ASSORTIVE MATING. 



239 



in the position of the head and pronotum, which would increase or decrease 

 the body length in individual cases without having any significance, and 

 would, therefore, introduce a large error into the result. The unit chosen is 

 not subject to any such error, and is a reliable index of size. 



Table 105. — Assortive mating found in L. oblongata, size being taken as the selective 



character. 



Sex. 



Normal range of variation— Class. 



I 



2 



3 



4 5 



6 



7 



8 



9 



10 



Males 



Per ct. 

 2 



Per ct. 

 I 

 3 



Per ct. 



Per ct. 

 6 

 4 



Per ct. 



1 



Per ct. 

 40 

 12 



Per ct. 

 38 



Per ct, 

 10 



19 



Per ct. 



6 



10 



Per ct. 



1.5 

 3 



Females 



Per cent of males of each class which mated with individuals in the various classes into 

 7vhich the females were grouped as regards size. 



Class of females. 



I, 

 2, 



3. 

 4. 

 5. 

 6. 



7. 

 8. 



9. 

 10. 













Class of 



males. 











I 



2 



3 



4 



5 



6 



7 



8 



9 



10 



Per ct. 



Per ct. 



Per ct. 



Per ct. 



Per ct. 



Per ct. 



F^r cf. 



P^r cf . 



P^r c^. 



P^r cf. 



90 



10 



2 



. , 



. , 







. . 



. . 



. . 



6 



70 



6 



•• 















4 



13 



71 



13 



I 

















7 



12 



74 



10 



5 

















8 



12 



76 



10 



5 



I 













I 



I 



II 



70 



II 



2 



3 









_ 



• • 



•• 



2 



13 

 2 



82 



2 



85 



10 



2 



6 



88 



2 



I 



3 

 6 











•• 











I 



90 



In the above table there is shown to be a striking correlation between the 

 size of the sexes in each pair, matings between males and females of very un- 

 like size not occurring at all in the series. Moreover, the number of unmated 

 individuals in any population is far greater among the extremes than among 

 the more mediocre individuals of the population. 



It is highly significant that extremely large or small beetles are unable to 

 find partners among the more moderate individuals, and are thus effectually 

 prevented from becoming parents to the following generation, and thereby 

 perpetuating their more or less extreme conditions, should they happen to 

 possess heritable qualities. This inability of extreme individuals to find part- 

 ners does not seem to be due to any aversion to them exhibited by the more 

 mediocre part of the population ; it may rather be attributed entirely to the 

 fact that the exremely large or small beetles are not able to effect copulation 

 with the others. In mating not the slighest appearance is found of a desire 



