stone string-courses
between the stages. These have corbel-tables under them, which are the
only enrichments in the building. All the brickwork is left to view
inside, and the light is admitted by a pierced arcade very high up in
the walls. The whole is domed over with an octagonal vault of brick, in
the centre of which is a small lantern, and the effect is exceedingly
fine and solemn, and enhanced very much by the grave sombre colour of
the bricks.

[Illustration: WINDOW-JAMB--PALACE OF JURISCONSULTS, CREMONA.]

Close to the baptistery is a building, called in Murray’s Handbook the
Palace of the Jurisconsults, turned when I first saw it into a school
for a not very polite set of children and teachers, who all apparently
felt the most lively interest

[Illustration: CHIMNEY AND BATTLEMENT--CREMONA.]

in my architectural pursuits. It was originally open below, but the
arches on which it stood are now filled up. This upper stage is very
simple and beautiful, and the whole is finished at the top with a
cornice and parapet, with battlements pointed at the top like those in
the Torrazzo, and not forked as we have been lately so accustomed to see
them. At one end of this parapet a chimney rises above the battlement,
which is, so far as I have seen, a unique example of the ancient Italian
contrivance for this very necessary appendage.[63] It is exceedingly
good in its detail, and coeval with the rest of the work. There is a
simplicity and truthfulness of construction about this little building
which make it especially pleasing after the unreal treatment of the
great transept-fronts of the Duomo.[64] By its side stands the Palazzo
Publico, out of one side of which rises one of those singular and very
tall brick towers, without any openings whatever in its walls, which
give such peculiar character to some Italian cities, and of which we
afterwards saw good store at Pavia. The whole of the building shews
either traces of arcades or perfect arcades upon which the upper walls
are supported; they are, however, so much modernized as to be
comparatively uninteresting, though enough remains to shew that their
detail was once very good. The building incloses a quadrangle, which is
rather small, but arcaded on three sides, and opens from the piazza by
open arches under the principal façade, and probably dates from the
middle of the thirteenth century, the date 1245 being given in an
inscription in the courtyard.

There are many churches in Cremona, all more or less appearing to be
founded upon the work in the transepts of the cathedral, but generally
very inferior to them in merit.

San Domenico has a west front singularly like theirs, but debased in its
detail. It has, however, a very fine campanile, lofty, very simple, and
pierced with pointed windows in each stage, one above the other. The
interior is completely modernized, and not worth notice.

SS. Agostino and Giacomo in Breda is another church of the same class,
with a west front which is again a very bad second edition of the
cathedral, and which has been horribly mutilated and modernized inside.
It is, however, to be remembered gratefully for a most lovely picture by
Perugino, representing the Blessed Virgin with Our Lord seated, with
SS. Augustine and James on either side. The Virgin is very calm,
dignified, unearthly, and very simple and stately. Our Blessed Lord, in
her arms, has perhaps rather too much the character of an ordinary
infant; and the two saints have more than is quite pleasant of the bend
in their figures of which he was so fond; the heads stooping forward,
and the knees considerably bent, are a little too evidently straining
towards a reverential posture. Such a criticism is a bold one to venture
upon with the recollection of so glorious a picture fresh in my
mind--one from which I really derived intense pleasure. The date of this
very fine work is A.D. 1494.

Sta. Agata is another church which still has its old campanile intact,
with round-arched windows, very simple and not large. The church which
has been built against it tells its story so well, that at first we all
mistook it for a theatre! So much for Classic symbolism.

Another church, dedicated in honour of Sta. Margherita, is a very poor
erection of brick, with a simple campanile. One or two other churches we
saw with fair brick campanili, which were not otherwise remarkable; and
one there was, San Luca, close to the Milan gate, which seemed to be
very singular in its arrangements. It had a projecting western porch,
with its columns supported on beasts; and at the north-west angle an
octagonal building of brick, of exceedingly late date, which appeared to
be a baptistery.

I enjoyed the architectural remains in Cremona very much indeed: its
rich array of buildings in elaborate brick-work is very striking; and
the campanile of the cathedral, towering up high above the many other
steeples, combines well with them in the general views, and helps to
convert into a fine-looking city what is, perhaps, in its streets and
houses generally, very far from being anything of the kind. The way in
which the old walls and towers of the Palazzo Publico combine with the
steeple of the cathedral is extremely fine, a large piazza a short
distance to the west of the palazzo affording perhaps the best point of
view.

From Cremona we went to Lodi, on our way to Pavia, and had a very
pleasant drive. The heat was intense when we started, and the drivers of
all the carts we passed were prudently ensconcing themselves in the
baskets swung beneath their carts, to escape its effects. Throughout the
Lombardo-Venetian territory there is a great traffic always going on,
and there is a much nearer approach to English arrangements, in the way
of harness and tackle, than it is at all usual to see on the Continent;
though, indeed, it ought in fairness to be said, that their carts are
much more scientific than ours generally are. Any vehicle with more than
two wheels is rarely if ever seen; and these two wheels are sometimes of
prodigious size--I should say quite ten feet in diameter--whilst the
length of the cart from end to end is immense. The extent to which they
are loaded is almost incredible, and of course it requires great care in
order to make the trim exact; but when loaded, the draught must be light
for the weight. It is impossible to talk about horses and carts without
thinking of the magnificent cream-coloured oxen which are everywhere
doing hard work on the roads and in the fields. They have most
magnificent, large, calm eyes; and this, with their great size and slow
and rather dignified motion, makes them look very grand. They are always
yoked to a pole, which rises up above their heads at the end, and has a
carved crosspiece attached to it, against which they press their
foreheads.

At Pizzighettone we crossed the Adda, here a very fine and full stream,
and then, changing horses, went on rapidly towards Lodi. Leaving the
main road, we travelled along a less frequented byroad, infinitely more
pleasant, and in many places very pretty indeed. We followed the course
of a small river, which was turned to good account for irrigation; its
stream being at times divided into no less than three channels, in order
to water the pasture-land on which are fed the cows whose milk is to
produce the far-famed Parmesan cheese. Some part of the road reminded us
pleasantly of English lanes and English scenery, but here and there a
distant glimpse of the Apennines far behind us, and of the Alps beyond
Milan before us, made us aware that we were indeed in Italy.

There is little to be seen in Lodi. It has a large and rather
shabby-looking piazza, at one corner of which is the cathedral, whose
only good feature is its doorway, which is, however, very inferior to
the western doorway of the cathedral at Cremona, to which it bears some
little resemblance.

Another church has a Gothic brick front. The real roof is one of flat
pitch, spanning nave and aisles; but in the façade the central portion
is considerably higher than the sides, so as to give the idea of a
clerestory. This is a foolish sham, and unhappily only too common in
late Gothic work in Italy. The centre division of the front is divided
into three by pilasters, which are semicircular in plan. In the central
division are a door and a circular window, in each side division is a
pointed window, and a brick cornice finishes the gable, crowned with
five circular brick pinnacles.

Another church in Lodi has a very beautifully painted ceiling; this has
been engraved by Mr. Grüner, but unluckily I did not know of its
existence until I returned home; it seems to be an admirable piece of
colour, and to be well worth careful study.

There seemed to be nothing else worthy of notice in Lodi; but, as in
duty-bound we walked down to the bridge,--a rough, unstable-looking
wooden erection over the broad rapid Adda, with nothing about it to
recall to mind the great event in its history, its passage by Napoleon
in 1796.

We left very early in the morning for Pavia: our way led us through a
country most elaborately cultivated, and irrigated with a great display
of science and labour; every field seemed to have some two or three
streams running rapidly in different directions, and the grass
everywhere was most luxuriant. No view, however, was to be had on either
side, as the road found its way through a very flat line of country, and
all the hedges were lined with interminable rows of Lombardy poplars. It
was a country which would have done more good to the heart of a
Lincolnshire farmer than to that of an architect!

The only remarkable building passed on the road was a castle at
Sant’Angelo; a great brick building, with square towers set diagonally
at the angles. The walls were finished with a battlement of the Veronese
kind, and there were several very good early pointed brick windows with
brick monials in place of shafts. A campanile, detached near one angle,
has fine machicolations in stone, now, however, partly destroyed. The
effect of the whole building was very grand.

We soon reached Pavia, and were, as we expected to be, well rewarded by
its churches. The general aspect of the city is singular, owing to the
number of tall slender brick towers which seem to have formed a
necessary appendage to almost every house in the Middle Ages. They are
entirely without openings or ornaments of any kind beyond the
scaffold-holes, and one can compare them to nothing that I know so well
as to the great shot-tower at Waterloo Bridge, save that they are always
square and not circular.

We did our best to see the cathedral, but were unsuccessful; it was
being repaired, and was so full of scaffolding that we could see
nothing. It contains a shrine said to contain the body of S. Augustine,
which I much wanted to see, but seemed in most respects to be an
unprepossessing church.

From the cathedral we found our way to San Michele, a very celebrated
church, and as interesting to an antiquary in search of curiosities as
to an architect in search of the beautiful. The west end is very
curious, and has a succession of sculptures, introduced in the most
eccentric manner, and with but little method in their arrangement. There
are three western doorways, and all of them are elaborately ornamented
with carvings, the central door having above it a very singular figure
of S. Michael.

San Michele, together with San Teodoro and San Pietro, seem all to be of
about the same date, and are of the same character; the most remarkable
feature being in each case the octagonal cupola, which rises above the
crossing of the nave, choir, and transepts: externally these cupolas are
arcaded all round under the eaves, and roofed with flat-pitched roofs,
and are far from being graceful; open arcades are introduced under the
eaves and up the gables, and everywhere there is a profusion of carving.
It is likely enough that this Lombard-Romanesque style, as we see it at
Pavia and elsewhere, did, as has been supposed, set the example which
was very soon after followed in the great churches at Köln and elsewhere
along the borders of the Rhine. In size, however, the children far
exceeded their parents, for San Michele is not remarkable for its
dimensions, except in the width of the transept.

The church consists of a nave and aisles of four bays, a transept of
great length, a central lantern, and a short choir with circular eastern
apse; small apses are also built in the east walls of the transepts. A
fine crypt is formed under the whole of the eastern arm of the cross,
and is entered by steps on each side of the thirteen steps which lead up
to

[Illustration: 46. SAN MICHELE, PAVIA. p. 276.]

[Illustration: 47. CASTLE OF THE VISCONTI, PAVIA.]

the choir. The nave aisles have a second stage or triforium, groined
throughout; and the whole of the church is vaulted, the transepts having
barrel vaults and the three apses semi-domes. The internal effect of the
lantern is extremely good; the pendentives under the angles are very
simple, and low windows are introduced in a stage between them and the
octagonal cupola or vault. The whole church is still left in its
original state with red brick walls and stone piers and arches, save
where, as in the eastern apse, the vault is painted with a Coronation of
the Blessed Virgin, executed in the fifteenth century. It is very
seldom, consequently, that a church of this age is seen to so much
advantage; and undoubtedly the fine, simple, but well-ordered
arrangement of the plan, and the dignified character of the raised choir
and the central lantern, would, even if the colour were not as
picturesque and agreeable as it is, make this interior one of extreme
interest. One of the best portions of the exterior is the east end. Its
extreme loftiness is enhanced by the groups of shafts which divide it
into bays, and rise from the plinth to the cornice. This part of the
building is mainly of stone, except in the fine gallery below the
eaves-cornice, where brick and stone are used together. In addition to
the west door already mentioned, there are two very elaborate doorways
north and south of the nave in the bay next to the transepts.

San Michele is, on the whole, the most interesting building in the town,
but is hardly superior to the grand remains of the old fortified castle
of the Visconti, which stands just on the outskirts of the city, close
to the Milan gate. This is only a portion of the original erection, only
three sides of a great quadrangle which is inclosed by the building now
remaining. The plan originally was a vast square with lofty square
towers projecting at the angles, and of these only two remain. The whole
front is still very nearly perfect, and is not far from five hundred
feet in length, the main building of two stories in height and the
towers of four, and all the old windows more or less intact. The whole
is crowned with a forked battlement, and the old bridge still remains
opposite the entrance with its outer gate, though the drawbridge has
given way to a fixture. This grand pile is now used as a barrack; its
most valuable architectural features are all towards the internal
quadrangle, which is of grand dimensions, more than three hundred feet
in the clear. Towards this court there is the same sort of arrangement
throughout (though many modifications of detail)--an open arcade of
pointed arches below, and a series of fine windows lighting a corridor
above. The lower arches are of stone, everything else of brick, and the
details everywhere are refined and delicate almost beyond those of any
brickwork that I know elsewhere. The original scheme is best seen on the
south side of the quadrangle, of which I give an illustration. This work
dates, I suppose, from about A.D. 1300, but it was soon found to be
inconvenient to have open traceries for the upper corridor, and the
arches on the other two sides were filled in before the middle of the
fourteenth century with very good two-light windows. Fortunately the
whole of this work is still in very excellent preservation, and deserves
much more notice and study than it has ever, I believe, received.[65]
The ordinary bricks used here measure 10½ in. × 5 in. and are 3 in.
high, whilst in San Pantaleone they are 3½ in. high and as much as 15
in. long. Here (as generally in the centre of Italy) the bricks have all
been dressed with a chisel, with which diagonal lines have been marked
all over the face. I can only assume that this has been done to improve
the texture

[Illustration: 48. CASTLE. PAVIA. p. 278.]

[Illustration: 49.--WEST FRONT, S. PANTALEONE, PAVIA. Page 279.]

of the bricks in appearance, and, perhaps where two bricks are side by
side on the same plane, to make a little distinction between them by
tooling the bricks in opposite directions. Two other features of Pavian
brickwork may also here be mentioned: one, that the depth of the
arch-bricks is almost always increased from the springing line to the
centre--the intrados and extrados not being concentric; the other, that
the arch-bricks do not radiate from a centre, but are arranged so as to
obtain a vertical joint in the centre. The first is a very defensible
practice, the second seems to me to be the contrary.

There are several other churches to be noticed here. The most
interesting to me after San Michele is that of Sta. Maria del Carmine,
or San Pantaleone (for it seems to rejoice in a double dedication),
which, in some respects, is more akin to our northern Gothic work than
any other Italian church I have as yet described. The plan and all the
details of the interior are exceedingly simple. The nave is divided into
four groining bays, each of which has two arches into the aisles; the
transept takes one bay and the choir one, and there are an aisle and a
row of chapels on either side of the nave, and chapels on the east side
of the transept. The only openings between the arches and the groining
are small circles by way of clerestory, ludicrously small as compared
with the immense space of blank wall below them, which seems to call
loudly for decorative painting. The whole of the interior is executed in
red brick, which has, however, been much daubed with a coloured wash;
its effect is, notwithstanding, very fine and well worthy of imitation.
As in Italian churches generally, the choir is very short as compared to
the length of the nave. The exterior is even better worth examination
than the interior; the design of the west end is an exaggerated example
of the mode of finishing west fronts not uncommon in Italian Gothic; it
is to some extent a sham, and therefore bad, but there is much which is
of value in its detail, as it is even more than usually elaborate; and
apart from the general outline of the mass, which pinnacles and
pilasters cannot redeem from ugliness, there is considerable beauty in
the group of windows and doors arranged so as to rise gradually to the
centre. The cornices are very heavy and elaborate, and the whole front
may be looked at as a masterpiece of terra-cotta and brick architecture.
It is purely Italian in its great breadth and general arrangement, and,
I confess, very far from being to my taste, though I could wish that we
had more often some of the same breadth and simplicity in our own
elevations.

It is very curious that this west end is the only elevation in this
church which is at all distinctly Italian in its design; for those of
the transepts and choir might much more reasonably be put down as
imitations of Northern work; they are very similar, and a description of
the latter will therefore suffice. It is flanked by massive buttresses,
and has two large and lofty trefoiled lancets, surmounted by a circular
window of great size; the whole is very richly moulded and executed
entirely in brick. The buttresses and roof finish at the top in a rude
temporary-looking manner, and it is therefore impossible to solve the
interesting question of their original terminations, which must, I
imagine, have been pinnacles. The ordinary bricks used here are about 10
in. × 3 in. in size, and laid with very wide joints of mortar; those
used for window-jambs and arches are of much deeper colour and finer
clay than the others. There is something quite refreshing in coming
suddenly and unexpectedly upon such a simple and English-looking
elevation, after the multitude of thoroughly Italian fronts it has been
our fate to see lately.

[Illustration: 50.--WEST END, S. FRANCESCO, PAVIA. Page 281.]

On the north side of the nave there seems to have been a fine row of
pointed windows, but they have been all destroyed to make way for
Renaissance improvements. There are very large buttresses dividing the
bays in this aisle--a feature which is unusual in Italy, and which, in
addition to the design of the choir and transepts, would seem to show
that this church was not entirely the work of an Italian. In the plan,
too, it is remarkable that, though the general arrangement is quite that
of the large Italian churches, such as the Frari at Venice and Sta.
Anastasia at Verona, in one particular it is unlike them. The groining
bays of the aisles are square, and not oblong; and as two of the aisle
arches make one bay of the nave, the groining compartments in both are
as nearly as possible square. This is an arrangement which occurs often
in German Romanesque, but is not seen so often in Italy.

There is a fine campanile between the south transept and the choir; it
has four stages above the roof of the church, and scarcely any opening
below the belfry windows: these are exceedingly good, of three trefoiled
lights under an inclosing arch, with two plain circles pierced in the
tympanum, the monials being shafts of white marble. A low spire of
circular bricks finishes, but does not improve, this very beautiful
belfry.

There is another brick pointed church at Pavia--San Francesco--which has
a fine west front redeemed from the common Italian character by the
grand window-arch in its central division; and though this has been
filled in with later and barbarous work, to the entire concealment of
the tracery, its effect upon the whole front is astonishingly good. The
detail is very elaborate, and in the arch a great number of terra-cotta
ornaments are introduced. The front is divided by large pilasters into a
centre and wings corresponding with the nave and aisles, but these are
again subdivided by smaller pilasters, each of which is composed of
three circles on plan, and finished rather nicely with a kind of finial
at the top.

San Francesco is lighted by a succession of small clerestory windows,
and the aisles have large buttresses, the greater part of the upper
portion of the west front being a mere mask to make out the desired
outline. I begin really to wonder whether I shall see a west front
before I leave Italy which is not a purely unnecessary and
unprepossessing sham!

Pavia is a busy and a pleasant city, and one that improves on
acquaintance; it is true that it was very hot and sultry, but to this I
have been fairly acclimatized, and so rather enjoyed it, except when a
piazza had to be crossed in the sun, or a walk to be taken along a
street unprotected by arcading, which by the way is much rarer here than
in Padua, Mantua, and other cities which I have been describing. The
main street of the city is very picturesque, with somewhat of a fall
towards the south, so that just a glimpse is obtained between the houses
of the distant Apennines.

From Pavia we went, on our road to Milan, to pay a visit to the renowned
Certosa. The road thither, which is also that to Milan, pursues a
monotonously straight course by the side of a canal, or canalized river,
and between rows of stiff trees, until, about four miles from Pavia, a
turning at right angles out of the main road soon leads to the gateway
of the monastery, and through this--which stands open apparently rather
through carelessness than out of hospitality--we drove into the
courtyard in front of the church. This, grown all over with weeds,
looked certainly very desolate and wretched, and but a poor preface to
the polished marbles of the west front, and the riches and paintings of
the interior of the church.

The west front is of great magnificence of material, though a kind of
design which seems to have proceeded upon the principle of setting all
established architectural styles and customs entirely at defiance. This
indeed may be said of the whole church, which is a kind of mixture of
Lombard-Romanesque features with some Gothic, and no slight dash of the
Renaissance spirit; altogether a most magnificent hybrid, but certainly
a hybrid. The doors stand wide open, and from the decaying and desolate
court in front of the church we enter into the nave, full of everything
that is magnificent in material, and all preserved with jealous care and
in admirable order; we look up to the lofty vault which spans the grand
width of the nave, and find the groining ribs arched overhead in pure
pointed form, and cannot help marvelling how far this one pointed
feature harmonizes--I had almost said sanctifies--the whole interior,
though in fact, save this one point, there is scarcely a single detail
throughout the church which would ever pass muster as really being of
Gothic character.

I think it is hardly possible to scan or criticize the architecture of
such a building; it is better to follow the guidance of the cicerone,
and look at the pictures behind the many altars set around with precious
stones, and inclosed within reredoses made of such an infinite variety
of marbles, that, with some degree of envy, one thinks how precious such
an array would be on this side of the Alps, even if spread through fifty
churches.

The nave and aisles are divided from the side chapels and from the
transepts by high metal grilles, and the transept is again divided by
another screen from the choir: this produces a very singular and unusual
effect, and makes the transept appear somewhat like a nave placed at
right angles to the choir. All the chapels on either side of the nave
communicate with one another, so that the monks are able, without
entering the nave, to obtain access to all of them, whilst females are
carefully excluded both from the chapels and from the transepts and
choir. Except a Perugino in one of the chapels on the north side of the
nave, and one picture in the sacristy, there seemed to be no pictures of
any very great value; in fact, travellers are asked rather to admire the
value of the stones which are used in the altars, and the marbles in the
reredoses behind them, than the paintings which they inclose. The
groining of the church, enhanced as it is in effect by the way in which
it is painted--with a blue ground, powdered very richly with gold
stars--conduces more than anything else to the very fine effect of
colour which the nave produces; and the beautiful pavements, composed
mainly of red and white marbles, laid in elaborate geometrical patterns,
increase not a little the general effect. This is an instance of the
superiority of decorative painting over pictures as far as improvement
of architectural effect is concerned.

South of the church are two cloisters; that nearest to the church of
ordinary size, but the other, to which it leads, prodigious in its
dimensions, and very singular in its effect, being surrounded at regular
intervals by the houses of the monks rising out of and above the regular
line of the cloister roof. I went into one of these houses, and found
its accommodation exceedingly ample; three rooms, closets, and a garden
being provided for each monk. The arches of the cloisters are
exceedingly rich in terra-cotta ornaments, and throughout the exterior
of the church and other buildings it is remarkable how very elaborate
these ornamental mouldings are; they are left in the natural reddish
colour, and, as the walls are whitewashed, they have a very singular
effect. We found here, as at other places, men busily engaged in making
casts for the Crystal Palace at Sydenham, whose managers certainly
seemed to have ordered casts of everything that could be modelled
throughout Europe!

[Illustration: 51.--CERTOSA OF PAVIA. Page 284.]

There are now[66] twenty-five monks at the Certosa, and the number
appears to have been gradually on the increase since the reconstitution
of the monastery in 1844; it was certainly very gratifying to see that,
whilst all the rest of the buildings looked forlorn and dilapidated the
church itself was most scrupulously well preserved, presenting in this
respect a great contrast to the fate of monastic churches generally in
the North of Italy.

A tedious drive by the side of a long straight canal, passing on our way
large well-managed farms and other signs of uncommon agricultural
activity, took us from the Certosa to Milan; and long before we arrived
there the white pinnacles of the Duomo, with the Alps in the far
distance, came in sight; certainly, seen thus, the Duomo is one of the
least satisfactory or imposing great churches I have ever seen, and does
but little in the way of imparting character--as most cathedrals do--to
the city which lies at its foot. At last we reached Milan, and entering
through a triumphal arch--the Ticinese Gate--and passing the front of
Sant’Eustorgio, we threaded our way down a very long narrow street, by
the side in one place of a row of Roman columns, still standing
tolerably perfect in the midst of the crowded highway, until at last we
found ourselves housed in a more luxurious hotel than it has been our
fortune to meet with for some days.




CHAPTER XI.

    “In the elder days of art,
       Builders wrought with greatest care
     Each minute and unseen part,
       For the gods are everywhere.”
                 _Longfellow._

     Drive from Padua to Ferrara--Monselice--Rovigo--Ferrara:
     Cathedral--Castle--Gallery--Road to Bologna--Altedo--Bologna:
     Cathedral--San Petronio--San Domenico--San Giacomo--Sta. Maria
     Maggiore--San Francesco--San Stefano--Leaning Towers--Casa dei
     Mercanti--Domestic Remains--Academy--Modena: Cathedral--Parma:
     Cathedral--Correggio--The Baptistery--Piacenza: The Palazzo
     Publico--Cathedral--San Francesco--Sant’Antonino--San Giovanni in
     Canale--Asti: Cathedral--San Secondo--Campanili.


Any one who has followed the route from Venice to Milan described in the
last chapter will do well, instead of following it on a subsequent
visit, to make a _détour_ from Padua to Ferrara and Bologna and thence
by Parma, Modena, and Piacenza to Milan. I shall give some notes of such
a journey in this chapter, the towns visited on the road completing the
subject which I have set before me in this volume, and leaving no
important city north of the Apennines, save Ravenna, undescribed. This
exception is serious; but the omission will be remedied naturally when,
as I hope I soon shall, I ask my readers to go with me to the towns on
the east coast of Italy.

The journey from Padua to Bologna is now, and I suppose always was,
extremely uninteresting. The only objects on the road which possess much
attraction for the artist are the towns I have mentioned, and these
certainly much more repay a visit than do those on the parallel line of
road which we have just travelled. Scenery there is none to speak of,
and fortunately a railway makes the journey a quick one now; but when I
first travelled it I retained no recollection at the end of the route
save of long straight and dusty roads lined on either side with tall
poplars, and wearisome to the last degree. At Monselice we found a
picturesque town domineered over by the ruins of a large castle, whose
walls climb the steep sides of the hill on which it stands. Here I saw a
good and little altered Romanesque church, with pilasters in place of
buttresses, and walls crowned with the usual eaves-arcade. At Rovigo,
further on, there is one of the tall brick towers so common here, with
its Ghibelline battlement perfect. Just before reaching Ferrara we
crossed the Po--here a large, unbridged, and dreary-looking river,
flowing rapidly between high artificial banks to the sea. Its bed is, I
suppose, now quite above the level of the plain, and year by year the
question becomes more urgent and yet more difficult of answer, what is
to be done with it? Certainly there are rivers and rivers; and this
great stream left none but painful and disagreeable impressions on my
mind. At length, after ten hours and a half of the slowest of drivers
and worst of vehicles, we reached Ferrara--a _trajet_ now performed by
railway in about an hour and a half, with advantage to every one’s time
and temper, and no counter-balancing loss.

The entrance to the city through a dirty suburb of tumble-down houses
was not prepossessing. I knew nothing of what I had to see, and my
delight was therefore all the greater when we drove into the piazza in
front of the Duomo, and I found myself gazing at a building which at
first sight looked as though it had been brought straight from the North
of Europe, and planted here in the thirteenth century as a warning
against Italian fashions! Further examination proved that I was not far
wrong in my first estimate of the west front; but it revealed also, I am
sorry to say, that this and part of the south side were the only parts
of the old cathedral which had been spared, when in the seventeenth
century the whole of the church was gutted and converted into about as
bad a Renaissance building as one could wish not to see.

The west front is a great screen, and does not and never did follow the
line of the roofs. It has three gables of about equal height covered
with arcading, which increases in depth and richness of moulding and
shadow to the top, where there are very fine open arched galleries,
stepped up to suit the raking lines of the gables. I know no Italian
work which imitates so closely as this does the extreme richness which
some of the Norman and English churches of the same period exhibit. The
arches of the arcades are carried upon clusters of columns which are set
with extraordinary profusion one behind the other. The centre of the
three divisions of the west front is almost wholly filled with a very
fine porch of three stages in height, and finished with gables on its
front and sides. The lower stage of this porch, as indeed of the whole
church, is round-arched, and belongs probably to the church consecrated
here in 1135. The knotted shafts which carry the front wall rest on
figures sitting on lions. The doorway is deeply recessed, and has
figures of saints with scrolls.[67] The tympanum of the arch has a
sculpture of S. George and the Dragon, and the lintel below it eight
subjects, beginning with the Salutation, and ending with the Baptism, of
Our Lord. At the top of this stage is an inscription which contains the
date given above.

[Illustration: 52. DUOMO. FERRARA. p. 288.]

