Charles VI Coat Armour Repro, Part 4: Sewing It All Together

This is the fourth in a series of blog posts describing the process I undertook to recreate the child-sized coat armour on display at the Musée des beaux arts in Chartres, France. This installment covers the sewing techniques used in attaching all the pieces and finishing all the edges. You can find the earlier installments here:

  1. Dyeing an Imperfect Lampas
  2. Recreating the Pattern
  3. Padding and Quilting on a Frame

Once I had all the pieces quilted and cut to shape, it was time to stitch everything together, make the buttons and buttonholes, and finish all the edges. I used the techniques I observed during my examination of the original. If you want to compare what I’ve done against the original, you should take a look at the photos in my article <link>, which the curator took at my request and which are not published elsewhere.

I acquired red silk quilting thread from Superior Threads and undyed linen sewing thread from Wm. Booth, Draper.

I started by connecting the lining pieces along the long seams, using a sturdy, undyed linen thread and a tight running stitch.

The lining assemblies sewn together into one complete assembly along side and back seams.

The lining assemblies sewn together into one complete assembly along side and back seams.

 

I then did the same for the outer assemblies.

The outer assemblies sewn together along the side and back seams to make one complete assembly.

The outer assemblies sewn together along the side and back seams to make one complete assembly.

 

Sleeve assembly came next. The original coat did not have any exposed seams, including inside the sleeves. This meant that I had to fold the sleeves in half length-wise and first sew the silk edge together, good sides facing each other.

The red stitches are where I sewed first, to finish the outside seam.

The red stitches are where I sewed first, to finish the outside portion of the sleeve’s long seam.

 

I then folded over the edges on the linen and sewed them together using an invisible running stitch that ran between the folds, out of sight. This would have been a stronger join if I’d used a proper ladder stitch, which is this situation’s equivalent to using stab stitch instead of running stitch.

Using an invisible stitch to finish the lining-side seam on the sleeve's long seam.

Using an invisible stitch to finish the lining-side seam on the sleeve’s long seam.

 

A sewn-up sleeve, inside out.

A sewn-up sleeve, inside out. The curved long seam is running down the middle.

 

Very carefully, I turned the sleeves right-side-out when finished.

The right sleeve, laid flat so that it folds along the curved long seam.

The right sleeve, laid flat so that it folds along the curved long seam.

 

A matching set of sleeves. Left sleeve on the left, right sleeve on the right.

A matching set of sleeves. Left sleeve on the left, right sleeve on the right.

 

I was beginning to see the shape of this tiny coat armour coming together.

 

The shoulder seams have not yet been sewn together, and the lining assembly has not been inserted yet, but already it was starting to resemble the real thing.

The shoulder seams have not yet been sewn together, and the lining assembly has not been inserted yet, but already it was starting to resemble the real thing.

 

Next, I put the lining inside the shell assembly, with the flatter, base sides facing each other. I stitched along the back seam on the inside, connecting the shell with the lining with a large overcast stitch. I had to be careful not to poke the needle all the way through to prevent the light gray thread from appearing on the crimson outside fabric. The original maker used undyed linen thread, so I used the same.

The overcast stitching on the original was as crude and obvious as that seen here. This was a brute-force stitch whose purpose was entirely practical rather than for style.

The overcast stitching on the original was as crude and obvious as that seen here. This was a brute-force stitch whose purpose was entirely practical rather than for style.

 

I then attached the lining’s shoulder seams and the shell’s shoulder seams.

A view of the back, shoulder seams attached.

A view of the back, shoulder seams attached.

 

Remember when I mentioned making a mistake in the last blog post? I said that using the undyed linen for outer assembly’s base layer caused a problem. On the original coat, the buttonhole flap was created solely from the outer assembly while the placket behind it was created from the lining assembly. The backside of the buttonholes were going to be visible, of course, and that linen needed to be the good stuff. Instead, the backside of my outer assembly—the base layer of linen that had been stretched on the frame—was the undyed linen that I had intended to hide. To fix this discrepancy, I basted a strip of good cream linen to the backside of the area where the buttonholes would go.

 

I folded the lining assembly back out of the way, and laid down a strip of the finer linen along the opening edge on the inside of the shell or outer assembly.

I folded the lining assembly back out of the way, and laid down a strip of the finer linen along the opening edge on the inside of the shell or outer assembly.

 

 

Before sewing any buttonholes, I had to make some buttons and practice buttonholes to confirm their width against the originals on the coat.

Practice buttons and a swatch for practicing buttonholes. Always warm up your buttonholes on a practice scrap before doing them on the good stuff!

Practice buttons and a swatch for practicing buttonholes. Always warm up your buttonholes on a practice scrap before working them on the good stuff!

 

Next I sewed the buttonholes, starting at the bottom. Tip: Always start your buttonholes or your lacing holes from the least-visible end of the area you will be covering. For front openings, this is the bottom. For sleeves, this is the area furthest from the wrist, unless your buttons extend all the way up your arm. In that case, the elbow area is the best place to start, working your way out from the center. You’ll get better as you go, and your earlier, less-wonderful ones will not be the first ones scrutinized by onlookers.

I left a sparse amount of cotton padding between the fabric layers under the buttonholes; just enough to give it body, but not enough to hinder the sewing process.

 

Some of my buttonholes were noticeably longer than others. Oops. Fortunately, this is barely noticeable on the front. Note the quilting stitches on the body piece are crimson. This is the backside of the shell or outer assembly, which was quilted with crimson silk thread.

Some of my buttonholes were noticeably longer than others. Oops. Fortunately, this is barely noticeable on the front. Note the quilting stitches on the body piece are crimson. This is the backside of the shell or outer assembly, which was quilted with crimson silk thread.

