My Adventures in Overshot

I went to lunch today with two of my fiber friends who commented that I have not updated my blog in a while. Even though I've been very busy they were right. So here you go Pru and Nadine, this post is for you.

I very much enjoyed a guild road trip to the Lost River Museum in West Virginia and learning about two Mennonite women (Ora and Lynn Tusing) who lived up on Branch Mountain near Mathias in Hardy County, WV. overlooking the Lost River Valley. You can see a photo album from the visit here.

I fell for the coverlet sample Priscilla Blosser-Rainey wove on the sister’s loom using the remaining warp that was on the loom when the museum acquired it from the Tusing sister’s estate. The warp is 40/2 linen and the weft is Harrisville 2-ply doubled. The draft is the sister’s favorite which they called Glaring Flower (a.k.a. Pine Burr/Pine Cone Bloom/Pine Bloom).

When I saw the pattern of the sister’s favorite draft embroidered on a piece of muslin hanging over the loom I knew I had to weave it. Did I mention I never wove Overshot before?

I set out to learn all I could about the Overshot weaving structure. I exchanged a few emails with Priscilla who said the coverlet is treadled “tromp as writ” and tabby is 1&2 and 3&4. I also had discussions with Beth, our guild president, who showed me, in the draft and the pictures I took, that Pine Bloom contains tables and crosses. Reading the draft (click any picture to enlarge) if you start where it says “Begin Here” you would start out treadling 1&4 - 8 times, followed by 2&3 - 4 times, next 1&4 - 8 times, and so on. The four Overshot blocks are A=1&3, B=2&3, C=2&4, and D=1&4 with tabby 1&2 and 3&4 as mentioned. Since I’m a visual person I wanted to graph out one full repeat. Here is what I came up with (click to enlarge).

I soon discovered the Tusing sister’s draft did not conform to traditional contemporary Overshot where the 4 blocks are 1&2, 2&3, 3&4, and 4&1 with tabby being 1&3 and 2&4. I also discovered all the blocks did not share a common thread between sequential blocks like traditional Overshot. Looking at the picture you can see the few blocks that do overlap are identified with a heavy X and blocks that don’t, have the adjoining threads boxed in with pencil. Underneath each block I wrote how many threads made up the block repeat and circled the number. I thought I must be doing something wrong! This is really weird! On a trip to The Mannings I brought along my worksheet and the Tusing sister’s draft to see if Tom Knisley could take a look. Tom said the way I was approaching it was correct.

My plan is to weave this on my Loomcraft loom which has a maximum weaving width of 45”. I needed to figure out how many repeats of the tables and crosses I could have. I settled on Partial Table, Cross, Table, Cross, Table, Cross, Table, Cross, Partial Table (reversed). All the numbers on the lower left of the worksheet contain my calculations of how many heddles I would need on each of the harnesses. The total came out to 1,067 ends. Sett @ 24 epi (12 dent reed – 2 ends per dent). I am using 10/2 cotton for warp, 20/2 cotton for tabby weft, and 5/2 cotton for pattern weft. Beth came over to help me wind the warp since she has a warping mill. We decided on 6 yards which would give me 3 separate squares of the pattern. The warping mill made a quick job of it and now I am a convert.

I was still feeling uneasy about this draft not conforming to traditional Overshot. Then one night I was looking at The Coverlet Book: Volume 1 – Overshot by Helene Bress. Bingo! I learned that this draft contained a combination of “on-adjacent blocks” (each block has one thread (or harness) in common with its neighboring or adjacent block) and “on-opposites” (opposite blocks are not neighbors and do not have any harnesses in common).

Helene went on to explain imperfections or doublers. If you remove the pattern weft you would find a Plain Weave ground fabric. When two adjacent warp threads ride side-by-side (threaded into the same harness) they interrupt the Plain Weave structure. The two adjacent warp ends rise or fall together. This is called “unsplit doublers”. They will never be separated or split up.

Can doublers be seen with the naked eye? It depends. Often you can’t or don’t but sometimes they show up as a warp-wise (vertical) streak. Helene goes on to say significant doublers are those that occur at the edge of a block, as one block progresses to the next. Doublers are most noticeable when they occur within a block as opposed to at the edge of a block.

One example of an unsplit doubler may be due to a threading error within a block where adjacent threads are threaded through heddles on the same harness. For example, the block should be threaded 4,3,4,3,4,3,4,3 but instead it is threaded 3,4,3,4,4,3,4,3. These threads will rise or fall together when tabby is treadled and tabby will be imperfect. Chances are good you will see a warp-wise streak. If this streak occurred regularly, i.e., within block D of every repeat then most likely there is something going on in the draft as opposed to a threading error.

The Tusing sister’s draft contains another type of doubler called a “split doubler”. Unlike the previous example where adjacent threads are going through the same harness you have adjacent threads going through different harnesses but the threads still rise and fall together. My tabby is 1&2 and 3&4. If you look at my worksheet starting at the top right and move left you start off treadling 2&3 - 4 times followed by 2&4 – 4 times. No problem transitioning blocks here. The blocks share a common thread. Next, we treadle 1&4 – 4 times. Now we have a problem. As discussed, this is an example of “on-opposites”. The adjacent threads are 2&1. When the pattern blocks are treadled the threads will be “split up” because they do not ride together but the doublers do occur in the tabby area since my tabby is 1&2 and 3&4. Since the doublers here occur in the tabby area only, and occur where there is a change of blocks they are less noticeable than unsplit doublers.

Helene says it’s not hard to find drafts with doublers in them and when there are consistent doublers, there is probably a specific reason for their existence. Half the coverlets Helene examined had doublers. She said some even had a tripler or two.

Here is my project on the loom. Whew! No threading errors. This has been a good learning experience.