(Chapter 16) Footnote 59:

Vesalius’ teleological digression, in the manner of Galen’s De usu partium, develops Galen’s comparison of the curved spine to a vault (4.78.10 a)/kanqan oi=(on yali/da). This metaphor reinforces an error of Galen that Vesalius did not discover. The human spine, unlike that of the apes, does not resemble a simple arch in its normal erect posture, but assumes a triple curve (convex ventral curvature in the cervical and lumbar regions and dorsal curvature in the thoracic) which is an adaptation to the erect posture of homo sapiens. This triple curavature was well known long before Galen (see, for example, the Hippocratic De articulis 45.24ff.). Though in chapter 39 Vesalius recommends bending the metal rod placed inside the vertebrae of an articulated skeleton to follow the natural lumbar and thoracic curves, his illustrations at the beginning of chapter 14 (where see note 3) and in the second skeleton at the end of book one minimize the ventral curvature of the cervical and lumbar vertebrae, while the third skeleton is stooped forward in a way that makes the spine curve in a single arch. In his instructions for the articulation of a skeleton in Chapter 39, he warns the student that “the greatest care should be taken with the iron rod that it not be ineptly bent backward and forward and hold the body upright in an unbecoming way.”

Vesalius points out in the margin that “the figure placed at the front of the volume shows various arches.” We see, in fact, two keystone arches in the title page illustration, one on each side of the picture. Architectural metaphors for anatomy in ancient sources are more than once echoed in Vesalius. Cf. the comparison of ginglymus [articulatio trochoidea] to a hinge (originally a Hippocratic metaphor) in chapter 4 and the re-use of the hinge woodcut in chapter 15 to illustrate the relation of the axis to the atlas. On metaphors linking architecture and anatomy, see Galison & Thompson 1999 and Hersey 1999.