358 Mr. H. J. Carter on some West-Indian 



Don ATI A. 



Lastly, we come to the only remaining species in " Group 

 14-," viz. Donatia hjncurium (after which it was named 

 " Donatina "), which, being a corticate sponge with a peculiar 

 structure and spiculation still allied to the family Suberitida, 

 will be best left where it is. 



Hence the emended classification would stand thus : — 



Order VI. Holoriiaphidota. 



Family 2. Suberitida. 



Grniip ]. Caveuxcsa. Group 5. Polymastina. 



2. StTBCOMPACTA. G. Xenospongina. 



3. CoMPACTA. 7. Placospongina. 



4. liAXA. 8. Donatina. 



It must not be considered that these are all hasty specula- 

 tions, which have only to be read and forgotten, but rather that 

 they are attempts to reduce to useful classification a number of 

 objects .which, although a part of Nature's creation, have 

 hitherto been almost contemptuously disregarded, not so much 

 perhaps from their having passed unnoticed, as from the ques- 

 tion whether they belong to the animal or vegetable kingdom 

 having been undecided. But now that they have been ad- 

 mitted to belong to the former, the subject must be seriously 

 gra]3pled with by the comparative anatomist, and a classifi- 

 cation developed for aiding the memory, which, as in other 

 instances of the kind, can only be produced by time, thought, 

 and experience extending over many generations, like that 

 of botany. 



Returning to a description of the sponges belonging to the 

 Liverpool Free Museum, I have now to add that of a curious 

 variety oi Donatia l7/ncuruon dveds;e(\ by Capt. W. H. Cawne 

 Warren in the harbour of Acapulco, after which a brief 

 history of the species of Donatia will be given. 



Do7iatia multifida^ n. sp. (PI. XII. fig. 22, a-e.) 



Membraniform, lacinulate, expanded, flat or erect, fan- or 

 vase-shaped, proliferous. Texture hard, tough. Colour now 

 pinkish. Surface even, presenting white lines radiating from 

 the excentric expansions to the circumference, which is fim- 

 briated by irregular lacinulate processes of variable length, 

 ending in thin expansions of attachment, by which tliey 

 become adherent, like the tendrils of a scandent plant, to 



