778 



PROCEEDINGS OF THE AMERICAN ACADEMY. 



being made in divisions of its scale, one division of which was equal to 

 0.017 mm. 



From the data thus gathered were constructed Figures 6 and 7 and 

 Tables XXII-XXVII. Table XXI is based on observations made by 

 F. W. Carpenter (see page 773). He counted the spines on the sex- 

 combs of both fore legs of forty males from each of three jars, A, B, and 

 C of the sfxth inbred generation of the A series, and also of forty males 

 from a stock culture considered " normal." The variation in number of 



Individuals 

 30 



25 



M 



20 



15 



10 



Spines 15 16 17 18 19 20 21 



24 25 26 27 



Figure 6. Variation in number of spines on the two sex-combs of one hundred 

 males from each of four different series ; A, from generation A 61 ; M, from M 31 ; 

 A T , from N 30; and A', from X 1. 



spines in each of these four lots is shown in Figure 5, and for the three 

 inbred lots combined in Table XXL That table shows also the degree of 

 correlation in number of spines which exists between the two sides of the 

 body in the sixth inbred generation of the A series. The same thing 

 is shown for the sixty-first generation of the A series in Table XXII, 

 and for the first inbred generation of the X series in Table XXIII. 

 A comparison of these three tables indicates that the correlation between 

 right and left sides is probably not increased by inbreeding, for though 



