1. Introduction

At the beginning of the 20th century, a new figure and a new type of corset became popular: the “S-Bend’. It features a straight front, wide hips, and buttocks pushed back from a tilted pelvis. Innovative corset patterns were created to attempt to achieve this figure.

Many corset patterns from this era are fascinating and make interesting challenges for the modern corset maker. I will focus on a specific corset from this period by Daniel Kops, patented in 1902. You can view and download the whole patent as a .pdf here. As soon as I laid eyes on this patent, I got fascinated by its unique features. I fell in love with the elegant bust lines. This article analyzes the original corset pattern and describes in detail how the pattern was adapted to my modern body. Part Two covers the construction and Part Three, the boning placement and finishing.

 

2. Analysis of the original corset pattern

2.1 Analysis of the pattern laying flat

The original patent provides an illustration of half the corset assembled and laying flat [see figure 1], and an illustration of the corset pattern [see figure 2]. This pattern can be described as having two main blocks: a “bust block” that is made of the pieces “a”, “b”, and “c”, each piece being contained into one another; and a “hip block” that is made of two under-blocks: the “front hip block” with an inverted V shape (“d”) and a gusset inside (“e”) and the “back hip block” that is a repetition of one big concave piece alternating with a small convex piece: “f”, “g”, “h”, “i”, “k”, “l”, ”m”, and “n”. And finally, there is a lacing strip at the back: “o”.

Other patterns from the same era feature a similar architecture (a “bust block” connected to a “hip block” with a long diagonal seam) such as the 1904 Savoye Corset, Patent 786,685.

Figure 1. Patent illustration of half of the corset assembled Figure 2. Patent illustration of the corset pattern

 

Based on the original illustrations, the following remarks can be made:

• The "j" piece is missing.
  None of the patent illustrations show the "j" piece. Each corset piece is numbered with a letter and those letters just go from "i" to "k". Where did the "j" piece go? The hip area depicts a pattern of one big convex piece alternating with one small concave piece. Based on this pattern, my guess is that no piece is actually missing in the hip area. The text of the patent also omits the "j" piece, it says "The hip portion of the corset [...] is composed of the sections f, g, h, i, k, l, m, and n in the order given". So, we will assume that this is just a numbering mistake, and that the "j" piece never existed and is not needed to make the pattern work. In order to avoid confusion, I will keep the original numbering of the pieces. That means that the “j” piece is still missing in my own work based on this pattern. [The letter "j" could have been omitted from the progression of letters in order to avoid confusion between letters "i" and "j" - in other words, there's no missing piece, they just didn't use letter "j" - Ed.]
• The seam allowances were initially included in the pattern illustration.
  The original corset pattern included seam allowances (if you look closely at the “f” piece both on the pattern and on the assembled corset, you will see that the shapes are definitely different). During the further developments, I had to exclude the seam allowances as it just makes pattern manipulations much easier. Removing the seam allowances can either be done by hand (after printing the pattern) or with a drawing software on a digitalized version of the pattern. I did it digitally.

Figure 3. The original pattern (in black) and the new pattern lines with the seam allowances removed (in blue)

 

• The center front seam is not straight.
  On the original pattern, the center front edge of the first panel is slightly curved [see piece “a” on figure 2]. This surprised me. Among existing corset patterns, the center front edge is usually straight and vertical as it is used to insert the split busk which itself is solid and straight. I wondered if that would be hard to sew. I am not sure if that curve is intentional or a mistake. I searched through other patterns from the same period and other periods as well. In my corset pattern books (Corsets and Crinolines by Norah Waugh and Corsets, Historical patterns & techniques by Jill Salen) I could not find any curved center front except for the “1844” corset from Corsets & Crinolines, but this corset has a closed front and no split busk so this special design eliminated the problem of sewing a straight busk on a curved seam. In the archives of the Foundations Revealed articles from the same period, I found two patented patterns that seem to feature a curved center front seam, both found in patent 845582 by Savoye. However patent illustrations need to be taken with caution as Joanne Arnett mentioned “Patent applicants did not have to prove that their inventions worked, only that they were new and unique, so [the inventor] did not have to present the Patent Office with a finished corset, just the schematics for his design”. All in all, the curved center front remains a mystery to me.

