DEVELOPMENT OF THE SKULL IN THE BATRACHIA. 
237 
enlarging in the cerato-hyal region ( c.hy .), and then becoming narrow again, without 
a lobe, as the retral hypo-hyal ( h.Jiy .). The basal plate is very similar to that of the 
last kind, but the thyro-hyals ( t.hy .) are still longer, and more divergent ; they are 
strongly curved upwards and outwards. The investing bones, also, are very similar in 
the two species, allowance being made for the difference of outline of the skull. The 
fronto-parietals (fp.) are better developed in the occipital and temporal regions, and 
as in the last, the parietal territory is anchylosed, and even somewhat adherent, to the 
bone below. They end in the frontal region so as to leave an emargination which 
slightly exposes the fontanelle (/o.); the right frontal is the longest, and yet it only 
touches the nasal of that side, the shorter left bone overlaps the nasal in front of it. 
These latter bones (n.) are very unsymmetrical; they are half their own width apart 
at the nearest point, and leave a large narrow-waisted bony tract bare, between them. 
Each bone has a thick rib, or boss, over its descending or facial process (figs. 6, 8, n.), 
and the whole facial edge (fig. 8) lies well down on the top of the maxillary fix.) ; 
it is gently concave there, and notched further forwards for the nasal aperture (e.n.). 
Of these convex, rounded bones, the left is one-fourth larger than the right, and is 
altogether further back. The premaxillaries (px.) lie completely under the cochleate 
snout; they form a more definite angle at this juncture than in the last kind; the 
nasal process is longer, and the labials ( u.V-.u.l 2 .) are larger. The maxillary (fig. 8, 
rnx.) is deeper, the short, unciform quadrate-jugal (q.j.) is a little more attached to 
the end of the maxillary ; it is but little united with the quadrate. The squamosal 
(sq.) is more normal, and the shaft is bent on the upper part at a right angle ; at that 
part the temporal region (fig. 6) is a large lozenge of bone, lying over the tegmen 
and horizontal canal; it is three times as wide as in the last kind. The postorbital 
process is very small. The rostral part of the parasphenoid (fig. 7, pa.s. ) is larger 
relatively to the wide wings; there is a definite hinder, triangular process, and a 
triangular apophysis looking in the other direction between the fore part of the wings. 
The whole bone, instead of being nearly as long as the skull, as in Dactylethm, is only 
two-fiftlis its length. Here, the left vomer (v .) is larger than the right; the two pro¬ 
tect the widely separated inner nostrils ( i.n .); and the nasal palate is extremely like 
the “hard palate” of Man ; all its elements are confluent. 
70. Phryniscus Icevis. —Adult female; If- inch long. Ecuador. 
This species, as far as my specimens show, is the largest of the three. 
This is a slightly longer skull than that of P. cruciger, the breadth and the length 
of it being equal. On the whole, it is as asymmetrical as the other two; it has a 
broader snout, and is less ossified; it agrees in several things with P. cruciger more 
than with P. varius, but like the latter species it has no columella, and no “ annulus 
its less intense ossification makes it a key to the difficulties in the other two. In one 
respect it is the most generalised of all the skulls of the Phaneroglossal types 
examined by me as yet, e.g.. the jugal arch is incomplete, as in the Aglossa. 
