When FR was looking for someone to tackle the corsets of the 1905 Savoye patent #845582, I jumped at the opportunity. I had never made an Edwardian corset, and I had never made corset from a patent before, so I relished the chance to develop my skills. The first design, with the embroidery, especially held my interest.

Enlarging the pattern

With my first look at the pattern I thought I would most likely use Cathy's Corsetmaking Revolution to enlarge the pattern, as I have had great results with it when drafting corsets based on historical patterns.

On the second look I thought about it a bit more, and while I was sure that this method would work well with the vertical pieces, it could be quite difficult to use on panel a, with its rounded shape.

I decided to give it a first go by simply enlarging it to "human" size by using Photoshop and printing it out. This way I would also see how the original pattern would look on a body and if an actual person would have fit into the corset from the patent at all. 

I had never enlarged a pattern in Photoshop before so it took me a bit to find out what was the best way. In the end I just did what seemed logical to me. 

First I had to find a measurement to use as scaling guide. The centre front seemed best for that.

Most of my corsets based on historical patterns carry a 30cm busk. So I opted for something close to this length. For confirmation I had a look in Robert Doyle's Waisted Efforts and on page 165 found a corset pattern from the Symington Corset Co. Market Harborough no. 670050 from July 21st, 1906. It has a similar silhouette and also carries a 12" busk.  

With some allowance to sew on the bias ribbon later, I would enlarge the centre front to a length of 32cm (12 9/16") in total.

 

After opening the pattern picture in Photoshop I first had to resave it as .jpg-file to work with it as with the original .gif-file I couldn't even create a duplicated layer. Having solved this issue I created a new empty sheet with the measurements that would approximately be needed to hold the whole life size pattern plus some centimetres to spare. I ended up with a sheet of 60x60cm (23 5/8"x23 5/8").  With rulers visible (Ctrl +R or View -> rulers) I added two guiding lines (View -> new guide), one at 10cm (4") and one at 42cm (16 9/16"), to create the space of 32cm (12 9/16") to fit the CF into.

 

Going back to the pattern-file I duplicated the base layer and dragged this one into the empty sheet. 

 

With Ctrl +T (or right click-> transform) and the shift key pressed (this will keep the proportions) I sized up the pattern until the CF line fit in between the guiding lines. 

To print it out I had to save it as a PDF file. I had to Google how to print in poster mode as Photoshop doesn't offer it for my printer. So with Adobe Acrobat Reader I managed to print out the pattern on eight A4 sheets. Next time I won't use the overlap option though, as it isn't needed and it would save two sheets of paper.

 

Checking the Original Pattern

After all pages were glued together I first checked the centre front to make sure it had the right size. Then I determined the waist line and from there marked the other horizontal lines according to my general body block.  When I first looked at the proportions I had thought that it wouldn't be very close to my body, but after taking the first measurements I was surprised that it didn't turn out too far off. The waist of the corset was 48cm (18 15/16"), so 5cm (2") smaller than my usual corset waist of 53cm (21"). Being a historical pattern this was to be expected, but it was no big problem to adjust. 

 

While the upper hip measurement of 67,8cm (26 11/16") was a good 2cm (3/4") too small, the lower hip of 89cm (35 1/16") was 8cm (3 1/8") too big. As I consider myself to have quite a big hip spring already, the figure intended for this corset must have been even curvier - even more so if you take the underbust measurement into account. I wasn't able to find the exact position, but I tried to get as close as possible and got an incredible 84cm (33 1/16"). My underbust circumference in corsets is around 63cm (24 13/16").  I can't deny that I'm small build in my upper body, but even being two clothing sizes bigger (which would meet my hip size) would end with 13cm (5 1/8") too little for this pattern. 

As it's really hard to imagine that a woman could naturally be build to fit into this corset, we must assume that this corset was intended to be worn over some form of padding to create the typical Edwardian "pigeon" bust.

Another look at the pattern and the sketch of the corset shows that Savoye seems to have made some mistakes on the pattern pieces. Just as Joanne Arnett discovered while making up the corset patent #786,685, that parts of the panels are missing, in this patent parts of the panels are mixed up. You might have seen already that the upper edge of panel d is rounded while in panel e it's straight. On the sketch it is the other way round.

At first I thought that Savoye (or whoever sketched the patent) simply mixed up the labelling of the panels. But looking at the shape of the bottom edges this couldn't have happened. So they obviously drew on the wrong top edges. Also panel f seems to be flipped vertically. In the pattern the side that is attached to the rest of the corset is longer than the centre back side, in the sketch the CB is longer.