 

2.2 Building original mock-ups

The next step was building a paper and a calico mock-up without any modifications yet. I did this in order to get a better understanding of how the pattern works in 3D [see figures 4 to 9]. The paper mock-up is not scaled. The calico mock-up is scaled-up (you can read more about scaling-up a digital small diagram and then how to print it in the “Corset Pattern Scaling” article by Sandra Stuart). At this point, the corset pattern was not at all adapted to my measurements. That is why doing the “fittings” of the calico toile on a mannequin and not a real human was good enough for me.

Figure 4. Front view of the paper mock-up Figure 5. Side view of the paper mock-up Figure 6. Back view of the paper mock-up

Figure 7. Front view of the calico mock-up Figure 8. Side view of the calico mock-up Figure 9. Back view of the calico mock-up

Figure 7. Front view of the calico mock-up with outlined corset parts Figure 8. Side view of the calico mock-up with outlined corset parts Figure 9. Back view of the calico mock-up with outlined corset parts

 

Based on the two mock-ups, I identified my references lines: the “Upper edge”, "Waist", "High hip" and "Low hip" lines. I decided that the “Waist” line would sit at the top of almond shaped parts, the “High hip” line would sit at the widest part of the almond shaped parts, and the “Low hip” line would sit at the bottom of the almond shaped parts. I then measured the corresponding circumferences and the height between those reference lines. The measurements were gathered in the following table.

“Upper-edge” half circumference	38cm
“Waist” half circumference	26cm
“High hip” half circumference	36.1cm
“Low hip” half circumference	40.5cm
Height between the "upper-edge" and the “waist” lines	12.4cm
Height between the “high hip” and the “waist” lines	5.6cm
Height between the “high hip” & “low hip” lines	6.7cm

Table 1. Measurements of the original pattern scaled-up.

 

3. Adapting the pattern to a modern body

3.1 Digitizing the pattern

I digitized and scaled up the original figure with computer drawing software. I was worried that scaling the original illustration would get me really thick lines (from enlarging the original lines). It would be hard to work with, especially when considering that the almond shaped pieces are really small (about 5mm [3/16"] wide!) So in this case, I thought that re-drawing the original pieces with a vector drawing tool would be worth the work. The beauty of vector drawing is that it ensures that the contour lines do not get thicker and thicker while scaling up. You can read more about working with corset patterns on a computer in these two Foundations Revealed articles, 1905 Savoye Corset, Pt.845582 #2 and CAD: “Corsets Are Digitised”.

 

3.2 Should we build an underbust, a midbust or an overbust?

The “b” piece gets extra-large at the top. To me, it suggests that there is some 3D shaping in the breast area and that the original pattern would be a midbust. However, the original heights are very short. Actually, the really short heights puzzled me for quite a long time. To me, there are two possible explanations. One way to see this issue is to consider that the heights are too short and the other way is to consider that the circumferences are rather big (compared to the heights). This suggests that this could be a pattern for over-sized ladies (although the text of the pattern doesn’t mention anything about this). Or the pattern provided in the patent could be more a technical illustration of a concept that is not meant to be used as a real pattern in its original state. I decided to make an underbust out of the original pattern.

 

3.3 Deciding what modifications were needed to match my desired measurements

I compared the measurements of the original pattern (scaled-up) with the measurements of a modern body (mine!). From this comparison, I deducted what modifications were needed to be made.

• The original “upper edge line” was much lower than needed if I wanted to get an underbust corset that sits right under the bust. So I needed to extend all the bust parts. Since a big portion of the parts was “missing”, I had to work mostly from scratch - but I kept in mind that the “upper edge” circumference and the height between the “waist line” and “upper edge line” needed to match the measurements I wanted.

Figure 10. Extending the bust pieces

 

• The original "high hip" line needed to be lowered so that the distance between the “waist line” and the "high hip" line increases from 5.6 to 8cm. In order to achieve this, the hip pieces (between the waist & the "high hip" lines) were stretched lengthwise.

Figure 11. Stretching (the upper part of)
one of the hip piece

 

• The original "low hip" line needed to be raised so that the distance between the "high hip" line and the "low hip" line goes from 6.7 to 4cm. In order to achieve this, the hip pieces (between the "high hip" & "low hip" lines) were shrunk lengthwise.