 

 

I then started the placket, or underlap. I laid down a strip of crimson silk lampas over the undyed linen of the lining assembly’s left front opening. This red strip would lay directly behind the buttonholes when finished. I did not attach it on both long sides of the strip; only on the inside long side initially. I took care to attach it at a deep enough distance from the opening to eventually hide the basting stitches between the layers. Also, I had to stitch only as far as the first layer of linen, to prevent the thread from showing on the outside of the lining.

 

I attached the silk placket piece with crimson thread along its inside edge. The outer edge stayed open to accommodate the addition of padding later. Note the quilting stitches on the body piece blend in with the linen fabric. This is the backside of the lining assembly—the undyed base linen—quilted with undyed linen thread.)

I attached the silk placket piece with crimson thread along its inside edge. The outer edge stayed open to accommodate the addition of padding later. Note the quilting stitches on the body piece blend in with the linen fabric. This is the backside of the lining assembly—the undyed base linen—quilted with undyed linen thread.)

 

At this point I had not yet finished any edges.

 

You can see the unfinished neckline, front openings (including buttonhole edge and placket underneath), and hem.

You can see the unfinished neckline, front openings (including buttonhole edge and placket underneath), and hem.

 

Everyone’s favorite part of sewing is sleeve attachment (ha ha). Just as with a normal sleeve insertion, I put the right-side-out sleeves into the inside-out armholes from the inside of the garment. I sewed the silk sleeve edges to the silk armhole edges.

 

The silk outer edges have been sewn to each other. You can see some of the red silk stitching on the part of the armhole closest to the top of photo.

The silk outer edges have been sewn to each other. You can see some of the red silk stitching on the part of the armhole closest to the top of photo.

 

 

To finish the lining edges, I turned them inwards and whip-stitched them to each other with a tight, tiny stitch length for extra strength, as the original had.

 

The finishing on the inside of the armhole.

The finishing on the inside of the armhole.

 

With an s-curve sleeve cap like this coat has, the long seam of the sleeve runs down the back of the arm.

 

The sleeve seam sits in the middle of the back of the sleeve. This combined with the S-curve of the sleeve cap angles the sleeve downward.

The sleeve seam sits in the middle of the back of the sleeve. This combined with the S-curve of the sleeve cap angles the sleeve downward.

 

Once the sleeves were on, I returned my attention to finishing the buttonhole edge and placket. The original placket was padded, separated by a line of quilting down its center. I added raw cotton between the silk strip and the lining assembly and quilted the central line. Then I finished the buttonhole flap’s edge, using an invisible running stitch.

 

Here you can see the inner channel of the placket is completed, and the buttonhole edge is in progress.

Here you can see the inner channel of the placket is completed, and the buttonhole edge is in progress.

 

 

With the buttonholes finally squared away, I turned to completing the placket. I placed more cotton against the inner channel and then whip-stitched the placket closed, cutting it to the abbreviated, angled length seen on the original.

 

 

The original maker was not terribly concerned about which thread was used to close hems. Finishing was frequently done with undyed linen thread.

The original maker was not terribly concerned about which thread was used to close hems. Finishing was frequently done with undyed linen thread.

 

I then went back to the buttonhole edge and sewed small stab stitches parallel to the opening, about 1/4″ in from the edge to reinforce it, as I realized the original had this, and I’d almost skipped over it.

 

The placket does not extend the length of the front opening, but instead angles inward at the top and bottom, leaving one buttonhole at the top and 3 at the bottom with no reinforcement behind them. I misjudged something somewhere, because the original's placket leaves 2 buttonholes unprotected at the bottom, not 3, like mine.

The placket does not extend the length of the front opening, but instead angles inward at the top and bottom, leaving a few buttonholes with no reinforcement behind them. Here you can see the left front upper assembly folded back out of the way. The crimson lampas lies on the inner-facing side of the left front lining assembly.

 

But then, a small disaster presented itself. I realized I’d accidentally sewn an extra buttonhole! The original had 27 buttonholes, and I’d managed to sew 28. I decided to patch over the bottom-most buttonhole. To disguise the patch, I lined up its top edge against the bottom edge of the buttonhole above it and re-stitched the bottom half of that buttonhole to hold the patch in place. Then I folded over the patch’s edge on its right side, and attached it with tiny stitches. Later, when I finished the left edge and the hem, it would blend completely.

 

You can't see the top edge of the patch at all, because it's hidden under the bottom row of buttonhole stitches. On the inside, there's also a patch of cream linen.

You can’t see the top edge of the patch at all, because it’s hidden under the bottom row of buttonhole stitches. On the inside, there’s also a patch of cream linen.

 

 

With the buttonhole hullabaloo finally behind me, I connected the two assemblies to each other next. I sewed the lining (with placket) and shell/outer (with buttonholes) together with an invisible running stitch, just past the buttonholes.

 

An invisible running stitch connects the inner and outer assemblies to each other. This also hides the original basting stitch holding down the strip of lampas.

An invisible running stitch connects the inner and outer assemblies to each other. This also hides the original basting stitches holding down the strip of lampas on the placket and the strip of good linen behind the buttonholes respectively.

 

 

The neckline isn't finished yet, but you can see the ingenious method by which the placket was originally designed. The maker used the lining assembly edge to create the placket, while the shell or outer assembly was used to create the buttonhole flap.

The neckline isn’t finished yet, but you can see the ingenious method by which the placket was originally designed. The maker used the lining assembly edge to create the placket, while the shell or outer assembly was used to create the buttonhole flap.

 

 

You can see the patch covering the inside of Rogue Twenty Eight, the useless extra buttonhole, in this photo.

You can see the patch covering the inside of Rogue Twenty Eight, the useless extra buttonhole, in this photo.

