Examining an Antique Length of Warren's "Skirtbone", Boning for the Hems of Mid-1890s Skirts!

Edited April 23, 2021

Today I have something special to add to the research on period methods of adding skirt fullness in the mid-1890s. A length of antique, unused "Skirtbone" produced by the Warren Featherbone Company of Three Elms, Michigan.

At just a quarter of an inch wide and about 1/16 inch thick (NOT 1/32" as I have it in the video), it's a springy, sproingy boning. It weighs, well, a feather, and you'd not notice any additional weight in your skirt, I believe.

To understand it, you really have to see it close up, see it move, and see the insides. It's really remarkable, and perhaps the most interesting thing about it, is that it's not made of wire, but the quills of poultry feathers, set parallel to one another. I hypothesize that the quills were woven together with black thread and probably glued in place, and then covered by interwoven black threads that are again glued or perhaps starched.

I've made a YouTube video so you can get as close as possible to experiencing the real thing.

Here's the reel that the length actually came from, below.

Warren Featherbone Company's "Skirtbone" hem boning. Photo from Annie's Antiques on Etsy.

The skirtbone in detail. The actual boning in only 1/4" wide, so the photo is quite magnified.
It's really quite small. Photo from Annie's Antiques on Etsy.

How would Skirtbone would be actually used? For that, if you haven't read it already, please see "1895 Outfit: Period Methods To Add Skirt Fullness, Part 5, Steels, Rattan, Candlewicking, and Dust Ruffles".

Skirtbone and Warren Featherbone's other products are a clever use of easily available and renewable natural materials. Back in the day, Featherbone was an alternative to whalebone, and made from materials not otherwise used, say in feather dusters. Skirtbone would have been an alternative to wire-based skirt hem products, too, which would have been subject to rust; somehow I can't see folks using stainless steel for a stiffened hem tape, do you? 

Warren's Featherbone was invented in the 1880s by Edward Kirk Warren. His application for a patent is copied in full at the bottom of this post, and it explains the construction of the first featherbone product. As Randy Miller of the Michigan newspaper The Herald-Palladium explained it,

"The process for making featherbone was published in magazines and newspapers across the country in 1883 and was described as follows:

'The first thing is to strip the feathers of their plumage. Rollers, with knives attached, split the quills in half. Sandpaper rollers revolving rapidly removed the pith. Then, a series of interlocking knives reduced the quills to fiber. In this state, the material is fed into a machine that forms it into a strong fine cord; at the same time it is being wound with thread. In another machine, four of these tightly wound cords are wound together with thread, in such a manner as to form a flat tape.'"

Here's an advertisement for it that appeared in towards the back of the August 1895 issue of The Ladies Home Journal (p. 27).

Warren's Skirt Bone advertisement, August 1895.
It appears that the product was pretty new. Other Warren's
products were already popular.

The sample of Skirtbone is still amazingly flexible, as we have seen. No, we don't know how how it stood up to sudden breakage or repetitive stress breakage. We do know that whalebone tended to become brittle, while this product isn't brittle at all, even now. That's some pretty good longevity, no? Am I going to bend it wildly or bash it to see how it takes rough treatment? Um, no. It's antique and a small but significant part of dress history. It goes into the collection.

Skirtbone wasn't the only mid-1890s skirt aid that Warren's produced. They also made "Bustle Bone", a wide bone strip that dressmakers were urged to shape into small hoops and install inside godet plaits to help hold the rounded funnel form. Here is an advertisement for it from December 1896, towards the end of the godet skirt's popularity, in The Delineator (Vol. 48, No. 6, , p. 108)

Warren's Featherbone Company "Bustle Bone" advertisement
December, 1896

For those among us who are vegetarian or vegan, the prospect of reviving the use of feathers for boning likely doesn't appeal. However, it is a nice alternative to plastic. I sure wish Warren's would consider bringing it out again. Are you listening, ladies and gentlemen of the Warren Featherbone Company? 

