438 ANIMAL NATURE OF DIATOME^E. 



the genus Schizonema, four (Smithii, corymbosum, apicu- 

 latum, ramosissimum] really belong to it in the sense we 

 have assigned ; three, again, belong to the genus Monnema 

 (rutilans, quadripunctatmn, Dillwynii} ; two are uncertain 

 (lacustre, Grateloupii) ; and one. (micans) must be referred 

 to another genus (Raphidoylcea) . 4th. The distinction 

 established by Greville is well expressed in the character 

 of the genus Monnema, of which he takes M. quadripunc- 

 tatum as a type, establishing, at the same time, as the 

 type of Schizonemce, the 8. Smithii. 5th. When Agardh 

 (1828) discovered the structure of Micromega, he found 

 that many species, which he had formerly ascribed to the 

 genus Schizonema, presented the same appearances, and 

 he referred some of them to his new genus in his ' Con- 

 spectus criticus' (1830). A greater or less rigidity of 

 cartilage is not a sufficient character for the distinction 

 of genera ; and, in fact, we find that it lends little aid in 

 the determination of species, when employed, as it is 

 by Kiitzing, in his subdivision of the genus Micromcga 

 into two sections. It varies, moreover, exceedingly in the 

 same species. 



In respect to Kiitzing's definition, we have only to 

 notice the fine crisped fibres which he seems to have 

 seen encircling the series of Naviculas sometimes, instead 

 of the characteristic partial tubes. I have never hap- 

 pened to see these fibres. In every species I have been 

 able to examine, I always saw, with greater or smaller 

 degree of difficulty, but always distinctly, the partial 

 tubes. Sometimes, certainly, I have seen these tubes 

 so fine, so transparent, so colourless, that, at first sight, 

 they only seemed to be fine margins. By the aid of 

 some reagent, especially of a solution of iodine more or 

 less concentrated, with or without the addition of sul- 

 phuric acid, moderating or interrupting its previous or 

 subsequent reaction, I have succeeded in clearly perceiv- 

 ing the partial tubes. I cannot suppose that the action 

 of these reagents would produce such a coagulation of 

 the surrounding amorphous mucous substance, as to 



