I love this gown. For now it will be the layer I wear over my linen shift, and once I have made another dress that is a bit more fitted, this one will get pocket slits and become an over dress. I feel absolutely beautiful when I am wearing it. Thanks to Elina and all of the Herjolfsnes Challenge supporters for inspiring me to become a weaver and make something this glorious!
The hem of Herjolfsnes 42 was not folded, it was finished with 'singling' and then 'slynging'. Singling is a serpentine stitching pattern sewn into the topmost threads on the reverse side of the hem, so that is is not visible from the right side. The singling stiches travel about 1-2 cm up into the fabric and then curve back to the edge. There are approximately 5 mm between each arch. Slynging is the attachment of a tabby-woven band along a reinforced edge. This can be done via heddles or using tablets if only two holes are threaded and the tablets are turned forward and then backward to open the same two sheds as you move along the edge (Ostergard, 2009, pp. 97-106).
I was very careful with the fitting of my pattern toile and the hem of my gown did not require much trimming before beginning the finishing techniques. I just needed to trim a few weft threads. Because the gore panels of the gown have bottom edges cut on curves, the weft threads do become a bit shaggy, even with careful handling. I could see the reason why reinforcement was required for the original garment. When I was done trimming I had just a small amount of wasted yarn.
In Woven into the Earth (2009, p. 99) Ostergard mentions the nearly invisible nature of singling stitches. They are more visible when light falls across the fabric (raking) than straight towards. This can been seen in the photographs of my singling stitches, below. In the first picture the singling is difficult to see, but in the second, veiwed at a slight angle, the stitches clearly stand out. You can see by the position of the needle in the first picture, that I left about two warp threads between every singling arch. This is a continuation of my 'rule of two' that I used for stitching seams and overcasting. My singling are approximately 17-20 mm tall. These stitches do a remarkable job of adding firmness and strength to this edge. I looped the lower singling arches around the last yellow weft thread to help negate any ravelling.
Slynging is a more complicated technique and one that requires a bit of practice to ensure evenness in tension for the warp threads and the weft. Since it is a simple tabby band, I decided to use my rigid heddle. Ostergard reasons that it was likely done using footweaving and string heddles at Herjolfsnes, and this method is still used in modern Iceland (Ostergard, 2009, 105). The warp needs to be attached in such a way that the band isn't so tight that it pulls in the hem, or so loose that causes the hem to ruffle. I tried various tensions for the weft thread and found that loosely pulling the thread to make the band lie flat on the fabric was not optimal. The band wasn't firmly applied, and the edge threads were not caught tightly in the band and looked ragged. Pulling the weft thread so that it lightly cupped the bottom edge was very tidey and it firmly encased the very bottom (and most fragile) edge.
In the photographs below I show the slynging from the right side and the reverse side. You can see how the little warp and weft threads from the raw hem are neatly caught up in the applied band. The reverse side shows the slynging weft threads as they cross underneath the band. Images of extant slynging on the lower edge of a hood have been provided in Woven into the Earth (2009, p. 218). In that case, the band measures about 7 mm in width (very similar to my band), and it looks rounded, as if it is cupping the the raw edge. The extant slynging looks like the band may be slightly longer than the edging length, causing a slightly ruffled look. Although, with the age and history of the object, that ruffling may have other causes. I'd like to think that it's been applied a bit ruffled, though, because I have more than a few spots on my hem that aren't perfect either!
There's no need to beat the weft during slynging. I used my fingers and thumb to pry the shed open and push each weft thread into place.
In slynging the weft is on a needle, so you are limited in the length of the weft by comfort in drawing the thread through. You probably should not cut a weft thread that is longer than your arm length when doubled or the first few passes will be a bit awkward. I was very carful when drawing my weft through each time to make a big loop (below), then pull it slowly smaller (next photo, below) to catch the little raw threads, then pull it snug (last photo, below).
So, that's the end of my gown - it has been completed!!!
My next post will show the finished product.
Ostergard, Else, Woven into the Earth: Textiles from Norse Greenland, 2nd Ed., Aarhus Universtiy Press, 2009.
