246 DISCOVERY REPORTS 



around the South Shetland Islands (see p. 240). Fig. 4 may be compared with a map of 

 the 'supposed migration routes of Humpbacks' drawn by Kellogg (1929, fig. 2, p. 473), 

 and it will be seen that the two agree fairly well. There is no reason to suppose that any 

 additional Pacific group of Humpbacks exists between the New Zealand and Bellings- 

 hausen Sea groups. There is no indication of any concentration of Humpbacks in this 

 position in the observations of the ' Discovery II ', and the coastal regions on both sides 

 of the South Pacific sector have already been related to groups in the Antarctic. 



Table 14. Provisional classification of the southern stocks of Humpbacks 



It is to be inferred then that there are five more or less self-contained stocks of Hump- 

 backs in the Southern Ocean. The existence of separate stocks was recognized in general 

 terms by Rayner, and the analysis of catch statistics and marking positions has now 

 enabled us to amplify his conclusions. The classification shown in Table 14 and Fig. 4 

 is of course liable to some modification. Perhaps a more exhaustive analysis of the posi- 

 tions of catches in the factories' log books would make it possible to define the limits of 

 the Antarctic concentrations in a good deal more detail, and might suggest subdivisions 

 of some of the main groups. It is possible that there is a slight infiltration from time to 

 time from one group to another, and it might, for instance, be found that the West 

 Australian group is less rigidly separated from the New Zealand group than from the 

 African group. There might also be some exchange between the ill-defined South- West 

 Atlantic group and the Chilean group. 



The colour varieties of Humpbacks referred to on pp. 203-4 are of interest in connexion 

 with this segregation of the stocks. Matthews (1937, p. 35) notes a difference in the 

 proportions of the different colour groups in Humpbacks of South Georgia and South 

 Africa on the one hand and those of New Zealand on the other, and infers ' the possi- 

 bility of some degree of segregation between the South Atlantic, Indian Ocean, and New 

 Zealand herds of Humpbacks '. 



It is difficult to compare the magnitude of the different Antarctic groups, but there 

 seems no doubt that larger numbers of Humpbacks resort in winter to the west sides of 



