VERTICAL DISTRIBUTION OF ADOLESCENTS AND ADULTS 325 



which simultaneous observations were made through each depth horizon. The 250-100 m. haul 

 takes 30 min. and the numbers taken in these nets have been corrected for a 20 min. haul for comparison 

 with the upper net. Although the two layers sampled are usually approximately loo-o m. and 

 250-100 m., in some instances the upper net fished from 150-0 m. and the lower net started fishing 

 as deep as 500 m. ; no allowance has been made for this variation. The number of observations in 

 each 2-hourly period varies greatly owing to the routine adopted for working stations. 



It will be seen that between 0600 and 1600 hrs. adolescent and adult E. triacantha are absent from 

 the upper 100 m. From 1600 to 2000 hrs. the numbers increase rather slowly, between 2000 and 

 2200 hrs. there is a more rapid increase and the maximum is reached between 2200 and midnight. 

 The descent in the early morning follows a similar by slightly steeper curve. The pattern of distribution 

 in the 250-100 m. layer is not so clear. Between 0800 and 1600 hrs. the numbers are low and the 

 major part of the population must be below 250 m. Between 1600 and 0800 hrs. there is a considerable 

 variation but there are two slight peaks between 1600 and 1800 hrs. and 0600 and 0800 hrs. and these 

 probably represent part of the population passing through the layer during their ascent and descent. 

 It is clear from this diagram that part of the population migrates upwards from a depth below 250 m. 

 but does not continue up into the lOO-o m. layer. 



Table 5. The number of observations in each ^-hourly period and 

 at each depth used in preparing Fig. 9 C. 



Time (hr.) 



There are not enough samples from i m. nets hauled below 250 m. to give helpful data, and, as 

 has been stated above, the vertical hauls with the 70 cm. diameter net do not provide adequate data 

 concerning the distribution of adult euphausiids of the size and comparative scarceness of E. tri- 

 acantha. In spite of this it was felt that some measure of the population below 250 m. was necessary, 

 however poor it might be, and these N70V data are presented in Fig. 9C. The observations have been 

 divided into six 4-hourly groups and the width of each block represents the mean catch per 250 m. haul. 

 The number of observations at each depth and in each 4-hourly period are shown in Table 5. Before 

 discussing this diagram it should be stressed that only the gross changes in vertical distribution can 

 be considered. To emphasize this point cross-hatching has been used where, before correction for a 

 250 m. haul, the total for the particular time and depth is only one individual. 



The gross daily changes in vertical distribution shown by this figure agree well enough with those 

 seen in the N looB data. In addition it is evident that, for 16 out of the 24 hrs., a large proportion of 

 the population is to be found between 750 and 250 m. There is, however, one anomaly. From 

 2000 hrs. to midnight the proportion of the population between 250 and 100 m. does not agree with 

 that shown by the NiooB; on the basis of Fig. 9C a peak would be expected between these hours 

 in Fig. 9B. This may be partly due to the oblique nets sampling the water between 500 and 250 m., 

 in which the population is reduced at this time. 



Thus it appears that during the day the majority of the population of adolescent and adult E. tri- 

 acantha is to be found between 500 and 250 m., part sinking below 500 m. for a short time around 



