18 The Mechanism of Evolution in Leptinotaksa 



PROCESSES OF EVOLUTION, 



The investigations of the past century have clearly formulated certain definite 

 evolution processes which may be characterized as ontogeny or development, 

 variation or heterogeneity, heredity, conservation, and equilibration. These 

 comprise groups of activities which are involved in the production, in deter- 

 mining the survival and final location, of the products of evolution. 



HETEROGENEITY. 



The process which has so long been characterized " variation " is distinctly 

 not a uniform one, but a collective name for the results of diverse reactions 

 whereby there are produced in natural things different degrees of unlikeness in 

 many directions, or heterogeneity. Because of this absence of unity of action, 

 it is more correct to think of the production of heterogeneity in nature rather 

 than of " variation," because heterogeneity is simply descriptive of the actual 

 results produced by many processes, while variation implies divergence from the 

 average or mean condition. This conception of variation reaches its climax in 

 Danvin's work and in the neo-Darwinian hypothesis is unduly accentuated by 

 the biometricians, but it is distinctly not applicable to the problems as under- 

 stood at the present moment. " Variation " in the physical sense implies simply 

 the finding of a range of values above and below some mean, divergence of 

 results produced by conditions external to the reaction, which cause it to 

 diverge and show error or departure from the mean or true value. Such arrays, 

 common in all physical measurements, are adequately expressed and formulated 

 according to the theory of error and arrangement in various types of frequency 

 curves. On the other hand, it is known at the present moment that the highly 

 miscellaneous differences which are found in any natural population are dis- 

 tinctly not the product in any population of one cause, but of many, and so the 

 diversity existing in any population is better described as heterogeneity. 



It is certain that within any natural population there may be produced 

 diversity not only as the result of the effect of the conditions of existence upon 

 the developing organism, but also as the direct result of the composition of the 

 organisms, and these again may interact to produce other added complications 

 in the array which the population presents, and when to this are added all of the 

 possibilities which came from the combination and recombination of the 

 numerous unlike factorial potentials in the gametes and the unknown series of 

 operations which may produce new factors in any generation, it is certain that 

 the problems of diversity in natural populations are among the hardest problems 

 to analyze at the present moment, and I have devoted in this report considerable 

 space to the attempt to analyze some of the problems of heterogeneity as it finds 

 expression in these organisms. 



In this respect the conception of the natural diversity of any population again 

 agrees with the diversity present in any non-living group. Their multifarious 

 conditions are the product of numerous unlike causes and not of a single opera- 

 tion, and out of the array present in any population two general types of result 

 seem to follow in the next generation. There seem to be distinctive conditions 

 present in the population which are not repeated in the next, and these are the 

 " fluctuations," according to the Weismannian point of view, which are 

 " somatic " in character and not " germinal," hence their " non-inheritance." 



