This is a record of communications to our client over the course
of the conservation and repair of their tapestry, which will be many years.
This post covers 2026 in its entirety.
In this post, when the speaker is “I”, Kate is keeping notes,
as she is the lead conservator, and this is a compendium of her
ongoing commentary (includes videos) of the process.
“We” is both of us working together. In videos Mitchell often comments,
and when he performs treatment then he will be named.
Note: Changes in color in images of the the same area are the result
of choosing the best image, with flash (more golden tones, sometimes
sharper images) or without flash (accurate color).
Make sure to click on images to see them in their larger format
and to read any notes attached to them.
DEFINITIONS (will be added to over time):
PRP = Previous Repair SOW = Scope of Work
SOW = Short for Outside SOW, outside the estimated and discussed scope of work for the project, however, these things had to be completed when discovered.
LF = left-facing RF = right-facing
I made notes regarding PROTOCOLS OR OVERVIEW items, and moved them to this area:
1) In order to match faded colors as closely as possible, some of the woolen yarns were lightened, and so there may be notes to that effect next to some of the yarn colors. Right, Cascade Superwash Col #354, both as-is and lightened.
2) “Over-wove” (I made that word up, along with over-weave, and over-weaving) is weaving over an area with frayed historic yarns, some of which cannot be completely removed. You can see “over-weaving” in the weft stitches which sometimes won’t release, or are twisted because of a fuzzy historic warp or weft yarn below.
3) Protocol is to follow the color of the historic stitches as closely as possible, though sometimes it doesn’t make sense to the image as we can see it, or it is a partial. We are honoring the tapestry artist’s pattern.
It is difficult to match faded yarn colors in any appropriate yarn, because fading makes for colors not available to purchase. Yarn sales (and therefore offerings) are based on the popularity of the colors in society.
In an area needing repair, ideally we have bits of yarns at least, or a completely obvious and logical pattern to follow, or whole areas of broken yarn to replace, and those situations tell us the colors and patterns to use during infill. But sometimes we have no clues as to the color, just a gaping hole! Then an educated guess as to the artist’s intent (to the best of our ability) is necessary.
My process was to look at the many images we shot during assessment of the tapestry in other areas that felt applicable to find clues that might apply to an area in question, while also stepping back to see the overall surrounding imagery. This takes time not calculated, but it is important, and it rendered me feeling more confident in making decisions in those areas.
If the historic artist’s intent was indiscernible, then I looked to my own artist’s eye as to what the story was in that area. (See the first video under “2026 6/24 – 7/10 RF20” as an example, below.) I want perfection in the recreation, and struggled with this in the beginning, but ultimately let go of it, because creating a beautiful scene to the best of our ability is important when in fact we cannot know the artist’s original intent.
4) I wish I could have pinned the tapestry down while working the many repairs, or suspended it, but in this case I had to continually look at the backside, so pinning was impossible, unless I was just infilling unencumbered warp lines. This made repairs go slower.
5) We did a test cleaning. An area with several different yarns, both silk and wool, was saturated with de-ionized water. We were surprised by an out-gassing of what smelled like mothballs, likely inert naphthalene activated by the water. We assume this was applied prior to the tapestry becoming part of our client’s collection. The smell drove us out of the studio; it was overwhelming and instantly gave us headaches. We aired out the studio until it dried, and then the bulk of the out-gassing lessened. Resolutions to this out-gassing of dormant naphthalene will be resolved during the immersion/washing phase of the tapestry once all reparations to the damaged areas are completed.
The next day we used coffee grounds to absorb any lingering strong smell.
Naphthalene is a bi-product of coal tar production in the late 19th and early 20th century. The introduction of naphthalene was a common practice in some European countries during the early 20th century. Proper cleaning protocols will assure the removal of this material.
6) I chose a warp thread that was smaller than the historic warp thread because I was concerned about damaging the intact historical weft threads during installation. Began installing warp thread using beading pliers to gently pull the warp thread through historic weft yarns, while holding both sides of weft area, which eases possible damage to weft. The warp thread was left a bit loose to allow the weft infill to pull it taut without over-tightening.
There are many sizes of historic warp threads, and some are close to the size of the warp thread we used.
7) In images you will see red thread, which I used to temporarily loosely bind an open area to keep it intact; red allowed me to easily see the thread for removal.
8) We are saving samples of all the odd bits of yarns used in PRPs, in case our client wants them in future for provenance; shown right, several types of PRP yarns and threads used by others.