 

Next came the button-side opening. I finished it by turning the rough edges of the outer assembly in to face the turned-in rough edges of the lining assembly. I whip-stitched them together using the undyed linen thread, as had been done with the original.

 

stuffed some cotton in the open edges of the two assemblies before attaching them to each other.

stuffed some cotton in the open edges of the two assemblies before attaching them to each other.

 

Finishing the neckline was the next logical step. The original coat does not have a collar. Instead, the neckline is bound with an on-grain strip of crimson lampas. I laid the binding strip good-side-down, against the good side of the neckline edge and stitched it with running stitches.

 

I started with a rather wide strip, to make sure I had enough. I could always cut it down if needed.

I started with a rather wide strip, to make sure I had enough. I could always cut it down if needed.

 

 

Then, I folded the strip over the rough edges and ratcheted it down tightly before overcast stitching the tucked-under binding to the lining.

 

The finished neckline. At this point, the buttons still need to be sewn on, and the cuffs and hem need to be finished.

The finished neckline. At this point, the buttons still need to be sewn on, and the cuffs and hem need to be finished.

 

The finished neckline, placket, and buttonhole flap.

The finished neckline, placket, and buttonhole flap.

 

Next came the buttons. The original coat had 11 spherical buttons and 16 flat-faced buttons, both covered in the silk lampas. The flat-faced buttons were stitched through in concentric circles. For the spherical ones, I was hard-pressed to come up with the right material for the stuffing. They were clearly not made from wood or metal, as they had some amount of give. It was also clear that self-stuffing them with the lampas would not achieve the desired firmness or smoothness. I know I’m going to get teased for this admission, but I ended up using lint from my clothes dryer’s lint trap. What can I say; it worked perfectly!

I made the flat buttons from a thick sheet of industrial, 100% wool felt from McMaster-Carr. I cut out disks and shaved off some wool at an angle on the underside to sharpen the edges and create the mild conical shape seen on the originals. Then I covered them with silk and stitched through in 3 concentric circles of stitches.

 

The buttons sit ready to be sewn to the front opening with a thread shank.

The buttons sit ready to be sewn to the front opening with a thread shank.

 

I sewed the buttons to the right front edge using a silk crimson buttonhole twist thread. To form a sturdy shank, I looped the thread between the backside of the button and through the edge 3–4 times before winding the thread around the loops as tightly as possible. I then poked the needle back through the edge of the front opening and tied a knot on the inside, between the outer and lining assemblies. Since the hem had not yet been sewn, I could easily reach up between the two assemblies and tie hidden knots.

 

The top button is flat, followed by the 11 spherical buttons, and then the remaining 16 flat-faced buttons.

The top button is flat, followed by the 11 spherical buttons, and then the remaining 16 flat-faced buttons.

 

 

Buttons in this time period were not off-set from the edge; they were placed directly on the edge and typically had shanks, rather than being the kind with holes cut in them for stitching them down flat on the surface of the garment.

Buttons in this time period were not off-set from the edge; they were placed directly on the edge and typically had shanks, rather than being the kind with holes cut in them for stitching them down flat on the surface of the garment.

 

 

Close-up of the flatfaced buttons attached.

Close-up of the flat-faced buttons attached.

 

On the original, the maker had done a running stitch about 1/4 inch in from the finished edge, presumably to keep the padding from bunching against it and reducing its sharpness. I did not take this step, but in retrospect I should have.

On to the sleeve hems. They were something of a mystery on the original. The original bottom edges had been worn away, so there is no way to know whether the current conserved finishing is accurate to the original. There may have been a binding like the neckline, or perhaps even a full separate cuff.

I chose the most straightforward course and folded in the rough edges, whip-stitching the hems closed using the red silk thread.

 

Maybe this is how the original coat's sleeve hems were done. It's plausible and in keeping with the techniques applied elsewhere on the garment.

Maybe this is how the original coat’s sleeve hems were done. It’s plausible and in keeping with the techniques applied elsewhere on the garment.

 

The skirt hem, with its shallow, wavy scallops, was the only remaining piece I had to finish. And man, it was the hardest part of all. This is because the deep fullness of the stuffed channels results in a superfluity of fabric at the hem. There’s no easy way to smooth out the rough edges before finishing them in the wavy shape seen on the original. Indeed, the hem on the coat armour is slightly gathered all around in order to make it lay flat rather than buckle.

Complicating matters further, the original assemblies had each been prick-stitched about 1/2 inch in from the hem edge, presumably to hold the padding up so it would not pool at the hem. The lining assembly had been stitched with linen, while the outer assembly was stitched with red silk. The short, widely-spaced stitches show on the front (i.e. on the lining and on the outside of the coat), while long floats must exist on the inside, out of sight. This means the maker had to finish the prick stitching first, and then attach the two assembly edges together last.

I was running out of time, as I was working on this last task in England, on my way to the International Medieval Congress at Leeds in July 2012. I was presenting my paper and planned to show the reproduction as a visual aid. As a result, after completing the prick-stitching on the lining assembly, I gave up on prick-stitching the outer assembly. I proceeded directly to hemming the two assemblies together with an overcast stitch. It was a challenge to wrangle the extra fabric to create smooth wave shapes. Later, I should have gone back, picked it out, and done the prick-stitching first to hold back the padding. Now, 5 years later, some padding has migrated south and is plumping up the hem too much. It’s on my to-do list.

 

View of the lining hem. Could be better. At the time, I had not differentiated between running and prick stitch, and so the stitches you see here are running.

View of the lining hem. Could be better. At the time, I had not differentiated between running and prick stitch, and so the stitches in linen you see here are running stitches.

 

View of the outer hem. Not quite right, but close.

