462 rOLYMOBPHISM AND LIFE-HISTORY IN THE DESMIDIACn:^, 



f.56) a siinilur same state of repeated division in the case of Sf. 

 ■brad datum. There is nothing abnormal in this kind of division. 

 Wherever tlie type is to he found, the same immature form 

 which is produced in this manner may be noted separately, some- 

 limes in quantity, but tlien it is a new species ! 



Finally, in tlie last-mentioned paper, it is shown'^Pl.iii., f.29, 

 :uid Fl.iv., f.43) that newly-divided cells may conjugate and pro- 

 duce a zygospore. If such cells can accomplisii that rare act, it 

 cannot be believed that they are unable, under natural condi- 

 tions, to manage mere vegetative division. 



In the second place, so far from growth having ceased when 



-cell-division is complete, it is only then that the development of 



the plant begins, continuing very slowly in the intervals of cell- 



-division. It is because the latter bulks so largely in the life of 



the Desmid, that degenerate forms of the species are to be found 



in such ai)undance and variety, while the fully-developed form 



itself is so very rarely seen. A true Desmid-species consists of 



an immense number of distinct polymorphic forms which ai'e 



partly successive modifications of the sporangial type under stress 



-of rapidly repeated cell-division, partly abnormal (but in no sense 



monstrous ) forms pi'oduced by unusual combinations of circum- 



.stances, and partly types arising from all these as the result of 



their struggle to develop upwards towards the perfect examplar 



of the species. 



. That spines and processes do develop on the cell, has already 



■been conclusively proved in "Some Sydney Desmids." Pl.xii., 



f.9, shows the well-known St. orbiculare with processes full-grown 



At the basal angles, and developing in pairs down the sides. 



According to Dr. West, St. orbiculare being mature, in the sense 



that both semicells are alike, cannot develop any furtlier. But 



we see tliat it does, and I have myself observed every stage of 



the development. 



Compare also Plate xii. of the present paper, where figs.9-14 

 exhibit the growth of the upper processes of St. sexanmdare, and 

 tigs.15-18 and 20, form a series illustrative of their very gradual 

 development (granule — spine — process), and of the resulting pro- 



