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.
Volume 30, Number 4
289
THE BIOLOGICAL STATUS OF NEARCTIC TAXA IN THE PIERIS PROTODICE-OCCIDENTALIS GROUP (PIERIDAE)
Arthur M. Shapiro Department of Zoology, University of California, Davis, California 95616
The taxonomy of Pieris protodice Boisduval & LeConte and its relatives in western North America has long been confused, with the relationship of that taxon to P. occidentalis Reakirt being a matter of particular contention. Both current "field guides" (Klots, 1951; Ehrlich & Ehrlich, 1961) treat them as conspecific, as does the North American check-list (dos Passos, 1965), while recent faunistic papers by authors in western North America (e.g., Garth & Tilden, 1963) have regarded them as separate species. Chang (1963) provided morphological evidence in support of the latter position and mapped the distribution of P. occidentalis. His analysis was followed by Howe (1975). The situation is complicated by the occurrence of named seasonal and/or alti-tudinal phenotypes in both, as well as one valid subspecies. The present paper summarizes the result of a series of field and laboratory studies in which the biological relationships among these entities have been clarified; these studies are cited individually in the text. It is not intended as a formal taxonomic revision; such a revision should, if undertaken, be on a world-wide basis. All of the taxa are abundantly illustrated in the papers cited.
The only previous American treatment of this group was by Abbott (1957) and Abbott, Dillon, & Shrode (1960). This incompetent study, which "synonymizes" the very different species Pieris beckerii Edwards and P. sisymbrii Boisduval with P. protodice and P. occidentalis with total disregard for their biology and extensive sympatry, makes no useful contribution to the taxonomy of the group. Those names in the group which were authored by William H. Edwards have been very thoroughly treated taxonomically by Brown (1973). McHenry (1962) has prepared a bibliography of the original descriptions of all taxa placed in Pieris in North America.
The biological entities recognized in this paper are:
I. Pieris protodice Boisduval & LeConte
f. vern./aut. vernalis W. H. Edwards nasturtii "Boisduval MS." (W. H. Edwards) II. Pieris occidentalis Reakirt
f. vern./aut./alt. calyce W. H. Edwards IIA. ssp. nelsoni W. H. Edwards
290
Journal of the Lepidopterists' Society
I. Pieris protodice Boisduval & LeConte (Fig, 1).
1829. Hist. Gen. Icon. Lepid. Chen. l'Amer. Sept. 1(5): 45-46; pi. 17, figs. 1-3. Type locality New York and Connecticut. Both sexes described and figured.
The type locality makes this name biologically unambiguous, as there is only one member of this species-group in the eastern United States. Pieris protodice is distributed over most of the U.S. generally below 2000 m but reaching 3000 m in New Mexico and Arizona. It is absent from the Pacific Northwest, north of the Central Valley of California, and from the northeastern states north of southwestern and southeastern Pennsylvania except along the immediate coast, north rarely to Massachusetts. It occurs in southern Ontario, at least sporadically. Its northern and upslope borders are extremely unstable; in cold-winter areas it is generally dependent on immigration and although it may breed, it overwinters only exceptionally. It is reported southward to southern Baja California, and on the mainland to Guatemala (Hovanitz, 1962), but is rare or absent in subtropical Florida although it has been collected on Cuba. It is striking that Boisduval recorded this species, along with Colias eunjtheme ("edusa"), so far north so early. We will probably never know whether they occurred naturally or had already been introduced from further south.
Pieris protodice is always associated with sunny, warm, and dry environments. It is a "weedy," "colonizing," or "fugitive" species commonly found in highly disturbed, early-successional habitats, especially on sandy soils—often by roadsides, along railroad rights-of-way, or in urban vacant lots. In the west it often occurs in the dry washes and around corrals. In the northeast it is frequent on beaches. In most of its range its preferred hosts are the annual or winter-annual cruciferous weeds Lepidium virginicum L. and L. densiflorum Schrad. In California at low elevations it breeds extensively on Brassica geniculata (Desf.) Ball where no summer Lepidium spp. grow, and on the southwestern and Great Basin deserts on native Lepidium spp. and Physaria spp. and on Sisymbrium altissimum L. It rarely breeds on Brassica nigra (L.) Koch (a misdetermination by Hovanitz, 1962) but is capable of accepting a great variety of crucifers in captivity.