View of the outer hem. Not quite right, but close.

 

At last, I was done (more or less). It turned out remarkably close to the original and I’m proud of the project. I finally understand how makers coaxed such sleek, structured silhouettes from puffy, quilted panels of fabric. With densely packed cotton built to varying and curving heights and widths with the help of a quilting frame, the body can be transformed into a new shape.

 

The final reproduction, back view.

The final reproduction, back view.

 

 

The final reproduction, back view.

The final reproduction, back view.

 

 

The final reproduction, inside view.

The final reproduction, inside view.

 

The inner and outer assemblies are only attached in a few places: the center back seam (the rough overcast stitching mentioned above), the armholes, and all the edges (front opening, neckline, and skirt hem). The original had a line of light stitching that ran horizontally around the waist on the lining side, but does not show through, obviously, to the outer fabric. This could have been an anchoring stitch at the time it was added, but felt as though it were disconnected by the time I examined it. I omitted it from my reproduction, unsure of its necessity or purpose.

 

Displaying my project in August 2012 at the Arts & Sciences Exhibit at Pennsic War, an event sponsored by the Society for Creative Anachronism.

Displaying my project in August 2012 at the Arts & Sciences Exhibit at Pennsic War, an event sponsored by the Society for Creative Anachronism.

 

I estimate that the time it took to create this garment once I had the materials in place was approximately 200 hours. Surely a more experienced maker could move considerably faster through the quilting and sewing. Even so, if the time were cut in half, that is still a significant resource sink. It’s clearer than ever that such garments could not have been cheap to purchase. In fact, they likely existed solely at the pleasure of the most monied classes. These were not cheaper substitutes for other forms of armour, but armour in their own right. Men-at-arms wore them in a complementary manner with other armour pieces, such as mail and possibly plate.

I hope you enjoyed this series and learned a few new things. I certainly learned a lot from the process, and as always, my mistakes taught me the most.

Charles VI Coat Armour Repro, Part 3: Padding and Quilting on a Frame

This is the third in a series of blog posts describing the process I undertook to recreate the child-sized coat armour on display at the Musée des beaux arts in Chartres, France. The prior installments can be found here:

  1. Dyeing an Imperfect Lampas
  2. Recreating the Pattern

The truly laborious stage was upon me now. I had dyed the silk lampas, mapped out the exact reproduction-sized pattern, and had acquired key materials:

  1. Appropriately weighted and twisted threads in crimson silk and natural linen
  2. High-quality, off-white linen for the lining (washed, dried and ironed)
  3. Medium-quality, natural linen for the interfacing (washed, dried, and ironed)
  4. A large amount of raw cotton
  5. A big, white, plastic quilting frame.

 

 

Clockwise from top left: undyed tabby-woven linen from fabrics-store.com; a warm, oyster tabby-woven linen from ulsterlinen.com, the crimson-dyed silk lampas, and raw cotton.

Clockwise from top left: undyed tabby-woven linen from fabrics-store.com; a warm, oyster tabby-woven linen from ulsterlinen.com, the crimson-dyed silk lampas, and raw cotton.

 

I deduced in the course of my examination of the coat that the complex padding and quilting design could best be executed with a quilting frame. The quilted channels are complex and variable in their size and curve. The cotton was placed between the quilt lines as well, rather than layered between fabric pieces and quilted through, sandwich-style. Given the variable nature of the channels, the least complicated way to proceed was to build the padding and quilting on top of a stretched-flat surface. Indeed, the original was made in two assemblies that bulged outward on their surfaces, but which lay relatively flat against each other on the inside. The give-away hint for the use of a frame is a D-shaped cross-section. Don’t forget to read my formal paper with my analysis if this topic interests you.

 

The lining and outer body assemblies were padded and quilted to separate base layers of linen.

The lining and outer body assemblies of the original were padded and quilted to separate base layers of linen. The silk lampas was reinforced with a layer of linen interlining. This would provide structure and support to the silk. It makes sense to use a backing of linen for any coat armour project involving a silk outer layer.

 

Since this entire garment was made from two mirror-images of padded and quilted assemblies, I decided to begin with the lining assemblies of the front pieces. First I prick-and-pounced the pattern stencil to transfer the lines and curves to a large rectangle of natural linen fabric. Then I reinforced the powdered ink with pen. This fabric would serve as the base upon which I would build the padding and quilting. The drawing of the pattern and quilt lines would be my guide. Because this fabric would never be seen anywhere on the finished garment, I used serviceable natural linen from fabrics-store.com.

 

Front piece patterns stretched on the frame. I went over the original prick-and-pounced outline with pen, though it appears faint in this photo.

Front piece patterns stretched on the frame. I went over the original prick-and-pounced outline with pen, though it appears faint in this photo.

 

I stretched the stenciled fabric on my quilting frame. I then laid another rectangle of high-quality cream-colored linen down on top of the stretched linen base. This linen was a perfect match to that of the original coat’s lining. I found it on ulsterlinen.com. It was tight plain weave with even threads, no slubs, and was simultaneously buttery to the touch and sturdy. Beginning in the middle, I pinned the top layer to the stretched layer along a central quilting line. This involved lifting the fabric and checking the stencil guides on the base fabric at regular intervals.

 

Smaller rectangle of creamy linen pinned to stretched base linen along a line of quilting in the piece's center.

Smaller rectangle of creamy linen pinned to stretched base linen along a quilting line in the piece’s center.

 

I stab-stitched the linen along this line, holding one hand above the frame and one below it to speed the stitching along. Stab-stitching is similar to running stitch, except that the needle is passed through the fabric at a perpendicular angle. Running stitch passes the needle at an angle well below 90°, because the needle is wiggling in and out of fabric layers before being drawn all the way through. Stab stitching is a sturdier stitch as a result because there is more thread moving between the layers for any given length of stitching.

