173 The Mechanism of Evolution in Leptinotaksa 



In Chapter IV, I have sho^^oi that the rate of development, the determiner, Ac, 

 and its values had a role in the production of certain interesting results in the 

 crossing of these species, and examination of the records of different matings 

 gives data which show that the same agent is in some respects concerned in the 

 production of the degree of sterility in these crosses. The data given in table 22 

 are drawn at random from the records of the crossing of the different species 

 and the stock matings, under the same set of cultural conditions, and in that 

 time the species attain the same Ac values, about 60, I find that in strains of 

 signaticollis in which the Ac value is artificially kept at about 40, that the degree 

 of fertility in crosses is lower and approximates the initial condition on being 

 taken direct from nature. 



Added data in the same line comes from the testing of one of the races that 

 arose from the crossing of signaticollis and undecimlineata described in Chapter 

 IV. This curious undecimlineata race arose as a pure-breeding one in F^ and 

 was characterized by its excessively slow ontogeny and long life-cycle, irrespec- 

 tive of conditions in the medium. This race I crossed repeatedly with signati- 

 collis having the Ac values between 40 and 60 as a test reaction to break it up 

 and so determine its composition, in which the crossing was successful in 11 out 

 of 131 trials, and then only with the Ac value at 60, never at 40, but even at 

 Ac 60 the cross is sterile. 



From these data, gathered by the wayside as the by-products of other experi- 

 ments, it is shown that the activities whose presence and action is symbolized 

 by Ac and its value expressed in elapsed time is an agent in the gametic complex 

 whose action is productive of changes in the degree of fertility in the crossing 

 of these species. The data accumulated show that when the values of Ac are 

 wide apart fertility is diminished, when equal fertility is increased, but that the 

 attainment of the same values in the Ac determiner does not raise the fertility 

 to the standard found in the pure species. 



This experience is in line with the collective experience of those who have 

 spent years in crossing organisms as was well stated by Darwin, Gartner, Kol- 

 reuter, and others among the older students of these problems ; but why do the 

 species from nature become increasingly fertile after they have lived under the 

 same conditions of existence for varying lengths of time, and what is it in the 

 species that changes ? 



In the species L. signaticollis, diversa, and undecimlineata it is the Ac deter- 

 miner values that change, made manifest to us by the rate of ontogenetic 

 development under the standard conditions of the experiment, although not the 

 least evidence is to be had as to what changes in the physical composition of the 

 gametes. Numerous possible causes can be suggested, but evidence is wanting 

 that it is any of them. It would be interesting to know how general the relation 

 between this changed rate of development and the degree of fertility is, but I 

 have not been able to find any data in the literature that has any direct bearing 

 upon the problem. 



Even though the rates of development be the same as can be made in experi- 

 ment, the fertility never reaches the point found in the normal cross, so that 

 there are other factors in addition to the Ac determiner that are not altered, or 

 not enough, by the changed conditions of life so that fertility reaches the normal 

 for the species. A point of interest in this connection is the interfertility of 

 extracted Fg types, such as are derived from the crossings of these species. In 



