316 
SIDNEY F. HARMER. 
outer body wall of the larva, lending support to the view that 
the anterior portion of the body is here reflected into the 
vestibule. If, therefore, the invaginated sac corresponds to 
the brain of Loxosoma, the budding regions in Cypho- 
nautes and Loxosoma will also correspond (by imagining 
an eversion of part of the vestibule in the former). In pro- 
ducing a single bud, however, Cyphonautes probably re- 
sembles the larval Pedicellina. 
From RepiachofFs account (18) of the development of 
Tendra zostericola, one of the Cheilostomata, it may 
be seen that the early processes are not unlike those of 
Loxosoma. At the morula stage there is a narrow blastocoel, 
the cells of the dorsal side being smaller than those of the 
ventral side. The latter are now invaginated, forming an 
archenteron, which loses its connection with the exterior. The 
account of the development may from this point be carried on 
from No. 21 in the list of references, aided by figures in the 
Russian paper already referred to. The epiblast thickens 
ventrally, and is invaginated as the “ Saugnapf ” (shown in a 
later stage ( v ) in fig. 22, a reproduction of RepiachofFs PI. ii, 
fig. 5). In front of this “sucker/’ which must not be con- 
fused with the larval foot-gland, appears a stomodaeum which 
meets the archenteron, and the latter gives off anteriorly at 
the point of union with the oesophagus, an outgrowth (hypo- 
blastic) which segments off as a mass of cells compared to 
Hatschek’s “ Entodermknospe.” Just in front of the 
oesophagus is an invagination of the epiblast (fig. 22, #), pro- 
bably (as I conclude from the figure) corresponding to the 
invaginated sac of Cyphonautes; whilst dorsally occurs an 
epiblastic thickening ( y ) representing the foot-gland of the 
larval Loxosoma. 
If this account of RepiachofFs is correct (it is doubted by 
Barrois, No. 23), it seems to me that it forms an important 
clue to thestructure of Cyphonautes and of other Ectoproctan 
larvae. (I am unable to say what view Repiachoff takes of the 
subject in his Russian paper.) The explanation which I would 
suggest is the following. The curvature of the alimentary 