BACKGROUND
The large tapestry was once a large woven entity, probably intended as a wall hanging, made of smaller woven areas pieced together to its original form using a string-like fiber, now brown from the years, as opposed to a woolen yarn.
At some point we surmise it went through a cataclysmic event, and this event ended up tearing the large tapestry, or someone cut it to save it — we possibly will never know.
We identified the pieces that could be disassembled, and did so by removing the stitches that bound them together, and then temporarily stitching them to keep them from unraveling where applicable. We then had the pieces we now will work on.
We began the conservation of the tapestry with the smallest piece, shown right, on a smaller workstation.
RF20
(NOTE: These large headings refers to the approximate quadrant of the repair,
though there is sometimes crossover into an adjacent quadrant…
in this case, RF19, because the camels head began in RF19.)
“Yarns” used in this quadrant, listed below. Includes the warp line which technically is not a yarn. Needles used were #22-#24 John James or Bohin.
Warp: Bockens Lingarn #003098, 12/2, 250 wt, color 0000 unlightened
Warm tones
- Cascade Yarn ARAN Col#511 (darker biscuit)
- Cascade Yarn SuperWash Col#1926 (grey)
- Cascade Yarn Superwash ARAN 100% wool, Color 349 (light biscuit)
- Soie d’Alger #4533 (a biscuit color)
- Soie d’Alger #4532 (a cream color)
- Soie d’Alger #3821 (a pale yellow color)
- Soie d’Alger #3823 (a medium bright gold color)
- Eugene Textile Center’s own brand:
- #084 2ply (brown)
- #855 2ply (brown)
Cool tones
- Cascade Superwash Col #354 (blue, both as is and lightened!)
- Cascade Superwash Col #856 (blue, both as is and lightened!)
2026 6/24 – 7/10 RF20 (previously reported)
Above, the first large repair performed, before and after. This involved warp infill, weft infill, and guessing the design of the camel and background.
Below, videos on process and discoveries.
If I could not decide what the historic artist’s intent was, then I looked to my own artist’s eye as to what the story was in that area. As I move forward, this will take less and less time as I become more and more familiar with the artist.
The oddly charcoal yarn did not seem a dyed yarn, but appears to have been scorched or altered by exposure to heat and/or hydrocarbons.
And below, my initial thought after finishing this first large repair above.
In the video below notice the uneven LF-edge running alongside the camel, which is pulled in approximately 1-inch. I initially planned on reweaving an infill for the entire edge to even it out.
As we progressed and made several large repairs in this area in September and October, the LF-edge relaxed and lined up properly. The pulling in of the edge was the result of poor previous repairs. This also means that there was not a large loss in the camel’s body nor leg, as postulated in this early video.
2026 8/13-8/14 RF20
Below Camel’s neck along LF-edge.
Recreated damaged and missing warp lines, anchoring and stabilizing them to re-weave damaged weft infill.
Tested lightening the first blue yarn, Cascade Superwash Col #354; it turned out nicely for matching of one color in the faded flowers. Below left, the original color and the color lightened, right, the color near the floral motifs. More lightened colors will be created.
2026 9 10 RF20
Historically the warp was both thin and fatter lines, and I could not see the reasoning behind the choice; example shown below. Note the thicker strip of soft brown historic wool in the image below left. In two areas the historic weft becomes fatter for no discernible reason, SOW.
Adjacent to a large completed repair was a medium repair with a color I could not match in any of the yarns I bought (even those I sampled then returned). I tried to mix two colors to see if from a afar they would look like the correct color, using a #24 John James needle in these two colors:
- Cascade Yarn ARAN Col#511 (darker biscuit) and
- Cascade Yarn SuperWash Col#1926 (grey)
To make the two yarns thin enough we had to strip down to one ply, and unfortunately on Col#511 the ply wasn’t strong enough to weave and kept breaking.
2026 9 11 RF20
I finished infill weaving of the two-color area, above left, but will not try it anywhere else because the single ply yarn is not strong enough to use. Had it been easier to find matching yarn colors I might have been able to find a yarn that was smaller in girth which would have worked.
Above right, three other smaller repairs, two of which needed both warp and weft infill, are shown, with some detailed images of those repairs below.
I still cannot cannot get the “wow” in the LF edge of RF20 to line up. I now believe it may be an issue in an area below (or to the right as the piece is on its side) this one. I want it to line up to make bringing the pieces of the tapestry easier, and to save me from having to add warp to create an infill along the LF edge. See video just above “2026 8/13-8/14”
2026 9 12 RF20
Camel’s neck and area just below.
