74 The Mechanism of Evolution in Leptinotaesa 



of crosses that I had there were 41 contrasting characters that were known to 

 be alternative and that had been proven to undergo metathesis in experiment. 

 This number of characters would, on the pure particulate basis, demand the 

 formation in Fg of thousands of classes of gametes and an impossible number 

 of combinations thereof. No such extensive metatheses were found, nor have 

 been found, in any organism that I know of, but there is a firm association of 

 these agents in the gametes, and only a few show metatheses in any one combina- 

 tion, and this is most important. The association of the agents, and the deter- 

 mination of their position and relation in the system of the gametic mass, receive 

 much-needed impetus from the recent developments in the study of sex-limited 

 and other characters in Drosophila by Morgan, and gives the first published, 

 clearly investigated data, upon this most important aspect of the problem. 



No portion of the entire subject is of so great import as is this point concern- 

 ing the position and relation of these agents in the system of the organic mass, 

 and the relation of these to the possible dissociations and recombinations that 

 may be produced during gametogenesis, and the number of kinds of gametes 

 that are produced. This is the basic problem in the investigation of the gametic 

 constitution of organisms in the future, and has to do with the analysis of the 

 composition of the system of the organism, the position and relations of the 

 agents, and the capacities for the dissociation, recombination, and replacement 

 under standard conditions, and also under divergent constellations of environic 

 factors. There is no reasonable doubt but that there are dissociation, recombina- 

 tion, replacement, and that in gametogenesis there are different factorial com- 

 binations produced in the gametes, such that when in combination in the 

 zygotes they produce in one or more " characters " unlike results. That there 

 is the random sorting of all the agents in gametogenesis, giving chance distribu- 

 tions of all of the agents and equal numbers of all possible kinds of gametes, is 

 so highly improbable that it is not worth any consideration at all. 



It is true that there is something acting in the production of gametes of 

 different kinds that gives wonderfully consistent and precisely repeated results, 

 and no doubt there is at the basis of the operations a series of phenomena with 

 which we are dealing in accurate operations, but of which our commonly 

 expressed conceptions are, no doubt, far from the truth. The mere fact that the 

 verbal description of the reactions and results has been crude, particulate, 

 vitalistic, or otherwise objectionable lies entirely in the wording and formula- 

 tion of the operations as we have tried to think of the mechanism " in biological 

 terms," and not at all in the operations or results — a condition that is not 

 limited to this field alone. One needs but recall the first formulations of the 

 agents operative in the field of immunity for an example of a professedly untrue 

 picturing of the mechanism and of the agents, but with no question of the 

 existence of the agents and the certainty and positiveness of the resultant 

 reactions. 



One of the most influential conceptions introduced by Mendel was the demon- 

 stration that the union of the gametes in fertilization was purely a chance one, 

 and that the resulting combinations or classes of zygotes produced followed 

 closely the expected array that results from the random union of the number 

 of classes of gametes present in equal numbers. It gave little chance for 

 " selective fertilization " and for " selection " or " choice " of any sort, pur- 



