Analysis of Hetekogeneity in the Population 339 



investigations in the Hawaiian Islands, with the interpretation of the meaning 

 and causes of the conditions in the popiU^ation of the Achatinellidae. In no 

 instance has the cause of the difference between the forms been determined. 

 In these instances, isolation or separation of some sort is assumed to be an 

 active and efficient agent in the production of the result. As far as the data 

 available are concerned, the isolation may as well be the result of the causes that 

 have produced the heterogeneity in the organisms as that it is the cause thereof. 

 A further unguarded assumption is that the conditions in the environment are 

 uniform, and so are not of any moment in the production of the results in the 

 population. It would have been profitable to have measured accurately even 

 some of the factors in the climatic complex in different mountain valleys 

 inhabited by these animals, and determined the identity or diversity therein. 

 I have been very desirous to find some mountain valleys that are identical in 

 their climatic complexes, but have not been able thus far to locate them. 



In the analysis of heterogeneity the problems center about the determination 

 of the factorial composition of the character or of the system investigated. 

 This lies at the foundation of the entire investigation, and without this deter- 

 mination the investigation of heterogeneity is meaningless and futile. The 

 problem is for the biologist as for the lithologist, crystallographer, or chemist, 

 the determination, first of the different internal and external agents that enter 

 into the production of a character of the simplest kind, then the relation of 

 these simplest characters in the diverse systems into which they enter. In 

 organisms, as in inorganic materials, these agents or factors are of diverse kinds, 

 in no instance bearers of anything, but determinative of ensuing or resultant 

 reactions and products by their presence or absence in the system at different 

 periods of its reaction. 



Having determined the factorial composition of any character or system, 

 one is then in position to experimenta,lly determine the laws of heterogeneity 

 through the influence of external incident agents, by metathesis, by the advent 

 or loss of factors in the system, and so to discover how heterogeneity in nature 

 may be produced, to produce it in experiment, and analytically to determine 

 the factorial differences between groups in nature. 



In this the categories of variation must vanish ; fluctuations will have to be 

 determined as to cause, whether due to states of stability, impurities, misplaced 

 factors with missing complements or loosely combined in the system and non- 

 productive of anything but erratic disturbances, or to the action of incident 

 forces. Continuity and discontinuity become terms that are merely descriptive 

 of the results of interactions of the factors, and unit-characters become obsolete. 

 In organisms there are unit-characters no more than in rocks or minerals. 



Had we the understanding of the constitution of living substance that we 

 have of the nonliving, it would be easy to isolate and determine the precise 

 nature of any or all factors, to take the entire organism apart and reassemble 

 it in the laboratory. At present no one factor is determined, nor has one been 

 isolated, in a chemical sense. 



We thus come to regard the heterogeneity of organisms in a new light, phys- 

 ical in expression, dynamic in action, nonvitalistic in conception. The 

 methods of investigation change from plausible arrangements to analytical 

 determinations of factorial constitution, action, and result, and in experi- 

 ment, as in nature, the responses in heterogeneity, whether definite or indefinite. 



