Investigation into a mystery flag in the collection at the Alaska State Museum. Research is ongoing as new information appears, and perhaps putting this on the internet will solicit additional expertise. At the time, the lab did not have a microscope powerful enough for fiber ID.
Alaska State Museum Conservation Report
Ellen Carrlee January 2008
Catalog No: III-O-495 Object: 36-star Flag
Materials: Textile (wool, cotton) Dimensions: 57” x 107”
Background Info: Although III-O-495, a 36-star American flag, is often referred to as the “transfer flag” or the “purchase flag” the attribution is uncertain. Documentation in the museum file suggests this flag was flown over Castle Hill in Sitka, Alaska on October 18, 1867. It was transferred from the Department of State in Washington DC to the Alaska Territorial Museum (now the Alaska State Museum) in 1933. However, subsequent research suggests this is might not be the flag from the 1867 transfer event, and historian Robert DeArmond wrote an article explaining this viewpoint. He includes the number of stars, the “unofficial” size, and the crude stitching as reasons why the flag III-O-495 is not the transfer flag. His research also indicated that people present at the ceremony later stated the flag used had the proper number of stars. On October 18, 1867, 37 stars would have been correct. The 36-star flag was the official US flag starting July 4, 1865 for the admission of Nevada, and was the official flag for two years until a star was added on July 4, 1867 for the state of Nebraska. The men sent to oversee the transfer had to sail from New York to Sitka through Panama (railroad, the canal had not yet been built) and left New York on August 31, 1867. Perhaps this was not enough time to procure a proper 37-star flag and a 36-star one was sent instead? Alaska was not even a Territory then, much less a State. Several of DeArmond’s criteria are not accurate, however, as explored in this report.
Scidmore (1885) wrote, “Brigadier General Lovell H. Rousseau was furnished with a handsome silk flag and many instructions by Secretary Seward, and left New York the same August in company with Captain Alexis Pestchouroff and Captain Koskul, who acted as commissioners on the part of Russia.” If Scidmore is accurate in her description of a silk flag, III-O-495 cannot be the transfer flag. She also mentions three American ships who “…were flying their colors in the harbor that gay October morning.” The Ossippee under the command of Captain Emmons, the Jamestown under the command of Captain McDougall and the Resaca under Captain Bradford. Could III-O-495 have been one of those flags? Was Scidmore correct in saying “silk?”
There is another flag in the collection of the Alaska State Museums (in the Sheldon Jackson Museum in Sitka, SJ V-C-42) also said to be the “transfer flag.” This was donated to the state of Alaska by Mrs. Caroline Davis Hall (Hale?)of Seattle, Washington in 1896. Apparently, it was known even at that time that it was not the transfer flag, largely because it only had 24 stars, typical of Arkansas admittance to the Union in 1821 and before the 25th star of Missouri in 1836. The color-bearer from the transfer ceremony stated that the flag used was proper in its number of stars (36), and thus an earlier version of the American flag was thought to be unlikely.
Records indicate the transfer flag was sent from Washington DC to the 1904 Louisiana Purchase Exposition (also known as the 1904 World’s Fair) to be displayed at the Alaska Building. It was sent back to Washington DC afterwards. Upon examination of the ASM flag, an oval stamp in red ink was found on the back of the hoist. It reads, “Bureau of Rolls & Library 11 Oct 1904 Department of State.” The stamp is much easier to read with UV light. A photograph from the fair shows a flag flying over the buildings housing the Alaskan collections, but the flag is larger and the arrangement of stars is not the same as III-O-495.
A great deal of useful information was gleaned from the website of Jeff Bridgman, and is referenced in this report. For example, “Prior to the Centennial of our Independence, the Stars & Stripes was simply not used for most of the same purposes we employ it in today. Private individuals did not typically display the flag in their yards and on their porches. Parade flags did not often fly from horse-drawn carriages. Places of business rarely hung flags in their windows. Some of these uses for started to take place during the patriotism that surrounded the Civil War, but civilian use of the flag was not wide spread until 1876.”
Paper removed from the back of the framing dates from May 23, 1967: “This is to acknowledge receipt of the first American flag to fly in Alaska after the take-over of Alaska by the United States, on loan to Mr. Don Dickey, Alaska State Chamber of Commerce; the flag is to be displayed at the Alaska Centennial Exposition site in Fairbanks for the duration of the Exposition, shown in Sitka for the Alaska Day celebration, and returned to this Museum for the opening of the new museum building. The flag is to be insured for $5,000.00”
Description: Large hand-sewn wool bunting flag made of separate pieces of fabric for each stripe (some stripes required more than one piece) and two separate pieces for the blue canton. The flag has 13 stripes in the field (7 red and 6 white) and 36 hand-sewn single- appliquéd cotton stars. The sleeve appears to be cotton duck (a kind of plain-weave canvas with double warps and double wefts) with whip-stitched grommets (no metal.) There is a handwritten ink inscription on the hoist near the upper grommet, but the word is illegible. It appears to be a single word, perhaps 6-9 letters long. Conservator’s best guess : “Crawford” May refer to a maker, an owner, a location, or something else entirely.
