236 The Mechanism of Evolution in Leptinotaksa 



Two of these gave at once the results that had already been obtained in the other 

 series, giving in the sublines d and c" fraternities of decreasing range in the 

 index and increasing isolation from one another. The third mating, d" , did 

 not respond to this treatment, and throughout the test retained its range in the 

 fluctuations and showed considerable shifting of the mode during the course of 

 the four generations. The c subseries shows, at any rate, that in the population 

 there were present minor conditions that could become fixed as minor races by 

 the use of appropriate and delicate methods of analysis and isolation. There 

 were clearly present in the same fraternity those that did not respond to the same 

 treatment, and which retained the range in the index and showed a shifting 

 mode. It is clear that the condition in the original c line was not a simple one, 

 and this may be entirely responsible for the failure to respond by decreased fluc- 

 tuations to the conditions of the medium when held uniform. 



From the a line four matings were made, only one of which was found to 

 respond to the conditions of the effort and become restricted in the range of the 

 fluctuations as in other series, while the other lines preserved their original 

 range of index and the shifting of the mode, so that the result, which is not repre- 

 sented in the figure, indicated only the complexity of the assemblage, the absence 

 of environmental influences, the presence of minute fixable differences, and sug- 

 gested the existence in the strain of agencies invisible in manifestation as to 

 characters in the usual sense, but productive of unstable conditions, admitting of 

 such diversity in the range of the indices under investigation. 



The entire series is suggestive of the conditions that may be expected in a 

 large number of characters, and the findings are more in line with the conditions 

 found in the simplest color-characters than might at first have been expected. 

 From time to time other series of determinations have been made in the effort to 

 analyze these problems of " variation," and thus far with identical results. In 

 the juvenile stages similar dimensions, especially the head-breadth, have been the 

 basis of some similar examinations, with entirely similar results. It is not nec- 

 essary to add to the already rather long mass of data presented, in that nothing 

 new in principle is gained and added cases are not of interest at this place. The 

 analysis of the meaning of heterogeneity in these simplest characters has shown 

 interesting developments in this material, but further discussion and analysis of 

 the situation is best deferred until after some data of heterogeneity in systems of 

 pattern and structure have been described and analyzed ; then it will be profitable 

 to return to an analysis of the entire problem of heterogeneity in the individual 

 and the population. 



