59.57.72D 
Article XXI.— EXPERIMENTS WITH DROSOPHILA AMPELO- 
PHILA CONCERNING NATURAL SELECTION. 
By Frank; E. Lutz. 
Most of the many discussions concerning Natural Selection have not 
only been purely theoretical but have postulated that the characters under 
consideration are in each case heritable ones. As a matter of fact Natural 
Selection is one problem and Inheritance another; combined they form an 
important part of a certain theory of evolution but one may be studied 
separately just as well as the other. Natural selection as applied to Homo 
sapiens has been carefully investigated for some time by the life insurance 
companies and in the only way in which such problems can be profitably 
studied — by the analysis of the death rate in populations. There have 
also been a few actuarial papers concerning lower organisms. Some of the 
more important of these have been reviewed by Harris.! 
Ordinarily we think of natural selection as changing the average by 
killing off mainly those creatures which have a given characteristic or which 
have it in a given degree. ‘Thus if very heavy men tend to die at an earlier 
age than those who are not so heavy natural selection is acting to decrease 
the mean, or average, weight of the population. However, natural selection 
may tend to kill off both the very heavy and the very light men in such 
proportions that the mean would remain the same. Natural selection is, 
nevertheless, acting and manifests itself by the decreased variability of the 
surviving population, it being largely made up of those who are neither 
heavy nor light. It is conceivable that natural selection might favor the 
very heavy and the very light but kill off first those of medium weight. In 
that case the average weight of the population might remain the same but 
the variability of the surviving population would be greater than that of the 
original one or of the one which perished. 
Finally, the weight, in itself, of the men might have nothing to do with 
natural selection and we could suppose that height, in itself, had nothing 
to do with natural selection but if those men who were short and heavy as 
well as those who were tall and light died earlier than the rest of men natural 
selection would be acting. The basis of its action would not be weight or 
stature but the correlation between the two and the effect would be to 
increase the positive correlation. It is easy to see that there might be cases 
1 J. Arthur Harris. 1911. ‘The Measurement of Natural Selection.’ Popular 
Sejence Monthly, LX XVIII, pp. 521-538. 
605 
