504 MR. A. W. WATERS OX 



may be no operculum remaining underneath it. Frequently a 

 tubular projection occurs in the middle of the supra-opercular 

 calcareous layer, and there is a closure of the same kind in Schizo- 

 porella unicornis and other Schizoporellidse and various otiier 

 genera besides some Tertiary fossils ; and a similar structure has 

 been described in '''Leprcdia"* syringopora Rss. and is found in 

 various Adeonidfef. It can be seen inS. nivea that these raised 

 ledges across the operculum are the basal lines for a superimposed 

 layer even though the layer is seldom completed ; and a South 

 African species which, though a larger and distinct form, is very 

 similar to S. nivea, explains the growth across the operculum 

 more fully. In it we have the commencement of a superimposed 

 new layer of zotecia, and in one specimen there are about fifty 

 zooecia together on which the walls for another layer are all 

 mapped out, and the wall passes over the operculum, nearly 

 always Avith the same curved line as in aS'. nivea. 



I described and figaired a similar growth in Meliceritiies% «n(l 

 apparently it occurs in various other cases. On the other hand, I 

 have specimens of 8cMzop>orella %inicm^nis Johnst. from Cape Verde 

 Islands and Zanzibar, in which there are many layers, and ea,ch 

 following layer is foi'med by the new walls, nearly always 

 growing exactly above the walls of the older layer. Nevertheless, 

 there is, in some cases, a curved calcareous ridge over the oper- 

 culum for which there seems no object. In S. xmiicornis there is 

 often a tubule on the calcareous closure. 



Some stained sections of Adeonella contorta Mich, in which 

 there are superimposed layeis, show this tubule as an inverted 

 funnel with a long tube (over the operculum) attached to the 

 tissues below the operculum. These closures can only be compared, 

 in a limited sense, with the tubules of the closures of the Cyclo- 

 stomata, as the operculum is unaltered and there is no perforation. 

 There are also membranous closures, and the subject deserves 

 further study. The Schizoporellidge and the Adeonidse are 

 apparently the two families most likely to throw light on the 

 closures of the Cheilostomata. Of course we do not find the 

 closures in the younger zooecia, only in the older ones. In both 

 these families blind cells are very frequent §. 



The Zanzibar forms ai-e not separated as varieties or species, 

 although the one with the tall mucro might perhaps be called var. 

 wasinensis (PI. LXX. fig. 1) on this account. It may be the Schizo- 

 porella linearis of Hincks, Ann. Mag. Nat. Hist. ser. 5, vol. vii. 

 p. 159, pi. ix, fig. 2, to which there is no description or locality. 



* Waters, " North Italian Br3'ozoa," Q. Journ. Geol. Soc. vol. xlvii. p. 20, pi. iii. 

 figs. 3, 4 (1891). 



t Waters, " A Structure in Adeonella contorta Midi.," Ann. Mag. Nat. Hist. 

 ser. 8, vol. ix. p. 498 (1912). 



X "On Cheilostomatous characters in fossil Bryozoa," Ann. Mag. Nat. Hist. 

 ser. 6, vol. viii. p. 52, pi. vi. fig. 4 (1891). 



§ Since the above was written Levinsen has published an important work, " Studies 

 on the Cyclostomata Operculata," D. Kgl. Danske Vidensk. Selsk. Skr. 7 K., Nat. og 

 Math. Af'd. vol. x. pt. 1, 1912, dealing with Melicerititidaj. 



What he calls the opercula of these fossils I should speak of as the closures, for 



