NORTH AMERICAN EARLY TERTIARY BRYOZOA. 335 



Evidently the hydrostatic and reproductive functions are identical in each 

 but they operate in a quite variable manner. These variations added to those of 

 calcification, which is also an important function, permits the establishment of a 

 large number of genera almost all rather natural however, which facilitate the 

 study of this very important family. The anatomical structure is given in detail 

 for each group. 



Group 1. SCHIZOPORELIAE Canu and Sassier, 1917. 



Bibliography (Anatomical). 1S77. BABBOIS, Recherches sur 1'embryologie des Bryozoaires, 

 pis. 7, S, figs. 8, 10, 11, 21: pi. 8, figs. 30, 33, 35. 1888. JULLIEN. Mission scientiflque du Cap 

 Horn, p. 48. 1892. WATERS, Observations on the Gland-like Bodies in the Bryozoa, Linnean 

 Society Journal, vol. 24, pi. 19, figs. 6-13. 1000. CAI.VET, Contribution a 1'Histoire naturelle 

 des Bryozoaires ectoproctes inarins, pi. 10. figa 8, 9 ; pi. 11, figs. 16, 17. 1900. WATERS, Bryozoa 

 from Franz Josef Land, Linnean Society Journal, vol. 28. pi. 9, fig. 11; pi. 12, fig. 5. 1902. 

 HARMER, On the morphology of the Cheilostomata, Quarterly Journal of Microscopical Science, 

 p. 301, figs. 48-52 ; p. 303, fig. 46. 1909. WATERS, Reports on the Marine Biology of the Sudanese 

 Red Sea, The Bryozoa, Linnenn Society Journal, vol. 31, pi. 12, fig. 11. 



The operculum is semilunar ; the proximal border bears a slit or rimule which 

 opens the compensatrix. The muscular attachments are two small, symmetrical 

 tuberosities more or less removed from the border. When the proximal border 

 of the aperture is linear it serves as a pivot for the operculum ; when it is arched 

 and the rimule very large the pivot of the operculum is formed by two projecting 

 interior condyles. 



The text figure gives a presentation of the anatomical structure of the species 

 of this group. 



The classification was attempted successively in 1899, 1904. and 1913 by 

 Waters in considering the muscular attachments of the operculum. In 1888 and 

 1903 Jullien formed many genera based on the nature of the frontal or on zoarial 

 peculiarities. We will continue the work of these authors by considering the 

 relationship of the operculum to the ovicell (passage of the eggs and escape of 

 the larvae) . 



After the elimination of the genera created by Jullien and Levinsen, approved 

 besides by Waters, a large number of very different species are still placed in 

 the large genus ScMsoporella Hincks. 1880. 



The group of Schizoporella viridis forming a part of the genus Watersipora 

 Neviani, 1895, is provided with an endozooecial ovicell, and belongs provisionally 

 to the family Hippopodinidae. 



The Schizoporclla sinvofta group, in which the operculum does not correspond 

 to the apertura and which is provided with a rimule spiramen, belongs to our 

 family Stomachetosellidae. 



The SchizoporeTla unicomis group is quite well characterized ; the orifice of 

 the ovicell is closed by a special membrane. This is our genus Schisopodrella. 

 Hincks did not designate a type for his genus Schisoporella; the first species 

 described is Schizoporella unicornis Johnston, 1S47. We could (but it is not 

 obligatory) preserve the name 'of SchiznpOrcTla for this group, but it seems to 



