DEVELOPMENT OP THE SKULL IN STURGEONS. 
179 
these things have to be carefully reflected upon before any classification of these ventral 
arches can be made. 
It seems to me that a cartilaginous arch developed inside a recently-closed “ head- 
cavity” must be a part of the “ splanchno-pleure,” and that a cartilage developed 
outside such a recently-closed cavity must belong to the “ somato-pleure.” * 
The ribs belong to the outer lamina of the body-wall of the embryo, so also, it 
appears to me, do the “ extra-branchials ” of the Lamprey, the Tadpole, and the Shark. 
If all this be true, the normal “ intra-branchial ” arches have no counterparts what¬ 
ever in the skeleton of the trunk ; they culminate in the class of Fishes, and are 
imperfectly developed, not only in the Abranchiata, but also in the degenerate 
Marsipobranchii, and the metabolic Amphibia Anura. 
The extra-branchials of some of these latter types, and of the Sharks, have some 
right and title to be classified as a sort of cephalic ribs, but perhaps that claim had 
better not be pressed for the present; it is safer for the Morphologist to keep 
certain things in solution, when any doubt remains, than to crystallize them into what 
may turn out to be obstructive error. 
In the metamorphosis of the larval Sturgeon the additional parts are easily 
explained, and are, for the most part, due to mere increase of certain tracts of tissue, and 
super-additions of secondary nuclei of cartilage, and of various centres of ossification. 
Moreover, the fact that the dermal scutes are largely dominated by the cartilaginous 
endoskeletal structures of the cranium and pharynx, however interesting from one 
point of view, is not of any great fundamental importance. 
With regard to the great “shoots” of cartilage that grow out, forwards, from the 
primary basis-cranii, these are parts that undergo a most extraordinary amount of 
modification in various types; they are specialised superadditions to the essential 
skull, of great importance in Taxonomy , but of little account in that which is funda¬ 
mental in Morphology. 
Comparison with Polyodon. 
The skull of the other principal existing Chondrosteous Ganoid type comes singularly 
near to that of the Acipenserine skull, and in some things is curiouslyAinlike it. Near 
as Polyodon approaches in the structure of its skull to Acipenser, it differs in having 
no complex metapterygoid plate, in the feebler ossification of its visceral arches, and in 
the presence of three pairs of “ endosteal ” centres not to be found in the Sturgeon. 
The anterior palatine ectostosis placed just where both the mesopterygoid and pala¬ 
tine plates meet in the Sturgeon, is probably the true homologue of the palatine bones 
of the Holostean Ganoids ; the presence of a small prootic and opisthotic brings us 
nearer in this case to those higher types. The much larger and more perfect ci orbitar 
The branchial artery lies inside the head cavity, and afterwards, when the branchial arch is developed, 
runs up a groove on the outside of it (see Balfour’s ‘ Elasmobranchs,’ plate 14, fig. 13, a, p. 208; and 
‘ Comp. Embryol.,’ vol. ii., p. 472, fig. 328). 
2 A 2 
