GROWTH 225 



overlap occurs. In the female specimens figured, for example, the most advanced stage 

 (Figs. 6e-h) had a carapace length of only 9-3 mm., whereas the preceding stage figured 

 (Figs. 6a~d) had a carapace length of 12-0 mm. 



The course of development of the pleopod here outlined for M. siibrugosa is followed 

 in almost identical fashion by M. gregaria. In the final form of the pleopods no distinc- 

 tion between the two species is to be found. 



GROWTH 



In the hope of obtaining an indication of the growth rates of M. siibrugosa and M. 

 gregaria, measurements of the length of the carapace of all the specimens brought home 

 and now in the Discovery collections have been made. The measurement taken was 

 from the mid-dorsal point of the posterior margin of the carapace, where a slight 

 indentation usually occurs, to the tip of the rostrum. The measurements were made in 

 millimetres by means of vernier callipers, and such measurements as included fractional 

 parts were referred to the integer immediately below. The smaller specimens (less than 

 15 mm. in length of carapace) were first measured in half-millimetres, but later these 

 were grouped in millimetres so as to conform to the rest. For the purpose of drawing 

 the frequency curves the measurements have again been grouped into two millimetre 

 classes. 



The numbers of specimens measured were M. siibrugosa 4457, M . gregaria 1951 from 

 the Falkland Islands region and, in addition, 137 immature specimens of M. gregaria 

 from New Zealand waters. The material, taken during the course of three trawling 

 surveys and at other times, was obtained over a period of several years and, although 

 large, it is not representative of every month of the year. Tables I and II show the times 

 at which the specimens of M. subrugosa and M. gregaria were obtained from the Falk- 

 land Islands region. Table III gives the time distribution of M. gregaria from New 

 Zealand waters. It will be seen that in the Falkland Islands region scarcely any 

 specimens of either species were taken in June, July and August, and that the numbers 

 of M. gregaria in October, November and December are negligible. In the ensuing 

 discussion of the length frequencies for M. subrugosa the figures have been grouped 

 into two-monthly time intervals. The June-July period is not represented, whilst the 

 inadequate catches in April, August and November cause the curves for the corre- 

 sponding bi-monthly periods to be not wholly representative of the stock. 



MUNIDA SUBRUGOSA 



The bi-monthly length frequency figures of this species have been reduced to per- 

 centages, and with the values thus obtained the curves in Figs. 8 and 11 have been con- 

 structed. The principle followed in drawing these curves is that suggested by Wollaston 

 (1929). The mukimodal curve which would normally be obtained by drawing a smooth 

 curve through the plotted points has been split up into individual "curves of error", 

 each of which is considered to represent a distinct group in the total stock. In the 

 figures each curve of error is shown in its entirety, and thus the catch of each bi- 



