16 A. Comfort 



fish exhibits a very characteristic increase in mortality with 

 age, which we have been able to correlate with its growth 

 pattern and reproductive rate under various experimental 

 conditions. These experiments are still in progress. Under 

 good conditions, at 23° C, less than 10 per cent of fish in a 

 cohort have died before reaching 1 year of age, and less than 

 10 per cent exceed 1000 days in the series so far kept. The force 

 of mortality in this species, both in the females, whose rate of 

 growth depends on feeding and space, and the males, which 

 cease growing soon after sexual maturity, approximates in its 

 age distribution to that of the laboratory rat. We are collecting 

 parallel figures for the rate of fin regeneration at different ages 

 and under different conditions of growth. I quote this work as 

 an example of the type of study in which it is possible to com- 

 bine observational with experimental work and direct them to 

 attack particular problems. Investigations of this kind must 

 necessarily be confined to short-lived species, and our ex- 

 perience with Lebistes suggests that even with an unusually 

 hardy and semi-domestic organism, and even in the absence of 

 epizootics and accidents, to which such research is extremely 

 vulnerable, the minimum time necessary to complete such a 

 study is twice the life-time of the organism. 



I have not, I hope, suggested in the course of this paper 

 that fundamental studies of the processes which go on within 

 organisms, or the cells of organisms, are either not worth 

 studying or inaccessible to study. The brilliant work of 

 Sonneborn's school on the senescence of clones in Paramecium 

 is an indication of what can be done in such fields — an apt one, 

 for my purpose, because decline in vegetative lines of Para- 

 mecium is an example of an effect which was long adduced as 

 evidence of a fundamental biological law, the necessity to 

 maintain cell vigour by sexual reproduction, and which 

 turns out to be peculiar to ciliates, since it is apparently due 

 to uneven distribution of the chromosomes of a polyploid 

 macronucleus. We require a larger amount of comparative 

 information because the greater our knowledge of the pattern 

 and distribution of age effects, the less will be the risk of over- 



