148 CJdorophycece 



I have attempted to give a reasonable classification of the 

 genera of Desmids based upon the evolutionary scheme previously 

 mentioned 1 and upon the recent and sound observations of 

 Liitkemtiller 2 . It is very intere3ting to note that Liitkemiiller's 

 conclusions, which he arrived at chiefly from the study of the 

 minute structure of the cell-wall, coincide almost entirely with the 

 scheme I put forward in 1899, which was based upon a compre- 

 hensive study of the external morphological features of these plants. 



The division of Desmids into filamentous forms and solitary 

 forms, such as is adopted by many writers even at the present 

 time, is no longer tenable ; and such an arrangement as that given 

 by Bessey 3 , in which he places Desmids under the three tribes 

 ' Desmidiese,' ' Arthrodieas ' and ' Cosmarieae,' is obviously one which 

 is not based upon a careful study of the Desmids themselves 

 and is therefore of no value. 



Most Desmids would appear to have had an origin from the 

 genera Mesotcenium and Cylindrocystis, which were themselves 

 derived by retrogression from ancestral filamentous Conjugates 4 . 

 The genera Gonatozygon and Genicularia have so little in common 

 with most other Desmids, and resemble so much some of the 

 present existing filamentous Conjugates, that they must be re- 

 garded as having had a distinct origin from filamentous ancestors. 

 They are however, more specialized than the genera Spirotcenia, 

 Mesotcenium or Cylindrocystis. 



The genus Penium is the most difficult to relegate to its proper 

 place. In the sense in which this genus is generally regarded ib 

 undoubtedly contains many widely different plants, and therefore 

 one hails with delight Liitkemuller's suggestion that the name 

 Netrium be again brought forward as a distinct genus to include 

 four species which obviously form a natural group. On the removal 

 of these four plants from the genus, the remainder do not fit 

 satisfactorily into any one place in the scheme of classification, 

 and the difficulty lies in the fact that our knowledge of the plants 



1 G. S. West in Journ. Linn. Soc. Bot. xxxiv, 1899. 



2 Liitkemtiller in Beitriige zur Biol. der Ptianzen, herausgegeben von Dr F. Cohn, 

 Bd viii, Breslau, 1902. 



3 Bessey in Trans. Amer. Micr. Soc. xxii, I'.lOl. 



4 Had these genera originated directly from Flagellate forms, it is scarcely 

 conceivable that the motile condition would not be frequently reverted to ; in fact, 

 it is reasonable to suppose that it would play a considerable part in the life-history 

 of these Algae. Such is the cose in /ill ijronps of Alr/tc with it direct Flagellate 

 ancestry ; and the entire absence of motile forms of any description from the 

 Desniidiaceas lends strong support to the view put forward from other considera- 

 tions, of the origin of this family of unicells by retrogression. 



