J. W. H. Harrison 251 



becomes blackish in extreme cases and suffused in those less so. 

 Naturally the acquisition of a wild female^ of this form, the first I had 

 ever taken in the birch wood, caused me to speculate as to its genetic 

 potentialities, so I paired it with a medium example of the barred form 

 reared indoors. Anticipating with some confidence its dominant nature, 

 I looked forward to an F^ brood entirely of latifasciata form and including 

 the first males of the type ever known. My expectations were utterly 

 falsified for in 1917 in a brood comprising 47 individuals I obtained 

 22 perfectly typical females without the faintest sign of a complete bar 

 and 25 males, all of which displayed the bar in the diluted form, showing 

 incomplete dominancy. This, prima facie, was evidence of that type of 

 sex-linked inheritance known as "criss-cross." Consequently, in addition 

 to pairing the F^ insects inter se, all of the crosses including this F^ 

 brood, the insect in its typical guise, and the allied species 0. dikdata 

 were made. All of the intraspecific crosses yielded fertile ova as also 

 did two of the matings of the Fi (/ and 0. dilutata $ ; on the contrary, 

 after a disheartening waste of material, none of the ova laid in the cages 

 containing i'^ $ ? and dilutata (^(f changed colour, and were therefore 

 infertile. 



My 1918 cultures therefore were the results of the following pairings : 



(1) F,^ xF,^. 



(2) Fi^ X 0. autumnata (^ . 



(3) 0. autumnata % x F^^^ . 



(4) 0. dilutata % xF^^. 



The first two of these were exceptionally successful and yielded 

 important results harmonising exactly with what was anticipated should 

 the inheritance be of the sex-linked type ; nor was there any discrepancy 

 between the results of the last two and one's expectations, only the 

 number of imagines bred in these two cases was very meagre — just 



TABLE IV. 



Number of Number Number of Number 

 latifasciata of type latifasciata of type 

 Brood females females males males 



Fj generation ex latifasciata $ x type <J — 22 25 



F2 generation. (Same cross) 18 21 19 18 



Fi$xtype(? — 43 — 36 



Type$xFi^ 2 1 1 3 



Dilutata^ xFi^ — — 2 



1 I have taken three latifasciata forms in Yorkshire and have heard of one taken in 

 Fermanagh, Ireland, and all were females ; the similarity of this with the case of Abraxas 

 grossulariata studied so exhaustively by Doncaster seems noteworthy. 



