132 ON BREEDING EXPEPaMEXTS WITH LEPIDOPTERA. [Feb. 20,- 



This hypothesis, with slight modification, leads to the I'esults 

 observed in the cross g7'ossula7nata x lacticolor. If we suppose 

 that among the eggs the male-bearing 'all carry grossulariata, the 

 female-bearing all lacticolor, while in the spermatozoa there is no 

 coupling, so that we have male- and female-bearing of both kinds, 

 we shall get the following results : — 



DR 2 gives eggs I) c^ , R $ . 



DR d gives spermatozoa D d" , R d , D $ , R 2 . 



Since in fertilization a male-beaiing gamete must always meet 

 a female-beai-ing, the possible combinations of egg and spermatozoa 

 ai^e as follows : — 



(«) Egg- D d" X sperm. D $ = zyg-ote DD J $ . 



„^ I) d X „ R 2 = ";, DR d 2 . 



„ R2 X „ 1)6 = „ DR2 J. 



„ R 2 X „ R d = „ RR 2 J . 



In the combination lacticolor 2 X heterozygous S we shall 

 similarly have eggs Re?, R 2 , spei-matozoa D d' , R c? , D 2 , R 2 

 giving combinations — 



(6) Egg R d X sperm. D 2 = zvgote DR c? 2 • 



,', R c? X „ R 2 = \, RR d 2 • 



„ R 2 X „ D d = „ DR 2 d . 



„ R2 X „ Rd = „ RR2 d. 



In the mating heterozygous 2 X lacticolor 6 the eggs are 

 I) cj , R 2 J tl^e spermatozoa R d" , R 2 ? ^l^e combinations — 



(c) Egg D d X sperm. R 2 = zygote DR d 2 • 

 „ R 2 X „ R d = „ RR 2 d . 



It is now ob-^'ious that if the sex bome l)y the egg is vuiifoi'mly 

 dominant over that carried by the spermatozoon, the results are 

 exactly in accordance with the obser^'ed phenomena. The sex 

 carried by the egg is that wi-itten first in the zygote colunnis, and 

 we get — 



In group (a) DD d , DR d , DR 2 , RR 2 • 

 {b) DR d , RR d , DR 2 , RR 2 • 

 (c) DRc? RR2. 



It is possible to get the same result in other wa^'s. e. g. by 

 assuming couj)ling Ijetween the tw^o characters and the I'espective 

 sex-determinants in both eggs and spermatozoa and random 

 conjugation between them ; but in any case, if Castle's assumption 

 of coupling be taken for gi-anted, it is necessary to assume that 

 it is always the egg which determines the sex. 



Note. — The sex-hypothesis here outlined seems at first sight to 

 be at variance wdth that i^ropounded by Wilson (Journ. Exp. 



