NATURAL SELECTION—DOBZHANSKY AND ALLEN 361 
present (see Andrewartha and Birch, 1955, and Lack, 1954, for rele- 
vant information). 
Nineteenth-century authors said simply that excessive production 
of progeny was balanced by wholesale destruction in the “struggle for 
existence,” in which “famine, war, and pestilence” were the principal 
factors. Actually, things are more complex. Thus, with many 
species of birds, the number of eggs in a clutch is such that under 
average environmental conditions the greatest number of young sur- 
vives to maturity. Larger clutches produce fewer, not more, sur- 
vivors, since the parents are unable to take proper care of their 
brood. Among insects, starving females or females that develop 
from underfed larvae deposit fewer eggs than do well-fed females. 
Scarcity of food, destruction by predators, disease, unfavorable 
weather conditions, and accidents of every kind are all involved. One 
or more of these factors may occasionally be decisive in different 
species or at different times and places in the same species. Struggle, 
in the sense of actual combat, is a rare occurrence among members 
of the same species, although it doubtless exists. To give just one 
example: adults and larvae of ladybird beetles, which normally feed 
as predators on other insects, resort to cannibalism when the food is 
scarce. 
Destruction of a large proportion of the progeny certainly does 
not by itself guarantee that natural selection will take place. The 
contrary may be the case. When death or survival and production 
or nonproduction of offspring are due mainly to chance, large-scale 
destruction actually hampers selection for anything except fecundity. 
Selection as an evolutionary force is most effective where each in- 
dividual’s success or failure in life is a consequence of his over-all 
excellence or imperfection. In precisely this situation, most nearly 
approached by higher animals, the number of young produced is 
usually small and survival rates are high (Schmalhausen, 1949). 
To put it simply, in order to be effective natural selection must be 
selective. On the average, survivors must be better fitted to live 
than nonsurvivors. The survivors must be stronger, or more intelli- 
gent, or better able to get along on little food, or more resistant to 
weather, or better able to escape from diseases, parasites, or preda- 
tors. But not even all these virtues combined will improve the qual- 
ity of the progeny unless the fitness of the survivors and the unfit- 
ness of the nonsurvivors are due to their genes. This proviso is 
obviously most important in human evolution. In man, individual 
and group success is often due to better means rather than to better 
genes. 
Natural selection is, then, brought about by the survival of the 
genetically fit, not of the genetically fittest. Spencer’s “survival of 
the fittest” was an effective slogan in the struggle for acceptance of 
