Click here for the original journal page (in Acrobat pdf format).

The text below is grayed out because it is not intended to be read. It is a necessarily imperfect OCR of the original and is only used by a search engine.

.962

Journal of the Lepidopterists' Society

5

THE GENETICS AND REPRODUCTIVE ISOLATING MECHANISMS OF THE PIERIS NAPI - BRYONI/E GROUP

by Z. Lorkovic

I. Introduction

The case of Pieris napi L. and bryonise Ochsenheimer has wasted much ink in arguments among taxonomists as to whether bryonise is a distinct species or only a subspecies of napi, without having brought about a satisfactory explanation. Even after the extensive work of Muller and Kautz (1938), a turning-point in the opinions on this problem, the situation remained rather unchanged for a more critical or biologically trained judge, familiar to some extent with the circumstances in this difficult group. This failure is mainly due to the fact that most early investigators as well as Muller and Kautz themselves have not used the genetic approach. The research of B. Petersen (1948-1955) offered a first refreshing exception, and his important conclusions anticipate some of those reached in the present paper. Nevertheless a more exhaustive genetic analysis of the napi - bryonise complex remained necessary. In fact, at the same time as I was about to publish my genetic research obtained some years ago, several articles of Bowden appeared (1955, 1956, 1957, 1958) containing also some genetical and experimental data about the questions involved, though a conclusive genetic survey has not been given. Besides this, his results differ in some respects so much from my own that it will be by no means unnecessary to show here the results of my investigations in the usual genetic way.

The Problem. The main attractive point for the taxonomist lies in the great color differences between the females of these two forms as well as in the geographic restriction of bryonise to the Alps, Carpathians, Scandinavia, and to some less known parts of Asia and North America. On the other hand, napi is widely distributed in both the plains and the mountains of the Palaearctic and Nearctic Regions but usually does not appear in the areas of bryonise. Because of its restricted boreal - alpine distribution and the occurrence of more or less common transitions towards napi, bryonise was in the past considered to be an ecological race of P. napi. However since Muller and Kautz have in their extensive work offered certain arguments in favor of the specific distinctness of the two forms, their view has gained more and more adherents among lenidopterists, so that finaly Forster and Wohlfart in their work Die Schmetterlinge Mitteleuropas (1955) quoted bryonise as a good species,

6                                          Lorkovic: Pieris napi — bryonige                       Vol. 16: no.l

although the problem has not reached its conclusive stage. My use here of the term "adherents" suggests that we are not concerned with a systematically simple case, but rather with one that is in some respect ambigous. Such cases of controversial opinions have much interest from the evolutionary point of view, because this fact itself suggests that we are dealing with an incomplete stage of speciation, which deserves a closer analysis, especially from the genetical and populational aspect. The present paper refers to my research concerning the genetics of the morphological characters of napi and bryonise, to the occurrence of recombinations of these characters in nature, to the possible and experimental hybridisation and to the degrees of reproductive isolating mechanisms between the two forms.

II. Genetics of Color Differences between P. bryonise and P. napi

Materials and breeding methods. The crossings were performed with the ssp. neobryonise Sheljuzhko that is abundant in the Alpine parts of western Yugoslavia and with napi from the lowland districts of the environments of Zagreb, where no bryonise occur. Some crossings were also obtained with Alpine napi captured in districts where they fly together with bryonise (Krnica Valley, Ratece, Planica). During 1947-1951 about 200 broods were reared to imagos and more than twice as many matings have been carried out. Previously (1926, 1931, 1954) broods from wild females occasionally captured at the lower situated localities of bryonise (Fala, Savinja valley, Krnica below Prisojnik) served rather as an orientation for further decisive research.

