The Project Gutenberg eBook of The Oriental Rug

This ebook is for the use of anyone anywhere in the United States and most other parts of the world at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this ebook or online at If you are not located in the United States, you will have to check the laws of the country where you are located before using this eBook.

Title: The Oriental Rug

Author: William De Lancey Ellwanger

Release date: July 13, 2010 [eBook #33144]
Most recently updated: January 6, 2021

Language: English

Credits: Produced by The Online Distributed Proofreading Team at (This file was produced from images
generously made available by The Internet Archive/American








Larger Image

Plate I.
Prayer Rug
From the Collection of Mr. George H. Ellwanger
Size: 3.10 x 6




Title Page Text




Copyright, 1903
By Dodd, Mead and Company
Published September, 1903



[Pg v]


That Oriental rugs are works of art in the highest sense of the term, and that fine antique specimens, of even modest size, have a financial value of ten, fifteen, or thirty-eight thousand dollars, has been recently determined at public auction. At this auction, several nations had a representative voice in the bidding, and the standard of price was fairly established. The value of rugs may have been imaginary and sentimental heretofore; it is now a definite fact, with figures apparently at the minimum. What the maximum may prove, remains to be seen.

Choice old rugs, therefore, to-day come into the same class with genuine paintings of the old Dutch School; with canvases of Teniers, Ruysdael, Cuyp, Ostade, or[Pg vi] whatever similar artist’s work may have escaped the museums. They vie in prestige with the finest examples of Corot, Diaz, Troyon, or Daubigny; and in monetary supremacy they overtop the rarest and grandest of Chinese porcelains.

And yet the Oriental rug, as against such competitors for the wealthy collectors’ favour, has hardly a history, and is practically without a name or a pedigree. Experts will tell you at a glance whether or not your Wouverman is genuine, or inform you where every true Corot was owned or whence it was bartered or stolen. In Chinese porcelains, the knowing dealer will easily prove to you not only under what dynasty but in what decade or year a particular piece was produced.

The painting has descent, signature, or the brush mark of a school to father it. The Chinese vase, bowl, or jar has its marks, cyphers, stamps and dates, and an undoubted genealogy to vouch for its authenticity. The rug must speak for itself[Pg vii] and go upon its intrinsic merits. It is its own guarantee and certificate of artistic and financial value.

The study of Oriental rugs, therefore, can never lead to an exact science or approximate dogmatic knowledge. Whoever is interested in them must needs rely upon his personal judgment or the seller’s advice. There is practically only one current book authority in the premises.

A new volume on the subject would thus seem to be well justified. It is the hope of the author that this book may prove itself sound and practical, and that it may help to make more clear and simple the right appreciation of a valuable rug.


Rochester, N.Y., 1903




Chapter  Page
I.The Mystery of the Rug 3
II.General Classification 13
III.Of the Making, and of Designs, Borders, etc. 21
IV.Of the Dyeing 35
V.Of Persian Rugs, specifically 43
VI.Caucasian Rugs, Daghestan and Russian Types 61
VII.Of Turkish Varieties 69
VIII.Turkoman or Turkestan Rugs 79
IX.Of Oriental Carpets, Saddle-Bags, Pillows, etc. 93
X.Auctions, Auctioneers, and Dealers 107
XI.Inscriptions and Dates 121
XII.General Observations and Particular Advice 131




I.Ladik Frontispiece
II.Koniah Facingpage22
III.Kazak ""36
IV.Sehna ""44
V.Chichi ""50
VI.Kabistan ""62
VII.Gheordez ""70
VIII.Koulah ""72
IX.Melez ""74
X.Beluchistan ""80
XI.Anatolian Pillows ""94
XII.Bergama ""124



[Pg 1]



[Pg 2]


[Pg 3]

The Oriental Rug


Chapter I


To judge of an Oriental rug rightly, it must be looked at from several points of view, or, at least, from two aspects; against the light and with the light. From the first standpoint, against the light of knowledge, speaking figuratively, there may be seen only a number of rude and awkward figures in crude colours scattered erratically on a dark or dingy-looking background, a fringe of coarse and ragged strings at either end, and rough frays of yarn at the sides. This is what is accepted by many people as an Oriental rug. And indeed this is what most rugs are.

[Pg 4]If, on the other hand, we view our rugs with the light of a better wisdom and happier experience, we will see the richest and softest of colours, the most harmonious shadings and blendings, medallions brilliant as jewels, or geometrical designs beautiful as the rose windows of a cathedral; or, again, graceful combinations of charmingly conventionalized flowers and delicate traceries and arabesques,—all these displaying new glories of ever changing and never tiring beauty. Each woven picture, too, is as soft to tread upon as a closely mown lawn, and caresses the feet that sink into its pile. These are Oriental rugs as their admirers know and love them.

Perhaps the chief charm of all such beautiful rugs is in their mystery. Their designs are odd and strange and full of hidden meanings, and their effects are often evolved from the crudest and clumsiest figures, hooks and squares and angles; they owe their wealth of colour to simple vegetable dyes from the woods and fields[Pg 5] and gardens, and yet the secret of many of these dyes is still a secret, or has long ago been lost. The places whence the rugs come, the people who make them and those who sell them, all are mysterious and hard to know and understand.

Moreover, broadly speaking, there are no experts on the subject, no authorities, no literature. He who would know them must learn them by experience. The rug dealers, for the most part, seem to treat their wares merely as so much merchandise, and what knowledge concerning them they are willing to impart is so contradictory as to be almost valueless. Few of them would agree upon the name of an example which might be out of the ordinary, or be able to tell where it was made. Ask of them what a “Mecca” is, and they will stammer in their varying answers. And yet the Armenians who handle most of the rugs in this country are often highly educated, and fully appreciate the beauty of their wares. Their taste, however, is[Pg 6] not always our taste, and all the Orientalists seem to retain their barbaric fondness for crude and startling colours. When we would turn to books for information in the matter we find that the authorities are not many. They might be numbered on your fingers and thumbs. These few books, moreover, have been published only in limited editions at high prices, and are not easily obtainable. One of the most important of such works is the sumptuously illustrated, elephantine folio, issued in Vienna in 1892 by the Imperial and Royal Austrian and Commercial Museum. And, elaborate as this authority is, the modest editor, by way of apology, says in the preface that “no pretensions are made toward perfection owing to the little information that we can fall back upon.” A recent authority on the subject is John Kimberly Mumford, and his volume on Oriental Rugs, published in 1900, has thrown much light on the subject. Too great praise cannot be given to this[Pg 7] work and to his later studies in the same field.

Still, no one knows it all, and the mystery of Oriental rugs only deepens as we try to learn. The little that any one may really know of them through experience, through questioning and elusive answers, through conversations with obliging and polite vendors, and through foreign travel even, is, when all is said, only a patchwork of knowledge. Consider how stupendous and hopeless would be the task of one who would dare endeavour to analyze, criticise, classify, and co-ordinate the paintings of the past five centuries, were no names signed to them or no appreciable number of pictures painted by the same known artist.

He who would write of rugs has a like condition to face.

And alas! also, whoever would write on this subject must now treat of it principally as history. The characteristic rugs, the antique rugs, the rare specimens,[Pg 8] are seldom to be bought. They are in museums, or in the hands of collectors who hold them in even a tighter fist.

Twenty years ago the warning was given that the choice old rugs were growing scarce; the years following found fewer still upon the market. Two or three years ago one of the largest wholesale houses in New York, carrying a stock of half a million or a million dollars, had no antiques to show. In the autumn of 1902, another large New York importer who had just returned from Persia, Tiflis, and Constantinople admitted that he had not brought back one valuable antique piece.

Nevertheless, the true enthusiast need not be discouraged. From wandering dealers, in odd corners, at the unexpected or by chance, one may happen on a choice specimen.

The very word “Persian” is a synonym for opulence, splendour, gorgeousness; and[Pg 9] “Oriental” means beauty and wonder and the magic of the “Arabian Nights.” From the Aladdin’s cave of the mystical East, therefore, we may still hope to gather treasure and spoil.

[Pg 10]



[Pg 11]



[Pg 12]


[Pg 13]

Chapter II


Most of the rugs of commerce in this country come from Persia, Turkey, Asia Minor, Turkestan, the southern part of Russia, Afghanistan, and Beluchistan; a few also from India. The rugs are named from the provinces or cities where they are woven, and to the uninitiated, the names seem to have been as fearfully and wonderfully made as the rugs themselves. They are spelled one way on the maps and every other way in catalogues and advertisements. In enumerating the most familiar ones it may be well to write their names as nearly phonetically and conventionally as possible. A few rugs have trade appellations only, without regard to[Pg 14] topography; and, often, unknown towns are called into requisition for fanciful titles to please the purchaser.

Of course the names of rugs may mean nothing to your man-of-all-work, whose duty it is to chastise them upon the lawn. But there is poetry in the names of the roses, and you cannot half enjoy their beauty unless you know a Mabel Morrison from the Baroness Rothschild; Cécile Brunner from the Earl of Dufferin; or can give the proper rank and title to Captain Christy, General Jacqueminot, and Maréchal Niel. And who would dare to talk of laces that could not give a French or Dutch or Irish name to them? Or, when painted pictures instead of woven ones were under discussion, who would venture to admit that he had heard for the first time the names of some of the Old Masters, or did not know any of the Flemish School, or could not at least touch his hat to a Gainsborough or a Romney? There were “old masters” in[Pg 15] wool as well as on canvas, as the Gheordez rugs most particularly prove, and though the artists’ signatures are missing or meaningless, their classification is important. Once learned, and then difficult to remember withal, rugs answer to their names like old and familiar friends. If Homer catalogued the ships, surely the masterpieces of the Eastern loom are worthy of brief nomenclature.

The Persians come first, and perhaps in the following order of excellence: Kirman, Sehna, Kurdistan, Khorassan, Serabend, Youraghan, Joshghan (Tjoshghan), Feraghan, Shiraz, Gulistan, Mousul, etc. The rug dealers frequently speak of a “Persian Iran,” but as Iran is the native expression for Persia, the name is as tautological as are the dealer’s laudatory adjectives. So far as the term “Iran” can be differentiated, it is now applied with some propriety to rare old Persian rugs of fine weave only, whose proper name may be in doubt.

[Pg 16]Among the Turkish rugs, which are mainly those from Asia Minor, the Yourdez (or Gheordez), the Koulahs, Koniahs, and Ladiks are by far the finest, and then come the Bergamas, vying often for like high honour, the Melez, and many others which are vaguely classed as Anatolians.

From Turkestan come the numerous Bokharas and the more uncommon Samarkands; from Afghanistan, the Afghans and the Khiva, and Yamoud-Bokharas. But the two rugs last named seem to have a doubtful paternity, and should perhaps be classed with the other Bokharas.

Beluchistan sends but one type, which is generally unmistakable, although Afghans, Bokharas, and Beluchistans all have a family likeness.

To Caucasia in Russia are credited the Kabistans, Shirvans, Chichis (Tzi-tzis), Darbends, Karabaghs, Kazaks, and Gengias, also the Soumacs, or so-called Cashmeres. The first four of these are somewhat similar in character, and not many years ago[Pg 17] were generally sold in this country under the indiscriminate title of Daghestans. We are more specific in our knowledge now, and can classify and differentiate an old Baku rug, or a Kuba, which is a Kubistan, and therefore what we used to call an antique Kabistan.