When finished the first line, I folded the linen back to one side and placed a line of shaped, compressed, raw cotton padding up against the sewn join.

 

 

The first channel's padding, laid out and ready to be secured with stitching.

The first channel’s padding, laid out and ready to be secured with stitching.

 

In retrospect, I included two errors in my handling of the cotton for this project. The first was the density I achieved in each channel’s padding. My results felt too pliable and airy, and I was never quite satisfied with it. In casual conversation in 2013 with a new friend, Christian Cameron, he wondered whether soaking the cotton in water and then squeezing it to the shape desired might be a better way to get increased density for each channel. I discussed this with another friend, Jessica Finley, sometime later in 2015, who immediately experimented with it. She generously sent me her sample results, based on the dimensions of the top half of a front piece of the coat armour I had studied. It turned out markedly more rigid and compressed than mine did. Unfortunately, you cannot experience this with a photograph, but here’s a picture of her sample:

 

Bowed and carded cotton rolled into tubes and then secured into channels by machine. Jess added the silk layer by hand afterwards. Remarkably stiff!

Bowed cotton (see below for more on bowing) rolled into tubes and then secured into channels with machine sewing. Jess added the silk layer by hand afterwards to avoid staining the silk with soak-through from the wet cotton. The result was remarkably stiff—more so than the original—but the intervening centuries and handling of it could have contributed to an increased laxity that wasn’t there in the beginning. It could also be that the original had slightly less-compressed padding, but the method used here by Jess is completely plausible and scalable for different densities.

 

To get the cotton in the shape needed to fill the flaring and curving channels, she did the following:

  1. Soaked bowed (see below), raw cotton in water until it was fully drenched,
  2. Drained it and then pressed it flat on a level surface (like a cookie sheet),
  3. Treated it like a sheet of thick paper and rolled it up tightly, pulling off or adding portions as needed to expand or contract the shape, and
  4. Used a rolling pin to squeeze out excess water from the rolled tubes of cotton.

She also tested two different oils soaked into the cotton and compared them to cotton that had been doused with water and then compressed as described above. OIL 1 was boiled linseed oil combined with carbon while OIL 2 was plain boiled linseed oil. OIL 2 made the tube incredibly stiff while OIL 1 did not, and the third tube—made stiff with water—had an ideal result. The cotton was packed densely but still had some give.

 

Channels hardened with oils and water. This was a Goldilocks experiment—Oil 2 was too hard, Oil 1 was too soft, and Water was just right.

Channels hardened with oils and water. This was a Goldilocks experiment—Oil 2 made the cotton too hard, Oil 1 left it too soft, and H2O was just right.

 

The second problem was lumpiness. I did not work hard enough at homogenizing the raw cotton to reduce lumps and clumps. Jess helped fill in a missing piece of this puzzle again when she found evidence of a historical method called “bowing”, which makes raw cotton fluffy and uniform. I think that would have greatly helped prepare the cotton for compression, had I known about it when making my reproduction. Here’s a video of her trying it out. She confirms that bowed cotton is far easier to compress into flat sections that can then be rolled into smooth, curved shapes.

 

Two 10 oz. lumps of cotton. The top lump has not been bowed while the bottom lump has. Bowing helps create uniform air pockets throughout the lump and provide improved fluffiness.

Two 10 oz. quantities of cotton after being stored in identical plastic gallon-sized bags. The top lump has not been bowed while the bottom lump has. Bowing helps create uniform air pockets throughout, which improves fluffiness.

 

I now think that bowing the raw cotton in combination with water-assisted compression is the best way to achieve the density required in the real coat armours of the 14th century. When applying silk on top of a water-damp wad of cotton, however, it will be up to future experimenters to decide whether to baste down the interlining until the cotton is dry and then more carefully stitch the silk on top (as Jess did in her experiment), or to take a chance and apply both the interlining and silk at the same time over the damp cotton. If any of you try these techniques, please do share your results with me.

Meanwhile back in 2012, I folded the fabric back over the line of wadded cotton and pinned down along the next quilting line. I replaced the pins with another row of hand-stitching. I repeated this process, working outward from the center. Working from the center outward helps to minimize the fabric’s grain distortion as it is sewn down over humps of cotton with varying degrees of height and width in different areas.

 

The first channel of the first front piece, stuffed and quilted down.

The first channel of the first front piece, stuffed and quilted down.

 

Three padded and quilted channels.

Three padded and quilted channels.

 

When I reached the center-front opening, I then worked from the center to the side seam.

 

The first front lining assembly, padding and quilting completed.

The first front lining assembly, padding and quilting completed.

 

I continued this work from the center outward until the lining assemblies of both front pieces were completed.

 

Both front lining assemblies completed on the frame.

Both front lining assemblies completed on the frame. Note the parentheses-shaped waves in the fabric bracketing the waist area. More of the fabric was taken up in quilting the wider areas than in the waist area. This causes buckling and grain distortion. If you work from one side to the other, by the time you get to the far side, your fabric will be stretching in an increasingly wavy pattern. Make sure you work from the center outward.

 

While the single piece of base linen was cut large in order to fit two pattern pieces entirely on the quilting frame, I cut the finer linen smaller for the process of covering over and quilting the cotton in place. I took a guess at how much width to allow, though there were two other ways I could have arrived at a more precise measurement for the fabric width.

  1. While examining the original, I could have molded a measuring tape over every hump in each pattern piece’s widest points to find the width of the original fabric.
  2. Or, I could have used math—π × diameter/2, where diameter = width of the widest channel—to find an approximate width of fabric used to create each channel. If one were to slice a channel in half to look at its cross-section, it would look similar to a semi-circle or the letter “D” turned sideways, hence the formula for finding a circle’s circumference, divided by 2.

