108 THE CRAG POLYZOA. 



But from all of these, as well as from the recent P.proboscidea, Milne-Edwards, and 

 another recent Austral form, for which the name P. delicatula is proposed, the diflPerence 

 in size and habit, besides which in this genus we have scarcely any available characters, is 

 so great as to leave no reasonable doubt that the Crag fossil is specifically distinct. 



2. P. PALMATA {n. sp.) PI. XVIII, fig. 2. 



Polyzoario palraato, compresso ; ramis ad basim compressis, superne cylindricis ; cellulis 

 obscure annulatis. 



Polyzoarium palmate, compressed, branches compressed at bottom, cylindrical above; 

 tubes obscurely annulate. 



Radiiaf.—C. Crag, S. Wood. 



The cylindrical branches at once distinguish this species from Entalophora pavo- 

 nina of D'Orbigny ('Terr, Cret.,' pi. dcxx, fig. 12), which is the only other fossil or 

 recent form with which it could well be confounded. Even in small fragments it is dis- 

 tinguished from the preceding by the unevenness of the surface and the uncovered con- 

 dition of the tubes. 



3. PUSTULOPORA SUBVERTICELLATA, {n. sp.) PI. XVIII, fig. 1. 



Polyzoario e ramis gracilibus, cylindricis seu subcompressis, pleruraque teretibus, 

 furcatis, e trunco communi surgentibus composite. Cellulis immersis, extremitate vix 

 prominulis. Superficie glabra porisque minimis ornata. Cellularum orificiis non raro 

 in verticello dispositis. 



Polyzoarium composed of slender, cylindrical or slightly compressed, usually tapering 

 forked branches springing from a common stem not thicker than themselves ; tubes com- 

 pletely immersed except at the extremity, which though uncovered scarcely projects above 

 the surface, which is consequently nearly level, and everywhere studded with fine pores ; 

 openings of cells often irregularly verticellate. 



Habitat.— G. Crag, S. Wood. 



This most abundant species might almost as properly be arranged with Cricopora, 

 Blainv., as with Pustulopora, but as, upon the whole, the openings of the cells are more 

 generally irregularly distributed than disposed in regular whorls, it is perhaps better to 

 leave it in association with Pustulopora. It presents a remarkable character in the fact 

 that the surface of the branches is pretty nearly level, owing to the space between the cells 

 being overlaid with what appears to be a secondary deposit, which covers the tubes com- 

 pletely, and allows only a small portion immediately below the opening to be seen. 

 This deposit, however, like the walls of the tubes themselves, is perforated by pores, which 



