least, if the fossil be not referable to Sagenodus , it belongs to Ceratodus ; and, 
judging from the geological age of the specimen, the latter alternative is 
most improbable. It may be added that the roof of a skull of Ctenodus has 
already been discovered in the Carboniferous of Victoria, Australia. 1 
Sagenodus laticeps, sp. non. 
Plate II, Figs. 1, 2. 
Obs. — Apart from a few scales, the type specimen (PI. II, Figs. 1, 1 a) 
is the only example of this species in the collection. 
Sp. Char. — A stout species probably attaining a length of about a 
metre. Each palatine dental plate consisting of four very divergent ridges ; 
and this pair of plates meeting in the middle line at an unusually wide angle. 
General Form. — The type specimen is distorted in front where the 
bead is turned round to the right side ; and the greater portion of the caudal 
region is missing. Beneath and behind the bead there are also scales of a 
Palaeoniscid and pieces of Eiasmobranch cartilage mingled with the fossil. 
It is thus impossible to determine the exact proportions of the fish. As 
suggested by remains of the palate, however, the bead must have been 
comparatively short and broad ; while the body seems to have been shorter 
and deeper than that of the European species in which the shape of the fish 
is known. 
Read and Dentition. — Unfortunately the top of the head is not 
exposed, and the only distinct fragment of the skull is part of the palate 
seen from the oral aspect (PI. II, Fig. la). The parasphenoid ( pas .) exhibits 
the usual great expansion in front, hut its exact boundaries are uncertain, 
and its sutural connection with the palatines is obscured. The palatine 
dental plates {pi.) are shown in horizontal section, and proved to be in direct 
contact in the median line. They meet at a much wider angle than is usual 
in the Dipnoans, and each dental plate cousists of four well-separated, acute 
ridges, which are too incompletely preserved to show whether or not they 
were denticulated. 
Vertebral Axis . — The elements of the axial skeleton of the trunk, and 
the fin-supports were evidently only superficially ossified or calcified, so that 
where broken in the fossil, they exhibit a central core of matrix or calcite. 
There are no traces of vertebrae, so the notochord was doubtless persistent. 
Ctenodus breviceps, A. S. Woodward, Mem. National Mus. Melbourne, No. 1 (1906), p. 15, fig. 3. 