Despite all the mismatching I decided to leave the panels in their original shape, cut them out and see how they would fit together before altering them. With the help of my tailor's ham, I taped the panels together. To confirm my suspicions that the top of panel d would fit into the back curve of panel a, I laid them together and they were, indeed, a perfect match.

I continued taping the panels together in their intended order, but had to leave out the CB panel as the back edges of panel a and e didn't meet. 

 

I pinned the paper toile on my adjustable dress form to get a first impression, but it looked very off. 

 

So I padded my other dressform to a more Edwardian shape giving her a larger chest and back hips. The paper corset still stood away from the body around the hips and the underbust area. It wasn't as bad as before so I thought that it might become even less once made up from soft fabric instead of stiff paper.

 

 

First Mock-up

Before cutting the first mock-up I had to change a few things. I switched the top edges of panels d and e and flipped panel f. (Fig.19)  Assembling the panels turned out to be easier than I had expected and the pieces fit together quite well, even when attaching panel a at the top. With a seam allowance of 1.5cm (5/8"), sewing around the curves was a bit tricky. It would get easier later when using a smaller seam allowance.

However, now panel f was somehow too long. I assume this happened because I sewed panel a too low on panel e. When measuring the paper panels the bottom curve of panel a was a bit longer than the top curve of panels b-e. And as it now seemed to fit perfectly, it probably took away the length now missing to fit onto panel f.  I would have to true the seams again before cutting the second mock-up.

Putting the mock-up onto the dress form, it already fit much better than the paper version and it molded better around the body. The general shape looked quite Edwardian, at the front top and bottom was too much fabric, which would be even more on my (unpadded) body. 

 

Lacing myself in, I found the top edge sitting rather low, too low if the roundness was there to "cup" the underside of the bust, and of course the top was much too wide. I pinned some darts into panel a to see where best to take out the excess fabric. At the front bottom the excess fabric pooled around my belly, which was weird as for me S-curve corsets were actually build to press the pelvis backwards. 

 

I made the following alterations to the pattern before cutting the second mock-up:

• Lengthened  panels b-f by 1.5cm (5/8") at the top and checked/adjusted the underbust width.
• Took the excess width out of panel a and made the bottom curve fit on the top edge of panel b-e.
• Straightened the bottom of panel a and took another centimetre (3/8") from the front of panel b to get the centre front closer to the belly.
• Adjusted  the hip to my actual measurements, which meant adding more room to the upper hip and taking some excess from the lower hip.

You can see the alterations on the draft: the striped mark where I took width away; the other areas (they actually carry little dots, not really visible on the picture) mark where I added something.  I also thought about adding 2.5cm (1") to the waist at each half, but then I decided to try for a tightlacing corset.

I traced the new panels on tracing paper and laid them over the original pattern so you can see how much the pieces changed, especially panel a. A lot of the roundness was taken away. 

 

Second Mock-up

This time I used a stronger, less stretchy fabric and also added boning to the mock-up to get as close as possible to the final corset.  The shape around the top was now a lot better, with most of the excess width taken out. Also the top edge was sitting just below midbust level and cupped the underside of my bust quite well and comfortable. There still as a little bit too much width in panel a, but it was easy to tweak 1cm (3/8") out around the side bust area.

 

I also decided to cut the CF a little bit lower to give the top a more heart shaped line. This would then need a shorter busk (28cm (11")), and luckily I had one in stock. With the boning in, another problem showed up. The front was reaching very low, so sitting might become difficult. I thought about cutting it higher, but then decided against it and just go with the corset as Savoye had intended in his patent. ("...extends [...] from the sternum to the pubian region when the corset it placed on the wearer." Line 34 to 36) I didn't plan on making a piece for actual wear, so sitting would not be a main concern in this corset.

 

The seam between panels b and c was quite curved and the bones twisted a bit in their case. I was thinking about using spiral for them, but as Savoye had only used whale bone and flatsteels ( "...whalebones h, and a steelplate i are arranged..." line 43), I stuck with them in hopes that it might help with the Edwardian posture. I added another few millimetres to the lower seams of the back panels as well as around the rib area above the waist, and smoothed out the top edge.  

Also with having enough room in the lacing gap I could make the CB panels 1cm (3/8") wider each, as with their original 2.5cm (1") they were too narrow to fit in two bones and the grommets in between. With 3.5cm (1 3/8") there would be enough space to build lacing panels.

In Savoye's description "f designates a narrow strip provided eyes secured to the rear vertical edges of the panels a e to receive the lace g." (line 40 to 43) Maybe it held only one bone, or narrower bones, maybe even no bones at all, as there are none visible in the sketch, so that 2.5cm (1") were wide enough. In my experience though, only one bone at the very CB edge doesn't provide enough stability in the back.

