1896.] MR. G. A. BOULENGER ON SCHLEGEL’S GAVIAL. 631 
hypapophyses on the cervical and anterior thoracic vertebre are 
less developed than in the other recent Crocodilians, and are not 
directed forwards; they are not developed beyond the eleventh 
vertebra (twelth or thirteenth in the others). The chevron-bones 
are all open dorsally. The first pair of ribs are inserted on the 
sides of the proatlanto-atlantic hypapophysis, or lower part of the 
atlas-ring, and separated from each other at the base by a wide 
interspace. The second rib differs from that of all Crocodilians I 
have hitherto examined (including the Gavial, of which I have 
examined the bone on a young specimen in spirit, and also the 
atlas and axis preserved in the Museum of the Royal College of 
Surgeons,—the Gavial-skeleton still being a desideratum in the 
British Museum Collection), Osteolemus excepted ; it is attached 
to the centrum of the atlas (odontoid bone), near its suture with 
the axis, by the capitulum only, the tuberculum being merely 
indicated by a small upward process at a distance from the base 
of the bone, and without any connection with the vertebre. 
It is well known that in Ichthyosaurus the atlas bears a forked 
rib, same as the axis and the other vertebre behind it. It seems 
that one Crocodilian at least presents an approximating feature. 
The late Mr. Hulke has first pointed out in Metriorhynchus (P. Z.S. 
1888, p. 419) the presence on the “ lateral pieces ” (neurapophyses) 
of the atlas of a tubercle situated in the level of the diapophysis on 
the epistropheus, and he concludes that this tubercle should rank 
as an upper atlantal transverse process or diapophysis. I have 
been able to verify the correctness of this statement on several 
well-preserved atlases of Metriorhynchus, still undescribed, from 
the Leeds Collection, which my colleague Mr. Andrews has kindly 
shown me in the Geological Department of the British Museum ; 
and I quite agree with Hulke that “the position of this little 
process in serial line with the upper transverse processes of the 
other cervical vertebre speaks distinctly in favour of its diapo- 
physial character.” We are, in consequence, justified in assuming 
that, although, as we know from one specimen, the first rib is not 
forked, it must have been connected with the diapophysis by 
ligament, its head being attached to the side of the hypapophysis 
(‘‘ basilar piece ”) of the atlas, or rather between the latter and the 
centrum (odontoid bone); and such a condition may be regarded 
as the most primitive known among Crocodilians, and as one from 
which, as Hulke has shown, the abnormal position of the first rib 
of recent forms may be derived and explained. 
The second rib in Metriorhynchus was attached by its capitulum 
to the anterior border of the lower surface of the centrum of the 
axis, or between the latter and the centrum of the atlas, and by its 
tuberculum to a process (diapophysis) of the neurapophysis of the 
axis. 
As regards recent Crocodilians, the information to be derived 
from books appears contradictory, principally from the fact that 
the various authors have dealt with different genera, and have in 
some cases generalized their observations to the whole group. © 
