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



Light by dark, ex F 6 (Ji). — F 7 to Fi . 

 Bark by light, ex F 7 (i). — F 8 to Fi . 

 Light by dark, ex F 7 (i). — F 8 to F w . 

 Dark by light, ex F 8 (</). — F 9 to Fi . 

 Light by dark, ex F 8 (/). — F 9 to F w . 

 Bark by light, ex F 9 (K). — Fi . 

 Light by dark, ex F 9 (k). — F w . 



Or in tabular form (see p. 139). 



The actual hiatus in the carrying out of this scheme will be seen from 

 the details which follow. The number of the dark x light crosses that 

 faded out would suggest some inherent tendency to weakness in this rather 

 than in the reciprocal cross ; yet the oldest hybrid of all was a dark x light (A) 

 and continued vigorous to the last. None of the hybrid strains was labelled 

 " B," this letter being reserved for the pure dark strain. 



In addition to these systematic crossings, a few quadroon broods and 

 complex crossings of hybrids were obtained. 



Most of the continuations of the broods were from single pairings, but 

 occasionally — as when a number of specimens had emerged simultaneously 

 and we could not be sure that they had not mated unobserved — we bred from 

 stock. The question of the influence of individual parental characteristics as 

 opposed to broader racial ones was not the least interesting in connection with 

 our work. 



It is to be remarked that the percentage of deaths in the early stages was 

 generally quite insignificant, and that many of the broods reared to the imago 

 state were so rich in individuals relatively to the fecundity of the species,, 

 that our statistics are incontrovertibly much more representative than those 

 obtained from Xanthorhoe ferrugata, where the pupal deaths were often 

 enormously numerous. 



Since it is impossible to forecast what statistics may ultimately assume* 

 unexpected importance, the following record tends to err, perhaps, in the 

 direction of over-completeness. Such deductions as we have been able to 

 draw from the mass of figures will be reserved for the next section of this 

 paper. 



Brood A. — This was started in duplicate, one of the strains being lost at 

 F 5 . The reciprocal cross (a) was not obtained, as we had no dark ? of 

 assured virginity. 



(1) In the generation Fi there were 62 specimens, all true intermediates 

 with variation inconsiderable. In F 2 , 66 specimens, variation considerable, 

 5 quite dark (perhaps less brownish than the pure Clapton race), others 



