Dye Analysis in Miscellaneous
Rugs of the Central Asian Group

[Ed. Note: Illustrations 2 and 4 appear with this article. Illustrations 1, 3, 5, 6, 8 and 9 appear in the companion article, Rugs of East Turkestan: Khotan, Yarkand or Kashgar?, by George O'Bannon. Illustration 7 is on the front cover of Volume 11/6.]

I have generally found that much of the older production of rugs and trappings in the Turkic tribal mainstream of this area involved a subtle and sophisticated use of dyes and color. This was a particularly notable feature of the range of reds and red-related colors of the Ersari, Salor, Saryk, Tekke and Yomud groups. By contrast, certain other Turkoman rugs have always indicated to me a rather wide use of certain synthetic colors, particularly in the reds. This characteristic is seen in examples which otherwise suggest a considerable age. Examples include Kirghiz rugs as well as pieces generally attributed to the eastern reaches of Turkestan.

2. Kirghiz Lattice Rug This is the same lattice pattern of the two East Turkestan Illustrations. The drawing is more angular and the ambiguity of the field and ground is less distinct in this nomadic rug. Here the patterns seem to be more definitely eight-pointed stars and crosses on a red field. Although clearly Kirghiz the border patterns can also be found in East Turkestan rugs. See Illustration 7. The weaver almost achieved perfect success at corner reconciliation of the border patterns. Size: 5' x 10' Beg. End: l 1/2" kilim, 3 1/2" knotted fringe Fin. End: Same Colors: RED, dark blue, light red, blue-green Knot: Persian, left; 6h. x 7v. =42 p.s.i. Warp: Z2,S, wool, beige Weft: Z2,S, wool, natural tan and light red, one shoot Edge: Selvedge, brown and blue wool

In this article, I report on dye analysis in certain rugs of the Turkoman group, representing both the western and eastern reaches of this vast rug producing region, with examples of the Kirghiz and such rugs of Eastern Turkestan as the Khotan sub-group. The examples chosen involved a collaborative effort with George O'Bannon (see his accompanying discussion), who provided the samples for laboratory analysis.

Dye analysis consisted of both dye-specific chemical behavior testing and chromatographic separation analysis using the technique of paper chromatography described in more detai1 in Note 5. Identification as to group or specific dye also included use of pure samples of authentic synthetic and natural dyes as comparison standards.

Of the various synthetic red dyes, all can be classified as acid azo dyes. These azo dyes vary considerably as to their hue, fiber loading and specific chemical structure and differ as to their time of likely introduction into rug-making areas. The principal red dye in both of the Kirghiz long rugs (illustrations 2 & 3) was the early acid azo synthetic, Ponceau 2R. I and others have found this dye to be probably the most cornmon colorant in the entire spectrum of Turkoman work and its use can be reliably traced via various accounts back to the 1880s and later.

The main red in the cloud lattice rug and the donkey bags has not been specifically identified, but it is an azo dye of a later type, behaving as an "indicator" type dye, i.e., its color is a function of the relative acidity or alkalinity of the medium in which it has been placed. The dye in question is bright red in alkaline solution and reverts to a dark blue-grey in acid. This dye also was identified as the red component in the amber samples of the single medallion rug (Illustration 4). This color dependence on the medium raises the obvious question as to how such a dye would have been applied to the yarn fiber. For example, was the dye applied from acid solution as is commonly done with acid dyes to give the dark yarn followed by careful treatment with weak alkali? It should be noted that ammonia solution only removes part of this dye while the dye remains fast to strong acid treatment. I have already encountered this type of dye, in eastern Turkestan rugs that appear stylistically to be from the first quarter of this century. This would jibe with the ready availability of this group of dyes by this time.

The caramel field in Illustration 1, the red in the medallion rug (Illustration 4) nd the rust areas of illustration 5 all seem to be derived from an acid azo red dye which has not been identified nor encountered earlier. The range in colors seems to mainly reflect the amount of fading or washout, being pronounced in the prayer rug.

