140 Messrs. Prout and Bacot. On Cross-breeding of [Jan. 8, 



approaching this or mottled or banded with dark; none approaching the 

 pure light form. In F 3 , 21 specimens, variation slight, follow their actual 

 parents very closely — i.e. all were intermediates. In F 4 , 54 specimens, variation 

 rather considerable in % 's, less in $ 's ; 2 ? dark, much as in F 2 ; several 

 quite as light as Fi, none pure light. In F 5 , 47 specimens, variation rather 

 considerable, none very light, 7 (4^, 3 $ ) dark. 



(2) In generation Fi all were true intermediates, though somewhat 

 variable ; a subordinate race-characteristic was perhaps adumbrated, which 

 became more pronounced in a few specimens of each brood from Ai to A$, 

 namely, a tendency to darkening in the outer area of the wings (a common 

 characteristic of some species of Acidalia, such as A. politata, etc.). In 

 F 2 -F 6 the variability increased, but according to no fixed rule; in F 6 (20 

 specimens) the range was from almost pure light (very weakly marked) to 

 almost pure dark, with intergrades. Two pairings were obtained in F6, one 

 of light x light, the other dark x light ; the former yielded, as F 7 , 47 

 specimens of remarkable constancy, most of which might be called pure 

 light, three perhaps light-intermediate ; the latter yielded, as F 7 (from fine 

 dark $ x light $ ), some half-dozen specimens only. F 8 , from the former of 

 the two F 7 broods, consisted of 13 pure light and one (J 1 ) intermediate; 

 another F 8 , from stock (of the latter of the two F 7 broods), also of 14 

 specimens, differed strikingly from its cousin brood in tone, all being inter- 

 mediate, and somewhat variable. F 9 was again duplicated, one batch 

 (labelled A ix 0), from light parents, consisted of 81, the majority light, but 

 only about 30 per cent, pure light ; the other batch {A ix vfc), also from light 

 parents, gave 6 only, all pure light. Fi (ex. A ix ©), 21 specimens, ranged 

 from light (1 or 2) to true intermediate (2 or 3), the majority light but very 

 slightly dusted. 



Brood B. — Carried only to the fifth generation (strictly speaking, the 

 fourth generation ; but, as shown in our " scheme," it has been thought better 

 throughout to give uniform numbering to collateral lines, i.e., to regard as F 2 

 the grandchildren of the original stock even though with strain B the 

 crossing of the two races only commenced a generation later, with strain C 

 two generations later, and so on). In generation F 2 there were 14 specimens, 

 variation inconsiderable, all intermediate, lines rather weak. In F 3 , 34 

 specimens, variation not great, similar to parent brood, but 3 or 4 distinctly 

 light — beginning to " throw back " towards ? grandparent. For F 4 a 

 duplicate pairing was obtained ; one brood, 31 specimens, averaged distinctly 

 paler than F 3 , several closely approaching the pure light ; the other brood, 

 52 specimens, also varied little, but more closely resembled the parent brood, 

 only 3 or 4 ( ? 's) being whitish the rest intermediate. In F 5 , from the 



» 



