203 The Mechanism of Evolution in Leptinotaksa 



has not yet been shown that in any of these simplest characters, or any of the so- 

 called unit-characters, there is a different composition where there is statistical 

 discontinuity. 



(3) Are the vaj-iations of the simplest characters delimited or undelimited in 

 the spots c, h, and d? The answer to this question has been to a considerable 

 extent given in the preceding sections. All of the variations thus far seen are 

 delimited in that they are all confined to certain well-marked directions of varia- 

 tion. These spots do not vary at random, but in conformity to certain rules ; 

 what these are need not now be stated, as it does not affect the situation. 



The delimited nature of the variations of these simplest characters shows cer- 

 tainly that they are not homogeneous or undifferentiated things, but each has 

 structure and relations that are its own, and entirely characteristic of it. That 

 is, in the areas called spot c or & or tZ there are forces and physical relations exist- 

 ing which center in the location c, h, d, and the same is true of all other color 

 areas in these organisms, which control the distribution of the pigment and limit 

 it to certain positions, which means that the underlying processes producing the 

 pigment are themselves thus localized and vary. For convenience of expression 

 it can be said that each of these spots has an organization peculiar to itself, 

 which delimits the variations which it shows, but it must be clearly understood 

 that this does not mean that the spot exists as an entity, independent of the rest 

 of the pronotum. Experience gives no indication of any such condition. All 

 that is known is the fact of the localization of the results of certain processes at 

 a fixed point on the surface of the pronotum, which there carry out certain reac- 

 tions from which, as a product, pigment is produced and distributed in the same 

 position as the processes themselves are localized. These localizing conditions, 

 as far as it is possible to analyze them, seem to be properties of the whole organ- 

 ism presenting most exact and complex arrangements. 



It is important to recognize that in these minute and unimportant simple char- 

 acters the variations, to use Elmer's phrase, are " law conforming " ; that is, are 

 orderly and not disorderly. This signifies of course a complete control of the 

 conditions in the smallest of the attributes that it is profitable at present to study. 

 These characters in organisms stand in the same relation to the organic body as 

 do angles, hardness, color, etc., in inorganic bodies. They are not entities in a 

 mosaic, but manifestations of physical and chemical interactions giving as an 

 end-result the things seen and used as a biological alphabet. 



(4) In local areas and in species as a whole, are the variations of the simplest 

 characters definite or indefinite f Whether variations are definite or indefinite 

 in Darwin's meaning of the term is a question of prime importance in the solu- 

 tion of many problems in organic evolution, especially in the formation of groups 

 and species in nature, and has been altogether neglected in recent years. 



In Darwin's conception of this response in " variation " there is a broader rec- 

 ognition of the general relationships of " variation " phenomena in organisms 

 that is found in the writings of his critics, or in any of his professed followers. 

 This aspect of the study of variation has been tested in spots c, h, and d in L. 

 undecinilineata Stal, L. multitceniata Stal, L. oblongata n. sp., L. decemlineata 

 Say, L. signaticollis Stal, and L. diversa n. sp., and in all the results have been 

 the same in principle, different only in statistical details. I shall present the 

 data of a few of the studies for consideration. 



