480 ANIMAL NATURE OF DIATOME^. 



61. LiTHODESMiUM. — Individua a latere secundaria 

 triangula in corpus prismaticum, articulatum conjmicta. 



The several frustules are united together by a shorter 

 intermediate body, in which Ehrenberg observed a granu- 

 lated surface ; this, he says, appears in young individuals 

 only, which are^ therefore, pliable when dried. He 

 regards this granulation as indicating the process of 

 siliceous ossification. 



The only species of this genus {L. undulatiuni) in- 

 habiting the North Seas, is too little known to exclude 

 all doubt as to its nature. 



63. Amphitetras. — Individua a latere secundaria 

 quadrangula {depressa) adangula istlimo niolli concatenata. 

 Catena; breviter stipitata adnata. 



Kiitzing himself observes, that the species of this 

 genus may easily be thought to belong to the Isthmia. 

 The only difference is that upon which the distinction of 

 the entire family is founded, that is, the angular figure of 

 the lateral surfaces. This character is, indeed, extremely 

 conspicuous, and in want of other data may serve very 

 well to distinguish genera. But that this is of itself 

 sufficient to separate, not families merely, but even 

 orders, will be conceded with difficulty in a natural 

 classification. 



I found the very beautiful A. ' antediluviana among 

 some specimens of Biddulphia quinquelocularis, from 

 Calvados, obligingly supplied by Brebisson. The smooth 

 fascia deserves attention ; it traverses the frustule in the 

 direction of its length. It is sometimes absent, some- 

 times double, or even triple. The appearance of the 

 areolation is different from that of the allied genera. 

 It seems produced by minute projecting papillae. The 

 ample round apertures are very evident, corresponding 

 with each of the four angles of the secondary surfaces, as 

 described by Ehrenberg, but omitted by Kiitzing from 

 his description as well as his figure. 



