ADAPTATION 359 



seen in the relation of many insects to their parasites. The 

 relative rates of reproduction at different temperatures may 

 be quite different, as in the observations of Webster and 

 Phillips (19 1 2) and others on the aphis Toxoptera graminum 

 and its hymen opterous parasite, Lysiphlebus tritici. Uvarov 

 (193 1, pp. 152-5) gives further instances. 



With more lowly organised animals the optimum is 

 probably equally or even more indefinite, and it is possible that 

 such species owe their survival to the existence of a number of 

 strains, at least one of which may be expected to thrive in any 

 likely combination of conditions. In this case variability, i.e. 

 lack of precise organisation, is required until, at higher levels, 

 the internal environment is better controlled. 



Besides the development of internal optima we may also 

 consider the optimum density for individuals of a species. This 

 is a subject on which our knowledge is still very slight. A dis- 

 cussion will be found in Elton (1930, pp. 25-35), and studies of 

 particular species will be found in the papers of Pearl (1927, 

 1932), Pearl, Miner and Parker (1927), and Maclagan (1932a). 

 From the present point of view certain broad general principles 

 are discernible. 



Until an animal has some control over its environment, 

 particularly its internal environment, it has little control over 

 its rate of reproduction, and this rate will vary in quite close 

 correlation with rapid environmental changes. This is known 

 to be true in soil protozoa and bacteria, and also of many 

 small insects (e.g. Smynthurus viridis (Maclagan, 1932)). Such 

 species undergo rapid fluctuations in numbers in the course of 

 the year, and are able to survive only on account of their ex- 

 tremely rapid rate of multiplication when conditions are suit- 

 able. There is easily recognisable in the evolutionary hierarchy 

 a tendency to lose this rapid rate of multiplication and to gain 

 an increased control over the reproductive rate. Not only does 

 the life of the individual become longer, but reproduction is more 

 under the control of internal relations. Even in small mammals, 

 where fluctuations in population-density are often extreme, 

 there is sufficient control to ensure that there is little response 

 to sudden environmental changes. The periods of the fluctua- 

 tions are measured in years rather than months. We believe 

 that further investigation of problems of this sort may show 

 that in the course of evolution the external environment is to 