Pieris protodice is not known to be a permanent resident anywhere where the breeding season is too short for multiple broods. It has four to six broods in central California and about four at New York and Philadelphia. It is always much commoner and more widespread in late summer and autumn than earlier in the season. Winter is spent as a pupa in diapause under photoperiodic control.
Stocks of P. protodice from California, Arizona, Colorado, Texas, New
Volume 30, Number 4
291
' s
O sUl
Jfe
H^
^
. **^
Wt
*Sg£' "ft**!®*
w * ■*
1
#
1
.■■••• . • :' ,{■-•*■
Fig. 1. Pieris protodice from New York (left-hand pair) and California (right-hand pair), summer phenotypes, dorsal and ventral surfaces. The New York male is phenotypically "intermediate," similar to the description of male nasturtii (see text).
292
Journal of the Lepidopterists' Society
Mexico, Pennsylvania, New Jersey, and New York have been crossed in various combinations with no significant loss of fertility or viability in continuous culture; wide geographic crosses may, however, disrupt the diapause response (Shapiro, unpublished).
Aspects of the biology of P. protodice in California are discussed in Shapiro, 1975a and of its reproductive biology in Shapiro, 1970.
Pieris vernalis W. H. Edwards (Fig. 2).
1864. Proc. Ent. Soc. Phila. 2(4): 501-502. Type locality Red Bank, N.J. (May). Both sexes described.
This is the phenotype of P. protodice which is produced when the larva develops on a long-night regime, regardless of temperature, and may be induced in the progeny of any female of any brood by appropriate treatment (Shapiro, 1968). Its production is not dependent on the occurrence of diapause, which may be inhibited under inducing photo-periods by high temperatures. Intergrades to the vernalis phenotype, and occasionally quite dark examples, are thus produced in autumn. This phenotype has not been reported for many localities where the summer broods of P. protodice occur; this is scarcely surprising given the fugitive nature of populations in this species. There are no differences in the expression of the phenotypes in Californian and northeastern stocks.
No type of Pieris vernalis exists. Despite information given Brown (1973) by New Jersey collectors, Pieris protodice is by no means "nearly extinct" in that state; in 1965 I collected over 500 in an afternoon in Camden. I have not, however, seen specimens of vernalis from Red Bank or elsewhere in Monmouth County, although I have several from nearby Staten Island. Because Staten Island is in New York state, I have refrained from designating any of these as a neotype in the hope that a genuine Red Bank specimen may turn up.
Pieris nasturtii W. H. Edwards.
1864. Proc. Ent. Soc. Phila. 2(4): 501. Type locality San Francisco, California. Both sexes described.
This was a Boisduval manuscript name, resurrected by Edwards. Its tangled history is given by Brown (1973). Apparently Boisduval applied it to an animal in the Pieris napi group, but the specimens sent by Behr to Edwards as "nasturtii" were "an odd variety of protodice'' as Edwards later wrote. No type, nor any specimen which can definitely be linked to this name, is extant.
The only member of this group found in or near San Francisco is P. protodice (except for the possible occurrence of P. occidentalis in the Santa Cruz mountains, see below). The descriptions strongly sug-
Volume 30, Number 4
293
Fig. 2. Pieris protodice from New York (left) and California (right), vernalis phenotypes, dorsal and ventral surfaces.
gest the late September-November phenotypes of Bay Area P. protodice, which are somewhat transitional to vernalis. The "coppery" color of the female is characteristic of old, faded-in-life specimens, as is the hyalinity. It thus appears that the name nasturtii W. H. Edwards refers to the
294
Journal of the Lepidopterists' Society
autumn brood of protodice and is, thus, infrasubspecific and without taxonomic standing. Its revival would in no way benefit taxonomy or biology, especially since variation from summer to winter phenotypes is essentially continuous in autumn in continuously breeding populations. In my publications the male description of nasturtii applies to my "intermediate" phenotypic grade.
