I have friends who joke about me counting threads in cloth.  I can readily laugh with them about it because textiles are a passion for me.  In part, because the historic construction of cloth fascinates me, but in part also because this is one of the few facets of Viking costume for which we have a definitive model.  Most of us who recreate these ancient bits of clothing know that we do so with great limitations regarding what patterning, cut and construction was used (particularly when it comes to women's dress).  There is always a great deal of guesswork involved on part of the costumer.

The cloth itself, however, we have thousands of examples of.  We know what this cloth looked like, we can see, even today, the texture and quality of textiles that the early Norse women crafted.  My patterning of garments is largely guesswork, but the cloth itself I can at least analyze and try to attempt to find textiles that bear similar visual qualities, including thread count.  Of course, one can bring up the argument that there are still faults with this (such as wrong breed of sheep or improper width of yarn leading to a more dense or loose cloth than that of a specific period example).  I still believe that studying the fabric yields us a wealth of information about dress in the past and can help us better reach the goal of more faithfully reproducing it.

One of the hardest things that I had to overcome (and I have mentioned this in past entries), is that notion that the Viking Era Norse were barbarians, that their craft was crude and clumsy.  While that might be true in some areas, the wealth of extant textiles does not bear this out.  There were fabrics, of wool, that had thread counts in excess of 150 threads per inch in the warp.  That cloth was exceptionally fine, even by today's standard.  A person of some status, who wore metal brooches and a string of beads, certainly did not garb herself in cloth as coarse as burlap.

Breaking those stereotypes should not stop with the fabric.  Another example (again, something that I have discussed at length in this forum, in online groups and in my classes) is that other elements could, perhaps, reflect the same sort of refinement and quality as did the cloth.

Long have I pondered how the large, open, and often poorly wrought (I have too many examples of this on my own costume) Herringbone Stitch has made itself so prevalent in Viking costume. 

I think part of this, has to do with some modern mentality of equating over-embellished design elements with wealth and status.  We love to see miles of trim or stitching on costume, and see it as a perfected work, rather than one that is over-wrought.  Interesting, I think, given that one of the staples of a modern woman's wardrobe is the rather understated, but always elegant, "Little Black Dress".  Simplicity can often speak volumes in the modern wardrobe, but we often bypass that concept in costuming because it, like the concept of fine wool fabric, does not fit with our own internal visualization of the past.

To look again at the scant evidence for the Herringbone stitch, see below:

• Hedeby harbor has a textile fragment that is hemmed using a herringbone stitch.  However, the herringbone is over the edge of the hem, and hence, that the part of the stitch we consider decorative was on the inside of the garment, with what would look like parallel rows of running stitch on the outside.  (As described by Inga Hagg)
• The find 6-8 Pavement at York has a fragment of a hemmed garment that is worked in Herringbone.  As with the Hedeby find, the herringbone is on the inside of the garment.  (Cordage and Fiber from 16-22 Coppergate, by Penelope Walton Rogers)
• The Mammen cushion has a complex variant of very closely worked herringbone over a seam (you can see that here http://www.cs.vassar.edu/~capriest/vikembroid.html).

• The Skjoldehamn tunic has a stitch similar to that used at Mammen at the neckline.  The issue with this one is that this find may or may not even be Viking and if it is, it seems to reflect a very different type of fashion than we find further south and that alone makes me leery of using it as a basis for embellishment on a Viking kit without further substantiating evidence.  Beyond that, the tunic itself is dated post-Viking era.  As with Mammen, this stitch is still much more narrow than reenactors use, and, again, it is more filled in.  (http://www.ceilingpress.com/Resources/Nye%20tanker%20om%20Skjoldehamnfunnet.pdf )

Breaking the Myth

So where did we go wrong with this?

I think the first fault lies in our own taste.  In our own desire to make something "rustic" that fits in with our own misconceptions of the capabilities of these people, we opt for things that fit in with this ideal, but also that are simple enough for anyone to add to their attire.  We also have this concept that "more stuff" means higher status.

The second issue is with the evidence, or rather, our misinterpretation of it.  One of the most circulated articles online is one entitled "Viking Embroidery" by Mistress Thora Sharptooth.  She has been a powerful inspiration in the SCA for digging deeper into Viking costume and textiles.  She has a series of works posted on-line that I think most of us would give credit as the best aids we had when starting out.  In her embroidery work there is a paragraph concerning the ornamentation of seams.

I feel the need to further elaborate on the statements contained in this measure, and help to provide context (as it is researching these items that helped me to better trace the origins of the myth).

1. Sutton Hoo - The author used this as a way to show possible evidence of a practice in Europe.  Note though, that this is not evidence of Vikings doing this (it is not Norse and it is too early).
2. Oseberg - The ring motif was used on a garment from the grave, but it does not, to me, appear to follow along a structural seam.  Perhaps it followed the edge of a neckline?  It is very unlikely that this followed something like the vertical seam of an aprondress (and even so, it is entirely unlike a herringbone stitch, in both form an execution for it to be considered valid evidence of that practice).  The rings are delicate and well formed, not something I would class in the same category with an oversized herringbone stitch.  I will note here that there is another small sample of embroidery from this grave that looks as though it could have followed a seam edge (or not), but it looks to be more akin to the type of work used at Mammen as it appears part of a larger decorative effort.  Regardless, this item again, does not cover a seam, but might have followed a line of the garment if the shape of the fragment represents a shaped piece of cloth (rather than something that degraded into this shape over time).
3. Hedeby and Birka - mentions cord and braid applied over a seam.  Unfortunately, I cannot actually find anything that really shows cord applied over a joining seam.  Rather, I have cataloged a number of uses of cord, braid or stitching used at the edges of a garment (where it can also serve an additional functional purpose of saving the edges from wear). The closest thing here that I see to our "seam embellishments" would be the braid laid over the darts at Hedeby. Once again, these are very different treatments than a herringbone worked over a joining seam (or on the outside of a garment in any fashion).  The edge treatments can all be found at the site here.  http://awanderingelf.weebly.com/blog-my-journey/viking-clothing-a-deeper-look-at-edges
4. Mammen cushion - Finally, we see an actual seam treatment from the Viking age.  This find is somewhat contested, as some researchers believe it to be an import (perhaps Saxon and from the same tradition as the Sutton Hoo cushion), while others see it as a local product.  There are also small hats from the Netherlands that have similar stitching that are dated to the same time period.  It has also been suggested that these hats might also be imported goods from the British Isles.
   

