303 



The Mechanism of Evolution in Leptinotaesa 



both sexes. These conditions in the population are shown in figures lO? and 

 108. This behavior of the population in this season is difBcult to account for. 

 The season, as far as the recorded conditions are concerned, was not below those 

 of the two preceding years, and on my visits to the location in the middle and 

 end of the season nothing was discovered that would indicate that the climatic 

 conditions had anything to do with the array presented by the population and 

 the practical obliteration of the major portion of the biotypes in lines 1, 2, 3, and 

 4. I saw the second generation twice in its life— once during the time when it 



JW-73 '^-; 



^ -SSt— 31 



" -" * to/^~ '^'-^ —81 



.^fr- -^ ti W'^> '' ^ -108 



21- ' 0^ ,^*"i<- 219 46 -'W-5'fiu^ 1 



18- 5# -_3. .----...---m&L^^^^^^^^^ 



*v \ m - - 89 / -:\^.^^ 96 



\ -V ^,.|, 49 / \ '^ 83 



\ V*<^- 12 / \-:m- 4t 



31 36 l'9 \^ 



2106 Males. Total 3995. 1889 Females. 



Fig. 107.— Census of first annual generation at Puebla In 1907, showing reduc- 

 tion in array In population and also absence of Isolated groups. 



21— ^» •^' 319 ^° a|'^' ^— 291 



38 ^^^ 58 41 



2048 Males. Total 3955. 1907 Females. 



FiO. 108. — Census of second annual generation at Puebla In 1907, showing an 

 extensive reduction, as far as the array Is concerned, with Isolated groups com- 

 posed of biotypes 1 and 2 In both males and females. 



was mainly in the larval and pupal stages and again when in the adult condition, 

 and at no time was there any indication of an epidemic of disease or of parasit- 

 ism which might have been responsible for the condition found. Further, I 

 have never seen disease or parasitic attacks select out a portion of the population 

 in this exact maimer. The condition found is more probably the result of some 

 condition in the population itself than entirely the product of external condi- 

 tions, which may have been a contributing factor. Whatever the cause of the 

 condition, it is all too evident that only experimental analysis of the situation 

 and its experimental duplication can solve the problems involved. The census 



