DEATH AND THE DURATION OF LIFE 89 



degrees centigrade. Such experiments can be carried out 

 only in organisms where accidental death by infection is 

 excluded and our aseptic fruit flies satisfied this condi- 

 tion. These experiments were made by Dr. Northrop 

 and the writer, and consisted in putting a number of 

 newly laid eggs of aseptic flies on an abundance of steril- 

 ized yeast (which is their natural food) in a flask plugged 

 with cotton. These flasks were put into incubators the 

 temperature of which was kept constant within 0.2 of a 

 degree centigrade. The temperatures selected for the 

 purpose were 10, 15, 20, 25, 27.5, and 30 C. It is not 

 possible to go into the numerous precautions which it was 

 necessary to take and the many technical difficulties in- 

 volved in this investigation. The result of a large number 

 of experiments was that the duration of life of such asep- 

 tic flies was a definite one for each temperature, which 

 means that all the flies died at practically the same age 

 when kept at the same temperature. Thus, for instance, 

 the total average duration of life of such flies was 21.15 

 days at 30 C. The overwhelming majority died at that 

 age, but a few died a little earlier and a few a little later. 

 When the number of flies of a culture which die on suc- 

 cessive days is plotted in terms of percentage of the 

 original number of flies, we get that curve of death rates 

 usually given in life insurance statistics. But this curve 

 is very steep in our case owing to the fact that the ma- 

 jority of flies die at about the same time for a given con- 

 stant temperature. The following table gives the average 

 duration of life of the fly in days for different tempera- 

 tures. 



TABLE I 



Temperature, Average Duration of Life of the 



C. Fly from Egg to Death, Days 



30 21.15 



25 38.5 



20 54.3 



15 123.3 



10 177.5 -f x 



