352 



The Mechanism of Evolution in Leptinotaksa 



females, distributed in the four classes common to this biotype, and one group of 

 6 males and 1 female having some of the characters of primary biotype 3 (see fig. 

 40e). These divergent individuals were mated, the single male and a female, 

 and gave a progeny in F5 of 21 males and 26 females, which were, however, all of 

 the primary biotype 2 character. From the normal group in F4, 2 of each sex 

 were taken at random from each of the classes, mated as the parents of F5, and 

 gave a progeny of 260 males and 274 females, with the usual distribution. In 

 F5 each of the two groups were bred from and both gave in Fg from the normal 

 131 males and 92 females. In the group B, 36 males and 35 females, all in 

 both sets typical of the strain. The lines A and B were continued, giving in F^ 

 in (A) 233 males, 191 females, (B) 65 males and 60 females, all true to type, 

 and in Fg (A) gave 248 males and 279 females, (B) 131 males and 123 females, 

 likewise true to type. At the end of Fg the culture was killed off. 



Fio. 40. — Showing breeding record of a primary biotype 2 of L. mitltitwniata 

 ttirough a series of generations and tlie constancy of the pattern produced in these 

 biotypic lines. 



In this test with primary biotype 2, when reared as a group culture with 

 matings at random and perfectly free interbreeding of the parent groups, the 

 culture remained stable throughout. Only in generation 4 did the population 

 overstep the normal range, and these, separated and reared independently, 

 proved not to be a gametic change, but only a somatic ( ?) disturbance, as shown 

 by the giving of typical biotype 2 pattern all through the line (B), with no 

 reappearance of the divergent examples of generation 4. In a culture of thia 

 character there is no intentional selection and no evidence of selection of any 

 sort. The conditions of life were kept as uniform as possible and care was 