India provides us only with some fine large carpets mostly of modern make, and also with many imitations of Persian rugs, made in part by machinery like the current substitute for a Turkish towel.

[Pg 18]



[Pg 19]



[Pg 20]


[Pg 21]

Chapter III


In order to appreciate the beauty of rugs, it is well to remember how they are made, and with what infinite patience the bits of wool are knotted onto the warp one after another, knot upon knot and tie after tie, until the perfect piece is finished. Yet, no! Finished it may be, but never perfect. Deliberately, if necessary, it must show some defect, in proof that Allah alone is perfect. Such at least is the poetical version of a crooked rug as the seller tells it. Yet never was a vendor but will expatiate fluently on the[Pg 22] merits of a rug which lies true and straight and flat upon the floor, as a good rug should. It is a common sight nowadays in shop windows to see some wandering artisan plying his trade for the edification of the passer-by. In his own home it is generally a woman who does the weaving, and very commonly the whole family take part in it. More often still the rugs were woven by an Oriental maid for her prospective dowry, and the practice yet obtains. A specimen of her handicraft in textile art was a bride’s portion and marriage gift; it was considered as essential to the proceedings as the modern trousseau. This offering was a work of love and often a work of years. It is but natural, under such circumstances, with dreams, hopes, and fancies for inspiration, and the stimulus of rivalry, too, that masterpieces should [Pg 23]result. These Eastern marriage portions correspond to the “linen chest” of our ancestral Puritan Priscillas; and similar customs now survive in many countries. Except that the “accomplishment” of the Oriental maiden is so much more important, it might also be compared to the beadwork so diligently done by our grandmothers. If the Persian bride gave infinite toil and pains to innumerable knots and ties, our belles of the last century were also unwearying in their tasks, and strung more and smaller beads than any would care to count or finger now. The designs on these bead-bags were mostly crude and “homely,” and their art was very simple. But though the handiwork of the Orientals was expended in a better cause with worthier skill, both linen and wool, and even[Pg 24] beads, bespoke a labour of love in such employments; which, alas! is out of date to-day. Rugs of this character, gathered from house to house, together with some few stolen from mosque or palace, were the first ripe spoils of twenty years ago. Of course the supply was soon exhausted. It is an interesting question whether it might not be possible, in the East, to revive this high class of work among the girls. Instead of establishing great factories for machine-made products from set designs, could not the most skilful of the girls be induced by good prices to create original pieces and rejuvenate the old art?


Larger Image

Plate II.
Prayer Rug
From the Collection of Mr. George H. Ellwanger
Size: 3.5 x 4.7


[Pg 25]


“Serabend” Border Persian, Caucasian
Feraghan Leaf Design Rhodian or Lily Border


The method of weaving is most simple. The warp is stretched on a rude wooden frame, and this warp is either wool, linen, or cotton. The knotting is begun at the bottom and worked from right to left. A bit of woollen yarn about two inches long is deftly twisted between the strands of the warp, then tied in a secure knot, and the ends left as they are. This knot of yarn is then secured in place by one or more twists of the end of the warp, and then another knot of yarn is tied and the process repeated ad infinitum until the bottom row is finished and another row begun. Not till the rug is all made are the ends of the knots cut, according to the length of nap desired. Such, at least, was the original method, although the various knots are all a mystery to any but the initiated, by whom they are generally[Pg 26] classified as two only. When one square inch of rug is completed, according to the quality of the rug and the coarseness or fineness of the yarn, there have been thus laboriously tied from one hundred to five hundred knots, not uncommonly a thousand and more in some museum pieces. And all this while the weaver is working with his brains as well as with his fingers and keeping true to the design and colour scheme which he carries only in his head. Except in the few intentioned copies, specially made, they had formerly no patterns to follow. Each particular weaver, however, was wont to keep to the general design and colouring which distinguished his particular locality.


Koniah Field  Koulah Border


Of designs it may be said, generally, that they were originally individual trademarks, and, of [Pg 27]themselves, stamped the locality of their weavers. Later, as knowledge and civilization spread and tribe grew to communicate with tribe and nation with nation, local designs came to be used indiscriminately. For example, you will find in the semi-antique Feraghans or Shiraz, or Kiz-Killims as well, the distinctive and unmistakable Sehna models. On the other hand, certain definite, primal, and unchanged designs, both in the field and border, mark some rugs absolutely and exclusively; as the Bokharas and Afghans. In many, their classification is fixed, or at least approximated, rather by their borders than by the figuring of their fields. There are many border designs surely determining their origin and the region to which they properly belong. These borders may have been borrowed or stolen, or may have naturally spread to other regions, even in the old[Pg 28] time; and they may be adapted to various other makes to-day. Their evident individuality of design tells its own history just the same.

It is not difficult to master the characteristic features of the borders of many types; and, once known, they make a fair foundation of knowledge for the collector. They are often truer and safer guides to classification than are the designs of centre or field. Indeed, the study of borders, inner, middle, and outer borders, and borders characteristic, modified, or exceptional would make a book of wondrous artistic interest and beauty of design. Even the item of selvedge, particularly in the Beluchistans, shows great skill in colouring and pattern.

Turtle Border Crab Border

The consideration of characteristic[Pg 29] patterns in field and border is so involved with verbal description and specification in the various classes of rugs that an attempt at complete pictorial illustration of such figures in their proper place is practically impossible. A few reproductions are shown in this chapter which may serve as examples. Some of them are more particularly considered elsewhere in the text, as reference may show.

The Serabend border is referred to on p. 50, and is quite unmistakable; and the Persian border (p. 23) is familiar to every one, and appears frequently on Caucasian rugs of every quality and every age. The Feraghan leaf design is noticed on p. 52, and[Pg 30] wherever used in the drawing, determines its class as absolutely as any figure may. The Rhodian border is referred to more particularly on p. 72, and the Koniah design and Koulah border are described in their proper place, p. 72. Other Persian borders are most interesting, although they may not particularize any class or locality. Such are the turtle and crab borders (pp. 28 and 29), and the lobster design, at the head of this page. The origin of these strange forms of ornament as applied to carpet-weaving adds only another mystery to the subject. But dyes were derived not only from leaves and roots, but also from insects, molluscs, and crustaceans. It must be that the origin of the colour originally suggested these symbols of marine or insect life for decorative effect. The more they were used, however, the more conventionalized and meaningless they appear,[Pg 31] recent weavers not appreciating what they represented. Old pieces show more clearly the evident model. But old pieces also often show original creations in border and design, far more artistic than the usual types. The Kazak border of the titlepage is an example. The discriminating collector, when a choice offers, will do well to avoid the commonplace.


[Pg 32]



[Pg 33]



[Pg 34]


[Pg 35]

Chapter IV


The dye, the tone, the richness, and colour value of a rug was, and still is, an essential characteristic of the weaving of each class and region; and it was formerly not only essential but exclusive, the dyes being often trade secrets or, more truly said, tribe secrets.

Of course every one knows that the colouring of the yarn of the best Oriental rugs is derived only from vegetable or animal dyes, and to this is due their beauty and durability. It may be noted also, in parenthesis, that it is the yarn and not the wool that is dyed. Alas, that modern weavers, Oriental and Occidental, have learned to substitute mineral or aniline dyes! These not only destroy the wool[Pg 36] and fade badly, but when the fabric is cleaned or wet by any chance the colours run, and leave their stains and blemishes. Of course, too, they fail to give the richness, depth, and lustre of the good old method. Generally, their manifest crudity bespeaks the poor quality and coarseness of their make. Some vegetable dyes also fade, but they fade only into softer and more pleasing shades, and more delicate and harmonious blendings, as witness, in many antiques, the soft and beautiful tones of pink, salmon, and fawn which come from raw magentas, as the back of the rug will prove. But that magenta dye was of the old school. Modern magentas seem never to fade away gracefully and becomingly. It must be noted, however, while speaking of the dyes used in the fine old rugs and in the best rugs of to-day, that for one or two colours resort was, and is, had to mineral dyes. Many of the best old Turkish specimens have thus suffered in their blacks and browns, and many a[Pg 37] museum exhibit is eaten to the warp where these colours occur. It may be well to remember this, as some varieties of Mousul and of Turkish weave, thus worn to the warp in spots, leaving the other figures raised and in relief, are palmed off on the innocent purchaser as rare, “embossed” pieces. Iron pyrites is the mineral from which these black dyes are made, and some Turkish weavers seem to know no vegetable black or brown. In some of the best Persians, Serabends particularly, the green which is used in the borders has the same fault as the Turkish blacks and browns; and if it does not “fade away suddenly like the grass,” at least it leaves the nap “cut down, dried up, and withered.”


Larger Image

Plate III.
From the Collection of Mr. Erickson Perkins
Size: 5.9 x 7.2


The subject of the various dyes might be extended to a separate monograph, for really the whole history of rug making depends upon the dyes used. The day that the aniline, petroleum dyes came into use doomed the perfect making of carpet or[Pg 38] rug; and not all the strictest laws of the Medes and Persians—which is to say, the Shah of Persia—have availed to prevent the use of the mineral dyes, and the complete demoralization of modern weaving. You may find even in choice, closely woven, artistic Shirvans and Kabistans of fifteen and twenty years ago some few figures in certain colours which are clearly and manifestly aniline. They are the strong reds and especially the bright orange. And in some modern Kurdistans, which should be free from guile, a few figures betray the same telltale glaring media. Used with a sparing hand, as they are, they do not ruin a rug, but they are none the less a blotch upon its fair repute. The theory is, so far as concerns the new Kurdistans, for instance, that these few mineral dyes are bought by the weavers from some traveller or agent by chance and inadvertently, and without knowledge of their character. Otherwise they would hardly be used for[Pg 39] a few figures in a finely woven piece, where all the other dyes are vegetable.

One expert Armenian has a sure test for mineral dyes in his tongue. When in doubt he cuts a bit of wool from the rug, nibbles it a minute or so, and then pronounces his sure verdict. But the test is a delicate one, and the fruit of knowledge is, presumably, bitter.

Again, in speaking of colours and shadings, it may be interesting to know why solid colours so often come in streaks, changing abruptly, for instance, from dark blue to light blue, or dark red to light red. You may have any of several explanations: that the weaver, dipping his wool into the dye, stopped, for any trivial word or interruption, and the wool took on a stronger hue; or, that another hand or one of the women or children took up the work; or, again, that the plant, from which he bruised that particular hue, gave out in his back garden. Any of these reasons may be right. But the more credible one is to believe[Pg 40] that the artistic weaver knew how effective is this change of colour, and what a pleasing, changing, varying light and shade it gives to his masterpiece.



[Pg 41]



[Pg 42]


[Pg 43]

Chapter V


To describe in detail the characteristics of all the classes of rugs and carpets that have been mentioned would be hardly possible, even with a hundred object lessons. The peculiar features of some of them, however, may be noted. But first be it observed that the term “antique” as applied to rugs is generally sadly abused. A rug is not beautiful simply because it is old. It must have been fine when new, it must have been carefully preserved, and it must rejoice in a ripe old age. Time must have dealt kindly with it, and only softened and mellowed its original beauties. Let the antiques which are but rags and tatters, however valuable for their design, hang in the [Pg 44]museums, where they belong! The only merit of one of these genuine remnants of three or four centuries ago is in their originality of design. They were creations and not imitations, and made by true artists and not merely skilled weavers. Choose you, instead, a more modern rug of fine quality which will improve from year to year as long as you may live to enjoy it.

