304 DISCOVERY REPORTS 



phenomenon of seasonal vertical migration is connected with temperature. Somme 

 (1934) suggests that light intensity is also an important factor. 



From the somewhat scanty data set forth above it seems possible to conclude that 

 Rhincalaniis, in the Bellingshausen Sea current and the Antarctic water of the West 

 Wind Drift generally, undertakes a seasonal vertical migration similar to that established 

 by S0mme (1934) for Calanus finmarchicus and Calanus hyperboreus in the northern 

 hemisphere, spending the summer months November or December to about February 

 at the surface and descending in April into the 250-100 m. layer and sinking still farther, 

 below 250 m., at the beginning of May. It reappears in the 250-100 m. layer once more 

 in October. Thus it follows that the winter months are passed in the warm deep water, 

 in water of higher temperature and salinity, which is moving southwards and will there- 

 fore tend to carry the animal southwards. In the summer the species inhabits the north- 

 ward-flowing Antarctic surface water, which will tend to carry it once more in a north- 

 ward direction. 



This theory of the seasonal vertical movements of Rhincalanus gigas receives con- 

 firmation from the work of the 'Discovery II ' on her third commission (1933^5)- Dr 

 N. A. Mackintosh, under whose direction the work was carried out, has given a pre- 

 liminary account of some of its more immediate results in Nature (1935). Part of the 

 ship's programme during the seasons 1933-4 ^^'^ ^934~5 involved the repetition at dif- 

 ferent times of the season of a line of plankton stations along the meridian of 80° W 

 (western end of the Drake Passage). This line of stations was taken in December 1933 

 and March, September, October and November 1934. At each station a series of six 

 vertical hauls was made, using the 70-cm. closing net, from various depths between 

 1000 m. and the surface. A preliminary examination of the samples obtained from these 

 hauls reveals vertical movements on a large scale, not only of R. gigas but of several other 

 macroplanktonic species. Mackintosh outlines the seasonal changes in the bathymetrical 

 distribution of Rhincalamis as an illustration of these movements. In December the 

 species was mainly concentrated at the surface above the 250-m. level, as was suggested 

 by the hauls with i-m. nets taken in 193 1-2 and 1932-3. In March it tended to sink, 

 larger catches being taken around the 500-m. level, especially north of the convergence. 

 In September it was practically confined to the warm deep water below 500 m., but in 

 October and November it regained the Antarctic surface water above 250 m. 



Weddell Sea (Table IV b) 



There are no observations extending over a long period by which to judge the seasonal 

 vertical movements of Rhhtcalamis in the Weddell Sea, but stations were taken in this 

 water in December and January 193 1-2 and November, December, February and 

 March in the season 1932-3. Fig. 13 shows the percentage of the total catch taken in 

 the upper and lower nets at all the stations in Weddell Sea water during the two seasons 

 and is constructed on the same plan as Fig. 12. 



In December and January, in the season 193 1-2, the majority of the catch was in the 

 lower nets at all stations taken east of South Georgia in the oldest type of Weddell Sea 



