JOURNAL OF THE LEPIDOPTERISTS' SOCIETY

Transcription

1 JOURNAL OF THE LEPIDOPTERISTS' SOCIETY Volume Number 4 IS PAPILlO GOTHICA (PAPILIONIDAE) A GOOD SPECIES C. A. CLARKE AND P. M. SHEPPARD Department of Medicine and Department of Genetics, University of Liverpool, England Remington (1968) has named the Papilio zelicaon-like swallowtail butterflies from a restricted geographical range (the high mountains of New Mexico, Colorado, and Wyoming) Papilio gothica Remington. Since the criteria used by Remington for claiming the existence of this newly named species are chiefly genetical and ecological rather than the usually used morphological and behavioural ones, it seems desirable to examine the genetic evidence more critically than Remington appears to have done. Genetical Evidence Obtained by Hybridization Experiments Remington showed that P. zelicaon Lucas and P. gothica are morphologically very similar and he also indicated that a number of specimens cannot be classified unless the place of their origin is known. However, the Fl hybrids between P. gothica and P. polyxenes Fabr. on the one hand, and P. zelicaon and P. polyxenes on the other, are distinguishable, as are thc Fl hybrids when P. bairdi is substituted for P. polyxenes. P. polyxenes and P. bairdi Edwards are much blacker insects than P. zelicaon. They show a great reduction in the amount of yellow on both wings and body. Clarke and Sheppard (1955, 1956) have demonstrated that the marked difference between the color patterns of P. polyxenes and P. zelicaon and, in fact, between P. polyxenes and the yellow European species, Papilio machaon L., is due to the presence of a single major gene which is dominant or nearly dominant in effect. The P. gothica x P. polyxenes hybrids differ from the P. zelicaon x P. polyxenes hybrids in that those involving P. gothica have a reduction in the yellow markings on the upper side of the wings and on the abdomen, even more marked than in the P. zelicaon hybrids (Remington, 1968). Consequently, it can be concluded that the P. gothica insects that have been

2 230 JOURNAL OF THE LEPIDOPTERISTS' SOCIETY Fig. 1. Male offspring from the mating P. polyxenes «X P. machaon ~. Three offspring are from a single mating, the fourth (top left) is from a second mating. Note that even full brothers can differ more from one another than the polyxenes X gothica and polyxenes X zelicaon illustrated by Remington. tested contain different modifiers of the effect of the major dominant allelomorph of P. polyxenes from those of P. zelicaon. However, genetic work that has been undcrtaken in the Lepidoptera (Ford, 1955b; Clarke and Sheppard, 1963, 1968; Sheppard, 19K}) shows that different populations of a single species and even different individuals within a population often contain modifiers having a much more marked effect than those reported by Remington. Furthermore, the F 1 hybrids illustrated by Remington show differences no more extreme and of exactly the same type as those found by Clarke and Sheppard (1955) between individuals when they hybridized P. polyxenes and P. machaon (Fig. 1). Thus, there is good reason to' believe that the differences between the F 1 hybrids reported by Remington are no greater than those usually found within a single species. Furthermore, Remington's data do not suggest that the differences necessarily apply to all populations which he has designated as being composed of P. gothica but only to those from a limited area.

3 VOLUME 24, NUMBER Sex Ratio in Hybrids Remington also supported his argument by claiming that P. gothica and P. zelicaon "have different hybrid sex ratios in their crosses with P. polyxenes, gothica X polyxenes being nearly lethal for the sex of the polyxenes parent whereas zelicaon X polyxenes had more nearly equal sex ratio although deficient for the sex of the polyxenes parent." Even if Remington's claim were validated (it is not supported by his data, the statistical procedures he used being wrong) the mere fact that sex ratios can be profoundly different in reciprocal hybrid matings (Haldane, 1922) should warn the unwary that very small genetic differences may profoundly altcr sex ratios in hybrids. Furthermore, even a cursory knowledge of Lepidopteran genetics would acquaint one with the fact that very different sex ratios may appear even between the progeny of sibs of the same sex when hybridization is involved. In fact, the extreme sensitivity of sex ratio to minor genetic differences in the parents is demonstrated by Remington's own data. Thus, the two crosses he reports between female zelicaon X male polyxenes show significantly different sex ratios among the progeny (P = 0.004, Fisher's exact test). It is because of this heterogeneity that it is illegitimate to combine the two broods as Remington has done in his Table IB for comparison with the single cross of a female gothica X male polyxenes. If the correct statistical procedure is followed we find that one of the two zelicaon hybrid broods is significantly different from the gothica brood (P = 0.(007) but the other is not (P = 0.122). Thus, there is no evidence as yet that the two types of hybrid (those involving zeliraon and those involving gothica) generate different sex ratios. Even if such evidence is eventually forthcoming this would not indicate that the two forms are genetically very dissimilar. Polymorphism for Larval Spot Color Remington noted from his limited experience of wild P. zelicaon larvae (less than 50 independent observations) that they are polymorphic for the color of their sub-dorsal spotting. Previously, we had shown (Clarke and Sheppard, 1955, 1956) that this polymorphism is mainly controlled by a single gene. More extensive sampling in Napa, Yolo and Eldorado Counties (Sheppard, unpublished) is in agreement with Remington's view that the polymorphism is widespread in lowland Californian populations of P. zelicaon. On the other hand, in all of about 20 independent observations (Remington, 1968) the larvae of P. gothica were monomorphic, being homozygous for the allelomorph producing yellow spotting. Remington suggests tbat in this respect they are unusual. However,

