The _De Partibus Animalium_ becomes in form a comparative organography, but the emphasis is always on function and community of function. Thus he treats of bone, "fish-spine," and cartilage together (_De Partibus_, ii., 9, 655^a), because they have the same function, though he says elsewhere that they are only analogous structures (ii., 8, 653^b). In the same connection he describes also the supporting tissues of Invertebrates--the hard exoskeleton of Crustacea and Insects, the shell of Testacea, the "bone" of _Sepia_ (ii., 8, 654^a). Aristotle took much more interest in analogies, in organs of similar function, than in homologies. He did recognise the existence of homologies, but rather _malgre lui_, because the facts forced it upon him. His only excursion into the realm of "transcendental anatomy" is his comparison of a Cephalopod to a doubled-up Vertebrate whose legs have become adherent to its head, whose alimentary canal has doubled upon itself in such a way as to bring the anus near the mouth (_De Partibus_, iv., 9, 684^b). It is clear, however, that Aristotle did not seek to establish by this comparison any true homologies of parts, but merely analogies, thus avoiding the error into which Meyranx and Laurencet fell more than two thousand years later in their paper communicated to the Academie des Sciences, which formed the starting-point of the famous controversy between Cuvier and E. Geoffroy St Hilaire (see Chap. V., below). Moreover, Aristotle did not so much compare a Cephalopod with a doubled-up Vertebrate as contrast Cephalopods (and also Testacea) with all other animals. Other animals have their organs in a straight line; Cephalopods and Testacea alone show this peculiar doubling up of the body.