564 
ME. W. K. PAEKEE ON THE STEUCTTJEE AND DEVELOPMENT 
But a broad tract of cartilage still divides the two sides of the occipital ring, both 
above and below (Plate 24. fig. 6, and Plate 27. fig. 6, s.o., b.o.). 
The ear-masses now stretch outwards as in the Frog and Lizard, and their divergent 
growths are extended into drooping arms by the addition of the transverse suspensoria, 
now forming even more than a right angle with the axis of the skull. 
The elegant form of the canals of the labyrinth can be well seen in the solid bone 
that has been fashioned as a mould over them (Plate 27. figs. 6 & 7, pro ., e.o.). They 
are, however, partly hidden by the parietals and squamosals (p., sq.). The bulging, 
bony floor of the vestibule forms a widely crescentic bulla, and in the notch the 
fenestra ovalis contains a small lenticular stapes, the centre only of which is ossified. 
The spiracular ligament fastens the stapes to the back of the top of the suspensorium 
(Plate 24. fig. 6, st., st.l.). 
Further forwards there is an unossified alisphenoidal tract between the fifth and optic 
nerves (Plate 24. fig. 6, and Plate 27. fig. 7. 5, 2), but the lateral walls are ossified from 
the optic foramen up to a short distance behind the inner nostrils (i.n.). The rest of 
the endoskeleton is unossified ; the cornua trabeculse lessen considerably after that the 
vomers have fairly taken their form and pattern ; the internasal cartilage ( i.n.c .) is seen 
below, with its rounded notch in front of a mimetic notch on the fore margin of the 
parasphenoid ( pa.s .). Outside (Plate 27. fig. 7, e.n .), the external nostrils are seen to be 
surrounded by the nasal roof-cartilage. 
The roof of the skull is now exquisitely imbricated — all the dense, polished, well- 
shaped bones overlapping from before backward. 
The premaxillary margin, arcuate instead of transverse, passes into the lateral sweep 
of the long maxillaries (compare Plate 27. figs. 2 & 6, px ., mx.). The palatal portion 
of these foremost bones is but little developed, only showing a headland for the close- 
set teeth. The nasal processes of the premaxillaries (Plate 27. fig. 6, px.) are thus 
visible below (Plate 24. fig. 6). They are well approximated, leaving a chink w'here 
the median passage was, and they are large, flat splints, and prefigure, very accurately, 
those of a bird. The maxillaries (mx.) have a narrow palatine edge within the teeth. 
They have a good face-plate (Plate 27. fig. 7), on which rest; the nasal and ecto-ethmoid 
bones (n., e.eth.). As in the Frog, a small rhinal ossification, the septo-m axillary (s.mx.), 
rests upon that part of the maxillary which is articulated to the dentary angle of the 
premaxillary. The zygomatic process now rivals in height and length that of a 
high type, such as a Lizard or a Bird ; and, exceptionally in this type, it is surmounted 
at its end by a small seed-like bone, manifestly the jugal (j.). There is no quadrato- 
jugal as in the Frog ; but the Urodeles are not of the same stock as the Batrachia. In 
their ascent, however, they often run almost parallel with them. The nasals (n.) are 
thin shells of bone, with a concave margin towards the nostril, a straight one towards 
the nasal processes of the premaxillaries, and a ragged, wedge-like hinder margin, which 
overlaps the frontals, as their outer edge does the ecto-ethmoids (e.eth.). These latter 
bones are now much larger than in the Siredon form ; they rest upon, and are somewhat 