More About Warren's Featherbone Company and Its Products

The Warren Featherbone Company is still in business, although it's no longer in Three Elms, Michigan, but in Georgia. There's quite a bit out there about the company and its history. Here are a few good examples, and if you run a search, you'll find much more:

How turkey feathers changed Three Oaks' destiny (The Herald-Palladium;
Aug 20, 1999 Updated Nov 15, 2011)
• The Legacy of Warren Featherbone
• Exhibit - From Corsets to Philanthropy: The Warren Featherbone Company
• Information from the archives of Western Michigan University

Warren Featherbone US Patent Application

Copied from Google Patents, number US286749A. Text is optical scan from original document, available from that page as a PDF file.

(Model.)

E. K. WARREN.

CORSET STIPFENER. No. 286,749. Patented 00's. 16, 1883.

' iff/y1.

WITNBSSES www# BY uw ATTORNEYS.

NITED STATES PATENT OFFICE.

EDWARD K. WARREN, OF THREE OAKS, MICHIGAN, ASSIGNOR lOF ONE-HALF TO GEORGE R. yHOLDEN, OF MICHIGAN CITY, INDIANA.

CORSET-STIFFENER.

SPECIFICATION forming part of Letters Patent No. 286,749, dated October 16, 1883.

Application filed January 9, 1883. (No model.)

.T all whom it may concern.'

Be it known that I, EDWARD K. WARREN, of Three Oaks, in the county of Berrien and State of Michigan, have invented certain new and useful Improvements in Corset-Stiffeners, of which the following is a full, clear, and exact description.

This invention has for its object the utilization as a rib or stiffener for corsets and other articles of dress or fabrics of the stalks, stems, or quill portions of feathers after they have been stripped-as, for instance, the feathers of turkeys, geese, chickens, and other fowlsmuch of which kind of stock has heretofore had little or noK commercial value. These I propose to use, among other purposes or uses, as a substitute for whalebone in corsets, Waists, dresses, abdominal supporters, surgical appliances, and other articles of wear and use. The growing scarcity andfincreased cost of whalebone for these and other purposes has led to the employment of various substitutes, including bones, horn, rubber, steel, and rat tan; also, the fibers of tampico held together by an exterior binding-as, for instance, by a wrapping of wire or thread. Many of these are expensive and much inferior in numerous respects to whalebone. My improved bone or rib, which I term featherbone, has many advantages, and the same may be made to form a compact stay or bone, which in some respects is better and has more enduring elasticity than whalebone, as it 'is not so liable to breakv or Warp, nor will it be injured by prespiration or boiling Water in washing. Such elastic substitute for whalebone I generally propose to make by splitting or otherwise reducing, either by hand or machinery, but preferably by machinery, the quills, into splints. These splints or bers may be held together by any suitable external binding. Thus I wrap them with either wire or thread by winding, braiding, twisting, or they may be cemented or be otherwise put up together to form a featherbone of any desired shape in` its transverse section, and may either be simply inserted in the article to which they are applied or be Woven therein, or they may be woven together, so as to form an elastic or 5o iiexible fabric.

`their entire form.

Reference is to be had to the accompanying drawings, forming part of this specication, in which similar letters of reference indicate corsponding parts in all the figures.

Figure 1 represents any exterior longitudinal 55 view of a featherbone in which a number of quill splints are held together by winding a wire or thread around them. Fig. 2 is a similar view of a number of quill splints held together by a woven covering. Fig. 3 is a View 6o in perspective, upon an enlarged scale, of a portion of one of the quill splints detached; and Fig. 4 is a cross-section of Fig. l, and Fig. 5 a crosssection of Fig. 2, both being on an enlarged scale. Fig. 6 is a perspective view of 65 a stiffener made of the entire quills in a suitable wrapper.

In the drawings, b b represent a number of split-quill splints arranged longitudinallyto-` gether in any suitable manner to form afeathl 7o erbone, A, of any desired form in its trans versesection, and which maybe held together either by a wound wire or thread external binding, c, as in Figs. l and 4L, or by a woven external covering, c', as in Figs. 2 and 5, or 75 which may be otherwise held together, as hereinbefore specified.

