Mr. J. Ralfs on the Diatomacese. 275 



p. 6; Grev. in Hook. Br. Fl. p. 404 ; Kutz. Syn. Diatom, p. 585. 



Odontella aurita, Ag. Consp. Diatom, p. 56 ; Harv. Br. Alg. p. 201 ; 



Berk, in Eng. Bot. t. 2842. fig. 2. (excellent). 



On marine algae. East-Bourne, Sussex, Mr. Jenner. 



Frustules very minute, quadrilateral, varying from nearly 

 square to tmce as long as broad. This difference depends upon 

 the breadth of the central portion or band, which is narrowest in 

 the newly-formed frustule, and broadest at the time two new frus- 

 tules are formed, whilst the lateral portions remain nearly of the 

 same size in all the stages of growth. 



The reticulations of the frustides are so very minute, that they 

 can only be perceived by emplojang the highest power of the mi- 

 croscope after the colouring matter is removed*. 



The lateral portions are constricted at their junction with the 

 central one, and are destitute of striae ; I have never seen more 

 than one convex projection between the angles, which are also 

 more elongated and less rounded at the point than those of the 

 preceding species. 



It will be observed that this species differs from pulchella in 

 the same manner as Isthmia enervis does from /. ohliquuta, namely 

 in the absence of striae. But Biddulphia (mrita also differs from 

 B. pulchella in its smaller size, in the obscure reticulations of the 

 frustule, and by the constriction at the junction of the lateral 

 portions with the central one. 



A specimen of this species from the Cape of Good Hope, given 

 me by Mr. Harvey, has the cellulose structure rather more evi- 

 dent than the Sussex specimens. 



Plate VIII. fig. 4. Biddulphia aurita. 



Amphitetras, Eh. 

 Filaments 4-sided ; frustules reticulated, cubiform, cohering by 

 one of the angles and forming a zigzag chain. 



Filaments attached by one of the angles of the basal frustule ; 

 frustules cubiform, rectangular, or more frequently with all the 

 angles slightly produced, reticulated, cohering irregularly by one 

 of their angles so as to form a zigzag chain. 



This genus agrees with Isthmia and Biddulphia in the reticu- 

 lated structure of the frustules ; in the great size of the lateral 

 surfaces, between which the central portion appears like a baud ; 

 in the produced angles, which are situated entirely in the lateral 

 portions ; and in cohering at the angles and thus forming a zig- 

 zag chain. It also agrees with these genera in their peculiar 



* Mi's. Griffiths employs nitric acid for the purpose of removing the co- 

 louring matter. This method is on some accounts preferable to that of burn- 

 ing, especially as it is less liable to injure the frustules ; it takes however more 

 time, whilst the other requires only a few minutes. 