Context

Putting the evidence into context is of paramount importance.  Understanding something being referenced in that article is not necessarily proof that all of our notions of "seam embellishments" are documentable.  It is absolutely not documentation that a standard Herringbone, or catch stitch, was ever used in such a manner either.

Looking deeper at the sources and meanings of each item is valuable as we move beyond beginner and progress on a journey as a costumer who is attempting to recreate the past. 

My own opinion is that seams were likely not embellished.  Not only due to lack of archaeological proof, but because of the time it would take to craft something that would make reuse of a textile less likely.  It would be hard to take in that garment if it were so embellished.  I often also see people citing that it "reinforces" the seams.  Why then would you need to reinforce the seam of an aprondress (a garment that Hagg describes as only "slightly fitted"), when you are not taking time to sew that stitch around an armhole of a dress, or down the seam in the back of the sleeve at the elbow.  Both of those areas are far more likely to break from stress than a side seam of a slightly fitted dress.  Why not instead use these complex stitches (and the stitch from Mammen is indeed time consuming) on a functional textile such as the cushion cover that you will not need to alter, and that might possibly need the strength from the applied needlework?

Scale

And finally, I need to mention Scale, because I think a lack of understanding regarding scale is one of the main contributors to the reenactorism of "decorative herringbone".  As I mentioned above, the only thing close to the reenactor's use of herringbone is the Mammen cushion cover.  It is important to view that piece in proper perspective.

What we typically see of this find is the diagram in the lower right corner.  This usually appears on our screen at a size of about a half an inch wide.  The weave of the needlework for the diagram is open, to better show how the stitches interlace.  What many people have not seen, is the actual photos of the cushion, which I have included here.  Note that the stitch is tightly worked, with no negative space visible to the eye.  In fact, it looks at first like applied braid.  And perhaps the most important thing here is to consider the scale of the piece.  That line of braid is a mere 3mm wide.

I think the single biggest perpetrator of the "herringbone seam embellishment" myth is a misunderstanding of that diagram (or rather, lack of understanding about its context).  If indeed you were to opt to use this type of stitching, I think the only credible way to do so would be to apply it in the same manner of the original.  While I personally do not think it would have been used over a dress seam, I can at least understand where the tradition stands historically and it reads as something that might be plausible for the Viking age.

I recently finished my first pleated-front aprondress.  I made this garment because Grave C27997 from Kaupang had layered smokkrs (aprondresses), with the innermost layer being a pleated garment.  (Ingstad, 162)  I love the look of the pleated dress layered under my handwoven, unpleated dress, but also decided that I very much like the garment on its own as well.

This version was made from lightweight linen, with the idea of being able to wear the garments layered even in warmer weather.  The cutting layout I chose was quite simple and produced a garment that has plenty of room for movement, while not giving too much volume to the hemline.  I added a bit of tailoring to the top of the side gores, but this is sill a loose, comfortable garment.  After making my dress, and wearing it, I realized there is s something exceptionally simple and functional about these pleated garments.  Pleats would be the absolute simplest manner of taking in a dress that is now too large, or letting out one in which you might need more room. 

Hilde Thunem has written rather extensively about the Kostrup dress, and her article can be found here: http://urd.priv.no/viking/kostrup.html  I will not reiterate her evidence in my piece, but intend to only add my thoughts on pleating that I see used in recreations, and what makes sense to me and what does not.

Before crafting my pleats I did some experimentation to get the size of the pleats correct.  Even though my dress is linen, I did experiment as well with wool.  I also did some research on my own on the extant pleated smokkrs from Kaupang, Kostrup and Vangsnes. 

The dress at Kaupang had pleats that were 4-5mm deep and was made from fine broken diamond twill (approx 88 threads per inch in the warp and 38 in the weft).  This textile is highly fragmented. 

Vangsnes B5265 has a larger fragment of pleated fabric as can be seen in the image below.  This was a wool tabby that had pleats 2-3mm deep.

Kostrup has the most compelling fragments and Hilde Thunem has the best images of those here:  http://urd.priv.no/viking/kostrup.html  That garment, like Vangsnes, was also a wool tabby.  The thread count is approximately 66 threads per inch in the warp and 25 in the weft.  The pleats there, according to Hilde Thunem, are 2-3 mm deep and 3 mm wide.  There is not enough left of the fabric to let us know how far down the pleats ran.

AnoThe Problem with Pleats

None of these dresses give much in the way of clues about how the pleats were crafted or secured. Things we do know are:

• There is no evidence of thread or stitching holding the pleats.
• There is no evidence of any sort of band sewn to the top or back of the pleats to let them keep their shape.
• There is lack of noticeable wear on the pleats and the fabric still has visible weave and texture.
• Two of the pleated fragments are hemmed at the top before pleating.
• The pleats are very narrow (2-5mm across all three samples).  Kostrup and Vangsnes are clearly stacked closely together and do not appear to lay flat.

Using Thread to Draw the Pleats
One theory, suggested by Rasmussen and Lønborg is that a linen thread was used to draw up and secure the pleats.  While no thread remains, linen might well have degraded in the grave leaving little or no trace.  Further, if indeed the garment was made in this manner it makes sense that a thread used for that purpose be of linen, rather than wool as wool would try to stick to itself and would catch making it difficult to form the pleats.  Linen, especially if waxed, glides though wool fabric easily.

