308 Heredity of Melanism in Lepidoptera 



I have given all care to the technique of the experiment such as 

 profuse note taking (including many points besides those referred to in 

 this paper), frequent counting of the specimens in all stages, double 

 labelling, and so on ; and Mr Doncaster has kindly made various visits 

 of inspection during its progress, and is willing to accept it as reliable. 



In a survey of the specimens reared in the experiment, certain 

 points are worth noting: 



(1) In melanic specimens the abdomen, legs, and antennae are black 

 or nearly so ; and the wings are usually black, but in some cases have 

 a slight brownish suffusion as in (12.52); and in some cases a patch of 

 tan on the fore-wing (as in 12.11). 



Type specimens have light coloured abdomen, legs, and antennae. 

 The wing colour varies between light brownish-yellow and dark brown, 

 and many northern specimens tend to be banded, but all show the 

 normal markings, and are always easily distinguishable at a glance 

 from melanic specimens. 



(2) In both type and melanic forms the male specimens are darker 

 than the corresponding females ; there is no other sign of influence of sex. 



(3) I cannot point out any difference to the eye between homozygous 

 and heterozygous forms. 



Now if we classify the results from the point of view of the constitu- 

 tion of the parents we note that : 



(1) Type by type breeds type only, no melanic specimen cropping up 

 among the 621 type specimens bred (F). 



(2) Segregation is clearly shown in 9 families (D) where 25 per cent, 

 of type specimens appear in the offspring of melanic parents. 



(3) Five families (C) are cases of a pure homozygous melanic paired 

 with a type, giving all heterozygous melanic offspring. 



(4) Nine families (E) are cases of a heterozygous melanic pairing with 

 a type, the resultant progeny consisting of type and melanic specimens 

 in equal numbers. 



(5) The fact that the melanic specimens are usually slightly in 

 excess of the expected percentage seems to show that they are more 

 hardy and fewer die during the early stages, and this is borne out by 

 the fact that the melanic specimens usually emerge more early in the 

 year than their type brethren, and will even emerge in temperatures 

 between 1° and 5° C. 