II. Pieris occidentals Reakirt (Fig. 3).
1866. Proc. Ent. Soc. Phila. 6: 133-134. Type locality "Rocky Mountains, Colorado Territory, California." Both sexes described.
Pieris occidentalis occurs upslope and northward of P. protodice in western North America. The two species have been found sympatric at various locations from 1000-2500 m in the Rocky Mountains and Sierra Nevada, e.g., Donner Pass, California, where P. occidentalis is a permanent resident and P. protodice a breeding immigrant (Shapiro, 1975a). Pieris occidentalis ranges from Arctic Alaska and adjacent Canada south at increasing elevation to the southern Sierra Nevada of California and the Colorado Rockies (both above 2000 m), east to the Black Hills of South Dakota. It may extend into northern New Mexico. Its eastern limits in Canada are poorly understood. Its southern extent in the California Coast Ranges is unknown; it may reach Santa Cruz County, where it was apparently collected (by M. Doudoroff?) in 1930 (U. C. Berkeley collection). However, I have not seen any other Coast Range specimens from Mendocino County southward.
As noted in the introduction, Chang (1963) has described morphological differences between these two species. There are useful color and pattern differences; both sexes characteristically are more heavily and completely marked in P. occidentalis than in P. protodice; the wings of P. occidentalis appear thicker and more heavily scaled; the body is proportionally larger and usually hairier. The larva is more contrastingly colored and the pupa tends to be shorter and broader than in P. protodice. The chaetotaxy is univestigated. In areas of sympatry occasional interspecific matings may occur, but only 2 of 339 specimens collected at Donner Pass in 1973 were phenotypically ambiguous (Shapiro, 1975a). Although conspecific pairs are readily formed in cages, interspecific ones are not.
Pieris occidentalis is characteristically found at low densities in mountainous regions, where most captures are of "hilltopping" males. It normally breeds on montane crucifers such as Arabis and Streptanthus species and Thlaspi alpestre L., but becomes "weedy" and breeds at high density when presented with the opportunity, as in the railroad yard at
!
Volume 30, Number 4 295
Fig. 3. Pieris occidentalis from the California Sierras, summer phenotypes, dorsal and ventral surfaces.
Donner Pass where its host is Lepidium virginicum. It is double-brooded at moderate elevations southward and possibly partially triple-brooded at its lower elevational limit in both Colorado and California (see below). Winter is spent as a diapausing pupa. Pieris calyce W. H. Edwards (Fig. 4).
1870. Trans. Amer. Ent. Soc. 3, signature 25: 189, no. 1. Type locality "Nevada," restricted by F. M. Brown (1973) to vicinity Virginia City, Nev. Male described.
Edwards (1876) speculated that calyce might be a spring form of Pieris occidentalis, and in this he was correct. Brown (1973) demonstrates that the type was probably collected by Henry Edwards at Virginia City (elev. 1921 m) in March 1868 or 1869. At Virginia City a March specimen would be a very early example of the phenotype emerging from overwintering pupae in a bi- or trivoltine population of occidentalis. In such populations Shapiro (1973) has shown that calyce is the equivalent of vernalis, a seasonal, photoperiod-induced phenotype (although in occidentalis the control is less absolute). It can thus be obtained in the laboratory from the progeny of any female occidentalis. In its original
296
Journal of the Lepidopterists' Society
Fig. 4. Pieris occidentalis from the California Sierras, calyce phenotypes, dorsal and ventral surfaces.
sense, then, although proposed as a species-group name, calyce clearly refers to a seasonal phenotype and is infrasubspecific.
The name calyce has been used by some authors (e.g., Garth & Tilden, 1963) in a subspecific sense to apply to the univoltine animal of this group which occurs at or above tree-line in the Rockies, Sierras and Great Basin ranges. This animal is phenotypically indistinguishable from specimens collected one or two months earlier 1000 m lower. When stock of univoltine "calyce" from Loveland Pass, Colorado (3600 m) was reared under continuous light at 25°C it did not diapause, but developed directly in less than a month and produced light, occidentalis "summer" phenotypes (Shapiro, 1975b). These animals were successfully crossed with P. occidentalis from Donner Pass, California, with no decrease in fertility or viability. An apparently spring-univoltine "calyce" stock from Haystack Mountain in the eastern foothills of the Colorado Rockies has also been studied (Shapiro, 1976b).
