MR. T. W. BRIDGE ON THE OSTEOLOGY OF POLYODON FOLIUM. 
693 
series to the end of the rostrum, and, as is the case with the medio-dorsal elements, 
they are separated from one another by the interposition and mutual apposition of the 
two pairs of splints (r 3 , v ?> ) which serially repeat the vomers along the ventral surface 
of the rostrum. Hence, just as the elements of the medio-dorsal and supero -lateral 
series — the supraoccipital and dermo- ethmoid, the parietals and frontals — are serially 
repeated along the dorsal surface of the rostrum, so there is a similar serial repetition 
of the parasplienoid and vomers along the ventral surface. 
There is also on each side a ventral series of rayed marginal ossicles (m.o.) which 
overlap the vomerine series. The rays of these ossicles are connected with those of 
the star-shaped spicules that support the ventral layer of the flexible coriaceous 
margin of the rostrum. In the interspace between the cartilage of the rostrum and 
the upper and lower layers of its flexible margin I noticed two large nerves running 
forwards to the extremity of the rostrum ; these nerves are probably the rostral 
branches of the first and second divisions of the Trigeminus. 
The Chondrocranium. (Plates 56 and 57, figs. 5, 6, and 7.) 
When stripped of its investing parostoses the cranium presents the appearance 
represented in Plate 56, fig. 5. 
Its most striking feature is the enormously elongated prenasal rostrum. The skull 
with its rostral prolongation is about one- third of the entire length of the axial 
skeleton, the rostrum alone being one-fourth of the length of body, and at least twice 
as long as the rest of the chondrocranium. 
About eight or nine of the anterior vertebrae have so completely coalesced with 
each other and with the skull, that the only indications of the primitive segmentation 
into distinct vertebrae is the serial arrangement of the obliquely disposed pairs of 
foramina for the exit of the anterior and' posterior roots of the spinal nerves, the 
anterior root leaving the spinal column below and a little in front of the posterior root. 
The posterior limit of the cranium seems to be approximately indicated by a promi- 
nent ridge-like elevation of the lateral walls of the cartilaginous mass formed by the 
coalescence of the occipital region with the adjacent vertebrae. This ridge (Plate 57, 
fig. 7, a), which is mainly an outgrowth from the cartilage of the two anterior ver- 
tebrae, first becomes evident beneath the Vagus foramen, and then passes obliquely 
upwards and backwards between the foramina for the exit of the anterior and posterior 
roots of the first spinal nerve to terminate behind the postero- external angle of the 
skull in a free, decurrent, wing-like process (fig. 7). The vertical lamina of the post- 
temporal bone is applied to the outer side of this process, and its anterior margin 
just overlaps the oblique ridge (fig. 3). In front of the ridge the occipital region 
is slightly constricted, and behind it the bulging notochord and paired foramina for 
the spinal nerves become visible. When the chondrocranium is viewed from above 
(fig. 5) the ear capsules are seen to form conspicuous lateral outgrowths terminating 
