338 The Mechanism of Evolution in Leptinotaksa 



produced in nature. It might seem as if the proper arrangement of experi- 

 ment and investigation in nature would give the proof of how the heterogeneity 

 arose, but in most of the conditions investigated, even in the simpler ones, the 

 production of a specific end-result is attained through diverse combinations of 

 processes. 



At Chalcicomula there was a precisely restricted type of pattern-system 

 which was permanent in the strain. There and elsewhere it never produced 

 the arrangements of the pattern that were found in the population at Chapul- 

 tepec, especially biotypes 1, 2, and 3. Why did it not produce them? The 

 answer is easy as to why the production of these pattern conditions was not 

 found; there were not in the race the requisite agents in its germinal endow- 

 ment to produce the reactions that are necessary to bring them into existence. 

 These I could introduce most easily into the race through the medium of cross- 

 ing. It is probable that some of them might have been produced in other ways 

 without the crossing, and so the " why " of the difference between this and 

 other locations is easy to discover. How did the locations become different? 

 That is another and most difficult question and one that can not, in any instance 

 that I have had to do with, be answered. At Chalcicomula I can never know 

 the source, condition of the original population, the events that have taken 

 place in the colony during its history, any or all of which may be of vital im- 

 portance in the production of the present condition. From a race taken at 

 some other location I might produce the same conditions by environmental 

 agencies, by elimination, by utilitarian selection; but it is no proof that the 

 conditions in the colony at Chalcicomula arose in the same way and as the 

 result of the same forces. So long as evolution was regarded as a progressive 

 series of events, of monophyletic origins, and dichotomizing schemes of phyletic 

 relationship, there might be some basis for this plausible array of causes and 

 results having necessary relationship. Even upon this basis there is only the 

 poorest of plausible suppositions of cause and effect. I have most earnestly, 

 in this investigation, in numerous instances, made effort to certainly discover 

 the productive agents of conditions found in nature. A successful termination 

 would have been of practical aid in other portions of my problem and otherwise 

 satisfying in being able to determine exactly the antecedent cause of present 

 conditions in a state of nature. In no instance have I been able to eliminate 

 some possible agents, and so restrict possibly the probable efficient factors to a 

 lesser number than at the start, but in no instance thus far have I been able to 

 attain to the desired end of a proof of the actual cause of the conditions 

 observed. 



It is possible to determine methods of producing different races — the agents 

 that must enter into the several compositions — and in this it is possible to 

 arrive at an understanding of how in nature the heterogeneity may have been 

 produced in these localized habitudinally different races. In plants and in 

 animals with low migratory capacities instances of the close proximity of 

 nearly related " species " have been described. Classic examples are the 

 Achatinellidae in the Hawaiian Islands, described by Gulick, and the species of 

 Partula discovered by Mayer and recently investigated by Crampton in the 

 Polenesian Islands. These and other instances that exist of remarkably re- 

 stricted and localized races in nature have been the basis upon which exten- 

 sive discussions are based. Gulick has presented the combined results of his 



