242 DISCOVERY REPORTS 



Thysanopsetta is one of the important forage species for larger fish, so that the restriction to a particular 

 type of bottom deposit, later described in detail, affords a good illustration of the potential value of 

 Matthews 's work. 



HYDROLOGY 



The water movements over the area covered by the trawling surveys are comparatively simple. Over 

 the whole of the plain of the shelf and eastwards beyond the shelf edge, relatively cold sub-Antarctic 

 water flows northwards in what is known as the Falkland current. To the east and north of the area 

 the warmer subtropical water of the Brazil current flows southwards, and in the region of the con- 

 vergence between these two (corresponding to the subtropical convergence in the open ocean still 

 farther to the east) hydrological conditions are more complicated. Here streams of sub-Antarctic 

 and subtropical surface waters may alternate, giving rise to large differences in salinity and tem- 

 perature within a few miles. Klaehn (191 1) was able to trace the southward movement of the Brazil 

 current as far as 49° S to the north-east of the Falkland Islands, but Deacon (1937, pp. 5S-9) has 

 shown that south of about 431° S, the subtropical water is becoming more and more mixed with sub- 

 Antarctic water. The influence of the Brazil current is strongest in summer, when relatively unmixed 

 subtropical water may extend some 4° farther south than in winter. The complicated conditions 

 around the southern extremity of the Brazil current rarely impinge on the trawling area, but this is 

 the probable route by which fishes that normally live in warmer seas occasionally reach the north-east 

 coasts of the Falkland Islands. The main facts concerning the southward limits of the Brazil current 

 have been ably summarized by Deacon (1937) and an attempt has been made to depict them in Fig. 4, 

 which should be regarded as a pictorial representation of the current system, and not an exact hydro- 

 logical study. 



The Falkland current, which bathes most of the shelf, is composed of sub-Antarctic surface water. 

 Deacon (loc. cit., p. 51) has described how the main west-wind drift of the south Pacific is com- 

 pressed while passing through Drake passage. This augments its speed, and the sub-Antarctic portion 

 of It IS swollen by a relatively small amount of warmer, poorly saline coastal water flowing southwards 

 down the south-west coast of Chile. The resultant of these forces is the so-called Cape Horn current, 

 which IS really a local intensification of the west wind drift. It sometimes reaches a speed of as much 

 as 40 miles per day. 



To the east of Staten Island the Cape Horn current divides in the form of the greek letter y the 

 esser branch swinging north round the Falkland Islands, but mainly between them and the main- 

 land, to form the Falkland current; and the main branch proceeding north-east and then east until it 

 merges into the main easterly (i.e. 'west wind') drift of the open ocean south of the Atlantic 



The Falkland current itself flows most rapidly on its right flank, well oflFshore and beyond the shelf 

 edge, outside our immediate area. It is here that the coldest water is found, but the lower tem- 

 peratures are not caused solely by the greater speed of flow than that obtaining over the shelf. Both 

 Krummel (191 1) and Klaehn (191 1) postulate upwelling as an additional source of the cold water. 

 Where the Brazil and Falkland currents are flowing in opposite directions alongside each other it is 

 natural to suppose, as did Klaehn, that the dynamic disturbances so set up favour the creation of 

 whirls with consequent upwelling. Deacon (1937) considers that Klaehn 's demonstration (191 1, pi. 35 

 hg. 4) of isolated patches of relatively cold water towards the northern end of the Falkland current is 

 proof of upwelling, but our own data did not then provide any fresh evidence of its mechanism.^ 



^^\^^^^^:^^X^£^^^ -' ^- -"-d by the Dis^very Cc.iUee's 



