SUMMARY 481 



increasing day lengths associated with the north and south migrations. Animals remaining in one 

 latitude or undertaking small migrations are only subjected to one period of increasing and one of 

 decreasing day lengths annually. 



This hypothesis agrees both generally and in detail with the type of female sexual cycle outlined 

 above, but for males it is necessary to assume that a refractory period follows the extended winter 

 period of sperm production, as in some male birds, and that the spring migration does not initiate a 

 second period of rut. 



The evidence strongly suggests that the female fin whale is not polyoestrous, as earlier workers 

 have assumed, but seasonally monoestrous. 



20. An explanation of the assumed biannual formation of ear-plug laminations is advanced. It 

 seems likely that the biannual hormonal cycle, regulated by changes in day lengths associated with 

 the long migrations, may be responsible. Thus during each migration there is a rise in the amounts 

 of circulating oestrogens, associated with oestrus, and it is suggested that this is responsible for the 

 formation of a keratinous layer in the epithelium contributing to the growth of the ear-plug. 



21. The possibility of variation in fertility with age is examined. It is shown that with increasing 

 age there is probably a slight increase in the occurrence of multiple ovulations. Details are given of one 

 case in which 13 corpora lutea were formed as a result of a multiple ovulation, but this is exceptional. 

 In 22 out of 27 cases of multiple ovulation only two corpora lutea were formed. 



The proportion of sexually active females in different age groups has also been examined, and 

 it is shown that age changes in fertility are slight; the youngest females show slightly lower fertility, 

 which is probably maximal at ages of about 1 5-25 years, and may decline at ages in excess of about 

 30-40 years. 



Evidence is presented which suggests that up to a point fertility is directly proportional to the intensity 

 of whaling, and has increased to a maximum level above which it appears that no further increase is 

 possible in this species. The conclusions reached in this paper about the incidence of, for example, post- 

 partum conceptions, probably apply to fin whale populations in which fertility is maximal. With 

 post-partum conceptions assumed to be 18%, the average conception rate per female is 0-59 per year. 



22. The annual rate of ovulation can be calculated from the type of sexual cycle which has now 

 been established. These calculations suggest an annual rate of ovulation of 1-43, which is very close 

 to the higher estimate obtained from a consideration of the regression of corpora albicantia. Allowing 

 for possible causes of error it is concluded that the average annual increment of corpora is 1-43 ±o-i. 



No firm evidence about the rate of accumulation can be obtained from the recovery of whale marks 

 together with ovaries for 1 o female fin whales, owing to the small size of this sample and the uncertainty 

 about the age at marking. 



23. For estimating individual ages of females taken in the Antarctic the number of corpora is 

 divided by 1-43 and to the result is added 4! years to allow for the immature period. Allowing for the 

 variation in the age at puberty, the accuracy of the method is such that about 90% of females should 

 be within ± 3 years of the estimate. 



Ages determined in this way are in close agreement with the results obtained by other methods, 

 namely baleen plates and ear-plugs. When individual ear-plug ages are plotted against the corre- 

 sponding ovarian ages, it is found that 72 % of the latter lie within ± 3 years of the regression line 

 showing an exact correlation, and are symmetrically distributed; 82% He within ±4 years. 



In drawing up survival curves or growth curves based on large samples the variation in individual 

 ages becomes unimportant. 



24. Survival curves are given for area 11 females which, when plotted logarithmically, approximate 

 to the diagonal type, and this implies a more or less constant mortality with respect to age. It appears 



