DEVELOPMENT OF EUPHAUSIA SUPERB A 103 



The figure shows the great range of average size which can be encountered at any 

 one period. For instance, in the month of November the range in average length (not 

 individual length) is between ii-8 and 22-5 mm.— this from samples of the same year 

 class. This range is almost equal to the increase in average length for the whole period 

 under discussion, namely the second half of August to the first half of January. Although 

 there is this great range in size for any one period, there is a definite and fairly regular 

 increase in the half-monthly average size. In the second half of August the average 

 length of the larvae is 1070 mm.; in the first half of January it is 22-18 mm. These 

 figures involve results from two diflferent year classes, but inspection of the figure 

 shows that in those instances where observations were made over the same period in 

 diflFerent seasons, the half-monthly average length shows such a small variation that it 

 seems justifiable to assume that the rate of growth is approximately the same in different 

 years. 



ANOMALOUS LENGTH FREQUENCIES IN 

 ADOLESCENT FORMS 



In his reports to the Discovery Committee on the results of the circumpolar cruise, 

 John drew attention to the length grouping of the young krill in the catches obtained. 

 His remarks have been incorporated in the account of the circumpolar cruise on p. 137 

 below. The larvae from St. 954 fell into two well-defined groups having their maxima 

 at 13 and 18 mm. respectively: this was in September. Later, in October-November, 

 John depicts (Fig. 61) the difference in constitution between krill taken in the Bellings- 

 hausen Sea, in the ice-free water near South Georgia and in the Weddell Sea. The two 

 sets of data involve two different problems : (a) the occurrence of two distinct length 

 groups simultaneously in the one locality, and (b) the occurrence, in different localities 

 at approximately similar times, of young E. superba differing widely in length frequency. 



There is no obvious explanation for the occurrence of two different length groups 

 in the same locality simultaneously, and the statements made here must be regarded as 

 conjectural. 



John's report of this occurrence led to the reconsideration of certain records from 

 i-m. net catches where groups of £■. superba had been measured and discarded from the 

 material used in this paper. They were regarded as being outside the range of larval 

 and adolescent lengths here dealt with and were given the provisional designation of 

 "small adults ". Thus at St. WS 279, 13. ix. 28, including the " small adults ", a bimodal 

 curve was obtained having maxima at 12-37 ^^^ 27-4 mm. respectively. There were 163 

 larvae in the first group and forty-eight in the second, the two groups being separated 

 by 8 mm., from 16 to 24 mm., within which no larvae were found. Again at St. WS 288, 

 19. ix. 28, the euphausians were arranged in two groups, the one with average length of 

 13-13 mm. and the other with average length 25-18 mm. There were over 20,000 of the 

 smaller larvae and only thirty-seven of the " small adults " ; the two groups were separ- 

 ated by a space of 5-5 mm. 



If the three sets of results detailed above from Sts. 954, WS 279 and WS 288 are 



