TO discuss the subject of this chapter in any but a most cursory way would require a volume of many pages. We hope, however, to implant in the mind of each one of our readers - who has not already felt it - the desire to know more about the beautiful works of the weavers' art as shown in the older Persian, Caucassian, and Turkish rugs and carpets which are constantly being imported to this country.

To one of artistic feeling there is a natural appreciation of the superiority of the old vegetable dyes, and however clever the modern artisan may be he has not yet been able by the use of his analine and acid colors to successfully reproduce the dull, soft tones possessed by the rugs made from the wools treated with these vegetable dyes.

The books treating of the history and technique of these are few. Probably the best is that by John Kimberly Mumford, published in 1900. To this volume we owe many of the facts herein stated, and from which we will quote occasionally.

To gain a knowledge of rugs, one must see them. It is through study of the actual fabric that this knowledge can best be secured. We, therefore, advise frequenting trustworthy establishments where such goods are carried. In examining them, one becomes familiar with similar characteristics existing in different rugs, which are not necessarily of the same name.

Plate XLVII. Caucasian - Kazak.

In the larger cities one may visit the auction rooms where, throughout the fall and winter, sales are held, and where frequent bargains may be secured, if one is able to judge quality under the conditions usually existing at such places.

For the amateur it is safest to rely upon the judgment of a dealer who has established a reputation for artistic feeling and honesty. However, for educational purposes, to spend an hour or two before a rug sale begins, examining the stock as displayed, with catalogue in hand, familiarizes the eye with certain features of peculiar types of rugs, and the effort to identify these characteristics in the rugs listed in the catalogue is a great help to the student. Of course, it is not always possible to succeed in classifying, for, as Mr. Mumford says: "The latitude for error is boundless, even to the best judges, since manufacture for market has become the rule instead of the exception. Hence, no writer, no authority so called, no dealer in rugs, may lay claim to infallibility. Patterns, figures, designs are largely discarded as a means of identification after the eye is able to distinguish some of the most usual types. Secure such books on the subject as are authentic and endeavor to get close to the heart of the weaver by knowing what many of his patterns and designs mean in the rug language, for however distorted the art may have become through falling under the thrawl of commercialism, the older rugs tell on their faces romances and histories as truly as if the story was woven in letters of our alphabet."

A judge of Oriental rugs will probably look at the back of the rug first, and then examine the details of the texture, the kind of knot used, the material of the warp, weft, and pile, and the length of the pile. On the back he will count the number of knots to the inch for the warp measuring horizontally, for the weft measuring perpendicularly. The more there may be, the longer it has taken to weave the piece, and an additional value is given it. By folding the face of the rug sharply to the back, between the thumb and first finger, the pile is opened up and the kind of knot is exposed. The mental notes the connoisseur so takes probably place the birthplace of the rug accurately in his mind before he examines its face for the color value. If the pattern and design confirm his conclusion, the rug is listed and passed. If, however, peculiarities occur which are not usually found in this type he begins at the beginning again, remarshalling his facts and readjusting circumstantial evidence until he makes them agree. If this cannot be done, it is listed without type or name, and designated as a "freak."

Plate XLVIII. Caucasian - Soumack.

Mr. Mumford classifies the Oriental products in four principal types, Caucassian, Turkish, Turkoman, and Persian.

In his book it will be seen that the four principal types are divided and subdivided until the number is many times multipled as to names, at least, these being derived from the name of the city or district where they are made, or from the name of the tribe making them. Hence, the search is pushed into the more remote districts, owing to the demand for these treasures. The quality gives a rug the name derived as above. It comes to market under that name, yet in reality is embraced under one of the four principal types in the foregoing classification. The result, however, is confusing to the layman, as the multiplicity of new names met with in catalogues, and the stocks of dealers, serve to shake his confidence in his newly acquired information, Therefore, it should be remembered that under whatever name a rug may be listed it is pretty safe to assume that it can be sifted down to one of the four principal types, and to one of the subdivisions of it, with considerable accuracy.

Plate XLIX. Caucasian - Karabagh.

Plate L. Caucasian Shirvan Prayer Rug.

Plate LI. Turkish - Ladik.

Plate LII. Turkish - Ghiordez Prayer Rug.

Plate LIII. Turkish - Melez.