SEAVIR PLT PARE LE We: Fe OEE Wy ep 7 Varese fat Masa oh~ UO rasan tate ride te Shyer ane ty ae nee aes F Oe? 4. Ooi fet g r ; hs : as; “ io - DR eters tye) 2 RES” é *MeDalie Sena bef ete Sy DE: he ets as bret ts Wien esas 4 haan hela: PRAIA ARAM Ae veh t Raghed RM VELA AY ELPA OF ee SPIE DARREN ee tT pitta te Tots SS a atan etwas tr eetet rete pee peek tert Ss wa) Senin ny ha ag A a ee To era Sa terres” Oe a eek Se a te ee = - ~ J e : i 3 = : ; - < 2 : > : : : HARVARD UNIVERSITY na LIBRARY OF THE Museum of Comparative Zoology Behe Sane ay) i Vi PAS hi | ita @theh ew r 3 “> oa © 3 reid & ye vies (thi! : 1 “6 (4. a4 1 aa y ae : Py ye atte T sPs)i Papen Ab a Ve a ee frovet 10 29th) IO Oe, | PATE. Pe | Bienes OGe i ¢ 7 —- 7 ? Perr |) Oe Brive ~ Sie they 4 pe) ore 7 n 7 " = 7 oe ow, ol 4; it : ee ee ew Sah e, jy valet a rAd < Soap rite, ha Te Pee ae - Tei? pelle. >t ae ey an gO 9: Os ho Mis We: \Eaee DA, i ; 1 Mev, ae mae i i i el a al oe aey ov an i ate? a co IVE ods SN LN ie LIBRARY SEP 6 1968 Postilla PEABODY MUSEUM OF NATURAL HISTORY YAGE UNIVERSITY, NEW HAVEN, CONNECTICUT, U.S.A. Number 119 10 May 1968 A NEW SIBLING PAPILIO FROM THE ROCKY MOUNTAINS, WITH GENETIC AND BIOLOGICAL NOTES” CINSECEA, LEPIDOPTERA) CHARLES L. REMINGTON DEPARTMENT OF BIOLOGY AND PEABODY MUSEUM OF NATURAL HIsTory, YALE UNIVERSITY ABSTRACT Papilio gothica (Papilionidae) is described as a new species from montane regions of Colorado and adjacent states. It is a sibling of the lowland Californian P. zelicaon and is characterized by subtle color and pattern differences, univoltinism, Pseudocymo- pterus montanus (Umbelliferae) as larval food, and probably dis- tinctive flight behavior. The chromosome number is n= 30. Phenotypically P. gothica and P. zelicaon are nearly alike, but their F, hybrids with P. polyxenes are unlike. They have dif- ferent hybrid sex ratios in their crosses with P. polyxenes, gothica polyxenes being nearly lethal for the sex of the polyxenes parent whereas zelicaon >< polyxenes has more nearly equal sex ratio although deficient in the sex of the polyxenes parent. The evolu- tionary meaning of the principle of phenotypic stability with phyletic divergence is discussed. Phenotypic, biological, and geo- graphic characters are summarized for the North American taxa 23 Postilla No. 119 similar to gothica, especially P. brucei, P. oregonia, P. rudkini, and P. hudsonianus. Photographs are given for all of these as well as P. gothica and P. zelicaon and hybrids of both crossed with P. polyxenes and P. bairdii. The foodplant of P. hudsonianus in, Manitoba is Zizia (probably aurea) (Umbelliferae). Egg hatch- ability and embryonic viability are tabulated for P. gothica, P. zelicaon, and several hybrid combinations. The principal purpose of this paper is to validate the name and describe the characteristics of a presently unnamed population, stock from which has been extensively used in my studies of nat- ural and experimental hybridization, caryotypes, foodplant specif- icity, and population ecology of the polyxenes—machaon complex of Papilio butterflies. The “new” species has for many years gone under the names zolicaon Boisduval, zelicaon Lucas, and brucei Edwards, but early in my work on this group it became apparent that the abundant, higher altitude, Umbelliferae-feeding entity in Colorado is biologically very different from the low-altitude, Umbelliferae-feeding true zelicaon of California and from the mid- altitude, Artemisiu-‘eeding true brucei of Colorado. My associates and I have informally used the following new name in our re- search protocols for several years. The formal naming has been delayed in the hope that one or more all-or-none characters would