The Freshwater Algae of Africa. 497 



that only tliose cells which come to lie near the opening undergo' 

 further division. 



Division into more than four parts has not been observed in the 

 case of initial cells, but the production of eight daughter-cells is not 

 at all uncommon in the later divisions of the colony (Fig. 3, i). I 

 have not seen a case of division into more than eight parts. I am 

 unable to say what, is the exact fate of the products of such division 

 into eight. The mother-cells containing them (Fig. 3, i) are generally 

 very irregular in outline, and it is not uncommon to find cells of this 

 form empty and with a very irregular split in the membrane, as often 

 as not situated laterally (jf. the left hand upper part of Fig. 3, k). 

 I am inclined to take the view that the products of division into eight 

 are liberated in toto from the mother-cell and constitute the repro- 

 ductive cells. It can hardly be doubted that the species under 

 discussion propagates by means of zoospores, the epiphytic habit 

 being difficult to understand on any other assumption. The products 

 of division into eight were, however, generally clothed with a distinct, 

 though thin, membrane ; only in a few cases did they appear to be 

 naked (c/. Fig. 3, I). The sediment, in samples 56 and 57, was 

 examined, and a considerable number of free cells of Ecballocystia 

 ramosa were encountered ; all of these had a thin membi'ane. At the 

 same time it should be mentioned that the smallest attached initial 

 cells were of approximately the same dimensions as the products of 

 division of cells into eight. It would only be possible to settle the 

 method of reproduction fully on fresh material, and I am unable 

 to go beyond the suggestion above made. 



Rejuvenescence is an occasional feature of this species, although it 

 is not nearly so frequent as in E. simplex (cf. below). In this case 

 the contents of a cell acquire a new membrane without any division 

 taking place, the original membrane of the cell breaking open apically 

 and later appearing as a cup within which the new cell is attached 

 (Fig 3, /, and upper left hand portion of /.■). 



As regards the individual cells, they are oval or elliptic-oblong in 

 shape, the larger ones not uncommonly being slightly curved (Fig 3, h). 

 There appears to be a single plate-like chloroplast, which in the young 

 cell invariably contains two pyrenoids, placed symmetrically in the 

 middle line, either near the two ends of the cell (Fig. 3, h) or, more 

 rarely, near the centre. Prior to division, each pyrenoid splits 

 longitudinally into two {cf. Fig. 3, g), so that for a time the cell 

 contains two pairs of pyrenoids. In some cases big cells have been 

 encountered with quite a large number of pyrenoids ; I suspect that 

 these are cells about to divide into eight parts. There is a single. 



