196 BIOLOGY OF DEATH 



out very thorougMy the cause of this increased vigor of 

 the first hybrid generation and show that it is directly 

 due to the mingling of different germ plasms. 



The average duration of life of the Type I original 

 parent stock is 44:2 ± .4 days. The average duration of 

 life of the short-lived Type IV flies is 14.1 ± .2 days, or 

 only about one third as great as that of the other stock. 

 The average duration of life of the first hybrid genera- 

 tion shown in the F^ h Une is 51.5 ± .5 days. So that 

 there is an increase in average duration of life in the 

 first hybrid generation, over that of the long-lived parent, 

 of approximately 7 days. In estimating the significance 

 of this, one should remember that a day in the life of a 

 fly corresponds, as has already been pointed out, almost 

 exactly to a year in the life of a man. 



When individuals of the first hybrid generation are 

 mated together to get the second, or Fg hybrid generation 

 we get a group of flies which, if taken all together, give 

 the mortality curve shown in the line at about the middle 

 of the diagram, labelled "All Fg L." It, however, tells 

 us little about the mode of inheritance of the character 

 if we consider all the individuals of the second hybrid 

 generation together, because reaUy there are several 

 kinds of flies present in this second hybrid generation. 

 There are sharply separated groups of long-lived flies and 

 of short-lived flies. These have been lumped together to 

 give the "AU Fj Ix" Hue. If we consider separately the 

 long-lived second generation group and the short-lived 

 second generation group we get the results shown in the 

 two lines labelled "Long-lived Fg Segregates h," and 

 "Short-lived Fj Segregates Ix." It will be noted that the 

 long-lived F^ segregates have a mortality curve which al- 

 most exactly coincides with that of the originalparent Type 

 I stock. In other words, in the second generation after 