The next stage is later, and has three arches with traceries carried
upon chevroned and twisted shafts, with a statue of the Virgin and Child
in the centre. This stage forms a sort of balcony. The upper stages are
covered with sculpture. In the gable is Our Lord as Judge in a
vesica-shaped aureole; around and above Him are saints and angels, and
below a group of angels dividing the good from the bad. The subject is
continued on to the wall on each side of the porch, where on one side
are represented the souls of the just in the lap of God, and on the
other the descent of the wicked into the jaws of Hell. The whole scheme
is a picturesque and unusually disposed treatment of the Last Judgment.
There is, however, not much merit in the sculpture as a work of art,
though as an enrichment of the buildings it is most effective. The
statue in the centre is said to be by Nicola Pisano. The south front of
the cathedral has a long range of arcading carried on engaged columns,
each arch inclosing a small arcade of three divisions. Above this, below
the eaves, is a very fine arcade of later date all carried on groups of
shafts, and treated in the Venetian manner with ogee-arched labels above
round arches. This stage is built of red marble; the lower part of the
wall, of brick and stone.

A large font of white marble is the only relic of the original church
which is still to be seen inside.

There is not much more to see here. The castle is remarkable as standing
surrounded by its moat in the very midst of the city. The old portions
of it are much like the castle at Mantua, and present the same boldly
battered base, and the same heavy machicoulis and battlements. I saw
also two or three old churches here, but of no interest; they were of
the latest phase of Gothic--bordering on Renaissance--and very poor in
their detail.

The picture-gallery is in a rather magnificent palace of the D’Este
family. Garofalo--the best of Ferrarese painters--is represented by a
fine Adoration of the Magi, and other works. But the collection
generally is not interesting. There are also in the churches a great
number of works of other Ferrarese painters, and most travellers go also
to see the house of Ariosto and the prison of Tasso. But on the whole
the attractions of the city are not great. The streets are grass-grown
and deserted, lined with palaces of coarse and bold but uninteresting
design, and I was in no way sorry to leave it for Bologna, expecting to
find there much more to interest and occupy me.

The drive from one city to the other was very wearisome. The land was
rich with vines, mulberry-trees, and rice-fields. The grapes were being
gathered, and everywhere along the road we met vast casks borne in grand
waggons drawn by white oxen, carrying home the grapes to the wine-press.
These waggons have an elaborately carved tree from back to front, and
the wheels and casks are also similarly adorned. They are really very
handsome, and quite carry one back to those old times when even in
utilitarian England the adornment of a carriage was not thought beneath
the notice of an artist.

On the road we changed horses at Altedo, where I sketched the good brick
campanile of the church, which has windows much like the example figured
at page 262.

Bologna is, I fear, only known to many Englishmen at the present day as
the one station in Italy at which you may always depend on getting some
food, and not less as a station which seems to be on the road to and
from every part of the peninsula. There is no excuse, therefore, for not
visiting it; and if the general feeling is not one of enthusiasm for
what one sees, there is still very much that is worth seeing, and in San
Petronio a fragment of a church which, had it been completed on the
scale intended, would have been one of the finest and, I suppose, almost
the largest in Italy.

[Illustration: 53. SAN PETRONIO. BOLOGNA. p. 291.]

The streets are not of the best. They are very narrow, and often arcaded
with pointed arches supported on circular columns, but contain few
houses of any interest, whilst the churches have been a good deal
altered and damaged in modern times.

The cathedral is a recent building of no beauty. Two lions which once
supported the columns of a Lombard doorway now carry the holy-water
stoups; and in a passage near the sacristy is a carved, painted, and
gilded rood of the twelfth century--a piece of furniture which is rarely
found remaining in Italian churches. The grandest church in Bologna is
undoubtedly San Petronio, which is well placed on one side of the Piazza
Maggiore. It was not commenced until 1390, so that we must not be
surprized to find the faults in detail which mark the period. But the
general scheme of the church is so magnificent that these faults do not
strike the eye at all offensively. As it stands even now, with only its
nave and aisles finished, it gives a vast idea of size and space, though
this is hardly appreciated at first, owing to the enormous dimensions,
the fewness of the parts, and the extreme simplicity of all the details.
The west front is of immense size and width, but its only finished parts
are the plinth and doorways, the whole of the rest being left in rough
brick. The detail of this finished part is of poor character, and later
than the fabric. It is rather richly carved with figures, which are
sometimes much praised, but which seemed to me (with the exception of a
Pietà over the south-west doorway) to be of poor style and character.
Going round to the side of the building, the design is of earlier date
and much more interesting. The aisle-windows are noble designs of
four-lights in each bay, separated by buttresses and surmounted by
steep-pitched gables. The detail is an extremely good combination of
brick and stone, whilst a magnificent plinth of stone and marble gives
great force to the work. The transept was never built; but at the point
where it was intended to be connected with the aisle, there is a curious
conceit--a window at a projecting corner with half its arch and tracery
facing south, and the other half facing west. It is, so far as I
remember, a unique example in a church, but is just a little like the
angle-windows in some of the Venetian palaces, though these never
indulge in such an absurdity as is the construction of two halves of
pointed arches over such an opening.

[Illustration: PLAN--SAN PETRONIO, BOLOGNA.]

[Illustration: SECTION--SAN PETRONIO, BOLOGNA.]

The interior is very magnificent. The columns, arches, and walls
generally are of brick, now coloured and whitewashed (but originally
intended to be seen, as is evident from parts of the incomplete work
where the internal brickwork is still exposed and is executed with the
greatest care), the capitals and bases being all of stone. The columns
of the nave are bold clusters; they are about sixty feet from centre to
centre, rather short in proportion to the height of their capitals,
which are carved with stiff foliage. Above these is a large pier running
up to carry the groining, and there are pointed arches opening to the
aisles of very lofty pitch, but which, owing to their great size,
certainly look very attenuated. Two chapels open into each bay of the
aisles: these are lighted by the large four-light windows already
mentioned, whilst both nave and aisles have no windows except cusped
circular ones of no great size, placed as near as possible to the
groining, which is very simple throughout the church. There is scarcely
a horizontal string-course or a label to be seen, and the mouldings are
few and simple; yet, nevertheless, the effect is grand. Such a church
may well trouble the mind of the English student who thinks that no
building is complete which has not its arcade, its triforium, and its
clerestory. One of our puny churches would stand--nave, aisles, chancel,
tower, and spire and all--within one of the bays of the nave and aisles
here; and there is a grand sense of restraint and simplicity about this
work which impresses me more each time I see it. At the same time the
interest is of this grand kind--there is a sense of the immense and
infinite, but no condescension to the love of detail and delight in
dainty variety which undoubtedly strikes us in most good Gothic works,
and makes them so enjoyable.

The church which inspired the design of this was, no doubt, the
cathedral at Florence. But of the two the design of San Petronio seems
to me to be the more beautiful. The addition of chapels beyond the
aisles and the traceries in their windows make the design a little less
bald and insipid, and also give a somewhat truer impression of the real
scale than one has at Florence. But at the best such work does not
create enthusiasm. The principal effort of the architect was to build
something very big, and he succeeded; unfortunately he so contrived as
very nearly to prevent one from quite realizing how vast his work is,
and I hardly know a more serious charge that can be made against an
architect than this.[68]

[Illustration: 54. SAN PETRONIO. BOLOGNA. p. 294.]

From this church I went to San Domenico, famous for a very elaborate
tomb or shrine of the saint by Nicola Pisano. This is not a work that
entirely pleases me. It is a high coped tomb covered with sculpture on
the sides, erected behind the altar. The history of its erection by
Nicola Pisano in 1265 gives it value as a dated work by a great artist.
He seems to have been a good deal assisted by his scholar, Fra Guglielmo
Agnelli, whose work is by no means equal to his master’s. The tomb alone
is the work of these artists, the rest of the work about the altar
having been frequently added to and altered in later days, and each
sculptor employed having done his best to glorify his own skill and
dexterity instead of thinking, as Nicola Pisano evidently did, simply of
telling his story in the most straightforward way.

The stories represented in the bas-reliefs had more than common value
for a sculptor of original power. In the centre is the Madonna; on one
side the resuscitation by S. Dominic of a youth who had been thrown from
his horse, and on the other the saint disputing with heretics and
burning their books. At the angles are the four Doctors, and at the back
two more subjects from the Life of S. Dominic, probably designed but
certainly not executed by Nicola Pisano.

The church which contains this tomb has been ruthlessly modernized; but
in the open piazza on the north side of it are two monuments of much
interest. One of these has a square basement of brick, supporting
detached shafts, above which are round arches, the whole being finished
with a brick pyramid. Under the canopy thus formed is placed the
sarcophagus, marked with a cross at the end, and finished at the top
with a steep gabled covering. The detail of this is all of late
Romanesque style. The other monument is of later date and much finer
design, though keeping to the same general outline. In place of the
brick basement of the first this has three rows of three shafts, which
support a large slab. On this are arcades of pointed arches, three at
the sides and two at the ends, carried on coupled shafts, and within
this upper arcade is seen the stone coffin carved at the top, and with a
stiff effigy of the deceased carved as if lying on one of the
perpendicular sides. This monument is also finished with a brick
pyramid. The whole design is certainly striking; it has none of the
exquisite skill that marks the best Veronese monuments, but it is a very
good example of the considerable success which may be achieved by an
architectural design without any help from the sculptor, without the use
of any costly materials, and with only moderate dimensions. The upper
tier of arches is kept in position by an iron tie, and, in spite of its
slender look, still stands after five hundred years’ exposure, in
perfect condition.

San Giacomo is another example of a modernized church, which has,
however, some interesting features. On the exterior there is a rather
good treatment of the polygonal apse, with steep gables and pinnacles
over the windows on each side. This is somewhat like the apse of San
Fermo, at Verona. The campanile on the south-east of the nave is a very
lofty late Gothic erection, finished with an incomplete Renaissance
belfry-stage. The old portion is divided into a succession of stages of
equal height, and is mainly striking on account of its good
colour--being all of red brick--and simple outline. The west front is,
as is usual here, a great gable divided into three parts by pilasters
and half columns; the doorway is of the thirteenth century, and its
columns rest on lions’ backs. The detail of the windows at the ends of
the aisles is extremely good, and seemed to me to be of the same date.
The windows are of two-lights, with shafts for monials, and a broad
transome of plate tracery; the tympanum of the arch at the top is
similarly filled: here, though the jambs are of brick, the whole of the
rest of the

[Illustration: 55. MONUMENT. BOLOGNA. p. 296.]

design is executed in stone, and the likeness to some of the later
Venetian church windows in the Madonna dell’Orto and SS. Giovanni e
Paolo is too great not to be observed. The cornices here are very good.
I noticed not only bricks of unusual shapes, very well arranged for
effect, but also disks of earthenware, set with the convex side in view,
and of brilliant glazed colour, generally blue or green. Their effect is
extremely good.

The church of Sta. Maria Maggiore is less altered internally than the
rest. It is of great length, groined throughout, and in general effect
and proportions seemed to me to have a somewhat less Italian air than
the other churches. The apse has an aisle and chapels round it of which
the brick detail externally is effective. Here there are large tiles cut
to a trefoil shape placed flat against the wall-face as an ornament.
They are good-looking, but have not stood well.

In front of this church there is a court or atrium, surrounded with a
perfectly open arcade, which is continued all along the north side of
the church next the road and over the public footpath. The columns are
very slender and of marble; and though the arches which they carry are
segments of circles (not an agreeable form), the whole effect is
extremely light and graceful.

San Francesco _was_ one of the finest churches in the city. It is now so
shabby and decayed outside, and so covered with painters’ and
decorators’ work inside, that all its good effect is ruined. Its
interior was the victim of what was, I daresay, a very well-meant
restoration some years ago, and little of it has been left in its
original condition. It has a chevet with aisles and chapels, and
externally the rare feature (in Italy) of flying buttresses to support
the choir vault. They were built by men who had never seen one before, I
suppose, and are as crude and misshapen as they well could be.

Two campanili close to each other, south of the choir,

[Illustration: CLOISTER--SAN STEFANO, BOLOGNA.]

group strangely with it in the perspective view. They are unlike in size
and design, and illustrate the perfect indifference with which all
mediæval architects viewed the gravest departures from laws of symmetry.
The larger of the two has rather rich Gothic details, the belfry-stage
having traceried windows of three lights with spiral shafts for monials.
The gables in this church have large white marble crosses let into them;
these are rounded at the ends, and each end has a bright green tile disk
inserted. The west front is of the usual description--a great sham front
of hideous outline. Most of its windows are lancets, new, but probably
copied from old examples, and the doorway under a canopy is of good
character.

I have left to the last what, I suppose, is in fact the oldest of the
Bolognese churches--San Stefano. It is a collection of seven churches,
rather than one church, and, in spite of modernization without end, it
is still a most curious and interesting jumble of old buildings. The
churches are dedicated to--1, San Stefano; 2, San Lorenzo; 3, San
Sepolcro (a circular church); 4, The Corte di Pilato (a cloister); 5,
Sta. Trinità; 6, SS. Pietro e Paolo (with three Romanesque apses); and
7, San Giovanni. No. 3 has an aisle round the circular portion, and was
probably a baptistery, and there is still an old ambon in it. One gets
fairly puzzled in this nest of queer little churches or parts of
churches, and I found but little of architectural--as distinguished from
antiquarian--interest in them. The brickwork in the cloister and in some
of the external walls is extremely good. Some of the latter are diapered
or reticulated on the face with square yellow tiles with dividing lines
of red brick, and the cornices are of the same two colours also. In the
cloister the columns and inner order of the arches are of stone, the
rest of the walls and cornices being of red and yellow bricks, and in
one part there is a course of red, green, and yellow tiles alternated.
The effect of this work is extremely pretty.

Probably travellers remember Bologna more by its two leaning towers than
by any other feature. One comes here however, from either side, after
rather a surfeit of this sort of thing. On the one side is Pisa, with
its leaning tower, and on the other we may see them at Rovigo, Ferrara,
and elsewhere. The soil here is generally bad for foundations, I
suppose, and these plain brick towers without any projection at the base
are the most ill-contrived constructions for such foundations that one
can conceive. In this case it is possible to get views of the two towers
which shew an apparently impossible amount of overhanging on the part of
the smaller one, and I confess to a strong preference for walking to
what one may call the windward side of such an erection! There is no
beauty in these towers, their only features being the vast array of
putlog holes in their walls, and the machicolated battlement of the
higher of the two.

Of domestic buildings of the Middle Ages there are not many remains. The
Casa dei Mercanti, though it dates from the end of the fourteenth
century, is not pleasing. The front has two lofty pointed arches on the
ground story, and a canopied Ringhiera of very poor design above. But as
the whole front has been restored, the bricks painted bright red, and
the stonework cleaned and repaired, I am disposed to believe very little
in the antiquity of any of the details. Certainly I found it not worth
sketching, which was the more disappointing as I had heard of it always
as a fine building. The Pepoli Palace has some old brickwork and a
Ghibelline battlement of unusually picturesque outline. The Piazza
Maggiore in front of San Petronio is certainly the best feature in this
not very striking city. As always in Italian towns, it must be visited
early in the day if it is to be seen to advantage. In the morning it is
crowded with fruit and vegetable dealers, sitting under bright-coloured
umbrellas; in the afternoon it is _triste_ and deserted, save by the
cabmen, who pursue the stranger with their importunities. One side of
this piazza is occupied by the Palazzo Publico--a large pile of building
altogether Gothic in its inclosing walls, I fancy, but they have been so
much altered from time to time that not much detail remains. It seems,
however, to have been much like the Castle at Mantua in its character,
with bold machicoulis at the top of the walls, and a well battered-out
base to the whole building. In a tower here there is a window which I
engrave, because it shews well how good an effect may be produced by the
skilful use of the very simplest materials. The combination of stone
with brick adds much to the effect; and though this is in itself a very
small and unimportant work, it appears to me to be exceedingly
suggestive.

[Illustration: BRICK WINDOW--PALAZZO PUBLICO, BOLOGNA.]

The Academy of the Fine Arts will be visited by every one who cares for
Francia, and by many who fancy they care for the Caracci. As one of the
former class, I recommend it very heartily. It contains a large
collection, gathered in papal times from convents and churches, and
mainly by painters of the Bolognese school. This school has the
redeeming virtue of counting Francia on its list of painters, which may
atone for much. He is seen here to advantage; and one trio, consisting
of two of his pictures on either side of one of Perugino’s, forms the
noblest group in the gallery. Here one sees how much in common the two
men had in spite of Francia’s more forcible character. The same love of
pure and rich colour, the same well-defined grouping, the same religious
feeling mark them both. And never, I think, has the Madonna--pensive,
lowly, and simple--been more beautifully represented than here by
Francia. It is a face that one hopes to remember. Here, too, is
Raffaelle’s S. Cecilia, no doubt a very great work, but not great quite
in the sense one wishes to understand the word at his hand. S. Cecilia
and S. John are very fine, but S. Paul is a sort of burly ruffian. The
heavenly choir is very mundane, and the whole work somewhat academical
in its design and treatment. The earlier pictures here are very
disappointing. The early Bolognese painters seem to have painted
coarsely and heavily, and to have drawn badly, and there is no
comparison for a moment between their work and that of the early Sienese
and Florentine painters on the one hand, and those of Padua and Verona
on the other. One has indeed in their works a sort of foretaste of what
one has from their followers--the Caracci, Domenichino, Bagna-Cavallo,
and to some extent Guido. Dismal colour, great striving after
chiaroscuro, violent and distorted attitudes, and a purely conventional
and academical style, are not great recommendations of any school, but
they are things which must be accepted if one is to enjoy these works at
all heartily.

A very short railway journey takes one from Bologna to Modena. The one
object of interest here is the cathedral, but this is a building of
extreme historical and architectural value, and has fortunately been
left with so few alterations that we can make out its history with fair
certainty. At Bologna everything was built of brick--here at a short
distance we find our eyes rejoicing again in the sight of stone and
marble.

The ground-plan of the cathedral consists of a nave with aisles
terminated at the east end by three semicircular apses. There are a
sacristy on the north of the choir-aisle, and a tower to the north of
this. There are two doorways on the south side, three at the west end,
and one on the north side. A grand crypt with arches on slender shafts
occupies the whole space under the eastern part of the church. The
access to the choir from the nave is by stairs against the aisle walls
in the same position as at San Zenone, Verona. Here the stairs and their
handrails are not later than the thirteenth century, and the choir is
divided from the aisles by screens of the same age; solid below, and
with a continuous cornice carried on coupled shafts above. The cathedral
is said to have been founded in 1099, but an inscription on the south
wall gives the date of the consecration of the building by Pope Lucius
III. in July 1184. I believe that the former date represents the age of
the plan, and of most of the interior columns and arches still
remaining, but that before the later date the whole exterior of the
cathedral had been modified, and the groining added inside. The work of
both periods is extremely good and characteristic. The columns of the
nave are alternately great piers and smaller circular columns of red
marble. The great piers carry cross arches between the groining bays,
and each of these in the nave is equal to two of those in the aisles.
The capitals here are very close imitations of Classical work, with the
abaci frequently concave on plan. The main arches and the triforium
openings of three lights above them are seen both in the nave and aisle,
the vaulting of the latter being unusually raised. There is also a plain
clerestory, and the vaults are now everywhere quadripartite. The outside
elevation of the side walls is very interesting. Here we seem to have
the old aisle wall with its eaves-arcade added to and raised in the
twelfth century, and adorned with a fine deep arcade in each bay,
inclosed under round arches, which are carried on half columns in front
of the buttresses or pilasters. These arches shew exactly what the
original intention was at Ferrara, where it will be recollected they
still in part remain. Certainly they would have made the side walls very
rich in their effect, even if there had not also been two porches, a
projecting pulpit, and various bas-reliefs inserted in them.

All the doorways deserve special mention. The eastern of the two on the
south side, with the porch of two stages in front of it, is remarkable
for the extreme skill and delicacy of its enrichment. The shafts are of
white marble, and the mouldings which separate them of red, while the
former are all carved in the most delicate manner. The porch is mainly
built of red marble, and is carried on detached shafts, cut out of one
block knotted together and resting on lions. The whole of this work is
evidently an addition to the aisle, and dates from about A.D. 1180. The
other doorway on the same side may probably be a work of the original
foundation in 1099. It has the twelve Apostles on the jambs, and rude
shafts carrying a canopy in front of it. The west doorway has also a
porch, and sculptures of the twelve months on its jambs. It is covered
with carving of foliage and figures executed by the same Wiligelmus who
was employed on the western doorway of San Zenone, Verona. Among other
figures are those of King Arthur and his knights, inscribed with his
name,[69] “Artus de Bretania,” above his head. The west front is very
remarkable. The ends of the aisles have two arches inclosing small
arcades similar to those in the bays of the side walls, and the end of
the nave has the same arcade on each side of a porch of two stages in
height, the lower of which is carried on detached

[Illustration: 56. DUOMO. MODENA. p. 304.]

shafts resting on lions’ backs. The upper part of the porch was altered
in order that a great wheel window might be inserted, sometime in the
fourteenth century.

This rose window fills the whole upper part of the western gable, and
is, like many Italian examples, very unskilful in its design. The vast
number of divisions or spokes, and the very slight prominence of the
arcuated part of the filling-in, make it look in very truth a wheel
window and nothing better. Above it are an insignificant figure of Our
Lord and the Emblems of the four Evangelists sculptured in low relief.
The lower portion of the walls is covered in the most promiscuous manner
with bas-reliefs, and a medley of mural tablets, the number of which
would delight the eyes of an English parish clerk; but nevertheless the
rich character given to the work by the fine shadows of the arcades in
the lower half of the front, is worthy of special notice and
recollection. The tower and spire are very lofty. The former has six
stages of nearly equal height, all round-arched, and on the top of this
two octagonal stages crowned with a modern spire. The lower stage of the
octagon is old, and was finished in 1317 by Enrico da Campione, one of
the family of architects of whom I have before spoken. The tower has
pilasters at the angles, and two intermediate on each face, so that
there is a triple division in elevation, and all the horizontal
string-courses are marked by arched corbel-tables. The repetition of
these very simple features, and the absence of all openings in the lower
part of the steeple, shew how simple the elements of a good work may be.

I found nothing else of any interest in Modena, and made my way from
thence to Parma, impatient to see not only the cathedral and the
baptistery, but also Correggio’s treatment of the decoration of the
former. In spite of the great fame of these works, I fear I must at once
confess that they took away most of the pleasure which I had
anticipated from my visit to the cathedral at Parma. This is a grand
Lombard church, fairly perfect in its architectural details and
arrangements, but entirely ruined in its architectural effect by the
frescoes with which most of its walls and roof have been covered. These
have been painted without the slightest thought of the requirements of
the building, and as a matter of course they have entirely ruined its
effect. The frescoes in the dome are by Correggio, and are amongst his
most celebrated works. Like all the rest of the paintings here, they
present, when regarded from below without the assistance of a glass, a
confused mass of distorted figures and limbs, not at all relieved by the
dark and dismal colouring in which they are executed, and which
doubtless is not what it once was. It is true that when examined in
detail, and still more when examined in Toschi’s careful engravings,
they are full of beautiful drawing and skilful chiaroscuro, but the
impression they have left on my mind is mainly one of the extreme risk
of attempting to decorate a building without previous training in and
knowledge of the requirements of architecture. As an example of Lombard
architecture the cathedral at Parma is almost ruined, whilst it would be
difficult to conceive a worse-fitted building for the display of
Correggio’s fancy and skill. The ill-assorted union, in itself ruinous
to both, has been aggravated by the bad state of repair which has
damaged and no doubt altered the colour of the frescoes; and the
impression now produced is that of simply the gloomiest interior in
Italy.

The church is cruciform, with a central cupola and apses to the three
eastern arms of the cross. The nave and aisles of seven bays are
vaulted, and there is a large and striking crypt under the whole eastern
part of the church which goes far to redeem its otherwise barren
character. The effect here is remarkable, owing to the complex
perspective and great number of single slender marble shafts carrying
the vaulted roofs, and in part also to its unusual height. The capitals
are all carved--frequently with coarse volutes; the church was founded
in 1058, and no doubt this crypt is of about that age.

Very near to the cathedral on the south-west is the now much more
interesting baptistery. This is on the exterior a large and lofty
octangular building, adorned in a succession of stages by small detached
shafts carrying the cornices and strings which divide the elevation.
Internally the scheme is very different. The eight-sided interior is
subdivided, so that sixteen shallow apses are set around the inside face
of the walls. These are separated by columns, and above them on each
side are two stages in height, each subdivided into three divisions,
which are again subdivided by smaller columns. A great vault or cupola
covers the whole, and from its height gives an air of solemnity to the
interior. It seems never to have been treated as a real dome, being
covered with a flat roof, resting on the external walls, which are
carried up far above the vault. The paintings with which the walls are
covered are arranged without any order or general scheme of design. They
seem to have been given by various donors, and each gave what best
pleased his fancy; but owing to the early date of most of the work,
there is in parts--especially in the vault--a fine effect of colour.

[Illustration: PLAN OF BAPTISTERY.]

[Illustration: SECTION OF BAPTISTERY.]

This baptistery is said to have been commenced by the architect
Benedetto di Antelamo in A.D. 1196,[70] who is also credited with many
other works here, and specially with much of the early sculpture in the
Duomo and baptistery; it was not completed until 1260.

There are three great doors to the baptistery. On the northern is
sculptured the Tree of Life, and over this twelve prophets carrying
medallions with half figures of the Apostles. Below are subjects from
the lives of Our Lord and S. John Baptist. The western door has a
sculpture of the Last Judgment, and the southern a not very
intelligible, though no doubt symbolical, figure of a man seated in a
tree and gathering honey. Inside there are various sculptures, and among
them a series of illustrations of the labours of the months.

My day in Parma was pleasantly concluded with a visit to the Gallery,
and then, finding no more mediæval remains, I pushed on to Piacenza.

This is a city of no small interest, and remarkable above everything
else in the possession of a Palazzo Publico of unusual and striking
design--a building of special value and interest to me, since it is a
capital example of the use of brick and marble together. Before looking
at any of the churches I devoted myself to this building with the more
satisfaction when I found that it was really, in some respects, one of
the very best works of the sort that I had ever seen.

An inscription carved under a banner on a square stone, in the front,
records the commencement of the work in 1281, and I think we may assume
that no part of it is of much later date than this. It consists, as do
most of these buildings, of a lofty open ground story, and a principal
story

[Illustration: 57. PALAZZO PUBLICO. PIACENZA. p. 308.]

above this. The façade is very dignified in effect. On the ground level
are five lofty arches, very slightly moulded, and resting on square
piers just rounded at the corners. The material of this stage is marble,
mainly white, but with just a line of red and another of grey near the
string-course which divides this from the next stage. From this point up
almost the whole work is executed in brickwork of very elaborate and
delicate detail. The two stages have no kind of uniformity or connection
with each other, six windows being arranged above the five arches. In
the centre of the first floor is the old doorway to the Ringhiera (which
was altered in the seventeenth century); the windows on each side are of
three lights, inclosed under a round arch with a deep archivolt very
slightly recessed--all the enrichments being on very nearly the same
face as the wall. These windows agree in size, but vary very much in all
their details. Some of the subordinate arches are pointed, some round,
and the tympana are everywhere filled with fine brick diapers. Above
this stage the walls finish with a good marble cornice of intersecting
arches, and then with a forked battlement. At the four angles of this
are raised turrets, and the ends are finished with battlemented gables,
very quaint and picturesque, as will be seen by the illustration which I
give. The niches between the two arches in the principal story seem to
have been intended for paintings. The marbles used here are red and
white. Red is used for the arcades under the cornices, for the middle
order of the rose window, and the inner order of the main arches.
Elsewhere the marble is white, except the one grey course below the
principal string. The whole of the lower stage is open below on all
sides and groined--in brick, I think--though it is now plastered;
indeed, save the parts already described as being of marble in the
principal fronts, the whole of this building is built of red brick.
Behind the open ground story there remains a portion of an internal
quadrangle, which, incomplete as it is, shews, nevertheless, the same
delicate attention to detail which is conspicuous on the façades. I know
hardly any detail of Italian brickwork which is so refined and good as
that in the arches and some cusped circles between them in this
quadrangle.

[Illustration: BRICKWORK--PALAZZO PUBLICO, PIACENZA.]

