126 BRITISH DESMIDTACE.E. 



distinguishing features of a Cosmarium is the entire outline of 

 the cells, which are likewise destitute of spines. There is no 

 lobulation of the margin of the semicells at all comparable to 

 that which forms such a feature of most species of Euastrum, 

 and there is a complete absence of an apical notch. 



Among the large number of species at present included in 

 the genus there are several transitional forms which could 

 almost equally well be referred to other genera. A few are 

 closely related to certain species of Euastrum, and others can 

 scarcely be distinguished from some of the constricted species 

 of Penium. Spines are as a rule absent from species of 

 Cosmarium, but a few of the smaller species closely approach 

 the genus Xanthidium in the possession of minute, marginal 

 or submarginal spiues. 



Kuntze (' Revis. gen. plant/ 1891, p. 922) revived the name 

 Ursinella Turpin (' Aperyu organograph/ 1828, p. 316) which 

 he substituted for Cosmarium Corda (1834). Turpiir's descrip- 

 tion is exceedingly bad and scarcely applies to the genus 

 Cosmarium as now understood. For this reason the name 

 " Ursinella'' has been universally rejected. Vide Nordst. in 

 1 Hedwigia/ 1893, p. 152. 



Several attempts have been made to subdivide the genus 

 by the character and disposition of the chloroplasts. Lundell 

 (1879) instituted the subgenus Pleurotseniopsis to include 

 those species of Cosmarium with parietal chloroplasts, and 

 in 1887 Lagerheim placed it as a genus. Gay (1884) also 

 described the genus Cosmaridium to include Cosmarium 

 Cucumis, a species characterized by its parietal chloroplasts. 

 With regard to the chloroplasts in the genus Cosmarium, it 

 must be borne in mind that the axile condition is unquestion- 

 ably the priuiitive one, and that the parietal condition has 

 been independently acquired by a few scattered species of the 

 genus. The removal of these species from all their nearest 

 relatives and their inclusion in a separate genus is very inex- 

 pedient, as it results in the association of species having 110 

 affinity with each other and which are not in any way closely 

 related. Moreover, as transitional conditions are sometimes 

 observed between truly axile and truly parietal chloroplasts 

 (vide PL LX. fig. 4), the disposition of the chloroplasts cannot 

 well be taken as a basis for the primary subdivision of the 

 genus. 



Many of the slightly constricted species have been placed 

 by various authors in the genus Dysphinctium Nag. (1849). 

 This genus, with which must be identified Calocylindrus 

 (Nag.) Kirchn. (1878), has been differentiated from Cosmarium 



