213 The Mechanism of Evolution in Leptinotaesa 



festation of the " general property of the whole " and always in the same place. 

 Moreover, c does not come back into such a strain except as it is put back by the 

 reactions of interchange of gametic factors between a c-less race and a race 

 that has c, which may be either another homobiotypic strain or any heterobio- 

 type that has c present. 



Not for a moment h cot I ox d or any other area to be thought of as a single 

 thing or entity. It is absolutely not a unit-character in any sense. The pig- 

 ment of the spot is the product of a black determiner and a color-factor; the 

 pattern rests upon two distinct sets of agents — a general pattern-factor of the 

 pronotum and the pattern determiner of the particular area, and this pattern 

 determiner is always, except in pathological examples, precisely localized in the 

 pronotum, but from this fixed position in space it is able to, and does, form 

 different combinations by movements mutual between it and nearby areas. 

 These combinations are always specific for the species, can be fixed and isolated 

 in biotypes, and are alternative in crosses of such biotypes. For example, 

 L. diversa, angustovittata, and rugosa have no directions of fusion; signati- 

 collis forms c + h and c+e ant., but never both at the same time in the same 

 individual. L. undecimlineata and panamensis may have c wanting or form 

 c + e ant., c + d,OT c + t combinations. L. multitceniata shows the greatest num- 

 ber of combinations, oblongata only c-\-e ant., decemlineata c + e ant. and no 

 other, while juncta, texana, tumamoca, and dejecta have not thus far shown in 

 nature or in cultures any fusions or heritable directions of variation. This 

 behavior has a strong analogy to the condition of many molecules or radicles 

 where there are one or more free bonds at times, at others with no unsatisfied 

 bonds. It is not for a moment to be supposed that the reactions shown are 

 such ; only the type of action seems to have something in common in the effort 

 of the localized mass of matter to place itself in the system with all bonds 

 satisfied and no free bonds unsatisfied, which at any moment may become a 

 possible entrance-point for incident dissociative forces. 



These simplest characters, while not capable of disruption into lesser areas 

 or characters, are each the visible product of the interaction of the factors and 

 determiners above described. These are the common normal agents most con- 

 cerned in such areas. There are others present, active, and important, which 

 need not be considered here. All react to produce the character " simplest " 

 observed, and the observed product varies in space, volume, direction of com- 

 bination, quantity of pigment produced, and area occupied. Variation in these 

 characters is not entirely of any one aspect, but is quantitative or qualitative, 

 continuous or discontinuous, depending upon the aspect, method and point of 

 view, and responds in delimited changes which may be definite or indefinite in 

 the action of the observed population. 



There are very many simplest color-characters in the adult that are precisely 

 like those discussed. The head, abdominal segments, legs, and wings present 

 color-marks and arrangements of coloration that would be serviceable in the 

 investigation of this problem of the nature and diversity of simplest characters. 

 I have in the course of this work gone into the study of more than a dozen of 

 these with greater or less completeness, but always with the end-result found in 

 those presented, and no purpose would be served by presenting a larger array at 

 this place. It might be asserted that the instances chosen and presented are 



