Embroidery of an Elizabethan Book Binding in Silk and Metal-Wrapped Threads
Free-Form Embroidery (Group II) – Beginner
Merewen de Sweynesheie
Goal: To do my first planned embroidery project, to embroider with silk, to work with metal threads, and to create something that I could use in the future.
1. Examination of Extant
2. Materials and Techniques
3. Deviations from Extant, Decisions, and Assumptions
|Photo by Master Eirik Andersen.|
I found a book binding in the online database of the British Library.
What it is - binding of a Greek New Testament dedicated to Queen Elizabeth I
Date - 16th century (estimated 1576)
Binding materials - listed as satin by the British Library website. A separate website I found identified the fabric as white ribbed silk. The binding has a front cover made of embroidered silk satin, set into a leather or wood binding; this is not original. Originally, the binding was entirely in silk, with the same image on the front and the back.
Description - There are roses, vines, strawberries, and leaves embroidered around the outside, the arms of Queen Elizabeth I in the center, and a great deal of silver-wrapped thread. The roses are of two types - the white rose common in works belonging to Queen Elizabeth the first, and the Tudor rose. Queen Elizabeth's arms are fully outlined in silver, and the lions on the arms are embroidered in gilt thread.
Size - Approximately 12 x 7 cm
Metal threads - There was some discrepancy in my sources as to the nature of the metals used. The British Library website lists the metal as silver. An alternative source, which did not have a photograph, but only a drawing, of the manuscript, listed the metal as gold. My own inspection of the photograph leads me to believe the British Library's claim. The majority of the metal-wrapped thread is either blackened (presumably tarnished) or silver in colour. The only gold I see is in the lions on the Queen's arms, which are a yellowish colour and not tarnished. the metal threads are used extensively in this piece: they comprise the vines, thistles, and borders, outline and mark veins in the leaves, encircle the roses, fill the roses' centers, and line the roses' leaves, outline every part of the Queen's arms, and outline the strawberries and their leaves. Most of the silver threads seem to be of the same thickness. The only exception is the square-ish border around the outer edge, which appear to be bullion-style metal-wrapped thread. The centers of the flowers appear to have been created by threading the coiled metal onto the embroidering thread and then stitching a shorter length than the length of metal. The flowers' tiny petals appear to be the same. The rest of the metal thread appears to be surface couched.
Other embroidery - When I originally viewed the British Library's website, I did not see many stitches on the binding. I assumed that this was because they had rotted away. I did see green stitches around the edges of the flowers and their tiny leaves, and what I believed were remnant colours from the coloured stitches that must once have been there. I did not know at first, but I later realized that the green stitches were done in stem stitch. I proceeded on this basis until I found the second source, which claimed that this was the only known extant binding to have been painted with watercolour painting and not embroidered with coloured thread. The coloured center of the thistles did not have much around them - the only indication of any stitches was a small amount on the right-hand side of the top thistle. I finally decided that this was likely either something that had come loose from another area or a thread outline. One other note - the crown at the top of the arms of Queen Elizabeth had one pearl remaining on it, with a few other holes where pearls must have been. From the quantity of the precious metals and the use of pearls in the piece, plus the use of silk as a base fabric, I assumed that the coloured threads used were silk.
Elizabethan embroidery stitches - from my online research, I found that common stitches used in Elizabethan embroidery included coral stitch, long and short stitch, split stitch, chain stitch, surface couching, satin stitch, backstitch, detached buttonhole stitch, tent stitch, double-running stitch, and laid and couched.
2. Materials and Techniques
|Photo by Master Eirik Andersen.|
a) Base fabric - indigo silk taffeta.
This was the only form of smooth silk I was able to find in town that would hold small enough stitches. It has some slubs, which are not so period, but is relatively smooth. In looking at the extant sample, I noticed several things. First, there is a small amount of white on the (mostly bare) fabric base. This either indicates that the satin used in the binding was white or that the satin was covered with white fill as a base. As I wanted to allow the blue silk to show through, I chose not to create a white background fill. This was my thought process before learning more about the base fabric.
b) Thread - silk, metal-wrapped, and
I used what I had in my own stash and what was available in town. This was of varying thicknesses and qualities. When using the larger yarns, I separated the plies to allow for finer embroidery work. The green, maroon, and blue were of such short staple that I had to re-twist and soak in boiling water after separating the plies so that they would hold a twist again and not break frequently.
• Red (18): YLI 100% silk thread, #100
• White: Wolves - YLI 100% silk thread, #100 (17), used because the wolves had detail too fine for the other white silk I had available to me. Flowers and shield - single ply untwisted from 3-ply silk yarn (3). 4 shows what the yarn looked like before untwisting.
• Green (7, 8), maroon (9, 10), and blue (11, 12): Short-staple silk yarn, 25g per 120 yards. 8 shows what the yarn looked like after separating the plies if I didn't re-twist and soak the yarn.
• Safflower couching thread: 20/2 2-ply silk yarn. 5 shows the yarn untwisted, 6 shows the yarn as it came originally.