Satisfied with all alterations made to the pattern, it was time to think about a design. I had this lovely atlantic blue linen left over from making a dress for a friend, and it was calling to become a historically-inspired corset. Combined with crème coloured embroidery, the corset would get a nice Wedgwood-inspired look. I liked that idea very much, so I went to work.

 

Making the Corset

Materials used:

• 1m (1¼ yard) atlantic blue linen (including bias tape)
• 1m (1¼ yard) white fusible woven cotton interfacing
• 70cm (¾ yard) white coutil
• 70cm (¾ yard) white cotton lining
• 5m (5½ yard) 7mm (¼") flatsteel
• 50cm (9/16 yard) 11mm (½") flatsteel
• 28cm (11") busk
• 80cm (7/8 yard) 1.5cm (5/8") wide cotton twill tape
• 50cm (9/16 yard) 2cm (¾") wide cotton twill tape
• 2.4m (2 2/3 yard) 3cm (1 3/16") wide cotton twill tape
• 24 5mm (7/32") grommets
• 4.5m (5 yard) cotton corset lace
• 10m (11 yard) cream white viscose embroidery floss

Before I cut out the pieces, I had to design the embroidery pattern. I wanted to stick very close to the sketch, so I made up a very simple vine with small leaves. 

I had interfaced the linen with cotton woven fusible to get more stability, even enough to add the embroidery before flatlining the pieces to coutil. This would prove Savoye's intention in the patent "that it is easy to produce on the said piece, preferable before proceeding to assemble the pieces, a continuous ornamentation..." (line 59 to 61) to be true. Although of course this would also be the case after flatlining, yet I thought embroidering without coutil would be easier for my hands.

I cut all pieces from linen, coutil and cotton for lining. The CB panels (f) I cut on fold to include the back facing. For the front facing I cut the vertical part of panel a from linen and the horizontal part from cotton. All seams got an allowance of 1cm (3/8"). Only the CF got 2cm (¾"), so I was able to add a front modesty panel. The a panels from linen I only marked with allowances, but I left lots of spare fabric around to allow me to fix them into the embroidery hoop.

To transfer the embroidery design onto the linen, I traced the drawing with a soft pencil. As the back of the linen was interfaced in white, the pencil was easy to see after tracing the pattern again with a harder pencil from the other side of the paper. To get the markings to the linen side I traced the pattern with a knitting needle with white carbon paper underneath the fabric. To improve visibility I went over the white marks with a white water soluble pencil.

For embroidery I used artificial silk floss, but it turned out to not have been a very good choice as the threads separated and tangled while working with them. Next time I'd use either real embroidery silk or cotton floss.

For the stem I used the full 4-thread floss, while for the leaves and smaller twigs I separated it and used only two threads. In the end I used up two packages of 5m (5½ yard) each for the whole embroidery. 

 

While still working on the embroidery, I started assembling the remaining pieces. First I flatlined the linen panels onto the coutil to build the shell or fashion layer.  Then I joined the lining panels and panels b-e of the shell. For both layers I pressed the seams, graded the seam allowances and top stitched them. On the lining I only top stitched at the seam between panel a and the vertical pieces as the cotton stayed in place otherwise.

 

After having finished the embroidery I also flatlined panel a and attached it to the other panels.  To be able to sew the tight curves I had to clip the seam allowances. I usually avoid clipping seams before sewing them, but as the coutil was way too stiff I had no choice and tried to clip as little as possible, just enough to stretch them around the curves. 

 

With panel a finally finished by grading, pressing and top stitching, I joined the lacing panels to the shell the same way. I would attach it to the lining only later after the busk and boning channels were sewn in. 

Before setting in the busk I basted on the waist tape. In this corset I used a 1.5cm (5/8") wide cotton twill tape as the waist section is rather short and a wider tape would have been harder to put in.   I always start setting in the busk with the hook side, I find it easier to adjust the position of the stud side after the hook side is fixed. I basted around the busk piece before sewing it in using the zipper foot.  As squeezing through the stud never works without any ripped threads (even with an awl) I secured the holes with a bit of Fray Check after the studs were in. I also included a small modesty panel on this side. 

