120 BRITISH DESMIDIE^. 



The end vievv varies in form : in most of the species it is trian- 

 gular or quadrangular, and the angles are either rounded or elongated 

 into rays ; in some it is circular with five or more processes forming 

 marginal rays ; in a few species it is compressed and the extremities 

 terminate by a process. 



Ehrenberg in his great work has distributed the plants, which I 

 shall here describe, among different genera according to the number 

 of angles or processes seen in an end view. Thus he refers those with 

 three angles to Desmidium, and those with four to Staurastrum, and 

 he formed his genus Pentasterias for the reception of a plant with 

 five rays. But this arrangement appears unnatural; not only be- 

 cause it separates nearly allied forms, but also because the number 

 of rays, as Meneghini remarks, is not constant even in the same 

 species : Professor Bailey says, when describing an American spe- 

 cies, "The number of arms is usuall}' three, but I have met with 

 specimens in which one corpuscule had three and the other four arms, 

 others in which both had four, and others again in which both had 

 five arms." I have myself seen fronds of Staurastrum paradoxum 

 and S. dejectuni, one segment of which had four, and the other only 

 three rays. I have generally found the Pentasterias margaritaceum 

 of Ehrenberg with six rays, although not uafrequently with five, and 

 occasionally with seven rays to each segment. Since we already know 

 that about half the British species vary in the number of their rays, 

 and specimens with three rays conjugate with four-rayed ones, the 

 number of rays is not only altogether useless as a generic character, 

 but it does not distinguish species and scarcely indeed varieties. 



Staurastrum contains more species than any other genus in the 

 family. M. de Brebisson enumerates forty in a list which he has 

 sent me, to which several must be added that have been figured by 

 Professor Bailey and others. The species exhibit a great variety of 

 forms, and but little affinity can be traced between many of them : — 

 as, for instance, between Staurastrum tumidum and ^. paradoxum, 

 or between S. dejectum and S. bacillare. Nevertheless, desirable as it 

 would be to separate plants so little allied, it seems better to keep 

 them together for the present than prematurely to attempt their 

 separation. 



I should greatly have preferred Kiitzing's name Phycastrum to 

 Staurastrum ; for the latter is applicable only to a limited number of 

 the species. Still, as it had been previously employed by Meneghini 



