13 
in Eryops and Cricotus, gives the following result: The glenoid cavity is 
an excavation in two codssified elements, of which the inferior and posterior 
is probably coracoid. The latter is then much smaller than in Reptilia and 
Batrachia anura, but resembles that of the salamanders. The scapular 
arch proper resembles that of the Urodela. The pelvis is intermediate 
between that of the anurous and urodelous Batrachia. There is no obtura- 
tor foramen, and the common symphysis is deep. The humerus closely 
resembles that of the Pelycoswurta, differing chiefly in the non- -enclosurg 
of the supracondylar foramen; and as in that sub- order, some genera 
possess condyles and some do not. 
Prof. Owen proposed the order Ganocephala chiefly for Avahercsaur Us, 
but he included in it also the genera Denderpeton and Pelion (Paleontol- 
ogy, p. 182-3). This division has not been generally adopted, the genera 
mentioned being usually placed in the Labyrinthodontia. Of the eleven 
characters given by Prof. Owen in evidence of the existence of this order, 
one only does not belong also to the Labyrinthodontiu ; this is the absence 
of occipital condyles. On this account I thought that the group should be 
retained, but not as an order. Besides this group and the Labyrintho- 
dontia, there were the types called Microsuuria by Dawson, ‘some of 
which have simple enamel, all agreeing in general characters, and differ- 
ing from other Batrachia. I therefore combined the three groups into 
one order, the Stegocephalt. (Proceedings, Academy, Philada., 1868, p. 
209.) This order was most distinctly characterized in the Report of the 
Geological Survey of Ohio, Paleontology, ii, p. 854, 1875. 
Von Meyer has given us enough of the characters of Archegosaurus 
to enable me to refer the forms of the Texan Permian to the same order. 
Prof. Owen, in his discussion of the affinities of that genus (1. c., p. 170), 
remarks, that the vertebrae and numerous very short ribs, with the ‘‘indi- 
cations of stunted swimming limbs, impressed me with the conviction of the 
near alliance of the Archegosuurus with the Proteus and other perenni- 
branchiate reptiles.’ As it is now well known that perennibranchiate 
batrachians belong to three different orders of the class (7'rachystomata, 
Proteida and Urodela), the above expressions lose point, and especially as 
the characters mentioned as indicative of affinity are of the most subordi- 
nate importance, or as in the structure of the vertebra. are totally distinct 
from what is found in those orders. When we read later (p. 173), that the 
fact that the superior ‘‘ossifications of the skull have started from centres’ 
more numerous than those of the true vertebral system, gives the charac- 
ter of the present extinct order of Butrachia ;’’ we find that Prof. Owen 
has quite failed to perceive either the definitions or affinities of his new 
order. He commits an error in describing a distinct pubic bone; an ele- 
ment which Von Meyer states (Paleontographica, vi, 179, 1858) that he 
had not discovered. Von Meyer describes the codéssified inferior elements 
of the pelvis as ischia. My numerous Texan specimens show that each of 
these bones includes both pubis and ischium. 
In now defining the Ganocephala anew, T confine myself to characters 
