Annals of the Transvaal Museum. 
95 
for the maxilla to be moved about the prefrontal through the transverse 
bone, the reduction of the hinder maxilla would proceed apace, as this 
function is best performed with the transverse bone directly attached to 
the fang-bearing portion. 
On the other hand, the origin of vipers from an opisthoglyphous type 
would demand more profound changes. Firstly, the Opisthoglypha have 
simple grooved teeth, and tubular teeth are quite unknown in the group. 
Secondly, and this is more important, the fang-bearing portion of the 
maxilla is situated far behind the prefrontal bone, and whilst it is difficult 
to see how the evolution of the viperine character can have commenced 
when these two structures were not even in contact, there does not appear 
to be any tendency in the Opisthoglypha for a forward movement of the 
fang-bearing portion. Nor is the tendency to get rid of the solid teeth so 
pronounced as in the Proteroglypha, and in no case have they entirely 
disappeared. It will therefore be necessary for the advocates of this theory 
to show how the posteriorly situated fangs have been moved forward up 
to the prefrontal bone, and how subsequently the anterior portion of the 
maxilla has lost its teeth and has been tilted or otherwise changed into the 
upright maxilla of a viper. But in any case, the evidence of allied forms 
points to the conclusion that in vipers a vertically elongated maxilla is of 
secondary origin. The maxilla of Bitis has a short and slender backwardly 
directed process situated immediately above the attachment of the transverse 
bone ; this no doubt represents the original articulating surface with the 
prefrontal, a connection which became unnecessary as the maxilla 
gradually crept upwards and found a new articulating surface on the 
anterior side of the same bone, thereby giving a greater freedom of 
movement to the maxilla. 
It is not possible for me to deal with the mutual relationships of these 
several groups from other points of view, but it may be mentioned that 
the various published accounts of snake venoms seem to show that there 
is more in common between the Proteroglypha and the vipers than between 
the Opisthoglypha and either of the former groups. And, again, the vipers 
and proteroglyphs agree together in the possession of hypapophyses 
throughout the vertebral column, whereas in the great majority of 
opisthoglyphs these processes are absent from the posterior body vertebrae, 
though in a few genera — but all with the primitive type of maxilla — these 
posterior hypapophyses are present. 
The facts of distribution could probably be explained in terms of 
either theory : both Yiperidae and Proteroglypha are comparatively recent 
in origin, as they are absent from Madagascar, and the vipers at anyrate 
were not able to make use of the Brazil-West Africa connection of the 
early tertiary period, whilst the absence of vipers from Australia may be 
best explained on the assumption that they are still more recent than 
Proteroglypha. 