The various populations of univoltine "calyce" are completely disjunct from one another on mountaintops. Present evidence implies that
Volume 30, Number 4
297
f *
• #^V
Fig. 5. Pieris occidentalis nelsoni, Fairbanks, Alaska, dorsal and ventral surfaces.
each is independently derived locally from the multivoltine populations found downslope from it. Whether or not one is willing to accept "poly-topic" subspecies, the use of calyce as a subspecific name is rendered inappropriate by the holotype data presented by Brown and should be discontinued. The name calyce is appropriately applicable to a phenotype, not a population.
Univoltine high-elevation populations have been found associated with Thlaspi alpestre and Smelowskia calycina (Desv.) Meyer in Colorado and Erysimum perenne (Wats, ex Cov.) Abrams at Sonora Peak, California. No definite host records are known to me. Erysimum is not a normal Pieris host (Chew, 1975), though P. rapae has been found on it at least once (Shapiro, 1975a).
IIA. Pieris occidentalis nelsoni (Fig. 5). Pieris nelsoni W. H. Edwards.
1883. Butt. North America 2(11): 71, pi. 15, Pieris I, figs. 6, 7. Type locality St. Michael's, Alaska. Male described and figured.
This entity has been "lost" since its original description, although many reports of P. occidentalis in Alaska have been made. In July 1974, I studied a population at Fairbanks, Alaska in which "nelsoni" is the most frequent male phenotype, and subsequently bred it in the laboratory, crossed it with Sierran occidentalis, and obtained genetic data which are being reported elsewhere (Shapiro, 1976a). The nelsoni phenotype is
298
Journal of the Lepidopterists' Society
expressed with or without diapause. Ventrally, nelsoni is about as dark as calyce, but some specimens reared without diapause are as light as summer occidentalis (Shapiro, 1975c).
St. Michael's, now St. Michael, is on the south coast of Norton Sound in western Alaska. Now that nelsoni has also been found in interior Alaska, it appears that it is not an aberration but a valid and recognizable geographic subspecies, and I so treat it. It remains to be seen how far it extends into western Canada and the nature of its contact with nominate occidentalis. Some nelsoni characters are recognizable in populations of occidentalis as far south as Modoc Co., California (Shapiro, unpublished).
P. o. nelsoni has some resemblance to the vernalis phenotype of P. protodice, especially in the male. This is presumably the basis for its classification as a subspecies of protodice by Howe (1975). Although nelsoni has not been tested genetically with protodice, its conspecificity with that taxon can be ruled out on Chang's characters, on the shape of the pupa, and on geographical grounds—nelsoni is distributed some 2000 mi. from the nearest known protodice populations, and in a totally different climatic and vegetational region.
The only confirmed host of nelsoni is Lepidium densiflorum, a weed in the railroad yard at Fairbanks. As noted in Shapiro, 1975c and 1975d there is reason to suspect that this population may be facultatively bi-voltine.
Relationships with Palaearctic Taxa
Higgins & Riley (1970) treat P. occidentalis (and by implication P. o. "calyce") as subspecies of the Palaearctic Pieris callidice Hiibner. Brown (1973) follows this usage, under which there would be two Nearctic subspecies, occidentalis and nelsoni, as interpreted in the present paper. There are no genetic data bearing on the relationship. It is evident that there is great phenotypic similarity among these taxa, particularly between callidice and nelsoni. All populations of callidice known to me have a yellow ventral hindwing, a character unknown in any North American population; in this respect they parallel many Palaearctic members of the Pieris napi complex. W. H. Edwards wrote Henry Edwards concerning Nelson's specimen, March 15, 1882: "There is 1 male Pieris which I think is certainly Callidice. The upper side agrees perfectly with a male Callidice I have from Europe. The underside is not so heavily green dusted on the nervures and branches. If this is Callidice, it is the first American example I ever saw." The plausibility of the conspecificity hypothesis is increased by a report from K. M. Philip,
Volume 30, Number 4
299
of the University of Alaska, that he has a male callidice from the River Omolon, Magadansk Oblast, NE Siberia—bridging the gap between the Alaskan populations and the nearest callidice recorded by Higgins and Riley, in Mongolia. It is very likely that further study will confirm the Higgins-Riley-Brown usage.