I have seen this building, full as it is of eccentric departures from
ordinary rules and customs--piers being placed over openings, and round
and pointed arches used indifferently--quoted for our benefit as a
remarkable example of a public building erected in the Middle Ages in
the most regular and formal fashion! It is, on the contrary, if the
plain truth is to be spoken, an example of a very bold disregard of such
fashions indulged in without any detrimental effect.

Piacenza cannot boast of any very fine churches. They have been much
ruined internally by modern alterations, and externally they do not
seem ever to have been very attractive. The cathedral is a Lombard
church of fine size, and with some good points. There are three western
porches of two stages in height, according to the usual Lombard fashion.
The doors are rudely sculptured; that in the centre with the signs of
the Zodiac, and the northern door with the Annunciation, Salutation, and
other subjects on its lintel. The columns rest on monsters, griffins,
and men. The whole front is now finished with one large low-pitched
gable, which has an arcade stepped up to suit its cornice. This is,
however, a fourteenth-century alteration of the older church, as is also
a rose window in the centre. A brick tower of late date rises at the
north-west angle of the church, and is finished with a circular spire
built of round-ended bricks. The external walls of the Lombard church
were all built of stone or marble with shallow buttresses; the windows
were very simple, and the effect was mainly due to the fine open arcade
below the eaves, carried now on shafts, now on figures. The north-east
view is very picturesque, owing to the number of angles and apsidal
terminations which are seen together. These are produced by a plan of
most unusual character, the transepts as well as the choir being
finished with three apses, and an octagonal brick lantern rising out of
the centre. Internally the cathedral, in spite of alterations, is still
very interesting. The nave opens with eight arches to the aisles and
transepts; but the three arches which open to the latter are much
loftier than the five western arches, so as to allow of the transepts
opening to the nave to nearly its full height. The groining of the nave
is divided into three bays of sexpartite vaults--each bay being equal to
two bays of the aisles, and there is a lantern over the two eastern
bays. The transepts and aisles have quadripartite vaulting.

Under the choir is a crypt of great interest and beauty, owing to the
vast number of delicate columns which carry the vault. It is planned on
a cruciform arrangement, with the principal altar in an octagonal
central compartment. Here the priest celebrates on the eastern side with
his back to the choir stalls in the apse, and his face toward the west.
The access to this crypt is by two staircases at its south-west and
north-west angles.

[Illustration: PLAN--SANT’ANTONINO.]

The large brick church of San Francesco is of the fourteenth century; it
is very simple, but not striking in effect. The east end has been
planned irregularly--to suit the site, no doubt--with an apse and aisle
round it, and irregularly shaped chapels beyond the apse. The effect is
bad, owing to the unskilful way in which the work has been done. The
west front has the favourite sham gable adorned with circular windows,
some of which are absolutely above the roofs of the aisles! San Giovanni
in Canale is another church which has nothing of interest, save a few
remnants of old brickwork.

Sant’Antonino is a remarkable church, hopelessly modernized with plaster
enrichments. Like the cathedral, it has a lantern in the centre
crossing, carried on eight columns from the ground, which produce
internally a very new and really striking effect. The lantern is
finished above the roof with three stages, each of which is lighted with
a two-light window in each face. There is a fine early marble doorway to
the north transept, with men and monsters supporting the shafts, and
some delicate carving. In front of this, at the end of the fourteenth
century, was built a lofty porch with a great open archway to the
north. It is finished with brick pinnacles and cornices, and is higher
than the transept. The hinges on the west door here are very good, and
the windows in the aisles--lancets with seven cusps in the head--are
quite worth notice. Piacenza struck me, both in its churches and Palazzo
Publico, as a town which had possessed a very distinctly developed
school of architecture of its own. The churches are peculiar, not to say
eccentric, in their planning, and the Palazzo Publico is quite unique in
its design and general treatment.

[Illustration: SECTION--SANT’ANTONINO, PIACENZA.]

The only other old work I noticed here was a house in the Strada San
Marco. This is all of brick, and has arches of slightly horse-shoe
shape. The bricks are all axed on the face, and are of large
size--11¼ inches long by about 2¾ inches high.

Not very far to the east of Piacenza is Asti, a dull city,
distinguished, however, by some remarkable features in its churches. The
most important of these are the Cathedral and San Secondo. They are
extremely similar in general design: they have naves with short choirs,
transepts, low octagonal vaulted lanterns over the crossing, and apsidal
chapels in front of the transept gables, and at San Secondo to the
several bays of the aisle. Their towers are on the east side of the
transepts. The peculiar feature of their detail is the very elaborate
way in which brick and stone are counterchanged in the jambs and arches
of windows and doorways. The moulded members of a jamb are alternately
of brick and stone, and in each course stone comes above the brick of
the courses below. San Secondo cannot, I think, be earlier than circa
1400, but at first sight looks like a building of 1200. The cathedral is
probably somewhat though not very much earlier. Its plan was evidently
derived from that of the cathedral at Piacenza. Its proportions are bad,
and it is only redeemed by the picturesqueness of some of its details.
Another church has an octagonal campanile; and another, one of sixteen
sides. This is of brick, except the upper stage, which is coursed in
brick and stone. Its sixteen sides have alternately a window and a shaft
running up to the cornice, and in the stage below it there are eight
windows below the shafted sides of the belfry. The composition of this
tower is certainly very good. Another fine lofty tower with bold cornice
and Ghibelline parapet recalls the Veronese towers to mind, and there
are besides not a few remains of mediæval domestic work, so that a day
may be well spent at Asti by an architect.




CHAPTER XII.

    “Launce! by mine honesty, welcome to Milan.”
         _Two Gentlemen of Verona_, act ii. sc. 5.

     Milan: the Cathedral--Sant’Ambrogio--Sant’Eustorgio--Sta. Maria
     delle Grazie--Certosa of Chiaravalle--Novara--Vercelli--Monza: the
     Cathedral--The Broletto--Sta. Maria in Strada--Como: the
     Broletto--The Cathedral.


Milan is better known to the generality of English travellers than,
perhaps, any other city south of the Alps; and its older portions afford
a fair idea of some of the most salient points of Italian manners and
customs, whilst its new and much-vaunted arcades and streets seem to be
not only so cosmopolitan as to be un-Italian, but so bald and poor in
design as to be repulsive wherever they are! Its narrow busy streets,
though they are wanting indeed in the arcading so characteristic of very
many other towns we had passed through, have that peculiar charm which
life and bustle always give to strange places; the crowds of
foot-passengers threading the narrow ways, with no protection from the
omnibuses and carts which jostle against them, are full of animation,
and lively and picturesque in their costumes. Elbowing our way between
them, we soon found ourselves in a piazza, with the Duomo rising before
us in all the magnificence of its white marble walls. If it be indeed
true that it was designed by a German,[71] there is on the outside even
more cause for astonishment at his work than if it had been the work of
an Italian. The west front is quite modern, but the rest of the
exterior, all in its original state, is as little German in its
character as any building I have ever seen, and--shall I add it?--as
little really grand as a work of art. I had just caught a glimpse of its
general outline and effect by the bright moonlight, on the evening of my
arrival, with the music of an Austrian band sounding pleasantly in my
ears, and, thus seen, there was certainly something wild and striking in
its effect. I saw the brilliantly light colour of the white marble in
the full brightness of the moon, and little of the poverty of moulding,
or the heaviness of traceries, or the preposterous tenuity of pinnacles,
which daylight revealed, to the destruction of any belief I might still
have in its beauty; and the more I examined it in detail the less
satisfactory did it appear; for neither in its general mass, nor in its
detail, does it bear examination. Its walls are panelled all over, the
panelling having a peculiarly painful kind of pendulous, unsupported,
and unconstructional character, and the string-courses are marked by a
continuous trefoil arcading on their under side, which recalls the
frequent Italian string-courses

[Illustration: GROUND-PLAN--MILAN CATHEDRAL.]

in brick. The buttresses are bold in their formation and scale, but poor
and weak-looking in their design, and finish at the top with pinnacles,
whose thin outline, seen against the deep blue sky, is painfully bad and
unsatisfying. The panelling of the walls is continued up to their whole
height without any decided line of parapet or cornice, and finishes in a
rough serrated line of small gables, which is particularly restless and
wanting in repose. Great flying buttresses span the aisles, and then in
the clerestory is repeated exactly what we have already seen below, the
same panelling, the same parapet, and the same light pinnacles; the
windows, however, are here very small and insignificant, whilst those in
the aisles are remarkable for their large size and for the singular
traceries with which they are filled. All the lower windows are
transomed with a line of tracery, surmounted in each light by a
crocketed canopy running up into the light above. In the apse this
tracery fills the four outer lights only on each side of the two centre
lights, the others being continued without interruption to the sill; and
in these windows it is remarkable that each light is subdivided with a
small monial below this band of tracery.

Altogether, an effect of a prodigious number and repetition of vertical
lines is produced, and yet, notwithstanding this, the effect of the
entire building is decidedly rather horizontal and depressing than the
contrary; this is not more owing to the absence of all visible roofing
than to the way in which the parapets, with their irregular gabled
outline, attract the eye to a markedly horizontal feature.

Upon the whole, therefore, the exterior is in no respect more Italian
than it is German in its style; it belongs to no school, and has no
fellows: from the beginning it has been an exotic, and to the end of
time will probably remain so, without a follower or an imitator in the
singular development of which it is the only example; and there does
appear, if we consider the matter, to be some intrinsic probability that
such a building must have been designed by a foreigner rather than by a
native. It has, in fact, all the appearance of having been the work of
a stranger who was but imperfectly acquainted with the wants or customs
of Italian architecture, working to some extent with the traditions of
his national school before him, but, at the same time, impressed with a
strong sense of the necessity under which he lay, of doing something
quite unlike what he had been taught to consider necessary for buildings
in his native land. It will be found upon examination that there is
absolutely hardly one point in which this vast building follows the
traditions of Italian pointed work. Its plan has not any Italian
characteristics in its arrangement or proportions. Its windows have
moulded monials instead of shafts; its walls are buttressed instead of
being marked with pilasters here and there; its pinnacles are Northern
in their idea, for strength rather than (as Italian pinnacles were) only
for ornament; its walls have no cornices, and there are no sham fronts
or attempts at concealing the necessary features of construction; the
walls are panelled instead of being arcaded, and there is a constant
endeavour to break up plain surfaces of wall, unlike the predilection
for smooth broad surfaces so usual in thoroughly Italian work, and
destructive of the kind of breadth and dignity which this last generally
has; finally, if rest and life may be taken--as by some they are--to be
respectively the distinguishing features of Italian and Northern pointed
work, then, assuredly, the lack of repose in Milan Cathedral--caused
mainly by the degree to which the system of panelling is carried over
the whole building, and the extent to which the use of the simple
horizontal line is carefully excluded--goes far to consign its exterior
to some school of life and restlessness of the most unsatisfactory
character. The one point in which an Italian model has been at all
followed is the section, where the proportions of the aisles to the nave
and the pitch of their roofs certainly shew a knowledge of San
Petronio, Bologna. But its architect appears to me to have been shocked
at the necessity under which he lay of sacrificing the steep lines of
roof so dear to him in his native land, and to have striven with all his
might to provide a substitute for their effect by the vertical lines of
his panelled buttresses and walls, by the gabled outline of his
parapets, and by the removal even of such a slight horizontal mark as
the commencement of the traceries of his windows on one line. And his
work is a most remarkable standing proof of the failure of such an
attempt; for, despite all these precautions, and I incline to believe in
consequence of them, the general effect is, after all, entirely
horizontal. Extremes meet, and so the attempt to avoid absolute
horizontal lines has completely failed, because in their place we have a
succession of vertical members placed side by side in such endless
numbers that we really think more of the horizontal arrangement of these
members, slight though they are, than we should of the simplest defined
horizontal line.

And the same consequence followed the same kind of work in England. I
know no building in which the horizontal line is more painfully
predominant than in Henry the Seventh’s Chapel at Westminster, where,
nevertheless, it is broken endlessly by vertical lines of pinnacles, and
where the walls are covered in all parts with perpendicular lines of
panelling.

Indeed, I should have but little respect for such a building as this
exterior of Milan were it not for its rare material (used though it is
in a prodigal manner, and without particular reference to its nature)
and its immense size--though this is far less in appearance than in
reality. But my detraction and harsh criticisms must end here; for if,
having first made the circuit of the entire church, the flight of steps
which leads up to the west door is last of all mounted, the first
feeling must be one of perfect amazement and delight--amazement that
the same mind which conceived the exterior should have been able also to
achieve anything so diverse from it as is the interior, and delight that
anything so magnificent and so perfect should ever have been reared on
the southern slope of the Alps, to exhibit, to the eyes as it were of
enemies, the full majesty and power of the pointed architecture of the
North. And mark, upon consideration, how very natural this was. Its
architect had been tied down in his exterior by the wants, or supposed
wants, of a climate unlike his own, and a material to which he was
unused; his genius had thus been fettered and kept under; but here all
shackles were undone, and he was free to carry out to its very greatest
perfection what he had learnt or dreamt of in his Northern home. And
what a result has he not achieved!--absolutely and without doubt one of
the grandest interiors in the world is, I do believe, this noble work of
his; its grandeur amazes one at first, and delights all the more
afterwards as one becomes on more intimate terms with it, and can look
at it with less emotion than at first. And how shall I describe it?--for
to say that it has so many bays in length or in width is not sufficient;
all this, and even the detail of its design, were familiar enough to me
before I saw it, but still the reality was so very far beyond any
description, that I felt, and feel still, averse from attempting it.

A few only of the most noticeable points, therefore, shall be touched;
and, as it seems to me to teach less of Italian art or architecture than
of Northern, I shall feel myself acquitted of neglect, inasmuch as the
main, if not the sole, object of my Italian notes is as to the
development of really Italian work. I was struck at first by the
prodigious width and height of the building. The nave is enormously
wide, and has two aisles on either side, those next to it being also of
great size and height, and having clerestories in their outer walls.
The outer aisles are lighted with large traceried windows, filled with
stained glass, which gives the church a character very unlike that of
the generality of Italian Gothic churches, in which coloured glass is so
rarely seen in large masses.

There is, therefore, a regular gradation in the heights of the five main
divisions of the church, which are well proportioned to their respective
widths; and, resting as these divisions do upon four rows of clustered
columns of immense size and height, a more magnificent internal effect
is produced than I can recollect even any approach to in any other
church; for not even in Köln or in Amiens is there any effect so
magnificent as that of this forest of prodigious piers. They are
finished at the top with capitals peculiar to this building, and quite
unlike anything I have before seen--I suppose at least twelve feet deep,
with a kind of arcade of tracery surmounted by a crocketed gable on each
side, and finished above and below with courses of foliage, and with
figures standing in the niche-like panels of the arcade.

Such a contrivance would be clumsy and absurd if attempted on a small
scale, but here, at the summit of these immense columns, it is
thoroughly successful, and, I believe, the only contrivance which could
have been successfully adopted. The capitals vary very much in detail
and in merit; but one of the finest is seen close to the west entrance,
and recalls forcibly the best Northern pointed work, having indeed,
neither in detail nor in design, one single trace of an Italian
influence.

So grand are the columns that the excessive poverty and lightness of the
arches which divide the nave from the aisles, which is perhaps the
greatest defect of the interior of the church, is not for a long time
noticed; they have, however, but little apparent work to do, and so
their lightness may perhaps be a virtue.

Throughout the interior there is a very remarkable love shewn everywhere
for a moulding which was never used by Italians, but was so much used
among ourselves--the wave-mould. Then, again, all the shafts are
filleted, and the fillet turning off gradually into the round line of
the shaft, produces the same moulding; a fulness and richness are
consequently very noticeable in all the interior, very different indeed
from anything else that I saw in Italy. The solitary blot upon this
otherwise noble work is one for which its architect is in no way
responsible--the cells of the groining are all filled in with painted
imitations of elaborate traceries in brown colour, an abominable device,
which never ceases to offend and annoy the eye more and more every time
it is observed.[72] The window tracery throughout is meagre, confused,
and unmeaning, and the traceries introduced at mid-height most
unsatisfactory; but the glass with which it is filled, though poor and
late in its character, contains much rich colour, and gives the entire
building a very grand and warm tone.

Compared with other foreign churches, there is in this--the largest of
them all--a singular absence of side altars. There are indeed some three
or four; but so great is the prominence given to the high altar in the
choir, that one scarcely thinks of the possibility of the existence of
any others, and has to look about a good deal before discovering where
they are to be found.

In common with so many of the Italian churches the pavements here are
very fine, and impart no small degree of additional magnificence to the
interior.

Many are the works which might be catalogued in this great church, but
I must content myself with noticing one only--the great bronze
candelabrum. This is, I suppose, the most precious work of its kind in
Europe. It was placed in the cathedral in 1557, a date too late for its
execution by some three hundred years. The supports are four winged
dragons, heads downwards, and with tails intertwined. The shaft has
bosses in the centre, and branches out above into a profusion of
branches full of exquisite leafage. This is all conventional in
character, and entirely like thirteenth-century work. In spaces formed
for the purpose and on the bosses are various figures and subjects all
wrought with singular skill, and a lavish profusion of ingenuity which
cannot be too much admired. Among these are the Temptation and the
Expulsion from Paradise, the signs of the zodiac, figures illustrating
architecture and music, and a variety of others.

After I had seen the cathedral I had, by way of enjoying the pleasures
of a contrast, to devote myself patiently to all the troubles attendant
upon personally procuring my passport. At Milan personal application was
absolutely enforced; and there, in a small room crowded by vetturini and
other Italians of the lower order, relieved by one or two unfortunate
travellers like myself, I had for nearly two hours to endure heat,
noise, and all the other annoyances of such a work, before I succeeded
in getting my passport _visé_, with permission to abide three days only
within the walls. This system is certainly most vexatious to an
Englishman; but at the same time, if people will make a point of keeping
the rulers of a city constantly on the _qui vive_ with plots, or rumours
of plots, I do not see what other resource their rulers can have. Some
idea of the state of the population may be drawn from the way in which
the soldiers patrol the streets, in companies of three or four, during
the whole day, giving an impression of a place where men keeping guard
singly would be in danger; and then again the great number of the
troops at the gates and elsewhere throughout the city certainly presents
very much the impression of a place whose normal condition is one
perpetual state of siege.[73]

Next to the cathedral the great architectural attraction of Milan is the
famous church of Sant’Ambrogio, which has equally as great claims upon
the antiquary and upon the ecclesiologist as upon the architect.

Entering from the west, it presented a most striking and, to me, most
novel effect. In advance of the church is an open atrium, surrounded by
a cloister of Lombard-Romanesque character, the columns having quaintly
and stiffly carved capitals of stone, and the wall and arches being
built of mixed bricks and stone.[74] Three arches, open from the atrium
to the west end of the church, and above them three other arches of
similar plan, and arranged in triplet fashion, that in the centre being
the highest and widest, nearly fill up the great flat pediment of the
church, on either side of which rise towers--that on the north divided
into stages by means of arcaded strings, like most Lombard belfries;
that on the south perfectly plain and rude, and perhaps therefore of the
very early date at which some antiquaries appear to place the building
of the main structure of this portion of the existing church.[75]

This arrangement, borrowed though it may be from heathen days and civil
buildings, is nevertheless uncommonly satisfactory; it serves to prepare
the mind for the entrance to the church itself, and, instead of the
abrupt transition commonly made from the world to the holy places, here
the intermediate atrium gives time and space to throw off all worldly
thoughts, and to enter entirely into the religious feelings proper in
such a place.

The entrance by the west doorway is of great interest, for in its very
ancient doors are inserted some still more ancient fragments, said by
tradition--and there is every reason to believe it to be true--to be
portions of the gates which S. Ambrose shut against the Emperor
Theodosius.

The interior has suffered much at various times, either by repairs or
alterations; and as most of the groining is of pointed character, the
first impression is that of a very low, simple, and solemn-looking
pointed church of early date. Upon more close examination, however, it
is found that the main walls throughout are Romanesque, and that this
groining was subsequently inserted, and again in later times
strengthened and supported by great piers and arches of Classic
character.

The plan, after passing the atrium, is a nave of three bays, each
subdivided into two by arches into the aisles, a lantern over the altar,
and three eastern apses. There are no transepts. The most striking
object in the interior is the magnificent Romanesque baldachin above the
high altar. This is supported on four marble shafts, and has a
semicircular arch on each face, with figures and foliage in the four
flat pediments or gables which finish it above. Three flights of five
steps lead up to the altar from the north, south, and west, and the
whole is protected both on the east and the west by high metal screens.
Here is a splendid shrine for the relics of the tutelar saint, of the
same age as the early church, which is however so jealously concealed
that I have never seen it.[76] The front of the altar itself is very
magnificent, executed entirely in metal, and containing subjects from
the life of Our Blessed Lord and from that of S. Ambrose. A modernized
dome rises above the baldachin; and behind this an apse decorated in
mosaic upon a gold ground very grandly finishes the interior of this
interesting church.

[Illustration: 58. BALDACHIN. S. AMBROGIO. MILAN. p. 326.]

On the north side of the nave is a very curious pulpit, coeval with the
church and remarkable for its carvings, and for a Roman sarcophagus
which occupies the space between the columns which support it.

From Sant’Ambrogio I made my way under a burning sun to what I expected
to find a very interesting church, that of Sant’Eustorgio. I was,
however, very much disappointed; the interior is abominably modernized,
though still retaining enough of its old Romanesque features to be
intelligible if carefully studied, and remarkable for many ancient
monuments on its walls. It has nave, aisles, and chapels beyond the
aisles, the whole groined, and there is a prodigious ascent to the
choir, which is raised upon a crypt.

That for which it ought to be visited is the exquisite monument to S.
Peter Martyr, executed by G. Balduccio da Pisa, in 1339. It consists of
a sarcophagus supported on pilasters, in front of which are statues
representing virtues. Few works of the kind have ever been executed, in
which the skill of the sculptor has been more happily united with that
of the architect than in this. It deserves all praise.

Sta. Maria delle Grazie, which I next visited, is a church known
generally as that in the refectory attached to which is to be seen all
that remains of L. da Vinci’s painting of the Last Supper. I know not
how much this may have suffered within the last few years, but really,
when I read the kind of remarks which I so frequently see about it, I
cannot help fancying, rather strongly, that they have perhaps been
written before instead of after seeing the veritable picture. It is in
fact in the last stage of decay, with scarcely any of its colouring or
drawing intelligible; and has probably been entirely repainted since
Leonardo’s death. Visitors who go to admire and do admire the Leonardo,
might do worse than in examining and admiring the extremely fine and
fairly well-preserved earlier fresco or distemper painting by
Montorfano which covers the opposite wall of the refectory, and
contains a good and busy painting of the Crucifixion, painted in 1495.
The church is of very late pointed date, entirely of brick, with a large
and ugly dome added by Bramante. The nave arcades are pretty
good--pointed arches springing from the square Classical-looking
capitals of equally Classical-looking columns; very much as in the
church of San Francesco at Venice. Elaborate brick cornices and the
usual sham front leave the same kind of impression on my mind in respect
to this church that it has of all late Italian pointed work in
brick--one of a tasteless, unreal, and unsatisfactory school of art.

South of the cathedral there is a fine late brick campanile attached to
the church of San Gottardo, which rises from behind some of the great
public buildings in which the city abounds. This is a very elaborate
work, octagonal in plan and covered with arcades one above the other,
and finished with a low spire. It is a fourteenth-century building at
the earliest, but in spite of this most of its arches are semicircular.
It is certainly a rich and picturesque tower, and well deserves
inspection.

The remaining churches in Milan seemed to be all Classical, of different
grades of merit and size. There were indeed some very late examples of
brickwork of some value, but really, save the cathedral, there is not
much architectural art to be studied or dwelt upon in Milan. The
cathedral, too, teaches little; its main office is, rather, to prove the
consummate beauty and magnificence attainable by the pointed style,
carried out severely and simply on the very grandest scale, and this its
interior does triumphantly beyond all cavil.

A visit to Milan had always been looked forward to by me with great
interest: first, from curiosity as to the real effect and merits of the
Duomo; and, secondly, from a longing

[Illustration: 59. PIAZZA DEI MERCANTI. MILAN p. 329.]

to see the magnificent Sposalizio of Raffaelle, which is the gem of the
collection in the Brera; and this famous gallery was therefore one of
the first objects of my curiosity. The careful examination of the
pictures which adorn its walls was, however, when we were there, much
hindered by an exhibition of modern Italian pictures, hung in the same
rooms as, and in most cases in front of, the old works. We were able,
fortunately, to get a fair view of what I believe to be not far from the
greatest work of one of the greatest painters in the world--the
Sposalizio being in a room unoccupied by other pictures and unmolested
by the modern exhibitors. The man who could so paint at the age of
twenty-one must, assuredly, have been almost matchless, for never have I
seen a painting more thoroughly noble and delightful, in every way
recalling to mind, it is true, in every figure the manner of his master,
the great Perugino, but not the less enjoyable on that account.

The modern pictures were almost invariably worthless, and shewed no sign
of any revival parallel to that which I trust I am not too sanguine in
believing that one sees at the present day in England.

In the Piazza dei Mercanti is a much-altered building of the thirteenth
century (its date is said to be 1228), which looks in its arrangements
like a Palazzo Publico, and which in its original state must have been
very charming. On the lower stage are five open arches, which have been
modernized, I believe; above this a line of square panels, inclosing
shields, forms a bold string-course, the central portion being brought
forward on corbels for a Ringhiera. The _piano nobile_ consists of five
pointed arches carried on delicate shafts, and above was a deeply sunk
line of arcading, each division garnished with a statue. There was once
evidently a canopy over the Ringhiera. The materials of this front are
black and white stone, but they are used with such moderation that
there is nothing at all bizarre in the general effect of the façade.

In addition to these buildings there is still to be described another
very grand brick domestic building of late date--the Ospidale
Maggiore--which contains a great deal of very rich detail, half
Renaissance and half Gothic in its character, though the general scheme
of the building is wholly Renaissance. This building has been much
extolled, but I think those who have praised it so much have mistaken
clever manipulation of detail for good architectural design. The
terra-cotta enrichments with which it is so richly set are hardly
surpassed in their way by any of the same period in Italy, but I cannot
admire the building as a whole.

With this building my architectural notes in Milan must end, but I
should advise all students of architecture to include in their visits
one of about half an hour’s drive to the Certosa of Chiaravalle just
outside Milan. The church here has been much modernized, but over the
centre of the crossing still rises a brick lantern and steeple of
singular interest. My engraving will explain what the character of this
work is, better than any words. The construction is singular. Behind the
base of the second stage of the great octagon a spire is constructed
which carries the upper steeple, and the whole of the walls pierced with
the second, third, and fourth series of windows are really only
screen-walls or parapets in front of the spire. The height of the whole
lantern from the ridge to the base of the spire is about ninety feet,
and the effect of the complicated brickwork is not bad. It is somewhat
difficult to say exactly when it was built. The Certosa itself was
founded in 1135, and consecrated in 1228, but my impression is decidedly
that though the whole steeple is built with round arches, it is not
really a work of earlier date than about A.D. 1370 to 1400. The steeple
of San Gottardo in Milan, which also has round arches everywhere, and is
in some other

[Illustration: 60. CERTOSA OF CHIARAVALLE. p. 330.]

respects somewhat similar in detail to this, dates from 1339, and it
would be a great mistake to argue for an earlier date in either case
from the mere use of the round arch. If I recollect right, the low third
stage in height at Chiaravalle is modern.

Milan was on the whole rather disappointing to me in my architectural
capacity, though pleasant enough in every other; and after I had lounged
and driven about, first in one direction and then in another, and had
really enjoyed my last great Italian city very much, finding that little
more was to be done but to eat ices, look at smart carriages on the
Corso, and long for more chance of a clear view of the Alps than the
hazy sultry weather afforded, I made up my mind to leave earlier than I
had originally intended.

No architectural student should turn his feet homeward from Milan
without having first of all visited Vercelli. It is easily reached by
railway, passing on the road Novara, where there was only a few years
back a fine Lombard cathedral, which has unfortunately been lately
supplanted by a modern Italian fabric, even more than usually vapid and
uninteresting, and where there is still an old baptistery so plastered
and painted as to have lost almost all its old interest. The traveller
has therefore to content himself with the views--which become better as
he proceeds--of the snow-capped Alps (including their noblest peak,
Monte Rosa) to the north of the railway, which, in clear weather, are
most glorious in their effect.

