ME. A. W. -VVATEES ON MEMBBANIPOEIDiE. 657 



called a variety. However, the Lepralia Otto-MuUeriana, Molly 

 is closely allied to L. Pallasiana, and in previous papers I have 

 called it a variety of L. Pcdlasiana, and have described the 

 ovicells ; and certainly, whether looked upon as a variety or a 

 species, generic separation cannot be thought of. Cases in other 

 genera could be given, as warning against placing too much 

 reliance upon the presence of an external ovicell. In the 

 Elustridse four cases of an external ovicell are known, and 

 Membranipo7Hi tenuirostris, Hincks, and curvirostris, Hincks,- 

 both have prominent ovicells, whereas in M. gregaria, Hell., 

 which is apparently allied, they are immersed. 



One group, of which M. Flemingii, B., may be taken as a type,, 

 has a definite area which is less raised than the rest of the 

 ovicells ; and another, with M. lioieata, L., as the type, has a rib 

 across; but these two forms of ornamentation seem to pass into 

 one another, as may be seen in M. aurita, H., where a rib 

 surrounds the area. 



Before leaving the ovicells it may be mentioned that the 

 M. serrata of MacGillivray has an immersed ovicell entirely 

 similar to those in the group Flustra dentigera, F. denticulata, 

 and -F. spinuligera. MacGrillivray called attention to its similarity 

 to jF. denticulata, but since then F. spinuligera, H,, has been dis- 

 covered, and the avicularia and the mandibles of this and 

 M. serrata are quite similar. Then, further, M. serrata has about 

 10 lateral rosette-plates, and 6 distal, as in the above-mentioned 

 group of Flustra. I am not aware of any other Memhranipora 

 with similar rosette-plates, though both M. gregaria. Hell., and 

 M. sceletos, B., in the rosette-plates and other characters are 

 much like Flustra. 



I have elsewhere shown that the division which Dr. Gregory 

 makes in his British ' Palaeogene Bryozoa,' and in his ' Jurassic 

 Bryozoa,' of genera with internal and those with external ovicells 

 is frequently unavailable, and do not need to further enter into 

 the discussion. 



It has not previously been noticed that the avicularia of 

 Membranipora usually have no complete bar (PI. 47. fig. 2), 

 though there are exceptions, as in M. crassimarginata, Hincks, 

 M. armata, Haswell, M. sculpta, MacG., M. acuta, Hincks, 

 M. aperta, B. In most genera the avicularium is divided by a 

 strong bar. This should be borne in mind, although it may only 

 be another way of sajipg that Memlranipora is not usually 

 strongly calcified. 



