125 
attaching itself to the struggling bodies of other animals. But apart 
from the fact that the presence of jaws implies a mode of life from 
which no known Gnathostome has thought it worth while to depart, 
and which therefore it is very improbable that the Marsipobranch 
ever experienced, it is obviously impossible, if the ancestral Marsipo- 
branch had employed its jaws as organs of fixation, that they could 
either have become transformed into a piston cartilage or have become 
so degenerate or modified as to be unrecognisable”. This purely 
a priori argument is superfluous because advocates of Gnathostome 
ancestry may maintain that although Marsipobranchs arose from 
ancestors possessing true visceral arches, yet the first pair of these 
arches may not have at that stage of evolution assumed the function 
of jaws; it is also not strictly valid since Mr. Tarr ReGan informs me 
that the numerous species of Loricariidae and also Gyrinochirus, one 
of the Cyprinidae, among the Teleosts, do possess labial suckers which 
they employ for temporary attachment to stones in much the same 
way as lampreys (Gyrinochirus and probably the other genera pos- 
sessing gill-clefts which are both inhalent and exhalent).!) Neither in 
these highly-specialized fishes however, nor in the degenerate Sturgeon 
have the jaws lost their original function and if functional jaws had 
been present in the ancestral Marsipobranchs, it seems unlikely that 
they would have done so then. 
Under this heading I may perhaps add one more purely morpho- 
logical fact which to me appears to show that the skeletal elements 
surrounding the mouth of the ancestors of Marsipobranchs were 
probably arranged on quite a different plan to the visceral arches of 
Gnathostomes and this fact is that, whatever views may be held 
concerning the origin of the mouth cavity in Chordata, the buccal 
cavity of Marsipobranchs is either not homologous with that of Gnathost- 
omes or is so differently related to surrounding parts as to be equi- 
valent to a new cavity. The relative positions of the hypophysis in 
the two groups alone proves this. Also it is notorious that even in 
the two sub-classes of the Marsipobranchs themselves the buccal 
cavities are not entirely homologous. “The stomodeum in the hag- 
fish is not the entire homologue of the structure in the lamprey ” 
(Dean 7), since, according to Hauer (19) and von KUPrFER (25), 
1) Certain tadpoles living in mountain streams of Natal, the Himalayas 
and other regions have also acquired this habit and suctorial lips (vide Nature, 
March 13, 1913, p. 33). 
