Press "Enter" to skip to content

How do you Reproduce the Alcega Farthingale Pattern?

How Accurate is the Drawing?

This is a very good question to be asking. The measurements seem to be far more accurate than the specifics of the picture, but the measurements take some decoding. Start with this: Alcega’s directions state that the narrow portion of the pattern is cut on fabric folded in half along its length, while the wider portion is the remainder of the farbic, folded in half along its width.  The wide bit should be twice the measure of the narrow bit, basically.

The narrow bit is 7/8″ wide. The wide bit is 1 5/8″ wide – fully 1/8″ less than twice the narrow bit, and we’re working on a reduced image. This is not encouraging.

On the other hand, the two narrow panels both measure 3 3/8″, which is good, because they’re both meant to be the same length.

The two narrow panels are labelled with letters that don’t seem to line up or relate to each other. These labels are actually indications for how long to make the lines. (This is why I insist that this is a cutting diagram with drafting instructions, rather than an actual pattern!) The lengths of the narrow panels are given at “bm”, or 1 1/2 bara, or 49 1/2″.

The top of the front panel is 3/4″ long. By our scale, that equates to an 11″ line (14.666667*.75). The top of the panel is labelled with a length of “t”, or 1/3 bara: 11″. That seems like a good sign. (But really, only “seems”, as we’ll see in a minute….)

The top of the back panel is labelled with a length of QQQ: 3/4 bara, or 24 3/4″. The length of the top measures 15/16″, or 13.75″. Not good, on two levels: first off, we’re not getting the right measurement off the picture.  Second off, the measurement given is wider than the silk the pattern was meant to be laid out on.

In f.8, Alcega tells us that “…it is important for the reader to know that all of the silks woven in these Kingdoms are two-thirds of an ell wide…” These pattern pieces were meant to be laid out on fabric that is 22″ wide, and folded in half.  That’s 11″, for those of you keeping score. How do you cut a 24 3/4″ long line on an 11″ wide fold of cloth?

Also visually troubling is the fact that both narrow panels are listed as having a hem of “sb”, or 5/6 bara, or 27 1/2″. The top line of the back panel is of “QQQ”, or 24 3/4″. And yet, the hems measure out at 7/8″ and the back top measures at 15/16″. Scaled up, 1/16″ accounts for just less than a full inch (.9166667″).

I’m going out on a limb and saying that this drawing is not proportional.

Unfortunately, that means that you can’t simply enlarge the pattern and get the right proportion to a support skirt.

If Everything is All Wrong, What’s Left?

I think the real story is in the numbers and specific clues in the layout. While the proportion is shady, there are some (surprisingly) sophisticated details of the pattern drawn out.

Now is time to take a good look at the gores. They’re not labelled with sizes. Perhaps the 24 3/4″ measurement is the full top of the back gore, or perhaps it’s the line drawn for patterning.  This would also serve as an implicit instruction: You’re going to want 24 3/4″ at the back waist, so cut what you’ve got on the large panel (22″ wide), and use the remainder (2 3/4″) as the top edge of the corresponding gores.

The  front and back hems, likewise, are both listed with the size “sb” or 27 1/2″. This is wider than the fabric available in the 22″ width, but there are gores. If we make up the difference with the gores, following the notion above, we’ll have 27 1/2 at the front hem and 27 1/2 at the back hem.

Ahem. At 55″, that’s an inconveniently small skirt. Especially given that Alcega’s description quite curtly notes that the farthingale is “…a little more than 13 hand-spans (palmos) wide, which seems to me to be sufficient for this farthingale…” (translation, pg 49). In her calculations in Patterns of Fashion (pg 7), Arnold lists a hem of approximately 117″ based on an estimated 9″ handspan. The chart on 12b of the Alcega translation notes that the measure Q, 1/4 bara, is “one hand span”. One quarter of 33″ is, in fact, 8 1/4″. I’m inclined to agree with this measurement, as it actually is my own handspan. (My next article will be on how vanity personal bias compromises research…. ;) )

For the sake of argument, this means we’re looking for a hem measurement between Arnold’s (9″ x 13 hand-spans) and the Alcega translation (8.25″ x 13 hand-spans): something between 117″ and 107 1/4″. Right now we have a 55″ hem being indicated. Unless, of course, the measurements listed are for the line to be drawn, not the finished piece. The line is being drawn on folded fabric, so the actual measurement would be double what’s listed.