The cathedral at Vercelli is modernized, and I believe not worth
visiting; I confined myself to the remarkable church of Sant’Andrea,
which is fortunately close to the railway station, and of unusual beauty
and interest. The interest is historical as well as architectural. The
church was built by Cardinal Guala de’ Bicchieri, who had been in
England as legate from the Pope at the very beginning of the thirteenth
century, and is said to have brought back with him to Vercelli a French
or English architect. The evidence of the building itself is in favour
of a French rather than an English influence, but neither is felt
anywhere save in the interior, and the outside views shew to my eyes no
trace of any but an Italian hand. It is the square-ended choir probably
which has made some writers say that it was designed by an English
architect. This choir is short, and on the east side of each transept
are two chapels, which are apsidal. The whole church is groined; the
columns between the nave and aisles are well clustered, and all the
mouldings and details are well and skilfully drawn. The church is well
designed to suit the climate, all the windows being small except at the
east end, where there is a fine triplet, with a circular window filled
with well-designed tracery above. Over the crossing of nave and
transepts is a raised lantern, groined, and lighted by very small
windows high up. The angles under it have extremely well-designed
pendentives with carvings of the Evangelistic emblems in the centre.
Brick is used for the main portion of the work, counterchanged in many
parts with stone, and the proportions and details of the interior are so
good that I found myself in the rare state of mind (in an Italian
church) of admiring without grumbling! The dimensions are good without
being imposing, the total length being a little over two hundred feet,
and the width across the transepts about a hundred and twenty feet.

The exterior has a great bald west front with three round arched
doorways, and a false gable between two small but lofty flanking towers.
The walls are arcaded under the eaves, and over the crossing rises an
octagonal lantern, which is gathered in after a rather ungainly manner
above the lowest stage. This steeple is finished with a low circular
brick spire adorned in the most curious fashion with small

[Illustration: 61. SANT’ANDREA. VERCELLI. p. 332.]

circular brick pinnacles--one over each side of the tower--which look
extremely like a collection of chimneys.

Detached from the church, and standing at an angle to it, is a simple
campanile of later date (1399), and of four stages in height. The
combination of this eccentrically placed tower with the rest of the
church is very remarkable, and appears to have been simply a caprice.
Its effect certainly does not warrant the sort of admiration which
should lead any modern architect engaged in studying the church to
recommend the copying of the relative positions of it and its
bell-tower. I ought not to forget one feature--the buttress--which is
treated here in quite the French or English fashion, with bold
projection and good steep sloped weatherings instead of in the usual
Italian fashion as a mere pilaster.

North of the nave of the church is a good cloister, on the east side of
which is a fine square chapter-house, divided by groining piers into
nine bays. The original triple entrance to the chapter-house--a door
with window on each side--still remains, and there is a communication
also through a groined sacristy with the north transept. I know few
churches which shew more just sense of the best treatment of a good
Gothic interior than this does. In its original state, when the
brickwork was exposed in its natural colour, the effect was of course
much better than it is now; but the effect is still so good that I may
safely assert that, were the exterior equal to the interior, there would
be few more beautiful churches in Northern Italy.[77]

There is an interesting monument in the south-east chapel, corbelled out
from the wall and decorated with sculpture and painting. In the gable is
the Coronation of the Blessed Virgin, below a figure kneeling before
her, and said to represent the architect of the church, who died in
1246, being Abbat as well as architect. Immediately opposite the church
is a hospital founded by the same cardinal. It has been much altered,
but there are still some ancient portions, which I could not get leave
to see.

From Vercelli we retraced our steps to Milan, and halted no longer than
was necessary before going on to Monza on our way to Como. Here we were
well rewarded and most agreeably surprized. We found a very curious and
good Broletto, a cathedral of fine and elaborate brickwork with a great
west front of marble, and another brick church of most elaborate detail.
The west front of the Duomo is a very fine example of Italian Gothic in
marble; it is divided into five divisions in width, those in the centre
and at the sides being the widest, and is constructed in yellow and dark
grey marbles in alternate courses, the former very deep, the latter
generally shallow, but varying without much rule.

All the roofs are flat, but finish at the west at different levels, and
not in one continuous slope. The eaves have heavy cornices, and under
these, all the way up, is an arcade resting upon shafts supported on
corbels. The windows are all filled with traceries which are certainly
not at all equal to English tracery, as they are very flat in their
effect, and have no proper subordination of parts. There is a large rose
window in the centre division treated in a better manner than is usual,
and set within a square line of moulding, with small circles in the
spandrels, and a line of square panels on each side continued in the
most unpleasing way above; five other smaller circular windows are
similarly treated. In some parts of the wall the courses of black marble
are continuations of the black arch-stones of the windows, which, though
not uncommon in Italian pointed work, is never satisfactory in its
effect. In the upper part of the front this is not the case.

[Illustration: CATHEDRAL--MONZA.]

The central division has a porch resting upon detached shafts, and with
a semicircular arch, which is, however, richly cusped; and throughout
the front semicircular and pointed arches seem to have been used quite
indiscriminately. The buttresses which divide the front were originally
finished with pinnacles, of which one only now remains; this is
certainly very beautiful, of precisely the same type as the pinnacles on
some of the tombs of the Scaligers at Verona, standing on detached
shafts, with gables on either side, supported on trefoiled arches, and
with small pinnacles between the gables, all of which are crocketed; the
mouldings are very flat, but in the pure white marble seen against the
deep blue sky of Italy this flatness is as much a virtue and a beauty as
its counterpart executed in stone in chilly England would be
poverty-stricken and tame. All the remainder of the exterior of the
Duomo is of red brick, with some particularly good detail. I give one
window from the south side of the choir as an example.

There is a large low cloister on the north side, and from this the
central tower is best seen; it is an octagon of two stages in brick and
stone, a good deal arcaded, and has a pyramidal tiled roof, with a
square turret in the centre. This forms a dome internally, which is
however (as is the whole of the church) miserably modernized.

The cornices under the eaves here are of brick of good detail, but of
the common arcaded character. I noticed an inscription on the east end,
referring to the erection of the church in 1390, a date which tallies
very well with the character of most of the external detail. Internally
there is a rather remarkable pulpit or ambon on the north side of the
nave. It is a gallery measuring some twenty feet east and west,
supported on detached shafts with a projection in the centre, and of
combined Gothic and Renaissance detail; so large a pulpit certainly
suggests rambling discourses.

On the whole, such a church as this is very interesting, and to some
extent striking; but, much as I admired the conjunction of the two
marbles, and the more than usually Gothic character of some of the
details, there was yet enough in the ungainly outline of the sham front,
and in the capricious use of round and pointed arches indifferently, to
damp my pleasure, and make me cease to admire the work very decidedly.
There is a difference in construction which ought to be noted between
this marble front and the marble work at Venice; for, whilst the latter
is not at

[Illustration: 62.--BROLETTO, MONZA. Page 337.]

all constructional, this is entirely so, the marbles not being veneered
on to the wall, but forming a portion of its substance. I need not say
that in this respect, when we wish to use marble in England, the paucity
of our supplies will probably always compel us to imitate the Venetians
rather than the architects of Monza, Como, or Milan.

The sacristy here is worth visiting. Among other relics are two
fifteenth-century chalices of unusually good workmanship. They appeared
to me to be by a German rather than an Italian hand.

[Illustration: BROLETTO--MONZA.]

The Broletto, which stands near the cathedral in the centre of the city,
is very interesting. It is raised upon open arches of stone, two at
either end and five at the sides. In my sketch of it the southern end is
shewn with the projecting Ringhiera in the second stage; the northern
end is very singular, the tower rising out of one side, with the
steep-pitched roof of the other half abutting against it. The detail of
the windows is very good, the arch-stones in some of them increasing in
depth towards the centre, with an effect of very great strength. All the
windows are shafted. The dimensions of this building are forty-two feet
from east to west, and sixty-four feet from north to south.

The only other ancient building which I could find was the church of
Sta. Maria in Strada; the most elaborate example of late work in brick
and terra-cotta that I have anywhere seen. The effect is not
satisfactory; for when, as here, carvings are imitated and repeated in
terra-cotta, and traceries entirely executed in it, one begins, I
confess, to long much for a little of the fire and spirit which some
mark of the individual artist might have given such an amount of
elaborate decoration in stone. The west front is the only part of the
church of any interest, the interior having been thoroughly modernized,
and retaining no traces of its original character.

The door and windows in the lower stage have been interpolated, and
besides this there is a strangely ugly window above them, about
which--as this is the last of its class we shall see--I wish to say a
word. In starting on a continental journey, between London and Croydon
on the South-Eastern Railway you used to pass under several great
semi-circular arched bridges. When first built, the engineer chose, in
order to gratify some odd fancy, to prop these up by two piers of
brickwork, dividing the arch into three, and putting the whole in great
jeopardy. It is curious that this singularly supported and divided arch
finds a counterpart in almost every large church in the North of Italy.
It was the one great idea of the Renaissance builders, and, until they
had taken out one of the old windows and inserted in its place one of
these hideous contrivances, they were never satisfied. In Venice every
church, even the noble church of the Frari, has them, and I believe
scarcely a large church in the North of Italy is without at least this
one evidence of the delicacy of taste which characterizes the
Renaissance age!

The skill which is shewn in making and fitting together brickwork such
as that in the front of this church is very great indeed, but, after
all, I fear, rather mistaken, for the effect is most unsatisfactory, and
every one must see that throughout the façade there is an evident
attempt to satisfy the eye by the exceedingly elaborate character of the
detail, rather than by the fitness of the thing itself, or by the beauty
of the proportions. The insufficiency of the windows for the extent of
wall is an obvious fault, and not less so is the fact which I am almost
tired of referring to, that the whole front is sham and designed without
any reference whatever to the wants of the building, to which it forms
the street-front.

In the evening we left Monza for Camerlata, a village within about a
mile of Como, availing ourselves again of the Milan and Como railway.

The Camerlata station was soon reached, and after some little delay we
found ourselves ensconced in one of those long omnibuses so fashionable
in Italy, and driving down a long hill, planted on either side with
trees, towards Como. Above us, to the right and to the left, we could
see, by the bright moonlight, the shapes of the mountains which hem in
this arm of the fairest of lakes; whilst just above us, proudly perched
upon a crag, were the ruins of a castle, which lent, when we saw it by
daylight, an additional charm to the otherwise beautiful view.

Soon we were in the outskirts of the town; but it was long before we
reached the borders of the lake, after following the windings of an
almost interminable street, passing the guardhouse, and, to our sorrow,
parting again for the last time with our passports, then crossing the
piazza in which stand the Duomo and the Broletto side by side--for me
the main attraction of the place--until at last we were fairly
discharged at an hotel on the very edge of the water.

We had heard an Austrian band as we rolled across the piazza, and so
without delay thither we returned in time to hear the last of Austrian
music, and to revel by moonlight in the beauty of the many-coloured
marble front of the fair Broletto. We stood listening to the music for
about a quarter of an hour, when suddenly a word of command was given,
the men who held the lanterns marched to the front, the band formed
behind them four abreast, the lights were extinguished, and, suddenly
breaking into a lively march, the band disappeared, and the crowd soon
left us in quiet possession of the piazza, whose old houses still rang
with the wild and clamorous echoes of the beautiful music.

We returned to our inn, and had infinite trouble in the attempt to find
a voiturier to take us to the Lago Maggiore in time for the steamer on
the morrow. The route by which I had fully intended to return was to go
by the lake to Menaggio, thence to cross to Porlezza and by the Lake of
Lugano, and then from Lugano across the Monte Cenere to Bellinzona; to
my great annoyance I found, however, at Milan that, owing to the
long-standing quarrel between the Emperor and the Confederation, no
travellers were allowed to pass immediately from the Austrian into the
Ticinese territory, or _vice versâ_, and we were obliged, therefore, to
defer seeing the beautiful Lake of Lugano, to go instead to Laveno, and
thence to Bellinzona, by the Lago Maggiore; and our difficulty at Como
was to find any true account of the time that it would take us to reach
Laveno, or of the time at which we were likely to find the steamer to
Magadino. In the end it was decided that we should start at seven the
next morning, and accordingly soon after five I was out in the piazza
taking notes and sketches of my last Italian building, the Broletto of
Como.

In general character this is somewhat similar to the Broletto at
Bergamo, but in real beauty it is scarcely inferior to any one building
I have seen in Italy. Towards the

[Illustration: 63.--BROLETTO. COMO. Page 340.]

piazza it has four arches on the ground story, which is divided from the
next stage by an arcaded string-course. This second stage has three
windows only over the four arches below; and another very noticeable
irregularity is, that one of these windows, and that not the central,
has a pedimental canopy above its arch, and has more shafts than the
others. The central window has been modernized to some extent, but this
was the Ringhiera, and the balcony still remains, though looking more
modern than the rest of the front. Some of the arches of these windows
are very noticeable; for though they are semicircular, the back of the
stones which form them is cut with a different sweep, so as to produce
an outer pointed line, and thus to leave an impression on the eye of
absolutely pointed windows. Another arcaded corbel-table finishes the
façade, or rather ought to finish it, for above this some barbarian has
added another stage, nearly to the destruction of the effect of the
building.[78] North of this façade a great plain tower of rough stone
recalls to recollection those of Bergamo and Brescia; it boasts of an
immense clock and some faint traces of painting, and is left unfinished
at the top. The whole of this façade (with the exception of the
campanile) is built of red, white, and dark grey marbles, which are very
carefully and effectively contrasted in their arrangement; the courses
are very irregular in their widths, and apparently arranged upon no
systematic rule. The opposite (east) side of the Broletto is very
similar, but one of its windows is remarkable for the way in which the
shafts are knotted together in the centre. This is not at all an
uncommon feature in Italian pointed, and I have often wondered how it is
that the eye is not at once disgusted with it, instead of being, as it
usually is, pleased. I take it to be a justifiable device on some such
ground as this: it takes much labour and skill to cut several shafts out
of one block of marble, but all this labour and skill is unthought of,
if they are entirely separated, or held together by a band which might
perchance be made of some other material; this knot therefore is devised
as the only means of explaining to us that the shafts so carved have
really been accomplished with a very great expenditure of time and
patience and skill, and do not depend upon any artificial band for the
firmness with which they are all united in one. The capitals of all the
columns in this Broletto are very well carved.

By the side of the Broletto stands the Duomo, the bad character of whose
west front, even though it is of late Gothic, hardly tempted me to go in
to see the effect of the interior. I did so, however, and found a large
but uninteresting church, with groining of pointed section, which gives
considerable character to an otherwise insipid work. The west front has
doorways of Lombard character, and above them a large rose window; but
every part of the exterior and interior seems to have been so much
altered that little remains of the original work.

Internally works of restoration were going on, and these permitted me to
see that the whole church had a great deal of colour introduced on the
walls and over the groining, though I was unable to ascertain anything
satisfactorily as to its age or character.

About ten minutes’ walk from the cathedral is the fine Lombard church of
Sta. Maria. This has unfortunately been much modernized, but its east
end with an apse arcaded outside, and finished with a fine eaves-cornice
rich in shadow, is still extremely striking and almost unaltered. It is
built of black and white stone. Here, alas! I remember that I thought at
the time of my first visit, ends my hurried study of Gothic architecture
in Italy. But if at that day it was

[Illustration: 64. STA. MARIA, COMO. p. 342.]

somewhat sad to leave Como after an all too rapid journey, it has been
my happiness to revisit the quaint old town again and again since, and
each time with increased pleasure. Italian scenery, Italian art, Italian
travel, afford some of the happiest recollections of well-spent days of
travel, which, if they have never been able to exceed in pleasure or to
approach in profit the remembrances of travels in my own and other
lands, undertaken for the same purpose, are nevertheless full to
overflowing with lessons in art which no true architect could afford to
despise or wish to forget.




CHAPTER XIII.

    “And now farewell to Italy--perhaps
     For ever! Yet, methinks, I could not go,
     I could not leave it, were it mine to say
     ‘Farewell for ever!’”
                   _Rogers._

     Departure from Como--Varese--Lake of Varese--Italian
     Boatmen--Intra--Laveno--Lago Maggiore--Magadino--Road to
     Hospenthal--The Dazio Grande--Airolo--Hospenthal--Ascent of the
     Furca--Valley of the Reuss--Lake of Luzern--Luzern--The Unter
     Hauenstein--Strasburg.


There was great delay in leaving Como; the passport officer was asleep,
and no one dared to awaken him for our convenience; at last we
determined to start, and went off to the passport office, and, after
waiting nearly half an hour, the dilatory clerk arrived, and our
passport having been stamped with the “Buon per partire,” so uncivilly
glad to get rid of you as it seems to be, I mounted the carriage, and we
were soon on our way.

All Como was astir, and bedecking the houses and churches, and building
triumphal arches across the roads, for some religious fête whose nature
we did not discover; but we soon left its streets and hills behind, and
began to look out anxiously for our first view of Monte Rosa and its
attendant Alps; but, alas! the weather, instead of clearing, rapidly
became more and more cloudy, and ere long we felt that we must give up
all hope of getting even the most distant glimpse of the monarch of
this portion of the Alpine chain.

Without this view, from which we had promised ourselves so much
pleasure, the road is tame and uninteresting all the way to Varese,
where we changed our horses and carriage. It is an uninteresting town,
with a good many villas and gardens, belonging, I believe, to
inhabitants of Milan, who come out here for the mountain air. None of
their houses are free from that general look of dreariness and lack of
care which seem to afflict most Italian villas. Passing through Varese,
we soon saw on our right a very famous pilgrimage church crowning the
summit of a considerable hill, and approached by a succession of
chapels, somewhat as in the still more famous pilgrimage church of
Varallo, and so popular that round it there seems to have grown up a
small town for the accommodation of the pilgrims.

Farther on we passed the lake of Varese, and from one point in the road
had a view of no less than about five different lakes, one of which was
Lago Maggiore. The Lago di Varese is a tame, uninteresting sheet of
water, surrounded by low flat woody country, except at one point on the
north, but even there the hills do not rise immediately from the lake.

The only approach to old buildings that we saw were one or two brick
campanili of early date, and the remains of a castle, near Varese,
finished at the top with the favourite forked battlement.

We had much ado to make our driver understand our desire to reach Lago
Maggiore without delay, and, to say the truth, there was something too
much like cruelty in the attempt to compel our poor steeds to any such
feat of speed and strength as the performance of some six miles an hour
really appeared to be. As we neared the lake the scenery improved; and
woody hills, with here and there a dashing streamlet finding its way
down the hill-side, and a glimpse now and then of the blue water of the
lake, made the way pleasant. At last we reached the outskirts of the
village of Laveno, and were immediately chased by all the male
population of the place, who explained their eager pursuit when at last
we stopped on the beach, by vying with one another for the privilege of
conveying us across the lake to Intra, where we had to join the steamer.
I asked their charge, and they rather astonished me by demanding twelve
francs and a buono-mano; of course I blandly offered them five francs,
much to their disgust, and with shrugs of their shoulders and grand
looks of contempt they turned away. However, I was determined not to
submit to so palpable an imposition, so, when I was having my passport
_visé_, I asked the courteous passport-officer what the fair charge
might be? “Four and a half lire,” was the answer. “But how am I to
compel the rascals to take me?”--“Oh! bring them up to me,” he replied;
so down I went to where all the boatmen were discussing together the
atrocity of my offer, and, taking two of them by the arm, I quietly
walked them up to the friendly officer; the rest followed, and then
commenced one of the most amusing scenes I ever witnessed. The
passport-officer told them to take me for the four and a half lire, upon
which, they all, standing with their right arms extended towards him,
answered with a furious volley of Italian ejaculations, quite
unintelligible to me, but sufficiently absurd when contrasted with the
quiescent state of their antagonist. Their eloquence was, however, all
in vain; for, after a short attempt to reason them into submission, my
friend sent them off, and threatened to send a soldier with us if they
did not start at once. Before I could reach the beach again the luggage
was all in the boat, and in another minute we were afloat, propelled by
three sturdy fellows, who, after having tried in vain to make me pay
fourteen lire and a buono-mano, were really not apparently much annoyed
when I paid them the legal fare, being about one-fourth less than at a
guess I had first of all offered! They were evidently true philosophers.

I fear that my experience of travelling in Italy obliged me to look upon
the proceedings of these men as by no means unusual or peculiar to
boatmen; wherever you go it is the same, and, unless you wish to pay
much more than the rest of the world ever thinks of paying, you must
make a point of disputing hotel accounts, shop charges, and voiturier’s
charges; the result always is, that you pay about twenty per cent less
than you otherwise would, and are evidently looked up to with infinitely
more respect.

About half an hour sufficed to take us to Intra, the Sardinian port
opposite to Laveno, just a glimpse being obtained of the famous Isola
Bella as we crossed. A Sardinian soldier welcomed us to his liberal
Majesty’s dominions, and, as we told him that we were going on by the
steamer, allowed us to go into the town without shewing our passports.
There was, however, nothing to see, except the pretty view of the
opposite hills--they are scarcely mountains--and of the long sheet of
water stretching up and down for many a mile, and commanded almost more
completely hence than from any other place in its whole extent.

We dined at a very miserable inn, with a pretty lookout, and, as it
happened to be a _jour maigre_, could get nothing fit to eat; the
landlord took, however, a convenient view of the matter, and, assuring
us that he never made any difference on this account, charged us as
though we had eaten all the delicacies of the season. Here, again, as I
had time on my hands, I amused myself with lowering my host’s demands,
and finally paid him a fair valuation for a very _maigre_ dinner.

This business was no sooner satisfactorily finished, a sketch of the
opposite coast having been secured during the argument, than the steamer
arrived, and in a few minutes we were ploughing our way along the fair
expanse of water, leaving the not very honest people of Laveno and Intra
behind us and forgotten, all our attention being devoted to the
gradually developing beauties of the upper end of the lake.

We were amused at Laveno by the warlike demonstrations of the Austrians,
who had there a very smart little war-steamer for the protection of
their interests on the lake, besides a small fort. They had, in fact,
less territory on the lake than either Sardinia or Switzerland; luckily,
however, for the peace of the water, these two last states seemed not to
think it necessary to keep up a rival force, and so the little
war-steamer at Laveno remained, untouched and uncared for, and her
officers passed their lives in smoking cigars and longing for some
change of place and duty, which was provided for them at last by the
abrupt conclusion of Austrian rule in these parts a few years after the
time of this visit.

Our steamer kept very much to the Sardinian shore of the lake, and, as
there are two or three great bends in its course, a view of one small
portion only of the lake is obtained, until, upon reaching a promontory,
and rounding it, as it were a new lake and new scenery are disclosed;
and happily, as the water is ascended towards Magadino, each turn brings
more beautiful scenery than the last, until, as the head of the lake is
neared, the view is very grand--not equal, certainly, to the head of
Lake Como--but still exceedingly beautiful. The sun was just setting as
we reached Locarno, and then our steamer, skirting along the sedgy
shores of the lake--where the Ticino, at its entry, brings down a
continually increasing deposit of mountain refuse--brought us in a few
minutes to Magadino.

Our principal companions on the steamer were a large party of English,
whose travelling-carriage and horses blocked up half the boat, and a
very pleasant old Italian woman, whose elaborately neat hair and
magnificent array of pins, each filigreed at the end, and all radiating
like arrows from a central knot of hair, through which two larger and
more magnificent pins were passed horizontally, forced upon our
notice--but not more strongly than it had been forced before--this
wonderful smartness and elaborate treatment and get-up of the hair, so
common among the middle and lower classes in the North of Italy, and so
unlike the customs of a similar class in England: I am bound to say,
though, that the result of the elaborate straining and dressing of the
hair seems generally to be, that, by the time women are fifty, they have
no hair at all, or, at best, some two or three stray locks, which are
then brought carefully together, and tied up, with a bold disregard of
effect, in a knot at the top of the head.

When we landed at Magadino we found a diligence waiting, and, securing
the _coupé_, jumped in, and were soon trotting off rapidly on the road
to Bellinzona; in little more than an hour we passed through its gateway
without having our passports demanded (how pleasant a change after
Italy!), and were soon comfortably ensconced at the very respectable
Albergo dell’ Angelo.

We started from Bellinzona very early the next morning, determined, if
possible, to surmount the worst part of the road and to sleep at
Hospenthal, on the northern side of the pass. The view of Bellinzona on
leaving it is very striking; three old castles perched on crags above
give it an air of picturesque antiquity, and these, with the mountains
rising grandly on either side of the Ticino, and sloping down in the
distance to the bosom of Lago Maggiore, make a most beautiful picture.
The situation is not, however, to be compared to that of Chiavenna,
whose wall of mountains, clothed with Italian luxuriance of foliage, is
pierced here and there with a chasm only, for the passage of some
headlong river dashing down into the broad valley below the town; here
the valley continues to be of considerable width for some miles above
the town, whilst there one scarcely sees in what way any road is to
escape across the mountains.

The first portion of the road is not very interesting. The pass of the
Bernardino soon turns off to the right up a valley which allows a
partial view, and from this point the S. Gothard road is sole possessor
of the valley. Our first change of horses was at Bodio; and from thence
the road gradually became much more beautiful. Many churches are seen
scattered here and there on the summits of the inaccessible-looking
mountains on either side of the valley, all of them whitewashed and
generally distinguished by their tall campanili, and sometimes by the
small cluster of houses and the patches of cultivated ground around
them, betokening man’s labour as well as man’s religious love, on the
summit of these forbidding-looking steeps. And whilst the distant
prospect was so fair, the scenery close to the road was embellished by
vineyards and magnificent chestnuts, growing in some places among great
rocks shivered from the mountain-side above, and, in others, in groves
on either side of some beautiful stream descending in a silver fall over
the grey precipices which overhang the road.

The villages through which we passed were pretty and picturesque, and
the villagers all very busy in the fields bringing in their hay, and
gathering their grapes, which are always trained here over rocks and
roofs in the most picturesquely irregular way; and altogether the
valley, rife with so many signs of industry and activity, bore
thoroughly the appearance rather of a Swiss than of an Italian district.
The upper slopes of the mountains, on either side, were clustered with
fir-trees, and the deep blue water of the Ticino, here gently murmuring,
there hastily dashing over some rocky impediment, made grateful music in
our ears and imparted additional beauty to the way.

At Biasca and Giornico there are ancient churches, the exteriors of
which are, however, of no interest; though the interior of the latter,
with its crypt and curious paintings, well deserves a passing visit; but
besides these all seemed new, and the houses as well as the people and
the scenery soon began to remind us of Switzerland. There were those
particularly large well-to-do looking inns in every village, with white
walls and windows resplendent with green wooden shutter-blinds which are
so common throughout this country; and here and there were to be seen
houses with a display of well-carved or craftily-framed woodwork, which
gave proof of our rapid approach to the land _par excellence_ of
carpentry.

But it was not till Faido had been passed, and the increasing barrenness
of the hills, the entire absence of vineyards, and the only occasional
appearance of some grand old chestnut-tree, weatherbeaten and rugged
from conflict with many a storm, or, may be, some frightful inundation
such as the Ticino loves at times to indulge in, shewed how rapidly we
were rising into mountain regions, that the scenery became really
striking. Then the road seems suddenly to arrive at the end of the
valley, but presently as we advance, a narrow gorge in the mountain is
perceived, and we enter this, the most magnificent portion of the Val
Levantina, called Dazio Grande. The road is admirably engineered,
carried through two or three short tunnels, and in excavations in the
rocks above the torrent; the dark blue water leaps from rock to rock,
and here and there dashes down in a fine waterfall; and the scenery is
altogether so striking that, on the whole, I am much inclined to give
the preference to this portion of the valley--that is to say, from the
commencement of Dazio Grande to within a short distance of Airolo--over
any portion of similar length in the whole course of the Splügen. The
first narrow defile passed, the valley opens out again, and, with
occasional glimpses all the way of the old road winding below near the
margin of the stream, and destroyed some years since in a storm, ere
very long we reach another defile as beautiful as the last, but much
shorter; for here, after crossing the stream and mounting a short
distance, a projecting rock is pierced, the river finds its outlet
beneath through a chasm not twenty feet in width, and then, the valley
opening out again, Airolo is seen just before us, and beyond the little
cluster of houses which marks the village rise the mountains, so grandly
and abruptly closing in the head of the Val Bedretto, up which our
course now lies, whilst every here and there on their rugged sides or
summits some snowy peak or glacier edge tells not uncertainly of their
grand elevation.

