123 
or not the branchial skeleton of Marsipobranchs is homologous with 
the branchial skeleton of Gnathostomes becomes of importance. Bat- 
FOUR’S contention that the Marsipobranchs do not possess and never 
have possessed a branchial skeleton homologous with that of Gnatho- 
stomes has been supported on various grounds. It has been pointed 
out for many years that the Marsipobranch gill skeleton is developed 
quite apart from the gut wall (being situated just under the external 
myotomes), and thus essentially differs in position from true visceral 
arches which are internally situated, that it is developed external to 
the branchial aorta and gill vessels instead of internally as in Gnatho- 
stomes, that in development is appears long before those head cartilages 
(sub-ocular arch, piston cartilage, &c.) which, on the hypothesis of 
Gnathostome ancestry, are assumed to represent anterior arches of 
the visceral skeleton, and quite independently of them, and that it is 
only developed to any extent in the Petromyzonts in which the gill 
apertures frequently serve for ingress as well as egress of water and 
so need a specially developed musculature and is therefore probably 
merely a secondary structure developed for a special purpose. The 
first of the above “grounds” has been criticised (SCHAFFER 33, 
CoLE 6) on the score that mere remoteness from the gut wall cannot 
be considered as by itself excluding the possibility of the Marsipo- 
branch gill skeleton being homologous with true visceral arches, 
especially since parts of the skeleton do come into contact with the 
gut, and this criticism is a just one. The second “ground” has also 
been held to be inconclusive, since although the ventral and lateral 
portions of the branchial basket differ radically from those of the 
visceral arches as regards relationship to the ventral aorta and 
branchial vessels, yet dorsally they are very similar, but personally 
I cannot admit the force of this objection and believe this second 
“sround’ to be conclusive in disproving all homology between the 
Marsipobranch gill skeleton and visceral arches. It is curious that 
many advocates of Gnathostome ancestry have on more than one 
occasion adopted the somewhat feeble expedient, when faced with 
such facts as those disproving the homology of the sub-ocular arch 
and of the branchial skeleton, of saying ‘“ Well, if the greater part 
of the structure is not homologous, at least what is left may be” 
and forthwith assuming that it is! 
Another fact supporting Batrour’s contention is that cited by 
BasHrorD Dean. He says “ As far as my own studies have gone the 
