384 DISCOVERY REPORTS 



Clupeoids are important, but our methods, aimed at investigating the ground-fish, could not demon- 

 strate their relative quantity, even roughly. It is very clear however that they are a staple food of the 

 larger mid-water and ground-fish, and, indeed, of all larger creatures in the region capable of swallowing 



them. 



Systematically, the two most prominent groups of ground-fishes were the Nototheniiformes and 

 Zoarcidae. The Nototheniiformes are a large group whose metropolis is the Antarctic Zone, but the 

 Patagonian species, although concentrated mainly on the southern part of the shelf, show only two 

 examples common to both regions. This suggests that dispersal between the Antarctic and sub- 

 Antarctic Zones has for long been difficult, and the hydrological barrier presented by the Antarctic 

 convergence may be the prime factor involved. The Zoarcidae (eel-pouts) prominent elsewhere in the 

 North Pacific and in the Antarctic Zone, provide a large number of species, but were taken in such 

 small numbers that they are thought to be of little ecological importance. 



Two Merlucciidae, a true hake and the ' long-tailed hake ' (Macruronus), were dominant among the 

 larger fishes, and were surpassed in numbers only by the most widespread of the Notothenias 

 (N. ramsayi), a much smaller species of little potential value to man. Gadidae and Heterosomata were few 

 and small. Elasmobranchs also were less important than on most other roughly comparable grounds, 

 dogfish being particularly scarce. The Rajidae provided a profusion of local species difficult to deter- 

 mine, but their numbers were few and sizes small. In general the fish fauna resembled that found in 

 the North Pacific more closely than that of other known fishing grounds, if we regard the Notothenii- 

 formes as filling an ecological niche similar to that occupied by the smaller ' rock-fishes ' of the North 

 Pacific. Jordan (1905, il, p. 501) had pointed to this last similarity. There are two outstanding and 

 discouraging differences between the two areas: the North Pacific abounds in salmon and sizeable 

 Heterosomata, but on the Patagonian shelf, salmon are absent and the only common flatfish too small 

 to be of use. However, the absence of salmon is a drawback shared by the whole of the southern 

 hemisphere, except where small-scale artificial introductions have been made. 



Apart from the hakes, another Patagonian fish is to be found in some plenty in summer, and might 

 eventually be exploited with profit. This is the 'spotted pomfret' (Stromateus maciilatus). Near 

 relatives of this species, whose commercial value has already been proved, are the butterfish of north- 

 eastern U.S.A. and the silvery pomfret of China seas. 



The fish fauna of the Patagonian Continental Shelf is notably poorer in species than that of roughly 

 comparable areas elsewhere : nearly twice as many species are recorded from the Gulf of Maine, and 

 more than three times as many from British seas. Unfortunately, it would seem that there is a corre- 

 sponding lack of quantity of fish that can be trawled, apart from the three most promising species. 



The study of distribution, migrations and general biology of demersal fishes has proceeded apace 

 wherever commercial fisheries are pursued. A vast body of evidence concerning these subjects has 

 accrued, but, with notable exceptions (such as Meek, 191 6; and Kyle, 1926), little attempt at synthesis 

 has been made. Certain broad similarities of behaviour, especially in respect of migration, are common 

 to many demersal fishes of the most diverse philogenetic origin. One can hardly call these tendencies 

 'rules', for there are too many obvious exceptions, but they are followed more or less closely by a 

 majority of the species inhabiting extra-tropical waters, and thus provide an invaluable aid to the 

 ordering of our thoughts on the bionomics of the several members of a fish fauna. Fisheries workers 

 will be very familiar with the tendencies I have in mind, and many more beside, but it is desirable to 

 state a few of them here, because they form a convenient guide to the general biological usefulness of 

 the data presented in the main body of this report, that is to say the notes on the individual species. 

 The main value of such work, from the general biological point of view lies, not so much in discoveries 

 of odd individual deviations from ' normal ' behaviour, as in seeing how far most members of a fauna 