We arrived at Airolo by about two o’clock, and here we had a rest of an
hour and a half for dinner, followed by a ransacking of a collection of
Swiss woodwork, ending--as such an operation always does--much to the
advantage of its proprietor.

With fresh horses we were soon on the road again, and now the weather,
which had been unpromising and occasionally wet, seemed inclined to
improve, and we commenced the real ascent of the mountain under rather
more promising circumstances than we had at first anticipated. The road
soon leaves the river, and, turning to the right, winds and twists about
in the serpentine fashion known only to Alpine roads, and quite
incomprehensible until one has seen them, keeping the church and village
of Airolo in view, first on the right, then on the left, for hours. Here
and there a straight bit of road gives hopes that the zig-zagging is
over, but the thought is no sooner expressed than it is contradicted by
another ascent worse than before, and one begins to envy the electric
telegraph carried here in straight lines from point to point, where one
would have thought it impossible to gain footing for its supports, and
giving fair idea of the directness and speed of the communications of
which it is the channel. The head of the Val Tremola, as the valley
along which the road finds its way is called, is nearly reached, and the
last glimpse of the mountains, at the head of Val Bedretto, is caught,
when a stream is crossed and the last flight of zig-zags is commenced;
these are both numerous and intricate, and as one looks down upon them
from above, their interlacings produce a most singular effect. At last,
however, these are surmounted, greatcoats and plaids are in requisition,
and we all begin to feel uncomfortably cold. The cold grey colour of the
wild mountains of granite, great blocks from whose sides strew the
ground thickly on either side, seems to harmonize well with the scene,
and when presently we pass the Capuchin hospice our driver tells us that
we are at the summit. Two or three dark deep-looking pools or tarns
stand close to the hospice, and reminded me in their gloomy and cold
aspect of the tarn which gives so much character to the hospice of the
Grimsel. The same kind of scenery accompanied us on the now rapid
descent; the sun went down, and the stars were soon out shining
brilliantly upon the mountain road, when at last a sudden turn brought
us in sight of lights, and then, descending a few zig-zags, we saw below
us the roofs of the houses of Hospenthal, and, in less time than it
takes to describe, were standing on the steps of one of the best inns
even in Switzerland, the Goldner Löwe, and superintending the unpacking
of our goods.

In such an inn as this everything proves forcibly that one is in
Switzerland; the rooms are all very clean and very small, and there is
a certain homely air about everybody and everything which is the
especial charm of the better class of Swiss country inns, and in which
they excel, perhaps, all but the very best English inns of the same
kind.[79]

We spent an amusing evening, having for companions a Frenchman with his
wife and two daughters, all very lively and exceedingly loquacious: the
walls of the modest _salle à manger_ rang with hearty laughter until
after the time at which early travellers generally go to bed, and so we
paved the way doubtless for a hearty night’s rest.

The first thing to be done in the morning, after the discussion of the
excellent trout and honey put before us, was to take a stroll up to the
old castle which lends so much picturesque character to the village. The
weather was glorious; the perfectly blue sky overhead, the bright green
of the valley, the luxuriance of the lower slopes of the mountains, and
the view up the pass of the Furca closed in with a white line of snow,
combined together to make us all regret our determination to push on
rapidly for Luzern; and no sooner was the regret felt than--like idle
school-children enjoying themselves while they may--we made up our minds
to ascend the Furca, sleep on the summit of the pass, and return early
the next morning. No sooner said than done; our horses were taken out of
the carriage, and in half an hour, with a guide and horses for the
ladies, we were on our way for a mountain excursion, full of that
elastic feeling which the treading of a Swiss mountain-path always
gives, and bent upon enjoying ourselves to the full.

The contrast with the flat dusty roads and the sultry weather to which
you are so often forced to submit in Italy made the walk especially
pleasant; and though, compared to many other mountain excursions, it
was of slight interest, under the circumstances it presented more than
common attraction to us. The path was one of those pleasant ways so
common in Switzerland--a paved narrow road between banks of fields or
low walls, gradually rising and falling, now crossing the dry bed of
some glacier torrent, and now bridging the stream which descends the
valley to feed the Reuss. The fields were rich in colour, and bright
with various and lovely flowers, and the lower slopes of the mountain
were tinted a rich purple with the bloom of herbs, cropped gratefully
here and there by small and melancholy looking sheep.

The small village of Realp is soon reached, and then the ascent begins;
this is rather stiff, and it has taken us, when we reach the summit,
just four hours and a half of hard walking from Hospenthal. We found
dinner going on at the little hostelry at the top, and, after partaking
of it, started again to ascend the Furca-Horn, a mountain rising above
the summit of the pass, and, as we had been told, quite worth the
trouble of the ascent. There was no kind of path, and in places the
mountain-side was so steep that I began to think it was no place for
ladies to scramble up; however, they thought otherwise, and after divers
tumbles in the snow, and surmounting rather formidable-looking
obstacles, we reached the summit at last, and, sitting down on the edge
of a great rock, spent a long time in enjoying the glorious view.

Just under us was the vast glacier of the Rhone, and then beyond it we
looked down the long valley of the same river until its shape was
obscured by mist, and traced the path by which we had walked in a
previous journey up the steep Meyenwand to the Grimsel. Immediately in
front of us were the vast peaked mass of the Schreckhorn, the whole
course of the glaciers of the Aar, and the peaks of the
Finster-Aarhorn, the Jungfrau, and the Mönch; above our heads rose the
Galenstock, and opposite us the Mutthorn and Monte Fiudo; whilst the
summit of Monte Rosa, discerned with difficulty among a marvellous array
of distant peaks, completed one of the finest views of snow-covered
mountains which it has ever been my good fortune to behold.

Long time did we keep our elevated seats, scanning again and again the
glorious panorama, and at last, most unwillingly, commenced the descent;
this was more difficult, though much more speedy, than the ascent, as
the side of the mountain was both steep and slippery. We reached the
summit of the pass, itself about eight thousand three hundred feet above
the sea, in little more than an hour, the ascent to the Horn having
occupied about two hours and a half; and here we found our French
friends of the previous evening, who had in vain endeavoured to follow
us in our ascent, but had been one and all obliged to give up the
attempt.

Late at night we all went out again to look at the most glorious
moonlight effect it is possible to imagine; the peaks of the mountains
and the vast fields of snow or glacier lighted up by the bright light of
the moon had a charm about them peculiarly fascinating.

Very early the next morning we started again on our way back to
Hospenthal, and got down to the inn in good time, had breakfast, and
then, mounting our carriage, we were soon off again down the valley of
the Reuss. The Devil’s Bridge was ere long reached, and the glorious
scenery with which it is surrounded amply redeemed the expectations I
had always formed of its extreme beauty; indeed for grandeur, combined
with luxuriant cultivation of the lower slopes of the mountains, and for
the wild beauty of the course of the river itself, nothing even in
Switzerland surpasses the narrow valley through which the turbulent
Reuss finds its way from Andermatt to Amsteg.

Here the valley widens considerably, and orchards full of fruit-trees,
covered with bright-looking apples, spread half over the valley, on each
side of which the mountains are very grand in their outline. Before long
Altdorf is reached, and all the scenes so dear to Swiss freemen are
rapidly passed, until at last our carriage sets us down on the very edge
of the lovely Lake of Luzern, where the half-hour which we have to wait
for the departure of the steamer is spent in attempting a sketch of the
rocks which descend so precipitously into the deep recesses of the lake.

Much harm is done by overpraising beautiful scenery, and even the Lake
of the Four Cantons suffers from this; for so much has been said and
written about its unmatched loveliness and grandeur, that the result is
perhaps a slight disappointment with the reality. One great beauty, no
doubt, is the succession of entirely distinct views which different
portions of the lake afford, though at the same time the irregular
outline of the water very much diminishes its apparent scale; I doubt,
too, whether there is any one view so grand as that at the head of Lake
Como, though otherwise I know no lake which can be preferred to it.

On the voyage down the lake, Tell’s Chapel and the famous Grütli were of
course seen; a distant view was caught of the old cradle of Switzerland,
the little town of Schwytz, with the grand peaks of the Mythenberg
rising proudly behind it; the Righi, and the black-looking Pilatus, were
each in turn passed; and as evening drew on, the flat shores of the
lower part of the lake were neared, and presently the spires and turrets
of Luzern came in sight. In a few minutes the bustle of landing being
over, the immense Schweizer Hof received us within its capacious walls,
just in time for one of those accommodating late tables-d’hôte so
acceptable after a long day’s travel.

Three or four hours spent the next morning in strolling about served
only to convince us that there was not very much of architectural
interest in the city itself. A line of old wall, broken at short
intervals by picturesque and irregular towers, and a very long covered
bridge across the lake just where the Reuss runs out of it, are the only
noticeable features. The bridge is ornamented with an immense number of
oil-paintings--two to each principal rafter--illustrative of the history
of the place and country, not valuable as works of art, but curious in
themselves, and giving much additional interest to the structure.

The principal church has two western towers and spires; the latter are
of metal and managed in the way so common in this part of the world,
though never, so far as I know, attempted in England,[80] with the
angles of the spire over the centre of the cardinal sides of the tower.
The whole of the rest of the church is modernized; and there is a
singular modern cloister, which nearly surrounds the churchyard, and
contains an immense array of graves and grave-crosses.

We left Luzern at eleven for Basel, in the diligence, and had a very
pleasant ride over often-travelled ground, of which, therefore, the less
said perhaps the better. The rich luxuriance of the crops, the careful
farming, the vast barns, and the great loads of produce which are
constantly met upon the road, remind the traveller more of England than
any other portion of the Continent ever does. The Lake of Sempach was
soon passed on our right, and at last a pause was made in the good
old-fashioned way, for a very comfortable dinner at the little old town
of Zofingen. At Aarberg the dashing Aar was crossed, and soon the
ascent of the Unter Hauenstein range was commenced; and here we enjoyed
the most extensive of all the day’s views of the Alps--the last and
grandest. Gradually as the summit is reached peak after peak is seen
rising up above the mist which shrouds the lower slopes of the
mountains, their white outlines tenderly relieved against the blue sky:
we recognized one after the other almost the entire range of the
grandest mountains in Europe, seen before nearer but not in fairer
guise, but wherever seen leaving the same lasting impression upon the
mind. Suddenly we overpassed the summit, and began rapidly to descend
the northern slope of the hills; but the last link that bound us to the
land in which we had been voyaging, as we hope, not for the last time,
is never to be remembered but with affectionate regard, touched, as
every one must be, in viewing such a panorama, with the extreme glory of
the scene.

We had now done with mountains and with mountain scenery, though the
road was still interesting and very pretty, and at last late in the
evening we reached Basel. The moon was shining brightly on the Rhine as
we went to our beds for the last time, in this journey at least, in
Switzerland.

We were amused, on our way to Strasburg, by the comparative
insignificance in our minds of the chain of the Vosges, which, on our
outward journey, had impressed us as really very striking in their
outline. So much for the effect of a recent acquaintance with grander
mountains!

Strasburg Cathedral was visited, not for the first time and with a
consequent increase of pleasure. Such magnificent architecture as that
of its most exquisite nave is truly refreshing after Italian work, and,
small as its scale is compared to that of Milan, it in no way lost its
effect upon my mind. I was particularly struck by the vast difference
between the delicate art shewn in the design of the traceries, in the
softly rounded contour and dark recesses of the mouldings, and in the
vigour and beauty of the carving of all the capitals, heightened as they
are by the flood of coloured light let into the interior by its immense
windows filled with some of the noblest stained glass in Europe,
compared with that shewn in the rude traceries, heavy carving, and
plainness or absence of moulding, which characterize almost all Italian
Gothic work. No more strong or decided example need be desired of nearly
all the points of contrast between the best work, north and south of the
Alps, than this, the first great northern Gothic church seen on the
homeward way, presents, when compared with all the work which has
nevertheless been studied with so much pleasure and advantage on the
southern side of the Alps.

It is only fair to say that the first impression produced by the west
front of Strasburg was one--felt, indeed, before, but much more strongly
now--of the smallness of scale and narrowness of the whole. I have not
at any time had any especial love for this front, but, just after seeing
the simple unbroken façades of Italian churches, with their grand
porches and their simple breadth of effect, there is something so
entirely destructive to all repose in a front covered as this is with
lines of tracery, panelling, niches, and canopies in every direction,
that it leaves, I confess, a painful feeling upon the mind, of the
restless nature of its designer’s thoughts. But this is true only of the
west front, for, on entering by the door into the nave, all such
thoughts are banished on the instant, and you stand awestruck at the
beauty and solemnity of the art in which hitherto Northern architects
alone have ever approached at all nearly to perfection, and convinced at
the same time that, with such a work to refer to, we need never doubt,
between the comparative merits of Gothic architecture north and south of
the Alps.




CHAPTER XIV.

    “Alas! of thousand bosoms kind,
       That daily court you and caress,
     How few the happy secret find
       Of your calm loveliness!”
            _Christian Year._

   Concluding Summary--Classic and Gothic Architecture--Italian
   Gothic--Shafts--Cornices--Monuments--Cloisters--Windows--Brickwork--Colour
   in Construction--Truth in Architectural Design and Construction.


I propose in this chapter to sum up, as shortly as I can, the
information which I have gathered in the course of my tours in the North
of Italy, on the subject of mediæval art. In doing this I shall have to
remark, not only on the beauties, but also on the failings of Italian
Gothic architecture, and to give expression to the thoughts which arise
in examining its remains as to developments which are possible to us in
the same direction, as well as to suggest some of the lessons which may
be learnt from them.

I think it may be gathered, from what has been already said, that it
will be useless to look for anything like the completeness in the
development of the style in Italy which our ancestors attained to in
England; and this is easily accounted for. In England there were no
Classic buildings to find here and there an admirer, or perhaps a
cluster of disciples, as in Italy; and men worked, therefore, when the
Gothic style was thoroughly established, freely, and in their own way,
and apparently quite untrammelled with a suspicion even that there had
ever been another style brought to perfection by people above most
others civilized and refined in their habits and tastes, and one
moreover which was distinguished by certain broad and strongly marked
lines of separation from the style which they inherited from their
fathers, and practised and brought, as nearly as they could, to
perfection. In England, therefore, in the Middle Ages, we may look in
vain for any evidence of active sympathy with a more ancient and
venerable style than that which was then in the fulness of vigour, life,
and constant development; and consequently, if it be likely that any
infusion of the art practised by the ancients could have aided the
Northern architects in their work, we must not expect to find here any
trace of such assistance or such advantages. But in Italy the case was
far different: the love for the remains of earlier ages was never dead,
but only slept, ever and anon to break forth in some new appropriation
of ancient materials, or some imitation and reproduction of an ancient
form or idea. So in Venice, in the thirteenth century, whilst pointed
arches were being reared by some to support the walls, not only of the
churches, but of the houses also, other hands were busy with the task of
raising aloft those two Classic shafts, with their antique capitals and
detail, which, even to the present day, stand peculiar and
well-remembered features of the Piazzetta of S. Mark; standing proofs,
if such are indeed wanted, that there had been artists in earlier days
whose art was noble and well worthy of the emulation of men in all ages.

And this Classic seed fell into not ungrateful soil; for though there,
as elsewhere throughout Europe, the value of the pointed arch as a
feature in construction--independently, that is, of its intrinsic
beauty--must have been well known, and was boldly recognized where it
could not be avoided, there were nevertheless in a hundred ways proofs
that men still remembered the lessons and traditions of the past, and
used it with a certain degree of caution and unwillingness, and
associated with features which, rightly or wrongly, were at the time
eschewed by all Northern architects, either as being contrary to its
spirit, or through ignorance of their existence. This fact would seem,
therefore, to place the Italian works of the Middle Ages in the ranks of
hybrid and mixed styles, and to debar them from competition with the
more pure contemporary works of Northern Europe.

There will, however, always be much profit in the careful examination of
such works as these in Italy, because their authors stood in the same
position that we do now, and, conversant to some extent with the
beauties of the best Gothic architecture of the North and the best
Classic examples of Italy, took what they deemed best from each, and
endeavoured to unite the perfections of both.

Classic architecture is that of the lintel and impost, involving the
idea of rest: Gothic is that of the arch and the flying buttress,
involving the idea of life and motion. The two ideas are absolutely
opposed to each other. Classic architecture, directly it admits the
arch, ceases to be true to itself in any real artistic sense; yet if it
refuses to use, and to exhibit the use of the arch, it denies itself
wilfully the use of the best known mode of construction. Gothic
architects may still, on the other hand, as they always have done, gain
much from the teachings of Classic buildings. And if sometimes there is
too great liveliness and want of repose in their works, they may
usefully study those of their predecessors who undoubtedly obtained more
breadth and repose in them by some knowledge of Classic examples, than
they would have had if they had not known them.

Gothic architecture was essentially the work of scientific men; the
most consummate skill being displayed in arranging thousands of small
blocks of stones, any one of which might be carried upon a strong man’s
shoulders, into walls rising far in height above anything ever dreamt of
by the Greek, bridging great openings, and providing by the exactest
counterpoise of various parts for the perfect security of works whose
airiness and life would seem to have lifted them out of the region of
constructive skill; and yet all these wonderful works were executed in
materials as ponderous in their nature as those which the Greek had
handled so rudely in construction, and so delicately in ornamentation.

The natural result of this excess of science was, perhaps, that less
delicacy and beauty of detail became necessary; for when the plain rough
walls, without carving and without ornament, were nevertheless of
necessity so beautiful in their intricacy of outline and delicacy of
structure; and when, too, so little (comparatively) plain surface
remained to be looked at or dwelt upon, men cared less for the choicest
examples of the sculptor’s art, and were less obliged to satisfy the eye
with them. Much, therefore, as Art gained in most ways by the invention
of the arch, she at the same time lost something which had been until
then possessed, and which, too, was essential in the highest order of
work.

This was the case, speaking generally; but, as need hardly, I suppose,
be said, there are examples scattered here and there throughout the
North of Europe, and particularly in France and England, which shew
distinctly enough that their artists had grasped this necessity in its
very fullest extent; and were in no degree satisfied with anything less
than the greatest perfection in the sculpture, and other decorations
with which they adorned their works.

Italian architects stood, in the thirteenth and fourteenth centuries, in
a position in many respects very different from that held by their
Northern contemporaries, and the marks of this difference are everywhere
to be seen. It was natural to them to reconcile in their works, so far
as they could, the principles of two styles which we are too prone to
deem irreconcilable; and where they have achieved a real success, it
ought to be a lesson that the course which they pursued is still open to
us, though with larger opportunities and greater knowledge. At the same
time they committed faults which we ought especially to beware of
imitating in any respect.

They ignored, as much as possible, the clear exhibition of the pointed
arch, and, even when they did use it, not unfrequently introduced it in
such a way as to shew their contempt for it as a feature of
construction; employing it often only for ornament, and never hesitating
to construct it in so faulty a manner that it required to be held
together with iron rods from the very first day of its erection. This
fault they often found it absolutely necessary to commit, because they
scarcely ever brought themselves to allow the use of the buttress; and
this reluctance was a remarkable proof of their Classic sympathies.
Classic architecture was as distinctly symbolic of rest as Gothic was of
life; the column and lintel of the one were as still and symbolical of
perfect repose, as the arch of the other, sustained by the strong arm of
the flying buttress, was of life, vigour, and motion. Italian Gothic
architects then, in never resorting to the buttress, avowed their
feeling that a state of perfect rest was the only allowable state for a
perfect building, and they preferred almost always to use the arch for
its beauty only, and avowedly not for its constructional value, which
they evidenced by tying it together with iron bars at its base, which
there was no intention of disguising, and therefore no shamefacedness in
the use of. From this, I believe, at least one beauty arose. It is
obvious that the pointed arch, descending upon the capital of a shaft
without any visible stay or buttress to retain it in its place, would
look weak and thin; and it was soon perceived that, in order to overcome
this difficulty, the only course was to add to its substance by cusping
it on the under side. Thence came the trefoiled arches so frequent in
Italy, and always so very lovely. As so much depended upon them, no
pains were spared in bringing their outline into the very purest form.
To this we owe the absolute perfection which characterizes some of the
trefoiled arches in early Italian work; of this we have an example (to
take one instance among many) in the tombs of the Scaligers, at Verona,
in which the mass of the trefoil descending from the arch conveys to the
eye the impression of a firmness which, in part, it certainly gives, but
which would nevertheless be insufficient in reality for the stability of
the work, without the aid of the connecting rod of iron below.

In northern Gothic architecture, arches invariably tell their own story,
and do their own work, avowedly and without any disguise; in Italy this
is the exception, not the rule, and the commonest exception is seen in
the arcades dividing the aisles from the naves of churches, and in the
large open arches forming the arcades upon which public and private
buildings so often stand, though even these are sometimes (as in the
second stage of the Ducal Palace at Venice) formed of continuous
traceries in place of arches, and are dependent for their stability,
like the single arches of tombs and monuments, upon the assistance of
iron ties.

Another proof of the reverence for ancient tradition in art is furnished
by the extent to which, throughout the time of the prevalence of pointed
architecture in Italy, the round arch was also used. Examples of this
are very frequent, and of the most capricious kind. At Cremona, e.g.,
the south transept has semicircular arches throughout, and the north
transept pointed arches; the date of both being, however, as nearly as
possible the same. At Verona, in the old house given in plate 14, some
of the bearing arches are pointed, some round; and again, in some
buildings the main arches are round, and the ornamental pointed. Again,
one of the most common cornices in all the mediæval brickwork of Italy
is a continuous arcade of round arches intersecting one another, and
forming at their intersections pointed arches; and in the Ca’ d’Oro, one
of the most elaborately ornate late Gothic buildings in Venice, some of
the entrance arches are pointed and some round. It needs not, however,
that examples should be multiplied, for almost every building exhibits
some trace of this kind of confusion.

The Venetian love for the ogee arch, which spread thence to Verona,
Vicenza, and Ferrara, has been already mentioned; no doubt we must look
to the East for the very early introduction of this feature in Venice,
and I only incidentally refer to it here as proving, from the way in
which it was perpetually used, that its adopters had no peculiar love
for the form of the pure pointed arch, of which it is certainly a most
vitiated perversion.

With this, by way of preface, let us now go on to look at the features
of these Italian works in detail. We shall find that we have two
influences brought out, even more strongly than in most mediæval works,
in Italian buildings. These are, first, local, and next, personal
influences. The Venetian, the Bolognese, and the Veronese, for instance,
are all distinctly local styles, in which early traditions were
preserved, to some extent, from first to last; and to these might be
added the Florentine, the Genoese, and the Pisan. On the other hand it
is impossible not to notice the very great personal influence exercized
over their descendants, as well as over their contemporaries, by some of
the greater Italian architects, of whom I may adduce Nicola Pisano as
the most eminent example.

There is a third influence which must not be overlooked--that of foreign
architects. Milan Cathedral, designed by a German, is very unlike any
other Italian building in style; San Francesco at Assisi was also the
work of a foreigner; and the western front of Genoa Cathedral owes much
of its peculiar and extremely beautiful character to contact with, and
knowledge of, French art; whilst Vercelli is another instance, as far as
its interior is concerned, which may be included in this list. Generally
speaking, however, the influence of the foreigner seems to have been
confined pretty much to the particular church designed by him.

In the Middle Ages the Italians led a life more akin to our own, at the
present day, than other people. The country was populous, the cities
numerous and rich, and the people full of emulation and individuality.
This was precisely the condition of things that would lead, most
certainly, to the employment of artists from a distance, and to the
establishment of a professional practice of the art earlier than
elsewhere. It led to the employment of the family of the Pisani, of the
Campione clan, the sculptors of Como, the Cosmati in Rome, and many
others. In England and in France it is much more difficult to point to
any facts proving the employment of the same architects in various parts
of the country, and there is, at first sight, therefore, less of what is
obviously personal in their art in the Middle Ages.

The history of our old architects in England and France is very
peculiar; they were, as a rule, each confined to a certain district, in
which they wrought in what was in fact a merely local variety of their
national style. In point of artistic talent, they were very equally
matched, so that it is difficult to assign the pre-eminence to any one
over the rest, though in England we may point to our Yorkshire abbeys,
and in France to the buildings in the Isle de France as being on the
whole the finest examples. We have had no Nicola Pisano here; our old
architects’ work is singularly equal in its character in each period;
and whether it was displayed in the little village church lying
concealed on the banks of a rippling stream, or in the vast abbey of
some sequestered valley, or in the cathedral church of the busy city,
there seems to be matter for equal admiration in all. In Italy, on the
other hand, we see a number of individual architects exercizing each his
peculiar influence, varying very much in their skill and power, and
having, moreover, the doubtful advantage of a constant recollection of
the works of a different style of art, from whose traditions they never
escaped. Placed, in short, very much in the position that we are at the
present day, they never wrought with the same absolute and joyous
freedom that marks their contemporaries’ work in France and England; and
thus, though their architecture may be inferior, it is of very special
value to us; for we may, perhaps, see better the cause of some of our
own shortcomings, when we investigate theirs, and so we may hope to
excel the works which they executed under conditions so similar to those
under which we labour, even if we cannot quite rival the complete
perfection of the greater mediæval architects of the North. And, seeing
that all the faults of Italian Gothic artists arose from their
incomplete devotion to the new art, and their lingering fondness for the
old Classic forms, we shall be led probably to recognize the paramount
importance of throwing ourselves heartily and entirely into the study
and practice of the one great and national division of our art, and
then, not venturing to attempt to design in some base imitation of
Classic one day, or in a pretended Gothic the next, as is only too much
the custom, we shall make our sense of art so completely a portion of
our inmost selves, as never to do anything new in any but our own one
special style. We shall, in short, recognize, as the greatest danger to
the progress of real art, that eclectic spirit which the Italians never
escaped from, and which, in our own day, leads men to design their work
in the style which they or their clients fancy for the moment, and not
in that which is the truest result of previous experience, and most
fitted to the country in which it is to be executed.

In examining the features of any national school of architecture, it is
worthy of notice how distinctly some of its peculiarities and prejudices
are marked from the very first, even in the ground-plans of the
buildings it produced. This is notably the case in the ecclesiastical
edifices of France, England, and Germany. Each country, after the art
had become settled and, so to speak, acclimatized, had its special
arrangement of plan, which was seldom departed from and was handed on
from age to age as a precious heirloom. And, going to Italy, we shall
find that the same feature strikes us there in almost all the buildings
of the pointed style, and that the buildings from which they were
directly derived are the same as those from which we indirectly derived
our own. Their plans are all derived from two ancient types, both of
which are of venerable antiquity. It was from the basilica, converted
into a church, with its nave and aisles terminated at the end, by an
apsidal projection from a sort of transept, that all of one class of the
Italian Gothic churches with transepts were copied. Indeed, if we look
at the ground-plan of S. Paul without the walls, at Rome, and compare it
with the fully developed church of Sta. Anastasia, at Verona, we shall
see that absolutely the only difference is the addition of small chapels
on the east side of the transept; so that in place of the one apse,
which marks the former, we have the central apse and subsidiary chapels
on each side of it. The church of San Clemente, at Rome, with its three
aisles ended with parallel apses at the east end, is a variety of this
type, followed in such churches as the Cathedral of Torcello, and indeed
in all later Italian Gothic churches without transepts. And even when,
as in the thirteenth and fourteenth centuries, the Italian architects
endeavoured to secure an immensely wide unbroken area of nave, they
still looked back fondly to these old precedents, and finished them at
the east, with three parallel apsidal chapels.

The other model was the Byzantine plan of S. Mark’s at Venice, itself
imported from the East, which was copied closely in the south-west of
France, and produced no slight effect in modifying the simple Romanesque
plan, until it assumed all the characteristics of the complete Lombard
style, and from the North of Italy sent out vigorous off-shoots, of
which the most important were those which we trace all along the valley
of the Rhine, and throughout the centre of France.

