466 DISCOVERY REPORTS 



frequency curves for the number of corpora in baleen group v females, and for the number of corpora 

 at the attainment of physical maturity. Thus, in the curve showing the number of corpora at the 

 attainment of physical maturity (Text-fig. 27), 81 % are in the range ±4 corpora about the mean. As 

 pointed out earlier (p. 392) it is unlikely that all females attain physical maturity at exactly the same 

 age, or even at exactly the same number of years after puberty. The discrepancy between the 

 estimated range of variation in a single year class and the actual range of variation at the attainment 

 of physical maturity is probably largely to be explained in this way. Also, the mean age and the age 

 range at puberty may be slightly greater than our small sample suggests; ear-plug laminations may 

 not always be formed biannually in immature females; and a further fact to be allowed for is the 

 incidence of multiple ovulations (p. 454). 



The close agreement between the expected frequency distribution of ovarian corpora, as calculated 

 from the age variation at puberty, and the actual frequency distribution at physical maturity, supports 

 the earlier conclusion (p. 384) that there is very little variation in the annual rate of ovulation and 

 accumulation of corpora. This is further evidence against the conception of a polyoestrous sexual 

 cycle in which, during an ovulatory period, a variable number of ovulations may precede that which 

 initiates pregnancy. 



It is considered that the number of corpora in the ovaries of a fin whale female, of any particular 

 age, will probably be within the range of the mean number of corpora expected at that age ±4; this 

 corresponds to an estimated age which will probably vary from the true age by up to ±3 years. In 

 exceptional cases the apparent age may differ from the true age by more than this. For instance, the 

 combined effect of an early puberty and a number of multiple ovulations would be considerable. 



For estimating individual ages the number of ovarian corpora excluding corpora atretica (p. 382) 

 and pathological bodies (p. 343), is divided by 1-43, and to the result is added 5 years, to allow for the 

 immature period. For animals taken in the Antarctic half a year is then subtracted. Thus first pregnancy 

 females taken in the Antarctic with an average of 1-43 corpora are on average estimated to be 5 \ years 

 old. The result should then give the probable age to ±3 years with about 90% accuracy. For example, 

 a female with 28 corpora in the ovaries is estimated to be 19 ±3 years old. 



In using the counts of ovarian corpora to determine the age for purposes of average growth curves, 

 population studies, etc., this individual variation can safely be ignored, because in a large sample 

 individual variations caused by early or late puberty, multiple ovulations, etc. will counterbalance 

 each other. 



Comparison with other methods 

 The method of age-determination based on the external ridges of the baleen plates which was 

 developed by Ruud (1940 and later papers) has already been referred to (pp. 335-37). Unfortunately 

 this method is limited in application to the younger age groups, and is, therefore, complementary to 

 the method based on the ovaries of mature females. Over the range of ages where the methods over- 

 lap, the latest work of Ruud (1958) suggests that the annual rate of accumulation of corpora, according 

 to the ages determined from baleen records, is about i-6. This is in fairly close agreement with the 

 present estimate (1-43^0-1 corpora per year). 



After the present work had begun a new method of age-determination was suggested by Purves 

 (1955). This arose out of studies on the physiology of hearing in Cetacea, in the course of which it 

 was found that the ear-plug in the external auditory meatus of baleen whales is of a laminated structure. 

 The core of the plug consists of a number of concentric laminations which follow the curvature of the 

 so-called 'glove-finger'. Each lamination consists of epidermal elements derived from the zona 

 corneum of the 'glove-finger'. From the presence of imperfectly keratinized epithelium in each 