A few flag terms:
Hoist = the height of a flag while flying, also called the width. The hoist edge that attaches to the halyard of the flagpole is called the sleeve. Sometimes it is also called the hoist.
Fly = the edge opposite the sleeve
Canton = the upper left area of decoration
Field/Ground = the area between the hoist and fly that does not include the canton.
Vexiollogy = Study of flags
About the stars:
There are 8 stars in the top row and 7 in each of the remaining 4 rows. Because of the single-appliqué technique, it was not possible for any of these stars to have been added to a 35-star flag. The 8 stars in the top row are evenly spaced, indicating that the 8-7-7-7-7 arrangement was by design. Until an Executive Order of June 24, 1912 there was no specific order of stars or proportions of the flag proscribed. In this 1912 order, a single point of each star was to point upward. These rules were not in place for the 36-star flag, but other examples show that six rows of six stars each is much more common. Also common were 8 stars in the first, third and fifth rows and 6 stars in the other two rows. The alternating arrangement of the star points seen on III-O-495 is sometimes called “dancing stars” or “tumbling stars” by collectors.
The flag is almost certainly made from wool bunting. “Bunting” is cotton or wool cloth woven in a plain open weave similar to cheesecloth, dyed in the piece (dyed after being woven, as opposed to dying the threads first.) The word comes from the old English “Bunt” which is a sieve. The cloth was stretched over a sieve or “bunt” in the processing of grain at the mill. “Bunt” is also the German word for “bright colors” which are characteristic of bunting. Wool bunting is woven from “worsted” yarns. Worsted cloth is woven from fine yarn spun from combed wool to remove the short fibers. Only the long fibers are used, making a smooth lightweight fabric.
According to Jeff Bridgman, “The wool bunting used in the making of most all naval flags and garrison flags, as well as some other, smaller flags for outdoor use, was not used in clothing or other household goods. It was a cheap, open weave that was great for flags but did not lend itself well to other purposes. The typical dry goods store would not have carried it. It could probably have been ordered, but maybe not in small quantities (only a couple yards would be necessary to make a common sized, single flag). So it was a specialty item and not readily available.”
The use of wool bunting, the pieces construction, and the hand-sewing (as opposed to machine-sewing) suggests the flag was made in a cottage-industry setting. Bridgman believes homemade flags (as opposed to cottage-industry) were more frequently made from cotton, which was more readily available to individuals.
The 7th red strip from the top is made from more than one piece of bunting. The canton is pieced together from two sections of bunting. And the 5th white strip from the top has a rectangular patch on its seam with the red strip below, not far from the fly edge. It appears to be made of the same fabric as the original flag. All three of these “patches” are likely part of the flag’s manufacture. Another patch in the upper corner of the canton is probably later, but still historic. The thread used is slightly different and it overlaps a star.
Why not cotton?
From historic flag expert Jeff Bridgman:
“…cotton absorbs water, which makes it heavy and causes it to rot. Wool sheds water, which is why it is the choice for almost all nautical flags made during and prior to WWI. Silk was the choice for land-use military flags in most instances, especially in the first three quarters of the 19th century and prior, because while more absorbent than wool, it is extremely light weight , can be readily painted to include text and other elements, and is beautifully decorative. So wool and silk were the primary mediums used by professional and cottage industry flag-makers before 1900, especially in the stripes and cantons of flags with pieced and sewn construction.”
Why not silk?
Some references apparently have suggested that the official transfer flag sent by Seward was made of silk. If so, not only would the fabric have a different appearance, but the damage expected would be markedly different. Silk made in the late 19th and early 20th century typically shows damage in the form of long cracks known as “shattered silk.” Silk at that time was bought by weight, not length, and producers took advantage of this by using metal salt mordants to set the dye to the fiber. Silk lost up to 25% of its weight during the degumming process used to remove the waxy substance sericin to enhance dyability. Tin chloride was often used for light colors, and iron oxides were used for dark fabrics. Weighting was done up to 300%. Industrial weighting began around 1860 and lasted until 1938 when the Federal Trade Commission disallowed weighting over 10% (15% for blacks.) If slits occur preferentially in one direction, that might mean the warp was weighted and weft was not. Weighting made the fiber more photosensitive. Acid hydrolysis is the likely damage mechanism. We do not see this kind of damage on III-O-495.