The matings were obtained always under the experimentator's control either inside the laboratory, mostly at the windows, or outdoors in the. net, sometimes also free in the field. Many pairings could be realised only by the authors' artificial copulatory method using clamped females and immobilised males (Lorkovic 1948, 1952). The mating usually lasted 1 to VA hours. The egg-laying females were kept in cages provided with a cotton net and the eggs were deposited on the potted food plant Roripa silvestris, where the larvae remained until the last molt. Subsequently the larvae were transported to plants kept in water. The plants must be changed every one or two days whether they are eaten by the larvae or not. By this way the losses caused by diseases were reduced practically to zero. The pupation occurred on the walls of the cage or in cardboard boxes, where the larvae were placed after they had stopped feeding. Emerged male imagos usually were kept in cardboard boxes about 15 X 15 X 30 cm for two or three days at a lower temperature (15°

1962

Journal of the Lepidopterists' Society

7

- 18°C) wherefrom they were liberated once daily for some minutes for feeding. In such a manner the males can be kept in good condition before they become ripe to mate; this usually takes two or three days before one may with great probability count upon the mating ability of a male. On the contrary, the females are able to mate as soon as their wings grow strong enough for flying. Older females are more convenient only in the case when the artificial pairing methods must be used.

Genes affecting the wing colour of hryonise and napi. The first clear and doubtless assertion about the inheritance of the main bryoniae character, i. e. the dark melanic markings along the wing veins in bryoniae females (these are missing in napi) has been expressed in Petersen's paper (1955), where on the strength of my unpublished research he reported that the bryoniae character is dominant over that of napi. A short note appeared previously to the paper mentioned in the Proceedings of the IX. International Entomological Congress (1952), where I declared in the discussion of Hesselbarth's report that the bryoniae pattern is dominant, whereas the inheritance of the brownish yellow color is intermediary. All other previous opinions about this matter, so far as is known to me, were uncertain, which is no wonder, for without reared F2 generation or back-crosses such a decision could not be reached. In fact, nearly all previous breeders, as well as some recent ones, suffered great losses caused by diseases, owing to the faulty breeding methods, so that the poor F2 broods were not sufficient for an accurate genetical analysis. One exception was Main's crossing between 1907 and 1909, but in spite of a satisfactory breeding method he reached a quite erroneous conviction "that the inheritance of bryoniae characters passed almost entirely through the female" (quoted after Bowden, 1956). After this failure it was not earlier than in 1956 that Bowden published the first successful research on this matter, and his results agree in general with my own. In respect to the lacks in the genetic interpretation of the crossings in his paper I shall give here a genetical analysis of my crossings in connection with the question of the reproductive isolation of these two butterflies.

The following crosses have been carried out by the author:

napi <S X bryoniae 9                              Brood

1.   1932: c? Podsused (Zagreb); 9 Rogovilec, Savinja

valley, Karawanken Alps         R X P

2.   1947: cT Maksimir (Zagreb); 9 Vrsic, Julian Alps,

1400 m.            Ill

3.   1948: cf Maksimir (Zagreb); 9 Vrsic, ex larva 1947                1 bn

Lorkovic: Pieris napi — bryonise                       Vol.16: no.l

bryonise tf X napi $

4.   1935: d Mangart, Julian Alps; $ Podsused (Zagreb)            M X P

5.   1948; cT Vrsic, ex larva 1947; $ Pericnik, Vrata valley,

Julian Alps, 900 m.           3 nb

6.   1948: cf Krnica, Julian Alps, 1100 m.; $ Mt. Sljeme

(Zagreb), 600 m.          (27)5

Besides the listed crossings, several bryonise broods from various localities of the Julian Alps and the Karawanken Alps in the Yugoslav part of the southeastern Alps have been carried out, i. e. from: Zirovnica, 700 m.; Fala near Maribor, 300 m.; Rogovilec in Savinja valley, 650 m.; Vrata valley in the Julian Alps, 800 m.; Vrsic and Mojstrovka in the Julian Alps, 1400-1800 m.; and Bovec in the Trenta valley, 600 m. P. napi were mostly used directly from nature, since napi is the species of the environment of Zagreb, ranging eastwards to the line Maribor -Zidani most approximately, where the last small populations of a light bryonise can be found.

The most important genetic results were obtained from the cross "III" which could be brought up to the 14th generation without introduction of any new bryonise blood, but with 7 refreshments of outdoor napi.

