BUFFALO SOCIETY OF NATURAL SCIENCES 17 
bone, which was already firmly connected with the cranium; a 
trasformation the last stages of which can still be seen in recent 
fishes. 
The skull of Eusthenopteron seems to me to show conclusively 
that Dr. Sagemehl was wrong in this assumption, at least in so far 
as the Crossopterygii are concerned, and that rather the opposite 
was the case is the more easily imagined when we recall that the 
para-basisphenoid in Permian ganoids and in modern snakes is a 
compound bone resulting from the fusion of at least three 
elements. (Henry Day, 1915). 
The metapterygoid (Plate 8, fig. 2; Plate 9, fig. 4) is a long 
subtriangular bone bearing on its thickened upper margin two 
processes, one of which may have articulated with the median 
bone just described, while the other suggests that it articulated 
with the side of the skull more or less after the fashion of an 
amphistylic shark. (Goodrich, 1909, p. 190, fig. 159). It is 
never in contact with its fellow of the opposite side. It oc- 
cupies a nearly vertical position and is welded by a very long 
suture to the entopterygoid. ‘This is a thin sheet of bone, concave 
outwardly and terminating in front in a spatulate process which 
rests upon the expanded prevomers, while behind, it lies against 
the inner side of the quadrate. It agrees in a general way with 
Watson’s description of the pterygoidal element in Glyptopomus 
and Rhizodopsis. The bone is nearly vertical in its upper and 
posterior portions, but suddenly curves outward in a nearly hori- 
zontal flange to meet the ectopterygoid and palatine to which it is 
firmly united by suture. On its lower surface it is covered with 
minute denticles. On Plate 8 it may be seen in dorsal and lateral 
views together with the metapterygoid, ectopterygoid and palatine. 
It is also shown on Plate 9, fig. 4. 
The ectopterygoid is a long quadrangular and robust bone, bear- 
ing on its oral margin an inner row of laniary teeth with a row of 
smaller teeth outside of them. Fig. 3 on Plate 7 shows a cross- 
section through this bone and its teeth and on Plate 17, figs. 2, 3 
and 4, I have figured cross-sections of various teeth to show the 
labyrinthine infolding of the walls. It becomes more and more 
complicated toward the base. 
In front of the ectopterygoid is the palatine, a similar bone but 
shorter. It also bears two or three tusks and is connected by a 
finger-like process to the prevomer. There is a foramen (shown 