It may also be premised that the sizes of rugs run from about three feet to six feet wide by four to ten feet long. Few rugs approach squareness, and rugs wider than seven or eight feet are classed as carpets.

Some of the most beautiful pieces used to come, and still do, in the form of “strips,” “hall rugs,” or “stair rugs,” according to trade parlance. They are worthy of a better name, which is their Persian term, “Kinari.” They were made in pairs to complete the carpeting of a Persian room, being placed on either side of a centre rug, with two shorter strips at the top and [Pg 45]bottom. More fine specimens of these long strips are now to be found than of smaller sizes, and they should not be neglected by the collector. By artistic arrangement and device they will accommodate themselves to almost any house, somewhere, and few choicer prizes can be bought to-day.


Larger Image

Plate IV.
From the Collection of the Author
Size: 2.4 x 3.1

This is apparently one side of a pillow. The other side, which is also in the possession of the author, is exactly similar, except that the colours are reversed, the medallion being red and the corners blue. This mat has 33 to 36 knots to the running inch, making over 1,000 to the square inch, or more than a million knots in the small piece.


The Persians are eminently the best rugs to buy. They are usually finer and more closely woven than the others, and more graceful in design, and seem to show a more refined and aristocratic art. The Kirmans would be the first choice, and are to the rug dealer what diamonds are to the jeweller, a staple article which he must keep in stock, and which finds a ready sale. But even were it possible to buy a true diamond Kirman, the very catholicity of taste to which diamonds and Kirmans appeal detract from their merit in the eyes of those who seek for more individuality. For the new Kirmans, fine, soft, and clean as they look, are all very much alike, and mostly[Pg 46] copies or variations of a few particular antique forms, with a floriated medallion in the centre, or a full floriated panel, and floriated corners. A familiar design is a vase of flowers in graceful spread, with birds perching on the sprays. Or, again, they show some adaptation of “the tree of life.” This symbolical figure appears in many forms, now denuded of its leaves like the “barren fig tree,” and covering the whole rug, and now in smaller form as “the cypress tree,” or the sacred “cocos,” three or more to each rug, in full foliage and looking for all the world like certain wooden fir trees. It needs only the combination of these trees with the stiff wooden animals, far more wonderful than Noah ever knew, and tiny human figures, which might be Shem, Ham, and Japhet, all of which adorn these rugs, to remind one of the Noah’s ark of childhood. Representations of birds, men, and animals never appear on Turkish rugs, the explanation being that the Turks, as Sunna Mohammedans, the orthodox sect,[Pg 47] are opposed to them on religious grounds; while the Shiites, the prevailing sect in Persia, have no such scruples.

But before leaving the subject of the Kirmans, be it well understood, by the wise and prudent, that not one out of a thousand, or indeed ten thousand, of those on the market to-day (and they are as common as door-mats) has any pretence to genuineness. They are faked in every way. They are washed with chemicals to give them their soft colourings, they are made by wholesale and, it is said, in part by machinery, and they are no more an Oriental rug than is a roll of Brussels carpet or an admitted New Jersey product. To the credit of whom it may concern, it must be stated that the dipping, washing, and artificial aging of these commercial pieces is mostly done by cunning adepts in Persia before their works of art are exported. Only an expert’s advice should be relied on in buying a Kirman, to-day, and even that should have a good[Pg 48] endorser. The distinction between Kirmans and Kirmanshahs was founded in fact and was important. But the latter term as now used in the trade is only poetical. It is the same new Kirman euphemized. No other rugs except silk rugs, which come under the same ban, have proved such a profitable swindle to unscrupulous and ignorant vendors, and have given a bad name to the dealers who try to be honest in their calling.

The Sehnas are highly prized by the Orientals and Occidentals. Old examples are uncommon and are very choice. “Their fabric gives to the touch the sense of frosted velvet. They reveal the Meissoniers of Oriental art,” says a writer on the subject. Some of these come in very small sizes, like mats, two feet by three. They have a diamond design, the centre being a graceful floriated medallion on a background of cream, yellow, red, or green, with floriation at the corners, making the diamond. They are the most exquisite of[Pg 49] Persian gems, and are further considered in another chapter.

The Sehnas have the nap cut very close, wellnigh to the warp, and are therefore often too thin for utility. They do not lie well on the floor, and by reason of their short nap look cold and lack richness and lustre. If you can find a choice one, however, and if, happily, as sometimes occurs, it may have a little depth of nap, you will own a pearl of great price.

The Khorassans are very soft and thick. They generally show the palm-leaf or loop design in their borders, and are altogether desirable. Their colouring almost always inclines to magenta, but time subdues this to a delicate rose. Time has also subdued most of the specimens offered, to the sad detriment of their edges and ends. The ends are very seldom perfect, and age seems to bite into the borders of the Khorassans with a strange and voracious appetite. It is well to consider these defects in your choosing.

[Pg 50]The Serabends and their class have one border peculiar to themselves and a centre of double, triple, or multiple diamonds in outline, in which are scattered irregular rows of small figures, generally palm leaves, so called. This peculiar figure has three or four different names, the palm leaf, the pear, the loop, etc. It was originally worked into the fabric of the finest Cashmere shawls, and represents the loop which the river Indus makes on the vast plain in upper Cashmere, as seen from the mosque there, to which thousands made their pilgrimage. It was thus intended as a most sacred symbol and reminder. The Serabends are firm in texture, lie well, and are most satisfactory. Sometimes, however, the green in them shows the faults of an aniline dye. Their designs are peculiar to themselves, but never become monotonous. The palm-leaf pattern is of course common to many kinds of rugs. But the varieties in the form and size of it are infinite.


Larger Image

Plate V.
About forty years old
From the Collection of the Author
Size: 3.6 x 5.10


[Pg 51]The Shiraz rugs are warm in colour, lustrous, but rather loosely woven. Many of them show the “shawl pattern,” small horizontal or diagonal stripes. These striped rugs, however, are always wavering and irregular in design and soon tire the eye. They are well passed by. Reproductions of the old Shiraz designs with the centre field filled with innumerable odd, small figures used to be common a few years ago. They were very rich and handsome. Almost all of them, however, have the great defect of being crooked. They will puff up here or there, and, pat, pull, or pet them as you may, it is hopeless to try to straighten them. They are frequently called Mecca rugs, on the generally accepted statement that these are the rugs usually chosen to make the pilgrimage to that shrine.

The Youraghans and Joshghans (Tjoshghans) possess the general excellences of the best Persians, but they are not commonly seen. The Joshghans will show in[Pg 52] their field a light lattice-work design with conventionalized roses, or graceful diaperings and patternings, of the four-petalled or six-petalled rose. The Persian rose is single, of course, and appears in many simple forms. The Joshghans might be the prototypes of some of the old Kubas or Kabistans, except that floriation was replaced by tiling and mosaic work in the Daghestan region.

The Feraghans are not as finely woven as the Serabends, and on that account, primarily, yield to them in excellence. But old Feraghans often come in smaller sizes than the Serabends and in more desirable proportions. On the other hand, while Feraghans are generally of a firmer quality, there are also antique Serabends heavy and silky. Between the two it would be little more than to choose the better specimen. While the Feraghans have no accepted border to distinguish them, they have a most marked characteristic in the decoration of the field. It is[Pg 53] a figure like a crescent, toothed inside; it might be a segment of a melon. But more than likely it was originally a curled-up rose leaf; for the rose, variously conventionalized, is most common to this class. There is generally an indication of a trellis, on which the roses are formally spread. But the curled leaf is almost always in evidence, however varied or angular it may be drawn.

The Persian Mousuls are perhaps the best rugs now to be had for moderate prices. The region where they are made, being partly Turkish and partly Persian, gives them some of the characteristics of each nation. But the choice ones are always offered as Persian; and the designs of most of them are distinctively of that country, with frequent use of Serabend borders, Feraghan figures, etc. Their centre field sometimes contains bold medallions, but generally it is filled with palm-leaf or similar small designs, which[Pg 54] in themselves are quite monotonous, except as they are diversified and made beautiful by graduated changes of colour in both the figures and background. Sometimes these streaks of varying colour make too strong a contrast, but generally they shade into each other most harmoniously, and, the nap being heavy and the wool fine, these rugs are eminently lustrous and silky. They have no rivals in this regard except among the Beluchistans and treasured Kazaks. As you walk around them they glow in lights and shades like a Cabochon emerald. One of their distinguishing designs is a very conventionalized cluster of four roses, the whole figure being about the bigness of a small hand. There is a rose at top and bottom and one on either side, with conventionalized leaves to give grace. The design is recognizable at a glance, and is wellnigh as old as Persia. For the rose is conceded to be Oriental in origin, and if it is not primarily a Persian[Pg 55] flower, it belongs surely to her by virtue of first adoption.[1]

The designation of certain rugs as Kurdish or Kurdistan has been used indiscriminately, yet they are by no means the same, and between the two classes is a well-marked distinction which should be recognized. Kurdistan is a large province in northern Persia, with a protectorate government both Turkish and Persian, and with the Turkish inhabitants in the ratio of about two to one, according to the geographers. The Kurds constitute only a small but most important part of the population. They are generally spoken of as “a nomadic tribe,” or more frequently as “that band of robbers, the Kurds.” Regardless, however, of their morals or habits, by them are made [Pg 56]characteristic, coarse, strong, and often superb rugs which are properly called “Kurdish.” On the other hand, the Persians in Kurdistan make a finer class of rugs and carpets, which are known as Kurdistans. These latter have been praised by an eminent authority as “the best rugs now made in Persia and perhaps in the East.” They are certainly bold and splendid in design, beautiful in colouring, and of great strength and durability.

The Gulistans are thick, heavy, and handsome, with striking designs, frequently like the flukes of an anchor, on a light ground. They are not common now even in modern weaving.

There are many other Persian rugs which might be further specialized and considered. But such old commercial names as Teheran, Ispahan, etc., can in fact only be differentiated by an expert; and when experts disagree, as will frequently occur, and when they are at a loss to decide whether an important specimen[Pg 57] is an Ispahan or a Joshghan, classification becomes obscure to the layman and even to the collector; and he will wisely avoid the complexities of such discussion. So, also, Sarak rugs are rarely seen now save in modern reproductions, and must be passed by with the same criticisms as apply to the new-made Tabriz.

[Pg 58]



[Pg 59]



[Pg 60]


[Pg 61]

Chapter VI


The Daghestan rugs of Caucasia are only second in importance to those from Persian looms. An opinion is reserved, nevertheless, regarding antique Turkish weaves, which are hereinafter considered.

If history does not satisfactorily prove that the Caucasus was originally the northern part of Persia (as may have been, under Cyrus), Persian dominance and influence may be demonstrated, in textile art, by rug borders, patterns, and designs. The Shirvans, Kabistans, Chichis, Darbends, Karabaghs, all exhibit pronounced Persian characteristics, and show the educational power of the mother country of this [Pg 62]handicraft. Fineness of weave, delicacy of hue, and chaste simplicity of design are distinguishing features of this group. But, as contrasted with the Persian patterns, the Persians use for their detail roses, flowers, palm leaves, etc., while the Caucasians gain similar effects from geometrical figures, angles, stars, squares, and hexagons, with small tilings, mosaics, and trellisings. The true and the beautiful was never better demonstrated by Euclid through angle, square, or hypothenuse. An old Chichi rug, like a drawing of Tenniel’s, will prove what grace may come without a curve and by angles only.