The stars are made of cotton, hand-sewn, and single-appliquéd. Single- appliquéd means the star was applied to one side of the canton, then the blue fabric was cut from behind each star, folded over, and under-hemmed so that one appliquéd star could be viewed on both sides of the flag. The stars would appear smaller on one side. This type of stitching often results in stars that are more irregular in their appearance. Collectors consider this an enhancement to the folk qualities of a flag.
A curiosity from flag expert Jeff Bridgman:
“The 36th state, Nevada, entered the Union during the Civil War on October 31st, 1864. The last Confederate general surrendered on May 26th, 1865. The 36 star flag became official on July 4th of that year, but makers of printed flags would have begun adding a 36th star to their flags in 1864, even before the addition of the new state occurred. Lincoln pushed Nevada through just 8 days before the November election to increase his support. While the population of the state was quite small, its primary industry was silver mining and the political alignment was generally Republican. The 36 star flag was replaced by the 37 star flag in 1867, with the addition of Nebraska. 37 star flags are very scarce, due not only to their age, but also to the lack of major patriotic events during the period they were used, which followed the Civil War yet preceded the 1876 centennial (our nation’s 100-year anniversary of independence). While the 37 star flag was still official in 1876, it was well known that at least one more state would be joining the Union that year. This caused flag makers to cease production in favor of 38 and 39 star flags, which were made as early as 1875 or even prior. The 37th state, Nebraska, joined the Union on March 1st, 1867. The 37 star flag was official from that year until 1877, although it fell out of use in 1876 with the addition of Colorado.”
The transfer flag ought to have been a 37-star flag.
III-O-495 is a little under 9 x 5 feet in size. At the time, there was no proscribed size for an “official’ flag” just as there were no rules governing the arrangement of stars. The Smithsonian’s original Star Spangled Banner from 1814 (the one that inspired our anthem) measures 30 x 34 feet.
Flag expert Jeff Bridgman states,
“Today, in the 21st century, a flag with 5 foot (60inches) length is common and some would even consider that to be large. In the 19th century, however, that was not the case …Printed flags (called parade flags or hand-wavers) were generally 3 feet long or smaller, but flags with sewn construction were generally between 8 and 10 feet long and larger. Even infantry battle flags in the Civil War and Indian Wars period were 6 x 6.5 feet. This is because flags needed to be seen from a distance to be effective in their purpose as signals.”
The stitching on the flag appears to be entirely hand-sewn, which would be rather unusual since most flags of the era included at least some treadle-sewn stitching. The stitches on this flag are very irregular in size and spacing and there are knots in the ends of the threads. The hand-sewing, use of wool bunting, and the pieced construction suggest a cottage-industry flag maker. By the time of the Civil War (1865) Bridgman estimates 60-80% of flags had some machine stitching. By the Centennial (1876) a treadle machine was used on almost all the stripes of American flags with stars still hand-sewn, but by the Spanish American War (1898) most flags no longer had any hand sewing at all.
Analysis identified the red as natural madder. No simple test is available for identification of the blue, but indigo was the standard dye used for this dark color of blue. Woad and Prussian Blue were other blue dyestuffs, but produced a lighter, grayer blue color more like cornflower.
Condition: A letter in the file from 1944, states, “The flag, 5’ x 8’ is of wool with the exception of the stars which are of cotton. There are thirty-six blunt-pointed stars (5 points), eight in the upper row and seven in each of the other four rows. It was in bad condition when received, due no doubt to mildew, so Father Kashavarof, the Curator, had a native woman back it with white cloth.” Upon examination, this appears accurate although mildew as the cause of damage is unlikely due to lack of staining.
A photograph in the museum file from 1967 shows the flag III-O-495 in its current framing with three panels of Plexiglas across the front.
The flag is not particularly faded. Madder and indigo are among the more lightfast natural dyes of the period. However, bunting is dyed by the piece instead of dyed in the individual wool yarn (dyed in the wool.) The latter is more stable. The bottom red stripe appears darker, but this is more likely from staining than fading because of observed irregular darkened areas. At a distance of approximately 3 feet, the discerning eye can barely made out two ¾” yellow vertical lines that correspond to the mullions on the exhibit case separating the three panels of Plexiglas.
There are numerous irregular holes and losses to the flag. These damages have irregular ragged edges. Although it has been suggested the damage is from use, this is unlikely. Most flags become tattered on the fly end from wind shear, but most of the damage to this flag is not in that location. Furthermore, there is serious damage to the flag at the hoist edge next to the sleeve, but no corresponding damage to the sleeve, suggesting insect damage instead. Moths, for example, would have eaten the wool preferentially to the cotton duck. A letter in the museum files suggests mildew damage, but the dark staining expected with mildew “rot” is not present. Furthermore, the locations of the most heavily damaged areas seem to suggest the damage occurred while the flag was folded. The flag has very minimal staining of any kind, only a few areas. The flag is unevenly soiled, and the white stripes have a grayish appearance. There is a greenish L-shaped area between the top two rows of stars that seems as though a chemical came in contact with the fabric and altered the blue color. The tears in the canton are predominantly horizontal, which generally means a stronger weft and weaker warp. The reason for it in this case is unclear. The wool sheds fibers readily.