Here's what they looked like before they were attached to the armscyes, and the wrist slits:
Based on my experience fitting the the muslin toile for this pattern, I was disappointed when I found that the sleeves could not be sewn in without much puckering of the armscye. This may be due to a number of reasons: cutting the fabric slightly larger than the outline of the muslin pattern, sewing narrower seam allowances in the gown than the toile, or the use of a thicker fabric for the gown. At any rate, I was left with a decision: take in the width of the gown at the side, upright seams to make the armscyes smaller; or add a gusset to the sleeve to make the sleeves wider. I dove back into Woven into the Earth for help, and it wasn't long before I found a lovely Herjolfsnes solution: garment #45 has two-part gussets at the underarm of each sleeve, described by Ostergard as "a 155 mm long, 120 mm wide two-part gusset...The sleeve is placed such that the bottom gusset seam is opposite the side seam of the garment" (Ostergard, 2009).
So, I used some small fragments left over from the woven fabric to cut four pieces that would fit the underarm of the sleeves and be similar to those described by Ostegrgard, above. Here is a photograph of the new, two-part underarm gusset, from the reverse side and the right side. The angled sides of the gusset form continuous seams with those from the side-front and side-back gores.
Here is the what a sleeve, with new gussets, looks like from the front, and back of the gown (reverse side).
I am so happy that I added the gussets! With fabric this thick, I would have been unhappy with tighter armscyes, and I think they would have been uncomfortable. Plus, taking in the sides of the gown would negate my wish to keep this garment with a comfortable amount of positive ease around my torso.
Ostergard, Else, Woven into the Earth: Textiles from Norse Greenland, 2nd Ed., Aarhus Universtiy Press, 2009, p. 180.SaveSaveSaveSave
But, I'm getting ahead of myself and think I should post some of the photos of the project on its way to becoming a gown.
At the end of my last post I was finishing up the fabric weaving and beginning the toile of the pattern. I scaled up the diagram from Medieval Garments Reconstructed (MGR), using the smallest size for the most part and making sure that none of the pattern pieces were wider than the finished width of my fabric (about 41 cm). I was most careful with the hemline, because I did not want any of my fabric to go to waste. Since the gown is not hemmed, but rather edged with singling stitches and then a tablet woven band, there was no need to add extra fabric. Here's the toile pieces laid out on my completed, washed and pressed twill:
After the cutting there was very little waste. If this was a garment that required buttons my scraps would have been completely used up. As it is, I think that I may make some buttons from the waste, just in case I want to alter it some time down the road and require buttons. I will keep the larger fragments for future mending.
If you are familiar with the Herjolfsnes 42 garment, you may notice one change in my pattern pieces. I found that I could cut the front and back waist-height gores out very economically with the front in one piece, and the back in two pieces. In the original garment, both were cut in two parts.
The long body seams are sewn mainly from the right side, with small and invisible stitches. However, the centre side seams were sewn with right sides together and stand upright. The text describing this treatment of the seams reads thusly:
"...aside from the seams that are laid into the fabric, there are seams where the two raw edges are overcast with very close overcast stitches, but instead of lying flat, the edges stand up."(Fransen et al., 2011)
The two longest of these upright seams occur at the exact sides of the garment. Could this be for ease of adjusting the size of a garment? It would be much easier to take in this type of seam, or insert a filler piece.
To sew the inlaid seams, I used a line of basting stitches to mark the seam line, while a 1 cm seam allowance of the top layer was pressed down and pinned in place for sewing. The seam allowances in the original garment were "never greater than seven mm" (Fransen et al., 2011). For my fabric, in which the weft threads are not as tightly packed as the original, I decided to add 3 mm to this width because the weave is looser and may come apart if a smaller seam allowance is used. Protect the seams! They are the most vulnerable part of my garment.
Here's an image that shows the two types of long seams: inlaid (sewn from the right side) on the left, and the upright seam on the right. All the seams were sewn using white, plied wool thread, les than 1 mm in diameter; while the overcasting was done using single-ply brown warp wool left over from weaving the fabric. I sewed the seams and did the overcasting using what I have dubbed 'the rule of two'.
The length of the sewing stitches and the distance between the overcast stitches are just two warp/weft threads each. You can see my little white stitches below.
The front and back central gores each have a rounded insertion point. This is an engineering marvel! Instead of on small point on which all the strain and weight of an entire gore has been placed, this pressure point has been increased to spread out the stress and weight. This makes the top of the gore very strong.
These gores were intentially cut about 2-3 cm longer than the slit so that they could be fit without worrying about the top edge fraying. Here are some images from the wrong side, showing the insertion of the back centre gore. The extra length can be smoothly trimmed away, leaving an intact and strong seam allowance for overcasting.