It is unfortunate that the best rugs of the Caucasus come from the large province of Daghestan, and that that general term is applied to them indiscriminately. Twenty or more years ago most of the Oriental rugs which were sold here to an uneducated and unappreciative public came by way of Tiflis, and for lack of knowledge were all branded with the common name of Daghestan. [Pg 63]Thousands of beautiful Kabistans, Shirvans, Bakus, etc., were then sold for a song under the one arbitrary title. They would be priceless to-day, and yet the former commercial, vulgar use of the name leaves it in undeserved disrepute. As used in this chapter, it is intended to mark a distinction between certain of the Caucasian types, which it properly represents, and the Russian types from the same region, which are illustrated in the Kazaks and Yourucks.


Larger Image

Plate VI.
Thirty or forty years old
From the Collection of the Author
Size: 4.5 x 5.6


What may have become of all the fine Kabistans, which were forced upon the market years ago, is a question. Are they all worn to rags and lost to the world? Or do they still turn up at chance household auctions? Many fine specimens may be so discovered, dirty, disguised, and disreputable, but easily reclaimable and made anew by washing. There is a theory, also, that many choice pieces came to San Francisco in the ’seventies and ’eighties, and are lost to sight and memory somewhere in [Pg 64]California. A collector might well explore this home field.

Too great praise cannot be given to the old Shirvans, with their “palace” or “sunburst” pattern; to the Chichis, with their mosaic work, worthy of Saint Mark’s Cathedral; to the Karabaghs, with their flaming reds; or to the Kabistans, with their soft, light tones of colour, made softer still in contrast with ivory and creamy white. These are the despised Daghestans which were, and for which the collector may now vainly search abroad.

It is not always easy to distinguish between an old—or middle-aged, may we say?—Shirvan or Kabistan. Many of their designs are common property, and it is the cleverer weaver who executes them the better. This broad statement may be made by way of a test: the best of the Shirvans are rather loosely woven and thin. The Kabistans are of finer weave, are firmer and heavier, and lie truer on the floor.

[Pg 65]Two classes of rugs from the Caucasus have been referred to as Russian, the Yourucks and Kazaks. There is no authority for the distinction except in the rugs themselves. They prove their case from their thickness and iron durability, from their sombre or strong red colouring, and from their daring crude and simple designs. In their utility they bespeak an article of warmth and weight, and in their art they represent a barbaric simplicity like a Navajo blanket. Kazak and Cossack are almost synonymous terms; and the Cossacks, the Kurds, and the Indians have something of kinship in weaving, at least. But the Kazak rugs are not all crude, by any manner of means. If strength is their first characteristic and strong primitive pigments in rare greens, reds, and blues; and if their patterns are simple and angular;—none the less, in antique specimens, much originality was shown in the drawing of their borders, and soft browns and yellows with ivory white appeared in their colouring.

[Pg 66]Of the Shirvans, Chichis, etc., ordinarily offered, there is nothing to be said. They are cheaply and roughly woven, and made only to sell. They are disposed of by the thousands at auctions, and piles and piles of them fill the carpet and department stores. Be it said to their credit that they will outwear any machine-made floor covering; that they are good to hide a hole in an old carpet; that they help to furnish the bedrooms of a summer cottage; that they are most useful in the back hall; and, in fine, that they are better than no rugs at all. Yet, on the other hand, be it well understood that they are not, as frequently advertised, “exquisite examples of textile art,” and that fine Oriental rugs are not to be bought at “$6.98” apiece.



[Pg 67]



[Pg 68]


[Pg 69]

Chapter VII


Babylon or Egypt may have woven the first carpets or floor coverings, and China of course worked early in the same field. But Persia acquired the art quite independent of China, and well in the beginning of the long ago. Indeed, the Chinese industry practically ceased to exist many centuries back, and was transferred to northern Persia, where the history of this handicraft has its true beginning. From Persia all other countries have drawn their knowledge and inspiration, and however much they may have endeavoured to create and to evolve new figures and new designs, even the oldest examples of their art must concede something to Persian influence.

[Pg 70]The Turks, above all others, have shown themselves the most apt scholars, and indeed in many lines have improved upon their teachers. The choicest specimens of Turkish weave are as rubies to the other precious stones, rarer, more brilliant, and more costly than diamonds. Though not so closely woven as some of the Persians, they are wonderfully beautiful in artistic picturing and in their own Oriental splendour of colour and design. Such in particular are the antique Gheordez, as splendid in rich floods of light as the stained-glass windows of a cathedral. They are the finest woven and have the shortest nap of their class.

Here is the description of one taken from a catalogue of twenty-five years ago: “Antique Gheordez Prayer Rug. Mosque design, with columns and pendant floral lamp relieved on solid ground of rare Egyptian red, surmounted by arabesques in white upon dark turquoise, framed in lovely contrasting borders.”


Larger Image

Plate VII.
Prayer Rug
From the Collection of Mr. George H. Ellwanger
Size: 4.6 x 5.11


[Pg 71]Another is pictured as: “A flake of solid sapphire, crested by charming floral designs in ruby on ground of white opal. The mosaics and blossom borders are toned to perfect harmony.”

These word pictures are in no way exaggerated, and only help to portray the glories of the old Gheordez, with their graceful hanging lamps, as wonderful as Aladdin’s, in a vista between pillars of chalcedony or onyx. They came in the form of prayer rugs generally, and a pronounced feature of those more commonly seen is a multiplicity of small dotted borders. The older and finer examples show borderings of far more graceful and artistic drawing.

The antique Koulahs and Koniahs, though not so finely woven, have mostly the same superb centres or panels of solid colour as the Gheordez, and vie with the latter in the splendour of their hues, if not in the delicacy and intricacy of their designs outside the central field. The[Pg 72] Koulahs may generally be recognized by a narrow border, which is peculiar to themselves and is almost invariably found on them. This consists of a broken line of little tendrils or spirals quite Chinese in character, and looking much like a row of conventionalized chips and shavings. It is so odd and distinctive that once seen it can never be mistaken. The Koniahs also have little figures which are quite their own, and which usually appear somewhere in the central design. They are small flowers each on a single stem, and the flower has commonly three triangular petals, like an oxalis or shamrock leaf. It is quite unlike the blossoms which besprinkle other rugs. With this, often come crude figures of lamps like miniature tea-pots. The Ladiks display all the colours of an October wood, and complete the group of Turkish old masters. Not a few of them have also a unique border in the form of a small lily blossom. Experts speak of it familiarly as the “Rhodian border,” but its origin is altogether obscure.


Larger Image

Plate VIII.
Prayer Rug
From the Collection of Mr. George H. Ellwanger
Size: 3.11 x 5.6


[Pg 73]These words in testimony to the beauties of Turkish rugs may be offered simply by way of guide-posts to lead to some museum. A few battered and torn war-flags of Gheordez or Ladiks are occasionally offered on the market, but the best of them lack all character and colour, and show only the bold design and holes and strings and naked warp.

Just which particular Turkish rugs are properly classed as Anatolians it is hard to say, Anatolia being so large a province. The term as commercially used is only as comprehensive and expressive as “Iran” applied to the Persians. It is generally misapplied to an uncertain class of old, worn, and tarnished remnants or new coarse prayer rugs, ruinous of harmony with their magenta discords. Yet many of the “mats” are rightly called Anatolians, and, premising a later chapter, one of the greatest delights of collecting was to look[Pg 74] over a pile of them, with the never-failing hope of finding some bright particular gem. And these mats are truly the little gems of Turkish weaving, and in accordance with the Oriental fondness for jewels and precious stones the suggestion that they represent inlaid jewelled work has been well imagined. But here again we cry, “Eheu fugaces!” They have gone. It is idle to look over the pile. There are no good ones for sale. One explanation of their scarcity is in the fact that the Armenian dealers have a weakness for these small pieces themselves, and are wont to indulge their fondness for colour and sheen by keeping the choice ones for their own use. So the mats of commerce are either new, coarse, and crude and offensive with arsenical greens and aniline crimsons and magentas; or they are but soiled patches and bits of old rugs sewn together. Caveat emptor! and let the buyer look at their backs before purchasing.

The old Melez rugs, with characteristics [Pg 75]peculiar to themselves, are of almost like importance to the Koniahs and Koulahs. Frequently they have a suggestion of the Chinese in their figures and decorations. You will find symbolized dragons pictured on them, also the cypress tree; while in colour they form a class by themselves, and exhibit shades of lavender, heliotrope, and violet such as no other kinds may boast. Whatever this dye may be, and whatever tone of mauve or lilac it may take, you will find it only in the Melez, a few Bergamas, or in some old Irans, whose race is practically extinct. Worthy modern Melez are still to be had, and will improve as they wear; if only they are firm in texture and do not flaunt the battle-flag colours of Solferino and Magenta.


Larger Image

Plate IX.
Forty or fifty years old
From the Collection of the Author
Size: 3.10 x 5.3


The Bergamas come mostly in blues and reds, most prominently set out by soft ivory white. One of their recognized patterns is quite individual, and readily marks their class. It is a square of small squares marked off like a big checker-board.[Pg 76] Other small pieces are almost square, with the field in mosaic-work or flower blossoms. In the fine old specimens, which used to be, the Bergamas rioted in superb medallions or in a floriated central figure like a grand bouquet. As a class, their merit is softness and richness. Their defect is that of the Shiraz, a proneness to curl and puff themselves with pride. The fault is caused by the fact that their usually artistic selvedge is too tightly drawn. Skilful cutting of the selvedge and new fringing will correct the error.

Some old and some excellent new Bergamas have lately been in evidence in the stocks of the Oriental dealers. Howsoever or wheresoever they come, the collector may well take courage from their appearance and apply himself to the chase with renewed zest.



[Pg 77]



[Pg 78]


[Pg 79]

Chapter VIII


The geography of the carpets and rugs thus far considered has included a very considerable area.

Any traveller or collector who may have journeyed in fact to the regions where they are made may well have stories to tell, for his wanderings will have led him into strange lands and wild places.

But the remaining classes of rugs, which we are wont to see lying gracefully in front of our hearths, as tame and peaceful as kittens, have come from still farther and wilder regions of the world; and the wonder is that we see them at all or are permitted the privilege of treading on them. The Turkestan class, so far as our subject is concerned, carries us east from Persia,[Pg 80] through Afghanistan and Beluchistan even into China. They are Oriental in very truth, and at first blush, it would seem, should be more crude and barbaric in their art. But as compared with the bold, rough, and rude weaves and patterns of the Russian Caucasians, they are, as a class, most refined and delicate in design and fine in texture.

It has been said that “whoever has seen one Bokhara rug has seen them all.” Their set designs and staple colouring have been so long familiar that we have lost respect for them. There are the well-known geometric figures for the centre, smaller similar figures for the borders, and a mosaic of diamonds or delicate traceries of branches for the ends. Choice examples, like the stars, differ from one another in glory only. The variations evolved from the one conventional design are almost infinite; and the many shades and tones of red which are used bring to mind the paintings of Vibert and his [Pg 81]wonderful palette of scarlets, carmines, crimsons, maroons, and vermilions.


Larger Image

Plate X.
From the Collection of the Author
Size: 4.10 x 8.3


Some of the rare old Bokharas come in lovely browns and are almost priceless in value. Sad to say, it remained for an American vandal to discover a process of “dipping” or “washing” an ordinary rug so as to imitate these rare originals, and many dealers unblushingly sell these frauds. To wear imitation jewelry is far less reprehensible. Happily the trickery is generally distinguishable because the “dip” or stain, whatever it may be, is apt to run into the fringe or otherwise betray itself. The wise buyer will reject with scorn any rug, under whatsoever name offered, which shows no other colouring than various shades of chocolate brown. No such uniform brown dyeing ever characterized any class of rugs. Even the brown Bokharas which are in museums show some other tints with their brown tones.

