STUDIES OF VARIATION IN INSECTS 



255 



faulty performances are due to congcjiitally inadequate equip- 

 ment, and if an individual takes unto itself as much of the en- 

 vironment as it can adopt, its differences from another individ- 

 ual which may have adopted more of the environment, are still 

 congenital and comparable since the environment has differed 

 only in so far as the congenital differences among individuals 

 have forced it to differ. (For an example of this kind of blas- 

 togenic difference, see our account of the differences in weight, 

 etc., of silk worms bred under identical conditions, Science, 

 vol. XVIII, N. S., pp. 741-784, Dec, 1903.) 



Taking now a character from the venation of the wings, we 

 measured accurately the length of the third longitudinal vein 

 from its origin at the anterior cross vein to its termination in the 

 outer margin of the wing (the coalesced fourth and fifth 

 branches of radius from the point of their intersection with the 

 radio-medial cross vein to their termination in the outer margin) 

 (see in fig. 45, from a to h) and found the following conditions : 



Fig. 45. 



In the 50 males the range in length of this vein is .96 mm. to 

 1.32 mm., the mode 1.15, the mean 1.148, the standard devia- 

 tion .59, and the coefficient of variability 5.15. 



In the 50 females the range is .87 mm. to 1.17 mm., the mode 

 1.02, the mean 1.03, the standard deviation .62 and the coeffi- 

 cient of variability 6.03. 



In this venational character the females show more variation 

 than the males. To test the relation of the variation in length 

 of this part of a vein with the variation in size of the wing it 

 should be noted that while the coefficient of variability of the 

 males for length (= size) of wing is only 2.85 the C. V. for 

 the venation character is 5.15 ; while in the females the propor- 

 tion is 2.76 to 6.03. Thus it is obvious that the venational 

 character varies not only quite independently of the size char- 