I moved on to creating the linings for the back pieces and the sleeves next. When I completed all lining assemblies, I cut out the body pieces with a generous seam allowance included. I left the sleeves attached to the original base fabric rectangle because I would eventually pad and quilt the other side of that fabric with the silk crimson assembly. More on that later below.

 

Front lining assemblies completed and cut out with a generous seam allowance.

Front lining assemblies, completed and cut out with a generous seam allowance. I could not create the scalloped hem until later when I had the silk assemblies done and everything could be sewn together.

 

One of the back lining assemblies, completed and cut out.

One of the back lining assemblies, completed and cut out.

 

I then repeated the entire process for the outer body piece assemblies topped in crimson silk lampas. I used the undyed linen as the base layer as well as for the interlining mentioned above, which reinforced the fine silk.  I used the modest, undyed linen for these layers of fabric because they would be out of sight once the garment was finished. Turns out I was incorrect on this matter. In the next post in this series, I’ll show you exactly where I went wrong and how I fixed it.

 

A line of padding applied on a front assembly made with an interlining of linen and the outward-facing silk lampas.

A line of padding applied on an outside back piece assembly made with an interlining of linen and crimson silk lampas. Here, the linen and lampas are folded back on the right.

 

 

A channel in the middle of one of the front pieces.

A channel in the middle of one of the back pieces, loaded with cotton and pinned for the next line of stitching.

 

Progress on one of the back pieces.

Progress on one of the back pieces.

 

 

Back piece assemblies, still on the frame.

Back piece assemblies, still on the frame.

 

On the right front piece, I mis-judged the width of the fashion fabric needed. I had to patch a small section as seen on the left in the photo below. On the finished garment, this mistake is virtually invisible.

 

Outer front pieces, still on the frame. Due to an under-estimation on fabric width needed, I had to patch the most bulbous part of the chest.

Outer front pieces, still on the frame. Due to under-estimating the fabric width needed, I had to patch the most bulbous part of the chest.

 

For the sleeves, I had to flip over the sleeve lining assemblies finished earlier and re-stretch them on the frame. The original coat’s sleeves were made this way—both the lining and outer assemblies were attached to the same base linen piece. This is why the sleeve was made of 6 layers of material while the body pieces were made of 7 layers. The lining and outer body pieces each had their own base layer of linen. As with the body assemblies, I used the fine crimson lampas and an interfacing of undyed linen for support to complete the outer portion of the sleeves. I stitched the quilt lines slightly off-center from the lining’s quilt lines to avoid making the crimson thread visible on the lining side.

 

The sleeve lining assembly and outer assembly were both attached to the same base layer of linen. The quilting thread of the outer layer had to be slightly offset from the quilting thread of the lining layer to avoid showing through on the lining side.

The sleeve lining assembly and outer assembly were both attached to the same base layer of linen. The quilting thread of the outer layer had to be slightly offset from the quilting thread of the lining layer to avoid showing through on the lining side.

 

 

Both sleeves completed.

Both sleeves completed on the frame.

 

Both sleeves, cut out. One is shown outer-side-up and the other is shown lining-side-up.

Both sleeves, cut out. One is shown outer-side-up and the other is shown lining-side-up.

 

At last, I completed all assemblies and was ready to sew them together!

 

All body pieces cut out and ready to be sewn together

All body pieces cut out with a generous seam allowance.

 

These garments were hefty and highly structured. Quilting frames enabled makers to put together thick and complex padded and quilted assemblies with relative ease. Embroiderers used frames in this time period for the same reasons pourpointiers, gambeson, and coat armour makers did—they make it extremely easy to control the materials involved. Without a tense base, you have to wrestle your materials quite a bit more and may even require special tools to control them. This is a clear-cut case for the use of Occam’s Razor. In addition, a D-shaped cross-section for any padded and quilted assembly reveals frame usage.

If you have the ability to make an adjustable frame that can be held on a sturdy base, all the better. I made due with a fixed-size frame and had to waste a lot more base linen as a result. Your frame needs to be held by something other than you, because you will need both hands free for the quilting.

In the next installment, I’ll show you how I sewed everything together, added buttons and buttonholes, and finished all the edges using the techniques found on the original. Stay tuned for that monster of a post next week! Update: Sewing It All Together, the final post.

Curved front vs. a straight front dress – thoughts

Sometimes I completely forget about things I’m not supposed to forget. This blog post is an example of that. I had this post 95% done sometime in 2014 and then promptly forgot all about it. I’m finally finishing it. I apologize to the person who inspired the post and I hope she still gets to see it, all this time later.

************************************************************

I got a query from a reader, Lady D of A Stitch in Time and Space, which I felt merited far more than a comment reply. The questions were:

  • Is there a reason to choose a curved front over straight front?
  • Does body shape make a difference?
  • Is one easier than the other?
  • Does it depend what closures or neckline you want to do?

Each one of these tempts me to fill reams of electronic paper with my thoughts, so I’ll do my best to answer them succinctly, based on my own experience and tailoring preferences.

Is there a reason to choose a curved front over straight front?