The problem I have with this is after experimenting with this is that those pleats have a tendency to shift along the string with wear (unless they are very compact to begin with).  It is also very easy to break the the pleating thread or create a small pull or hole in the ground fabric.  However, it would be a fairly quick way to make the pleats and to secure them.

Steaming the Pleats
Quite a few people in online forums have have suggested that the pleats were set by steaming, causing the wool fabric to felt, thereby holding the pleats.  While steam can definitely be used to help set a pleat, it would not felt a textile to an extent that it would hold the shape of the pleats during wear at the edge of a garment (not and still have the individual threads in the weave still be plainly visible).

What do I mean by this?  If you make a wool skirt, create pleats and set them into a waistband and then steam it, those pleats will last some time.  However, without the waistband there, they will absolutely not hold shape to the waist, or in the case of an aprondress, hold shape at the bust.  The only way to get the pleats to maintain that shape would be to take the felting process much further than the evidence shows us as you can still see the weave in the cloth on these finds.  Even if one thought the garment was felted, I do not think there is enough felting to maintain shape without some other form of structural support.

Another option for steaming would be a more loose garment.  However, over time the weight of the wool itself would pull the pleats open, possibly in a sagging U shape to some extent.

Weaving/Steaming the Pleats
Another, more recent theory, suggested by Nille Glaesel is that the pleated portion of the dress is created on the loom.  The idea is that while weaving the dress fabric an additional linen weft is inserted periodically and then drawn together after weaving to create the pleats.  The textile is then steamed to set the the pleats and the strings are removed.  The problem with this is the same as I mentioned above.  The pleats are still not capable of holding shape at the bust.  Nille had to add a band to secure hers.  (Further, as a weaver, I find this far more tedious of a process than simply drawing the pleats on a thread after the garment is crafted.)

Bands at the Top or Behind
I see no clear evidence of a band sewn to the top of the garment, or even from the back, to keep the pleats from stretching open with wear though Nille Glaesel believes that blue linen found in the grave served that purpose.  My issue with this theory is that the reasoning behind weaving the pleats in to begin with is that there were no threads or holes left from a drawing thread or stabilizing thread.  Yet an assumption is being made that all of the threads holding this band on have since disappeared. 

Stabilizing Stitch
While there is no more evidence of this stitch than of the drawing thread, this is a practical solution that, to me, is at least plausible.  A careful seamstress could sew in the drawing thread without piercing the yarn of the textile (or, at least, without often doing it).  Likewise, Stitches added from behind could penetrate the web of the textile, without impaling individual weaving threads.  While I think it unlikely that all of this linen yarn would disappear in the grave, I find it far more likely than there being an entire band, plus its stitching, that disintegrated over time.

This is the solution I ended up using (and it is the same conclusion that others have come to as well).  We do not have the evidence to say that it was definitively done in this manner, but I feel that it is still more than plausible (and more plausible than some other possible solutions). 

I chose to use a method similar to that used on later Norwegian clothing where the pleats are drawn up on a thread, and whip stitched in place.

Unfortunately, I have seen no evidence for any of these types of pleats on any Viking garments.  (If you have seen something like this, I would love to see the sources.  I have looked through my collections of materials to no avail.)  While I do recognize that the "Swedish Poofy Pants" likely had some sort of pleating at the waist, we do not what what form that pleating took.  It might well have also been some sort of very tiny pleats or gathers.

I have to think that sometimes modern ideals guide these choices.  I know that there was a time that I never would have worn a pleated front dress (which looks to me very much like a maternity gown), preferring instead garments that were more tailored and flattering to a modern eye. 

Another reason that might prompt these choices is the wrong choice of fabric (and sometimes, this is unavoidable, as the right wools are frequently difficult to obtain).  The thread counts in the extant garments are very high.  These were fine wool textiles.  What we often buy as reenactors are much more coarse, and often our cloth also often has a brushed surface.  Thick fulled fabric will produce very bulky pleats that would appear quite off if they could even form a small enough pleat at all.  The bulk produced by these textiles can be very off-putting to many.  The fabrics to sometimes tend to conform nicely as deep, flat pleats, and that might be the reasoning behind some choices.


I plan to make another pleated dress out of a mid-weight linen for Pennsic and then eventually a wool one.  I enjoy the dress I made more than I thought I would and hope to explore additional possibilities for both the pleating and over all construction of the dress with each iteration.

References:

Blindheim, Charlotte. Kaupang-funnene, bind II, (University of Oslo), 1999.

Glaesel, Nille.  "The Kostrup Aprondress", 2015. 

Holm-Olsen, Inger Marie. “Noen gravfunn fra Vestlandet som kaster lys over vikingtidens kvinnedrakt”, Viking, 39, 1976.

Ingstad, Anne Stein.  “Two Women’s Graves with Textiles from Kaupang,” Universitetets Oldsaksamling 150 år, Jubileumsårbok, 1979

Rasmussen, Liisa and Lønborg, Bjarne. Dragtrester i grav ACQ, Køstrup, 1993.

Thunem, Hilde.  "Viking Women: Aprondress."  January 2011. http://urd.priv.no/viking/smokkr.html

Thunem, Hilde. "The aprondress from Kostrup (grave ACQ)."  April 2015.  http://urd.priv.no/viking/kostrup.html

Vedeler, Marianne, ‘Pleated Fragments of Clothing from Norway’, NESAT VIII, 1997.

I have made some updates to my Viking Textiles - A Deeper Look at Plaids, Stripes and Checks article.  Added in a bit more information I got about several pieces, as well as some possible weaving drafts of those for which I did not have images. 

Additionally, there is now a downloadable PDF at the end of the document that can be printed out to give a a rough sense of the scale for these plaids (since all monitors will show items differently, and many of the photos from the academic works are enlarged already).

http://awanderingelf.weebly.com/blog-my-journey/viking-textiles-a-deeper-look-at-plaids-stripes-and-checks

Silks were expensive imports and reflected the wealth and status of the wearer.  Many of these silks were cut into very narrow strips and applied as bindings or trim to garments.  The silks used in this manner were typically cut .5-1.9 cm wide (some were cut in the warp direction, some weft).  They were applied to both wool and linen garments. 