The recent analysis of these crossings confirmed the author's previously mentioned remarks about the inheritance of the main bryonise and napi characters. The three taxonomic distinctions between these two forms are controlled by three independent pairs or groups of genes: 1) An autosomal gene pair controls the extension of the dark melanic color along the wing-veins; its dominant allele, B, causes the bryonise females to be heavily infuscated along the veins, whereas with the recessive allele, b, characteristic for napi, the darkening along the veins is either absent or reduced to the outer (distal) part of the veins (Fig.l $ $ ). In the male sex the manifestation of these alleles is only slight, it is similar in both forms and therefore markedly sex-controlled. 2) The brownish-yellow ground color of bryonise females, and the white one of napi controlled by another pair or group of genes, Y, y, and its expression is strongly sex-controlled, since the males are completely white. 3) A third pair of alleles, W, w, affects the white or greenish-yellow ground color of the underside of the hindwings and the apex of the forewings, especially in males, and is thus partially sex-controlled.

We shall now turn to the more detailed analysis of each of the mentioned gene-pairs.

1962

Journal of the Lepidopterists' Society

9

Fig. 1. Melanic wing color pattern in females and males of bryonise (left) and napi (right) which are controlled by the allele pair B and b.

The B, b alleles. The main controversy concerning the allelic pair Bb which controls the principal feature of bryonise against napi is the question of its dominance. Because of the fact that the bryonise X napi hybrids show a less pronounced bryonise pattern than the pure bryonise it was believed earlier that this character behaves intermediately. Such opinions rest on a misinterpretation of the meaning of dominance. This term does not denote full identity with the parental phenotype, respectively the homozygous one, since there are all transitions between the complete dominance to none at all. In our case, by pairing homozygous bryonise and napi, the Fi generation of both reciprocal crossings is always of the bryonise pattern, although a somewhat slighter one. In the F2 or any later generation, when individuals heterozygous for the genes Bb are mated, bryonise and pure napi segregate in a 3 : 1 ratio. In back-crosses the segregation is likewise clear, and the ratio is 1 : 1. In Fig.2 the succession of 14 generations obtained in the stock "III" is represented diagrammatically. The genotypes of each brood or of the parents as

BB,Bb

10

Lorkovic : Pieris napi — bryonise

Vol.16: no.l

8. 27 VIII.

9 4. IV 7- 1949.

10.     7. V

11.     9. VI.

12.   1QVII.

13.    (5.VIII.

14 5. IX 4.IV1950

MAKSIMIR tf $ VR§|6 l_

MAKSIMIR tf

KRNICA <j>

C? KSAVER

|                  33T30               I

tf $ cW99 d1 $

|<^Hl4|                                |7/lll4|

, 5?n \i^ 1 (fd1??^?? cV9?

|(V'ii3l                |4/in

I 3^ I V*6-----

12/111 .

r

5LJEME (5» $

I 4/Mlx I

1J = 3

I TTi I

•" $ t? 9

i-

15: 14-

? a* 9 d» $

d*

--------1

|3/HI7j

t? £ cf 9 c?1 ? (? 9 «ftf £ 9

IV'»+I I                          27:20                           1

tf $«? ? c?9 d1 9 ? d*

|V"jI |2/"'irl 5/IH5I |4/i»sl ™ M 7M I        I 0(5 ir~6T2------1

<? ¥ d"*V$9 d" 9 tf**o*9

I 4^0 I I              9-4-

c^fcf^d'©**^ 9

|2/niei |P/iiie|             i^'al i Vim

r~S/T23-8" I I ' |\ I itf.21 I

c* ^cf?9t c*o^\V?o*9

1 44^0 iff ¥

I<y'"81 l Vug I

GIELp Y_L

Rt^

STENJEVEC cT I 154

I          35 0 I

MAKSIMIR ff ? c* «f ?