That would give us a very convenient 110″ hem measurement. But that’s very convenient, and quite a jump to make without anything other than a coincidence to back it. The good news is that this does seem to be the convention throughout the book. F. 63, for example, is a skirt and bodice of cloth. The cloth is folded lengthwise, and the hem of the skirt front (which goes almost entirely across the folded panel) is listed as ‘b’ – one bara. In f.vii, Alcega states that “fine cloths” are “usually two ells wide”. The designated length ‘b’, in this case, is given for a skirt cut on the fold across nearly two full baras. Half the finished measure is listed. Other folios in which the pieces are cut on a lengthwise fold (63a, 64, 64a, etc) seem to follow the convention: the listen hem length is half that of the finished hem. In items which are not cut on the fold, such as in f.66, a skirt and jerkin of silk, things work out the same. As mentioned, silk is a 22″ wide fabric. This cuts the fabric widthwise at the middle, resulting in two panels, each still 22″ wide. (So that patterns can run the same direction.) The skirt front hem is given as ‘b’. It is cut clear across the panel, and has a side gore with a hem equal to it’s waist measurement, ‘t’. (22″ of farbic plus the 11″ t does give us the full 33″ needed.) On pieces without a fold, the measurement given indicates the hem of each individual piece. This makes sense, as the finished skirt will still require both a left and a right front, resulting in a finished measurement that is twice what is listed.

So the measurements seem pretty solid. I mentioned some sophisticated details in Alcega’s diagram. Let’s take a look at the hems:

comparison of the front and back hems
The front and back panels of the Alcega farthingale show different hem curves.

If you look closely, you’ll notice that they are visibly different even before the join with the gores. The front (conveniently on top in the photo) has a greater angle to the side gore and a greater curve at both the hem and the waist. (In fact, I followed the waist curve to mark the hem on the gore.) Assembling the gores per Arnold’s reconstruction, straight off the facsimile, results in a pattern where the critical side joins at the waist and hem are absolutely square.

The other thing that becomes obvious from the difference in hem curves is that the difference in side angles on the gores is deliberate. This is because gores with less of an angle have less of a hem curve, and gores with a greater angle have more of a hem curve. (Compare your standard gored skirt with the quarter gore of a circle skirt – the circle skirt’s sides are at 90 degree angles and its hem is a one quarter arc of a circle, while the standard gore doesn’t really look like a circle at all at the hem and has a far lesser angle to the sides. The more extreme the slant on the sides, the curvier the hem.) Why do this this way? I don’t know, but I can posit guesses: it could be because the pleats of an overskirt skirt add bulk to the back, it could be because the gathering of the back waist pulls so much of the farthingale to the back that the imbalanced gore angles are needed to compensate (so the front doesn’t lie flat along the legs), it could be to show off a forepart…. Or something else entirely. I just know this is a surprising detail to include.

It seems to me that the Alcega farthingale plan reads like this:

“Here is a farthingale layout for a farthingale that is 49 1/2″ long. Cut the main front and back panels on a piece of 22″ silk, folded lengthwise, and the remaining gores on a piece folded in half widthwise. The front and back main panels are of equal length, and the back side gore is the same length. You need to curve the hemlines of the front and back panels, as shown. Each hemline should be marked out 27.5″ on a curve, as shown (the remainder is made up in the gore). The waistline of the front should be 22″. Begin the side slant at 11″ from the fold, and angle down to the edge of the fabric. The back waistline is 24 3/4″ from the fold, curved for a gore. Square the waist to make the side line of the gore, as one does in skirts. The gores may be cut of the remaining (folded widthwise) fabric. Draft the back gore first. From near the top of the back gore, begin the front gore, connecting it with a point as far down the side as seen at the front waist. When you’re all done, the pieces should look about like this and you should be able to lay it out for cutting as shown. Please don’t hesitate to call for technical support.”
Next: Final Notes and Summary

Pages: 1 2 3 4


  1. Anna-Carin
    Anna-Carin October 13, 2010

    Maybe Arnold simply has a typo in the metric equivalent of 49 1/2 inches, saying 127.5 cm where 125.7 was intended?

    My personal impression is that metric equivalents in UK/US books are wrong as often as not… While I haven’t seen Arnold’s conversion table for Alcega, I think it’s probably just a matter of rounding errors. If 1/48 bara was exactly 11/16″ then it converts to 1.74625 cm, which was neatly rounded off to 1.75 cm for the book. The difference between 11/16″ and 9/13″ is only about 1/200″ or 1/8 mm; a quite acceptable rounding error in itself, though the rounded numbers give very confusing fractions when converting back to inches.

