No. 384.] CHELONIAN CARAPACE AND PLASTRON. 941 
costal plate, the half of the plate next in front, and the half of 
the plate next behind. The neural scutes are similarly disposed, 
covering sometimes wholes or parts of from two to four neurals. 
The marginal scutes are only as long as the marginal bones 
which they cover; but, instead of coinciding with the latter, 
they “ break joints ” with them. Neither do the plastral scutes 
coincide with the bones of the plastron. It is evident that the 
scutes have had a development wholly independent of the bones 
beneath them. How has this occurred ? 
The skin of the adult Dermochelys is wholly devoid of divi- 
sion into areas resembling scales or scutes ; but in the young, a 
fine specimen of which I have been permitted to examine in 
the National Museum, the skin is everywhere, on body and 
limbs, broken up into small polygonal areas. Along the dorsal 
and ventral keels these areas are considerably larger than else- 
where. It is quite certain that these areas coincide with the 
osteodermal plates which are, or .will be, developed in the skin. 
When the bony plates have increased in size, the overlying 
scute has become correspondingly extended. 
I conclude, therefore, that the earliest turtles were covered 
with numerous small horny scales, possibly overlapping like 
those of lizards; and that in the dermis beneath these scales 
there were produced osteodermal plates. From such an ances- 
tor, land-inhabiting, and having limbs fitted for such a life, 
there arose a race that has culminated in our leather-back 
turtle. This race early betook itself to an aquatic life, and its 
limbs suffered profound modifications. Possibly also the epi- 
dermal structures and the underlying bony plates became more 
or less modified. Quite certainly the deeper carapace and 
plastron underwent considerable reduction. The nuchal bone, 
however, remains to the present day. 
From the same primitive ancestors that gave birth to this 
athecate tribe there arose another vigorous race, whose mem- 
bers tarried longer on land. In the members of this branch 
of the Testudines the elements of the more deeply developed 
shield were probably present, but in a somewhat rudimentary 
state. To such an animal, with probably a broad and inflexible 
body, slow of movement, and with few defenses, it would have 