M85 1                |1B2|

MAKSIMIRI -1514 I

a* ?          2d1

I 205 1                  I 20<5 I MAKSIMIR

1 I 40-0

. «T $ ©* f f o71 '"8

184)                      L25J |9£J         I 99 I

fl8^o1            I 27: 0 I I 10-0 I I 13:0         I

f •* f «" ? «V? «7* $

M49I             M50I M57 I

I 4.0 I         ' 90 I I 32 I

17 20 I I 12 12 I

tffcfS cf ? o*9

Fig. 2. Diagrammatic representation of 14 successive generations in the stock "III" of the napi $ X bryonise 9 crossing. During this time only napi were introduced from outside. The actual ratio of phenotypes is indicated for each brood; the expected genotypes are indicated by black (BB), black and white (Bb) and white (bb). Male and female symbols show approximately the given ratio. The bold black lines indicate broods in which all individuals were of bryonise phenotype, consequently at least half of them were homozygous.

indicated in the diagram result from actual ratios obtained in the progeny of each brood as well as by the symbols of male and female phenotypes. The four possible ratios which occurred in 54 broods are summarized in Tables 1 — 3, each brood containing not less than 15 individuals; broods with a lower number of descendants are not included, except a few which are of special interest.

1962                                Journal of the Lepidopterists' Society                                     11

12                                    Lorkovic: Pieris napi — bryoniae                    Vol.16: no.l

The Chi-squared test and the probabilities (P) of the ratios 3 : 1 and 1 : 1 convincingly prove that the difference between the bryoniae and napi pattern depends on only one single pair of alleles and that the bryoniae pattern is dominant over that of napi. Accordingly, the napi individuals are always homozygous, bb, and consequently pure-breeding; bryonise can be either homozygous or heterozygous. It can be seen from the diagram of Fig.2 that the mother of the stock "III" was heterozygous, since her progeny segregated in 6 bryoniae and 3 napi, a quite unexpected fact for a high-alpine bryoniae, usually considered to be true-breeding. It must be pointed out that in all these broods a clear-cut discontinuity between the bryoniae and napi patterns was absolute, although bryoniae vary considerably. The best but by no means the only decisive diagnostic character which allows the clear distinction between bearers of B and b has been found in the so called "margin-streak" ("Saumstrich" of Muller

1962

Journal of the Lepidopterists' Society

13

and Kautz or Bowden's "bryo-streak"), i. e. the dark suffusion along the rudimentary vein Ai, extending from the posterior discal spot to the outer wing margin (Fig.l, ms). This streak is so rarely to be found in napi, that when it is present it is more than probable that it originates from a previous mixture with bryonise. It is present nearly always not only in each similarly patterned subspecies or species as adahvinda Fruhst., bicolorata Petersen, pseudobryonise Vrty., camtschadalis Rob., and ochsenheimeri Stgr., but in Pieris melete Men. too. Spring napi females which show a more marked gray suffusion along the veins than is usual in this species are also distinguishable by the lack of this streak. The distinction between a slightly marked heterozygous bryonise female and a more strongly marked napi would be sometimes difficult without the bryo-streak present in the former and absent in the latter. In general, there is a good correlation between the vein markings and the bryo-streak in bryonise, the streak being a more reliable diagnostic character inasmuch as it is a qualitative one. One exception to this rule will be considered below. It may be also emphasized that the occurrence of the bryo-streak does not depend upon the presence of the posterior discoidal spot, since the streak can be well developed even when the spot is completely missing.

Despite the fact that napi were three times introduced from outside into the stock "III" during the first six generations (not counting the parental crossing), no particular diminution of the bryonise pattern could be noticed. On the contrary the homozygous BB individuals which appeared first in the 10th generation were obviously darker and had broader vein markings than the mother. This is also proof that the mother was, in spite of her high-alpine origin, heterozygous.

In spite of the well-proven monohybrid inheritance of the bryonise -napi pattern one could hardly explain the extreme range of variability of bryonise by only a twofold nature of its genotype. Moreover, the same phenotype can be either homozygous or heterozygous; crosses of napi with heavily dark bryonise females give rise to hybrids which do not differ from specimens usually considered as normal bryonise (brood 3/III-3), while on the other hand several stocks of homozygous BB individuals bred from heterozygous parents show a less pronounced bryonise pattern than the just mentioned primary hybrids. In respect to the clear monohybrid segregation there can be little doubt that this very variable shading of the bryonise pattern depends on polygenic factors independent of the B gene. This is most clearly shown by the observation that among the progeny of the two similarly heterozygous bryonise sisters but with different napi fathers (back-crosses), both phenotypes

14                                    Lorkovic: Pieris napi — bryonise                    Vol.16: no.l

bryonise and napi may be considerably different in the dark shading, depending on the intensity of the dark pattern of the napi father mated to the bryonise mother (brood 278). If the napi father has a dark pattern, the entire progeny which segregates into bryonise and napi, will have a dark one, too. Similarly a paler napi father gives rise to brighter individuals of both forms, bryonise and napi (Fig. 3). Although the shading varies in both crosses, the bryonise pattern is preserved, since even in the palest females the bryo-streak is fully developed. Such pale females can always be reliably distinguished from every napi individual. One can compare the differences described to more or less exposed or developed copies made from the same photograph negative.

