438 DISCOVERY REPORTS 



corpus albicans in the early stages of regression, occasionally with some yellow glandular tissue 

 remaining, in no case is such a young corpus albicans associated with a corpus luteum of ovulation. 

 This is valid evidence for a monoestrous cycle at this time, as in females at puberty (p. 416). 



The records of the incidence of diatom infection show that the ovulation which gives rise to this 

 corpus luteum occurs at about the time of the entry into antarctic waters of the so-called ' resting ' 

 females. Diatom infection has been assessed as absent, incipient, moderate, and extensive, and the 

 results are shown for 268 lactating, recently ovulated or 'resting' females in Table 21. Females in 

 which the diatom film is absent or incipient have probably been south of the Antarctic Convergence 

 for a month or less (Hart, 1935). Of lactating females 70-9 ± 10-5 % fell into one or other of these two 

 categories, and of recently ovulated females 7i*4±iS"3%. In contrast to this low incidence of diatoms 

 in lactating and recently ovulated females, only 46-2±8-2% of 'resting' females fell into these two 

 categories. This difference is statistically significant at the 95 % level. 



These figures strongly suggest that this ovulation occurs at or very soon after the end of lactation, 

 and the approximately 5 % of lactating females which had recently ovulated presumably represent 

 a premature ovulation at this time as in the humpback whale. 



Table 21. Incidence of diatom infection in three classes of fin whale females 



Table 22. Monthly frequency of recent ovulations in mature non-pregnant, 

 non- lactating fin whale females {Antarctic) 



It is now possible to test the hypothesis that female fin whales experience a monoestrous cycle at 

 the time when they are weaning or have just weaned the calf, shortly after entering antarctic waters, 

 by examining the monthly incidence of recent ovulations in the Antarctic. This is set out in Table 22 

 for a sample of 410 non-pregnant, non-lactating fin whales, taken between October and April over 

 a number of years. The proportion of females with a corpus luteum of ovulation in each month declines 

 from one-third in November to none in April. The standard errors of these percentages have also 

 been calculated, and the results are shown graphically in Text-fig. 47. Owing to the small size of the 

 monthly samples the standard errors are large, and the only samples which show a significant dif- 

 ference in the incidence of ovulations in successive months are those from January and February, 

 March and April. However, when the percentages and standard errors are considered as a whole the 

 decline from November to April cannot be disputed. 



A straight line has been fitted to these points by inspection, and extrapolated to give a value for 

 September. The justification for extrapolating for the September value will become apparent during 

 the ensuing discussion. There is, of course, no certainty that a straight line best describes the monthly 



