568 An no Is of fh<' South African Museum. 



tion is complete, the filaments dissociate in such a way that the paired 

 ends of the conjugating cells remain, for a time at least, in connection 

 with the zygospore. Unfortunately, however, the material did not 

 contain any stages older than those figured, although an occasional 

 tendency to split between the successive cells of filaments with mature 

 zygospores was observed. In a very few cases I have seen stages in 

 which it looked as though a dense strip of membrane might also be 

 developing round the zygospore opposite the lumina of the conjugating 

 cells. It is of course possible that none of the zygospoi-es in the 

 material had attained to complete maturity, which would also explain 

 the highly irregular shape of the contents (Fig. 34, b. d). That the 

 very peculiar thickening of the membranes above described plays a 

 great part in the constitution of the Avail of the zygospore can, how- 

 ever, scarcely be doubted. Non-conjugating cells, in threads which 

 are conjugating, are not thickened in any way. 



The species is specially distinguished by the method of conjugation, 

 by the spiral twisting of the conjugating threads, by the very complex 

 membrane-thickenings of the conjugating cells, and by the character 

 of the zygospore. It is possible that the peculiarities in the method 

 of conjugation, combined with those of the zygospore, will ultimately 

 warrant the establishment of a separate genus, but a more detailed 

 study is necessary to settle that point. 



A remote comparison may be instituted with Z. pachyderimwn, 

 W. & G. S. West (On some Freshw. Alg. from the W. Ind., Journ. 

 Linn. Soc, Bot„ xxx, 1894, p. 266, PI. XIII, figs. 1-16), a species 

 which West and Starkey now regard as synonymous with Z. ericetorum 

 (loc. cit., p. 203). There are many marked differences. Somewhat 

 similar membrane-thickenings occur in some species of Debar ya (e. <j. 

 D. hardyi, G. S. West). 



(Note. — Sterile species of Zijgnema were present in samples 1, 22, 

 24, 46, and 50.) 



(4) MESOCARPACtiAE. 



Genus MOUGEOT1A Agardh. 



1. ? Mougeotia gracillima (Hass.), Wittr. ; Borge, op. cit., p. 43, 

 fig. 70. 



Sample 34 (sterile). 



Diam. fib, 6-7 /a; cells 15-20 times as long as broad. 



(Note. — Sterile species of this genus were also observed in samples 

 1. 3. 8, 19, 37, 38. 41, 50, 56, 57.) 