By using the separated fibers or splints of the stripped quills or stems of feathers, produced by splitting or disintegrating them, the 8o resiliency of the quills is preserved without exposing them to breakage` in bending, to which they would be more or less liable if used in However7 the entire quills make avery good stiifener lwhen stripped and 8 5 bound together, as shown in Fig. 6 ofthe draw-- ings.

Having thus fully described my invention, I claim as new and desire to secure by Letters- Patentl 9o A corsetstiffener formed of quills or quill splints stripped of the feathers and bound together, as shown and described.

EDWARD KIRK VARREN.`

Vitnesses:

HENRY A. MOGANN, RETTA HOLLETT.

Additional Research

Extensive research on mid-1890s skirt stiffeners is contained in a series of related posts available on this blog's 1890s Costumes & Research page. Primary sources include books, magazines, newspapers, film clips, and extant garments.

Okay, I've interrupted the petticoat project. Unless something else interesting shows up, which as we know from this blog sometimes occurs, we're back to it next time.

Wishing you all health and safety during very dark days, literally and figuratively...

An 1895 Godet Petticoat With Boning and Stiffened Frills: Making Unusual Seams

Where were we? It's been a little while since I worked on this project. Oh yes, we're cutting the pattern to petticoat length and finally constructing the thing. The project has turned into yet another test of period materials and methods to see how they work. Nonetheless, if the petticoat works, it will give my cotton skirt the regulation -- to use the popular term of the time -- flare. 

This time we're cutting and seaming, but the seams are not normal ones...

Cutting the Pattern to Petticoat Length

Petticoats are usually a little shorter than outer skirts. Therefore, I subtracted 1 1/2" inches (6.35cm) off the pattern, which is 40 inches/100cm in front. That works out to 38" long when finished, but leaving a 1/2" seam allowance in place. That allowance will be needed to seam the bottom facing and a hem binding to. Hem binding on a petticoat that doesn't touch the ground? What an oddity, but that's what's suggested. 

So I copied the pattern pieces onto fresh paper (oh joy), and cut off the bottoms. This way, I have the original skirt pattern plus the petticoat pattern to go with it. 

(In case you missed any posts in this very long petticoat series, you can find them on the 1890s: Costumes, Research, and Documentation page.)

Cutting the Fabric

The next, rather obvious step was to cut out the fabric pieces.

I used silk shantung from some old silk curtains I made years ago.

The old curtains were rolled up. I used an entire panel, about 4.5 yards.
First I had to deconstruct the entire panel, removing the lining (bonus fabric!)
and the plastic curtain facing stuff at the top. Forget what it's called.
Fortunately, only bits of the silk were weak or shattered, so I could
use most of the fabric.
Oh, and yes, our hallway library doubles as a free-weight lifting station.
Pandemic creative space use :}

The silk fabric has body and resistance to wrinkles, although probably not to the degree that moreen has. The latter fabric was a favorite among dressmaking writers for silk petticoats (Davis, Mallon), as we have read. As it happened, the two larger pieces were cut on the bias, but not on the perfect bias and another was pieced, because I placed them puzzle-wise to conserve fabric.The bottom of the petticoat is to be faced, hence the 1/2 inch (1.27cm) seam allowance.

The side seams were cut with a 1/4 inch (.635cm) seam allowance. Note: I would have used all metric measures and forgone inches and feet, but the presser foot and markings on the treadle sewing machine are in inches, and they are wonderfully natural sewing guides.

The top of the petticoat has a 1/2-inch seam allowance, but the plan is to cut a yoke for the petticoat, so the extra is just in case I change my mind :}

Stayed...and Bound...Seams

Next, all the skirt pieces were seamed together, but watch out! The seams are emphatically not normal ones. Because just about every seam is on the bias and thus at the fabric's weakest position, we have to sew in a fabric stay on each one so that the fabric doesn't stretch at the seam and cause ripples and puckers and a poor skirt hang. Sophie Klug writes in the 1895 book The Art of Dressmaking, "Where two bias edges are to be joined in one seam, a stay tape or strip of lining must be basted at one side and sewed in with the seam to prevent stretching." (p. 35)

I was going to use bias tape for the job, but big thanks to Quinn of The Quintessential Clothes Pen for pointing out that using bias tape would have been bias over bias and therefore not much of a help. I looked back at documentation on an 1890s skirt in my collection, and sure enough, there's a straight-grain stay there. Why I'd forgotten such an interesting detail, haven't a clue.