Some silk strips, as shown at Birka and with the Pskov cuffs, were used as a narrow binding on the edge of the fabric, while others were used flat against a garment (many of the stripes from Oseberg show this type of use).  In addition to these these sites, graves at Mammen, Haugen Rolvsoy, Tuna and Turinge also have silk strips.  (Vedeler, p7)

Both plain weave and samite silks were found in graves, with the latter being elaborately patterned.

Birka
Graves 464, 834 and 835 all have fragments of silk that were possibly attached to the tops of wool smokkrs.  464 is described as a narrow binding of silk with straps also made of silk twill attached to a dark blue-green wool smokkr.  834 had a scrap of silk binding that might have been attached to the dress or that might have been used as a strap to hold tools.  835 had a wool smokkr with linen loop.  A silk band was found there that might have been on the smokkr or might have been used to hang tools.

Birka III, p29, p78-79, p158-159;  Kvinnodräkten i Birka, p69.  Image from Kvinnodräkten i Birka and photo from Birka III (shows fragment from 834)

Oseberg
This rich grave yielded numerous narrow bands of silk samite that were cut into long strips.  These strips have the long edges folded in and there are needle holes show that they were once applied to fabric (traces of both wool and linen have been found on the bands).

Vedeler, p7.  Photos from Unimus.no.

​Hedeby
Grave 159/1960 has a dense, shiny tabby silk that possibly was a decorative band at the top of an aprondress.

Pskov
This Viking age grave contained both a dress and an underdress.  The over dress (referred to sometimes as a sarafan, and sometimes as a smokkr) had strips of silk applied in bands to the top.  That decorative portion of the dress was then trimmed with a very narrow, blue silk binding.  The binding was .7cm wide.

The underdress had wide silk cuffs and narrow bands of silk at the bottom edge of those (and potentially a band of silk at the hem as well).  The band used to form the hem was 3cm wide and folded around the edge as in the diagram below.  The sewn edge was only .5cm wide.

Reference and photos: Zubkova, et. Al.

York - High Ousegate
A sleeve cuff of chevron twill wool has Osenstitch applied to the very edge.  The yarn used is glossy, smoothly spun wool, plied Z2S.

Resource and image: Archaeology of York, Web Series, No. 3, The Textiles by Penelope Walton Rogers

Christensen, Arne Emil and Nockert, Margareta. Osebergfunnet: bind iv, Tekstilene (Universitetet i Oslo), 2006.

Geijer, Agnes.  Birka III, Die Textilefunde aus Den Grabern.  Uppsala,1938.

Hägg, Inga.  Die Textilefunde aus dem Hafen von Haithabu (Karl Wachlotz Verlag). 1984.

Hägg, Inga.  Die Textilefunde aus der Siedlung und us den Gräbern von Haithabu (Karl Wachlotz Verlag). 1991.

Historiska museet 2011.

Holm-Olsen, Inger Marie.  “Noen Gravfunn Fra Vestlandet Some Kaster Lys Over Vikingtidens Kvinnedrakt,” Viking Bind XXXIX, 1976.

Ingstad, Anne Stein.  "The Textiles in the Oseberg Ship". http://forest.gen.nz/Medieval/articles/Oseberg/textiles/TEXTILE.HTM

Ingstad, Anne Stein. “Two Women’s Graves with Textiles from Kaupang”

Jones, Heather Rose. "Archeological Sewing". 2004.  http://heatherrosejones.com/archaeologicalsewing/wool.html

Skre, Dagfinn.  Things from the Town: Artefacts and Inhabitants in Viking-Age Kaupang (Aarhus University Press), 2011.

Thunem, Hilde.  "Viking Women: Aprondress."  January 2011. http://urd.priv.no/viking/smokkr.html

“Universitetsmuseenes Fotoportal,” 2013.  http://www.unimus.no/foto/

Walton, P. "Textile Production at 16-22 Coppergate." The Archaeology of York Volume 17: The Small Finds.  1977.

Zubkova, E.S, Orinskaya, O.V, and Mikhailov, K.A. “Studies of the Textiles from the Excavation of Pskov in 2006,” NESAT X, 2009.

Zubkova, E.S, Orinskaya, O.V., and Likhachev, D.  “New Discovery of Viking Age Clothing from Pskon, Russia.”  (Notes and summary by Perer Beatson) http://members.ozemail.com.au/~chrisandpeter/sarafan/sarafan.htm

Walton Rogers, Penelope. “The Textiles,” Archaeology of York (28-29 High Ousegate), Web Series, No. 3.

As I stated some time ago, I recently started looking deeper at certain items that I see in proliferation in the reenacting community.  While many of these treatments are lovely they seem to have little evidence among archaeological finds.  An example of this is the herringbone stitch that we commonly see decorating every seam of a garment (you can see the results of that research HERE).  While on my mission to find use of that stitch, I came across several fantastic ways to embellish or strengthen garments that are, in fact, well documented to the Viking age.  This article is to share my discoveries in this area.

For this research project, the use of cord, braid or binding on garment edges, I chose to look at how often certain things we do know of occur in extant objects.  The criteria I used to weed through my data is as follows:

• Dated between 800-1050AD
• Location had to be Scandinavia or areas that had Viking settlements such as Dublin and York
• Use of textile material only (metal thread work and possaments are not included in this survey)
• The fragments have to be from a garment or that are assumed to be part of a garment by the professionals that surveyed them
• This survey does not include tablet weaving applied as trim at the edge of a garment

Types of edges included:

• Cord/braid applied to the edge, or near the edge, of a garment
• Binding
• Stitching

This survey is, of course, a work-in-progress.  I will be adding items to it as learn about them and as I have a chance to experiment with these finishing treatments myself.   If you have additions, comments or questions, please drop me a note so that I can look into the sources!

