144 Inheritance in Philosamia Hybrids 



gens for full spots were isolated and mated. It is clear that forms 

 which are homo- and heterozygous for the full number of spots may 

 appear either as 8 or as R8 larvae. It is possible that P. cynthia 

 may have two forms, one with the full range of spots and another or 

 others in which certain spots are missing ; no picture or descripti(jn 

 indicates that this is the case, but as entomologists seek " type" such 

 variations may well pass unnoticed. 



No families have been bred which produced only RS tbrms. Eleven 

 niatings have been made between RS parents in F.^,F:,,Fi, and F^ genera- 

 tions, nine of which produced all types of larvae, the proportion being 

 as follows : S 16-2 : R8 87 : P 90, or .S' x RS:P = 2i9 : 90, approximately 

 the same proportions as were found in the F^ generation. 



Review of literature. 



Lefroy at Pusa(8) evidently made experiments to segregate spotted 

 and unspotted larvae, for he writes, p. 22 : " The offspring of either has 

 been seen to be spotted and unspotted mixed. If only unspotted ones 

 are bred, the majority tend to be unspotted. On the other hand, if 

 spotted ones are bred, the majority tend to be spotted... it is possible to 

 eliminate either spotted or unspotted worms wholly. ..of worms from the 

 eggs of these moths (spotted), more than oO'/'o (but not all) were black 

 spotted." These observations were made from mass matings, not from 

 individual niatings as in this present work, hence the somewhat confused 

 statements. It would be surprising to find that Lefroy was right in his 

 statement that spotted larvae ajjpear in the progeny of unspotted forms. 

 He makes no distinction between completely and partly spotted larvae. 

 Kellogg(7 ) crossed white (plain larvae) and moricaud types of silk- 

 worm, and as a result of his work he says, p. 16 : " White is regularly 

 recessive to all the other larval colour pattern types. And white larvae 

 mated with white never produce any but white larvae." He does not 

 distinguish between a homo- and a heterozygous type and indeed, con- 

 sidering the nature of the markings, it does not seem possible to do so, 

 except by breeding and judgment from the offspring. Whilst Kellogg 

 found that white cocoon colour might be dominant or recessive, according 

 to race, the white colour of the larva was always recessive. But 

 Tanaka(ll) states, p. 24, "the quail factor is partially dominant and 

 partially recessive to the plain. These facts show that the relation of 

 dominance and recessiveness is more complex than is generally sujJ- 

 posed." The Japanese workers have made the following general state- 