4 232 JOURNAL OF THE LEPIDOPTERISTS' SOCIETY although P. polyxenes is polymorphic for red spotting in some populations, as Remington states, the allelomorph responsible is often absent or at such a low frequency in P. polyxenes populations that these cannot be considered to be different from those of P. gothica with respect to the polymorphism on the available evidence. The gene frequencies in the populations of P. zelicaon and P. gothic([ so far reported are clearly different. However, since P. gothica has merely become monomorphic for a character polymorphic in other areas the difference does not indicate any profound genetic change. Parallel examples are common in the Lepidoptera in situations where there is no question of two different species being involved (Clarke and Sheppard, 1963, 1968). Even within the rnachllon group itself in North America the different frequencies of the fonds cornstocki and clarki in populations of P. rudkini Comstock may illustrate the point. Ecological Evidence P. gothica is strictly univoltine in Colorado, whereas P. zelicaon, as well as many other members of the rnachaon group, is multivoltine. This genetic difference has not yet been analyzed in detail. However, the difference is not surprising since P. gothica is from the high mountains where a single generation a year would seem to be ecologically advantageous. In fact, the difference from P. zelicaon merely suggests that both are ecologically adapted to their respective environments. Thus there are two ecotypes, as is so often found when montane or northern and lowland or southern types of a single species are compared. Examples from Lepidoptera in the British Isles which illustrate this point are the butterflies Aricia agestis Schiff., and Polyomrnatus icarus Rott. (Ford, 1945) and the moth Lasiocarnpa quercus L. (Ford, 1955a), the latter having races with a two year and a one year life cycle. P. gothica also differs from P. ZeliCllon in that thc larvae of the former appear to be found in nature only on Pseudocyrnopterus montanus (A. Gray) Coulter and Rose (five larvae and an unknown number of eggs reported by Remington, 1968) whereas several specics of Umbelliferae are utilized by P. zel'icaon. However, this restriction of P. gothica but not P. zelicaon to a single larval food plant, even if it is substantiated by more evidence, would not indicate that P. zelicaon and P. gothica are different species, for such ecotypic variation is common in the Lepidoptera and is found evcn within the machaon group itself. Thus, the P. rnachaon race from eastern England not only shows marked behavioural differences compared with the continental race but the larvae are confined to the milk parsley, Peucedanurn palustri L., whereas the con-

5 VOLUME 24, l\u!viber tinental form is not so confined (Ford, 1945). This is a striking parallel within a single species to the difference reported by Remington. All the genetic evidence presented by Remington suggests that gothica is only a minor high mountain ecotype of P. zelicaon and is not a good species in its own right. Unless evidence can be produced to show that the two forms arc sexually isolated if they come together in nature (a matter that should not be difficult to study experimentally), then the use of gothica as a specific name should be discontinued. Its continued use will only confuse and not clarify the taxonomy of the machaon group. Although genetic evidence is valuable in taxonomy it must not be interpreted in a parochial way. Only if it is considered in the light of studies on other material from other lands will it be useful in clarifying taxonomic relationships. We fully agree with Remington's (1968) statement that the machaon group "is much too complicated for grand conclusions based on scanty breeding experiments or on specimen samples from a few distant, randomly-chosen localities." Literature Cited CLARKE, C. A. AKD P. M. SHEPPARD A preliminary report on the genetics of the machaon group of swallowtail butterflies. Evolution 9: A further report on the genetics of the machaon group of swallowtail butterflies. Evolution 10: Interactions between major genes and polygenes in the determination of mimetic patterns of Papilio dardanus. Evolution 17: T he genetics of the mimetic butterfly Papilio memnon L. Phil. Trans. Roy. Soc. 254: FORD, E. B Butterflies. Collins, London. 1955a. Moths. Collins, London. 1955h. Polymorphism and taxonomy. Hcredity 9: HALDANE, J. B. S Sex ratio and unisexual sterility in hybrid animals. J. Genet. 12: REMINGTON, C. L A new sibling Papilio from the Rocky Mountains, with genetic and biological notes (Insecta, Lepidoptera). Postilla 119: SHEPPARD, P. M Evolutionary genetics of animal populations: the study of natural populations. Proc. 12th Int. Congo Cenet. 3: A NEW RECORD FOR NORTH AMERICA OF A SWALLOWTAIL BUTTERFLY (PAPILIONIDAE) While visiting with Mr. J. E. Lipes in El Salvador, I was privilegcd to examine some of the material collected by him. Among his material were three male specimens of Papilio (Gmphium) philolaus Esd. Mr. Lipes informed me that all three were taken in Texas. However, only one specimen had complete data, which are as follows : July 21, 1958, Padre Island, Port Isabel, Cameron County, Texas, leg. J. E. Lipes. Thc specimen is being donated to the Allyn Foundation eollection, Sarasota, Florida. RAYMOND J. JAE, 1286 South Umatilla Street, DenveT, Colorado.