ii6 DISCOVERY REPORTS 



from their attachment, so that females which have shed their eggs, generally have empty thelyca 

 (stage H). 



Stages 7 A and 7B, comprising gravid females and those which have spawned, have been combined 

 together as stage 7 in all analyses of the catch, and in the tables of minimum and maximum length, 

 and in the average length estimations. This has been done for two reasons: in the first place, the 

 number of gravid females (i.e. females at stage 7 A) obtained was very small, and in the second place, 

 combining stages 7 A and 7B makes it easier to compare the adult females with the adult males. 

 The external stages G and H have been treated similarly. But in order to describe the life history 

 as fully as possible, I have, where necessary, analysed the catch of adult females into the two groups 

 of gravid and spawned. 



Out of 200 specimens at stage 6, 6 were at stage D, i6 at stage E, 35 at stage F and 143 at stage G. 

 The smallest specimen, measuring 32 mm., was at stage E; the largest, measuring 60 mm., was at 

 stage G. No specimens at stage 6 occurred in May, June, July, August, September or October. The 

 monthly average length varied between 47 mm. in November through 50 mm. in January and February 

 to 47 mm. in April. 



Stages 7A and 7B were represented by 491 specimens, of which 57 were gravid, i.e. at stage 7A, 

 and 434 had spawned, i.e. they were at stage 7B. The time of their occurrence ranged from December 

 to April. The smallest specimen, measuring 41 mm., was found at the end of the season; the largest, 

 measuring 64 mm., in February; both were at stage G. The average monthly length decreased from 

 57 mm. in December to 47 mm. in April. 



Table 10. Occurrence of gravid and spawned females 



Analysis of the total catch of these stages shows that the majority of gravid females (48) occurred 

 in February, and that all except four of the spawned females occurred in February, March and April. 

 Attention has already beeen drawn to the scarcity of the gravid stage. Of the 57 specimens obtained, 

 40 of the large catch in February were taken from deep nets (250-100 m., 750-500 m.); 3 of the 

 remaining 17 specimens occurred in the surface layer (0-5 m.) and 14 at depths varying between 137 

 and o m. In the catches examined, the marked failure to take gravid females at the surface in any 

 number seems to point to deep spawning. The euphausian egg contains a large quantity of yolk, 

 sufficient to feed the young larva for some little time after hatching, and the external food supply is 

 therefore not of immediate importance to the larva when it first leaves the egg. Presumably, this 

 quantity of yolk makes the egg heavy, for Eraser found that the bulk of the eggs and early larval stages 

 occur at depths below 250 m. Evidence from later commissions of the R.R.S. ' Discovery II ' confirms 

 this. Eraser suggested that the eggs become concentrated in basins on the submarine ridges, where 

 the homogeneous water column provides the uniformity of temperature and density necessary for 

 their development. A large catch of 95 gravid females in a vertical net, fished between 250 and 100 m., 

 in February 1930 off South Georgia rather confirms this idea. The vertical 70 cm. net is small and 

 usually catches very few fully grown specimens of E. superba. The occurrence of 95 in one haul is 