I found three more broken warp lines (small areas) and infilled. Many are hard to even see without taking a small crochet hook to find the warp lines. 1-2 additional warp lines were secured to perform proper weft infill, SOW.
Note: The larger PRP below is not one of ours, but is stable.
Weft infill on the camel neck, plus added small bits of warp line:
- Cascade Yarn Superwash Col #1926 (grey) and
- Soie d Alger #4533 (a biscuit color)
“Over-wove” (I made that word/phrase up, along with over-weave, and over-weaving) is weaving over an area with frayed historic yarns, some of which cannot be completely removed. You can see “over-weaving” in the weft stitches which sometimes won’t release, or are twisted because of a fuzzy historic warp or weft yarn below.
2026 9 15 RF20
Continued with weft infill along camel neck and surrounding area; did not image all parts of this long neck infill.
Occasional infill with Cascade Yarn ARAN Col#511 (darker biscuit) which is keeping smaller weft issues intact surrounding the area.
Again noting that historically two different sized warps were used, shown right. They have not disintegrated, but are simply of varying widths. It is currently unclear why various sizes of warp lines were used, but current assumptions based upon other tapestries warp thicknesses often have to do with the sizes pf the weft yarns or spinning techniques.
When they were hard to see on the front, the back sometimes best showed where the historic placement of the yellow-golden tones of the Soie d’Alger yarns needed to infill, shown right.
Protocol is to follow the color of the historic stitches as closely as possible, though sometimes it doesn’t make sense to the image. We are honoring the tapestry artist’s pattern.
Also, it is difficult to match faded yarn colors in any appropriate yarn, because fading makes for colors not available to purchase. Yarn sales (and therefore offerings) are based on the popularity of the colors in society.
2026 9/17-9/18 RF20
Completed the weft infill along camel neck and surrounding area, examples shown below:
- Cascade Yarn Superwash Col#1926 and
- Soie d Alger#4533 (a biscuit color)
Note: The image below was taken on 10/15 after many other damaged infill areas were completed which will be discussed later, to show the long neck and head completed.
Next I treated a huge PRP on the camel’s body at the bottom of the neck as it moves into his chest, shown right. Note the poorly woven taupe weft yarn which did not match the historic surrounding yarns of both field and motif.
I added several warp lines.
This PRP crossed half over a heavy linen patch that was approx. 7” x 15”. Having the repair over this large patch hindered my ability to perform repairs. I had to remove the back patch, but did so initially in stages, shown below, until finally deciding to remove the entire patch.
Additionally, there were several additional larger PRPs (shown left), and they were hindered by the large patch as well, plus whomever applied the heavy patch performed repairs THROUGH the linen patch. The patch pulled on the fragile disintegrating fibers, and was too heavy for the historic tapestry weave.
Remember the way the LF side of RF20 was uneven, shown left? This was due to the PRPs and the heavy linen pulling the entire LF side up in different places, causing the uneven pull, causing additional damage to the tapestry.
As I could, I cut a bit more of the heavy patch off and then secured the PRPs, sometimes temporarily. Truly some of the worst repairs, created in a heavy taupe or cream colored yarn, where someone stitched areas up tightly, binding them in an unnatural way and causing the tapestry to bunch up. These types of PRPs, attempting to camouflage losses, are the primary cause of the asymmetrical “wow” of the salvaged edge.
Some repairs had folded under and into the repair, and these heavy folds had to be released and the tapestry was carefully handled so as not to cause more damage until the PRPs could be properly repaired.
2026 9 24 RF20
Continued working on disassembling the rats nest of a poor PRP. There are many knots in thin brown string, a taupe yarn, and a cream yarn, none of them historic. I don’t understand the choice. Whomever did these repairs also pierced the historic warp yarns, weakening them, and caused additional warp replacement.
I have to move slowly through these repairs, as much as I’d love to just cut the knots out, because in the mess I often discover:
1) information about historic yarn colors, and
2) intact warp or weft which may be able to be saved.
Heavy linen patch after removal, shown above.
2026 9 25 RF20
Videos are an excellent way to communicate:
Above, a video after the removal of the heavy linen patch on the back;
Mitchell and I discussing what we believe are the various parts of the camel, saddle, etc. Below, more of the PRPs removed, and I talk about how I was moving gently
into one repair and then had to undo so many loose bits, becoming chaotic!
