726 
MR, T. W. BRIDGE ON THE OSTEOLOGY OF POLYODON FOLIUM. 
segment. On the other hand, nearly all Ganoids in possessing mento-Meckelian 
ossicles, and the Ganoid Polyodon in possessing an ‘ ‘ orbitar process,” agree with most 
adult or embryo Anura. In Polypterus among Ganoids, in all Anura, and in not 
a few Urodela, there is a tendency towards the development of a more or less exten- 
sive interorbital ossification comparable to the “ os en ceinture ” of the Frog. But 
perhaps the most important of all the characters by which the Ganoid-Amphibian 
stem is distinguished from the Elasmobranch stem is the presence of a coecal diverti- 
culum from the fore part of the alimentary canal, which in its varying degrees of 
development forms the functional lung of the Amphibian, or the swim-bladder of 
the Ganoid — a structure which, though partially respiratory, is mainly mechanical in, 
function. 
Finally may be mentioned the existence of such transitional forms as the Dipnoi, 
which in so many features are connecting links between the Amphibia and the more 
generalised Crossopterygian Ganoids. It may now, perhaps, be permissible to assume 
that the primitive ancestors of the Ganoids and Amphibians were differentiated from 
the Elasmobranch stock prior to the evolution of their two main groups of descendants. 
The fundamental similarities between the main divisions of the Amphibia lead us 
to believe that they have been derived from some one ancestral form. That the 
divergence of the Anura and Urodela took place at a very early period is evident 
not only from the differences observable when their adult structures are compared, 
but also from the extreme dissimilarity of their larval forms. Even if we admit that 
differences which now exist between the larvae are the result of adaptive modification 
in one or other of them subsequent to the evolution of the two types, there will still 
remain those existing between the adults which are scarcely less important. 
As far as our deficient knowledge of the development of the cranium in the Urodela 
will admit of any attempt to make general statements, it would appear that the very 
characteristic metamorphosis of the mandibular arch of the young Tadpole which 
results in the attachment of the forwardly inclined mandibular pier to the prefrontal 
region by a vertically disposed palato-pterygoid bar, and in the development of an 
“ orbitar ” outgrowth from the quadrate, while the non-functional Meckelian cartilages 
are superseded in their use as jaws by the labial cartilages, never occurs in any of 
the caudate Amphibia. In the latter also there are no labial cartilages, and even 
when horny jaw sheaths are present, as in the adult Siren, they are not supported 
by labial ■ cartilages, as in the Tadpoles of the Anura, but by the dentary and 
premaxillary bones. But the wide difference between the two types is most decisively 
shown when such Perennibranchiate forms as Proteus and Menobranchus are compared 
with any of the Anura. 
In Menobranchus* the suspensorium retains its primitive position at right angles 
to the cranial axis, and the palatine and rudimentary pterygoid processes are widely 
sundered ; the trabeculae retain their embryonic condition as simple nearly parallel 
* Huxley, Proc. Zool. Soc., 1874. 