Thus, the intensity of this bryonise character depends not only on the B gene itself but also on the genes which control the appearence of melanin in general. Apparently we have here an analogy to the expression of the dark markings in the Nun Moth, Lymantria monacha, as it was postulated by Goldschmidt (1921, 1928).

Apart from such multiple chromogene genes there are also other genes which control the extent of the dark suffusion along the veins. It is well known that this suffusion varies from very narrow lines, hardly broader than the veins themselves, to such an extent that the dusky suffusion of two neighboring veins joins together and the whole surface of the wings becomes dusted with dark scales. A good support for the opinion that this suffusion depends on separate genes has been obtained in a cross between a heterozygous cf bryonise X napi and a $ Pieris ergane female, reared in 1935. The dark pattern of P. ergane is identical to that of P. napi or P. rapse; but similarly to P. manni the spots have less-defined contours and the outer part of the forewings is often largely dusted with dark scales. In the female hybrids (bryonise X napi) d X ergane 9 the dark markings along the veins are strikingly broader on both the fore-wings and the hindwings than was the case in the father's (bryonise X napi) sisters, that show the width of the suffusion usual for bryonise X napi hybrids.

Consequently, the expression of the dark melanic pattern in bryonise would be dependent upon three independent groups of alleles: 1) the dominant allede B controls the extension of melanin along the veins, and its recessive allele b causes the absence of melanin; 2) an unknown number of additively acting genes (M, m) control the intensity of the darkening; whereas 3) a probably low number of genes would be responsible for the width of the vein-marking.

Now, after having got acquainted with all the possible genes influencing the dark pattern of bryonise females, we again must turn to the question of the dominance of the gene B. Bowden (1956) was right

1962

Journal of the Lepidoptcrisis' Society

15

in his doubt about the dominance of this gene so far as some mixed lowland bryonise - napi populations are concerned. Instead of a clear-cut distinction between the presence and absence of the bryo-streak, as was found to be the rule in our experiments, here its variability produces all transitions ranging from a common dark streak to a few scarcely visible dark scales or perhaps none. The same seems to apply also to the dark vein markings, which can be as narrow as in some more strongly marked napi specimens of pure napi populations. In such cases one cannot decide whether such a transitional specimen carries only two b genes or also a B gene. In my last crossing experiments no such specimens were available, and the crossings carried out many years ago were not complete. However, by a more thorough genetical investigation of this character perhaps very interesting detections could be done. Apparently, two possible explanations may be taken into consideration: either there exist two or even more multiple B genes with different manifestation of the bryo-streak, or the recessive gene b of these populations is not identical with the gene b of the pure napi populations, bringing about a stronger vein suffusion than commonly found in napi, so that sometimes a slight bryo-streak appears too. Such a supposition could lead to the conception that the gene b in bryonise populations is not derived from hybridization with napi populations, but that it takes part in the specific gene pool of Pieris bryonise as a distinct entity. The confirmation of such a supposition might turn our opinion in favour of the conviction of those authors who consider bryonise as a separate species. However, in the absence of experimental evidence for any of these hypotheses we must for the present emphasize the genetic proof that the napi pattern segregates from heterozygous bryonise,

Finally, it should be remembered that the intensity of the melanic color depends also upon environmental factors, especially on temperature and humidity. It is known that low temperature and high humidity lead in Pieridae to spread of melanin, so that it is more than probable that the form "concolor" Rob. is at least partially due to these environmental influences. Pupas kept in the refrigerator at the time of development gave imagos with extremly wide vein suffusion, while conversely, heat and dryness gave rise to pale and bright individuals.

