ANIMAL NATURE OF DIATOME^E. 4S9 



division ought to be that of true Diatomese and Actiniscese. 

 Yet, in the actual state of science, we may characterise 

 the first as completely loricated, the second reduced, in 

 the dermal skeleton (dermoscheletio] , to the presence of 

 a simple dorsal shield. It is indeed true, that, in other 

 classes of animals (Mollusca), the characters derived solely 

 from the dermal skeleton have not so high a taxonomic 

 value. It is possible that we may come to establish 

 the basis of classification, even in Diatomeae, upon the 

 particulars of their intimate organic structure ; but since 

 this may never be realised, I believe that, provisionally 

 at least, this character should be considered the prin- 

 cipal one. 



The order, therefore, of Appendiculatae (Tripodiscieae 

 Biddulphieae, Angulatse, Actiniscese) is the most artifi- 

 cial of all, and even in that respect inconsistent. By the 

 same title that Zygoccros and Triceratium are inserted, 

 we ought also to insert AmpJdtetras and Ampldpentas 

 (if they be true Diatomeae). The Biddulphiese constitute 

 a very natural group, when we unite the Anguliferae 

 and exclude Lithodesmium and the Angulatse. There 

 would remain some doubt as to the genus Odontella, 

 in respect to the want of areolation in the shield. Tri- 

 podiscus would remain intermediate between this group, 

 and the Coscinodiscege. Finally, the Actiniscese, in my 

 opinion, ought to be entirely excluded. 



Thus the entire tribe of Areolata (Disciformes, Appen- 

 diculata,) would be reduced to three groups only, 

 (Coscinodiscese, Tripodiscese, Biddulphiese,) to which, 

 this denomination really belongs. And here, we cannot 

 but adduce some considerations on the organic condition 

 that gives origin to it. In the Biddulphiese, the struc- 

 ture of the shield seems evidently cellular. In the Cos- 

 cinodiscese we have supposed it to be such on account of 

 the optical phenomena it presents, and examination 

 confirms this supposition. Comparing, then, the tribe of 

 AreolataB with the preceding, under this point of view, 

 and independently of the animal or vegetable nature 



