144 The Mechanism of Evolution in Leptinotaesa 



modes were the modes of the parent forms from which pure lines could be 

 grown, pure for the parent form-index, showing that the two parental basic 

 gametic systems did not intermingle ; but while they might react in common in 

 the heterozygote, as they must or perish, at the first opportunity they separated, 

 each into gametes that were pure for the species gametic system in question. 



In the cross between diversa and signaticollis this is all that took place, the 

 two systems maintaining their integrity throughout the series, and no indica- 

 tion of metathesis being found in the series. This to me seems the simplest 

 reaction that one can find in the crossing of two forms of organisms and is the 

 simple interaction of two systems that are sufficiently like in substance, struc- 

 ture, and reactions so that in combination they are able to react as one, pro- 

 ducing the compromise end-results of the heterozygote, which in most respects 

 are intermediate between the two parent forms, but not able to form any 

 permanent association as far as experience has shown, and between which no 

 interchanges of agents or groups of agents take place in the interactions that 

 go on during ontogeny and in the gametogenesis of F^. 



Quite different are the reactions and results in the series shown in the crossing 

 of signaticollis and undecimlineata, in which there is the same retention of the 

 integrity of the gametic system ; but in the reactions incident to gametogenesis 

 in Fi certain interchanges of groups of agents occur. The array has the appear- 

 ance of a trihybrid, owing to the fact that the three pairs of masses are present 

 and interacting: (1) The non-dissociated gametic mass, characterized by 

 form-index, elytral pattern, etc.; (2) larval pattern; and (3) larval color. 

 These present the array of three independently interacting characters. In 

 reality the elytral system is not in any of these dissociated from the basal 

 gametic complex, and is the indicator of its presence, a fact that can and has 

 been tested by breeding and by biometric measurements in the diverse lines 

 with full confirmation in all respects. That which happens in this instance is 

 the separation of the basal systems in gametogenesis into two great groups, 

 distinguished by the elytral-pattern sign, and then the interchange of the two 

 other groups of agents in metathesis, such as would in all reactions take place 

 between unlike complex substances when in mixtures and under conditions 

 that permit of interactions and interchanges between them. 



With considerable possibility of truth one can symbolize these substances and 

 reactions in one's imagination as two collodial mixtures, each a highly organized 

 system in itself, with structures and reactions such that they may for a time 

 interact in common to produce the soma of F^, but are not able to form per- 

 manent combinations. We think of this simple relation in terms of a single 

 pair of reacting groups of agents, and all of the evidence from Fa separations, 

 measurements, and breeding of extracted lines gives support to the symbolic 

 conception of the two gametic systems retaining their identity and integrity 

 throughout the operation. There is no evidence that the gametic systems in 

 this cross are impure or contaminated in any way by the passage through the 

 cross, nor are extracted stocks and parent species different in any point that 

 has been detected. 



The cross of signaticollis X undecimlineata shows in essence the same general 

 operation as in the simpler cross, with the exception that in gametogenesis the 

 two gametic systems which retain their individuality in the divisions of the 

 germinal materials into the two main types of the parental stocks, with respect 