Everything said so far applies only to the female sex. We must, however, also take into consideration the male sex, although it is not so important for our purposes, since here the phenotypic differences of the genes in question are only slight. Careless observers or novices mostly think that there are no differences between the males of bryonise and napi, which may be true only exceptionally for the spring brood. In the

16

Lorkovic : Pieris napi — bryonise

Vol.16: no.l

summer broods bryonise males are distinguished by gray streaks along the veins on the upperside, particularly on the hindwings, where these markings gradually diminish from the margin of the wing to half way toward the discocellular vein. Rarely the darkening is of such a width that the neighboring markings join near to the wing margin. Usually the markings are developed only as narrow "vein-streaks" as shown in fig. 1, left. In napi these gray vein-streaks are either absent cr are present as minute dark triangles at the ends of the veins (fig.l, right). In the spring brood the triangles can be prolonged about one millimeter or two at most; this is the only case when males of bryonise cannot be distinguished with certainty from the napi ones.

The vein-streaks are present in every bryonise male which carries at least one B gene, i. e. not only in homozygous but in heterozygous individuals too. This circumstance is of great value in the establishment of more precise ratios of phenotypes in segregation, because it is possible to determine the presence or absence of the gene B in almost every brood in males as well as females, so that the numbers usable in determining ratios are twice as large. The phenotypic differences of this character correspond in almost every case very well to the expected ratios of 3 : 1 or 1 : 1 (see Tables 1 and 3).

Of course, the stronger vein darking in bryonise males is present in the forewings as well as the hindwings, although in these the difference from napi is not so apparent, because the apical spot always sends shorter or longer vein-streaks towards the inside of the wing.

The demonstration that gene B, which controls the dark pattern of bryonise, is dominant over the recessive napi gene, so that pure napi traits segregate from heterozygous bryonise, is of decisive significance. It reveals that napi must be always homozygous recessive, bby while phenotypic bryonise can be either homozygous, BB, or heterozygous, Bb (Table 2). Muijler and Kautz (1938) considered specimens with bryonise pattern as pure bryonise or as its ssp. flavescens because they did not know this. They also did not know or underestimated the fact that from heterozygous bryonise, although very dark, specimens of the napi phenotype segregate which do not differ from individuals of a pure napi population. Muller even quoted from the literature some cases o( the appearence of napi specimens within bryonise broods, but he explained them as individuals accidentaly introduced with the food (I.e.: pp.36, 37). Corresponding examples in their monograph are: Tab.6, fig. 15; tab.5, fig.11; but also 6, 7, and 8, except for their yellow color. It is true that Muller and Kautz count such napi individuals as bryonise ssp. flavescens Wagner, but there is no proof whether they belong to this subspecies or

1962

Journal of the Lepidopterists" Society

17

rather are descendants of a cross with napi. It is only certain that such napi individuals belong to the population of Modling, which is composed of both flavescens and napi. Whether they belong to napi or flavescens cannot be resolved without breedings or crossings, as will be shown below. We shall learn also that bryonise X napi heterozygotes are sometimes present but concealed by the dominance of the gene B even in the so called "pure" mountain bryonise populations.

The Y y alleles. It was mentioned above that the factor for the brownish-yellow ground color of the bryonise females is dominant, although it is not sure that it is really so, since this color shows graduated intensity in both the crossings and in nature. Owing to its strong sex-controlled manifestation it is not easy to decide whether this character depends upon one or more pairs of alleles. Since the males are entirely white, without any trace of a yellowish color, the choice of males in crossings is entirely by chance when pure stocks are not at one's disposal.

The back-cross (R X P-1932) between an intense yellow bryonise Bb female from Rogovilec (see below) and a pure napi male from Zagreb, where no yellow specimens occurred, seem to be a rather decisive one: 28 yellow and 30 white females appeared, the yellow specimens being represented by all transitions from nearly the same yellow color as the mother had to the very pale one; all white females had the same white color. This cross indicates that the gene for the yellow color is likely to be incompletely dominant, Y, its expression being variable, and that there is no more than one allelic pair for this color, a rather unexpected result, since such a gradual color variability usually is attributed to polygeny. However, the characters which are dependent on the cumulative action of polygeny can by no means result in a ratio of 1 : 1, providing, of course, that the estimation of color was correct.