  2. Anna-Carin
    Anna-Carin October 14, 2010

    Actually, I was so carried away with the oddities you pointed out in your interesting article, that I forgot to also mention how much I enjoyed reading it. Sorry! :-(

    At first I thought it seemed strange to attach the flared edge of the back side gores to the back piece, while the front pieces were joined on straight grain, but afterwards it struck me that if you picture the pieces lined up side by side in the order of assembly, like Arnold does in her books, it makes perfect sense. In her 19th century examples, each side gore still has the front edge cut on grain and a flaring back edge; only the proportions of the gores are different. (BTW, she has an interesting remark on the pattern diagram facing page 28 – cutting gores so they flare towards the front instead of the back seems to have been a common mistake.)

  3. missa
    missa October 14, 2010

    Oh, oh, I wanna play! Page 28 of which book? :)
    Thanks for your comments, Anna-Carin – I agree the the numerical differences between what’s in Arnold’s work and what’s in the Alcega conversion table (provided just after the beginning commentary) is probably just a rounding thing, but it’s a rounding thing that’s always bothered me. It’s too small to make a huge difference, sure, but it doesn’t *feel* right. And when the basic numbers in a system don’t feel like the right numbers to be using, it makes me suspicious of the whole thing. It’s sort of like being served a blue egg for breakfast….

  4. Anna-Carin
    Anna-Carin October 14, 2010

    Oops – I did have a feeling I’d left something out… :–( I meant Patterns of Fashion 2.

    It would have made more sense if Arnold had stuck to the base unit of the original text, but then I guess it would’ve confused and frustrated lots of people who were less able to sort it out!

  5. Monique
    Monique November 30, 2010

    I have an issue with my farthingale. I tend to trip over my skirts and I get caught on the bottom hoop.

    Where should my length be? How far from the ground without showing my ankles should I allow?

    Thanks for ANY help you can provide.

  6. missa
    missa November 30, 2010

    Hi, Monique,
    The problem isn’t the length of your hoops. It’s that you walk like a modern western woman. We tend to have a fairly long stride, and to lift our toes as we step. If your stride is longer than radius of your farthingale, you’re going to risk stepping on it. Additionally, if you raise your toes as you step, you’re at risk for catching the hoops.

    The solution is to take smaller steps, and walk more like a small child than a model in an athletic shoe ad. You can also just shuffle along- your skirts can’t get under your feet if your feet don’t leave the ground. (I can “run” over uneven ground in a skirt that’s 6″ too long and a heel with the shuffling trick. No joke – I have witnesses. It’s also the key to successfully backing up in a trained gown. While we’re at it, it’s completely possible to walk up stairs *without* lifting your skirts from the front. There’s an easy way, and an advanced version that leaves your hands free. You can use a hand to press on your hoops from behind, thus raising the front slightly without mussing all your layers. If you don’t have a free hand, you can take a sort of ballet approach by running your toes under your hoop from the side, sliding your foot forward until you locate the stair in question. Then, draw your toes smoothly up the face of the stair and across the stair into position to take a step. If you’ve kept your toes under the edge of the hoops and skirts the whole time, you’ve just gracefully managed to get your foot on the stair without your skirts being under it. I strongly recommend practicing before taking on a full flight…. )

    Practice really is key. You need to know exactly how long of a step you can take, and where your toes will run into the hoop. (I measured, actually, then marked out the living room floor for practice. I’m sort of a geek like that.) You’re going to be safest in your costume when it’s as familiar as clothes. It’s no different to learning to walk in heels, really. You just have to adjust your movement (balance, length of stride, the angle your foot hits the pavement at) slightly. You’ll end up with a walk that’s safer, and an overall look that’s far more natural.

    That all said, I make my hoops so they’re about 1-2″ above ground level so I’m not replacing the bottom hoop casing all the time.

    I’m not sure that’s quite the answer you were looking for, but I hope it helps!

  7. Michael
    Michael August 11, 2011

    While I won’t lie, I got tripped up in all the math but that’s probably because I’m just a tad bit tired at the moment! Loved the work put into this. I also have had issues with the whole “the extra length is for creating tucks/casings for the reeds/boning” issue simply because of the pain it is to create a smooth casing on a conical surface. Now I’m tempted to tackle a scaled down version of this to see if it’s any better!

  8. […] I’ve been reading up heavily on farthingales (and many thanks to the sempstress for her in-depth mathematical analysis. If you didn’t guess from the excel charting of the […]

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.