An overast inset gore (front pattern piece):
An overcast inset gore (back pattern piece):
All the long seams have been completed, and the next step is the neckline and sleeves.
Fransen, L., Norgaard, A., and Ostergard, E., Medieval Garments Reconstructed: Norase Clothing Patterns, Aarhus Universtiy Press, 2011, p. 30.
The first piece was removed from my loom a couple of weeks ago. I washed it and the yarns fulled a bit and bloomed very nicely. The fabric is cushy, but drapes very nicely because it is twill, and twill drapes like a dream!
This is the first of two lengths that I require for the dress. It's about 41 cm wide and 4.5 m long. The front side it weft faced and shows more of the mustard yellow, whereas the back is warp faced and shows more brown. The second piece is underway as I write this...with only a couple more metres to go!
There's quite a difference in the openness of the weave when the cloth is stretched on the loom versus the state after it has been washed. The threads spring back towards each other, and bloom and full in the wash. This fills in all the little gaps. Do not fear the gaps!
I'm now making a muslin pattern for the Herjolfsnes 42 (acc. no. 10584) gown. Last night was the sleeves. I love drafting sleeves! This is a mighty project. It is filled with bliss.
The top edge of the hood rises in a peak at the face opening, just like many of the Herjolfsnes hoods. This enables the edge to be turned back without ugly ripples forming in the hood sides and, probably more importantly, keeps the hood opening from being stretched and stressed. The pattern was cut so that the face opening is a selvage edge. This has also been found in some of the Herjolfsnes hoods. No further seam finishing or hemming was required.
The hood required almost every small scrap from my fisinshed twill piece (1.5 m x 0.45 m). There are 14 buttons on the front opening and two spares that I have tucked away in my sewing kit, just in case. I finished the bottom edge of the hood with singling stitches. These are small running stitches that are only visible from the back side of the fabric and run in a serpenting pattern from the bottom edge, up about 1.5 - 2 cm, over a few weft threads, and back down again, all the way around the edge. You can just make out the top turning of the singling in this photo:
The singling stitches reinforce this edge and will help to keep it strong, and from stretching out. My original plan was to add a woven edging at this singling edge, and I used tablets, threaded in all four holes for the edge tablets and two holes for four middle tablets, but did not like the result: too bulky. So, snip, snip. I removed the whole edge. In these picture you can see the bulky edging, from the right side of the hood, and the back side. In the brown back side photo, you can see how the edging grabs on tho the singling stitches. But, in the end it was too stiff, and not pleasant.
This is one of the best parts of making things by hand. Learning what works, and what does not. Trying new things, and not being afraid to rip out 10 hours of work. The experience of producing this hood is every bit as important as the final object.
I am more please with a turned hem, stab stitches and overcast edge. It is neat and lies smoothly on my shoulders and across the top of my back:
Because the hood used almost all of my fabric, the liripipe required some piecing. I love this part of the hood! Use all the scraps, make it work, no waste! Can you see the piecing?
Here are some details showing the gore insertion, buttons and button holes. I love my new hood!
Presenting my challenge to the Queen. Photograph: Kyle Andrews (Two Ravens).
I have been struck by what I will describe as a 14th-century slow fashion movement. I want to make my own clothes from the yarn up, slowly and deliberately. I have decided, having almost no weaving experience, to weave my own cloth for most of my clothing and accessories from this point onward.
My inspirations include:
- the video for the Reconstruction of the Hammerum Girl's Dress (produced by Museum Midtjylland, Denmark, (www.museummidtjylland.dk/). Although the entire video is amazing, it was the warp weighted weaving that I really enjoyed watching: the slow and deliberate placement of each weft thread, followed by skilled manipulation of the string heddles.
Images from Reconstruction of the Hammerum Girl’s Dress (Museum Midtjylland, Denmark). These details show the manipulation of the three string heddles in order to produce the 2/2 twill cloth for the gown. The fourth image show the loose placement of the weft in the warp shed in order to accommodate the over/under path of the weave.
