There was not, in my experience any great difficulty in getting the various forms to pair in captivity. Some attempts were made to see whether individuals of either type selected mates of their own type in preference to those of the other, but the results were inconclusive. There were some indications of such a preference; though, from the impossibility of judging how much of this may be due to other circumstances, I could not come to a positive conclusion on the rather meagre evidence.
Recently Schima[17] has given a careful and detailed account of all the forms found in Lower Austria which he enumerates under 14 distinct varietal names. He gives full references to previous accounts, especially to the beautiful plates lately published by Roger Verity.[18] Examination of these and of my own specimens strongly suggests that the several forms are due to the recombination of the factors I have named. Among those which I have bred are representatives of most if not all the types enumerated by Schima in addition to other curious forms. For example I have bryoniae markings on a ground practically white; the dark veins with spots almost obsolete; meridionalis on a yellow ground; the intermediate amount of black on a white ground, etc. The last-named may occur wild and I have one from Macugnaga as well as one given me by Mr. F. Gayner from Lulea (Lapmark).
To obtain really exact knowledge of the number of factors and their properties it would be necessary to repeat the work. After the beginning, I made a mistake in using British napi instead of meridionalis and the results were much confused thereby. The contrast between meridionalis and the various dark forms is much greater and classification of the types would have been therefore easier. The British form is presumably meridionalis plus the factor for the basal pigmentation. The problem is greatly complicated by the differentiation of the seasonal forms. The first point to be determined is whether bryoniae is capable of producing a second brood when it is thoroughly pure-bred, and whether such a second brood is, as I suspect, normally intermediate in character.
In the Alps generally there is no definitely intermediate population; nor I believe, is any such population met with in the north where the arctic bryoniae meets napi, but as to this I have no precise information. One curious fact, however, must be mentioned, namely that there is a population that can probably be so described with fairness established at Mödling near Vienna. This is not in any sense an Alpine locality, and does not, as I am told, differ in any obvious way from the other suburbs of Vienna. Dr. H. Przibram was so good as to send me a set taken at this place, representing a second brood, and they were decidedly heterogeneous, ranging from an intermediate form such as bryoniae fertilised by napi usually produces, to a light yellowish second-brood type with little dark pigment. There are also two actual bryoniae. Whether true napi also occur there I do not know, but I have no doubt they do. It would be well worth while to investigate the Mödling population statistically, and to breed from the intermediates which might not impossibly prove to be heterozygotes. There are also records of such intermediates being occasionally found in some parts of Ireland, in the north of Scotland, and in south Wales,[19] but I do not know of any regular colony of these forms. We can scarcely avoid the inference that one or more of the factors which make up bryoniae may be carried by these intermediates. It is not clear why their interbreeding does not produce actual bryoniae occasionally. If this occurred, the probability is that the fact would be known to collectors, at least in the British localities. The absence of true bryoniae must, I think, be taken to mean that some essential factor is absent from these intermediates.
To sum up the evidence, the facts that are clear may be thus enumerated:
1. Napi and bryoniae, or in the Italian Alps, napaeae and bryoniae frequently meet each other.
2. They cross without difficulty, producing fertile offspring.
3. But in the levels at which they overlap there is no intermediate population, and only occasional intermediate individuals.
4. In certain parts of the distribution of napi similar intermediates sometimes occur, and at one place (Mödling) they are so frequent as apparently to constitute a colony.
5. As to the genetic relations of the two forms there is no complete certainty. Indications of segregation have been observed in some cases, but there are several factors concerned and they are liable to some disintegration.