326 DISCOVERY REPORTS 



It will be seen that the difference is slightly greater than twice the standard error, so 

 that one may conclude the difference has some slight significance. 



Thus the general conclusion: that while silicate reduction and rise in temperature 

 combine to favour an increased proportion of the spineless-chain form in the Corethron 

 population, silicate reduction is the more important of the two factors; appears to be 

 justified. 



SPORE FORMATION IN ANTARCTIC PLANKTON DIATOMS 



The recent experimental work of Gross (1937-40) has shown that in the future 

 it will be necessary to make more observations upon spore formation in the endeavour 

 to understand the relations between populations of marine plankton diatoms and their 

 environment. Most important points arise in the consideration of the conditions 

 leading up to auxospore formation and the formation and germination of resting spores. 

 Gross's observations led him to doubt the existence of microspores among centricate 

 diatoms. Among the Antarctic solenoid species Karsten has described probable micro- 

 spore formation in Corethron criophilum, and both Hendey and I have seen stages similar 

 to those described by him, as I have described in earlier sections of this paper. I have 

 also seen a very similar appearance in Rhizosolenia polydactyla Castracane (Hart, 1937, 

 p. 436). It will be an important task of the future to prove whether these ' appearances ' 

 really are microspores. 



In working up large numbers of plankton samples from a general point of view, 

 proper investigation of spore formation is not possible, but some incidental obser- 

 vations of spore formation in the solenoid group, etc., have been included in the notes 

 on the species. As the whole problem deserves separate study in the future, it seems 

 desirable to summarize these observations here. 



In preserved material, auxospore formation is most readily seen in the solenoid 

 diatoms. In Corethron criophilum it was fairly frequent in the upper water-layers at 

 and just after the period of the main increase, in all regions and areas, usually at 

 stations where the species was abundant. At these stations the process was actually 

 taking place in from i to 10% of the population, and very rarely the proportion was 

 higher. Notes on the possibility of microspore formation in this species have already 

 been given. 



Of all plankton diatoms Rhizosolenia alata exhibits auxospore formation most 

 frequently. In the Antarctic zone, at some 10% of the stations worked at all seasons, 

 up to 50% (rarely more) of the individuals showed this phenomenon. It would appear 

 to be most frequent in late summer, however, as seems true of most other members 

 of the genus, even in the northern hemisphere (cf. Wimpenny, 1936). Other Rhizo- 

 solenia spp. which have frequently been observed forming auxospores in the Antarctic 

 zone, chiefly in late summer, are Rh. bidens, Rh. chunii and Rh. truncata. 



On one occasion auxospore formation of Dactyliosolen antarcticus was observed, as 

 shown in Fig. 15. This evidently represents a stage beyond that shown by Gross (1937, 

 pi. 3, fig. 16) in Ditylum brightwellii. The cell wall of the new broad cell formed from 



