Based on the terminology used in the devis, the rakes should be the horizontal distances from the ends of the design keel, points C and B, to vertical lines dropped from the outside top corners of the stem and sternpost—points F and M' in Figure 21. Both the stem and the sternpost are sufficiently preserved to indicate that the rakes to the outside of La Belle's endposts were greater than those given in the devis. Reconstructing in detail how the stem and sternpost were originally drawn gives a very plausible explanation for the apparent discrepancies with the devis that does not simply discard them as mistakes. I believe these measurements are for the rakes to the rabbets—groves cut into the stem and sternpost for fitting the planking.

Stem

The procedure for studying the design of La Belle's stem is representative of the approach used throughout the hull's reconstruction. Since the use of circular arcs predominated in the design methods of the time, all archaeologically preserved curves were analyzed to determine whether they could have been drawn with one or more arcs of a circle. There are several ways to proceed with this analysis that are independent of determining the historical methods used to find the centers of such arcs. The most common one uses the geometric theorem that a perpendicular bisector of a chord passes through the center of the circle (Figure 24). This work requires great accuracy and was done at a large scale, 1:10, numerous times. In addition to this method, a transparent template with concentric circles drawn with radii spaced at half a French foot was superimposed on the archaeological recordings to help identify component arcs. For La Belle's reconstruction, none of the center points found using such methods were considered valid unless a plausible sequence of steps could be presented that would have allowed the original designer to locate the same points. Such "plausible design sequences" were derived from historical methods described in various shipbuilding treatises.

Geometrical analysis of the remains of La Belle's stem indicated that two separate center points, G and H in Figure 25, were used to draw the curves of the outboard face and rabbet. However, the radii for these two arcs are identical and they have a very specific relationship to the 12-ft height of the stem listed in the devis. Point E in Figure 25 is located on a 12-ft perpendicular line raised at point C', 4 ft 6 inches beyond the end of the keel. These 4 ft 6 inches are the rake measurement, and the 12-ft stem height is measured from the bottom of the keel (Dassié 1695:73; Duhamel 1758:165–166; MnM 1686:PH11393, 11409–11410). The shipwright located point F away from point E by the desired timber dimension for the stem beyond the rabbet. In Duhamel du Monceau's treatise from the mid-eighteenth century, where the length is given between rabbets, he confirms that a necessary step in the sequence of drawing the stem is to establish how much the stem projects beyond the rabbet (Duhamel 1758:164). NEXT