OF THE MARSIPOBRANCH FISHES. 
4 L5 
The subocular bar should be compared with that of a Tadpole during transformation, 
when the hinge of the mandible is below the emerging trigeminal nerve (see Phil. 
Trans., 1881, Plate 4, figs. 8, 9,) that is when the quadrate condyle has left the front 
of the head and has retreated to the postorbital region. In the adult Lamprey we 
have the pedicle (pel.) growing forwards as the pterygoid cartilage (pg.) ; these two 
regions form the hinder, and half the lower, boundary of the subocular fenestra (s.o.f .); 
the rest of the lower boundary is the postpalatine region (pt.pa.), and the front 
boundary the ethmo-palatine ( e.pa .), from which grows the prepalatine spur (pr.pa.). 
But there is no quadrate condyle, that is completely suppressed; in the Myxinoids 
(Plate 9, fig. 2 ; and Plate 16, fig. 1, q.) there is a rudimentary quadrate tract, but 
without any condyle ; in these it is still further back than it would be, if developed, 
in the Lamprey. The slight elbowing of the subocular bar under the hind margin of 
the fenestra is the only sign of it in this Fish (Plate 18, fig. 1). That which, for some 
time, kept me from seeing this matter fairly was a comparison of the Lamprey’s skull 
with that of a Tadpole before transformation ; I then mistook the prepalatine spike 
for a rudimentary quadrate tract. Now if these parts in the Lamprey and the 
Myxinoids be compared with those of the transforming Tadpoles of the species of 
Rana (op. cit., Plate 4), we shall see that it is the subservience of the free mandible 
(Meckelian rods) in the Tadpole to the suctorial function—they are mere carriers or 
supporters of the incomplete “annulus”—that makes the necessity for the forward 
position of the condyle of the quadrate. Indeed, nothing in morphology is more 
marvellous than the behaviour of those condyles, which, lying at first in the front of 
the head, gradually swing themselves round, and, in the Bull-frogs, come to project 
some distance behind it (op. cit., Plate 8). 
Here, in the Lamprey, the perfect annulus, or suctorial disk, is functionally free 
from cranio-facial trammels, and, being so huge, causes the once short, backwardly- 
placed lower lip to project in front of all the other structure; in the Tadpole (see in 
the huge Pseudis, op. cit., Plate 1) it is almost as forward as in the Lamprey; but in 
the other kinds it is more under the face. 
The forward position of the large quadrate condyles in the Tadpoles of the “Anura 
Phaneroglossa” masks the prepalatines for a time, turning them inwards, as a boundary 
to the internal nostrils. (See the various figures in my 3rd Memoir “ On the Batra- 
chian Skull.”) 
In Dactylethra (“ Batrachian Skull,” Part II., Plates 56, 57,) the condyle of the 
quadrate, although well in front of the head, is a small, sessile, selliform condyle on 
the edge of the enormous suspensorium, and the prepalatine is confluent with the 
cornu trabeculse. 
But in the larva of the other Aglossal Batrachian (Pipa, same memoir, Plates 60, 
(namely, Nov., 1880), lie himself worked out this subject afresh, and found that there was no difficulty, as 
a true 2nd branch of the trigeminal does pass over the subocular arch. My later dissections confirm this 
important discovery. 
3 H 2 
