THE GREEN ALGJE 63s 



mucilage and cyst-membranes, giving the reactions of cellulose 

 in the Cryptomonadineae. On the other hand, there are some 

 marked differences (e.g. the origin of the cilia from within a 

 depression and the obliquity of the cell in the Cryptomonadineae) 

 which point to the relationship not being too close. On the 

 basis of the existing evidence we may perhaps assume a hypo- 

 thetical ancestral group of Flagellates (having several equal 

 cilia, a basin-shaped chloroplast, starch as the product of assimi- 

 lation, no pyrenoid and a limited power of change of shape), 

 giving rise on the one hand to the Cryptomonadineae and on the 

 other hand to the Polyblepharidaceae, the Chlamydomonadaceae 

 branching off during the evolution of the latter. Unfortunately 

 it is as yet impossible to trace the origin of Chlamydomonas and 

 its allies more completely than we have done, for Cryptomonas 

 is, of course, in some respects more highly differentiated than 

 the Polyblepharidaceae, which I should prefer to regard as still 

 being Flagellates. The Flagellate origin of the Isokontan series 

 (of which Chlamydomonas is the starting-point) must, however, 

 be regarded as fully established. The most serious obstacle 

 lies in the absence of sexual reproduction in the Cryptomona- 

 dineae, but in view of the recent discovery of a process of 

 conjugation rather similar to that of the Chlamydomonadaceae 

 in the Euglenineae (Dobell 22) this difficulty may very probably 

 be removed when our knowledge of these groups improves. 

 Sexuality has only quite recently been established in a member 

 of the Polyblepharidaceae (viz. Dunaliclld). 



We may now once more return to the Chlamydomonadaceae 

 and advance from them along the lines of evolution that are by 

 slow degrees becoming manifest. One very obvious line has 

 long been recognised ; the forms that constitute it are classified 

 as the Volvocaceae, and may be looked upon as a glorification 

 of the motile Chlamydomonas-individual. The series is so well 

 known (cf. Oltmanns 53) and has so frequently been described 

 that it will suffice to indicate certain salient features. It would 

 seem that this evolution in the colonial direction had started 

 from two or three different points, for in Spondylomorum (Stein 

 64) (fig. 2, a) we have a colonial Carteria, in Stephanosphcera 

 (Hieronymus 33) a colonial Hcematococcus, while the familiar 

 Gonium (Migula 50) (fig. 2, b) is a colonial Chlamydomonas. 

 Only one of these attempts at colony-formation has, however, 

 proved successful, namely that of Gonium, from which we can 



