628 SLGIQMES, JA OW STAI OVEIN TES 
zygomatic arch, which is formed by the molar and the squamosal 
bones, and in the separation of those vacuities from each other 
by a long, narrow, parietal crest. The orbit of the eye, how- 
ever, is separated from the zygomatic vacuity by a postfrontal 
bone, so that the structure is distinct from that which obtains 
in the ‘mammals.’ 
“There are two well-defined occipital condyles at the back 
of the base of the skull, united to each other inferiorly in a way 
that is closely paralleled in some mammals. The hard palate 
formed by the maxillary and palatine plates terminates trans- 
versely in the middle length of the molar teeth in a way that is 
remarkably like the dental condition of certain Marsupial mam- 
mals. 
“So far as is known, there is no fundamental difference in 
the skeleton to separate the Gomphodontia from the Cynodontia, 
which may be regarded as related inthe same way as are groups 
of the Marsupials with similarly differing dentition.” 
The group Gomphodontia, as the author of the suborder says, 
does not obliterate the interval between the mammals and the 
reptiles, but it does close up the gap to a large extent, and we 
are well able to see what must have been the few final steps 
that led to the formation of the Mammala. If the quadrate 
should entirely disappear and the teeth take on ever so little a 
difference in form, if the bones of the squamosal region should 
coalesce, as they have almost done already, there would be no 
place where we could say this is reptile and this is mammal. 
The known genera of this suborder are: Z7itylodon, Triracho- 
don, Diademodon, Gomphognathus and Microgomphodon. 
Tritylodon was originally described as a mammal by Owen. 
It is known only from the imperfect upper portion of the skull. 
The skull was long and rather narrow; the anterior nares were 
joined in one external opening ; there were no incisor teeth, but 
a large pair of canines; the molar teeth were broad and sup- 
plied with three rows of tubercles; between the canines and the 
molars there was a long diastema. The specimen is from the 
Permian of Basutu-land in South Africa. From the same region 