Perhaps. The two cuts will shape the body differently because they both have different tailoring and rely on the concept of negative ease (tight fabric pressing and molding the body). The curved front provides a more rounded bust silhouette and it tends to sit lower than the straight front dress, which provides more of a shelf/corset effect and tends to push the bosom high. So, if you’re going for a particular shape to the bosom, keep these differences in mind when picking which style to make. If you care more about which will be easier to get a flattering, proper fit from, I’d argue that adding curve to the sides and the front seams provides a lot more leeway for imperfect fit which will still result in a good bust shape than a pattern which only adds curve to the side seams. For the straight front, you really have to get it just right, or it’s very easy to make the bust look terrible. I’ve seen way too many straight front dresses with an unfortunate effect that squishes the bust on the lower outside quadrant in a diagonal that gives the lower curve an unnatural “V” shape. When done well, it’s magnificent. But far fewer people actually do it well than think they do. It takes either a serious affinity for tailoring or a LOT of practice to make it look right, in my opinion. If any of you got it in one, congratulations — you have a serious affinity for tailoring.

Does body shape make a difference?

It can, if you wish to be that discerning about it. Every rib cage and bust is different, and it’s possible that a curved front works better for one while a straight front works better for another. I believe that for large-busted women, a curved front is more forgiving. For a very small bust, it’s hard to make the case for a curved front, as the fabric does not need to swell out much at all to accommodate a modest cup size. But, even so, there are small-busted women who prefer a curve and there are large-busted women who swear by the straight front. This is the kind of thing each woman needs to decide for herself, in my opinion, possibly through trial and error.

Is one easier than the other?

Easier to fit? I’d say the curved front is easier to fit for a good look. Sewing-wise, the straight front is a bit easier, because your eyelets do not need to be distributed over curves. Also, if you lose weight, a straight front is a lot easier to take in without needing to redo eyelets. If you gain weight, it’s a tie, because proper expansion of the dress needs to occur along the side seams, armholes, and sleeves. The center front is the one thing that can get away with not being changed until there is a pretty large amount of weight gain.

Does it depend what closures or neckline you want to do?

If you’re fitting for serious bust support, you pretty much have to use lacing instead of buttons for your closure. If you’re making a fancier top layer, buttons have a better chance of laying down without the edge ruffling when that edge is straight. That said, we do have 14th century examples of extant garments (albeit men’s) with serious curve in their center-front openings where buttons are used to good effect. (Two of my favorites, if you are wondering: the coat armour attributed to Charles VI of France and the pourpoint attributed to Charles de Blois.) Neither of these tailoring styles rule out any particular closure type. I will point out, however, that if you are planning to side-lace, the curved-front design is risky, because when you sew your two curved panels together, it’s very easy to create a baggy pocket which does not get properly filled by the bust. People can easily misjudge how tight the bodice needs to be when doing this, and so I recommend using straight front tailoring for dresses which will side lace. Also, it will allow you to cut the front in one piece, creating a smooth expanse across the bosom. Here’s an example of me wearing a dress which side-laces and was cut in one piece across the front. The fit was nicely done by Charlotte Wurtzel Johnson. I used this dress as the base layer under my black dress with gold accents for my 1480s English Lady impression.

Straight front side laced square neck

This dress has a straight front, cut in one piece, it side laces, and the neck is a softened square.

Side view of a straight front dress

The side view shows the shape of the bust using this fitting style. A very experienced and talented fitter, Charlotte, draped me, so it looks pretty good.

The neckline does affect the fit of the bust, and to repeat what I’ve said above, the harder fit is the straight front. So, the wider and deeper the neckline, the more challenging it can be to get the bust to fit well. I recommend you attempt a neckline that is as wide as you can manage it without doing away with the shoulder seam altogether. I also recommend covering the whole bust — not dipping too low into the cleavage that tends to occur with a tightly-fitted dress like this. The dress above has a square neckline with slight rounding at the corners.

Check out this past post about necklines for a survey of necklines on fitted dresses in the art of the time.

Anatomy of my bodice pattern for 14th century bust support

I’m going to break down the tailoring details of how I typically attain 14thc century bust support, and more specifically, the kind that looks right for the last two decades of that century in Western Europe. Please note, though I talk about a specific method in this post, I happen to also like other methods with different tailoring techniques and placement of the lacing. I also can’t wait to see the work that develops from the Austrian bra-like finds. There has long been an undercurrent of rumor that visible historical costumers like me are dogmatic in our approach to such things, but I find that silly. By the very nature of what I am seeking to learn, there is no corner on the market of knowledge because so much is still yet unknown. All I have ever done is conduct experiments based on viable theories. And what fun it has been!

***

Henri de Mondeville, before his death in 1320, wrote in his work Cirurgia (translated into English):

“Some women, unable or unwilling to resort to a surgeon, or not wanting to reveal their indecency, make in their chemises two sacks proportioned to their breasts, but shallow, and they put them on every morning, and compress them as much as they can with a suitable bandage. Others, like the women of Montpellier, compress them with tight tunics and laces…” (bolding mine)

I first became aware of this gem from Will McLean’s blog, A Commonplace Book.

I suppose that makes me very much a “woman of Montpellier”, because I prefer to wear dresses requiring lacing and a tight fit in order to wrangle my bosom. Although, one really must ask, what did a surgeon have to offer the amply-endowed 14th century woman? I am both intrigued and afraid to know!

As with any journey from Point A to Point B, there is more than one way to solve this whole bust support issue in European 14th century clothing, with more than one result in comfort, aesthetics, and silhouette. I am partial to a front-lacing dress with curves built into the center front edges and the side seams. Of course, as discussed in a previous blog post, I still fudge the silhouette with generous shoulder seams, but I am of the opinion that changing that aspect of my pattern will not affect the rest of it enough to invalidate the version I’m sharing in this post.

This comfortably traditional pattern of mine for bust support has evolved through the years. It’s a silhouette that works well for the 1380s into the early 15th century, judging from the artistic portrayals across multiple geographic locations and media usage. I could stand to adjust the shoulder seams and neckline shape as already mentioned, but otherwise, I’m quite happy with it. To see how different this pattern is from my earlier efforts, take a quick look at the final pattern shape I came up with for the curved front seam method in my old photo essay on this topic. Keep that page open, and then compare to the photo of my more recent pattern below. Virtually unrelated!

