The female _bryoniae_ is totally unlike the others. The ground colour is a full yellow, and each nervure is thickly irrorated with a brown pigment often spreading so far as to hide the ground almost entirely in the fore-wings. The males corresponding with these females are not certainly distinguishable from those of our own _napi_. Both sexes have the green veining of the underside of the hind wing fully developed, rather more than is usual in the lowland races, but this is not really diagnostic of the variety. The first serious difficulty arises in regard to the second brood of _bryoniae_. It is stated that there is only one brood,[16] but I feel fairly sure that a second brood is sometimes produced, and that the females with a yellow ground and diminished irroration of the veins, not very uncommon in the Italian Alps in July to August, are generally representatives of it. Such insects would of course be classed with _bryoniae_ in collections. My experiments began with eggs of true _bryoniae_ females caught at about 2,500 feet early in July. These emerged in August-September as intermediates with yellow ground and about half as much black on the upper surface as _bryoniae_. They are exactly like the intermediates usually found in nature and in the light of later experience I regard them as natural F_{1} forms, and I think the mothers had been fertilised by _napi_ males, though I admit that in view of the rarity of natural intermediates there is a difficulty in this suggestion. Three of these females were mated with males raised from thorough _meridionalis_ females, and three families were produced. Two of them showed distinct evidence of segregation, some being yellow and some white with various intergrades, some being no blacker than _meridionalis_ and some ranging up to a dark intermediate type. Part emerged in the same autumn; and part overwintered, emerging as the spring _meridionalis_ or as the peculiar type which I afterwards learnt to know as the spring F_{1} form. The distinctions were fairly sharp between the several forms. But the offspring of the third female gave a series practically continuous from _meridionalis_ to the F_{1} type. The work of subsequent years gave results similarly irregular which could only be described adequately at great length. The outcome may however be summed up in the statement that there is evidence that both the yellow ground and the dark veining are due to factors, but that there are several of these and that imperfect segregation is not uncommon, producing various reduction-stages. The yellow ground may be due to one factor, and the several shades may be the result of irregularities in dominance, but the black markings when fully developed cannot I think be the result of less than three factors, one for the basal darkening, one for general irroration, and one for the margins. Probably also the enlargement of the spots is produced by a fourth factor.