- Riina Rammo’s PhD thesis on the TEXTILE FINDS FROM MEDIEVAL CESSPITS IN TARTU: TECHNOLOGY, TRADE AND CONSUMPTION (Rammo, 2015). The majority of these 13th and 14th century textiles from Tartu, Estonia were either tabby weave or 2/1 twill. There are really wonderful close up images of some of these textile fragments on the Hibernaatiopesäke blog (http://hibernaatio.blogspot.ca/2015/1
2/aarreaitta-tarton-keskiaikaiset.html) . In her thesis, which is available for download in a Finnish and English version, Dr. Rammo talks about the differences in the coarseness of the fragments and whether it can be assumed that some of the more more coarsely woven examples were produced locally from rural home looms, as opposed to being imported. The discussion concerning the post weaving treatment of the cloth (fulling, teasling, and shearing) is also very interesting. And, Dr. Rammo provides thread counts for both fine and coarse fragments.
- I also became very interested in producing my own archaeological-type textiles in my home, using a simple loom. I found abstracts online from the NORTH EUROPEAN SYMPOSIUM FOR ARCHAEOLOGICAL TEXTILES, held in Copenhagen in 2008. At the conference, Riina Rammo gave a presentation on the Tartu textiles, and Barbara Klessig from Holmbold State University gave a poster on re-creating archaeological textiles using a rigid heddle loom:
Barbara Klessig (Humboldt State University, USA)
Re-creating archaeological textiles: a comparison of tools, techniques and structures
“… By using techniques developed in weaving with multiple rigid heddles, archaeological textiles can not only be re-created but can also show the mechanics of how they could have been created on a warp weighted loom. This poster will show a comparison between a warp weight loom and the rigid heddle loom using multiple rigid heddles. Along with photographs and diagrams comparing the actions of the two types of looms, there will be samples of archaeological textiles re-created with multi rigid heddle techniques. Although the ends per inch or centimeter are limited, the use of this type of loom can give one insight into the workings and structures of archaeological textiles and how they could have been woven on early historical looms.”
I became convinced that a rigid heddle loom, which can easily do 2- and 3-shaft weaving, would be the perfect instrument to reproduce the tabby weaves and 2/1 twills that were recovered from the Tartu archaeological site. My hope was to produce coarser cloth similar to what was made in the medieval household - slowly and deliberately purpose-crafted. As it is possible to buy up to 15 dent heddles, this means that tabby weaves with 30 ends/inch (EPI), and 2/1 twills with 22-23 EPI are possible if two heddles are used. This density can be increased further when the cloth is fulled.
The Challenge: weaving the cloth
The gown that I intend to make is based on the Herjolfsnes 42 dress (Accession no. D10584) (Ostegard, 2009; Fransen et al., 2011). This is a loose dress that can be pulled on over the head, yet it is relatively close-fitted to the waist. The long sleeves are close-fitting, but without lacing or buttons. Instead, there are short slits in the lower arms, which would allow the sleves to be foldes back during messy work. The centre front gores are inserted into a rounded slit shape. The fullness at the bottom is aided by two gores on each side which run from the underarm to the floor (one set of side gores on the original dress is actually one piece with a false seam running through the middle to make it appear as though it is two).
Herjolfsnes 42, Accession no. D10584
I calculated that I require (after washing) approximately 10.5 m of 40 cm wide cloth. I will be using Vuorelma Vippela yarn (100% wool) from Finland, single ply, tight z-twist in brown for the warp and mustard for the weft, 380 m/100 g. Requirements: approximately 1250 g. I have sketched the pattern cutting design based on my measurements and one that has the least waste of the woven fabric. I think that this is the perfect width of cloth to produce the pattern pieces I need, with minimal waste. After washing, the cloth shouldl be 40 cm wide.
Here is a photograph of the fabric in progress:
The Challenge: sewing the dress
I have been practicing sewing seams, seam finishing and edge finishing as discussed by Ostegard (2009). My attempts to make tiny invisible stitches in linen thread for the seams have turned out very pleasing, although it takes a bit of time to get used to sewing the seams from the front of the cloth. I’ve also practiced my seams finishing and singling edge finishing. In this latter technique, small running stitches are made in a serpentine pattern on the inside surface of a cloth edge for reinforcement.
As I weave the cloth for my challenge, I have decided to construct a hood from the brown and red 2/1 twill. This will give me a chance to practice these sewing techniques on home-woven coth and I hope to iron out all my inexperience before I move on to the dress!
Fransen, L., Norgaard, A., and Ostergard, E., Medieval Garments Reconstructed: Norase Clothing Patterns, Aarhus Universtiy Press, 2011, 66-72.
Ostergard, Else, Woven into the Earth: Textiles from Norse Greenland, 2nd Ed., Aarhus Universtiy Press, 2009, pp. 171-173.
