76 
t It is perfectly possible that the month and anns might ex- 
change functions during the evolution of a group of animals, 
or that one or both might be replaced by new openings, and 
Semper (Stammverwandsehaft der wirbelthiere und wirbel- 
losen), und Dohrn (Ursprung wiebelthiere), have given very 
convincing evidence that such a change has actually taken 
place in the vertebrate mouth during the evolution of these 
animals, but there is not the least reason for believing that 
anything of the kind has taken place during the evolution of 
the classes of Molluscs, but the whole of the evidence furn- 
ished by Comparative Anatomy and Embryology tends to 
show that nothing of the kind has taken place, but that the 
mouth and anus, and the shell-gland as well, can be homolo- 
gized perfectly in all the classes of true Molluscs, and that 
they are not only homologous with each other, but - must be 
perfectly homologous also with similar structures in the an- 
cestral form of which the classes of Molluscs are modifications. 
If there has been a time when all the classes of Molluscs 
were represented by a single form, a jproto mollusc , with a 
mouth, an anus and a shell-gland, which were homologous 
with the similar structures in all its descendants, this ancestral 
form must have been much later than the “gastrsea,” and if 
it was produced by evolution from a gastrsea at all, it is plain 
that the mouths, anuses and shell-glands of all the classes of 
Molluscs must bear the same relation to the organs and open- 
ings of this ancestor — the “gastrsea.” 
The fact that the blastopore of the gastrula stage does not, 
• according to our best information, bear any such constant re- 
lation to the body of the adult, therefore opposes the conclu- 
sion that this stage has a phylogenative significance, and we 
are fully warranted in the statement that the present state of 
our knowledge forbids the acceptance of the gastrula theory 
as an established generalization of scientific value. 
I do not think, however, that we are justified in going far- 
ther, and concluding the theory is disproved by the facts of 
molluscan development. 
The early stages of the development of the different classes 
*-of vertebrates presents, at first sight, few points in common, 