Thus, the ground-plans of Italian Gothic buildings were simply a natural
development from those of earlier date, and adhering, as they always
did, very closely to the older plan and arrangements, afford us scarcely
an example of those later developments of prolonged choirs, of which our
English cathedrals and abbeys afford perhaps the most magnificent
examples, or of the splendid French _chevet_ with its surrounding aisle
and chapels. The traces of Classic influence on the plan are indeed so
many and so clear, that it is hardly speaking too strongly to say that
Gothic planning was never developed by Italian architects, so shackled
were they by the ever-present influence of buildings in another style.
Hence, the more we study their peculiarities, the more we see how
curious a mixture there is in them of the character of Classic and
Gothic buildings, and how essentially clumsy all their planning is when
compared with the scientific work of the more thoroughly Gothic
architects of the North. If we compare an ordinary Italian groined
church with a French or English example, we shall find one very marked
difference in them. In the former each bay of the nave is square, and
hence each bay of the narrower side aisles is oblong, with the greatest
width towards the nave; in the latter, on the contrary, the aisle
compartment is square, and that of the nave oblong, with its narrowest
side towards the aisle. Hence, in the former, the points of support are
farther apart, and the plan loses much of its intricacy, and at the same
time, no doubt, the whole building loses much of its apparent scale. The
enormous church of San Petronio at Bologna, for instance, has but six
bays in the length of its vast nave, and the eye refuses to be convinced
by the practical measurement of the foot of the real dimensions of the
building. The object of the Italian architect always was to obtain as
few points of support as possible in a given area; but there is little
if any real gain in this. The points of support in the Italian churches
were larger, and the cost in the end was probably much greater than in
the apparently more intricate and complex plans of the French and
English architects. The science displayed in their planning was
therefore of a superficial and mistaken description, and not really
equal to that which marks the work of their Northern competitors.[81]

The same absence of subdivision is seen in the elevation of each bay of
an Italian church, where in place of the triple division in height of
our great Northern churches, with their well-accentuated proportions and
beautiful variety of detail, we have a singularly meagre design
perpetually repeated, and consisting generally of simple broad arches,
with a small circular clerestory window above them, and no other kind of
decoration, save where the painter has come with his ever-ready art to
the rescue of the apparently incompetent architect.

How this plainness and severity was corrected we have already seen in
the interior of Sta. Anastasia, at Verona--a church which, though it is
simple and unadorned in its construction almost beyond any large church
with which I am acquainted, has been made remarkable, owing to the skill
of its decorator, for being more than usually ornate and magnificent.

The invariable simplicity of Italian groining has been often noticed in
these pages, and need only be referred to here as illustrating the love
of simplicity upon which I have dwelt. No one feature gains so much by
it as this, and I can fancy the horror with which an Italian, or indeed
any other architect, in the thirteenth century, would have viewed such
fine works in their way, as our own examples of complicated
fan-traceried groinings. The painting of this simple groining at Verona,
at Lodi, and still more at Assisi, proves how it ought always to be
treated; and happily in England we have still more than one example of
the same kind, as, for instance, two chapels in the Cathedral at
Winchester.

So far for the plan. When we look at single features of the building we
shall find another peculiarity meeting us at every turn, and of the good
effect of which it is impossible to speak too warmly. This is the
constant use of the detached shaft or column. This is the one great and
lasting beauty which was derived from Classic examples. The Italians,
finding it used with luxurious profusion in their Classic and Byzantine
buildings, persevered in its use throughout the whole Gothic period. It
is true that, as time wore on, there was a somewhat less free use of it
than at first, but it was never altogether ignored. We cannot say as
much for our own ancestors: so long as the influence of Lombard and
Romanesque art is visible in French, German, and English Gothic, so long
the detached shaft was used, and just in proportion as in course of
time that influence decreased, so did the frequency of its use decrease.
Our fourteenth and fifteenth century buildings present nothing in its
place but combinations of mouldings, in themselves very beautiful, but
by no means so beautiful as to reconcile us to the loss of that which
they so entirely supplanted. One consequence of their introduction, to
the exclusion of the detached shaft, was, that the art of sculpture
deteriorated just in proportion as the art of moulding was developed.
There is no place in which architectural sculpture can be more fittingly
displayed than in the capital of a column. It is the most convenient,
and at the same time the most conspicuous, position for it. It is, too,
the most important feature in every design in which the detached column
is used. The gathering together of all the arch mouldings into one above
the capital, in order that their forces may be collected before being
transmitted to the ground, leads naturally to the laying of a special
emphasis on this point above all others; and it is one of the strongest
among the many reasons in favour of earliest Gothic, and against the
later varieties of the style, that in the former the use of shafts
involved the use of forcible and elaborately-cut capitals, so that this
point might be most distinctly marked, whilst in the latter, by the
disuse of the shaft and the constant practice of carrying the mouldings
of the arch down to the ground without any interruption, it was made as
little of as was possible.

In this respect, therefore, above all others, Italian Gothic
artists--having given way to change less than their contemporaries--are
worthy of our highest admiration. The love of variety, which is
characteristic of all good Gothic art, is conspicuous in the Italian
treatment of the shaft just as much as in the northern Gothic variation
of mouldings. When they are plain cylinders, and not banded (and the
band occurs but seldom in Italian work), they often taper slightly, and
with very beautiful effect. This was distinctly a relic of the Classic
entasis, and the examples given in a note will suffice to shew the
extent to which the shaft is reduced in proportion to its height.[82]
Wherever, however, the shaft is spiral or decorated with carving, or
when the occurrence of a band destroys the idea of its continuity from
base to capital, its monolithic character needs not to be marked, and
the shaft is then always made of the same diameter throughout. The
ornamentation is of various kinds. Circular shafts are inlaid, carved,
diapered, or made of marbles selected for their beauty of colour. Lucca
Cathedral affords, perhaps, the best examples of decoration by inlaying:
there the shafts are of white marble inlaid with dark green; some have a
diaper, others are girt with a succession of simple chevrons, others
with spiral lines, others with crosses, flowers, fleurs-de-lis, or
foliated circles; and one, at least, with a succession of imitations of
arcades one over the other; but I remember no example of the same kind
in the buildings described in these pages. At Lucca these inlaid shafts
are frequently alternated with sculptured shafts, of which examples are
common all over Italy. They are deeply cut with spiral lines of
mouldings, occasionally adorned with sculpture of flowers, as in the
doorways at Bergamo and Modena, and are often also inlaid richly with
mosaic.

The best examples of the richer kinds of shafts are seen frequently in
Italian Gothic buildings and monuments south of the Apennines. Such, for
instance, as those which, with rich iron work between them, form the
screen round the altar under the “crossing” of Sta. Chiara at Assisi,
or in the monument of Pope Benedict X. at Perugia, where the shafts are
of white marble, the spiral lines formed with a bead and fillets, and at
the base and neck of the shaft the beads are all connected together with
small arches, in the spandrels of which vine leaves are delicately
carved; whilst to add to the extraordinary richness of the work, small
figures are carved in the marble creeping round the shaft in the hollow
formed by the spiral construction, and the spaces between the mouldings
are filled in with glass mosaic patterns in red, green, and blue on a
gold ground. The shafts of Orcagna’s shrine in Or’ San Michele,
Florence, are of the same character, but these extraordinary decorations
occur generally only in the best and richest internal work, executed
under the direct influence of Florentine, Neapolitan, or Roman
artists.[83]

The arrangement of spiral shafts was generally, but not always,
symmetrical. In the Campanile of Florence, where they occur in pairs on
each side of a window, they are always arranged with the spiral lines
curving in opposite directions. In the porch of Sta. Maria Maggiore,
Bergamo, the three shafts in each jamb are all different: the first
spiral and carved, but not moulded; the second moulded with chevrons;
the third spiral and moulded only, and the central shaft is red whilst
the others are all white. In the spirally designed string-courses of the
Venetian palaces the spiral lines are always worked in opposite
directions right and left of the centre.

Another bold variety is seen in the beautiful coupled shafts at the
entrance to the crypt of San Zenone, Verona, where both the shafts are
quatrefoil in section, but entirely different in character, one of them
being quite straight, the other slightly, but yet conspicuously,
spiral. In a monument in the south aisle of the same church, four shafts
are cut out of one block of marble; and in order that this may be
realized they are knotted together in the centre; and a similar example
exists in the porch, of which I have given a view, at Trent, and in one
of the windows of the Broletto at Como. The shafts round the lower basin
of Nicola Pisano’s beautiful fountain at Perugia are in clusters of
three, chevroned, spiral, and fluted in the greatest possible variety;
and in later works, in which we find that the error of continuing arch
mouldings to the ground, instead of stopping them on the capital, was
occasionally committed--as, e.g., in the archway to the baptistery in
the church of the Frari, at Venice--we see an instance of a moulded
imitation of such a cluster of shafts as these of Nicola Pisano’s,
chevroned and spiral, forming jamb and arch alike. In the later
doorways, such as those of the western front of the cathedral of Como,
and many of those at Monza, Pavia, and in most of the Venetian churches
and buildings, the jambs and arches are identical in section; but even
in these cases there is always a capital interposed, and I hardly
remember an example of a continuous moulding without an impost, which at
least marked--if often in an incomplete manner--the line of the
springing of the arch.

I must not forget that which is, after all, the most charming of all
arrangements of shafts, viz. the use of them in couples, set generally
one behind the other, so that, in elevation, extreme delicacy is
secured, whilst, in perspective, there is beautiful light and shade, and
ample effect of strength. This is a favourite arrangement in cloisters
and in belfries; and, wherever it occurs, one regrets, as one looks at
it, that though in old days French, Spanish, German, and even Irish[84]
architects gladly followed the Italian example, it was so seldom that
an Englishman condescended to do the same.[85]

In the treatment of mouldings the Italian Gothic work is quite peculiar.
An Italian architect would have been surprized, could he have seen the
dark and piquant recesses of the mouldings in which his Northern
brethren in the thirteenth and fourteenth centuries so much delighted.
He rarely, if ever, indulged in a deep hollow, but made a point rather
of shewing the hard sharp outline of the square edges of his stones or
bricks, relieved only by the interposition of simple round members,
alternated with flat hollows; his mouldings, even when designed for such
grand works as the Ducal Palace, being bald and crude in design, though
they conduced no doubt to that breadth of effect to which he always
desired that everything should be sacrificed. There is but little skill
shewn in the way in which their contours are drawn, and the carelessness
with which they are fitted to the size of the capital that carries them
is a constant source of disgust to any one who has been trained by the
study of the exquisite English mouldings of the same period. The
architectural carving was designed with the same idea; for when it was
introduced elsewhere than in the capitals of columns, it was always very
flat and delicate, severe in outline, and not much relieved, and often
very decided in its direct imitation of nature. This, however, is mainly
seen in the earliest examples, for I am bound to say that later Italians
never rivalled the Byzantine capitals of S. Mark’s, or some of those in
the early church of San Zenone, Verona. Indeed, as time wore on, the
carving in Italian buildings became steadily worse and worse. Most of
the later Venetian capitals are bad in their outline, confused and
purposeless in all their lines, and shew no sense of that vigorous
petrifaction of the elements of natural growth and form, which was so
sensitively felt and expressed by French and English artists.

[Illustration: CORNICE--SAN FRANCESCO, BRESCIA.]

And, next, we come to the cornice, the feature which above all others
must most startle men who, for the first time, make acquaintance with
Italian work, and which most recalls in its idea its Classic prototype;
for, though its treatment in detail is as unlike that of the ancients as
it can well be, it is, nevertheless, so decidedly marked and so
prominent a feature (crowning not only the summits of walls, but even
running up the gables, and returning round buttresses), that it is
impossible not to regard it as another evidence of admiration for, and
imitation of, earlier work.

The ordinary northern parapet is never used, the eaves almost always
finishing with the common Italian tiles projecting slightly over the
deep cornice of the walls. We have nothing at all parallel to these
cornices in England, and I remember but few examples of anything of the
same kind in the North of Europe, save in the transept of Lübeck
Cathedral, and such churches as those of Bamberg and the Rhine country;
which last seem to be derived from the Lombard churches of Pavia, and to
have nothing in common with later pointed work, and to have exerted
little, if any, influence on its development.

I have said enough, I hope, to explain the grounds for my opinion that,
with the single exception of their use of shafts, we never find the same
kind of perfection in Italian Gothic buildings as in French or English
works of the same date; but I am not slow to allow, nevertheless, that
they do contain features of extreme beauty and purity; and many of them
peculiar to themselves. There is perhaps no country in the world which
excels Italy in the buildings which were erected for civic and domestic
purposes. The Doge’s Palace, and many of the other palaces at Venice,
the Broletto of Como, the other houses or palaces of which I have given
illustrations throughout this volume, are some only among many which
might be enumerated, any one of which would have not only an advantage
over very many of our own buildings, as a model of good architecture,
but at the same time the merit of fitness for the purposes for which our
domestic buildings in towns are at the present day required. Then again,
there are, as I have shewn, the exquisite porches; the peculiar, and
generally noble, campanili; the many-shafted cloisters; the perfect
monuments; the use of brickwork of the best kind; and, finally, that in
which Italian architecture of the Middle Ages teaches us more than any
other architecture since the commencement of the world--the introduction
of colour in construction--which is managed generally with such
consummate beauty, refinement, and modesty, that oven where it
accompanies faulty construction and unworthily sham expedients, it is
impossible to avoid giving oneself up altogether to admiration of the
result.

It will be seen that I am not by any means a blind enthusiast about
Italian architecture. Who, indeed, that has studied on the spot, as I
have done, not only a vast number of buildings in England, but also
nearly all of the best examples in France, Spain, and Germany, could do
otherwise than profess his truest allegiance to be due to the truthful
beauties of his own national variety of the style? I should think that
most students would agree with me, if they found themselves able to
institute such a comparison.

The first view of an Italian Gothic church, whatever its date, is
startlingly and, I think, disagreeably unlike anything that we are
accustomed to in our own old buildings. You may go to a great English
cathedral and find that from every point of view, inside and outside,
every feature is well proportioned to its place, and beautiful in
itself, whilst the _tout ensemble_ is also perfect in proportion and
mass. This can never be said of Italian work. It never produced anything
perfect both in detail and in mass; and one always finds it necessary to
make excuses for even the best works, such as one never finds necessary,
or allows oneself to think of making, for English works. There is
something really absurd in comparing even the best of the Italian
churches with such cathedrals as those of Canterbury or Lincoln, so
superior are the latter from almost every point of view. The Italian
church is usually a long, broad, rather low building, lighted with but
few windows, having but a small, if any, clerestory, and with scarcely
any irregularity in shape or plan; it has scarcely ever more than one
tower, and this is never combined with the rest of the design in the
manner common to us in England or France. There is no approach,
therefore, to such combinations of steeples as are familiar to us at
Canterbury, Wells, Laon, Reims, or Rouen, and undoubtedly there is very
much less external grandeur. The steeple, when it does occur, is often
detached; and when it is engaged, it does not open into the church, but
is placed in some irregular and abnormal position, where it is at once
felt that it is purposely not intended to be looked at in conjunction
with the main façade of the building. There is no attempt even to secure
a tolerable sky-line. The only relief to the monotonous outline of the
main building is at the crossing, where something in the way of a low,
mean dome is occasionally introduced, but this is always of but slight
elevation, and not intended to produce any good external effect, such as
was aimed at in our own central steeples. So also if we look at their
façades, we have a feature on which, in common with ourselves, they
often lavished considerable expense, and probably the greatest pains.
The treatment is very similar in its idea throughout the whole period
during which the style prevailed; and the effect produced is undoubtedly
oftener very disappointing than attractive. The commonest type of façade
is one in which the cornice, which is generally of slight projection,
but deep and marked in character, is carried up the flat gable of the
building, whilst the whole front, divided by vertical pilasters into
three or five divisions in width, is lighted either by a series of
circular windows, or by one large and important rose in the centre. This
class of front is common to most of the Gothic churches in Lombardy.

At Ferrara, we see another and very different design in the grand front
to the cathedral, which, save that it is entirely and shamelessly a sham
front, might vie even with that of a Northern cathedral in beauty,
intricacy, and richness of character; but this design is not really
very Italian in style, and is a solitary example. When an Englishman
sees the tympanum of the principal doorway of such a church filled with
a sculpture of S. George and the Dragon, he may be pardoned for
recollecting that our royal family have sprung from the same stock as
the D’Estes, once Lords of Ferrara, the front of whose cathedral is
almost the finest in Italy.

Other portions of Italian churches are even less satisfactory than their
façades. I have already explained that the clerestory is rarely lighted
by anything more elaborate than a succession of circular windows of
small size. The church of Sta. Anastasia, Verona, gives us the best
example of these, the windows there being of brick, filled in with very
good early plate tracery in stone. The east end is often more
picturesque than the west; and that of the church of San Fermo Maggiore,
Verona, affords a good example. The east ends of the churches of the
Frari and SS. Giovanni e Paolo, at Venice, are perhaps the two finest
examples of this portion of the building to be seen in Italy, and are
worthy of very high praise.

If we descend from generals to details, we shall find much more to
admire, and altogether much more to interest us. The doorways and the
porches, which protect without concealing them, are often especially
beautiful. I have already mentioned the doorways of Sta. Anastasia and
San Zenone at Verona, of S. Mark’s, Venice, the cathedral at Modena, and
the north porch of Sta. Maria Maggiore, at Bergamo, all of which are
full of beauty and full of national character. Another favourite and
beautiful type is that of the porch on the north side of San Fermo, at
Verona, which is arched on each face, and roofed with a flat-pitched
roof, gabled both ways. In the doorways, inclosed within these porches,
we shall hardly find so much to admire, and must not expect anything
like our own or the great French examples. Generally the opening is
square-headed, with a lintel often formed by an ingenious dovetailing of
stone together,[86] and the mouldings of the jambs are too often
continuous and very small and badly marked in their sections.

The exquisite monuments which so often occur against the walls or by the
sides of Italian churches are somewhat similar in idea to these porches.
Of these I have said so much that I only refer to them here, as among
the most charming features of Italian art. I think that our own
monuments, rich as our country is in them, will only be considered much
superior to the best of these by those whose patriotism warps their
judgment! One of the most perfect examples of the class is to be seen in
the Castelbarco monument close to the church of Sta. Anastasia, Verona.
Here the pointed trefoiled arch, on each face of the canopy over the
tomb, springs from four detached shafts, and fits very closely under
flat gables or pediments, above and from between which rises a perfectly
plain pyramidal mass of stone. The very simplicity of the design of
these monuments is their greatest charm; and so conscious were their
designers of this, that they seem to me to have lavished all their care
and refinement upon them. There are of necessity iron ties at the
springing of the arch; but it was felt necessary to give the eye a sense
of security beyond what the existence of these ties could afford, and
this was accomplished by adding on the under side of the arch a simple
and rather heavy cusp, generally proportioned with a degree of delicate
skill of which modern architects appear to me (if I may dare to say it)
never to dream. I believe that good architecture may generally be
detected at once, by the excellence of such apparently small matters as
the shape of a cusp. I am certain that good Italian architecture
invariably has cusping, which is both nervous in its curve and yet
delicate; and I believe that most modern cusps are drawn by the hundred
as a mere matter of routine, without care and without pleasure, and
consequently without good effect.

Not less beautiful than the porches and monuments is another feature
which occurs in Italy, as often as, and perhaps more perfectly than,
anywhere else--the cloister. This consists generally of an open arcade
supported on detached shafts, and is very frequently of two stages in
height. Notwithstanding their extreme simplicity and moderate amount of
enrichment, these Italian cloisters are always capital in their effect.
They owe this very much to the number of shafts which support their
arches; these are generally of marble, coupled together except at the
angles, where there are usually four. Some of the cloisters which still
remain in Verona, are among the most beautiful examples. Nor are those
of San Stefano, Bologna, nor at a later date those of the church of
Sant’Antonio at Padua and at Brescia, less worthy of study. They never
have the beautiful traceries of our Northern cloisters, for the
sufficient reason, that the climate rendered less protection from the
weather necessary; and the consequence of this is, that their arcades,
being always severely simple in design, no effort was ever spared to
make them as perfect as possible in their proportions.

The cloisters of the cathedral at Aosta are interesting as affording an
instance of the lavish richness of illustration which some of the
mediæval sculptors bestowed on their work; the capitals throughout being
sculptured with illustrations of subjects, all of which are made fully
intelligible by inscriptions incised in the stone--a favourite practice
not only of early Italian sculptors, but of their brethren in Germany
and France. In the cloister of San Gregorio, Venice, we have another
variety in which the shafts support the woodwork of the roof in a very
picturesque fashion, without any arches. This is a type of cloister
which might often be most useful in modern work, and is an evidence of
the extent to which a Gothic artist may, when necessity requires it,
trench boldly on what is ordinarily supposed to be the exclusive
province of Classic art--the use of the shaft and lintel; but here the
Gothic artist with his usual reality made his lintel of wood in place of
stone.

The Italian treatment of windows is especially worthy of note. The
drawings which I have given will shew how generally the tracery,
commonly called plate-tracery, was used. It is, indeed, a very beautiful
mode of treatment, but quite distinct from all fully-developed systems,
inasmuch as it deals only with the piercings here and there in the block
of stone which forms the window-head, and not with the intricate
combinations of lines which mark out the outlines of the spaces to be
pierced. The difference is great--the one kind giving that exquisite
depth and mystery and admitting of the infinite variety so
characteristic of northern Gothic, the other giving breadth and flatness
of effect, and leaving space on its broad surfaces for the play of the
brilliant sunshine, save where the black piercing of some simple
form--quatrefoil or trefoil--gives life to the otherwise monotonous
window-head. Of the two the former admits of infinitely more variety and
display of fancy and ingenuity; though the latter, perhaps, when seen at
its best, is really the more beautiful. Both of them are, however, so
good as to be equally usable, and neither of them to the exclusion of
the other.

In Italian Gothic traceries, it is difficult to shew the progression or
regular development which marks every stage of northern Gothic. There
are numerous examples of simple lancet windows, of cusped lancets, of
combinations of lancets cusped or uncusped, and oftentimes of windows
of plate-tracery, and then of more developed tracery which, however, was
still treated as plate tracery with the addition of mouldings. In the
later windows an unsightly effect is produced by the wide and bald plain
splay or reveal which is usually formed round the window, outside as
well as inside, and also by the placing of glazing behind the traceries
in a separate wooden frame, so that they are completely concealed from
view on the interior. The tracery of the second stage of the Doge’s
Palace at Venice is probably equal to any that has ever been executed,
and may well be the pride of the country. It has also the special
peculiarity, common to all Venetian domestic work, of being sufficiently
strong in its section and construction to bear an enormous weight of
wall without the aid of any discharging arch above it. There is another
class of traceries which seems to have been essentially an Italian
invention, and which is as objectionable as any tracery that I know.
Examples of this may be seen in the south transept of SS. Giovanni e
Paolo, at Venice. They give the idea of having been cut out to order,
from an enormous mass of ready-made tracery, kept in slab, sold by the
superficial yard, and cut to fit any opening required. There is no
attempt to finish the tracery where it meets the inclosing arch, and the
effect is consequently always rude and unskilful.[87]

Another feature is, the constant use at all dates of shafts in place of
moulded window monials; and another, the very frequent insertion of a
transome of tracery, across the middle of the height of the window. The
use of the shaft instead of the moulded monial does not seem to be so
admirable in ecclesiastical as in domestic work. It generally
accompanied the system already mentioned of fixing the glazing in wooden
frames, behind the stonework, and hence seldom looks well in
church-windows except on the outside. The deep transomes pierced with
tracery are of common occurrence. They are seen in almost all the
Venetian Gothic churches, but their utility and beauty are alike
doubtful.

No country affords more frequent examples of circular windows than
Italy. They occur in almost every church, and are of stone, marble, and
brick. The best example I know, of very early pointed character, and one
which is unsurpassed anywhere, is in the west front of the church of
Sta. Maria, Toscanella. Here there is an inner wheel, and the space
between the two wheels is divided by shafts, between each of which is
pierced a quatrefoil. The little window in the gable of the oratory of
San Zenone, Verona, is a favourable example of a smaller window. It is
of eight lights divided by shafts, and well moulded and proportioned in
all its parts. Other circular windows as at S. Mark’s, Venice, and in
the church of Sant’Antonio, Padua, are of the class already described,
where the tracery seems to have been inserted, without reference to, or
connection with, the inclosing circular line.

I have now, I think, exhausted the distinguishing features of an Italian
church, with the exception of its campanile. This, as I have said
before, is always a single steeple (with the perhaps unique exception of
Genoa Cathedral, which has two), generally detached from the main
fabric, planted wherever it happened to be most convenient; sometimes as
at Vercelli, not even at right angles to the building to which it is
attached, and with but little reference to its effect upon the remainder
of the building. There are no features of Italian buildings which are so
universally remembered with pleasure; and similar as they are in their
general scheme, there is nevertheless a considerable variety in their
treatment in detail. The first thing to be noted is, that they are never
supported by buttresses, and that the usual mode of ascent was by a
staircase in the thickness of the wall, and not by any such excrescence
as a staircase turret. The outline is severely square and simple, and,
unlike as it is to most of our English steeples, generally exceedingly
striking. No doubt the apparent height of these towers is enormously
increased by the absence of buttresses. The originals of all Italian
campanili seem to be the early steeples of the Roman churches, divided
into a series of stages of equal height, pierced with windows not
varying much in design throughout. The most exact reproductions of these
steeples may be found so far from the centre of Italy as at Susa at the
foot of the Mont Cenis, and in other very similar erections all over the
North of Italy. With certain broad general features of resemblance there
are, however, several local variations, which may be divided roughly
into the following separate classes or schools:--(1) The Pisan: (2) the
Venetian and Veronese: (3) the Genoese: and (4) the Florentine: of which
I need here only speak of the Venetian and Veronese, which are
undoubtedly among the finest. In the Veronese there is a very distinct
imitation of the pure Romanesque examples, whilst in the Venetian there
is generally less subdivision into stages, the walls being arcaded with
very slightly recessed lofty arcades, and the horizontal divisions being
much fewer in number. Of these, San Giacomo del Rialto is one of the
best examples; it is executed, as all the Venetian examples are, in
brick. In all these examples the pilasters at the angles are carefully
marked. But there is another variety of distinctly Italian towers in
which this is not the case. The finest example of this is the tower of
the Scaliger Palace, at Verona, where the simple unbroken mass of
masonry and brickwork shoots up into the sky all but unpierced until the
belfry stage. Of this kind of tower the examples are not unfrequent in
the smaller churches in Lombardy, and in the public halls and private
houses of many of the cities described in these pages. The tower at Asti
is one of the best of these.

There remain two other subjects which, more than any others, meet us at
every stage of our study of Gothic architecture in Italy. These are,
first, the use of brick; and secondly, the introduction of coloured
materials in construction, of both of which, as we have seen, there are
so many valuable examples.

It has been by far too much the fashion of late years to look upon brick
as a very inferior material, fit only to be covered with compo, and
never fit to be used in church-building, or indeed in any buildings of
any architectural pretension, so that I suspect many people, trusting to
their knowledge of pointed architecture in England, would be much
surprized to find that, throughout large tracts of the Continent, brick
was the natural, and indeed the popular material during the most palmy
days of architecture in the Middle Ages. Yet so it was that in Holland,
in the south-west of France, in Northern Germany, and the Low Countries,
in large tracts in Spain, and throughout Northern Italy, stone was
either scarce or not to be obtained, and brick was therefore everywhere
and most fearlessly used.

In all these countries, just as in Italy, it was used without any
concealment, but each country developed its practice in this matter for
itself, and there is therefore considerable diversity in their work.
They are all unlike, and far superior to what remains to us of ancient
brickwork in England, for I need hardly say that, with a rare exception
here and there, and in comparatively small districts, brick was not used
in England between the time of the Romans and the fifteenth century,
and, when used afterwards, was seldom remarkable either for any singular
beauty or originality of treatment. In this matter, therefore, we are
obliged to go to the Continent for information.

[Illustration: STRING-COURSE--PALACE OF JURISCONSULTS, CREMONA.]

