On the Construction of Women's Gowns:

When we make up a set of skirts and a bodice (or doublet) for period re-enactment, we commonly treat them as two separate garments. The bodice is made and finished at the bottom, and the skirts are made up and mounted on a waistband. The ensemble is then worn with the bodice covering (hopefully) the waistband. The two pieces might be held together by means of a series of hooks and eyes between the waistband and the bodice lining, or perhaps tied with laces through eyelets worked in both the bodice and waistband. But is this truly a documentable, period way of making up and wearing a skirt and bodice?

There are certainly sources that say so. Many commercial patterns contain instructions for making up skirts with a waistband, as does at least one book on the subject (the perennially popular, "Elizabethan Costuming" by Winter & Savoy). Looking at older works, or works with significantly more research backing their theories, however, there's scant evidence of it.

In, "A History of Costume" (Kohler, 1928 translation), Kohler states, "From the end of the fifteenth century onward the long, voluminous dress worn by women was no longer made in one piece. It now consisted of a bodice and skirt made separately." (pg 227) Two pages later, Kohler goes on to mention the skirts being sewn to the bodice: "The skirt when sewn to the bodice..." (pg 229). It is entirely probable that the first quote refers to the cutting of the fabric pieces for the skirt and bodice separately, comparing this to the dresses of preceding eras, where the panels for the body and skirts were cut as one.

"Patterns of Fashion" (Arnold, 1985) provides more direct evidence that gowns were constructed with the skirts sewn directly to the bodice. Of the five examples studied by Arnold that contain either a skirt or petticoat, there's no evidence of an original waistband. The gown of Eleanora De Toledo shows stitching holes at the waist of the bodice and portions of the top of the skirt. In addition, the side back openings of the bodice are echoed in the side back slits of the skirt. Arnold's own notes also indicate where the bodice and skirts were attached. The gown of Grafin Katherina Zur Lippe is slightly less conclusive, as no stitch holes remain in the bodice. However, Arnold's notes indicate an exact match between the gathered down size of the skirt pieces and the bottom edge of the bodice, and no waistband piece or partial piece is indicated in her notes. Arnold's notes do indicate where the side seam of the skirt met the bodice. The gown of Pfalzgrafin Dorothea Sabina Von Neuburg, again, shows a pleated size to the skirt top that exactly matches the bottom edge of the bodice. The bodice has a considerable drop to the waist in front, and the top of the skirt is turned down to match it. A skirt attached to a separate waistband, most commonly, is not shaped to the drop of the bodice at the front, as the bodice can easily cover the top of the skirt. Again, no waistband is present. This gown also includes fragments of a petticoat. No waistband is present and there is no indication of how the petticoat was supported or worn. The gown of Pfalzgrafin Dorothea Maria Von Sulzburg was, per Arnold's notes, made up with the skirt attached to the bodice: "The skirt was originally gathered into small cartridge pleats and attached to the bodice, from the evidence of the crease marks." (pg 115) The only waistband noted is on the last example of a surviving skirt or petticoat, listed as a petticoat from roughly 1615-20. In this case, Arnold's investigation indicated that "The pleated edge is bound with black satin ribbon, folded lengthwise, probably dating from the 1880s." (pg 117). Arnold also notes that this piece has been repleated, and may have been used as a domino in the nineteenth century.

Additionally, in the wardrobe warrants reprinted in "Queen Elizabeth's Wardrobe Unlock'd" (Arnold, 1988), skirts and bodies are never entered as separate items, but as complete gowns. In fact, the headings in the Stowe inventory, include separate entries for doublets, petticoats and foreparts, but skirts and bodices were always catalogued in sets, either as round gowns, french gowns, or simply gowns.

Beyond the burden of historical evidence favoring gown construction where the skirts are sewn directly to the bodice, I would also like to offer up a few observations from time spent making and wearing costume. When a costume is constructed with a bodice and separate skirt mounted on a waistband, the bodice and skirts simply do not want to stay smoothly together - bending, dancing, even strolling will cause the bodice to shift slightly, leaving the waistband at least partially visible if the bodice does not have tabs, skirting, or some other detail at the bottom edge. A bodice with a more than the slightest drop at the center front will also want to bunch at the waistline, due to the bulk of the skirts. (Unless the corset is worn over the skirts. That is another argument for another day, but I do not believe it to be the proper way to wear a corset.) Adding significant boning to the front of the bodice will control this, however, boning is only noted on one example of a bodice or doublet in Patterns of Fashion: The woman's doublet, dated 1585, contains significant boning at the front. Arnold supposes that these are to control the shape of the bust (pg 107). I note, though, that there is a single piece of boning shown at the side, running on what would be a true vertical when the doublet is worn, directly below the armscye. Vertical boning beneath the armscye is usually added to control wrinkling at the side of a garment. Sewing the skirts directly to the bodice also, in my experience, also solves the problem of wrinkling at the side of the bodice. The added weight of the skirts pulls the bodice taut, removing the need for boning beneath the armscye, or along the front waist drop.

Based on the evidence of surviving garments as detailed by Janet Arnold, the inference from the Stowe Inventories that the bodice and skirts were commonly treated as a single item, and reproducible results from my own experiences with trying to recreate and wear gowns in the period fashion, it seems entirely likely that skirts were normally sewn directly to bodices. The waistband, while in use on men's garments, does not appear to have been used for women's. Sewing the skirts directly to the bodice is not only a more period construction technique, it results in gown that is far easier to wear.