44 BULLETIN" 16 5, UNITED STATES NATIONAL MUSEUM 



colonies do not have real weight when covered by their ectocyst. 

 Moreover, all the integument contained in a cell having the density 

 of water, their absence or presence could not have any hydrostatic 

 consequences. 



Another more plausible explanation is therefore in order. This 

 zooecial epicalcification is more a means of reinforcement and con- 

 solidation, operating in places where a zoarial rupture would occur 

 under certain conditions. In fact, in all the Tertiary or Recent 

 Escharian Bryozoa, this epicalcification can be observed at the base 

 of all the zoaria as noted by Milne Edwards as long ago as 1838. 



Among our fossils we have found only fragments of fronds often 

 bifurcated and always without a base. We have not been able to 

 reconstruct the entire colony from these fragments, but it must have 

 been large (probably many centimeters in height), flat, and more 

 or less regularly flabellate. Such an ensemble would be fragile, 

 and the lateral kenozooecia assuring solidity by a thicker calcifica- 

 tion than the frontal and basal kenozooecia would avoid rupture at 

 the dangerous points. 



In 1920 we offered another hypothesis, suggested by the presence 

 of the vibracula of the Selenariidae. We thought that the colonies 

 of C oscinco pleura digitata were more or less floating and attached 

 to algae. Now that we think the Vincentown marl was accumulated 

 at a considerable depth of water where marine waves have little 

 effect, we consider this opinion no nearer the truth than the first. 



The fragments of this species occur in innumerable quantities m 

 most of the localities studied, but nevertheless the ovicells are rela- 

 tively rare. The larvae, apparently not being delicate, easily and 

 rapidly discovered the substratum necessary for their development. 

 Moreover, the colonies must have been very large, as a single larva 

 could give rise to a great number of fragments. An analogous phe- 

 nomenon can be observed in the Recent seas with Mymozoum trunca- 

 tuTYi Pallas, 1766, in which a single larva gives rise to a bushy colony 

 8 centimeters in height and of the same width. After death such 

 a colony can furnish hundreds of fragments such as those discov- 

 ered in the Miocene formations, where the species is very abundant. 



In general, this observation, which can apply to many other 

 Recent species, makes it possible to establish the principle that zoa- 

 rial fragments with very few ovicells always belong to large colo- 

 nies. In the Cyclostomata, this observation also holds true, and it 

 complicates the task of the paleontologist, who is thus frequently 

 deprived of the essential characters for classification and deter- 

 mination. 



In the Recent seas, large colonies after their death serve imme- 

 diately as a substratum for a throng of other encrusting Bryozoa. 



