110 Mr. J. Ralfs on the Diatomacese. 



more numerous^ and are then generally from six to nine in num- 

 ber, but they are never present towards the ends. This may 

 always be easily distinguished from the other species by the di- 

 stinct strife, and by their terminations forming puncta along the 

 margins. 



When young the frustules are filled with a minutely granular 

 fluid, but this is soon condensed into a few large granules, and 

 the rest of the frustule becomes nearly hyaline. 



There are generally about eight or ten granules w^hich are large, 

 and sometimes collected into a central mass, but more usually 

 arranged in one or two rows. 



"When the filaments of Diatoma vulgare are broken up into 

 single frustules, they greatly resemble those of F. hyemalis, but 

 in the latter the puncta at the ends of the frustules are more mi- 

 nute ; the strife of the lateral sm-faces do not appear to encroach 

 upon the front surface, and the puncta along the lateral margins 

 are more distant, fewer, and more distinctly marked. Diatoma 

 vulgare also turns green in drying, whilst this plant remains of a 

 brownish colom*. 



Plate II. fig. 5. a, states of/', hyemalis; h, single frustule; c, frustule, 

 deprived of its endochrome; d, views of the lateral surfaces. 



4. F. virescens. Frustules broad, with two evident puncta at each 



end ; lateral surfaces turgid-lanceolate, constricted near the ends ; 



strise none, or indistinct. Frag, pectinalis, Ehrenb. Die Infus. 



p. 206. t. 16. fig. 1 ; Pritch. Infus. p. 222. fig. 176. 



In freshwater pools. Cold Bath, Tunbridge Wells, Mr. Jenner. 

 Madron and Chyanhal Moor near Penzance. 



Plant green, not much altered in diying : when verging to 

 decay it is occasionally found of a bro^vn colour. 



Filaments elongated, attenuated, less fragile than the preceding 

 species. The frustules frequently separate and cohere by the 

 angles in a zigzag chain, from which circumstance it is doubtful 

 whether the plant is rightly placed in this genus. They are 

 often nearly square, but more frequently they are thi-ee or four, 

 and sometimes five or six times longer than broad. The endo- 

 chrome is greenish, and consists of numerous small granules, 

 either scattered or collected together in the centre of the frustule. 

 The lateral surfaces are rather broad, somewhat lanceolate, biit 

 constricted near the ends, which are obtuse. 



When the frustules adhere by their angles, this plant bears a 

 considerable resemblance to yormg specimelas of Diatoma vulgare, 

 from which however it may be distinguished by its green colour 

 when recent, by the fomi of the lateral surfaces, and by the ap- 

 \ arent absence of strise under a moderate power of the microscope. 

 In one or other of these respects it diflFers from any of the pre- 

 ceding species. 



