Now that I've got a pattern, I've got to figure out what materials to use. Marsh's patent specifically calls for leather for the busk, back and side pieces, with some sort of fabric for the rest. His preference is for kid leather, but he allows for “other fine leather dressed to the same condition”.

My initial reaction was to laugh and call the man nutters. The only experience I've had so far with kid leather is thin, brittle antique gloves. Surely such delicate material isn't suitable for corsetry?

Much to my surprise, a little research showed that kid leather is considered quite strong for its thickness in relation to cow and deer hide. The gloves I've handled in the past were brittle due to age, and the leather was skived to be much thinner than the original hide. The fact that such thin material lasted so long is actually a testament to how tough it is! A brand new kid hide at its original thickness should be quite strong indeed. Unfortunately, it's also a bit out of my price range. At $100+ a hide, I decided to test out other leathers in my stash instead.

The thin sheep and pigskins I had on hand all proved to be far too soft and stretchy. Veg tanned leather was too thick and tough, as were most types of hides intended for non-garment use. The garment leathers I tried were too flimsy to withstand the stress of corseting. Luckily, deep in a box of mystery leather samples I found a couple of large swatches of deep red that were fairly thin and very tough. I believe it was originally intended for bags or other non-garment uses. I've no idea what it is (cow, maybe?) but I'm going to use it!

Since I'm stash-busting so far, I pulled out some domestic coutil that was dyed red ages ago. Unfortunately the dye job has shifted in places over time, but I'm hoping the mottled effect paired with the leather will give me a nice steampunk feel. Rounding out the materials list is a slightly chipped black busk rescued from an old corset, as well as a variety of spiral and flat steels.

 

Stress Testing

Marsh claims that the life of a corset can be extended by using leather in place of fabric in areas prone to wear. I'm not so sure about this, so I'm going to test out the durability of leather vs. fabric. I don't have time to construct and wear a trial corset until a part fails, so I'll make small sample pieces and test them separately. Despite color differences, all leather and fabric samples are of the same type and quality that I'll be using for the final corset.

Friction test - Since Marsh has placed leather at the center back edges, I want to see how both materials stand up to the friction of laces rubbing against them. In the period, corsets would have also been subject to wear around the middle from waistbands, so this seems like an important point to test.

 

I stretched pieces of coutil and leather over the corner of a wooden box. Since my coutil sample is white, I scribbled a bit of red pencil over it so in the hopes that it makes any damage to the fabric more apparent. I then rubbed a piece of standard black cotton corset lacing over the coutil sample for two minutes. The black from the lacing rubbed off onto the fabric a bit, but didn't seem to harm the material at all. When I tried the same test on the leather sample, damage was visible almost immediately. After just ten seconds, the top layer of the hide had rubbed off. After the full two minutes had passed, a sizable area was worn at least halfway through the leather. Strike one for Marsh's idea!

 

Grommet test - While we're on the subject of the center back piece, what about grommets? Modern tightlacers complain of poorly set grommets popping out and Marsh specifically states that the leather will securely hold them without danger of wearing through. To test this, I created two lacing strips complete with bones and grommets; one of coutil and one of leather. Right away I find one potential issue: my grommets don't quite reach through both thicknesses of leather! Even so, I manage to hammer on the washers well enough that the everything is set tight. If I was to do this again, I'd probably source the type of grommets with extra-long shanks.

 

Now the fun part! I attempted several methods of destruction, including twisting, bending and on the strips as hard as I can. I pulled so hard that the lacing broke, but not one of the grommets budged. I upped the ante by stabbing an awl through the center of a grommet and twisting it forcefully every which-way in an attempt to leverage it out. Both the fabric and leather sides survived ten minutes of this abuse without issue. I event tried to pry them out using a flat headed screwdriver without much luck! This test appears to be more or less a draw; both materials performed equally well, though the fact that leather needs longer shanked grommets puts it slightly under the fabric.

 

Stretch test - A sturdy, non-stretch material is the foundation of a good corset. However, leather has a tendency to stretch. Marsh doesn't mention this at all in his patent, but I'd like to see how it stacks up to traditional materials.

 

I drew lines at one inch intervals for six inches on each sample, then subjected both to stretching. Afterwards, I measured the six inch sections to see how much they grew. As expected, the leather stretched the most (almost an inch!) but I was annoyed to find that my domestic coutil stretched about three eighths of an inch. Just for fun, I threw in a sample of higher quality spot broche coutil to see how it would stack up. Predictably, it showed next to no stretch. Strike two for leather!

 

The test I didn't do - Later I realized I should have tested which material best resisted the ends of steel bones wearing through. I suspect that leather would have outperformed fabric in this area, though the use of special bone casing might even the odds.

 

Joining Leather to Fabric

While I'm not convinced that leather is quite the solution Marsh makes it out to be, I still want to make his corset as described. However, I'm not sure what the best method of joining the fabric and leather sections, so I tried out a few seam variations to find what worked best for me.