In 1956, Assistant Curator Rosemary A. Allen responded to a research request by Valerie K. Stubbs of Washington DC and mentioned that the flag had been sewn to a linen backing under the supervision of the Territorial Museum Curator in 1933. This is consistent with the current appearance of the flag. The backing fabric is a creamy color and plain weave typical of linen. It also has “slubs” or small imperfections in the yarn where the diameter is irregular, making a long lump or nubby appearance. Linen is made from flax and is the strongest of the plant fibers, 2-3 times as strong as cotton. The stitching appears to have been done carefully and seems to have held up well. It is stable, has held the damaged areas of the flag adequately, and has only sagged/wrinkled a bit on the lower third of the flag, more of an appearance issue than a stability issue. This might be resolved by removing the flag from the old framing support. There is a bright white two-ply modern cord stitched through the flag (blanket stitch) along the top hem. It punctures both the flag and the backing fabric, and is visible in the 1967 photograph.
There is a series of large nail holes along the flag and backing fabric just inside the sleeve edge but absent from other areas, suggesting that sometime after lining (1933) but before the photo was taken in frame (1967) the flag was displayed vertically with the sleeve at the top.
Photography: Digital images, RAW format overall and jpeg details.
- The flag was vacuumed while still stapled in place, using a HEPA-filter Nilfisk vacuum cleaner with adjustable suction, brush nozzle, and a piece of nylon mesh to avoid sucking bits of the flag into the vacuum nozzle.
- Additional professional photographs taken by Ron Klein in April 2008.
- Flag laid flat in frame and staples were easily removed just with fingertips. Some were corroded. Appear to have been galvanized staples.
- To examine the back of the hoist, the basting was removed stitch by stitch and insect pins put through the holes as stitches removed in order to reuse the original holes. In the center back, the stamp of the Bureau of Rolls and Library was revealed and read by UV light. Inscription written in ink on front top corner of hoist seems to read, “Crawford.” Basting stitches on hoist were replaced with cotton-covered polyester thread.
- Cardboard tube was covered with Mylar and acid-free tissue paper. Flag was rolled onto the tube and covered with acid-free tissue and then unbleached cotton muslin. Tied in three places with twill tape.
Consultation: Considerable information was taken from the website of Jeff Bridgman. http://www.jeffbridgman.com. Bridgman is the world’s largest dealer in early American flags, a collector himself, and stocks at least 1000 variations of the Stars and Stripes at any given time. Since 1995, he has focused his business on 19th century textiles including flags and quilts. He worked closely for several years with Dr. Jeffery Kenneth Kohn, the American flag consultant for Sotheby’s Auction House. During this time, Bridgman and Kohn cataloged the Connelly and Mastai collections for auction, the two largest collections of American flags ever sold.
Analysis: Several pieces of loose blue fiber on the backing fabric were captured on a piece of tape and examined under the stereo-zoom microscope alongside fibers from a modern hat known to be wool. The magnification was not strong enough to determine the fiber for certain, but the fibers looked very similar. Furthermore, the matte appearance and fuzzy fraying at the ends of broken fibers is consistent with the deterioration expected for wool.
Under ultraviolet examination, the red stripes of the flag fluoresced bright orange, suggesting the dye might be madder. Alizarin, the synthetic version on madder, does not fluoresce orange. Madder was a very common dye in the 19th century, derived from the root of the madder plant. Madder is one of the most stable of the natural organic coloring matters. (Gettens and Stout, 1966)
A loose piece of red yarn was tested with 10% sodium hydroxide and readily turned purple, a positive indication for madder as the dyestuff. (Gettens and Stout 1966)
Ballard, Mary, Robert Koestler, Norman Indicator. “Recent Results Concerning the Degradation of Historic Silk Flags.” ICOM 9th Triennial Meeting, Dresden 1990.
Gettens, Rutherford J. and George L. Stout. Painting Materials: A Short Encyclopaedia. 2nd edition. Dover Publications, New York. 1966.
Horswill, Merrill, Raymond A. Young, Beverly Gordon, and Majid Sarmadi “Characterization and Preservation of Weighted Silk” Textile Conservation Newsletter. Spring, 1992.
Miller, Janet E. and Barbara M. Reagan. “Degradation in Weighted and Unweighted Historic Silks.” Journal of the American Institute for Conservation. Fall 1989.
Scidmore, Eliza Ruhamah. Alaska Its Southern Coast and the Sitkan Archipelago. D. Lothrop and Company, Boston 1885. p. 206
Conservator: Ellen Carrlee Date: January 2008