Figure 12. Shrinking (the lower part of) one of the hip piece

 

• The "high hip" circumference needs to be increased to go from 36.1 to 40.1cm. In order to achieve this, the almond shaped pieces ("f", "h", "k" and "m" pieces) were stretched width-wise. This follows the inventor's intentions. In the patent, Kops states that: "I do not limit myself to the precise shape of the plait-pieces f, h, k, and m, as they may be made wider at the central portion, so that the plaits are fuller and increase the fullness over and back of the hips". It convinced me that I was on the right track!     

Figure 13. Widening one of the hip piece

 

The result of the modifications of the hip parts can be seen in figure 14.

Figure 14. The original pattern (in black) and the modifications (in red)

 

I also straightened the busk edge of the center front panel as I didn't want to deal with curves in that area.

 

3.4 Truing the seams

The last step was truing the seams. That means checking that both edges of each seam have the same length. Measuring the length of a seam can either be done by hand with a flexicurve or with a computer software. I did it with a drawing software. Truing the seams in Victorian corset patterns is rather easy as one has only to check the vertical length of each panel (two panels lengths are checked and compared to ensure that they are the same). But in our case, it gets a bit trickier. To true the long diagonal seam connecting the two main blocks, I needed to make the sum of the length of the seam of several small pieces (see figure 15).

Figure 15. Checking the length of the long diagonal seam
connecting the “Bust block” area to the “Hip Block”

 

The result of all the modifications made can be seen in figure 16.

Figure 16. The original pattern (in black) and
the modifications (in red)

 

3.5 Testing the modified pattern

At this point, I was ready to test my pattern.  Since the pattern is rather unusual, and since a lot of modifications were needed and the construction of this corset requires much time and care (I am looking at you, pointy gussets! And you too, numerous tiny hip pieces!), I decided that it was safer to build another mock-up. This would ensure that my pattern was correct before spending more time in the final corset construction. Since I only wanted to check the pattern, I spent as little time as possible on the construction of this second mock-up: the gussets are not done carefully, the corset is not fully boned, etc. You can see the fitting of this second mock-up in pictures 17, 18, 19, 20 and 21.

Figure 17. Front view of the second mock-up Figure 18. Tree quarter view of the second mock-up Figure 19. Side view of the second mock-up

Figure 17. Front view of the second mock-up with outlined corset parts Figure 18. Tree quarter view of the second mock-up with outlined corset parts Figure 19. Side view of the second mock-up with outlined corset parts

Figure 20. Back tree quarter view of the second mock-up Figure 21. Back view of the second mock-up

Figure 20. Back tree quarter view of the second mock-up with outlined corset parts Figure 21. Back view of the second mock-up with outlined corset parts

 

The rib cage is pretty conical and the front is definitely straight, with not much space for the front ribs to expand. The hips are very round, especially in the back. This fitting tells me that the modified pattern is pretty close to being correct. The only thing that bothers me is the bottom front - “V” shaped edges tend to poke out. In that case, I want to make the “V” shape less pointy and also make the front hip a little less wide at the bottom. I will have to redraw slightly the pieces “d” and “e”.

 

Conclusion

This article described how I adapted a 1902 pattern by Kops to my measurements.

Does this pattern work? It is often said that the more seams a corset has, the more control you have on the shape. This experiment shows that the 1902 Kops Corset Patent 712,312 has enough seams to get a beautiful silhouette. It should be noted that the almond shaped pieces of the hip area provide a lot of control on the waist / hip ratio. However the long diagonal pieces in the “bust block” are trickier to work with. In their original form, they give more a conical ribcage shape and the fashionable straight front. I am not sure that they are meant to be transformed to get an accentuated hour glass torso or a cupped ribcage shape, if that was needed.

In this article, I modified the original pattern using mostly measurements, a calculator and the original pattern laying flat on a sheet of paper. Other methods could be used to work with this pattern such as “draping on the stand” or “drawing on an existing mock-up and then tracing a paper pattern from the cut-up pieces” which is explained in details in the “Y&N Diagonal seamed corset” article by Katarina Vukšic. 

In this article, I explored an underbust version of the pattern, but you can see a modern overbust interpretation in my 2014 competition entry. I had a lot of fun playing with this pattern!

You can also read about another patent by the same inventor Kops in a series of articles written by Jenni Hampshire on the "Hearts" patent #832390.

 

add