My first attempt was a standard seam, where the two materials are placed right sides together, sewn, then opened right sides out and the seam allowance pressed open. This turned out to be the worst method, as the leather resisted folding, and there was only one row of stitching holding everything together.

 

The second attempt fared much better. I sewed a standard seam as above, but instead of pressing the seams open, I pressed it to one side and topstitched through all layers. There is no need to fold the leather back, and it is stronger than the single row of stitching above.

 

A third option is to overlap the seams with the leather on top and topstitch through all layers. Again, the leather lies flat, and I can use two spaced rows of stitching for strength. The cut edge of the leather will show on the outside, but since leather doesn't ravel or fray, this isn't an issue.

 

Both the second and third methods seem like equally good options, but I decided that for comfort's sake I'd prefer the cut edge of the leather on the exterior of the corset, rather than on the inside where it might rub against the body and cause irritation. I'll use lapped seams wherever fabric joins to leather.

 

Sewing Leather

I know that sewing leather on a home machine with a standard foot often doesn't work very well, but my machine has done very well with garment leathers in the past when I use a walking foot attachment. Unfortunately, that doesn't appear to be the case this time. When sewing samples I noticed that it struggled and had trouble feeding the fabric through. In the example below, you can see that some stitches are tiny, some are huge, and others are skipped entirely.

 

Luckily, I have access to an industrial compound walking foot machine at work. Not only does the walking foot move in tandem with the feed dogs to move the leather through evenly, the needle also moves. Check out the slow-mo video to see sewing through leather like it was butter!

 

Assembling the Corset

As with my mockup, I save time and my sanity by laser cutting all of my pieces. The edge of laser cut leather is darker than when cut by hand, but I prefer the dark edge.

It seems best to sew all the fabric pieces together first, then attach them to the leather pieces last. I begin by inserting the bust gussets into piece 1 using Laura Loft's method. They look pretty wonky at first, but once I give them a good pressing over my ham they look great! To make them a little sturdier and add a decorative touch, I topstitch all around the outside of each gusset. Later I went back and reinforced the points with a little zig-zag topstitching as well.

 

Next up is fitting the hip gores to their respective panels. This proves pretty tricky, as it requires fitting a convex curve into a concave one. I've found that the best way to pin these together without everything ending in tears is to lay the piece with the convex curve over a rounded surface, then place the piece with the concave curve on top and align the edges. The two opposing curves will fit together much more easily than if you tried it on a flat surface, or with the pieces reversed. If they're extremely stubborn, you might also want to clip your concave seam allowance before stitching.

 

I press the seam to the side away from the gore, then topstitch around the outside to match the gussets. With the gores and gussets done, I can now join panels 1 and 2. Again, I press the seam to one side and topstitch for strength. Now all the fabric parts are assembled, and it's on to the leather!

 

I want a boning channel to run down the center of the side leather piece, and I decided it's easier to add that before attaching all the panels. To make things easier on myself, I'm using purchased bone casing. A bit of masking tape holds the casing in place while I sew it down from the other side. While I can't see the casing from the right side, I can feel the edges of it through the leather; this allows me to sew in the correct spot.

 

Now I start joining the fabric to the leather panels. Since I can't pin into the leather for fear of leaving holes, I draw in my stitching line in pencil on both pieces. By holding the leather up a few inches out from the foot so I can see the underside, I can align the penciled stitch lines as I sew.

 

The busk piece gives me some trouble. The leather doesn't want to fold easily, so I have to hammer it down with a rubber mallet to get it to lay somewhat flat. I forgot to laser cut the holes for the busk hooks to fit through, and my hand cut ones look messy.

 

It's also quite difficult to stitch closely to the busk with the walking foot, so I have to settle for topstitching along the very edge of the leather. Towards the bottom the piece is so skinny that I'm stitching almost right up against the busk, but at the top there's enough space that the busk can move back and make the hooks partially hidden inside the leather. Luckily the center back leather piece is more well behaved, and now I've got all the pieces assembled!

 

Since Marsh doesn't mention where the bones go or how many there are, I settle for placing them vertically and relatively evenly spaced. Flat white steels provide stability at the center back, while spiral steels throughout the rest of the corset allow for fabulous curves.

 

Grommeting and binding go smoothly, and I'm almost done! Mindful of the friction test earlier, I choose double-faced satin ribbon instead of shoelace style laces at the back.

 

Final corset

WOW. Not once during this process did I think that this corset would turn out so curvy and well fitting! I take back any quibbles I may have had with Marsh's pattern. There's some minor wrinkling to the side of the bust gussets, and I could bring the hip in just a touch at the front, but I do believe this is my new favorite corset!

 

Due to the leather, I don't think that I'd be happy wearing this as underwear; it's strictly an outerwear corset for me. I'm still not convinced of Marsh's claims that the leather will wear better than the fabric, but time will tell!

 

add