Good Bokharas, like good Kirmans, are undeniably beautiful and of great value,[Pg 82] but the mere fact that both are considered staples in the rug trade tends to detract from their artistic value; and that they are so generally doctored, disguised, and perverted puts them in bad repute.

The Yamoud-Bokharas come in larger sizes than the others of their type; are not so fine in texture, but thicker and firmer. Their designs are larger and bolder, and they show a most becoming bloom. They also display green and even yellow in their colouring, which is not usual in Bokharas. Their selvedge is beautifully characteristic. In Bokharas proper the adornment of the selvedge usually is on the warp; as in the Bergamas and Beluchistans. In Yamouds the selvedge is almost always carried out in wool with like skill as that given to the rest of the piece.

The Afghans are a coarser edition of Bokharas, and may be mostly considered for utility. They come in large sizes, and almost square; have bold tile patternings, and in the finer examples are plush-like[Pg 83] and silky. These are still to be had, but many modern ones are dyed with mineral dyes, and their bloom is meretricious. The chemist has waved his magic wand over them, not wisely but too well.

The Beluchistans are somewhat akin to the Bokharas, and like the latter rejoice in reds and blues in the darker tones, while they display greater variety in their designs. These are ordinarily crude and simple, but in the old exemplars they were of considerable variety, and their wealth of changing colours in sombre shades was rich beyond the dream of avarice. “Lees of wine,” “dregs of wine,” “plum,” “claret,” “maroon,”—these are terms which have served to describe their prevailing colours. The adjectives are still applicable and may give some idea of the colourful effects which are obtained from their stains of brown and red and purple. For decorative effect, their deeper tones make most harmonious contrast with the subdued and softened Persians and old Daghestans. In[Pg 84] many specimens, new and old, white, both blue white and ivory, is used in startling contrast. It makes or mars the picture, according to the artistic skill of the weaver. The wool used in the good Beluchistans is particularly soft and silky, and lends to them their unique velvety sheen. No other varieties show it so perfectly, although antique Kazaks have their particular plush, and the Mousuls with their depth of pile have a shimmer and shifting light which is their especial artistic feature. The distinction may not easily be formulated; but, nevertheless, the sheen of the Beluchistan is one beauty, while the play of light and shade on a Mousul is another pleasure to the eye.

In the Bergama rugs the weaver does not disdain to spend some toil and time upon the selvedge; and this, even in small specimens, is commonly four to six inches long, carefully woven in white and colour and with occasional ornamentation. In this selvedge a small, elongated triangle is[Pg 85] frequently embossed in wool, with the commendable purpose of avoiding the “evil eye.”

But in the Beluchistans the maker “enlarges his phylacteries, and increases the borders of his garments.” He goes even to greater pains and trouble in the elaboration and finishing of his selvedge. It is often prolonged to eight or ten inches in moderate-sized rugs, and is woven into most interesting patterns and stripes of colour. It is literally carried to extremes. It may seem an act of vandalism, but the wise and stoical collector will do well to eliminate all but two or three inches of it and have a skilful weaver overcast and fringe the ends. Selvedge, however adorned, is utilitarian only, and, like useless fringe, it must not be allowed to detract from the proportions and beauty of the piece itself.

For the comfort of the collector be it known that within the last year or two, many fine Beluchistan mats and small rugs[Pg 86] have been secured somehow by the wholesalers and are in evidence in the retailers’ stock. Beluchistan, evidently, is one of the remote regions last to be drawn upon, scoured, ravaged, and exhausted. The opportunity should be improved by the provident buyer.

The Soumac or Cashmere rug calls for no further description than a Cashmere shawl. With the exception of choice antique specimens which time has chastened and mellowed into pictures in apricot, fawn, robin’s-egg, and cream colours, the Cashmeres are rather matters of fact than of art.

What are known as Killims, or Kiz-Killims, the better class, are hard fabrics akin to the Soumacs except that they have no nap on either side, and are double faced. They are mostly Caucasian and Kurdish, with the bold designs of those classes, or they come in the beautiful, delicate patterns of the Sehnas. In their crudest and strongest Kazak figures they[Pg 87] appear in the most brilliant pigments, with soft reds, rose, lake, and vermilion for contrasting colours, splashed together as on a painter’s palette. Of course they lack the sheen of a rug, but their colour effects are marvellous. While generally used for portières and coverings, they are perfect rugs for a summer cottage, being most durable, and are worthy of attention. Moreover, fine antique examples are still to be had. Some collector might be the first to make a specialty of them and garner them before they pass; the end of the Oriental weaver’s pageant. The usual warning, however, must be given, that they are often cursed with the barbarous magentas hereinbefore mentioned, a colour which would ruin a rainbow.

The products of Samarkand are quite out of the ordinary, and thoroughly Chinese in character. Except by association and classification they have no resemblance to the Turkestan or any other division. They form a class by themselves, the[Pg 88] legitimate successors of the old Chinese rugs, long gone by. They are very bold in design, and in colour tend to yellow, orange, and various soft reds. An inferior make of Samarkands often appears under the title of Malgaras. They have neither quality nor colour to commend them.

But there are old Chinese rugs also. Most of them are in the conventional blue and white, with simple octagonal medallions, with no border to speak of, and with little strength of character. They are coarsely woven and have been so commonly imitated by machine reproductions in English carpetry that even blue and white originals have small merit to boast of. There were, and doubtless still are, Chinese rugs of far more importance. Many are noted in the catalogue of a sale in New York City no longer ago than 1893. From one item remembered, they showed various beautiful colourings, far beyond the simple white and blue, and in design displayed much of the[Pg 89] artistic strength, grace, and beauty of the old Chinese porcelains. It is a mystery where these rugs lie hidden. No one boasts of owning them or claims credit to even a modest $10,000 antique specimen.

[Pg 90]



[Pg 91]



[Pg 92]


[Pg 93]

Chapter IX


However a man may justify himself for collecting rugs, regardless of his success, of his needs, or of his income, there would seem to be no danger of any one making a specialty of buying carpets. Except to millionaires or for clubs and palaces, space would absolutely prohibit, if the housewife did not. The nearest that the enthusiast might approach to such an ambition would be in the accumulation of hall strips; which has its own temptations, quite within the possible.

And yet the term “carpet” is an elastic phrase, and any piece which exceeds six or seven feet in width and of greater[Pg 94] length, is entitled by courtesy to be named a carpet. It may be said that a rug, like a baby, ceases to be a rug at an uncertain size, and then becomes a carpet. But carpets in the larger dimensions, ten by twelve feet or more, as ordinarily understood, are only herein considered. They are really articles of utility first and always, and must answer to certain measured requirements. Such is the accepted theory and practice. The buyer is wont to think that the merit or beauty of a carpet is of secondary consideration if only it fit the room. Here is a heresy. It is far better that the room should be made to fit or adapt itself to the perfect carpet.

If you would buy one, the best that you can do is to choose wisely. They are all of modern make, with very few exceptions. If you have one that is antique, you yourself have made it so, or you have inherited a ragged and neglected example of bygone years. The modern carpets, nevertheless, those still made to-day, are many of them[Pg 95] superb pieces, far outclassing any small rugs of the same weaving.


Larger Image


Larger Image

Plate XI.
From the Collection of the Author
Sizes: 1.10 x 2.10, and 2.1 x 2.11


The Kirmanshahs would come first, of course; closely woven, beautiful and soft in colour, delicate and artistic in their designs, they are the most perfect floor coverings for the salon, reception or music room. If they were only real! But very, very few of them are. They have all been treated with chemicals, and their beauty of complexion is just as artificial as any rouged and bepowdered courtesan’s. Unless you have one out of ten thousand, it has not come from a palace, but from a scientific laboratory.

Many of the Tabriz carpets lie under the same suspicion, and those of soft tones, claiming to be antiques, may be wisely questioned. But new ones come in clean, rich colourings, in fine designs, and are textile masterpieces.

The Kurdistan carpets of to-day are by far the best of all. They are more loosely woven, but they are so much the heavier,[Pg 96] and that is to be desired in a carpet. And they are honest. Their colours are beautiful, varied, strong, and true. It is claimed for the Kurdistans that some of their dyes are still well-guarded secrets; and it is true history of some years ago that many a bloody feud and murder grew out of cherished Kurdistan secrets of dyeing. Their designs are bold and striking, with grand centre medallion and corners, and a field artistically adorned. Money cannot buy anything better than a fine new Kurdistan; and thirty or forty years of wear should leave it better still.

Next to be chosen would be the Gorovans. They also show brave figuring with a strong centre medallion, characteristic zigzag corners, and angular ornamentations which are most gracefully carried out. Their colouring is usually in fine blues and reds.

Modern Feraghans come in large carpet sizes, and some antique ones are still to be had. But the Kurdistans and Gorovans[Pg 97] far surpass them in two important particulars. The Feraghans appear only in their own peculiar, small-figured designs, which are without strength or character on a large floor space. Besides that, being more closely cut than the others, if they do not soon wear out, they soon wear down, and begin to show the suspicion of their warp and their loss of tone and colour. They are beautiful carpets, nevertheless, and will practically last a lifetime. But the heavier they are, the better.

There are few other modern Persian carpets in large sizes which come in appreciable numbers for classification. There is a rather indefinite order of Gulistans, under which title many good nondescripts are sold.

There are also current Sultanabads, in very large sizes, well woven, on old models, to meet present uses.

Most other carpets are of Turkish weaving, whatever their names, and come under the general title of Smyrnas. Smyrna is[Pg 98] the centre of distribution for a great variety of cheap and coarsely woven carpets; but poor in quality as these may be, they should not be confused with the American machine product also known as a “Smyrna.” In the same class come the Oushaks, Hamadans, etc. There is nothing more to be said for them than to testify that they will wear better than a Brussels carpet, and give some distinction to a modest dining-room.

It is a far cry from carpets to saddle-bags, and yet these latter are of greater importance and interest to the collector. More valuable pieces of Oriental weaving are to be found among the diminutives than in the grand opera of textiles.

Beginning at the bottom, we find plenty of the little pairs of bags, twelve or eighteen inches square. They are donkey bags, carried back of the saddle, and generally appear in Shirvan make or, most commonly, in Shiraz weaving. The Shiraz often have considerable beauty and sheen and dark rich[Pg 99] colouring. But these very small pieces have little real utility or available artistic beauty. They never lie well, and only litter up the floor. They belittle a well-arranged room as would a frail and useless gilt chair. They are recommended for pillows, but we Occidental infidels associate rugs too closely with the foot to find them easy to the head. They are also advised for use as hassocks. But the hassock long ago disappeared, with or under the “what-not,” or behind “the horse-hair sofa.”

Other bags, used on horse and camel, come in more important sizes, as large as two feet by six feet or more. Exquisite specimens of Bokharas are found among these; artistic, antique pieces, woven as fine as needlework. A number of these seem to have come suddenly on the market in some mysterious way; and they are of every size within their small limits; because, as an Oriental has suggested, there are pony camels also. Another mystery about those camel bags would seem to be[Pg 100] that some are beautifully straight and therefore most to be desired, while others are so curved as to be impossible of use unless around the foot of a pillar. Here is a case differing from that of the ordinary crooked rug, because these bags were originally made straight and true. Overloading and overpacking have only sagged down the middle. I dare not say that the more the curve, the greater the age and the more the value; but it may be that curved Bokhara saddle-bags, passed by, by the Levite, are prizes to be picked up by the good Samaritan, and may be easily restored to normal rectitude.