Italian brickwork is remarkable as being almost always executed with
nothing but red bricks, with occasional but rare use of stonework; the
bricks for the ordinary walling are generally rather larger than ours,
in no way superior in their quality, and always built coarsely with a
wide joint of mortar. Those used for windows, doorways, and generally
where they were required to attract attention and to be ornamental, were
made of much finer clay and moulded with the greatest care and skill.
The transepts and campanile of Cremona Cathedral are instances of red
brick used without any intermixture of stone save in the shafts of the
windows, and their effect is certainly very grand. The mouldings are
elaborate, and the way in which cusping is formed singularly successful.
This, it must be observed, was not usually done by means of bricks
moulded in the form of a cusp, but with ordinary bricks, built with the
same radiating lines as those of the arch to which they belonged, and
cut and rubbed to the necessary outline. Sometimes, as, e.g., in the
windows at Mantua,[88] which are some of the very best I have ever seen,
the points of the cusps and key-stones of the arches are formed in
pieces of stone, the alternation of which with the deep-red hue of the
bricks produces the most satisfactory effect of colour. This sort of
treatment is common at Brescia, Verona, Mantua, and Venice, but unknown
at Cremona.

[Illustration: WINDOW IN NORTH TRANSEPT, CREMONA CATHEDRAL.]

In nearly all cases where brick is used for tracery, it is in the shape
of plate-tracery. The tympanum of the arch is filled in with a mass of
brickwork, through which are pierced the arches over the several lights
of the window, and these are supported on marble or stone shafts with
carved capitals, instead of monials:[89] and above these sometimes, as
in the windows of Sant’Andrea, Mantua, are three cusped circles;
sometimes, as in the palace at Mantua, only one cusped circle; or else,
as in a beautiful example at Cremona, the plain brick tympanum is
relieved by the introduction of a panel of terra-cotta, bearing the
cross on a shield, whilst round its outer circumference delicately
treated though large cusping defines the outline of the arch.[90]

[Illustration: DETAIL OF WINDOW-JAMB--CREMONA.]

The windows at Coccaglio[91] and at Monza[92] are examples of tympana
left quite plain, or, as in the former case, pierced only with a small
opening of a few inches in diameter, which nevertheless gives much
effect to the design. In the latter case there is a feature which is
well worth notice, because it is remarkable in the best Italian
brickwork, and always very effective. Labels are exceedingly rare, but
their place is usually supplied by a course of very narrow deep red
bricks which surrounds the back of the arch. In the window in Monza
Cathedral there are two such courses--one about 4½ inches wide, the
other not more than 2½. They serve to define the arch and keep it
distinct in effect from the walling around it. Sometimes, as in the
Vescovato at Mantua, and in the houses at Asti, these narrow bricks are
introduced between a succession of rims of brickwork on the same face,
alternated very picturesquely with squares of stone, and sometimes, as
in some beautiful arcading outside San Fermo Maggiore at Verona,[93] to
define and enliven the lines of stonework; for in this case, though the
work is all in stone and no brick was really required, so great was the
appreciation of colour, that it was gladly and most successfully
introduced. In the early cloister of San Zenone we see it again,[94] as
also in all the very beautiful arches which still remain in the Broletto
of Brescia.[95]

[Illustration: BRICK ARCHIVOLT, VESCOVATO--MANTUA.]

[Illustration: ARCH-MOULD--CREMONA.]

But beside this there was another way in which Italian architects
produced a very beautiful effect: this was in the alternation of stone
and brick. We have one of the first examples of this in the magnificent
walls of San Zenone at Verona, in which a deep red brick is used in
courses alternating with a very warm-coloured stone. The brick is used
very irregularly: beginning at the base of the walls over the cloisters,
we have alternately with courses of stone, first a band of three courses
of brick, after this one course of brick, four courses, five courses,
two courses, one course, and then the cornice, which is mainly of stone,
but relieved by two courses of narrow bricks; in spite of the variation
in the height of the brick courses, those of stone in this case are
nearly uniform in depth. In the west front of San Fermo Maggiore[96] we
have brick and stone used in alternate and regular courses all the way
up; in this case the brick is used rather for colour than for any other
reason, though the side walls of this church are entirely of brick and
crowned with excessively deep brick cornices.

The churches at Asti afford examples of the worst kind of
counterchanging of stone and brick; in them the jamb of a window is
treated like a chess-board, being chequered with alternate red and white
both horizontally and vertically. To accomplish this the construction is
ingeniously twisted, and it need hardly be said that the effect is not
at all good.

The interior of San Zenone, Verona, is lined with brick and stone,
arranged just as it is outside, and the effect is most satisfactory;
indeed, this and the interior of the baptistery at Cremona, still left
in their original state, shew how noble an effect of colour may be given
by brick internally, and how mistaken we are when we cover our walls
with undecorated plaster. I have seen it maintained by men who possibly
had never seen such old works as these in their old state, that such
colour as this is savage and fit only for uncultivated men. If, however,
there is refinement in whitewash or plaster, I am unable to see it, and
I can see no more inconsistency in honestly shewing the real materials
of the wall inside than in doing so outside. This at any rate is beyond
all doubt what these old Italian architects did.

[Illustration: WINDOW--VERONA.]

The east end of the church of the Frari at Venice is another example of
coursing with stone.[97] There, however, the courses are far apart, and
seem to be intended to define the lines of the springing of arches, of
transomes, and the like; and this they do very satisfactorily. But,
perhaps, the very best example of mixed stone and brick is that which we
have in the window-heads of the church by the side of the Duomo at
Verona,[98] in which the arrangement of the two colours is quite
perfect. In this case the cusped head of the light is executed in stone,
not in brick; and this is, I think, as a general rule, by far the better
plan; for if an attempt is made to execute tracery in brick, we have the
example of the Germans before our eyes as a warning. They rarely (S.
Katharine, Lübeck, is almost a solitary exception) used stone in their
window tracery; and as they never developed the kind of brick
plate-tracery which is so characteristic of the best Italian work, they
built windows which were either bald and ugly in their simplicity, or
else, endeavouring to execute elaborate traceries by the use of bricks,
moulded into the forms of component parts of tracery, they produced
what are even more distasteful than any other kind of window; in part
because they consist of an endless repetition of small reticulations,
and in part because they lead naturally to the constant reproduction of
the same window for economy’s sake.

[Illustration: BRICK WINDOW, SANT’ANDREA--MANTUA.]

There can be no doubt that the best windows for brick churches are
either those beautiful Italian developments of plate-tracery in which
all the bricks are carefully cut and rubbed for their proper place, or
those in which, within an inclosing arch of line upon line of brickwork,
a small portion of stone is used for the traceries. And this last has
the advantage of giving much more opportunity for variety of form and
beauty of effect than any brick traceries can ever give.

There is one point in which a curious practical difference exists
between our old work and most old Italian. Here it was not the custom to
have keystones to pointed arches, whilst there it is quite the rule to
have them; this may have been partly, perhaps, because it was a matter
of convenience to mark the central stone in arches composed of alternate
voussoirs of brick and stone, and it may have been partly some relic of
Classic traditions: not only, however, is there a keystone, but
sometimes, as in the Broletto at Brescia, this is additionally
distinguished, above the rest of the stone voussoirs, by some small
ornament carved upon it. With one more fact I think I may end what I
have to say on this head: this is with reference to the mode in which
some of the Italian brick arches very beautifully follow the fashion,
not so uncommon in stone, of increasing in depth as they approach the
centre. In this manner, one sometimes sees an arch whose outer
circumference is pointed, whilst its inner line is a semicircle. This
was a fashion most popular in Florence, and not so common in Northern
Italy; still it is to be seen at Monza, Verona, and elsewhere. The
effect is always very good, and, though quite unknown to our forefathers
in England, may well be introduced in our works, as it gives great
appearance of strength, and is no doubt, at the same time, the strongest
possible form of arch.

No one can deny that the study of Italian brickwork must be useful to
those who are compelled, as we so often are, to use the same material in
buildings for whose good architectural effect and character we are
anxious. But, far as it is in advance of most ancient brickwork in
England, there are points in which we must refuse to follow it; we need
not, for instance, in attempting to rival its beauties, confound them
with faults which were essentially those of the whole Italian style, and
not specially of Italian brickwork. We may with the greatest advantage
emulate the Italian system of brick-window tracery, whilst we take care
never to imitate the equally common custom in Italy of erecting sham
fronts in order to display our traceries. Again, though they never used
any but red bricks, there is no reason why we should not enliven our
work with the contrast of other colours. Germany gives us examples of
green and black bricks, and, indeed, Italy affords some (e.g.,
Sant’Antonio, in Padua, Murano, and Torcello) of a yellowish brick; and,
no doubt, the effects producible by these contrasts of colour are such
as Italian architects were always ready to avail themselves when they
had the opportunity. This their parti-coloured works in marble
sufficiently prove; but at the same time it was seldom that they
ventured upon such works in brick; and as it must be admitted that there
is no sort of work which so much requires skilful handling, or which is
so liable to degenerate into vulgarity, as this, it is probable that
they advisedly abstained from it.

As to the question whether it be desirable or not to introduce brick at
all in ecclesiastical edifices, or generally in public buildings, one
might, a few years ago, have been anxious to say somewhat. I trust,
however, that the ignorant prejudice which made many good people regard
stone as a sort of sacred material, and red brick as one fit only for
the commonest and meanest purposes, is fast wearing out, and that what
now mainly remains to be done is to shew how it may most effectively be
used, not only in external, but also in internal works.

One word only as to its colour, for I think that we ought as much as
possible to insist upon this being taken into consideration. We do not,
as a general rule, I suppose, adopt any material in good works of
architecture simply because it is the very cheapest that can be
obtained; sometimes, indeed, we must, and then I should be the last to
contend against what is simply an act of necessity, not of choice; but
ordinarily, before, for economy’s sake, we determine to sacrifice the
colour of our work, and to use those detestable-looking dirty yellow
bricks in which London so much indulges, we ought to consider whether,
by some economy in other respects, we may not save enough to allow of
the use of the best kind of red brick for the general face of our
walls.[99]

At the present day there is, I think, absolutely no one point in which
we fail so much, and about which the world in general has so little
feeling, as that of colour. Our buildings are, in nine cases out of
ten, cold, colourless, insipid, academical studies, and our people have
no conception of the necessity of obtaining rich colour, and no
sufficient love for it when successfully obtained. The task and duty of
architects at the present day is mainly that of awakening and then
satisfying this feeling; and one of the best and most ready vehicles for
doing this exists, no doubt, in the rich-coloured brick so easily
manufactured in this country, which, if properly used, may become so
effective and admirable a material.

The other mode of introducing colour in construction, by means of the
use of marbles, deserves also some notice. In my notes upon the
buildings as they were passed in my journeys, I have described two modes
in which this kind of work was treated: the first was that practised in
Venice--the veneering of brick walls with thin layers or coats of
marble; the other, that practised at Bergamo, Cremona, and Como--in
which the marble formed portion of the substance of the wall.

These two modes led, as would naturally be expected, to two entirely
different styles and modes of architecture.

The Venetian mode was rather likely to be destructive of good
architecture, because it was sure to end in an entire concealment of the
real construction of the work; the other mode, on the contrary,
proceeded on true principles, and took pleasure in defining most
carefully every line in the construction of the work. It might almost be
said that one mode was devised with a view to the concealment, and the
other with a view to the explanation, of the real mode of construction.

I have already described, at some length, the main features of these old
works in marble, and I feel, therefore, that all that need now be done
is to point out the degree to which they afford matter for our imitation
with the coloured materials and marbles which we fortunately have in
Great Britain fit for the purpose, though not in very great variety.

There appears to me to be a certain limited extent to which we may
safely go in the way of inlaying or incrustation: we may, for instance,
so construct our buildings as that there may be portions of the face of
their walls in which no strain will be felt, and in which this absence
of strain will be at once apparent; obviously, to instance a particular
place, the spaces inclosed within circles constructed in the spandrels
of a line of arches can have no strain of any kind. They are portions of
wall without any active function, and may safely be filled in with
materials the only object of which is to be ornamental. All kinds of
sunk panels inclosed within arches or tracery would come under the same
head; the spaces between string-courses might also do so very
frequently, if, as in old examples, the string-courses were large slabs
of stones bedded into the very midst of the wall, and so capable of
protecting the thin, weak slabs of marbles incrusted between them.

In Venice we have some grand examples, at S. Mark’s, of this system of
incrustation filling in the whole of the space within large arches; here
it is lawful, because there is no weight upon it to thrust it out of its
place or disjoint it, as the least pressure most certainly always will.
So far in praise of the Venetian system. But in other parts of the same
building we have this system carried to a length which I cannot but
think most mistaken, and which, I most heartily trust, may never find
imitators here. In these the arches were constructed in brick, and then
entirely covered with marble. Of course there was some difficulty in
doing this, and the way in which the difficulty was met was extremely
ingenious; a succession of thin slabs of marble was placed round the
soffeit of the arch, having perhaps enough of the cohesion given by the
form of the arch to enable them to support their own weight, and
further supported by metal staples let into the joints of the brickwork.
The edges of these thin slabs projected sufficiently in advance of the
face of the brickwork to allow of their being worked with some kind of
pattern--generally, as has before been said, a sort of dentil--and of
their giving some support to the thin slices of marble with which the
walls were then covered. The whole system was excessively weak; and this
can nowhere be better seen than in the Fondaco de Turchi, where almost
the whole of the marble facing and beautiful medallions, in which it was
once so rich, have peeled off, and left nothing but the plain and
melancholy substratum of brick. Few architects, I should think, would
like to contemplate their work perishing in this piecemeal manner, any
more than they would enjoy the thought of a west front left unfinished,
like that of Sta. Anastasia at Verona, and prepared only for marble with
rough, irregular, and unsightly brickwork.

It would be unjust not to say that often, very often, this system of
incrustation, even when carried to the extreme limits of what seems to
be lawful, wins upon our love by the exquisite delicacy and taste of the
sculptured patterns, worked in low-relief, with which it is covered. The
men who did this work were, perhaps, more of sculptors than of
architects; and certainly it must be confessed that never in buildings
in which the construction is mainly thought of, is there, so far as I
know, so much elaborate thought and skill exhibited in the decorative
part of the work as in buildings such as these.

Sometimes, the sculptured medallions set in the centre of a plain
surface of marble are of exquisite taste and beauty; whilst here and
there, as e.g. in S. Mark’s and in one or two spots in the water-front
of the Ducal Palace, are examples of great beauty, of medallions formed
of marbles of various colours, arranged with great refinement in some
kind of geometrical pattern, which shew another and equally beautiful
mode of relieving plain spaces of walling.

The plain surfaces of the walls in Venetian work were commonly either
entirely inlaid, or else inlaid within a square inclosing border of
projecting moulding. The inlaying was composed of a number of
rectangular slabs of marble, not by any means always of the same size,
supported to some extent by the projections of the inclosing marbles or
by those of the archivolt, but always dependent mainly on metal cramps
let into the fabric of the wall; and, when possible, these marbles were
slabs cut out of the same block, and put side by side, so as to produce
a kind of regular pattern wherever the veining of the marble was at all
positively marked.

The other mode of introducing constructional colour in marble commends
itself to one’s reason as that which is most likely to endure for ages,
and as that, therefore, which ought, wherever it may be, to be adopted.
The first idea of the architects of these buildings seems to have been
to arrange their material with as much regard to strong contrasts of
colour as was possible. The first thing they did, therefore, was to
alternate the colours of every course of masonry, either simply as in
the Broletto at Como[100] and in Sta. Maria Maggiore at Bergamo,[101] or
as in the west front of the cathedral at Cremona, where very narrow
layers of white marble are laid between each of the other courses, which
are of course so much the more defined.

The description which I have already given of these works, as well as of
the porch at Bergamo, will shew how regular is the way in which this
system of counterchanging the colours was carried out by the purely
constructional school; this is, in fact, the great mark of difference
between the constructional and the incrusting school of Italian
architects, the whole arrangement of coloured materials by the two
schools being quite different, and producing singularly different
results. The most common fault of the Venetian system of incrustation
must have been that upon a general surface of plain wall you had here
and there a square patch of marble surrounding a window opening; that of
the other system would be, in the opinion of many, that you have too
stripy an effect of colour, and that all the divisions, moreover, are
horizontal.

The former must certainly have been the case wherever the incrustation
did not extend over the whole surface of the walls, which was very
frequently the case; but the latter is not really a fault; it was only
an elaboration in a more beautiful material of the same system which we
have seen pursued with such happy results by the builders of Verona in
brick and stone,[102] and which we find adopted by the architects of
Northern Germany in the frequent alternation of courses of red and black
brick, and sometimes by our own forefathers in the coursing of flint
work with stone, or in the counterchanging of red and white stones which
we see in some of the Northamptonshire churches. The system was a
thoroughly sound one, because it not only proceeded from and depended on
the natural arrangement of the material, but afforded the best possible
means for displaying the various colours which were to be used.

Probably all these systems are mainly useful now as shewing us certain
principles which we may work out and apply to our own somewhat different
circumstances; and surely one of our first objects ought to be the
discovery of the extent of our means and opportunities, which in this
matter are at the present day far beyond what is generally imagined.

It must never be forgotten by us that our forefathers had very limited
means of obtaining materials from one locality and transporting them to
another; and were moreover, to a great extent, unacquainted with the
materials which might, if necessary, be obtained. We have not this
excuse; we not only know what materials we may obtain, but we have at
the same time marvellous facilities for their conveyance between all
parts of the country; and we know also how much has been done of old in
other countries by using them in a proper way.

No excuse therefore can be found for us if we continue to neglect to
avail ourselves of them as though they were still undiscovered. We have
alabaster, which may be wrought at a really trifling expense; large
fields of marbles, of which those of Devonshire and Ireland are
particularly valuable for architectural decoration, and those of
Derbyshire and Purbeck for the formation of shafts and columns: we have,
moreover, an exhaustless supply of granites of various colours; of
serpentine; and, lastly, of building-stones of many tints, many of which
may be very effective when contrasted. In addition to these natural
materials we have every facility for making the most perfect bricks;
and, owing to the excellence already achieved by our manufacturers of
glass and pottery, we have no difficulty in making mosaics and tiles,
either for roofing, flooring, or inlaying, of any degree of beauty,
either of colour or form.

With such advantages we ought long since to have effected far more than
we have ever attempted, or apparently ever thought of. Our buildings
should, both outside and inside, have had some of that warmth which
colour only can give; they should have enabled the educated eye to revel
in bright tints of nature’s own formation, whilst to the uneducated eye
they would have afforded the best of all possible lessons, and, by
familiarizing it with the proper combination of colour and form, would
have enabled it to appreciate it.

And then, if ever the day shall come when our buildings thus do their
duty and teach their proper lesson to the eye, we may hope that we shall
see a feeling, more general and more natural, for colour of all kinds
and for art of every variety in the bulk of our people. At present it is
really saddening to converse with the majority of educated men on any
question of colour. For them it has no charms and no delight. The
puritanical uniformity of our coats and of all our garments is but a
reflection from the prevailing lack of love of art or colour of any
kind. A rich colour is thought vulgar, and that only is refined which is
neutral, plain, and ugly.

Perhaps in all this there may be something more than art can ever
grapple with; it may be ingrain, and part of the necessity of the
present age; but if so, oh for the days when, as of yore, colour may be
appreciated and beloved, when uniformity shall not be considered beauty,
nor a hideous plainness be considered a fit substitute for severity! Oh
too, for the days when men shall have cast off their dependence on other
men’s works, and the customs of their own days, and, like true men and
faithful, shall honestly and with energy, each in his own sphere, set to
work to do all that in them lies to increase the power of art and to
advance its best interests. All these aims and objects are more or less
bound up with the best interests of a people, however old and however
powerful, because they depend for ultimate and real success upon the
thorough belief, on the part of all its votaries, in certain great and
eternal principles, which, if always acted upon, would beyond all doubt
sometimes make great artists and always good men.

The principle which artists now have mainly to contend for is that of
TRUTH; forgotten, trodden under foot, despised, if not hated for ages,
this must be their watchword. If they be architects, let them remember
how vitally necessary truthfulness in construction, in design, and in
decoration, is to any permanent success in even the smallest of their
works; or sculptors, let them recollect how vain and unsatisfactory has
been their abandonment of truth in their attempted revival among us of
what in Classic times were--what they no longer are--real
representations and natural works of art; if painters, let them remember
how all-important a return to first principles and truth in the
delineation of nature and natural forms is to them, if they are ever to
create a school of art by which they may be remembered in another age.

Finally, I wish that all artists would remember the one great fact which
separates by so wide a gap the architects, sculptors, and painters of
the best days of the Middle Ages from us now--their earnestness and
their thorough self-sacrifice in the pursuit of art, and in the
exaltation of their religion. They were men who had a faith, and hearts
earnestly bent on the propagation of that faith; and were it not for
this, their works would never have had the life, vigour, and freshness
which even now they so remarkably retain. Why should we not be equally
remembered three centuries hence? Have we less to contend for, less
faith to exhibit, or less self-sacrifice to offer than they, because we
live in later days? Or is it true that the temper of men is so much
changed, and that the vocation of art has changed with it? I believe
not. There have been evidences enough that there is no lack of
liberality on the part of our employers, where there is any evidence of
skill and enthusiasm for his work on the part of the artist. The English
architect of to-day has opportunities as great as those of any of his
predecessors, if he will but use them. But he must use his art as one
who respects both himself and it. There is no real respect for an art
when it is treated, as it always has been by the Renaissance architects
and their followers, as a mere affair of display. No good building was
ever yet erected in which the architect designed the front, and left
the flanks or internal courts to take care of themselves. So also no
good building was ever seen, in which the exterior only was thought of,
and the internal decoration and design neglected. But this is almost
universal now, except in the few buildings in which the Gothic style has
been carefully revived. In such treatment of art as this, there is an
ingrain falseness, which is as demoralizing as it is ruinous. If
architecture is only an affair of outside display, no one will take any
real interest in it, for from the first it is the evidence of the
architect’s love for his work which has given the human interest which
is all in all to it.

It is this truthfulness only, in every line and every detail of every
part of a building, which can ever make great architecture--it is this
only which one would wish to extract from the works of our
forefathers--and this only which I have desired to discover in the works
of those Italian artists whose labours I have been considering, and
whose efforts I have endeavoured to set before my readers; and it is
this desire which can alone be my excuse for having undertaken the work
which I have now brought to a conclusion.

[Illustration: 65. KEY PLAN OF CAPITALS DUCAL PALACE VENICE.]




APPENDIX.


_Catalogue of the Subjects of the Sculptured Capitals in the Lower Stage
of the Doge’s Palace, Venice._

The capitals are numbered as in the accompanying key-plan, beginning at
the south-east angle on the Molo front, and going from right to left
until the last capital, near S. Mark’s, is reached. For those who wish
for something more than a catalogue, I need hardly say that in the
latter portion of the second volume of the ‘Stones of Venice,’ Mr.
Ruskin has given all that can be desired, with his usual felicity and
beauty of verbal illustration.

In the ‘Annales Archéologiques,’ vol. xvii. (1857), Mr. Burges has given
a very full and careful account of all the capitals, to which M. Didron
Ainé has added some supplementary notes.

Zanotto (‘Il Palazzo Ducale di Venezia,’ vol. i. pp. 209-355) has given
a still more full description, accompanied by rather rude outlines of
all the lower range of capitals. There is small need, therefore, for
anything more than a mere catalogue here, which it seems to me may be of
service to some of those who are able to look at the Ducal Palace, but
unable to carry with them any of the weighty volumes to which I have
referred.


_South-east Angle._--Above the capital is the Drunkenness of Noah. Above
this, on a level with the traceries of the upper arcade, the archangel
Raphael, with Tobit, who bears a scroll with these words--

    EFICE [=Q]
    SO. FRE
    T[=V]. RAFA
    EL. REVE
    RENDE
    QUIETŪ.[103]

CAPITAL I. Partly built up. Has three figures of nude children, one with
a comb and shears, another with a bird. The foliage is good; but the
nude figures have the appearance of semi-Renaissance work.

II. Partly built up. Large birds--one devouring a serpent, another a
fish, and the third pluming its feathers. The foliage here is not very
good, and the design of the capital in no way first-rate, the birds
being treated in a very naturalesque way.

III. Partly built up. Large heads, male and female. The man has a
helmet, partly of plate, but with chain mail round the neck.

IV. Partly built up. Children, nude, holding (1) a bird, and (2 and 3)
fruit.

V. Partly built up. Emperors. This is the first capital which has
inscriptions. Those visible are (1) TITUS VESPASIAN IPAT; (2) TRAJANUS
I[=NPE]; and (3) (OCT)AVIANUS IPATO.

VI. Partly built up. Large heads, alternated with tufts of foliage,
badly carved.

VII. Virtues and Vices. (1) Liberality; inscribed LARGITAS. ME. ONORAT.
(2) Constancy; COSTANCIA. SU. NIL. TIMĒS. (3) Discord; DISCORDIA.
SU. - - - DISCORDANS. (4) Patience; PATIENTIA. MANET. MECUM. (5) Despair;
DESPERACIO. MORS. CRUDELIS. (6) Obedience; OBEDIENCIA.A.D[=NO].EXIBEO.
(7) Infidelity; INFIDELITATE.--ILI.GERO. (8) Modesty; MODESTIA.ROBŪ
OBTINEO. This capital should be compared with No. XXVIII. The foliage
here is very beautiful; but the execution of No. XXVIII. is best.

VIII. Monsters, generally with musical instruments. A riding figure here
wears chain armour. There are no inscriptions on this capital, and its
intention is very obscure.

IX. Virtues. (1) Faith; FIDES.OPTIMA.IN.DEO. (2) Fortitude;
FORTITUDO.‘SUM.VIRILIS’ (Mr. Ruskin), or ‘INVINCIBILIS.’ (3) Temperance;
TEMPERANTIA. SUM. IN. OMIBU. (4) Humility; HUMILITAS.ABITAT.I.ME. (5)
Charity; KARITAS.DEI.MECŪ.EST. (6) Justice; REX.SUM.JUSTICIE. (7)
Prudence; PRUDENTIA.METIT.OIĀ. (8) Hope; SPĒ.HABE.IN.D[=NO]. The
differences between this capital and No. XXIX. are very slight. In the
latter, Prudence has a book, which she has not here; and Temperance has
a jug here in addition to the chalice which the other carries.

X. Vices. (1) Luxury; LUXURIA. SUM. IMENSA. (2) Gluttony; GULA. SINE.
ORDINE. SŪ. (3) Pride; SUPERBIA.PREESSE.VOLO. (4) Anger; IRA.
CRUDELIS. Ē. IN. ME. (5) Avarice; AVARITIA. ANPLECTOR. (6) Idleness;
ACCIDIA. ME. STRĪGIT. (7) Vanity; VANITAS.IN.ME.HABUNDAT. (8) Envy;
INVIDIA.ME.CŌBVRIT. This capital is very finely sculptured.

XI. Birds. Some web-footed, some not so; and with no inscriptions.

XII. Virtues and Vices. (1) Misery; MISERIA. (2) Cheerfulness;
ALACRITAS. (3) Folly; STULTICIA.. E. REGNAT. (4) CASTITAS
(CE)L(EST)IS.Ē. (5) HONESTY; (HO)NEST(ATEM. DILIGO). (6) Falsehood.
(7) Injustice; INJUSTICIA. SEVA. SŪ. (8) Abstinence; ASTINĒCIA.OPTIMA.E..
This capital is so much damaged as to be hardly intelligible without
comparison with No. XXXIII.

XIII. Lions’ heads, large and coarse, and with very poorly carved tufts
of foliage between them.

XIV. Wild animals. The whole of the beast, not the head only, is given.
They are poorly carved and designed.

XV. Damsels and Youths. Considered by Selvatico, and after him by Mr.
Ruskin, to represent Idleness. There are no inscriptions. More probably
they represent the youth of the higher class with marks of their
sportive occupations. This is an extremely well-carved capital.

