108 The Mechanism of Evoltjtiox in Leptinotaksa 



easily changed on introduction into a new habitat, without any change in the 

 characteristics of the organism. It is, as far as I am able to decide, a condition 

 in the organism, a state of being, that exists within the mass and is the product 

 in its values of the conditions in the medium in which the organism is living, 

 acting upon and directly modifying the reaction values made possible by the 

 gametic factors of composition, and is in this respect comparable to the con- 

 ditions of the mass that one finds in non-living materials. 



The ontogenetic group of developmental reactions are immensely complex, 

 but they have clearly the relation and values in these species, which I have 

 designated the ontogenetic rate determiner {Ac), and changes in the reaction 

 value of this agent, induced through changes in the medium, produce in the 

 organic mass changes in the rate of ontogenetic progression, which are measured 

 by the length of the time taken for the completion of growth, from fertilization 

 to the adult condition. The quantitative value of Ac in any species I have 

 expressed in terms of elapsed time demanded by the ontogeny of the species, as 

 expressed in days. Thus Ac*' or Ac^", simply signifies that in the stock from 

 which the individuals came under the conditions of their culture, this factor had 

 the value of reaction, such that the total series of ontogenetic reactions would 

 be completed, on the average, in 40 or 60 days. 



This agent serves as a most important factor in disturbances of the Mendelian 

 reaction in species crosses, and changes the character, as far as visibility is con- 

 cerned, of many individuals in the population for indefinite periods of time. I 

 have shown that the two species, L. dvversa and L. signaticollis, when crossed 

 under common conditions, and when the values oiAc are the same or essentially 

 the same Ac^", give perfect monohybrid reactions in all instances, and without 

 exception showing the production of gametes that are pure for the total system 

 of each of the two species, giving true signaticollis and diversa as extracted 

 products. On the other hand, it is shown in plate 9 that when the two are 

 crossed where the Ac values are different there ensues in Fj and in following 

 generations a different series of events, in which the fate of certain individuals 

 is permanently altered, and the F^ reaction suggests the interpretation that one 

 of the parents is heterozygous, the other homozygous. 



Thus far there has been no exception to the two types of results when the 

 conditions in the medium and the values of ^Ic are those stated,- and this repre- 

 sents the condition of the crossing of the two species fresh from nature, or of 

 fresh L. signaticollis crossed with the L. diversa that has been in the laboratory 

 for any length of time. Also, the same results are produced if L. signaticollis is 

 in experiment made to attain the value of Ac^" or thereabouts, regardless of 

 what its former rate of ontogeny has been. 



In the earlier experiments from 1904 to 1907, the series described were dupli- 

 cated over and over again, and many tests were made that gave profitable 

 indications for further work. From 1907 to 1914 the reactions found, and 

 represented in plate 9, have been extensively tested and an analysis made of the 

 products. Starting with a series that is easily obtained by the crossing of 

 L. signaticollis from Cuernavaca, when freshly introduced into the laboratory, 

 and L. diversa, either fresh or in stock, the cross made under the standard con- 

 ditions of the testing-room gave uniformly 50 per cent of the progeny that were 

 of the typical mid-type, and 60 per cent that were to all appearances signaticollis 