The genetics of this character thus seems to remain somewhat obscure, too.

The W, w pair. Especially interesting is the allele W which controls the white color of the underside of the hindwings and the apex of the forewings in the males of bryonise populations. It is completely dominant over the greenish-yellow color of the recessive gene w. The same applies to the female sex, although the gene w is manifested only exceptional by the white color, mostly by a pale buff one. The phenotypic manifestation of the recessive allele, w, is always the same in males and in females, in bryonise and in napi. W is a remarkable gene in that male sex-controlled polymorphism is very rare in Lepidoptera, as pointed out by Remington

18

Lorkovic : Pieris napi — bryonige

Vol.16: no.l

(1954). Although we do not deal here with a quite true sex-controlled dimorphism, since the difference appears in both sexes, the dimorphism is much more pronounced in the males than in the females, the latter being sometimes difficult to distinguish in respect to it.

It is worth nothing that the genetical identity of the white males and pale buff females has not been realized before the present analysis. Each of these two forms has been proviously considered to be a separate variant: ab. d* "subtalba" Schima and ab. 9 "subtochracea" Kautz. Muller and Kautz reported that "subtalba" can also rarely be found in males and females of napi, and they quote the find of a "subtalba" female in Pommerania in northern Germany. It is obvious that the white color of this female has nothing to do with the gene W, as was already suspected by Muller (1938).

Another interesting point of this gene pair is that the author has not as yet been able to obtain a brood homozygous for the gene W. As is shown in Table 4, the segregation is the rule and the ratio 1 : 1 prevails. Of the two 3 : 1 ratios, broods 2/4b and 1/bbn, the first is only apparently of this kind since the brood was raised from a back-cross, the mother of the second brood unfortunately escaped before her phenotype was noted. It is striking that 8 of 11 matings between Ww individuals were entirely sterile, whereas 3 others gave a total of only 5 individuals, 3 of them in one brood. Only the back-crosses were successful, and even

1962                             Journal of the Lepidopterists' Society                                 19

among these a great part produced no progeny. The writer is inclined to attribute this failure to inbreeding rather than to any other essential cause, except the Fj hybrid sterility, which will be discussed below.

No population of bryoniae is known in which the gene W is the exclusive allele. One of the most significant concentrations of this gene so far known to me occurs in the Julian Alps, where about 45% of all pheno-types are W bearers. According to Peter's (1950) count the proportion of the "subtalba" males in the Allgauer Alps is 19%. No white males had been recorded in the western Alps, as some authors have claimed.

Recombination in reared populations. The three gene pairs controlling the morphological features of bryoniae and napi, already discussed above, have been found to segregate and completely recombine in the F2 generation and back-crosses. The dihybrid back-cross "R X P -1932" already mentioned is especially illustrative. It yielded four female phenotypes: bryoniae pattern and yellow color (Bb Yy)7 bryoniae pattern and white color (Bb yy), napi pattern and yellow color (bb Yy)7 and napi pattern and white color (bbyy) in the ratio 14 : 14 : 16 : 14, which agrees very well with the expected dihybrid ratio 1:1:1:1. Besides these two gene groups free combinations with the Ww alleles have also been obtained, among them the formerly scantily-known combinations bYW and byW. Thus the three pairs of alleles are located in different chromosomes, all autosomes. This is only to be expected where the number of chromosomes is so large (n = 25). The recombinations are the same as certain "varieties" or "aberrations" from natural collections, already described, and it seems likely that the genetical explanation of these natural variants has therefore been found. I shall mention some examples of such natural "aberrations": bryoniae "obscura albida" (BB yy)7 bryoniae "albida" Mull. (Bbyy), bryoniae "flavida reducta" Mull. (bbYY or bbYy)7 and bryoniae "albida reducta" Mull, (bb yy). This shows at best how unsound it may be to give names to forms which are no more than recombinations of a few gene pairs.

However, before starting with the comparison of the experimental genetic results and the circumstances in nature we must briefly get acquainted with the other morphological as well as physiological and ecological differences between bryoniae and napi.

(To be continued)