OF THE SKULL IN THE UKODELOUS AMPHIBIA. 
593 
We now come to two causes of modification of the basal plates that are very important, 
namely, divarication of the paired tracts, and obstruction of the unpaired axis ; the latter 
is the most important of the two. 
The rapid growth of the brain-vesicles, as they turn over the bent skeletal axis, causes 
the floor of the membranous primordial cranium to bulge, and this bulging affects the 
hardening bands that are ready to begin to build the “ chondrocranium.” 
The middle tracts (trabeculae), although starting from the sides of the apex of the 
notochord, curve round the infero -lateral regions of the brain-sac, and in an early state 
are merely like two horns, gently curved outwards from their origin, and far apart at 
their tips (Plate 22. fig. 1, tr.). 
At their beginning, however, these outbent bars have the notochord between them, 
so that the posterior trabecular region has in it all the elements that go to form 
vertebras. 
Now, however, we come to obstruction. The posterior part of the oval space, marked 
out by the divaricating trabeculae, is occupied by a structure that effectually stops all 
further forward growth of the notochord. 
A process from the postero-inferior region of the fore brain, the infundibulum, here 
joins on to the pituitary body — a tear-shaped sac that passes downwards and backwards 
from its cerebral attachment, and rests upon the oral mucous membrane. 
Here is a perfect barrier; and although the sheath of the notochord may grow 
upwards between the hind and mid brain, and downwards below the pituitary body, as 
is seen in the Axolotl and Salmon , this axial growth is stopped along the basal 
mid line. 
In the Selachians it pushes itself against the barrier, and becomes crumpled and 
beaded ; it then smooths out its creases, and presses its fore end against the intrusive 
body, so as to flatten its point ; but it is there arrested. 
Therefore it is evident that we can no longer speak of vertebral division of the skull 
from this point, and this is no further forwards than the “ posterior clinoid ” region. 
From thence we have the highly modified mesoblastic tracts running forwards to the 
frontal wall, without their proper bond, and subject to conditions that cannot be found 
in any region of the axis of the body. 
I now come to consider another great factor in the specialization of a skull, as 
such, namely, the sense-capsules, the “ paraneurals ” of Professor Huxley. 
Behind the ear-capsules the basal plates are free to grow upwards, and to arch over 
the neural axis. Between them the basal plate is, as it were, cut away and bevelled ; 
and no upgrowth is possible, as these sacs are built into the sides of the cranium and 
form much of its side walls. 
Moreover, the basal tracts become confluent with these capsules ; and afterwards, when 
ossification sets in, it often works blindly, not keeping to the primary morphological 
landmarks. 
Then, while the neural arch or roof, that should grow up everywhere from the basal 
mdccclxxvii. 4 o 