But the term “saddle-bag,” whether for this animal or that, is confusing and altogether too generally used. It must be borne in mind that a bag was and is an article of universal utility to the Oriental. For all purposes of travel, journeying, or visiting, it corresponds to our valise or portmanteau of to-day; or, in aptest comparison, to our “carpet-bag” of fifty years[Pg 101] ago. And, according to the taste and means of their owners, these Persian, Armenian, or Turkish carpet-bags varied in size and beauty. A few rare old Caucasian small rugs can only be accounted for as valued personal rug-bags of their period.

Among these smaller pieces are alone to be found the most valuable of all the collector’s spoil, the small Sehnas. Very rarely they come in pairs, about two feet by three feet, and therefore could not have been used as bags for any purpose. They are pillows; and pillows of course play their important part in the ménage of the East. Besides the exquisite Sehnas, the finest of the Anatolian mats, as they are generally called, were used for pillows and not saddle-bags. The warp generally proves their purpose. When the warp runs vertically to the larger side, and ends in a fringe, that specimen was of course some sort of a saddle-bag. When the selvedge is at the shorter end you have the pillow.

[Pg 102]Among the other beautiful miniature specimens of textile art, which are still occasionally offered, are saddle-cloths. They appear mostly in beautiful Sehnas, and occasionally in fine old Feraghans and other Persian weaves. They are marred, however, for beautiful floor coverings by the necessary angular cut in them, through which the straps of the saddle passed. This is often skilfully filled in, in the case of choice specimens. But the blot remains. Their irregular shape also condemns them for the most part with the many admirable but irreclaimable crooked rugs.

These saddle-bags are frequently used for table coverings or for mural adornment. But in our modern house decoration rarely does a rug look well upon the wall. The Persians hang them instead of pictures, which is well. But they do not mix them with pictures on the wall, which is better, and shows good taste on the part of the Persians. A rug appears best upon the floor.

[Pg 103]The collector of small pieces to-day will do well to buy every bag or pillow of Bokhara or Beluchistan which may please his fancy. They are to be had now at modest prices, but unless all signs fail, they will soon become as rare as any of the other miniatures. You will look in vain for them with the vanished Anatolians and diamond Sehnas.

[Pg 104]



[Pg 105]



[Pg 106]


[Pg 107]

Chapter X


A justification of the method of selling rugs by auction has been offered in many forms and phrases. It is perhaps best expressed somewhat thus: Every number has a certain intrinsic value, and that is a basis price at which it should sell. But beyond that it may have an extra value, which, like beauty in general, is in the eye of the beholder. The beholder, therefore, who sees a rug to covet it should name his own price for it. It may be one of the specimens he lacks in his collection; it may fit this corner or that. Anyway, it is worth more to him than to the[Pg 108] lower bidder. Incidentally, the seller and the auctioneer gain the fair profits of competition.

Other arguments in favour of the auction have been advanced by the head of a great department store. His opinion is that the auction gives every one a chance to get the rug desired at a fair price. Tastes differ and prices differ, but the average of an auction is fair to both buyer and seller.

Regardless of theories, rug auctions, by whomsoever fathered or sponsored, thrive and flourish.

If the auction be the collection of such and such an Oriental, whatever his name, there will be a great deal of cheap stuff in his stock, and there will also be many choice pieces which he holds as the apples of his eye.

He buys from the wholesaler so many bales at so much per bale of say twenty pieces. In the bales of ordinary qualities the several items will average about the[Pg 109] same. But in the more expensive bales there is a good general average, with a few prizes added. They are like the two or three green firecrackers in the packs of our childhood. These special pieces in the high-priced bales give the seller his legitimate opportunity and profit. If these odd firecrackers please your fancy more than mine, and I am contented to choose the conventional red ones, it is for you to fix the value of the greens.

At an auction the apparent authority and ruler is the auctioneer, while the owner weeps cheerfully on one side and shrugs his shoulders in half-pathetic resignation at the sacrifice. In reality the auctioneer knows pretty well what he is about, and, if not, is quickly posted by the owner. It is no harm to say that if we cannot believe all that we read in the Bible, no more is it safe to take literally all that the auctioneer asserts. A recent skit in “Life” is pertinent (quoted from memory):—

[Pg 110]The wife. Look at this splendid bargain I bought for twenty dollars to-day. It’s worth two hundred.

The husband. Indeed! How do you know it is worth that much?

The wife. Why, the auctioneer told me so.”

A new plan of auction has been recently tried. You may buy in one or more lots at your own price, and if you do not wish to keep any, they may be returned within a certain number of days. You may bid ad libitum, recklessly as you choose; and if your choice be not all that your fancy and electric light have pictured it, you are under no obligation to keep it or pay anything on it; you may elect to change your mind and send it back. How this plan works in practice and finance has yet to be demonstrated. It would seem to be all on the side of the buyer and against the seller, who must lose many a bid from a bona fide purchaser at a lower figure. The matter of human nature doubtless figures in the problem, because there is some little[Pg 111] feeling of shame about returning an article bought in under competition, no matter what the guarantee may be.

As to the auctioneers, they are always glib of tongue, good-natured, and persuasive. That they are not canonically and absolutely truthful is perhaps not their fault. They certainly cannot know more about rugs than the few authorities who have made a study of the subject; and, as said before, they are generally prompted by the “consignor” of the collection. If only they would not call every rug an “antique and priceless specimen,” their individual consciences might be happier, and their audience less bored.

However, no matter what the audience, or how small it may be, there are always some there who will appreciate the difference between a four-dollar and a forty-dollar offering, and bid up the former to seven dollars and the latter to thirty dollars. Thus the auctions go merrily on and strike a general average. The skilful[Pg 112] auctioneer will feel the pulse of his audience with a quicker touch than the most renowned of doctors; and once assured of their class and position, wealth and condition, and what grade of merchandise they are willing to buy, the game is in his own hands, provided only that his audience is large enough. He should have at least a regulation pack of fifty-two in order to do justice to his own hand and skill, and in order to play off one card of his audience against another.

The auction has its own particular fascinations, and its own habitués and devotees in every city. The chronic attendants should be the most careful and conservative of buyers. But the artful auctioneer soon learns to know them, to recognize them among his clientèle, and to humour their whims, moods, and fancies. Sooner or later he will wheedle them into a bid against their better judgment, and then make good capital of the fact that such[Pg 113] and such a connoisseur had bought so great a bargain.

The question might be asked, impersonally and perhaps impertinently, What was the auctioneer’s influence at the Marquand sale? Was his the power? Was it due to the catalogue? or was it in the air; and the zeal of an eager audience?

The retail trade in rugs throughout this country is largely in the hands of Armenians, both fixed and peripatetic; but of recent years much of their business has been annexed by the department stores.

These various Armenian dealers are universally known for their shrewdness and cleverness as well as for other ingenuousness and natural courtesy. Except the heads of the carpet departments in some few large concerns, they know much more about their wares than other salesmen, and their personal, live knowledge gives a fillip of enthusiasm to the purchaser. They would control the retail trade in rugs, were it not that the department store has[Pg 114] brought against them its powerful weapon of per cent. The store asserts that it wants only its modest per cent on the cost of any article, no matter what its sentimental value may be. This may not be truth in its stark nakedness, but it has availed to draw to them a great deal of the trade in Oriental textiles.

The wholesale dealers are the most important factor in the question of distribution, for almost all the rugs sold in the United States must first pass through the hands of one or another of a dozen New York princes of the market. Large or small retailers may import some pieces directly from London, Paris, or Constantinople, but even the most important retailers buy heavily from the great Armenian wholesalers in New York City.

It is difficult to estimate and impossible to state absolutely the number or even the value of the Oriental rugs annually imported into the United States. The reason is that in the reports of the U. S. Treasury[Pg 115] as to “Imported Merchandise,” etc., Oriental carpets and rugs have no separate classification, but are included under the general heading of “Carpets woven whole for rooms, and Oriental, Berlin, Aubusson, Axminster, and other similar rugs.” It is quite a mixed company, but Oriental weaves as herein considered are at least distinguished as such, and differentiated from carpeting by the yard. They have also the distinction, with the others of their group, of paying a tax of ninety cents per square yard and forty per cent ad valorem, as against from twenty-two to sixty cents per square yard and the same forty per cent ad valorem for the various Brussels, Wilton, and Axminster floor-coverings coming by the yard, and not in one piece. And the duty on Oriental rugs, be it observed, is measured by the square yard, and therefore no record is kept of the number of pieces, or how many individual items of the four classes have been imported.

[Pg 116]Nevertheless, the statistics for the year ending June 30, 1902, show this general result: The total value of that year’s import of these “whole carpets, Oriental, Berlin,” etc., was a trifle below three million dollars. Two and a half millions of this value came to New York with only half a million left to divide between Chicago, Philadelphia, Boston, San Francisco, and other ports of entry. The supremacy of New York City as the Oriental rug mart for this country is easily manifest, although it is not so easy to estimate what proportion of the two and a half millions of value was in Oriental rugs and what in modern carpets. One expert figures the value of the Oriental rugs imported that year into New York as more than half the total, or perhaps two millions. It is as fair an estimate as may be had. Considerable as this amount may be, it seems much less than might be expected. It may perhaps indicate the cheap grade and low quality of most of our present acquisitions in this category.

[Pg 117]The gathering of the rugs by the buyers, in the first instance, involves great hardships, endurance, and even danger; and the deeper their incursions into new and strange territory and unopened and unexplored sources of supply, the more profitable their spoil, but the greater their toil. Beluchistan, as previously suggested, would appear now to be one of the remotest regions yet remaining to yield up a few new treasures to the persevering buyer.

These rugs so gathered to the centres of trade in Constantinople, Tiflis, and other distributing points, quickly find their way thence to New York, and help to make the magnitude and seeming wonderful complexity of the large wholesale depots. Whoever is fortunate enough to have the entrée to any of these great New York storehouses will be first among those who understand the importance, value, and appreciation of the Oriental rug.

[Pg 118]



[Pg 119]



[Pg 120]


[Pg 121]

Chapter XI


In addition to the many patterns, figures, devices and symbols, which are used for ornamentation, rugs and carpets are often embellished with hieroglyphic writing, somewhere in their field, and commonly at top or bottom. Not unfrequently complete borders are thus composed, as is evidenced in old Kirmans. These designs are so graceful in their many angles and occasional curves that they scarcely suggest mere lettering. Such they are, nevertheless; and our English script, with all its loops and turns and recurrent “lines of beauty,” would hardly avail for like effective results. It is but another proof of the artistic possibilities of angular lines and geometric figures, so often demonstrated by Oriental weavers.

[Pg 122]With few exceptions, all of these hieroglyphics are in the Arabic language, and are quotations from the Persian poets, with flowery sentiment, or from the Koran, in proper precept. But, as is more important, there will frequently be found in the corners of a choice piece, or elsewhere unobtrusively woven, the signature or cipher of the maker, with the date of the making. This at once gives distinction and value to such a specimen and exalts it above its fellows. It also calls loudly for an answer to the question of what such name and date may be. Very rarely can the dealer inform you, because he does not know. Here, then, is a great stumbling-block in the path of the collector. It may be worth while to go around it by way of a brief explanation.