Cord or braid was applied at the edge of several extant items including caps from Dublin and smokkrs (aprondresses) from several sites.  In the article ““Two Women’s Graves with Textiles from Kaupang”, author Anne Stein Ingstad stated that she believed that a fragment of cloth from Kaupang that had brocaded tablet weaving at one edge and a cord sewn to the outside of that edge was likely the hem of the garment because of the applied cord would take the wear from use and thereby protect the more painstakingly wrought woven trim.

While I find it more likely that such an item was at the top of the garment (as is also evidenced at Birka), I still see value in cord used at any garment opening to protect the edges of the woven textile from wear.

One common factor in these finds is that the cords or braids are quite narrow in diameter or width.  Many of the cords or braids from Birka are noted for being constructed of very delicate or fine cords.  I believe that those used at the edges of garments would have been of a sturdy wool, and well spun, even if the yarn diameter is thin.  The exception to this is one of the Dublin caps that had string applied to the edge that was described as being wooly and produced from yarn that was not combed.  This edge would have been purely decorative, and it makes sense given that there would be no wear on the edge of a cap.

Only a few finds have comments on how the braid or cord was attached to the garment.  The pants from the Hedeby settlement, the caps from Dublin and the smokkr from Kaupang all employ an overcast stitch as a means of attachment.  One of the two Dublin caps was noted as using longer stitches (6mm) set a steep angle to create a more decorative effect.

Unfortunately many of the reports just list that woolen string or cord was applied to a garment, and they give little details on the actual type of cord it actually was.  A few, however, offer clues and those are listed below.  One thing to note is that all corded edgings made use of very narrow cord only (only a few millimeters) and not coarse, chunky yarns.

Braids

• Birka Grave 465 there was a braid with a triangular cross section that was 2-3mm wide.  Three additional graves list similar cord that is 3-4mm wide.
• At Hedeby there was a six strand braid of of 2-ply wool dyed red and yellow (three strands of each).  This was applied over a dart on a smokkr/aprondress, but because of its use on a garment I am including it here.  1-2mm wide.
• Birka Grave 466 lists a cord or braid with a triangular cross section.  It is listed as similar to other examples or possibly made with a fork (I assume this is a reference to lucet, but the author said it is impossible to tell how it was made).

Cords

• Birka Grave 973 had a string consisting of fine 2-ply (Z) yarns that were then cabled to the left (S).
• Birka Grave 954 had a cord 2 mm in average, consisting of several threads around which another yarn is wound.
• Kaupang Grave B lists wool string made from twining several woollen threads together.
• The settlement at Hedeby has fragments (19E) from pants that has a doubled folded hem with a thin decorative thread whipstitched to the very edge.  Twining is mentioned in the description so it might just be a plied yarn.
• While not evidenced in seams, there are cabled cords found in York and the ends of some of the scarves from Dublin are cabled (twining plied yarns together).
• At Sadanger there was a cord with several threads as a core wrapped around with another yarn.

Birka 465
This item is a 2cm strip of braid 2-3mm wide, having a triangular cross section.  This braid is referred to as type D3 in the Birka materials and was additionally found alongside tablet woven bands in several graves. 

Birka III, p128, p159, plate 33;  Kvinnodräkten i Birka, p60.  Photo from Birka III.

Birka 466
Fine woolen cord found on a wool tabby smokkr that was triangular in cross-section.  Geijer suggests that is was manufactured as the other braids, or perhaps was made with a fork (which I presume references a lucet).  The technique, however, cannot be determined with certainty.

Birka III, p128; Thunem, Smokkr; Photo from Historiska museet 2011.

Birka 511
Diamond twill smokkr and with cord considered more coarse than similar pieces from other graves.

Birka III, p160;  Kvinnodräkten i Birka, p123.  Images from Kvinnodräkten i Birka. 

Birka 563
Linen smokkr fragment with cord applied over the seam from the hem.  Hagg suggests that the cord was used to cover the hemming stitches (stitching on wool was often invisible, but the seams show more clearly on linen). 

Thunem, Smokkr; Kvinnodräkten i Birka, p125.  Images from Kvinnodräkten i Birka.

Birka 838
This smokkr fragment There is a tabby binding 1cm wide, turned to the outside and stitched down with stem stitch.  Fine wool cord is applied to the outer edge.  The stem stitches are in a straight line and are of twisted woolen yarn (Z). - The seam goes through both layers of fabric throughout and consists of unusually long (6-7 mm) stitches on the top and almost invisible stitches on the other side.

Birka III, p25, p109, p168; Kvinnodräkten i Birka.  Image from Kvinnodräkten i Birka; Photo from Birka III.

Birka 839, 860, 1004
1004 has a wool cord is sewn to the edge of a fine, dark wool tabby garment. The cord has triangular cross-section but is more coarse than other samples(3-4 mm) and is sewn to a edge that was folded and hemmed.  The same type of cord is found in graves 839 and 860.

Birka III, p128, p173, p168; Photo from Historiska museet 2011.

Birka 954
Very dark wool tabby smokkr (likely originally lined) with woolen cord.   The cord, approximately 2 mm, consists of several threads together with another yarn wound around them.

Birka III, p128; p171; Kvinnodräkten i Birka.  Image from Kvinnodräkten i Birka, photos from Birka III and Historiska museet 2011.

Birka 973
Diamond twill wool smokkr lined with tabby.  Cord edge is described as several quite delicate, 2-ply threads (Z), which were then twisted together (S).  This would be cabled cord.

Birka III, p128; p173; Images from Kvinnodräkten i Birka, photos form Birka III and Historiska museet 2011.

Birka 1084
Wool garment with string.

Thunem, Smokkr; Image from Kvinnodräkten i Birka.

Birka 1090
Blue and maroon striped wool tabby (5mm stripes) has two cords (both the same color) that might have been applied to the garment or that might have been part of a now missing tablet woven band.

Birka III, p36; Thunem, Smokkr; Image from Kvinnodräkten i Birka.