• Black: 3-ply silk yarn, slightly lighter weight than the white. 1 shows the yarn after separating the plies, 2 shows the original yarn with one ply removed.
• Couching thread: silver-coloured stranded silk (16).
• Silver-wrapped thread: DMC Fil D'Argent Mi-Fin 5g (15), and Kreinik
thread metallic in silver
(14). For most lines, used two strands
of DMC and one strand of Kreinik couched together to better match extant
thickness. For bullion area, used five
strands of Kreinik thread twisted together and folded in half into a rope to
try to better match thickness and look of extant (19). For flower centers, used one strand of
• Gold-wrapped thread (13): Kreinik
thread metallic in gold, used five strands couched together to match more
closely thickness of extant. Japan
c) Mounting - hickory embroidery frame, tied with leather. Cotton buttonhole twist thread to mount to the frame. Linen backing basted on back for strength in mounting and stitching, then grosgrain ribbon added to edges for strength in mounting. I sewed the ribbon to the front, rather than to the back, because I wanted to make sure that the silk would not be damaged or pulled out of place by the string tying it to the frame.
d) Stitches -
• Satin stitch: white and red parts of roses, centers of maroon thistle parts, strawberry leaves, leaves
• Backstitch: white rose outlines, flower center outlines, black wolf outlines, black flagpoles, tree outlines, fess outline
• Stem stitch: Tudor rose outline, red flag outline, maroon thistle outline, black shield outline. After stitching around the white roses, I realized that the rose outlines should have been in stem stitch, so I proceeded in using stem stitch for the Tudor roses.
• Split stitch: red of strawberries, white of wolves
• Chain stitch: white trilliums on flags (loops held open with three stitches around loop)
• Long and short stitch: blue fess, tree fill
• Laid and couched: red of flags, white of shield
• Surface couching: all metal thread except rose centers. Based on the late time period of the piece, surface couching was more likely than underside couching, and the couching looked like surface couching.
• Modified bullion stitch: rose centers. My metal thread was not thick enough to thread onto my couching thread, so I had to find an alternate method of getting a similar look. I tried a number of different thread types and methods, and bullion stitch seemed like the best method. Unfortunately, when I tried to do this on my piece, the rose centers were not large or loose enough to take the needle down and back up through the fabric to complete the stitch. Instead of wrapping the thread around the needle, I had to complete a stitch and then repeatedly stitch through and around this stitch to create a similar look to the bullion stitch.
e) Marking method - slightly reconstruct using computer software, trace, prick and pounce, paint in watercolour.
First I used a computer program to slightly reconstruct the badly degraded image to make it easier to trace. I also made the changes to the center image via computer. I then used a light under a piece of glass (essentially a light table) to trace out the design on pattern paper, reconstructing the remaining degradation by hand, visually. I then used a needle to prick through the new pattern and the correct-sized modified version. I pricked both because if one didn't work well for pouncing, the other one should. I ended up using a combination of both to get the whole pattern. The pattern paper produced finer lines, and the regular paper let more pounce through. I painted with silver watercolour, brushing off and pouncing again when needed. The watercolour was used to draw the design out more permanently so that it would not fade out over the time it took to stitch.
f) Needles - Mostly John James Embroidery needles in size 8. Otherwise Singer assorted hand needles.
3. Deviations from Extant, Decisions, and Assumptions
|Photo by THL Lassarfhina inghean Uilleag|
a) Pattern - I did not want to create a book for Queen Elizabeth I, so I decided to place my own arms in the center instead.
b) Colours in stitches or watercolour - Before reading otherwise, I noticed that the flowers and some of the other items have coloured outlines, a second, inner, outline in silver-wrapped thread, and then fill in another colour. Specifically, the roses and their leaves are outlined in green. Looking further into colours, the colours in the extant sample are all showing directly on the base of the satin fabric. There do not appear to be any stitches. I decided that if someone were to go to the trouble of laying down all of this silver, they would not have simply dyed the fabric, so the stitches must have rotted away, leaving only a few of their coloured fibres behind in the base fabric. Later, I read that watercolours had been used to create the colours in the pieces after I had already mounted the blue silk on the frame. I could not paint the range of colours needed well enough on the dark blue silk. I also had chosen this project to embroider, not to paint. I decided to proceed as if the colours had been stitched by using common embroidery stitches from that era and locale.
c) Further embroidery - I want to have the opportunity to embroider the back and spine of the book binding at a later date (once I know the thickness of the book), so I left the silk large and in the frame.
d) Fabric choice - I could not find silk satin locally, so I looked to extant examples in choosing an alternative. Common bindings from the period for this kind of embroidery were either silk satin or silk velvet. Locally, cotton or synthetic velvet was available. Extant examples using velvet were generally in dark red or green. I did not want to use synthetic, and cotton velveteen was only available in a pinkish red, which would have been dyed with a cheaper dye if done in period. Since this book belonged to a Queen and was embroidered with expensive metals and pearls, I did not believe that this was the right choice. The alternative was silk taffeta or silk dupioni, which is not period. Both had slubs, but the taffeta had far fewer than the dupioni, so I chose the taffeta. I chose the blue silk, as it would work best with my arms. Because the silk is lighter weight than velvet, canvas, or leather commonly used in this application, I placed a piece of linen on the backside and embroidered through both.
