8TATTRASTRUM. 119 



extremities, and each often situated at the apex of a 

 mamillate or obtusely conical protuberance. 



The genus Staurastrum is primarily distinguished by the 

 radial symmetry of the cells as seen in vertical view. It 

 embraces species of more varied character than any other 

 genus of Desmids. 



Through some of the smooth species, and especially through 

 St. tort urn, St. cosmarioideSjSMid St. Clepsydra v'a,i\sibiriciuu,ihe 

 genus Staurastrum is closely related to certain species of Cos- 

 mar ium, and it is highly probable that most of the species of 

 Staurastrum were evolved along various lines from the genus 

 Cosmarium. Specialization has been carried to a much greater 

 extent than in the latter genus, and the development of long, 

 hollow processes at the angles of the semicells is a character 

 which is practically confined to the genus Staurastrum. 



All kinds of spiny forms occur in the genus, from those in 

 which a single spine is present at each angle to those in 

 which the whole surface of the cell is covered with spines. 

 All gradations occur from smooth to granulate species, from 

 granulate to asperulate and minutely-spined forms, and from 

 these to coarsely-spined forms. 



All attempts to split up this genus on natural principles 

 have entirely failed. The relationships of the numerous species 

 are too complex and close, so that only arbitrary lines of 

 demarcation can be drawn. Kven the genus proposed by 

 Halfs as Didymocladon completely breaks down on considering 

 all the known species with accessory processes, both smooth 

 and rough forms. St. furciyerum var. reductnm is almost 

 sufficient in itself to show how unnatural such a genus would 

 be, and the various forms of St. leptacanthum, St. Toliopekali- 

 gense, and others, clearly show how impossible it would be 

 to define the limits of such a genus. 



On the whole, it would seem most unwise to split up the 

 large genus Staurastrum, but, as in the case of the still 

 larger genus Cosmarium, it is more in accordance with our 

 knowledge of the genus to group the species into such 

 sections as will most clearly indicate their probable affinities. 



The submenus Pleurenterium founded bv Lundell in 1871, 



/ 



and raised to generic rank by Wille in 1890, is just as 

 untenable as Pleurotasniopsis amongst the Cosmaria. It was 

 founded to include those species of Staurastrum in which the 

 chloroplasts were parietal, and is in consequence one of the 

 most unnatural groups which could possibly be suggested. 

 It would include a few scattered species, having no relation- 