Dublin
Two of the wool caps found at Dublin exhibit cord sewn to the edge.  DHC30 is crafted from dark reddish brown tabby with an edging cord that is made of 6 Z spun strands that are S plied and then plaited.  The cord is 3mm diameter and is sewn on with Z spun, 2 ply (S) Black wool thread, approx 1mm wide using 4 stiches per centimeter.  Yarn is wooly and not combed.

DHC33 is a dark brown wool cap with edging dark brown edging cord, 5mm wide (same color as cap).  6 strands, 1mm each with each strand being Z spun, S plied (2ply) and then Z cabled.  It is whipstitched onto hem edge with long slanting stitches (to make it more decorative).  Stitches are 6mm in length, and there are 3 stitches every centimeter.

An additional fragment (DHC51) from Dublin has a cord inserted into a hem. The cord is 2mm wide, Z spun, 2 ply (Z), S cabled and then Z corded.

Heckett, p57.

Hedeby Settlement 
Fragment 19E is a fine wool tabby, likely originally blue or green.  This is believed to be from pants.  It has a double fold hem and with a cord attached with a whipstitch.

Photos and reference from Seidlung, p35, p108.

Hedeby Harbor
From the harbor there is a fragment of a brown wool tabby smokkr.  The applied treatment on this garment is not on the edge, but rather is sitting atop the darts that sit at the waist.  I include it here because there are decent descriptions of the braid available and it might be also suitable as an edge treatment.

The braid is made of made of six 2ply (Z spun, S ply) threads, three red and three yellow. interpret the text as describing a 6-way plait in which each thread consists of two z-spun plies which have been twisted together in s-form. The braid is only 1-2mm wide.

Instructions on how to craft a similar braid can be found here:  http://blog.eibeck.de/2014/plait-from-the-hedeby-apron-dress-fragment.html

Photos and reference from Hafen, p33-39.

Kaupang
Grave C 27997 B has a smokkr with a tablet woven band is sewn to edge of fabric with.  At the top of the edge of the band there is a firm, evenly twisted wool cord made from twining several S twisted threads together.  It is sewn on with an overcast stitch.

Thunem, Smokkr; Two Women’s Graves with Textiles from Kaupang.  Image from Two Women’s Graves with Textiles from Kaupang.

Køstrup
This find has a pleated smokkr that is accompanied by a tablet woven band with two woollen strings along each side of the band.  Unfortunately, they are uncertain as to how the cords, dress and band were attached (aside from a few whip stitches that attached the cords to the loop of the dress).

Hilde, Smokkr

Sadanger (Western Norway) Grave B 10720
This grave held remnants of a possible wool smokkr in a broken diamond twill.  There was a braided cord that is the to that found in Grave 838 in Birka. 

Birka III, p128; photo from ”Noen Gravfunn Fra Vestlandet Some Kaster Lys Over Vikingtidens Kvinnedrak”

Sadanger (Western Norway) Grave B 10722
This is a fragment of diamond twill that was edged with a cord made by twining a string around a core made from several strings. 

Thunem, Smokkr.  Photo from ”Noen Gravfunn Fra Vestlandet Some Kaster Lys Over Vikingtidens Kvinnedrak”

Christensen, Arne Emil and Nockert, Margareta. Osebergfunnet: bind iv, Tekstilene (Universitetet i Oslo), 2006.

Geijer, Agnes.  Birka III, Die Textilefunde aus Den Grabern.  Uppsala,1938.

Hägg, Inga.  Die Textilefunde aus dem Hafen von Haithabu (Karl Wachlotz Verlag). 1984.

Hägg, Inga.  Die Textilefunde aus der Siedlung und us den Gräbern von Haithabu (Karl Wachlotz Verlag). 1991.

Heckett, Elizabeth Wincott.  Viking Age Headcoverings from Dublin (Royal Irish Academy), 2003.

Historiska museet 2011.

Holm-Olsen, Inger Marie.  “Noen Gravfunn Fra Vestlandet Some Kaster Lys Over Vikingtidens Kvinnedrakt,” Viking Bind XXXIX, 1976.

Ingstad, Anne Stein.  "The Textiles in the Oseberg Ship". http://forest.gen.nz/Medieval/articles/Oseberg/textiles/TEXTILE.HTM

Ingstad, Anne Stein. “Two Women’s Graves with Textiles from Kaupang”

Jones, Heather Rose. "Archeological Sewing". 2004.  http://heatherrosejones.com/archaeologicalsewing/wool.html

Skre, Dagfinn.  Things from the Town: Artefacts and Inhabitants in Viking-Age Kaupang (Aarhus University Press), 2011.

Thunem, Hilde.  "Viking Women: Aprondress."  January 2011. http://urd.priv.no/viking/smokkr.html

“Universitetsmuseenes Fotoportal,” 2013.  http://www.unimus.no/foto/

Walton, P. "Textile Production at 16-22 Coppergate." The Archaeology of York Volume 17: The Small Finds.  1977.

Zubkova, E.S, Orinskaya, O.V, and Mikhailov, K.A. “Studies of the Textiles from the Excavation of Pskov in 2006,” NESAT X, 2009.

Zubkova, E.S, Orinskaya, O.V., and Likhachev, D.  “New Discovery of Viking Age Clothing from Pskon, Russia.”  (Notes and summary by Perer Beatson) http://members.ozemail.com.au/~chrisandpeter/sarafan/sarafan.htm

Walton Rogers, Penelope. “The Textiles,” Archaeology of York (28-29 High Ousegate), Web Series, No. 3.

One question that often comes up on message boards is whether or not plaids are appropriate for a Viking reenactor.  It is not an easy answer, but I have been putting together a list of things as I come across them and, in all honesty, I would have to say that at this point I personally would be very careful of what patterned fabrics I would use for a period impression.

If we look prior to the Viking age, we see a host of exceptional plaids.  Hulremose had an exceptional plaid from the an Iron Age bog find that is dated to the 2nd century BC.  Thorsberg, from the 3rd to 4th century CE has also yielded large, bold plaid textiles. 