Not only would the strip be a seam stay, but it would also bind the seam allowance for a neat, durable finish. Petticoats normally have nicely finished seams because they get so much wear. Twofer!

Making the Stay/Binding

So, I cut 1" strips of thin cotton (from the old muslin curtain lining fabric -- why waste it?) on the straight as long as each skirt seam. 

Then I ran the strips through a 1" bias tape maker widget (the Clover brand version; there are others).  

It's pretty easy to use. First, you start feeding the fabric strip into the channel.

Help the fabric through the channel by sticking a pin into the fabric and pulling...

...and pull at the tip to pull the fabric right out the end.

Voila! You have bias tape, or straight tape, in my case.

You will want to make sure the fabric strip is out of the way of the little wire handle.

After you start pulling the fabric through, go to your ironing board, pin a bit of the finished tape to the board, and heat your iron to steam heat level. Then slowly pull the gadget, feeding the fabric into it, and pressing the resulting tape immediately.

I found that the fabric wanted to go in wonky sometimes, so the pulling process was slow. Other times the steam from the iron was so hot that I just pushed the bias tape maker along in front of it, like so:

Sewing the Seams

To create a seam, the two skirt pieces that were to be seamed were laid, right sides together, ready to sew. On top, I laid the prepared cotton binding with its pressed edge at the fashion fabric edge and and the rest of it open and ready to fold over the top of the completed seam to bind it. All three layers were carefully pinned together, trying to lift, handle, and tug the fabric as little as possible. Every tug can stretch the bias-cut fabric.

Each seam was sewn from skirt top to bottom in case the ends should get a bit out of alignment. 

Immediately after sewing a seam, I folded over the prepared binding to wrap the seam allowance in it, turned in the already prepared fold so no raw edges would be showing, and sewed it down on the covered seam allowance to finish the seam.

Readers, I started sewing the seams with the Singer 28k handcrank machine. The 28k came out during the 1880s, so this was a logical choice. However, we all know silk can be no fun to sew by machine, and the long and slippery skirt pieces were determined to be naughty, and I only had my left hand to guide the fabric, because the right hand was cranking the machine! Here's a sample, videoed by my son, Christopher.

I even tested the antique Singer binder attachment. It worked; the attachments nearly always do -- alas that I don't have a video, because it's interesting! -- but again, trying to do all this with a hand crank on silk was annoying and error-prone. The process became so unfun that I chucked that idea, removed the first seam, and hand-sewed them all. The main seam I made in running stitch with a back stitch every five stitches or so, 6-8 stitches per inch. Then the binding was of course hemmed. I enjoy hand-sewing, so it was a relaxed effort.

Here I have pinned the two skirt pieces right sides together, and have pinned the stay/binding
to the edge. I am running-stitch the seam with a backstitch every few stitches.
Generally I pile up three or four running stitches on the needle, then pull the needle through, 
then take a back stitch. It becomes a steady, rhythmic pattern of movement.

Here is an example of a completed seam. The sewing thread blends in so well that it's practically invisible to the camera this afternoon.

The hand-sewing makes sense within the context of undergarments of the day. Hand-sewn underthings were desirable as being particularly dainty.

I wasn't the only one relaxing. Nutmeg settled in, too. Here we have evidence that some kitties will sleep on just about anything. For some unaccountable reason, she found my sewing box to be a good pillow.

Interested...

(Yes, I made sure she wasn't into anything sharp.)

Getting nappish...

Oh boy, time to go to sleeeeep...

It's January. A good time for napping. Wishing you safety and many naps this month!

Next steps with the petticoat? Facing the bottom and inserting the surprise stiffening. Yep, a mid-winter surprise for you.