794 REPORTS ON THE STATE OF SCIENCE.—1913. 
retract their neck in the same special way as the Cryptodira and 
belonged to this division of the Testudinata. 
As the Cryptodira possess a typical thecophorous shell, inherited 
from the Amphichelydia, this shell must also have been present in the 
ancestors of the Leatherback. This conclusion is confirmed by the fact 
that parts of this shell, in a reduced state, are still found in the latter 
animal: the nuchal and the plastron. These elements represent a 
deeper layer of dermal ossifications, a thecal layer, whilst the mosaic 
shell represents a more superficial, epithecal * layer. 
However astonishing in the light of the utterly different type of 
shell the assertion of Baur ** may seem, that Dermochelys is closely 
allied to the Cheloniide, the conclusion just reached, that the Leather- 
back belongs to the Cryptodira, like the Chelontide, makes this view 
less improbable. 
There certainly is in many respects a considerable resemblance in 
the skeleton of the Leatherback and the Cheloniide ; but as both these 
types are modified in adaptation to marine life, a large amount of 
parallel modification and of convergence seems possible; the modifica- 
tion of the legs into paddles, a short, not retractile, neck, and a reduced 
plastron can certainly not be considered as proof of relationship. And, 
apart from the entirely different construction of the shell, there are 
other important differences, pointing rather to a more distant relation- 
ship—for instance, the strongly developed first rib and the absence of 
descending plates of the parietals to the pterygoids in Dermochelys,*® 
and the different type of the sphenoid rostrum.’ Yet the investigations 
of Baur, Nick, and Volker show several resemblances in structure, 
where convergence seems more or less improbable, and which cannot 
either be interpreted as very primitive characters, that might have been 
inherited independently from Prochelonia. 
The most important points of common structure in Dermochelys, 
the Cheloniide and related turtles, are the following: 
1. The cervical vertebral column is not only of the cryptodirous 
type, but it shows the same modification in two details. The articular 
surface uniting the centra of the sixth and seventh vertebre is plane, 
and the strong spinous process of the eighth cervical forms a jointed 
articulation with the nuchal. Both these modifications lessen the 
flexibility of the hinder part of the neck, probably adaptations con- 
nected with the swimming life of marine turtles; but they are of such 
a special nature that it does not seem probable that this adaptation 
was acquired independently. 
2. The pelvis shows a cartilaginous ischio-pubic bridge; the ischia 
are very small, the pubica large. The resemblance in the pelvis of 
Dermochelys and the common turtles is not very striking, and might 
be due simply to parallel adaptation connected with the use of the 
hind legs as paddles. There exists, however, a striking resemblance 
5 These layers are called subdermal and dermal respectively by Hay (1898; 
1908, p. 17); as the deeper layer, however, is also formed in the cutis and is not sub- 
dermal in position, the terms thecal and epithecal are proposed (Volker, 1913, p. 475 ; 
especially Menger) ; compare Wieland, 1912, p. 299. 
5 1889 a, p. 180-191; 1889 6 ; 1890, p. 533. 
& Boulenger, 1888. 7 Versluys, 1909. 
