154 R. M. LAWS 



be less. The proportion of females estimated from their length to be sexually 

 immature, should therefore increase relatively to the observed proportion of 

 sexually immature females. The (smoothed) curves showing the proportion 

 sexually immature estimated by the length criterion and by inspection do not 

 show the expected divergence, and this is suggestive, but not conclusive, 

 evidence for an increase in the growth rate. 



In fish increased growth is often associated with earlier maturity and in 

 seals also accelerated growth appears to be associated with sexual precocity 



(Laws, 1959^)- 



If earUer puberty in whales is, as seems likely, secondary to an increase in 

 growth rate, it is possibly a result of lessened competition for food. There 

 will in any case be an interval of several years before its influence on the 

 reproductive rate is effective. Fig. 7 shows that the resumption of whaling 

 on a large scale in 1946-7 and 1947-8 did not have an appreciable effect on 

 the age at puberty until at least 1949-50. 



CHANGES IN THE SHAPE OF SURVIVAL CURVES 

 Owing to wear at the tip the baleen plate method of age determination can 

 only be used for the early age-groups. We can get a better idea of the 

 distribution of the older age-groups in the population if we study the ear 

 plugs or ovaries. The ear plug is better because it applies to all age-groups, 

 though there is still some uncertainty about the interpretation of the lamina- 

 tions. As regards adult female fm whales the results obtained by ear plugs 

 and ovaries are practically identical, assuming two ear-plug laminations per 

 year (Purves & Mountford, 1959) and 1-43 corpora per year (Laws, I959^ 

 and in press). For present purposes the ear-plug method has the disadvantage 

 that the first samples date only from the 1955-^ season, and I propose to 

 draw upon the ovary data for examples. 



For reasons which have been given at the beginning of this paper it is not 

 advisable to regard the survival curves as precise life table material. The 

 examples given have been selected from similar areas and are restricted to 

 the same season (January, February and the first half of March). My intention 

 is merely to show that the age composition has changed and that, even 

 allowing for the increase in the reproductive rate, recruitment is insufficient 

 to maintain the stock at the present level of exploitation. 



First, a few words on the method of constructing these curves. In convert- 

 ing the ovarian corpora counts to survival data, the age composition of the 

 samples of adult females is examined and survival (or mortality) is inferred 

 by a dynamic method (Laws, in press). All females in each sample have 

 survived to accumulate one corpus luteum or corpus albicans in the ovaries; 



