EXPERIMENTS ON LONGEVITY 



405 



loupe seedlings, which still further 

 confirm and extend our results. To the 

 experimental situation which has been 

 described above, there has been added the 

 determination of the carbon dioxide pro- 

 duced by the respiration of the seedling 

 throughout each period of the experiment, 

 and also determinations of the total meta- 

 bolic transfer of dry matter from the 

 cotyledons to the stem and roots, and of 

 water from the purified agar substrate to 

 the seedlings. I shall not weary you with 

 a description of the elaborate technique 

 required to measure accurately the respira- 

 tion of a single cantaloupe seedling grow- 

 ing in the dark. The plant physiologists 

 present know all about the tribulations 

 incident to such endeavors. The rest of 

 you would be bored with their recital. 

 The details must be reserved for pub- 

 lication in a technical journal, but I wish 

 to present here, for the first time, in a 

 preliminary way, some results of one 

 experimental run, which involved 7 seed- 

 lings. Of these 7 seedlings 3 lived, to the 

 beginning of death, 14 days; 3 lived 15 

 days; and one, 16 days. If we express the 

 average condition of each variable for the 

 whole group in this experiment as 100 

 per cent we then have the results shown in 

 figure 10. 



The number of seedlings involved in 

 this experiment is far too small to have 

 anything but suggestive value, taken by 

 themselves. But a great deal more work 

 has been done along the same lines, always 

 with the same results in principle, which 

 will in due time be published in full detail. 

 This particular experimental run is taken 

 for illustration simply for the reason that 

 it was the first definitive experiment in 

 which the hypothesis developed above as 

 to inherent vitality and duration of life 

 was tested biochemically. 



The general upshot of all this work is 

 that those seedlings which have a rela- 



tively rapid rate of CO2 production in 

 respiration during the growing period, and 

 which metabolize a relatively large 

 amount of dry matter and of water during 

 growth, live a shorter time in total than 

 do seedlings which have a relatively 

 slow C0 2 rate, and a relatively small me- 

 tabolic transfer of dry matter and water, 

 during growth. The relatively long- 

 lived plants lived at a slower rate than the 

 relatively short-lived plants. In figure 10 

 lines are inserted connecting relative 

 duration of life with approximately equal 

 relative metabolic rates. These lines run 

 across the diagram. 



Let us now bring together in sum- 

 marized form the results of the different 

 lines of experimentation that have been 

 briefly described in this lecture. Starting 

 with the fruit fly Droso-phila melanogastef 

 and its mutant form called vestigial, we 

 have seen that the normal wild type of 

 fly dies according to a characteristic, 

 normal sort of life curve. But under the 

 same experimental — that is to say, in this 

 case, environmental — conditions the ves- 

 tigial mutant form has a quite different life 

 curve, not only in respect of absolute 

 average duration of life, which is only 

 about one-third that of the wild type fly, 

 but also in the whole shape of the curve. 



Under these conditions of identical en- 

 vironment these differences between wild 

 and vestigial flies behave in inheritance 

 exactly like a simple Mendelian character, 

 when the two kinds of flies are bred 

 together. 



At the same time, however, by modify- 

 ing one element of the environment, the 

 density of population or degree of crowd- 

 ing, it is possible to convert the normal 

 wild type curve over into the vestigial life 

 curve. 



This apparent paradox makes it neces- 