The Arabic language has been the lingua franca of the East from the time that it succeeded Greek in the seventh century. It still retains its universality wherever Mohammedanism rules. Turkey may be[Pg 123] excepted from its sway, but, none the less, it is a most necessary language to-day in Constantinople. Its use by carpet-weavers is by reason of its catholicity; that it may be understood where their varying languages and unknown dialects would tell no story.

That Arabic is so generally known throughout the Orient is doubtless no greater marvel than that mere children in Paris speak French. But, however convenient, as an inter-racial and commercial language, Arabic may be to those accustomed to it, or naturally conversant with it, it is most difficult to learn by Western races. With ten years’ study one may become a good scholar, and proficiency may follow for the persistent few. This will explain why inscriptions, texts, and verses on rugs and carpets are meaningless, except to the most erudite; and except, also, to those who see in them only another phase of Persian ornament, strange, mysterious, arabesque, and beautiful.

[Pg 124]Regarding the date, often woven into an example which the artist thought especially worthy, it would seem that some simple formula might be given for its ready translation. This may be approximated, although it is not so easy a matter as might appear, and requires a few words on the subject of Arabian numerical notation. Their general system is similar to ours, and, corresponding to our miscalled “Arabic figures” of:

0, 1, 2, 3, 4, 5, 6, 7, 8, 9,

their digits are represented by

Both are read from left to right. These Arabic digits, however, are not always easily to be deciphered on a rug, on account of the spreading of the wool and consequent irregularity of outline, and also because they generally appear in modest size. The back of the rug will show the figures much more sharply than the face, [Pg 125]when there is a doubt. When the Arabic numerals are made clear, it remains to reduce this date to the corresponding one of the Christian era, by means of a complicated table.


Larger Image

Plate XII.
Thirty to forty years old
From the Collection of the Author
Size: 3.1 x 3.5


All Mohammedan dating (with exceptions not to be considered here, however interesting historically) is from the Hegira. The reckoning is not from the time of Mohammed’s “flight” from Medina (September, 622), but from a day about two months earlier; namely, the first day of that Arabian year. This beginning of the epoch, according to the best modern authorities, probably corresponds to July 16, 622.

Mohammedan chronology, however, is often expressed in other ways than by clear figures, and such florid records are most difficult to interpret. Again, in old manuscripts, on coins and on a few rare antique carpets, the date is written out in full, in so many words; as, for instance, “two-hundred-and-five-and-twenty-after-the-thousand.”[Pg 126] Intricate dates like these are to be solved only by an expert.

But when the year is in question, without regard to month or day, and when the year is written in legible figures, a rough formula for computing the corresponding Christian date is as follows: Subtract from the given Mohammedan year one thirty-third part of itself, and add to the remainder six hundred and twenty-two. Thus: A. H. 1196 = A. D. 1196 - 36 + 622 = A. D. 1782. This is accurate enough for all practical purposes, and involves no difficulty except the deciphering of the Arabic digits. The failure to subtract this essential one thirty-third part explains frequent misreadings by the ignorant dealer or uninitiated amateur. That six hundred and twenty-two must be added to the given Mohammedan date explains itself. But it must be remembered also that the Moslem year is lunar, and thus a little more than eleven days shorter than our solar year. Their reckoning[Pg 127] therefore gains one year in every thirty-three of our computation.

Modern commercial rugs of ordinary quality are occasionally provided with a date or other calligraphic figure to simulate the real signed and dated masterpieces. This trickery should never deceive even the most unwary, unless the piece is of exceptional merit; and then, there is no deception; or at least there is value received.

[Pg 128]



[Pg 129]



[Pg 130]


[Pg 131]

Chapter XII


Many kinds of rugs are made in part of camel’s hair, generally undyed and of a soft brown tone. They are praised as particularly desirable and durable, and antique specimens often showed a distinguished beauty. Modern examples are seldom improved by this addition to the wool. Camel’s hair, in the muggy days of summer, has the great fault of offending the nose and proclaiming not only that the “Campbells are coming” but that the circus and the whole menagerie is already here. If the camel’s hair part of your rug is soft and silky, it has been taken from young camels or from the camel’s belly, and the odour is hardly ever[Pg 132] noticeable. Of wool in rugs generally it may be said that the best is from the younger sheep, and the silkiness and sheen of the wool give those same characteristics to the rug.

Silk rugs, both antique and modern, fairly dazzle the eye with their beauty, but he who may afford one will needs afford also to furnish the surroundings for it in like magnificence. Otherwise all else grows pale and dull and leaden beside their refulgent glory. Place a piece of modern Dresden china side by side with a fine antique specimen of Chinese porcelain, and the garishness of the modern ware will give a pallid tone to the soft whites of the Oriental artist. But the fault is not with the older and perfect art; it is simply the old truth, in a new form, that evil colours corrupt and kill good colours.

Be that as it may, old silk rugs are almost priceless, and of value to a millionaire collector for their originality of design[Pg 133] and for their soft harmonies of colour which centuries alone can give. Modern silk rugs are mostly machine made, in part at least; are a detriment and a blot on any scheme of household decoration, and are always worth less than the price paid for them.

By experience we may best learn how to choose a rug. As, for instance: never buy a rug, least of all at an auction, without thoroughly examining it. See its back as well as its face, and so be sure that it has not been cut, and that there are no serious holes in it. Quite one-third of the good old rugs will show some rents or tears, often made by the grappling-hooks as the bales are shipped and transhipped. If these are no bigger than a silver dollar, a skilful repairer, of whom there are plenty, will readily remedy the defect. Also hold the rug up to the light to know that the moths have not eaten it. Look at the nap and see that it is not worn to the warp. Lay it on a board floor, if[Pg 134] possible, and apart from other rugs, and see that it lies flat and straight. None but those that are firm enough to lie well are desirable for use and general comfort. Of course many fine antiques are their own sufficient excuse for exception from this rule.

If in doubt as to whether a rug has aniline dyes or been doctored or painted, a handkerchief moistened with the tongue may sometimes discover the truth. Painting a rug is a device not unfrequently practised when the nap is worn down and the warp shows white.

Bear in mind that a good example may be so dirty as not to show half its merits. A sharp patting may scatter enough dust to display it in its proper colours, and you may thus, literally, unearth a treasure.

Remember, too, that rugs never look so well or show as clear and bright when hanging on the wall as lying on the floor. Therefore, test a rug spread out flat before[Pg 135] you in broad daylight. It is a trick of the trade to hold up one end of the piece exhibited and keep it waving to show its sheen. This is often a mere device to conceal its bad shape or other defects. If you are buying a rug for use on the floor, you should see it so displayed. Its sheen should be judged by walking around it and considering it in various lights.

Note that with few exceptions the fringe and selvedge on a rug were not made for beauty but for protection. When the fringe is ropy, long, or uneven, or the selvedge eaten into or ragged, do not leave the rug to its unkemptness, but trim it religiously. A man should have his hair cut and put in order at proper times; and the propriety of this observance is commonly preached on very many prayer rugs, where the comb is prominently pictured, to remind the devout that “cleanliness is next to godliness.” Indeed, the comb in various forms is so common a feature in the[Pg 136] angular arch of most prayer rugs that its suggestiveness almost detracts from their beauty. The counsel is most persistent.

Even the clean white fringe of a fine Persian is often so long as to need clipping. Two inches or so is a plenty. If more is left, the strings only curl under and show a ragged and broken line, and the rug never appears trim and orderly.

When the selvedge is gone, and the end borders or sides of the rug itself are encroached upon and sawed by the tooth of Time, more than half of the value and beauty of the piece is lost; but to preserve its usefulness it should be overcast and further damage prevented. Never buy a rug as a perfect or even choice specimen if any border at the sides or ends is gone beyond repair. Every border should have its corresponding end, and vice versa, or the piece is imperfect. Selvedge is of slight importance, but, like a woman’s skirts or a man’s trousers, it is unforgivable if worn or frayed. The side edges[Pg 137] which are otherwise still perfect are apt to become more or less ragged with wear. That is a detail, if the borders themselves are intact; and the edges only need overcasting before it is too late.

When the good housewife has the rugs and carpets beaten, let it be done on the grass, if possible, and not when they are hung on a line and so allowed to break with their own weight. Also let the severity of the beating be tempered with kindness and discretion. In winter, sweeping with snow will clean and brighten them most wonderfully.

This whole matter of cleaning is a neglected science and worthy of a thesis all to itself. The face of a rug will stand the slapping which is its usual punishment for being dirty; but do not forget, in the end, to stroke it, with the nap, and so soothe its feelings. Do not beat a rug or carpet on the back. That has no defence of nap, and you are liable to break the warp and loosen the knots. Frequent[Pg 138] sweeping is far better than the brutality of constant beating. The wool of a rug is really a sentient thing. However dead it may seem, it has a life and vitality all its own. It can be quickened, rejuvenated, and made alive again by proper washing.

Rugs in our modern houses easily accumulate dust and grime and smoke. But it is absurd to think that a rug is antique because it is dirty; or, more foolish still, that because it is dirty it is both antique and beautiful. Wash some of your treasures and you will wonder at their real glory and colour. Generally speaking, every rug should be washed about once a year. It is the Oriental custom; and carpets there are otherwise kept much cleaner than with us, by reason of many usages and observances. That the Orientals wash their rugs in cold water is not so. Wherever and whenever their laundering is done, the water is as warm as can be had, naturally.

[Pg 139]Milady washes her laces with her own fair hands, and delights in the task. The rug collector will do well, perhaps, to follow her example; except for the tender specimens, which must needs do without it, and the carpets, which are unmanageable. At all events, he will do wisely not to send his valued specimens to the ordinary carpet-cleaner. They may come back expurgated, but some virtue has gone out of them. The wool has lost its oil and life.

It is hardly within the province of this volume to prescribe the exact methods of washing. Wool soap will do wonders, it being always remembered to stroke softly with the nap, while the rug is drying. In Kurdistan and neighbouring provinces the rugs are first soaked in milk of some kind and then rinsed, cleaned, and rubbed dry. The milk gives back to the wool its essential oil, and it becomes at once soft, shining, silky, and alive with glowing colour. This process, simple as[Pg 140] it is, is kept as a profound secret by the few who know it in this country. Another Eastern method is to rub the rug with a mixture of rice-meal and oil, but the first recipe is by far the better.

Rugs must be cared for particularly as to moths. When they are in general use the moth will not corrupt, rust, or break through and steal, as may be paraphrased from the Scriptures. The criminal indictment against the moth in this regard cannot be drawn too strongly. He is the collector’s great enemy, because he destroys. Age and even wear only ripen the perfections of fine modern pieces. Carpets and rugs stored, or laid aside, are not mothproof, wherever they may be; unless they are treated as in the great wholesale houses, where they are lifted and moved once a week and protected with the odorous moth-ball.

When rugs have to be moved and packed frequently they should be folded differently each time, and not always in the[Pg 141] same creases. Otherwise, wear and tear will soon show in the folds. For many obvious reasons they always should be folded away with the nap inside.

