100 DISCOVERY REPORTS 



ANTARCTIC BOTTOM WATER 

 Conditions ideal for the formation of freezing water with a salinity greater than 34'5i%o occur during the winter 

 months on the wide continental shelf south and west of the Weddell Sea. In summer the water in the coastal current 

 is well above freezing point and diluted to less than 34 %o. But with the approach of winter the temperature falls 

 and the salinity rises as drainage from the land ceases, ice separates out, snow lies unmelted above the sea ice, and 

 mixing between the surface and underlying water is intensified by increasing convection. A month after midsummer 

 the salinity of the coastal current passing westwards across the Greenwich meridian is rising rapidly, and on the 

 wide continental shelf much farther south and west it feels the effect of winter to the full. 



Now it will be seen from Tables 13 and 14 that although there is often a rather wide vertical gap 

 between the deepest sample and the line I take as marking the transition from warm deep to Antarctic 

 bottom water, there are significant instances (notably at Stations 762, 823, 2594, 637, 638, 11 44, 1994, 

 1713 and 855) where the gap is much narrower. From this, and the probability that the normal habitat 

 of the Second Nauplius is below 1500 m., there would seem to be a distinct possibility that the still 

 deeper hatching eggs are in the bottom water itself and that it is directly from this cold deep stratum 

 that the new-born krill are climbing towards the surface. The remarkable west to east progression of 

 the larvae in and under the Weddell stream revealed in Table 17 could, therefore, it seems, be brought 

 about in two ways, in the first instance through the spreading of the bottom water northwards and 

 eastwards away from its place of origin, carrying with it the developing and hatching eggs, and in the 

 second instance, the developmental ascent having been accomplished, through transport in the surface 

 stream. 1 Somewhat speculative though this may at first appear, it appears less so when one takes into 

 account the major features of the distribution of the bottom water itself. In his chart illustrating the 

 potential temperature of this layer at depths exceeding 4000 m. Deacon (1937, PI. xliv) shows its 

 colder parts (including the very coldest) flowing directly below the Weddell stream, and it seems, 

 therefore, to be more than simply coincidence that, outside the East Wind zone, it is from directly 

 over these colder parts, conspicuously (Table 13, Stations 823, 2594, 1 138, 1 142, 1 144) from above the 

 very coldest, that the young krill (Table 13 and p. 200, Fig. 28) are rising, and that where such deep 

 water has escaped into the relatively shallow basin of the Scotia Sea there too we find them climbing. 

 We have, moreover, direct evidence, although it has only been obtained once, that it is in fact from 

 the bottom water that the larvae are coming up, evidence pointing to the overriding importance of 

 the role this deep current must be playing in the dynamics of the larval dispersal. At Station 2594 

 in 61° 51' S, 00° 11-7' E, a position not far from the southern boundary of the Weddell drift in this 

 meridian, there was cold water extending from the bottom in 5400 m. up to the abnormally shallow 

 depth of 1440 m. (Discovery Reports, Station List, 1937-9). The extreme upper boundary of the cold 

 layer may even have been higher, there being no hydrological observations between 1440 and 1000 m. 

 to show where it actually lay. In the 1500-0 m. vertical net,^ which clearly at this station (see 

 Table 13) must have fished for some unknown period in the upper part of the cold stratum, there were 

 3444 eggs, 1000 Second Nauplii and 400 Metanauplii of E. superba. No eggs were taken in any of the 

 six nets fished between 1000 m. and the surface, and this leads one strongly to suspect that those 

 captured in the deep haul must have occurred well below the 1000 m. level, an unknown proportion 

 of them, there is little doubt, in the bottom water itself. Of a total of 2880 eggs in which the develop- 

 mental condition could readily be determined, 2130, or 75%, contained clearly recognisable developing 



1 Below part at least of the Weddell stream the warm deep water (see p. 123) might also it seems be involved in the eastward 

 movement (see Deacon, 1937, pp. 90, 92). 



^ This net unfortunately failed to close at 1000 m. as intended and was fished open to the surface. In view of the absence of 

 eggs at all levels traversed by the other nets between 1000 m. and the surface, however, it is assumed that those that were 

 captured were in fact taken below 1000 m. Moreover, 75 % of the eggs recorded contained advanced or very advanced 

 Nauplii and this (see p. 183, Figs. 21 and 22) points strongly to a deep and not to a shallow concentration. 



