of two longitudinal folds extending above the line of insertion of the fore-and hind-limbs which have already made their appearance. Hence the carapace grows outward and over the limb-girdles which come to lie within the rib-like osseous supports. This ontogeny shows us clearly how the ancestral carapace may have been formed. Paleontology has not as yet confirmed this, but doubtless will do so.
Among birds one of the most interesting features is the occurrence of vestigial tooth papillae in the jaws of certain embryo parrots and owls—reminiscent of Mesozoic days when birds were toothed in their adult state.
Mammalian evidence is very striking in many details and much of it has recently been summarized by Hubrecht, who makes much of the character of the placentation and derives from it and other features some remarkable conclusions. Hubrecht abandons the idea of the derivation of the mammalia from the reptilian-insectivorean stem, but on the contrary derives them from amphibia-like animals of the Carboniferous. The character of the placentation, moreover, places man, the Anthropomorphæ and the hedgehog among the most archaic of living mammalian types, an idea also borne out by comparative anatomy and one which paleontology may some day confirm.
The most primitive mammals, the Prototheria, are still very suggestive of their old reptilian ancestry in many ways, especially in the method of producing the young inclosed within an eggshell. The position of the Marsupials is surely low in the scale of mammalian life, although they show in many respects remarkable specializations. Wilder compares them with the Prototheria in that they also bring forth their young at a very early state of development, though not protected by an eggshell. The period during which they are permanently attached to the nipples within the pouch is actually post-embryonic and properly larval. Vestiges of placentation have been found in at least one marsupial, a fact which gives color to the belief that in this respect they may be degenerate rather than primitive. Broom has shown that the modern marsupial shoulder girdle passes through a prototherian stage implying a relationship which is strongly supported in other ways.
The fœtal Sirenia and Cetacea, so far as known, show no greater development of hind-limbs than do the post-natal individuals. They do show a marked neck construction and the head bent at right angles with the trunk in a normal quadrupedal posture. The head of the fcetal manatee is very suggestive of the modern ungulate. Ryder has tried to prove the homology of the caudal flukes in the Sirenia and Cetacea with the integument of the hind feet, drawing his evidence largely from comparison with the seals. In the embryo the flukes appear as lateral expansions near the end of the tail, giving it a lancelike form when viewed from above. These flaps by transverse expan-