The Viking age, however, does not seem to show the same love for vibrant, large scale patterning.  The technology was certainly there to craft large plaids or stripes, but the evidence that they were used is lacking.

What we do have, however, is a small selection of woven patterns that show a much smaller scale.  Delicate lines comprised of a single thread, or pair of threads, creating a repeating pattern, is much different than their more tartan-like predecessors.  I have looked at the mass of articles and books in my collection to see what sort of patterning was used and after eliminating all silks (likely imports) and self-patterning (meaning that the pattern was either a result spin-patterning - using S and Z spun threads in the same textile - or using threads of different grist to weave cloth with a plaid or stripe) I have found the following examples of patterned cloth from the Viking era and in Viking areas:

• Hedeby Graves Linen - Windowpane plaid linen.  This type of textile was found in two graves.  One was blue with white patterning and the other was red with blue, both of a tabby weave.  The patterning was TINY, the white and blue resulted in 4mm checks.  These textiles were likely from shirts or serks (underdresses).
  

These items are excluded from my list above due to date, or in the case of the Skjoldehamn find, due to the fact that it could potentially not be a Norse item.  I still consider them when looking at textiles to use, but would not choose something based solely on these finds were I looking for a patterned textile for a more accurate costume.

I do recognize that there are, of course, a couple of issues with looking at historic textiles in terms of color.  Often textiles are not analyzed for color, and tests can also come back inconclusive.  So there could indeed be more patterned (or even colored) cloth than which we currently can consider.  That being said, the sampling we do have is so exceptionally small in terms of the overall volume of finds, that I believe that patterned textiles in this time period were indeed in the minority.

Does that mean that I would never choose to use a plaid or stripe for a new piece in my Viking kit?  No, but any purchases I make from this point on will take into consideration that the evidence does show me one clear thing and that is the importance of scale.  The stripes tend to be narrow, and the plaids or checks are tiny.  Things that read as modern, such as pinstripes, might actually be a much better choice than plaids that would be better suited to a kilt. 

Were I to come across a larger, two-tone plaid or check in a heavy fabric, I think it might suit well enough for a cloak or blanket, but I would not choose to use it now as textile for a garment such as a tunic, dress or pants.  The larger checks from Haugen in Rolvsøy, which is a coarse textile, as well as the coarse, loose wool from Værnes seem better suited to those types of applications.  (And I would, at this point, only look at items with two colors, as seen in those fabrics as well as the Thorsberg and Skjoldehamn cloaks, rather than a plaid of many colors.)

Examples of fabrics I have or will be using, and my current thoughts on the choices, are below so that you can follow along with my thought process on this if you so choose.

Does this mean that you should never use that great paid you have in your stash for a Viking costume?  Of course not!  We are free to use what we will and I always advocate for "Use What You Have" when possible.  It is, however, good to know what the extant pieces looked like, and how frequently they were found compared to other types of cloth, when we go out to purchase new fabric for a historical garment.

And please, if you have evidence of other stripes, checks or plaids, I would absolutely love to see them!!!

To help give a sense of scale, I have included a single page that can be downloaded and printed on 8.5X11 inch paper.  This printout should be close enough to the correct size to give a rough sense of scale for these historic plaids.

RESOURCES

Blindheim, Charlotte. Vernesfunnene og kvinnedrakten i Norden i Vikingtiden, 1945.

Christensen, Arne Emil and Nockert, Margareta. Osebergfunnet: bind iv, Tekstilene (Universitetet i Oslo), 2006.

A.W.Brøgger. "Rolvsøyætten. Et arkeologisk bidrag til vikingetidens historie" _Bergens Museums Aarbok_. 1920-21

Geijer, Agnes.  Birka III, Die Textilefunde aus Den Grabern.  Uppsala,1938.

Hägg, Inga.  Die Textilefunde aus der Siedlung und us den Gräbern von Haithabu (Karl Wachlotz Verlag). 1991.

Hägg, Inga.  Die Textilefunde aus dem Hafen von Haithabu (Karl Wachlotz Verlag). 1984.

Pritchard, Frances.  "Aspects of the Wool Textiles from Viking Age Dublin", Museum of London.

2.  I learned quickly enough that there is no evidence for large scale appliques or embroidery on Viking costume.  I've gone further to learn how few embroideries we have at all compared to the volume of total textiles recovered.  This tells me that the few items we do have were quite rare and given the context from other items in the graves, they were reserved for the very wealthy.  (I have written further about this here http://awanderingelf.weebly.com/blog-my-journey/viking-embellishment-and-embroidery-part-3 )  But what I more recently accepted is that the practices of seam embellishment (decorating an existing seam), is probably not as common as reenactors (myself included) would like.

See those two Dublin caps above?  I do not actually consider those seams embellished.  All rows of stitching are necessary and functional.  Even if I used a brilliantly colored contrasting thread, those seams still serve a purpose. (Of course, use of colored thread on seams was also a rarity, but that is a topic for another time.) Examples of true seam embellishment would be laying cord over the seam and stitching it down or doing a decorative row of stitching by using something like osenstitch to cover the seam joining.  I personally love this type of decoration, but plan to use it a bit more judiciously in the future. 

What brought me to this conclusion?  Well, as mentioned earlier, I have been in the process of reevaluating things I learned early on in my journey.  One of those things was that herringbone stitch is period for Vikings.  At the time (when making my first dress) that was good enough for me!  Looking deeper, I have discovered that the stitch so favored by reenactors was never found used as an embellishment on a garment (at least not that I can find).

What examples do we have of this stitch specifically from Viking era and place?