Rammo, Riina, Textile Finds from Medieval Cesspits in Tartu: Technology, Trade and Consumption, (Thesis) Tartu Ulikooli Kirjastus, 2015.
Here is the twill pattern of the loom state cloth (pre-washing):
Using a brown warp and a red weft. The front face of the fabric is very red, but the back is mainly brown. I love how this is going to look when both sides are visible - like in an unlined hood!
Before washing, I threaded 48cm width of warp, which during weaving (and off the loom) had a draw in to about 47cm. I was only able to weave about 145cm of cloth because of the way the warp yarns were wrapped around the cloth beam - the sheds became unnavigatable far sooner than I would've like. But, I have a cunning plan for next time that will give me several more prescious inches of fabric: crochet chaining across just in front of the warp dowel to squeeze the warp into a single plane.
After washing the fabric fulled up beautifully! Here is the recto:
And the verso:
With a bit of steam and pressing:
It's a nice medium-heavy weight cloth. Here are the dimensions:
width of warp threaded: 48cm
width on loom: 47 cm
length woven: 145cm
width after fulling (with soap in washer on delicate cycle): 39cm
length after fulling: 129cm
My plans are to make a hood with medium length shoulder cape: Norlund no. 67 (Acc. no D10598).
So, I ordered the book and have been carting it around the house for months, pouring over fabulous details and learning! My first goal was to practice making the tiny stitches that haven been found on the garments. I used scrap lengths of 2/2 twill and linen thread.
I learned to sew narrow (7 mm) seams invisibly from the right side of the fabric:
I learned how to make very small overcast stitches on the wrong side to fell the narrow seams:
I learned how to do singling stitches on the edge of the cloth, snaking the thread up about 1.5cm every other warp yarn, to reinforce the edge (used for hems and buttonhole edges):
All this practice was so much fun, and I feel that I have fundamentally changed the way I think about costuming.
So, I've been experimenting using 2 rigid heddles. 2/1 twill is 1.5 times more dense than plain weave. So, in direct warping the heddles you need to pull half again as many threads through the slots as you would for 1/1 tabby. Even though you will weave with two heddles, you direct warp with just the back heddle in the loom. Using the warp beam to tie on your warp, then pull one loop through the first slot in the heddle and wind it on the warping peg at the distance you have calculated. Then loop your yarn around the warp beam for the second loop and pull it AGAIN through the FIRST slot in the heddle, and place the loop over the warping peg. around the warping peg. You now have 2 loops (which wil become 4 ends) thorugh the first slot. The next slot in sequence will only need one loop. Continue this pattern to achieve the 1.5 density required for 2/1 twill. Here's a diagram:
Then wind the warp under tension, and thread it very carefully using the following pattern (keep your wits about you! Constant Vigilance!):
Then , ta da! You've got a three shaft loom set up.
shaft 1: back heddle up, front heddle neutral
shaft 2: front heddle up, back heddle neutral
shaft 3: both heddles down
The sheds are very generous with this set up. It's so much fun to weave 2/1 twill this way!
note about selvages: once every 4th pick or so you'll find that just slipping the shuttle through the shed will skip the outermost selvage thread. When you notice it's about to happen, just slip the shuttle underneath that first thread on it's way through the shed. Then having caught the outer thread, everything will be fine and smooth. No gaps!
So, here are some samples:
1. woven by changing the direction of the shaft at 2.5 cm intevals (1,2,3,1,2,3...1,2,3,2,1,3...2,1,3). 2. playing around with weft stripes. 3. 1,2,3 weaving with dark weft and light weft.
Here's what my fine wool 2/1 twill looks like. I've folded it over to show the back.
Specs for the red and brown fabric:
100% wool (Finland) 2-ply fingering weight, 320 m/100g
48cm threaded in heddle, approx. 285 ends, 1.9 m long warp
required approx. 175 g brown wool for warp.
progress: approx. 40 cm completed
draw in: 5%
Pasanen, Mervi, "Aarreaitta: Tarton keskiaikaiset tekstiilit / what a treasure: Medieval textiles from Tartu, Estonia," in: Hibernaatiopesake (blog), 2015.
Rammo, Riina, Textiles Finds from Medieval Cesspits in Tartu: Technology, Trade and Consumption, (Thesis) Tartu Uliooli Kirjastus, 2015.