118 
into two fleshy ‘lip-like structures” projecting into the ventral side 
of the mouth (fig. 1). As I have previously stated, this piston car- 
tilage at no time exhibits any of the characteristic features of the 
first visceral or jaw arch. 
(B) It is an elementary fact in Gnathostome morphology that 
the parietal lateral myotomes, which are so characteristic of the trunk 
region, have all disappeared as such in the region of the head. This 
disappearance is due, directly and indirectly, to the presence of the 
visceral arches and musculature associated with the large slit-like 
gill-clefts. Directly, since the pronounced development of visceral 
musculature in connection with all these arches (especially the first or 
jaw arch) and the bulk of the gill skeleton itself leave but little space 
for any other muscles, and indirectly. since the presence of this well- 
developed visceral skeleton (especially the jaws) has so contributed 
to the increase in size and complexity of the “head” as to have 
robbed it of all power of flexibility and so rendered superfluous the 
presence of flexor muscles. In all aquatic Gnathostomes locomotion 
is effected by flexion of the tail and trunk but not of the head, 
whereas in Amphioxus and Marsipobranchs, on the other hand, the 
head shares in the general flexibility and therefore locomotion of the 
body. It is worthy of notice that gill-clefts by themselves, provided 
they are small, are not competent to prevent the extension on to the 
head of the series of trunk myotomes, as we see e. g. in Bdellostoma, 
in which the gill-pouches communicate with the exterior by means 
of narrow ectodermal tubes; it is only when such clefts are large 
and associated with a bulky skeleton and musculature that the myo- 
tomes cannot pass externally to them.') Now in the Marsipobranchs 
we find that the lateral series of body myotomes does on each side 
persist in the head region, extending almost to the anterior end, and 
the question arises as to whether this condition is to be considered 
primitive or secondary. If the former then it is certain that a well- 
1) It is true that vertically-elongated gill-clefts supported by a skeleton 
coexist in the same body-region with external myotomes in Amphioxus but 
this is rendered possible by the formation (possibly for the purpose since the 
whole body of Amphioxus undulates in swimming and therefore myotomes 
must extend without interruption along the entire length of each side) of 
the atrial cavity—a secondary feature (GoLDscumipt 14) not found in other 
Chordata. I cannot agree with Wittry (“ Amphioxus and the Ancestry of 
the Vertebrates”) that the atrium of Amphioxus is merely a homologue of 
that of Ascidians. 