Acknowledgments
I thank: K. M. Philip for data on Alaskan and Siberian material; R. & K. Stanford and M. Fisher for help in Colorado; and F. M. Brown for many kinds of aid, including the letter from W. H. Edwards, quoted above. This study has been supported by Grant D-804 from the Committee on Research, U.C. Davis.
Literature Cited
Abbott, W. P. 1957. Ecology and geographic variation in Pieris protodice Bois-
duval and LeConte. M.S. thesis, Texas A. & M. University. 95 p. Abbott, W. P., L. S. Dillon, & R. R. Shrode, 1960. Geographic variation in
Pieris protodice Boisduval and LeConte. Wasmann J. Biol. 18: 103-127. Brown, F. M. 1973. The types of the Pierid butterflies named by William Henry
Edwards. Trans. Amer. Ent. Soc. 99: 29-118. Chang, V. C. 1963. Quantitative analysis of certain wing and genitalia characters
of Pieris in western North America. J. Res. Lepid. 2: 97-125. Chew, F. 1975. Coevolution of Pierid butterflies and their Cruciferous food
plants. Oecologia 20: 117-127. dos Passos, C. F. 1965. A synonymic list of the Nearctic Rhopalocera. Mem.
Lepid. Soc. 1: 1-145. Edwards, W. H. 1876. Catalogue of the diurnal Lepidoptera of America north
of Mexico. Trans. Amer. Ent. Soc. 6: 1-68. Ehrlich, P. R. & A. H. Ehrlich. 1961. How to know the butterflies. W. C. Brown
Co., Dubuque, Iowa. 261 p. Garth, J. S. & J. W. Tilden. 1963. Yosemite butterflies. J. Res. Lepid. 2: 1-96. Higgins, L. G. & N. D. Riley. 1970. A field guide to the butterflies of Britain
and Europe. Houghton Mifflin, Boston. 380 p. Hovanitz, W. 1962. The distribution of the species of the genus Pieris in North
America. J. Res. Lepid. 1: 73-84. Howe, W. H. 1975. The butterflies of North America. Doubleday, Garden City,
N. Y. 633 p. Klots, A. B. 1951. A field guide to the butterflies. Houghton Mifflin, Boston.
349 p. McHenry, P. 1962. The generic, specific and lower category names of the Nearctic
butterflies. I. The genus Pieris. J. Res. Lepid. 1: 63-72. Shapiro, A. M. 1968. Photoperiodic induction of vernal phenotype in Pieris
protodice Boisduval and LeConte. Wasmann J. Biol. 26: 137-149. ----------. 1970. The role of sexual behavior in density-related dispersal of Pierid
butterflies. Amer. Nat. 104: 367-372. ----------. 1973. Photoperiodic control of seasonal polyphenism in Pieris occidentalis
Reakirt. Wasmann J. Biol. 31: 291-299. ----------. 1975a. Ecological and behavioral aspects of coexistence in six Crucifer-
feeding Pierid butterflies in the central Sierra Nevada. Amer. Midi. Nat. 93:
424-433.
300
Journal of the Lepidopterists' Society
—. 1975b. Ecotypic variation in montane butterflies. Wasmann J. Biol. 32: 267-280.
—. 1975c. Photoperiodic control of development and phenotype in a subarctic population of Pieris occidentalis (Lepidoptera: Pieridae). Can. Ent. 107: 775-779.
—. 1975d. Notes on the biology of a "weedy" butterfly, Pieris occidentalis (Lepidoptera, Pieridae), at Fairbanks, Alaska. Arctic and Alpine Res. 7: 273-278.
—. 1976a. The genetics of subspecific phenotype differences in Pieris occidentalis Reakirt and of variation in P. o. nelsoni W. H. Edwards (Lepidoptera: Pieridae). J. Res. Lepid. 14: 61-83.
—. 1976b. Photoperiodic responses of phenologically aberrant populations of Pierid butterflies (Lepidoptera). Great Basin Naturalist 35: 310-316.