DISTRIBUTION AND GENERAL NOTES ON THE SPECIES 343 



the north, and W. F. Thompson took Notothenia jordani in Grande Bay. We took young A^. macro- 

 cephala in 48° 50' S in the open ocean, but large individuals were obtained by us only in southern 

 coastal waters. In New Zealand, however, adults of this species are known to range as far north as 

 the southern coasts of North Island. All the other shallow-water Nototheniiformes were taken (by us) 

 only in the southern region. 



The Nototheniiformes are preeminently an Antarctic group, completely dominant over all other 

 fishes in such small areas of the vast oceans south of the Antarctic convergence as are sufficiently 

 shallow to support any coastal fish fauna. Their abundance and variety in the Patagonian region and 

 their tendency to be distributed mainly near its southern limits, are consistent with the view that 

 they have spread northwards, but we have seen that only Harpagifer and Dissostichus have species 

 common to both zones. It thus seems probable that the Patagonian fish fauna has been insulated 

 against invasion from the south for a very long time, and that the hydrological barrier of the Antarctic 

 convergence provides too big a contrast in environment for the majority of such fishes to overcome. 

 Quite early in these investigations, when attention was focused on large species (notably hake) that 

 migrate over long distances in a comparatively short space of time, we found it impossible to gain 

 much by the study of depth distribution because of the very slight gradient on the plain of the shelf. 

 This necessitated laborious calculations of the distance of each observation from the coast before we 

 could attempt to follow the movements of such fishes. With the Nototheniiformes it is quite otherwise. 

 Here we have a group consisting for the most part of fairly small bottom-living fish with limited powers 

 of movement, and the study of depth distribution has helped a great deal in our attempts to gain some 

 insight into their probable way of life. 



The depth relations of the Patagonian species are summarized in Fig. 42, and the differences m the 

 mean depths recorded are given in Table 36, with data sufficient to determine their statistical signffi- 

 cance. The figure shows the effective mean depths and extreme ranges observed for all the species 

 except Dissostichus eleginoides, our few specimens of which were so widely dispersed as to demand 

 individual plotting. The distribution of all the species descending below the 50 m. level is also in- 

 dicated by the black polygons. The widths of these are proportional to the relative abundance of each 

 species within each 50 m. depth grouping. The littoral species are further indicated by stippling of 

 the rectangle covering the whole of the observed depth range of each. 



It will be seen that the species can be divided into three main groups according to their depth 

 distributions : 



I Deep-water species found mainly on the plain of the shelf and rarely beyond the shelf edge: 

 Cottoperca gobio, Notothenia guntheri, N. ramsayi, N. elegans, Dissostichus eleginoides and Champso- 



cephaliis esox. , 



II ' First-slope' dwellers, rarely descending below 100 m., where the plain of the shelf may be 

 said to begin: Notothenia canina, N. jordani, N. tessellata, N. hngipes and perhaps Harpagifer hsprnis. 



III Exclusively littoral species: Notothenia brevicauda, N. zviltoni, N. squamtceps, N. stma, 

 N. cornucola, N. macrocephala (adults) and Eleginops maclovinus {Bovichtus argentinus may fit m here). 



Within the first two groupings it proved possible to recognize further distributional trends, either 

 regional or bathymetric, which serve to differentiate the species still further. As a result of this (always 

 excepting the dominant and ubiquitous Notothenia ramsayi) most of the species show a distinctive 

 distributional pattern that tends to minimize territorial overlapping between them. The possible 

 ecological ' advantage ' of this in lessening competition between species of similar size, may be one of 

 the factors that has led to the slight modifications in food requirements and general habits that it must 

 have entailed. 