There are a number of signature features to this bodice pattern’s tailoring. By analyzing them individually, we can better understand a form of patterning that will be effective for the challenge of wrangling the female bust.

Here’s a photo of the most recent custom bodice pattern which I use for my own late 14th century-style fitted dresses:

Tasha's traditional bodice pattern

Tasha’s traditional bodice pattern

Keep in mind that the principle at work behind this tailoring method is “negative ease”, which means that it’s skin-tight. This is not a pattern intended to skim the upper body. This is what my body silhouette would look like if it were flattened out and quartered. (YUCK!) Now I will highlight specific shapes and explain their purpose.

Under-bust points on the front piece

Under-bust points on the front piece

On both the center-front edge and the side seam there is a sharply defined point where the bust curve ends and the lines below the bust begin. When laced close, this creates a band of strength just below the bust which prevents the bust from creeping downward. It also creates a sharply defined pocket within which the bust can rest.

Bust curves on front piece

Bust curves on front piece

Note also that both the side seam and the center-front edge have a defined curve in the bust area. This helps distribute the bust evenly across the entire chest and helps reduce the tendency for the bust to collect in the middle. I find the curve on the side seam especially useful and important, even for women of a smaller size. The less important curve is the one in the front, as long as the side’s curve provides room for the bust.

Under-bust flare on front piec

Under-bust flare on front piec

Notice that the lines extending out from under the bust points are mostly straight for about three inches of length before they flare generously. This is the aforementioned band of strength which stabilizes the bust and gives you confidence that everything is going to stay where it should for long periods of time. The subsequent flare allows the dress to skim over the tummy while maintaining the appearance of a defined waist. Much of the later 14th century and early 15th century figural imagery portrays women with higher waists than we modern folks are used to. If it is too close-fitting across the belly, every little variation in your curves will show, and that’s not always desirable (or comfortable!) either.

Armholes on front and back pieces

Armholes on front and back pieces

The armhole is deeper in the front than in the back. They are both cut with a strong curve and end right up in the armpit, though you can see that the back piece’s armhole is somewhat shallower than the front piece’s. This improves range of motion of the arm in its natural position (slightly towards the front of the body). It also places the armhole seam at the fulcrum of your shoulder joint’s movement, which keeps it from binding your mobility. My initial photo essays on the draping method for this style showed the armholes much larger than I would now deem ideal for this style of dress.

The curve of the center-back seam

The curve of the center-back seam

The center back seam closely follows the curve of the body all the way to the top of the rear end. This accentuates the S-curve at the bottom of the spine — a feminine feature much appreciated aesthetically. You may have also noticed that the back piece is much thinner than the front piece. This is because there is typically more of you on the front of your body than the back, if we are using the side seam as the splitting point.

Tasha, August 2012

Tasha, August 2012

Here I am, wearing a linen dress I made from the pattern above. Pretty comfortable and definitely supportive.

Other people have different signature bodice patterns that they are equally happy with, and I encourage you to experiment until you find the right one(s) for you.

The cotehardie that ate my brain!

My sweetheart’s birthday is in early November. He gently suggested I might make him something historically-themed and  fine to wear. I thought this was a capital idea, given my experience and interests. I decided to make him a cotehardie in the style of Charles de Blois. I have my own pattern for easy fitting, after all, and I even had enough of the fine lampas I used for the Charles VI recreation left over to use. I was off to the races.

Greg hanging out with a circa 1400 harness

Greg hanging out with a circa 1400 harness

I took measurements and then altered the sleeves on my commercial pattern for a size 42 chest. Greg has really long arms, so I knew better than to go with the pre-sized sleeves. I lengthened them by a good three inches, adding room to the upper arm piece, the forearm piece, and the cuff. Next, I ran the pattern up using a bubblegum-pink linen I had in my stash but knew I’d never have the heart to use for a finished garment. I don’t usually waste linen on a mock-up, but this color was hopeless with respect to my tastes, so it might as well be put to use.

Because Greg lives some distance from me, I mailed him this atrocity and had him try it on while we were Skyping. I took notes for alterations and he mailed it back to me. I reworked it slightly and mailed it to him again. This time the fit was correct. I proceeded to cut out the good stuff.

This is where most people quail. The fabric cost me $58 a yard (a bargain really; it originally retailed for about $300 a yard). But I’m weird and unafraid of cutting expensive fabric. I’m sure this s a fault, not a positive, but so far so good. Ironically, as gorgeous and complex as this silk lampas is, the pattern on it is not medieval. At best, it’s 18th century.

I sewed up the fashion layer and the lining, which incidentally is made from a gorgeous natural herringbone linen which I scored for $2 a yard in a private sale from a retired seamstress. This was the super-easy part. Next came hand-finishing, making all the buttons, and the dreaded buttonholes. By now, Greg’s birthday had come and gone and I was only getting started.

Somewhere around mid-November I began making buttons. And making them. And making them some more. All together, this cote ended up with 67 hand-made buttons. Which meant 67 buttonholes. *gulp*

A surfeit of buttons

A surfeit of buttons

I do believe the buttonholes drove me slightly mad. I had no time for anything else aside from making dinner, attending to my son’s homework and bedtime routine, and going to my day job. When I finally finished, it was the night before Greg was to wear it, this past weekend, at a Yule celebration held by friends of ours in Connecticut. But it got finished! And, it looked pretty spiff, if I say so myself. Greg was happy with it, which is what counts.

Tasha and Greg before the Christmas tree

Tasha and Greg before the Christmas tree