JAIC 1989, Volume 28, Number 2, Article 4 (pp. 97 to 115)
JAIC online
Journal of the American Institute for Conservation
JAIC 1989, Volume 28, Number 2, Article 4 (pp. 97 to 115)


Janet E. Miller, & Barbara M. Reagan


FIFTY-FOUR HISTORIC SILK FABRICS (10 to 400 years old) were evaluated and compared to new silk to determine the extent of degradation in relationship to the type of silk (cultivated [Bombxy mori] or wild [Tussah]), presence of weighting agents, and age. The silk samples, obtained by donations from universities, museums, and private collectors, varied in fabric count, denier, amount of twist, metallic elements present, and general condition (poor to excellent).

Of the 54 undyed, historic silk fabrics, 15 contained 1% or more total weight of metallic elements and were classified as weighted. All of the silk samples contained small amounts of iron. Other elements detected by NAA in various amounts were tin, aluminum, chromium, copper, barium, and arsenic. Combinations of metallic elements ranged from 0.02% to 14% of fabric weight. Combinations of metallic elements present differed within and between age groups. Additional research on weighting agents is recommended using a combination of methods of elemental analysis to determine if the amount and type of weighting agents might be a potential method for dating silk fabrics.

The highest element percentages were obtained for tin, which was commonly used as a weighting agent in the late nineteenth and early twentieth centuries. Six of the fabrics from 1890 to 1909 and five of the eight fabrics from 1910 to 1939 were weighted with tin in amounts of 3% to 14%. Tin weighting at levels above 3% appeared to be particularly detrimental to silk, resulting in low yarn tenacities, ranging from 0.0 to 1.06 gf/den., and somewhat lower elongation. In general, tenacity, elongation, and viscosity were inversely related to the age of the silk, based on regression and orthogonal trend analyses. The tenacity means for the silks in age groups representing ≥ 50 years of age (i.e., 4–9) were significantly lower than that for the new silk habutai (intrinsic viscosity of 3.1). Similarly, silks older than 80 years showed a marked decrease in viscosity. The majority of the historic silk fabrics had viscosities of <1.0.

The amino acid content of the new and historic silk fabrics were similar; however, some of the samples exhibited a decrease in bulkier side-chain amino acids (proline, valine, methionine, lysine) with a corresponding increase in the simpler amino acids (glycine, alanine, and serine). As in the other methods of analysis, considerable variability existed in the amino acid content of particular samples within age groups because of inherent differences and previous histories. Overall, the samples containing wild silk had lesser amounts of glycine, alanine, and serine and greater amounts of the bulky side-chain amino acids.

The variety of fiber fracture patterns in the historic silk also reflected the diverse histories and causes of deterioration in the naturally aged samples. However, few differences were observed in the infrared spectra for the new and the historic silks, except for the weighted silk. Some of the characteristic absorption bands for tinweighted silk were flattened and less pronounced.

Visual assessment of the general condition of the fabric did not always correlate with tenacity, elongation, or viscosity. Most of the fabric samples that were classified as in poor condition had low strength, but in some cases, fabrics that appeared to be in good condition were so weak that the yarns could not be raveled out for testing. Not all weighted silks were crisp or stiff, nor were cracks or splits in fabrics always associated with weighted silk. Hence, visual examination may be inadequate to judge the condition of historic silk fabrics.

Because of the weighting practices and severe manufacturing treatments reported in textile trade journals from 1890 to 1939, silk fabrics from this period may be more fragile and readily damaged than much older silks. Neutron activation analysis and other analytical methods may provide important information pertaining to the amount and type of weighting agents present in historic silks that in turn, may have a significant influence on the rate and extent of fiber degradation. Additional research is needed on the synergistic effects of weighting agents and other aging factors on silk degradation, the effects of wet and dry cleaning on weighted silks, and possible methods of retarding degradation.


THE AUTHORS EXTEND THEIR APPRECIATION to the following faculty, staff, and graduate research assistants at Kansas State University: Jack Higginbothem, Department of Nuclear Engineering, for his assistance with the neutron activation analysis; Charles yang and John Graham, Department of Chemistry, for their assistance with photoacoustic infrared spectroscopy; Scott Nelson, Department of Grain Science, for the amino acid analysis; John Kirchma, Department of Entomology, for his expertise in scanning electron microscopy; and Dr. Kenneth Kemp, Department of Statistics, for his assistance in the statistical analysis.

Copyright 1989 American Institute for Conservation of Historic and Artistic Works