7. Antique Lattice Rug Rugs of this type are usually attributed to circa 1800. Schürmann's Plate 89 has the same palette but different design and is dated 18th century. Certainly silk pieces with similar dyes are so dated. The softness of the reds and yellows are undoubtedly due to its great age. Size: 3'8"x6'4" Beg. End: None Fin. End: None Colors: LIGHT BLUE, red, brown, blue, pale yellow, white Knot: Persian, left; 8h. x 7v. =56 p.s.i. Warp: Z4,S, cotton, white Weft: Z1, wool, white and brown, three shoots Edge: Overcast red wool

The pale red in the antique cloud lattice rug (illustration 7), a natural vegetal dye, appeared to be more of a madder of the type found in India and other parts of eastern Asia than that typically encountered, i.e., Rubia cordifolia rather than Rubia tinctorum. This dye was found by me in certain antique rugs of Tibet.

Blue and Green Colors
The sole source of various shades of blue and the blue component of green areas in these examples is indigo. Light blue was derived from a combination of light dyeing with indigo and use of ivory yarn. In a number of cases, dark blue yarn was obtained from heavy dyeing on brown or grey yarn. Four green areas, of differing color depth, were tested in these example. In addition to indigo as the blue component, different sources of the yellow component were found. These are described below.

Brown, yellow/Gold and the Yellow Component of Green
In contrast to the dyes used for reds, all but two of the brown, gold and related colors were natural, either natural brown wool (for the browns) or applied natural dyes.

The brown areas in these pieces were from natural brown wool in three pieces and from a myra-balans tannin in the Illustration 7 brown, with no evidence for use of tannins mordanted with iron. The yellow areas in the studied examples are derived from either flavonols (illustration 4 & 5) or flavonols with tannin (illustration 2).

A flavonol/tannin source of the gold brown in the green of Illustration 2 was identified and this may have been derived from pomegranate rind admixed with other local yellow sources. Flavonol yellow was seen in the green of the medallion (Illustration 4) and the cloud lattice (Illustration 5) rugs. The pale green of Illustration 5 contained a tannin as the source of the yellow-brown.

Larkspur may be the source of the yellow in those green areas where flavonols alone were identified. This plant is the most common source of moderately clear gold or yellow-gold. No evidence for weld was obtained, consistent with its general absence in Turkoman work and the absence of any bright, vivid weld yellows in the present examples. Tannin could have been obtained from pomegranate rind or woody portions of local botanica.

4. Single Medallion Rug But for one feature, this rug in black and white could be mistaken for a much older rug., It avoids the rigid commercial look of many of these pieces from the turn of the century in spacing and different sizes of the various motifs. But the small patterns in what should be the plain outer border give its youth away. Size: 3'3" x 5'8" Beg. End: l" kilim, 2" uncut, fringe Fin. End: three shoots weft, 4 1/2" fringe Colors: BLUE, amber, light-green Yellow, maroon (faded) light blue, ivory Knot: Persian, left; 9 h. x 8 v. =72 p.s.i. Warp: Z7, S, cotton, white, severely depressed Weft: Z3, S, cotton, blue, two shoots Edge: Selvedge, blue wool

The medallion rug (Illustration 4) contained yellow from a flavonol as did the gold in Illustration 6. The flat gold in the antique cloud lattice (Illustration 7) contained mainly flavonol with a little tannin.

Two different early synthetic orange dyes were identified. The yellow of Illustration 3 was a faded synthetic orange, identified as Orange II, while the yellow of the saddle bag floral device (Illustration 6) was another early synthetic orange, Crocein Orange. Orange II is perhaps the most common early synthetic orange in diverse groups of rugs. Crocein orange was noted in many Caucasian rugs by Whiting.

2. Mushak, P. and O'Bannon, G.W., Chemical Analysis of aTurkoman Chuval. Oriental Rug Review 2, No. 10, 6-8, 1982.

3. Mushak, P., "A Rare Yomud Jewelry Asmalyk: Stylistic and Technical Analysis." Oriental Rug Review 8 , No. 2 , 30-33,1988. 4. Mushak, P., Selected dye analysis in two old Saryq main carpets. Oriental Rug Review 8: No. 2, 13-15, 1988.

5. Mushak, P., A rare Turkoman trapping in the S-group classification: Stylistic and technical analysis. Oriental Rug Review 8, No. 1, 18-22, 1987.

6. Mushak, P., Actual early synthetic dye use in the major rug classifications including new analysis of Caucasian and Turkish examples. Oriental Rug Review 5, No. 2, 7-10, 1985.

7. Mushak P., Dyes and related materials in Turkoman rugs and trappings. In: Vanishing Jewels: Central Asian Tribal Weavings (H.J. Swinney, Ed.) Rochester Museum and Science Center, Rochester, NY, 1990, pp. 59-65.