Analysis op Heterogeneity in Some Simplest Chaeactees 309 



the eastern slope of a large barranca and always moist, owing to constant seepage 

 of ground-water. The San Antonio colony was located on the west side of the 

 river above the Palls of San Antonio, and was the best drained and driest of the 

 three. All were about the same altitude and received the same rainfall. Evapora- 

 tion, however, was most intense at the San Antonio, least at the Quauhtemotzin 

 colony. Statistical determinations made by measurement and seriation were 

 used, frequently together, and the results show constant differences between the 

 three locations, between which there was little chance of interchange. They 

 have not been found to cross the elevated ridges from colony to colony, but work 

 up and down the barrancas, so there was little or no intercommimication prob- 

 able and none was discovered between them. In table 34 are given the results 

 from these determinations. It is evident that the Quauhtemotzin colony has 

 always the highest mode, that is, c is larger here than in either the Basoco colony 

 or the San Antonio colony, where it is smallest. As far as the biometrics are 

 concerned, here the story ends. Fo more information of any importance is 

 forthcoming as the result of the application of this method of study. The con- 

 ditions and any mass actions of spot c in the population are known ; the reasons 

 for these conditions can only be analyzed through experiment. At best, only a 

 plausible interpretation of the observed condition based upon a favored assump- 

 tion is possible, which is neither proven nor investigated; and to assume that 

 large water-content in the medium means much pigmentation and lesser 

 amounts mean less pigment is nonsense. This relation of water-content may be 

 an entirely correct conclusion, but by no means do the biometric data in this or 

 similar instances prove the conclusion at all. 



Tabu; 26. — Percentages of fusion or movements to produce fusion in the total popu- 

 lation of the Cuemavaca district that were observed in the three colonies at 

 Rancho Basoco, San Antonio, and Quauhtemotzin colonies with respect to 

 directions of fusion. 



Much more interesting and indicative of more significant conditions in the 

 population are the results from a census of the directions of variation of spot c 

 in the three locations. This I have shown in table 35, where the census is 

 recorded for each of the three stations at Cuemavaca for 10 consecutive periods 

 of observation. In this census a decided difference is evident in the San Antonio 

 colony, in that c X 6 is not found, and its presence is only indicated in 3 females, 

 neither of which was able to transmit the observed variation. Both were tested 

 by breeding. At the Basoco and Quauhtemotzin colonies all the variations were 

 found, but in different populations. In all, 6377 records were made at the 

 three colonies in the 10 censuses. Eegarding this as the population of the 

 Cuemavaca area, the three local populations show their differences in percent- 

 ages of the whole as shown in table 36. 