The repair in the pleated or folded area of the flower of the historic tapestry weave was another mess, shown right. The folds were trained over time by the poor repairs scrunching the tapestry.
I needed to remove the folds, and dampening the tapestry was not an option due to the inert, but toxic chemical naphthalene embedded within the historic fiber. I carefully ironed the tapestry.
Note: Naphthalene will be neutralized (removed) during washing processes after repairs are completed.
2026 9 29 RF20
Below, a video discussing the large nest of PRPs, twisting and bunching
and grabbing the tapestry, and why I decided to take the “ratz nest”
of repairs out before beginning to repair warp and weft infill.
Video below about disassembling the large hard thick linen patch,
and it took a very long day as I did not want to stop until I could
make sense of it. The LF-edge is straightening.
I decided removing the entire heavy linen patch was the best in the long run, as working around areas was tedious. I decided the mess would be better than fighting the linen and doing everything piecemeal. After carefully removing the patch, many PRPs which went through the linen patch were damaged. I pinned and stitched a temporary light cotton strip along the edge to secure it, shown in images below.
All the while I am vacuuming constantly to remove debris so I can work in a clean environment. It is surprising the amount of embedded large grit suspended in the historic fibers. All will be removed and/or dissolved prior to the backing installation.
2026 9/30-10/1 RF20
Warp lines were placed into the holes and disintegrated areas. I kept loops to tighten or loosen the warp lines as necessary, and will make those decisions as I fill in the areas with the weft yarns. As these are loosened, the LF edge of RF20 lines up again; showing it may have been the tight PRPs that caused the edge to become wonky.
Video above about the disintegrating weft which appeared stable from afar.
Warp is frayed and new warp, which was chosen as a slender warp line in order to go through historic weft. The historic silk is also frayed, and brittle. SOW.
Video above discusses the different warp threads in the tapestry,
the different ways they are pieced together historically, and the LF-edge,
which was indented 1-inch, straightened out when everything was released.
2026 10 12 RF20
Changed our treatment workstation to better facilitate work. Notes now will be entered directly into computer, with handwritten journal for items that have samples attached.
Finishing warp infill today on second large repair area, above and right, and 95% of these are previous repairs. Graduated to finger cots because I wore a hole in my finger pulling the warp yarns through the weft.
The first golden flower historic weft (above, matched by Soie d’Alger #4533) is shockingly brittle, almost as if a chemical was dripped onto the area. It looks intact but when touched it disintegrates, causing more work than anticipated; all the historic weft infill was removed.
Several small intact PRPs were in the golden flower, but due to the brittle disintegration, they had to be undone and the area reworked, which was essentially the entire flower. Outside SOW but must be treated.
Last week I recognized there had been an extra warp line dropped into position by the historic weaver/artist.
Today I understood at least one historic strategy for this. In the silk areas, (such as the golden flower), they apparently dropped extra warp lines in to account for the thinner silk yarns (my deduction). I found an extra warp line in the brittle golden flower.
Completed warp infill on large area of repair, shown right.
Going back up to the camel’s head at the top of the neck, some weft infill and stabilization, and to stabilize the connecting area of background to underside of jaw, and finally, eye area and nose, two colors weft yarn, outside SOW.
In images, if red threads are seen loosely tied, they are temporary stitches to hold areas from pulling open and causing more damage while I work in another area.
2026 10 13 RF20
Ordered more sample colors of Soie d’Alger.
Removed large area of damaged historic golden silk weft above and below the camel’s rein in preparation for new infill, shown left, discussed below.
Removing it allows me to check warps in these areas, and exposes the damaged silks. They easily lift when damaged. In this case, almost the entire area is damaged, so there will be little over-weaving.
I also continue to infill small areas outside the damaged area, SOW.
Below, a video discussing removing weft infill to see the warp lines,
and showing the historic weft as broken pieces.
I found 7 warp lines that were either disintegrated or broken, shown below, SOW. These were replaced. Much of this had to be performed from the backside, and jeweler’s pliers help pull the needles through in all cases. The last three had to be woven far above the historic golden silk area, as the warp was compromised in several places above the silk area.
I chose to show this complicated warp infill because it shows the treatment that must happen in many of these small areas that were unseen before.
At the end I removed the pieces of broken warp where I could, which helps when weaving the weft infill.
A smaller set of small holes to the right of the larger hole was infilled, both with new warp and weft, shown left.