• Hedeby harbor has a textile fragment that is hemmed using a herringbone stitch.  however, the herringbone is over the edge of the hem, and hence, likely that the part of the stitch we consider decorative was on the inside of the garment, with what would look like parallel rows of running stitch on the outside.  (As described by Inga Hagg)
• The find 6-8 Pavement at York has a fragment of a hemmed garment that is worked in Herringbone.  As with the Hedeby find, the herringbone is on the inside of the garment.  (Cordage and Fiber from 16-22 Coppergate, by Penelope Walton Rogers)
• The Mammen cushion has a variant of very closely worked herringbone over a seam (you can see that here http://www.cs.vassar.edu/~capriest/vikembroid.html ).  So here a decorative stitch is used over a seam of a household item, not a garment.  (And it is important to note that this stitching was only 2.5mm wide.  There is no open space in the stitching, making it resemble a tight, narrow braid.  This is entirely unlike the herringbone typically worked by reenactors over seams.)

• The Skjoldehamn tunic has a stitch similar to that used at Mammen at the neckline.  The issue with this one is that this find may or may not even be Viking and if it is, it seems to reflect a very different type of fashion than we find further south and that alone makes me leery of using it as a basis for embellishment on a Viking kit without further substantiating evidence.  Beyond that, the tunic itself is dated post-Viking era.  As with Mammen, this stitch is still much more narrow than reenactors use, and, again, it is more filled in.  (http://www.ceilingpress.com/Resources/Nye%20tanker%20om%20Skjoldehamnfunnet.pdf )

• Additionally, there is a 10th century reliquary pouch from York that uses a what Penelope Walton Rogers calls "catch stitch".  Typically catch stitch is another name for Herringbone, but in this case, the diagram in the materials shows a different type of blind catch stitch being executed. Stitches are predominantly on the inside of the item and this would not be a decorative stitch at all.  This item is described in an article in NESAT III and Textiles, Cordage and Fiber from 16-22 Coppergate, both by Penelope Walton Rogers.

Four samples only (out of hundreds of textiles), and none of them has the stitch displayed in the manner we commonly see it at events (and two of the four are different than a normal herringbone, two are inside the item, and one might not be Viking...).  This does not make a terribly good case in my eyes for extensive use of this decoration.

Now, will I remove all of my hard work from garments I have already made?  I cannot say that is the case.  I will, however, be more judicious in use of the stitch in future projects as I attempt to better recreate clothing of the past.

(The earlier segments of this article can be found here:  http://awanderingelf.weebly.com/blog-my-journey/viking-embellishment-and-embroidery-part-2  and here http://awanderingelf.weebly.com/blog-my-journey/viking-embellishment-and-embroidery-part-1 )

Embroidery, while evidenced, was not common, and often what we have is considered to be imported goods.

First, let me make a comment about how I differentiate types of embroidery that are often discussed among reenactors.  Often I see someone come onto a forum and ask about Viking embroidery.  Others quite often launch into long dissertations about how Vikings did not do embroidery.  What the poster sometimes means, however, is often not figurative, large-scale embroidery, but rather the seam embellishments that are frequently employed by reenactors.  For the purpose of this post I am going to categorize those three types differently as “seam embellishments”, "functional stitches" and “embroidery”.

Seam embellishment, while (in my opinion) over done by reenactors (myself included), is at least plausible and in evidence in several places.  The Hedeby Harbor aprondress fragment had a braid applied over a dart.  There is also mention of embroidery covering a seam in one of the Oseberg scraps and some of the apron dresses from Birka had narrow cording sewn to the top edge of the garment.

Functional stitches, on the other hand, are exactly what they sound like.  They are stitches that serve a purpose (to create a seam or to hold open the seam allowance) but because part of the stitch rests on the surface of the cloth, it can be quite attractive when stitched in a different color of thread. (Even something as simple as unbleached linen thread against a dark blue dyed linen garment can be striking in its simplicity.)

(Note 9/22/15: I was rereading Hilde Thunem's excellent paper on aprondresses and am really thinking hard about a remark the author made "Inga Hägg comments that the hemming stitches would usually be invisible on wool, but would show up clearly on linen. She believes that the reason that the band has been placed lower on the linen smokkr could be to cover the stitches."  I need to go back and look at Hägg's comments and some other things as well.  I want to chart out the extant seams from various finds and see what I can figure out about the visibility of the stitching and the threads used.  It could well be that they Viking era Norse preferred a very clean look, with little in the way at all of stitches showing.)

Ideas for using seam embellishment and functional stitches to create a more accurate impression:

• Dot not cover every inch of a garment with decorated seams.
• Fine cord or braided string as an edging has ample evidence (on apron dresses a string, especially if it were linen or silk, at the top might serve a functional purpose in helping the top of the garment to not stretch as much with wear).
• If the stitch serves a functional purpose, it actually makes more sense in use (example a running or overcast stitch on the surface that actually is part of a hem).  There is one example from the Hedeby Settlement (S 27) that uses buttonhole stitch to join two textiles.  Note though, that that was only one of several types of stitches used on that garment. (S 27 information added 9/30/15)
• Keep stitches narrow and compact, there is no evidence of an inch wide herringbone stitch anywhere.
• If you must add stitching, employ stitches that were used in period – examples of some stitches are running stitch, stem stitch, split stitch, osenstitch (though most of the the extant examples of this are metalwork).  Note here that while we as reenactors love herringbone, evidence for its use in period is actually quite limited.  It was used on a pouch from York and on the INSIDE of a hem at Hedeby.  A stitch similar to herringbone was used on the Mammen cushion cover (decorative item rather than a garment) and on the collar of the Skjoldehamn tunic (which may or may not even be Viking), but it was not the sprawling herringbone seen in the reenactment community.  For a deeper look at Herringbone see my post HERE.

And now on to the controversial topic of larger, decorative embroidery. 

The examples that we have are often are thought to have been Saxon, Rus or Byzantine imports (the first is thought to be the case with the larger Oseberg pieces in particular).  It is very important to understand that some of these were also considered to likely be from decorative goods, rather than garments.  This proves that the technique of heavily filling an area with stitching existed at the time, it does not at all indicate that this was applied to garments in period.  Also worthy of note is that all of these items were very, very high status.

Mistress Þóra Sharptooth has a good article about extant embroidery here:  http://www.cs.vassar.edu/~capriest/vikembroid.html