EECENT KESEAECHES ON THE BADIOLAEIA. 175 



character, existing, as it does, in such different forms as Spongo- 

 cyclia, Dictyoplegma, and Spongurus. Therefore Haeckel's 

 8pongurida seems to me an unnatural group. Accordingly 

 I would remove from that assemblage Dictyoplegma, Spongodic- 

 tyum, WiizospTicera, and Spongosplicera ; and associate them with 

 the above Ommatida as ENTOSPHiEEiDA, dividing the group into 

 four sections — (1) Ommatida, (2) Spongosplicsrida, (3) Cladococ- 

 cida, and (4) Ccelodendrida. 



At the same time I fully recognize that the mode of growth 

 (by absorption and redeposition) of the last-named form is ex- 

 ceedingly noteworthy ; and I should be inclined, on that account, 

 to make a distinct primary group of it, but that I suspect an 

 analogous mode of growth may exist, as yet undiscovered, in 

 some other forms. 



To Haeckel's Spongodiscidal ^!fou\di add the genus Stylospongia, 

 as one of discoidal or cylindrical Eadiolaria with spongy skeleton 

 with or without radii ; and I would unite these (as a subordinate 

 group) with Haeckel's very natural section Discida. As I have 

 said, I cannot think the spongy nature of the skeleton to be an 

 important character for the reasons already mentioned ; while 

 if such a form as Spongodiscus is to be associated with any other 

 primary group, I think it must be with the Discida. I think so, 

 because in them the skeleton is made up of a multitude both of 

 circumferential and radial parts, and the ^\e\eton oi Spongodiscus 

 is also made up of a multitude of circumferential and radial parts, 

 only these are quite irregularly arranged instead of being regu- 

 larly aggregated as in the typical Discida. To the Discida I 

 would further add the genus Lithelius : not that I do not attach 

 importance to the peculiarity of the structure of the latter ; but 

 I think that if we may add Coccodiscus (with its concentric sphe- 

 roidal shells) to the Discida, on the one hand, we may also take 

 into it Lithelius on the other. 



In this way we shall succeed in reducing the primary groups 

 from fifteen to seven, which may stand as follows : — 



1. Discida ; 2. Flagellifera ; 3. EntospJicerida ; 4. Acanthome- 

 trida ; 5 . Polycystina ; 6. Collozoa ; and 7. Vesiculata. 



Section I. DISCIDA. 



Eadiolaria mostly discoidal, sometimes elliptical, rarely cylin- 

 drical or spheroidal ; skeleton in part intracapsular, and consisting 

 always of both circumferential and radial parts, which may ^e 