Experience should teach the collector to appreciate and care for all fine examples which he may already have. There are few others to take their places. “Going! going! going!” has been said of them too often. Time, as auctioneer, now says of them, as of old Chinese porcelains, “Gone!” And that they should be even rarer than old china is quite understandable. The ravages of time deal more gently with porcelains than with rugs. Only breakage, not wear, moth, and abuse affects the former; and it is generally guarded in glass cases and dusted by the mistress herself. Your rugs are neglected, or left to the gardener’s heroic care and treatment. Use and abuse encroach upon the ends and edges of a glorious old masterpiece, and ere it is too late, it becomes but “a king of shreds and patches.”

[Pg 142]If there were new rugs to take its place, we might say: “The King is dead. Long live the King!” But there are no new ones worthy of succession. The royal line is virtually extinct.



[Pg 143]



[Pg 144]


[Pg 145]


Afghan rugs, 82
" "   , modern, mineral dyes in, 83

Anatolian rugs, 73
"     , commercial term, 73
"   mats, 73, 74

Angles, mystery of, 2
" , use of, 62

Angular ornamentation, 96, 121

Antique carpets, 94
"  rugs, not to be had, 6
" "  , term abused, 43
" "  , tones of imitated, 81
" "  , valuable for design, 44

Arabian digits illustrated, 124

“Arabic figures,” miscalled, 124

Arabic language, 122
"   "   , catholicity of, 122, 123

Armenians, appreciation of, 3
"   , as dealers, 113
"  hoard Anatolian mats, 74

Auction, arguments for, 107, 108
" , caution in buying at, 133
" , fascination of, 112
" , ways of, 109, 110

Auctioneer, 109, 110
"   , powers of, 111-113

Authorities, few available, 4

[Pg 146]
Babylon, first rugs woven at, 69

Bale, rugs by, 108, 109

Bead-bags, 23

Beluchistan rugs, 83
"   "   , silkiness of, 54

Bergama rugs, 75
" "   , defect of, 76
" "   , lavender in, 75

Bokhara camel-bags, 99
"    rugs, 80
"   "  , brown, 81
"   "  , Yamoud, 82
"    saddle-bags, crooked, desirable, 100

Border, Caucasian, Persian, 23
"  , classifying rugs, 27, 28
"  , crab, 29, 30
"  , dotted, Gheordez, 71
"  , Koulah, 27, 72
"  , Ladik, lily, 25, 72
"  , must have end to correspond, 136
"  , Rhodian or lily, 25, 72
"  , Serabend, 29, 50
"  "   , in Mousuls, 53

Borders in Khorassan rugs, defective, 49

Camel-bags, 99
"   "    , crooked, desirable, 100

Camel’s hair in rugs, 131

Cashmere, see Soumac

Carpet, 93
"  , room should fit, 94

Carpet-bag, 100, 101

Carpets, modern Persian, 94-98
"    "  Turkish, 97, 98

Caucasian rugs, characteristics of, 62
" "    , Persian influence on, 61
" "    , varieties of, 16
[Pg 147]
Chichi rugs, 61, 62
" "  , ordinary specimens of, 66

Chinese figures in Melez rugs, 75
" old, rugs, 88, 89
" weaving transferred to Persia, 69

China, first rugs from, 69

Collector encouraged, 76

Colours, brown, to be rejected, 81
" , chemical in carpets, 95
" , dark, of Beluchistans, 83
" , flaming red, of Karabaghs, 64
" , green and yellow, in Bokharas, 82
" , lavender, heliotrope, &c., 75
" , magenta, to be avoided, 75, 87
" of Beluchistans, 83
" "   Bergamas, 75
" "   Ladiks, 72
" "   Gheordez, 70, 71
" , red, of Bokharas, 80, 81

Comb, as symbol, 135

Cossack, like Kazak, 65

Crooked rugs, poetical version of, 21

Daghestan, confusing term, 61, 62
"   rugs, 61
"  "  , term distinguished, 63

Dates on rugs, 124, 127
"  " "  , Arabian digits for, 124
"  " "  , formula for reading, 126
"  " "  , intricate forms of, 125

Dealers, uncommunicative, 2
" , wholesale, 108
[Pg 148]
Design, checker-board, 75, 76
" , comb, 135
" , Feraghan, 24, 53
" , four roses, 54, 55 (note)
" , Koniah, 26, 72
" , lamp, 72
" , mosque, 70
" , no pattern for, 26
" , “palace pattern,” 64
" , “palm-leaf,” “pear,” loop, 50
" , “shawl pattern,” 64
" , “sunburst,” 64

Designs, as trademarks, 26
" , animals for, not on Turkish rugs, 46
" , geometric figures for, 80
" , Kazak, in Killims, 86, 87
" , mosaic-work in, 64, 76, 80
" , palm-leaf, in Mousuls, 53, 54
" , tile, 82

“Dipping” rugs to imitate antique, 81

Donkey-bags, 98

Dyes, aniline, mineral, 35, 36, 38, 134
" " "  , test for, 39
"   , black, 36
"   , brown, 37
" "  , imitated, 81
"   , green, 37, 50
"   , magenta, 36, 49, 87
"   , secret in Kurdistan carpets, 96

Edges, should be overcast, 137

Ends, importance of perfect, 136
"  , in Khorassan rugs, defective, 49
"  , should have corresponding borders, 136

Experts, disagreement of, 56
"     , no, 3
[Pg 149]

Feraghan carpets, 96
"    "  , small figures of, 97
" rugs, 52
"   "  , characteristic design of, 29, 52
"   "  ,  "   " "  illustrated, 24

Figures, see Design

Fringe of rugs, not for beauty, 135
"  " "  , trimming of, 136

Geography of carpets and rugs, 79, 80

Gheordez rug, 70, 71

Gorovan carpets, 96

Gulistan carpets, 97
"  rugs, 56

Hall rugs, desirable, 45, 93
" "  , Persian term for, 44

Hamadan carpets, 98

Holes in rugs, cause of, 133

India carpets, 17

Inscriptions on rugs and carpets, 121

“Iran,” as descriptive term, distinguished, 15
"  ,  a trade term like “Anatolian,” 73

Ispahan rugs, 56

Jewels, mats like, 74

Joshghan rugs, 51, 52
" "    , like Ispahans, 57

Kabistan rugs, 63
"   "    , distinguished from Shirvans, 64

Karabagh rugs, 64

Kazak rugs, 65
" "  , plush of, 84

Killims, 86, 87

Khorassan rugs, 49
[Pg 150]
“Kinari,” Persian term for “hall rugs,” 44

Kirman rugs, 45-48

Kirmanshah carpets, 95
"   rugs, trade name for Kirmans, 48

Kiz-Killims, see Killims

Knots, kinds of, 25
"    , numbers of, 26

Koniah rugs, 71
" "   , characteristic design of described, 72
" "   , "    " "    illustrated, 26

Koulah rugs, 71
" "   , characteristic border of described, 72
" "   , "    "  "    illustrated, 27

Kurdish rugs distinguished from Kurdistans, 55

Kurdistan carpets, 95, 96
"   rugs, 55, 56

Kurds, “a band of robbers,” 55

Ladik rugs, 72
"    "      , characteristic border of described, 72
"    "      , "    "   "  illustrated, 25

Lamp, Aladdin’s, 71

Lamps like tea-pots in Koulahs, 72

Malgara rugs, 88

“Mats,” Anatolian, 73, 74
"  , Beluchistan, 85

“Mecca” rugs, doubtful term, 3
"   "  , Shiraz, so called, 51

Melez rugs, 74, 75

Mohammedan dating, 125, 126

Moth holes to be looked for in buying rugs, 133

Moths to be guarded against, 140

Mousul rugs, 53
" "   , shimmer of, 84
[Pg 151]
Museums, best rugs in, 6
"   , brown Bokharas only in, 81
"   , guide-posts to, 73

Mysterious inscriptions, 123

Mystery of the rug, 2

Names of rugs, 8, 9
"       "     "  , commercial, 56
"       "     "  , importance of, 14
"       "     "  , unknown and fanciful, 14

Oushak carpets, 98

Pattern, see Design

Persia, inspiration drawn from, 69

“Persian Iran,” ignorant term, 15

Persian, means splendour, 6

Persian rugs, best to buy, 45
"   "   , order of, 15

Pile, depth of, in Mousuls, 84

Pillow, shown by selvedge, 101

Pillows, Sehna, 101

Prayer rugs, 71
" "  , comb in, 135

Rose, conventionalized, 53
" , Oriental origin, 54
" , Persian, 52

Rugs, beating of, 137
" , cheap, uses and value of, 66
" , cleaning of, 137
" , firm, that lie well, desirable, 134
" , folding of, 141
" , holes in, 133
[Pg 152]" , hung on wall, criticised, 102, 134
" , moths in, 140
" , much worn, to be avoided, 133
" , neglect of, 141
" , number annually imported, 114, 115, 116
" , painted or doctored, test for, 134
" , retail trade in, 113
" , tricks in selling, 135
" , washing of, 138, 139, 140
" , wholesale dealers in, 114, 117

Russian, types of, in Caucasian rugs, 63, 65

Saddle-bag, 98
" "  , shown by selvedge, 101

" "  , term confusing, 100

Saddle-cloth, 102

Samarkand rugs, 87

Sarak rugs, 57

Selvedge, cutting of in Beluchistans, 85
"  "   "  "  Bergamas, 76
" of Bergamas, 84
"  "  Beluchistans, 28
"  "  Bokharas, 82
"  "  pillows, 101
"  "  saddle-bags, 101
"  "  Yamoud Bokharas, 82
" should be trimmed, 135, 136

Sehna rugs, 48, 49
" pillows, 101

Serabend rugs, 50
"   "  , characteristic border of, 29
"   "  , "    "      illustrated, 50

Shiraz donkey-bags, 98
"    rugs, 51
"   "  , defects of, 76
[Pg 153]
Shirvan donkey-bags, 98
"   rugs, 64
"  "  , distinguished from Kabistan, 64

Silk rugs, antique, 132
" "  , modern, to be avoided, 48, 132, 133

Sizes of carpets, 44, 94
" "  rugs, 44
" "    "   , almost square, 82

Smyrna carpets, 97, 98

“Smyrna” carpets, so called, 98

Soumac rugs, 86

“Strips,” or “Stair-rugs,” proper name of, 44

Sultanabad carpets, 97

Tabriz carpets, 95
"  rugs, 57

Teheran rugs, 56

Tjoshghan, see Joshghan

Tree, cypress, 75
" of Life, 46

Trellis, rose, 53

Turkestan rugs, 79
"   "   , varieties of, 16
"   "   ,  " " , order of, 16
" weaves, like rubies, 70

Turkoman rugs, 79

Tzi-tzi, see Chichi

Washing of rugs, essential, 138
"   " "   , methods of, 139, 140

Weaving, done by women, 22
"  , method of, 24, 25

Wholesale dealers, buyers from, 108

Wool, camels’, 131
"  from young sheep, desirable, 132
[Pg 154]
Wool has life, is sentient thing, 138
"   , like plush in Kazaks, 84
"   , soft and silky, in Beluchistans, 84

Youraghan rugs, 51

Yourdez, see Gheordez

Youruck rugs, 65




[1] This ancient four-flowered pattern appears in as many forms as the loop or palm-leaf; but whatever bud or blossom may be modelled by the weaver, the design retains its strong distinctive lines. It is shown on the cover of this volume in one phase, and it appears in different form in the plate of the Beluchistan rug.










Author of “A Summer Snowflake”



New York:



Transcriber’s Notes:

Images have been moved from the middle of a paragraph to a nearby paragraph break.

The text in the list of illustrations is presented as in the original text, but the links navigate to the page number closest to the illustration’s loaction in this document.