e) Back side of the embroidery - The backside of the embroidery will not be very important because it will be applied to a book, provided that there are not large lumps. I tried to do a few stitches to hold the majority of the embroidery threads in place, rather than tying knots, but did have to tie knots in the stranded couching silk, as stitching a few stitches didn't seem to hold it in place at all.
f) Embroidery frame - While trying to decide between velvet and silk, I looked into how I could mark the velvet effectively. This lead to me learning that the velvet would crush if held in a modern, circular frame. Upon further thought, I realized that metal threads would not hold up well to that kind of frame if I had to move the work to complete it, so I realized that I needed a slate frame. My husband stepped in to the rescue by making me one out of hickory. I intentionally specified it to be large so that I could use it for larger projects in the future, as well as having the embroidered fabric basted to a larger lining fabric for velvet. It ended up a little too large, however, so he made me a smaller frame.
g) Leaf veins - I did not add the silver lines for leaf veins into my piece. I felt that in the extant piece they served to identify the leaves as leaves by adding texture. For my piece, I felt that that green satin stitch gave sufficient texture that the silver was not needed, and that the silver would end up covering up the majority of the stitching, so I decided to leave these out.
h) Metal thread - For affordability, I could not use real silver and gold wrapped threads. I also had French wire, but it was steel-coloured and not silver-coloured, so I could not use it with other silver threads, and thus could not use it. The metal threads I was able to purchase locally were much thinner than those used in the extant, so I got creative in my use of the threads to better match the extant's look. This included couching multiple strands and types together, creating ropes, and using some for other kinds of stitches.
i) Use of similar materials - I saw both gold thread and pearls on the extant example, in an area that I was switching out with my own arms. I wanted to try to use the same materials as the original broderer had used. Using gilt thread was easy, because the bezants in my arms were already meant to be yellow/gold. I could not find an appropriate place to use pearls, but I may decide to add them in the future.
j) Reds - Examining the extant, the roses appeared to be more pink than the red in the arms. This would lead me to believe that the dye used for the roses was possibly madder, while a more costly red dye was used for the red in the arms. This became a non-issue, as I only had one red to work with.
k) Wolf pattern - The detail in the pattern was too fine for the pounce to properly mark it, so I had to improvise a bit while painting the lines. This resulted in the right hand and left hand wolves looking different. Paint line degradation while stitching other parts of the piece didn't help. I tried to make the two wolves look more alike during stitching, but they still ended up with unique looks.
l) Filling rose centers - I did a lot of experimentation to determine the best method of filling the flower centers with as accurate a look to the extant as possible.
A. Bullion stitch using DMC silver thread and stranded silk. Stitch through fabric with the stranded silk and to the wraps with the DMC. This formed very large, uneven loops, unsuitable to the task.
B. Some regular fill stitch like backstitch using the DMC thread. The DMC thread was found to be unsuitable to travelling through the fabric.
C. Backstitch using
fill. This worked, but the result was
extremely close to the fabric and did not stand out much. Japan
D. Backstitch using
fill, with heavier silk "padding" underneath. This lifted the stitches a little, but the
result showed the background fabric and padding stitches through the
embroidery. Not ideal. Japan
E. Bullion stitch with the
Attempting to use the bullion stitch to the book cover, it became apparent that it would not be so simple. The flower centers were far too small to be able to bring the needle down through and back up through the fabric within the shape. I ended up using a modified bullion stitch, which was done the same way except that a single stitch was formed on the fabric, then the needle was brought back up through the fabric to the front, where I wound it multiple times around the stitch before taking it back down through the fabric on the other side. The result was a stitch very similar to bullion stitch (or detached buttonhole? which is period), just slower to produce.
m) Thistle centers - I debated whether to trace the outline of the coloured oval center of the thistle with silver. In three out of the four thistles, there is no indication of any threads, metallic or otherwise, around the outside. On the top thistle in the extant example, there is a small amount of twisted something on the right hand side. Visually examining as much as I am able to with the images I have, this twisted thread does not look the same thickness or build as the silver threads in the rest of the piece. This leads me to believe that, just like the flowers, this thistle part was outlined in stem stitch with regular silk, not metal-wrapped thread.
|Photo by Lord Alexander Gladstone.|
Corbet, Mary, Goldwork on Velvet - Transferring Embroidery Design, 2011. Needle 'n Thread, accessed February 2015, from
de Holacombe, Christian, Prose, poems, points, and purls: Embroidered Book Covers, unknown date. The West Kingdom Needleworker's Guild, accessed in February 2015, from
Muller, Robin E, Textile and Embroidered Bookbindings of Medieval England and France, unknown date.
College of Art and . Accessed in February 2015, from Design, Nova Scotia, Canada
Nouum Testamentum, 1576. Archived online by the British Library Database of Bookbindings. Accessed February 2015, from