120 
been developed at any period of evolution, since it is incredible to 
suppose either that jaws ever existed in a vertebrate devoid of visceral 
arches and with a head so ill-developed as to take a share in the 
flexion of the body involved in swimming or that these head myo- 
tomes were ever temporarily suppressed whilst a visceral skeleton 
was present!) and, on the gill-arches becoming suppressed and the jaws 
transformed into a piston cartilage, became redeveloped in such a form 
as to imitate in all respects (innervation, e. g.) a primitive condition. 
In the Petromyzontes different conditions obtain. At first sight 
(fig. 3) the myotomes appear to extend in a primitive fashion to the 
vicinity of the anterior end of the head, much as in the Myxinoids, 
but the study of their development (Kourzorr 24) and innervation 
(Near 27, Jounstox 22) proves, on the contrary, that these anterior 
sub-branchial myotomes merely represent a secondary extension for- 
wards on each side of the head of the ventral portions of post-branchial 
myotomes; in other words, these head sub-branchial myotomes of 
Petromyzontes are formed in exactly the same way as the true tongue 
musculature of Gnathostomes (though not from exactly corresponding 
myotomes) and like it are innervated by true hypoglossal fibres. 
Indeed, the only striking difference between these two sets of muscles 
is one of position: in the Gmathostomes the hypoglossal muscles are 
restricted to the mid-ventral line in the space between the two rami 
of the mandible (fig. 4), are relatively small and in most cases 
specialized in connection with the floor of the buccal cavity; in the 
Petromyzontes the sub-branchial muscles retain their primitive position 
on the side of the body, are relatively large, being almost equal in 
size to the myotomes of the trunk region and are segmented like 
ordinary myotomes (though, like that of true tongue muscles, this 
segmentation is quite secondary and does not correspond with that of 
the mother myotomes from which these sub-branchial muscles have 
arisen in the form of muscle buds). It seems therefore that in 
Petromyzontes, as in Gnathostomes, the primitive extension of the 
lateral myotomes on to the head region has been interrupted by the 
presence of the well developed skeleton (the branchial basket)?) and 
1) On the other hand, there is of course no reason why the development 
of a median ventral rasping dentigerous piston should ever have interfered 
with the myotomes in the head and the facts prove that the two can coexist. 
2) It is possible that the great development of the gill-cleft skeleton 
and musculature in Petromyzonts as contrasted with Myxinoids may be due 
