482 POLYMORPHISM AND LIFE-HISTORY IN THE DESMIDIACEyE, 



Var. sULCATUM(Nord.). (Pl.xiii., £.21). 



Long.37; lat.32; crass. 18/x. 



Collector, Moura. 



Cos. sulcatum Nord., Alg. Sandvich. T.i., f.lS, 19. This is 

 sciagraphically a variation of sexangulare (as is also papillahim, 

 mihi) with more pointed lateral angles and narrower base. 

 Exactly the same two forms occur here in Cos. hirehim. The 

 dimensions given by Lundell agree perfectly with Nordstedt'.s 

 highest for Cos. sulcatujn. The sulcae are not characteristic of 

 this variation, or indeed of any of them, being sometimes present 

 and sometimes not, in all forms of var. FinmarkicH and var. 

 cyclopeurn, besides. In none have they any permanence, being 

 simply a transitory device to allow for the considerable develop- 

 ment in thickness {crass.) characteristic of the highest forms. In 

 the lower forms they are often reduced to a single granule, or 

 refractive (but not incrassate) spot, generally just below the 

 apical margin but sometimes nearer the centre of the semicell. 

 These hexagonal forms do not constitute a separate life-history 

 within the species, or a distinct side-issue of it. In difFerenfe- 

 sizes they alternate in development with those forms having a 

 regularly arcuate apex. Like most other Desmids, so far from 

 being fixed unchangeable organisms, they are merely shapes, types 

 of outlines which occur over and over again with varying dimen- 

 sions in different stages of the life-history of the plant. The 

 species at its simplest is not a form, but a series of forms, through 

 which development moves. Tliis series, however, is, in Cosmi- 

 rium, obscured largely by a number of degenerate forms produced 

 by repeated division, and by a third set of forms the product of 

 environment. In tliis species, however, I have noted a well- 

 mafked series in major and minor sizes. Compare reniforme- 

 suhreniforme-cyclopeum-suhhexagonum, with rotundaluta-patere- 

 fovhie-australe-nodulatuni. There maybe in a species, a form a 

 'minima, forma minor-, forma major, and forma marAm^t, of the 

 same shape, but never under any circumstances does any one of 

 these develop directly into another — always through at least one 

 different form. 