XVI. Eight large heads alternately with tufts of foliage. The whole
finely carved and designed. Supposed by Zanotto to represent the
foreigners who traded with Venice. Selvatico describes it as
representing Latins, Tartars, etc., and as being in fact a repetition of
No. XXIII. (_vide infra._)

XVII. Philosophers. This is very much damaged, and the inscriptions are
nearly destroyed. (1) Solomon. (2) Triscian. (3) Aristotle. (4) Tully.
(5) Pythagoras. On a label carried by this figure Mr. Burges reads the
date 1344. Mons. Didron interprets it 1399; this is a date, however,
which he will not admit, believing the real date to be 1299. Mr. Ruskin
does not appear to have seen these figures, and I have been unable to
satisfy myself about them.

XVIII. This is the angle capital. Above it is the Temptation of Adam and
Eve; and on the second stage, the four winds on the capital of the
arcade, and the Archangel Michael above. The whole is a perfectly
beautiful group of sculpture, of an equally beautiful and well-selected
story.

Planets. (1) Creation of Man;
DE.LIMO.D[=S].A[=DA].DE.COSTA.FORMAVIT.ET.EVA. (2) Saturn;
ET.SATURNE.DOMUS.EGLOCERUNTIS.ET.URNE. (3) Jupiter;
INDE.JOVI.DOMA.PISCES.SIMUL.ATQ.CIRONA. (4) Mars;
E--ARIES.MARTIS.ET.ACU--E.SCORPIO.PARTIS. (5) The Sun; EST. DOMU. SOLIS.
TU. QUOQ.SIGNE.LEONI. (6) Venus; LIBRA. CŪ. TAURO. VENUS.--T. PURIOR.
AURO. (7) Mercury; OCCUPAT.ERIGONE.STILBONS.GEMIUQ.LACONE. (8) The Moon;
LUNE CANCER.DOMUT.P[=BET].Ī.ORBE SIGNO[=RIO]. The whole of the
sculpture of this capital deserves careful study. Mars is a figure in
chain mail. Venus, seated on a bull, and the Moon--a female figure in a
boat, with a crescent in one hand and a crab in the other--are both of
them exquisitely treated.

XIX. Artificers. Figures alternately crowned and uncrowned working at
parts of a building. The foliage is admirable. The pieces of stone on
which the artificers are at work are inlaid with porphyry. Mr. Ruskin
points out that all the architectural details represented are such as
would be found in the early part of the fourteenth century. It is
certainly very curious that among the workers one has ‘DISCIPULUS
OPTIMUS;’ another--‘DISIPŪLS INCREDŪL,’ over the head: a reference
to S. John and S. Thomas, which is not intelligible to me.

XX. Beasts. Eight large heads, well carved and set as knops below
masses of rather heavily treated leaves. The beasts have their names
inscribed: LEO. LUPUS. URSUS. MUSIPUL. CHANIS.[104] APER. GRIFO. VULPUS.

XXI. Trades. Finely carved. Inscriptions over the heads of the workmen:
(1) LAPICIDA SUM. (2) AURIFICES. (3) CERDO SUM. (4) CARPENTARIUS SUM.
(5) MENSURATOR. (6) AGRICHOLA. (7) NOTARIUS SUM. (8) FABER SUM.

XXII. The Ages of Man. A very interesting capital. (1) ╋
LUNA. D[=NA]T. IFANCIE. P. ANO. IIII. (2) MECUREŪ. DNT.
PUERICIE. P. ANO. X. (3) ADOLOSENCIE. (VEN)US. P. AN. VII. (4)
INVENTUTI.D[=N]T.SOL.P.AN.XIX. (5) SENECTUTI.D[=N]T.MARS.P.
AN. XV. (6) SENICIE.D[=N]T.JUPITER.P.ANN.XII. (7)
DECREPITE.D[=N]T.SAT[=N].UQ.AD.MŌTE. (8) ULTIMA.E.MORS.PENA PECATI.

XXIII. Nations. This is treated in the same way exactly as No. XX. The
heads are inscribed, LATINI. TARTARI. TURCHI. ONGARI. GRECI. GOTI.
EGICY. PERSII. It will be found that, counting from the south-west
angle, the second and fifth capitals on each front are of the same
description, i.e. Nos. XIII. and XVI. on the west side, and XIX. and
XXIII. on the Piazzetta front. I think this militates against Mr.
Ruskin’s view that this capital is not old, as it shews how very
regularly they are placed; and I do not see much to choose between in
these four capitals. It is worth notice here that this capital is, to a
considerable extent, a replica of No. XVI.; but the latter has no
inscriptions.

XXIV. Love and Marriage. This is a much larger column than the others.
It carries a cross wall above, so placed as to allow of rooms on the
sea-front, about sixty-three feet in width. It is one of the most
exquisite of all the series. The subjects are: (1) A young man with his
hand on his heart, admiring a damsel. (2) They meet and converse. (3)
She puts a crown on his head, and presents him with an apple. (4) They
embrace. (5) The marriage bed. (6) They hold their bambino wrapped in
swaddling clothes between them. (7) The child grows, and enjoys life.
(8) The child is dead, and his parents mourn over his body. The change
in character from the extreme smartness of dress in the earliest
subject to the carelessness about it in the last, should be observed.

Above this capital is a figure of Venice personified, in one of the
divisions of the tracery. She is seated on lions’ backs, with her feet
above the sea.

XXV. Labours of the Months. This is not a replica. (1) March; MARCIUS.
CORNATOR. (2) April and May; APRILIS ╋ MAGIUS. (3) June;
JUNIUS.CŪ.CERISIS. (4) July and August; JULIUS ╋ AUGUSTU. (5)
September; SEPTEBE. SUPEDITAT. (6) October and November; OCTOBĒ ╋
NOVEMBE. (7) December; DECEM - - - CAT SUUM. (8) January and February;
JANVARIVS ╋ FEBRUARU.

XXVI. Sports and Employments of damsels and young men. This is a replica
of No. XV. The foliage here is exquisitely carved. There are no
inscriptions. The figures, as in the other, hold: (1) A horse. (2) A
bird, and the leg of a larger bird (hawking?). (3) A distaff. (4) A dog.
(5) A flower. &c. &c.

XXVII. Fruit in baskets. This is not a replica. It is the finest of the
“knop” series of capitals. The fruit is admirably carved, and here the
exact imitation of nature in the fruit, combined with foliage in the
knops between the baskets, which is quite conventional and architectural
in its character, is extremely interesting and instructive, as shewing
how distinctly the sculptor knew the proper limits of conventional and
realistic representation. The fruit are described: (1) Cherries;
SEREXIS. (2) Pears; PIRI. (3) Cucumbers; CHOCUMERIS. (4) Peaches;
PERSICI. (5) Gourds; CUCHE. (6) Melons; MOLONI. (7) Figs; FIGI. (8)
Grapes; HUVA.

XXVIII. Virtues and Vices. This is a replica of No. VII. The differences
are very small. Here II[H? unclear?] is prefixed to the “ONO[G?]RAT” of
the other capital.

XXIX. Virtues. A replica of No. IX.

XXX. Vices. A replica of No. X., but well deserving study.

XXXI. Monsters. A replica of No. VIII., but very rudely carved, and very
inferior to the next.

XXXII. Students. This is _not_ a replica of No. XVII. The figures are
admirably treated, all looking thoughtful, and some holding foliage. It
seems to represent the more thoughtful side of youth as compared with
the idle side, represented in the capitals XV. and XXVI. Zanotto
supposes the figures to be the Seven Wise Men of Greece; but the eighth
figure--a woman speaking--does not lend itself to this. Mr. Burges most
ingeniously suggests that it refers to Nouvella d’Andrea, who, in her
father’s last illness, lectured from behind a curtain to his pupils; her
father died in 1348.

XXXIII. Virtues and Vices. This is a replica of No. XII.

XXXIV. Eight birds. A replica of No. XI.

XXXV. Nude children. A replica of No. IV. They have birds, fruit, etc.,
in their hands.

XXXVI. Justice and Lawgivers. This is the north-west angle column. The
capital has the following subjects: (1) Justice; JUSTITIA. (2)
Aristotle; ARISTOTEL. CHE. DIE. LEGGE. (3) Moses;  - - -
L.PUOLO.D.L.SUO.ISEL.RITA.[105] (4) Solon;
SAL.UNO.DEI.SETE.SAVI.DI.GRECIA.CHE.DIE.LEGGE. (5) Scipio;
ISIPIONE.ACHASTITA.CHE--E.LA.FIA.ARE.[106] (6) Numa Pompilius; NUMA
POMPILIO.IPERADOR.EDIFICHADOR.DI.TEPI.E.CHIESE. (7) Moses receiving the
tables of the law; QUADO.MOISE.RICEVE.LA.LEGE.I. SUL.MONTE. (8) Trajan;
TRAINO.ĪPERADORE.CHE.DIE.JUSTITIA.A.LA.VEDOVA.

The carving of foliage in this capital is, I think, simpler, and on the
whole better than that on the other great angle, capital No. XVIII.,
with which it invites comparison; but the carving of the figures is
quite inferior, and later in character. The foliage above this capital,
which supports the Judgment of Solomon, is inferior to that on the
capital itself, and is, I believe, considerably later in date, being no
doubt of the same age as the figures in the Judgment, i.e. not earlier
than 1430-1450. Above this angle is the figure of the Archangel Gabriel.

                                THE END

                               LONDON :
                  PRINTED BY WILLIAM CLOWES AND SONS.
                   STAMFORD STREET AND CHARING CROSS

FOOTNOTES:

 [1] In a view of Zurich, published A.D. 1654, these
 steeples are shown with octagonal spires rising above the gabled
 sides of the towers; the belfry stages and cupolas now existing must
 therefore be of a date subsequent to the publication of this view.

 [2] This division is seen clearly in one of the curious prints by
 Merian, which illustrate a most valuable and interesting book,
 entitled ‘Topographia Helvetiæ,’ published at Frankfort-am-Main,
 A.D. 1654, and full of most picturesque and exact views
 of Swiss towns; they are valuable, as proving beyond all question
 their state in the beginning of the seventeenth century, and as being
 executed with very much artistic feeling. That of Chur gives the whole
 town in the most complete manner; the castle, the churches, the walls,
 and the many watch-towers, with the magnificent mountains behind
 them, making one of the most picturesque _ensembles_ conceivable.
 Many of these views of Swiss towns are remarkable, as proving how
 very regularly the mediæval towns were planned whenever there was the
 opportunity, the streets all at right angles, and the great church and
 marketplace in the centre of the whole.

 [3] I grieve to say it does so no longer. When I last crossed the
 Splügen, in 1869, this bridge had disappeared, and one of iron had
 been erected in its place. It was a capital example of the skilful
 carpentry of the old Swiss bridge-builders.

 [4] Probably most travellers who pass by Chiavenna are now on their
 way to or from the Engadin by the beautiful Maloja pass. They will do
 well before they reach the top of the pass to notice on their left the
 ruined remains of a Gothic chapel of the fifteenth century, which may,
 I suppose, aspire to the honour of being at a greater height above the
 sea than any other Gothic church in Europe. Its architectural merit
 is not great, but still it has a certain value, as showing how well a
 simple little Gothic church looks among the wildest mountain scenery.

 [5] See illustration opposite page 104.

 [6] The church was built in A.D. 1134 by Maestro Fedro.

 [7] For a view of this porch, see the frontispiece of this volume.

 [8] “✠ MCCCLX · MAGISTER · JOHANES · FILIUS · C · DNI · VGI · DE ·
 CAMPILIO · FECIT · HOC · OPUS.” This Giovanni da Campione was one of a
 family of architects of much celebrity. See their genealogical tree in
 ‘Italian Sculptors’ p. 106.

 [9] The round church of San Tommaso in Limine, described by Mr. Gally
 Knight as similar in plan to San Vitale, at Ravenna, is only eight
 miles to the north of Bergamo, and ought, equally with Malpaga Castle,
 to be seen. I regret that I have never yet visited it.

 [10] S. Gereon, at Köln, is a magnificent example of a church upon
 the same kind of plan; a grand choir projected from a decagonal nave,
 the effect of which is capital. No doubt such a nave does much more
 than merely suggest the possibility of adapting the dome to Gothic
 buildings.

 [11] I say this advisedly, though knowing very well that some German
 antiquaries assert this cathedral to have been built in the time of
 Bishop Ulrich II., A.D. 1022-1055. Those who say so
 must, I think, be entirely blind to all architectural detail.

 [12] Anno Domini MCCXII. ultima die Februarii
 presidente venerabile Tridentino Episcopo Frederico de Vanga, et
 disponente hujus Ecclesie opus incepit et construxit magister Adam de
 Arognio Cumane Dioc. et circuitum ipse, sui filii, inde sui Aplatici
 cum appendiciis intrinsece et extrinsece istius ecclesie magisterio
 fabricarunt. Cujus et sue prolis hic subtus sepulcrum manet. Orate pro
 eis.

 [13] The church of the Capuchins, at Lugo, is a Spanish example of the
 same arrangement.

 [14] Admirable drawings of it have been published by Mr. Grüner.

 [15] An extremely careful chroma-lithograph of this wall and monument
 has been issued by the Arundel Society, accompanied with a notice of
 both, written by Mr. Ruskin.

 [16] I leave this passage as it originally stood. It is pleasant to
 feel how completely unnecessary it is now to use such language on the
 subject.

 [17] It is not a little remarkable that this should be the monument a
 copy of which the late Duke of Brunswick desired in his will to have
 erected over his grave!

 [18] The wings of these angels are of metal, though the figures
 themselves are marble.

 [19] This grille is worthy of especial notice. Instead of being hard
 and stiff, it is all linked together, so that it is more like a piece
 of chain mail than of iron railing. Its intricacy adds manifold to the
 effect of the group of tombs which it half conceals.

 [20] I need hardly say that all this is changed, and I hope changed
 for the better. The city looks more thriving than it did, and more of
 the old mansions are properly occupied than was the case in the time
 of the Austrians.

 [21] This cloister is said to have been built in 1123. This is, I
 think, at least fifty years earlier than its real date.

 [22] The sculptor of this work left his name--Adaminus--on the capital.

 [23] The most important of these is the interesting church of San
 Stefano in the suburb on the opposite side of the Adige. It has been
 much modernized, but has still, I believe, an early crypt and an
 octagonal steeple over the crossing of the same age as San Zenone.

 [24] See also ‘Pietro Brandolese, Scultura, Pittura, &c., di Padova.’
 Padova, 1795.

 [25] Since this was written the whole of these subjects have been
 published by the Arundel Society, and Mr. Ruskin’s notice of them
 has also been given to us: they are very valuable as exemplifying,
 as well perhaps as colourless engravings can do, the exceeding value
 and originality of this series of paintings. It is to be wished that
 they may produce some effect upon the minds of our modern artists, who
 much require to take home to themselves the lesson of sincerity and
 earnestness of purpose, combined with the highest kind of subject,
 which Giotto so eminently exhibits in all his works. An extremely good
 series of photographs of the whole of these paintings may now also be
 obtained in Venice.

 [26] The order of the planets attached to the seven ages is as
 follows:--I. The Moon. II. Mercury. III. Venus. IV. The Sun. V. Mars.
 VI. Jupiter. VII. Saturn.

 [27] This is the tradition, but it is one which is not, I think,
 supported either by documentary evidence or by the style of the
 building. Nicola left Padua four years before the church was
 commenced; and Fra. Carello is mentioned in the archives of the
 convent as one of the architects, of whom no doubt there were several
 before the work was finished.

 [28] The eastern chapel and dome are comparatively modern, and the
 coverings of the other domes appear to be also modern; but I suppose
 they follow the old outline.

 [29] The dimensions are worth giving to show how little this church
 owes to mere size. It is 245 feet long, 201 feet across the transept,
 and 170 feet across the west front. The height of the central cupola
 is 90 feet, and that of the west front 72 feet.

 [30] The inscription on the screen, which gives the date and the name
 of the Doge Antonio Venerio, gives also the names of the sculptors.

 [31] The tomb of Vitale Faliero and another.

 [32] The tomb of the Doge Marino Morosini.

 [33] ‘Sagornino Chronicon,’ p. 119.

 [34] ‘Il Palazzo Ducale di Venezia,’ per Francesco Zanotto, i. p. 9.

 [35] ‘Continental Ecclesiology,’ p. 306.

 [36] The mosaics here, as in Venice, are wholly of glass. The gold is
 covered with a thin film of glass, and the other colours used are dead
 white, black, dark and light blue, green and red. The very smallness
 of the palette was here, just as it was with the old painters on
 glass, a distinct advantage, saving them from the bizarre and confused
 effect produced in such works by the use of too many colours or shades
 of colours.

 [37] The red bricks are 2¼ thick × 9½ in. long, whilst the yellow
 bricks are 3¼ thick × 12 in. long.

 [38] The twelfth-century bricks here measure seven inches by two
 inches, and are built with a half inch mortar joint; they are of
 red and yellow colour, used indiscriminately, and, though good and
 lasting, extremely rough in their make.

 [39] The crockets on the monument of A.D. 1437 are
 exactly similar to those on the western gables of S. Mark’s, and prove
 that these are of about the same date.

 [40] I refer here to San Giacomo del Rialto. Its neighbour, San
 Giacomo del Olio, has also a brick campanile, but of inferior merit.

 [41] A view in the ‘Nuremberg Chronicle’ shews these three gables just
 as they now are.

 [42] I leave this description as it stood in 1855. Since then the
 whole of this interesting building has been so elaborately restored,
 that I doubt whether an old stone remains. It has lost all its charm,
 and this was once intense.

 [43] This house is in the Sestiere di Cannaregio, Parrochia San
 Canciano.

 [44] Zanotto, ‘Il Palazzo Ducale di Venezia,’ i. 39.

 [45] Ibid., i. 52-60.

 [46] “1362, die iv. Dec. Quia est magnus honor civitatis providere
 quod sala magna majoris consilii nova non vadat in tantam desolationem
 in quantam vadit cum notabili damno nostri communis: et sicut clare
 comprehendi potest, leviter potest compleri, et reduci ad terminum,
 quod satis bene stabit cum non magna quantitate pecuniæ; vadit pars
 quod dicta sala nova compleri debeat,” &c. &c.--Decree in Zanotto, i.
 72.

 [47] Mr. Burges, in his account of the capitals, ‘Annales
 Archéologiques,’ vol. xvii. pp. 74-88.

 [48] The capitals which are replicas of each other are the 4th and
 35th, the 7th and 28th, the 8th and 31st, the 9th and 29th, the 10th
 and 30th, the 11th and 34th, the 12th and 33rd, the 15th and 26th. The
 25th, 27th, 32nd, and 36th (north-east angle) are original, though
 they are in the northern portion of the Piazzetta-front. See Appendix,
 with key-plan.

 [49] The similar marble facing at Vicenza was executed between 1400
 and 1444. See p. 130.

 [50] Published by Mr. Parker, of Oxford, to whose courtesy I owe the
 use of this illustration.

 [51] See Appendix at the end of the volume.

 [52] I say “usual,” because it is really quite curious to see how
 repeatedly either the dog-tooth or the nail-head is used in this
 position. The commonest eaves-cornice consists of a simple chamfered
 stone--the chamfer covered with dog-tooth--supported on moulded
 corbels at short interval.

 [53] There is another traceried balcony in the canal near the Bridge
 of Sighs. It is the only other example I know in Venice.

 [54] They may be compared with the chevroned and spiral columns in
 the archway, leading from the north aisle into the baptistery of the
 Frari, erected between 1361 and 1396, which is probably about the date
 of the Ca’ d’Oro.

 [55] The Europa and Danieli’s.

 [56] This arrangement is not by any means unknown in Northern Europe,
 though certainly uncommon as compared with Italy, where it was almost
 universal. There is an example of the thirteenth century at Easby
 Abbey, Yorkshire, and another at Oakham Castle; whilst in France the
 ancient houses at Cluny all have it; and at Ratisbon, one of the most
 interesting cities in Germany, a great number of houses of the twelfth
 and thirteenth centuries, of prodigious architectural interest, have
 it.

 [57] It was only so used in the Ducal Palace.

 [58] I _have_ heard a polka played by the organist in S. Mark’s!

 [59] I take these notes of Grado from ‘Mittelalterliche Kunstdenkmale
 des Oesterreichischen Kaiserstaates.’ Stuttgart. 1858.

 [60] It is to be seen, however, in the church of San Petronio, Bologna.

 [61] See for an engraving of this archivolt, Chapter XIV. p. 393.

 [62] The two transepts are so very similar, that it seemed unnecessary
 to engrave my sketches of both.

 [63] The chimneys so common in Venice are ancient, but yet hardly
 redeemed from ugliness. They are cylindrical, with heads sloping out
 in a strange fashion, and in the form of inverted truncated cones. See
 p. 224.

 [64] This building has recently (1872) been restored, and with
 not much gain, though the barber’s shop which used to occupy the
 ground-floor has been removed.

 [65] Mr. Grüner has published some very careful drawings of these
 details, in which he has restored the painted decorations with which
 the coloured construction of the walls was enriched. The style of
 decoration was much like that of Sta. Anastasia, Verona.

 [66] This was written in 1855

 [67] Said to have been carved by Nicolò, who is supposed to be the
 same man who wrought on the west door of San Zenone, Verona. This does
 not appear to me to be likely; the work at Verona being, I think,
 earlier than that at Ferrara.

 [68] The complete church was to have been 800 feet long, and
 525 feet across the transepts, with a central dome 130 feet in
 diameter.--Fergusson’s ‘History of Architecture,’ ii. 210.

 [69] Mr. Perkins, ‘Italian Sculptors,’ p. 251, says that the
 nationality of Wiligelmus has been much disputed. Kreuser says that he
 came from Nürnberg; and the representation of King Arthur’s victories
 over the Visigoths is adduced as a proof that he was not an Italian.

 [70] The inscription on the north door is as follows:--

    “Bis binis demptis. Annis de mille ducentis
     Incepit dictus, opus hoc sculptor Benedictus.”


 [71] It is commonly said to have been designed by Heinrich von
 Gmünden in 1387; but in a most interesting note at p. 116 of ‘Italian
 Sculptors,’ Mr. Perkins gives the evidence for and against the claim
 of a German to be the architect of this cathedral. He believes that
 there is no longer any reasonable doubt that the first architect
 was Marco Frisone da Campione. Heinrich Adler von Gmünden, who has
 commonly been stated to be the architect, did not come to Milan until
 five years after the foundation of the church. Marco da Campione
 died in 1390; and the church was ready for divine service in 1395.
 The criticisms I have made in the text appear to me to be equally
 applicable to an Italian architect trained in Germany, or to a German
 working in Italy; and if Marco da Campione was the architect, one is
 compelled, by the logic of the building itself, to say that either he
 had studied north of the Alps with a view to perfecting his design,
 or that he depended very largely on the help given him by such men as
 Henry of Gmünden, whom he had called in to his assistance.

 [72] When I first saw this I thought it was an entirely modern device.
 In 1871, however, in passing through France, I found at La Fère a
 church of flamboyant character, with all the cells of its groining
 covered with sunk traceries; and at Chambéry the painting on the roof
 of the Sainte Chapelle is said to be old.

 [73] I leave this passage as it was written in 1855. Such troubles are
 now all passed and gone; but we run some risk of forgetting how much
 we have gained in this way by the political changes that have occurred
 since then.

 [74] See plate 18, p. 122, for an illustration of the cornice of this
 atrium.

 [75] The atrium was added, it is said, in the ninth century.

 [76] See, for description of it, Hemans’ ‘Mediæval Christianity,’ &c.,
 p. 305.

 [77] It is worth notice that the regular-looking bays of the nave are
 of very various widths. The two eastern are 21 feet; the next two,
 four feet less; and the fifth still narrower. The bricks here measure
 1 ft. ½ in. × 5 in., and are 3¼ in. high, and have all been chiselled
 on the face.

 [78] I think no apology is necessary for the omission of this modern
 stage in my view of the Broletto.

 [79] This old inn is a thing of the past. A large, bustling, and much
 smarter hotel has taken its place; and a magnificent carriage-road
 has put a stop to all need for walking by pleasant field-paths to the
 Furca.

 [80] The famous steeple of Antwerp is arranged in the same way.
 The spires of S. Elizabeth at Marburg, and the metal spires of the
 churches at Lübeck and Luneburg, are also somewhat similarly treated.

 [81] The usual plan is sometimes deviated from and improved by having
 two bays of aisle opening with two arches into each bay of the nave,
 so that every bay of groining throughout the church is very nearly
 square. This is a common plan in early German churches, and is one of
 the many indications of similarity between German and Italian work,
 which might be adduced were we to enter on this interesting question.

 [82]

                                   Height     Diameter   Diameter
                                  of Shaft.   at Base.   at Neck.
                                  Ft.  in.      In.         In.

  Cloister, San Zenone, Verona     3 11½       5⅛         4¼
  Cloister, Genoa Cathedral        4  5        6          5¼
  Cloister, San Stefano, Bologna   2  8½       4⅜         4⅛


 [83] The shafts used in the shrine of S. Edward at Westminster are
 of the same description, and shew that in one of the most exquisite
 works in England our early architects saw no incongruity between their
 beautiful but foreign character and the otherwise, at that time,
 purely national architecture.

 [84] Jerpoint and Cong Abbeys are Irish examples.

 [85] English examples may be seen in the western porch of Fountains
 Abbey, and among the extensive fragments of Egglestone Abbey, near
 Barnard Castle.

 [86] Numerous examples of masonry arranged in the same way occur in
 old English examples. The openings of fireplaces in particular are
 often so constructed.

 [87] There is a very curious example of Italian tracery in the church
 of San Giacomo, in Bologna. This is a two-light window, composed of
 a series of slabs of stones, pierced with geometrical figures and
 supported by shafts. It has, beginning at the sill and reckoning
 upwards--1, two lights divided by a shaft monial; 2, a slab pierced
 with two trefoiled heads to the lights; 3, a large transome panel of
 stone pierced with a quatrefoil and two trefoils; 4, same as 1st; 5,
 same as 2nd; 6, the arch, whose tympanum is filled in with another
 pierced slab of stone. It will be seen that the construction of such a
 window as this is altogether unlike that of any English window.

 [88] See plate 40, p. 256.

 [89] The windows in the Castle of S. Angelo, between Lodi and Pavia,
 are the only examples I met with of the use of brick for monials. In
 Northern Germany, on the contrary, where the shaft was almost unknown,
 brick monials are universal, and generally unsatisfactory in their
 effect.

 [90] See preceding page.

 [91] See pp. 64, 65.

 [92] See p. 335.

 [93] See plate 18, p. 122.

 [94] See plate 15, p. 111.

 [95] See plate 9, p. 72.

 [96] See plate 17, p. 121.

 [97] See Plate 24, p. 180.

 [98] See Plate 18, p. 122.

 [99] It would be difficult to give stronger evidence of the intrinsic
 effect of a good coloured material than is afforded by the fact that
 designs so really ignorant in their architectural detail as, e.g.,
 most of the buildings of the time of William III. and Queen Anne
 should nevertheless have a certain charm for us, solely derived from
 the beautiful colour of the bricks with which they are built.

 [100] See plate 63, p. 340.

 [101] See plate 1, frontispiece.

 [102] Some of the mediæval buildings in Greece have small patterns
 carved in low relief all round the walls in occasional courses, which
 are evidently intended to produce the kind of effect referred to above.

 [103] “Oh, venerable Raphael, make thou the gulf calm, we beseech
 thee!” This figure looks towards the sea and the port.

 [104] Mr. Ruskin founds an argument on the introduction of an
 H in this way in what he calls one of the Renaissance
 capitals of the Piazetta (the 33rd). He omits to notice its use in
 this undoubtedly early capital.

 [105] Unintelligible; but explained by Zanotto to be: “Mosè che die’
 legge al suo popolo Israelita.”

 [106] Zanotto reads this: “Scipione a castita che rende la figlia al
 padre.”

       *       *       *       *       *

Typographical errors corrected by the etext transcriber:

repnted=> reputed {pg 38}

cöeval=> coëval {pg 114}

just at it was=> just as it was {pg 169 fn. 36}

Christendon=> Christendom {pg 189}

PALACE OF THE JURISTCONSULTS=> PALACE OF THE JURISCONSULTS {Illustration 45}

duty, bound=> duty-bound {pg 274}