450 Mr. J. Ralfs on the Diatomaceae. 



The different use of the terms length and breadth by botanists 

 serves to perplex the student ; for when they are applied to Dia- 

 toma and to Fragilaria they have exactly opposite meanings, al- 

 though the only difference between the genera is, that in Fragi- 

 laria the separation of the frustules is complete, whilst in Diatoma 

 they still adliere at their angles. 



Those naturalists who regard the Diatoinacece as belonging to 

 the animal kingdom have been more consistent ; they consider 

 each fnistule as an individual, and apply the same terms to the 

 same parts in Fragilaria as in Diatoma. 



In the following remarks I shall employ the terms length and 

 breadth in the sense in which they are used in om- British works 

 and in Agardh. Thus the length of the tmstule will be the breadth 

 of the filament, and its breadth a portion of the length of the latter. 

 The ends of the frustule will form part of the margin of the fila- 

 ment, and its sides ^\dll be those sm-faces which were in contact 

 before any separation : thus in Plate VIII. fig. 5, where a repre- 

 sents the ends, b the sides, and c the front. 



The British species of this genus form two very distinct groups, 

 considered by Mr. Shuttleworth, and I believe rightly, to belong to 

 distinct genera, which are called by him Diatoma and Tabellaria. 



1st, Diatoma. — In this group no stride are seen on the front 

 sm-face, and there is no transverse canal*, whilst the lateral sm'- 

 faces have transverse striae, the ends of which appear along the 

 margins of the frustules when these are in then* usual position. 

 PI. YIII. fig. 6. 



2nd, Tabellaria. — In this group two or more longitudinal strise, 

 interrupted in the centre by a canal, are seen on the front surface, 

 but there are no strise on the lateral sm-faces. PI. VIII. fig. 7. 



The figm-es in Dillw^m^s ' Conf.^ t. 28. fig. B. and D, and in 

 Berkeley's ' Alg.' t. 6, which have been supposed to represent 

 changes of fonn, seem to be rather oblique \dews of the sides of 

 the fnistules united in chains. 



I am happy to find that ]\Ir. Bon-er and Mr. Shuttleworth have 

 arrived at the same conclusion, and still more that in a letter re- 

 ceived from Mr. Berkeley in April last, he ex-presses his o^\ti behef 

 that these changes are merely various Aiews of the thread. 



* Inner or lateral surface striated, without a transverse canal, and 



destitute of stria on its front surface. (Diatoma.) 



1 . D. vulgare, Bory . Frustules convex, three or four times as long as 



broad. D. vulgare, Kutz. in Linnsea 1833, p. 582. f. 6Q. D. tenue, 



Grev. Crypt. Fl. t. 354 ; Berk. Brit. Alg. t. 6! ; Hook. Br. Fl. 



v.ii.p.406 ; Harv.Man. of Br. Alg. p. 202. D. tenue, a.moniUfornie 



* I liere gladly adopt Mr. Shuttleworth's term canal in preference to rib, 

 by which I had previously designated it in my correspondence. 



