Analysis of Heterogeneity in the Population 333 



represents considerable change in the environment to more optimum conditions, 

 nevertheless the change did not put into the test series the agents to produce the 

 conditions in the pattern that are not present in the parent location to bring it 

 up to the same range that is found in the species native in adjacent portions of 

 the valley of Mexico. 



These two tests show that the condition in the population in at least two 

 locations is not temporary, to be shifted about at the will of the oscillating 

 environment, but is relatively stable and the product of the present and work- 

 able factorial composition of the population, and of no other source. The 

 unchanged condition shown by the tests of the Chalcicomula colony is in all 

 respects the result of this, and if by any means there had been introduced into 

 the test series the multilineata form-factor, the melanothorax pronotal pattern- 

 factor, an extension factor, and the factor and determiner for the anterior 



11 — ^_^ 



14 ^ ^ 12 ^'_'" 



S-— -» »---3i ^ ^_:::i 



" '""^^ — re ^ ^ ^_»-- - 



^ m^'r 86 T , r ^^ 



Tm ^--« / / ,^\^..; 



\ ^--« / / /m\ \ 



— 21 / / / \ \ \ 



/ ^\ \ \ ^^-11 / / / \ \ \ 



12 4 6 9 31 44 12 8 4. 14 



Population 434, First Generation. Population 322, Second Generation. 



Fig. 156. — Census showing conditions in first and second generations in 1909 

 at Chalcicomula test colony. 



marginal spots, there would have been produced in two or three generations all 

 of the missing biotypical conditions in the pattern. 



OTHER TESTS. 



As stated in an earlier place, a test was made with the colony at Tlalnepantla, 

 at a location in the pedregal about 0.75 mile from the Chapultepec test location, 

 but with identical results, which I have not thought it worth while to burden 

 this paper with. The population in the Tlalnepantla location showed some 

 interesting differences from the other locations in the valley at that time, 

 especially in the absence of biotype 3, in which the operation of an extension 

 factor plays a considerable role, and this seemed to be entirely wanting in that 

 location, and the test series showed the essential conditions that were present 

 in the parent colony. At diverse times tests have been made for two or more 

 generations with the stocks from other locations at points in the pedregal in the 

 manner indicated. Thus the location at Puebla was tested at Esperanza in 

 1906 and 1907, Ometusco in 1907 and 1908, Tlascala in 1906, and in none have 

 I thus far obtained any but the same kind of results as here shown. 



In this portion of the study I have been careful to use only the materials of 

 this species that are well within its central range, and at no point in the range 

 employed is there anything like geographical races, so that the tests have been 