 

With the busk secured, I could mark the placement of the boning channels. I was lucky that Savoye had included the boning in his sketch. He referred his way of boning to his former patent No. 774,537 from November 8, 1904, "in which the whalebones designed to stiffen the waist of the corset do not extend upward above the lower edge of the horizontal branch of the piece a, only a small number of whalebones extending up to the upper egde to bear up said piece and keep it stretched." (line 50 to 55)

In the 1904 patent he claimed that "by the use of short whalebones limited to the waist part a great economy is realized in the manufacture, greater lightness is obtained, and the corset is rendered much more comfortable, because the upper and lower parts of he corset are kept pliable and the intermediate and whaleboned part cannot penetrate into the skin, while being at the same time sufficiently stiff to properly maintain the upper part of the body." (patent no. 774,537 line 21 to 31)

I made the boning according to the sketch as double boning with 7mm (¼") steels and an 11mm (½") bone for the "steelplate i". Only the very first bone next to the CF had to be single, as there was no space for double boning left. 

 

I marked the top and bottom of the boning channels and then went on to attach the lining to the back facing. This done and the corset turned, I could proceed to stitch the CB boning channels.

 

After inserting the bones, I stitched the channels closed where necessary. I did it by hand, so it wouldn't show on the outside.  I stitched the top and bottom closed, trimmed the edges down and attached the selfmade bias binding, handfinishing it on the inside. 

 

The last steps to finish the corset were setting in the grommets and adding the lace. For the 5mm grommets I punched in 3mm (1/8") holes and widened them with an awl. After pushing the eyelet through I added some Fray Check and laid the washer on top.

I usually use the Prym Vario Pliers to fix the grommets, but as my hands are really weak the edge of the eyelet that is turned over the washer ends up rather rough. So I go over all grommets again using the tool and a hammer. Hitting it a few times will provide a smooth surface. 

 

 

The Finished Corset

When I put on the finished corset for the first time, I was very pleased that it indeed had an Edwardian look. And although the embroidery ended up not completely symmetrical, I really liked how the dainty shape wraps around the top. I got a good impression of how decoration can be included, and how different colours and patterned fabrics can create eye-catching effects.

It did feel quite comfortable, especially around the top, where the underside of my bust was nicely cupped and enclosed. There's no push-up effect like in many Victorian semibust corsets, but I guess as the original corset was intended to be worn with some form of padding anyway, this wasn't necessary - and it adds to the comfort. In addition there are no seams that could cause rubbing on the ribs, which can be an issue for slender women sometimes.

 

So Savoye didn't make any false promises when he claimed that the "corsets made according to my invention are not only well adapted to receive a harmonious and varied decoration, but they also afford a stylish appearance and great comfort owning to the continuity of the fabric all along the top and front of the corset." (line 96 to 102)

On the down side, I didn't like that the bottom front was still standing away from the body. Even after taking away quite a bit of fabric it didn't mold to the body. Also my hopes that using flatsteels as front boning would help pressing the pelvis back into an Edwardian shape, weren't fulfilled. Maybe more rigid boning or using an underbusk would have helped.

Still, next time I'd probably use spiral still in those curved boning channels to prevent the slight twisting, too.

The hip area could have used some light tweaking here and there, but overall it looked well. For piece a I would maybe try to distribute the width a bit differently, taking out a bit of space around the front and adding it in the back area. On the pictures you can see the top pulling towards the back a bit. On the dressform, with a fuller bust, it looked better. I could also try some padding to solve this issue. 

Otherwise I was a bit disappointed that the corset didn't hold the Edwardian shape I had hoped for. I can't tell how comfortable it is to wear with the pelvis actively bent backwards to get the typical Edwardian look. I tried to pre-bend the bones a bit, but this didn't help very much. [See Marion McNealy's article on the S-Bend in context for more discussion on whether or not Edwardian corsets really did "force" the figure into the fashionable posture.]

It was difficult to sit down in this corset. As the bones don't reach the very bottom, except for the first bones next to the busk, the lower egde folds up over the bottom of the bones. This might cause a lot of wear on these spots as well as creases and wrinkles. Also I can't say that sitting was very comfortable this corset. So for actual wear I'd probably cut the corset higher at the bottom front.

 

Conclusion

I liked working on this corset very much. The pattern Savoye provided with his patent did indeed give new possibilities for style and decoration. The continuing front and top piece is a nice addition to the usual vertical piecing.

I learned a lot about Edwardian corset patterns and how to get the typical look. It took a bit time and work to get the top piece to fit on an actual modern body, but once you've got this, style and design possibilities are endless. If you wrap your head around the pattern a bit, it's not as complicated as it might seem on the first look.

The corset creates a nice and smooth shape and might even be suitable for tightlacing. The pattern is not overly complicated to adjust, and assembling the panels works well. I will definitely use this pattern in the future again and try out different styles, because this design really is something special.

 

Sources

United States Patent 845,582 (Application Oct. 18, 1905; patented Feb. 26, 1907)

United States Patent 774,537 (Application Jan. 25, 1904; patented Nov. 8, 1904)

Robert Doyle, Waisted Efforts

add