Weft infill finally began. Keeping in mind, especially as it pertains to the silk yarns, it is essential to untwist the yarn strands every few inches to ensure they remain fat or plump.
Some background and camel infill was also performed. Colors:
- Cascade Yarn SuperWash Col#1926 (grey)
- Soie d’Alger #4532 (a cream color)
- Soie d’Alger #3821 (a pale yellow color)
- Soie d’Alger #3823 (a medium gold color)
2026 10 14 RF20
Lightened several more blues, trying to create several more “worn” or “faded” turquoise-blue colors.
I also lightened more of the blue that already works in the flowers, Cascade Superwash Col#354. The other yarns hung onto their color and did not lighten enough despite a long soak; more yarns on the way to test to try to get the correct infill colors for the many blue flowers in this and other quadrants.
Weft infill on large hole continued, begun 10/13.
2026 10 15 RF20
Weft infill on large hole, continued from 10/13, was completed, sample images above.
After infill of large area was complete, I used normal sewing thread thread to stabilize silk knots on the back to keep them from untying themselves.
(Remember that due to the extent of warp and weft damage the stabilizing backing cannot be secured to the tapestry until after repairs are executed. This is so we can gain access to the weave structures during warp restoration.)
In the next large reparation of the area that was backed by a thick linen patch, I started with the brittle first golden flower at the top of the long cut. The brittle golden silk was removed, and the open blue area at the top tacked with blue sewing thread.
2026 10 20 RF20
After considering the toxicity of the naphthalene, the tapestry was doused with long ago, I decided it was wise to wear a mask while removing fiber, especially the silk fibers which are sometimes dust they are so disintegrated. Eye-drops are also essential!
The first golden flower was a difficult infill. Removing the heavy linen backing possibly made working on the various infills more difficult, however, I saw no other way. Having that heavy linen was going to pull on the damaged areas as I repaired various areas, and made infilling the warp impossible as I could not pin the tapestry open.
2026 10 21 RF20
Completed the first silk flower, see above.
- Soie d’Alger #4533 (a biscuit color)
- Cascade Superwash Col #354 (blue, lightened!) as center
Defined the outstretched camel leg, at the knee, shown left. Also, infilled the line down into the shin, SOW.
I am into unknown territory in terms of what the artist intended, with only the damaged images such as those shown above to guide me. Some of it was repaired with huge strands of yarn overwhelming the historic fibers, and some of the area was simply missing. I will have to decide what I believe are the correct yarns to use to tell the story of what is happening behind the knee, above the knee, and in the lower area of the chest of the camel. Before the end of the day Mitchell came in to see what I finished and we talked about the mystery area, and talking it out allowed me to see the infill I needed to perform.
Began stabilization of an area above the knee, so tightened the warp lines previously loosely installed.
As I did this, I also decided to undo a small stable PRP (SOW) adjacent, shown above, and in the way of the reweaving of new silk and wool weft. After undoing the PRP I found missing and broken warp lines. I added new warp.
(NOTE: Finding missing warps under this stable PRPs, which we were not going to repair, makes me wonder if there are many broken or missing warps in PRPs. I decided to investigate a few of them — and repair those I undid at no charge as it is outside SOW; and discuss it with our client. See 2026 10 23 below.)
Above, images of the warp lines added, and the beginning of the weft being infilled over 10/21-10/22.
2026 10 22 RF20
Infill three-color background and second flower:
- Soie d’Alger #4533 (a biscuit color)
- Cascade Superwash Col #354 (blue, lightened!)
- Soie d’Alger #3821 (a pale yellow color)
- Soie d’Alger #3823 (a medium bright gold color)
The second flower is on a join line, and unfortunately the upper part of the panel is quite damaged. The two warps come together then separate, and again, I don’t know why they did this, but for stability I had to infill the weft over the double warp, and took it a 1/4″ longer than they had taken it.
Infilled below the second flower, Cascade Yarn Superwash Col #349 (lighter biscuit). Right, both first and second flowers completed.
2026 10 23 RF20
.
.
.
.
.
.
.
.
.
RF19
.
.
.
.
RF21
.
.
.
dkatiepowell @ aol.com / mitchellrpowell @ aol.com
(So sorry, but we got too much spam — please copy and remove the spaces!)
503.970.2509 / 541.531.2383
©MPF Conservation. May be printed for your own use.
Notify us if you repost, and use our url + copyright in reference.
