54 DISTRIBUTION AND DISPERSION 01? LEPTINOTARSA. 



the basis of variation and natural selection or on the basis of " mntation " 

 and ' ' segregation in the fittest environment ' ' ? 



It is almost impossible to obtain direct evidence upon these questions; 

 hence we must depend to a large extent upon the indirect and circumstantial. 

 Aside from experimentation, which is usually not possible, evidence from 

 two sources only is available in the attempted solution of this question— that 

 from distribution and migration and that from the study of geographical 

 variation. Herein only the facts from distribution and migration are con- 

 sidered, the data and arguments from geographical variation being presented 

 in the next chapter. 



If the distribution of these species has been produced by variation in direct 

 response to environment and natural selection, we should find evidences of 

 continuity of species differentiation, as the different races or groups were 

 disseminated over the country; but if mutation and "segregation in the fittest 

 environment ' ' has been responsible for the present distribution of the genus, 

 continuity of distribution could hardly be expected, and its general occur- 

 rence would be improbable ; hence continuit}^ of specific differentiation in 

 a given group of species in its distribution over a country should give us 

 some evidence as to what the evolution of the group in question has been. 



In the distribution of the species in this genus evidences of continuity in 

 specific differentiation, correlated with natural environmental complexes, are 

 found, the series of species in the lineata group presenting a striking exam- 

 ple of this phenomenon. The most generalized species and the probable 

 ancestor of the group, L. undecimlineata , gives rise, as the result of certain 

 environmental conditions in nature and in experiment, to a form, angus- 

 tovittata, which is a close approximation to signaticollis . In this series of 

 three species tcndecimlineata inhabits the low, hot, and moist country to the 

 south and east of the Mexican highlands, angustovittata occurs as a regular 

 member of the fauna of the lower portion of the escarpment of the Mexican 

 Plateau on the east, south, and west, while higher up on the southern and 

 western sides of the escarpment signaticollis is found. Between these three 

 species there are variations of an intermediate character, both in nature and 

 experiment. It is true that angustovittata arises as an extreme variation in 

 experiment, and also, as I have observed, in nature, and that intermediate 

 stages are rare ; but this fact is indisputable, that this species arises in 

 direct response to the stimulus of a certain environmental complex, and 

 that it is a stable species both in nature and in experiment. From the 

 distribution of this series of three species- — undecimlineata, angustovittata, 

 and signaticollis — in three closely placed environmental complexes, we have 

 strong evidence for the orthogenetic evolution of species as the result of 

 response to changes of environment. If they had originated as the result 

 of mutation and if their present distribution were the result of segregation 