be found by which every dead specimen could be infallibly recog- nized. Such characters have not yet been discovered, but the name is needed for use in various forthcoming papers and is now designated as follows: Papilio gothica, species nova Phenotypically extremely similar to typical Californian Papilio zelicaon and some typical Coloradan P. brucei, and not always distinguishable at present from these two on any single criterion. In this instance, suitable photographs are truly superior to many words of description, and the accompanying plates will stand in lieu of a routine description of P. gothica. Helpful wing charac- ters by which most gothica differ from most zelicaon or brucei are the following: SEE SeP 6 1968 HARVARD 1. Dorsal ground color sexually dimorphic. Males, Mustardy Yellow, Amber-Yellow, or Pinard Yellow (color terms from Ridgway, 1912) [approximately OY-17-11° of Villalobos (1947) |. Females distinctly paler, i.e., usually Straw Yellow of Ridgway. In ground color P. zelicaon and P. brucei are sexually more monomorphic; both sexes of P. zelicaon tend to be similar in color to gothica males, and both sexes of P. brucei tend to be similar to gothica females. 1968 A NEW SIBLING PAPILIO 2. Forewing below, in postmedian broad yellow band, with the anterior spot (pm/ in Fig. 1B) having an outer edge strongly offset from a line drawn through the outer edges of spots 2 through 9; in zelicaon and brucei the outer edge of the anterior spot tends to be in line with this edge of spots 2 through 9. 3. Forewing below with postmedian spot 2 (pm2 in Fig. 1B) tending to have its outer edge forming an angle with its caudad edge only slightly greater than 90°; in zelicaon and brucei this angle is much greater than 90°. 4. Forewing above, near costal edge, usually with two fine sm FIG. 1. Papilio gothica, illustrating characters used in distinguishing taxa in this complex. A, dorsal surface (upperside); B, ventral surface (under- side). Symbols: pm/ —first pale spot of postmedian row, pm2— second pale spot of postmedian row, etc.; c/— fine yellow costal lines; an anal marginal cell of hindwing; dc — discal cell; sm — submarginal row of pale spots of forewing; /ism/ first submarginal spot of hindwing; eye — “eye” marking of hindwing anal angle. 4 Postilla No. 119 yellow lines (c/ in Fig. 1A) opposite end of cell, one anterad and one posterad of the closely parallel veins Sc and R,; these lines are usually absent in zelicaon but frequently are present in brucei. 5. Hindwing above has cell Cus, near the anal margin (an in Fig. 1A), with basal dark color usually extending far distad of origin of vein Cu. from cell; this dark area tends to be greatly reduced in zelicaon males and in both sexes of brucei. 6. Forewing below with spots of submarginal yellow row (sm in Fig. 1B) large and tending toward fusion; in zelicaon these spots tend to be smaller and to be separated clearly by blackening of the veins crossing this row, and in brucei these spots are even larger and more fused than in gothica. 7. Forewing tends to be slightly more rounded at the apex and to form a more acute angle between anal and outer margins than in zelicaon; brucei tends to have this angle much more obtuse than gothica or zelicaon. 8. Forewing below with basal half of discal cell (dc in Fig. 1B) entirely black (extremely rarely there is yellow streaking, perhaps an indicator of introgressive hybridization with brucei); zelicaon is like gothica, but brucei usually has extensive yellow scaling in this area (see Fig. 35). 9. Hindwing below with eyespot at anal angle (eye in Fig. 1B) tending to have the black “pupil” moderately large, rounded, and thinner than the orange area just cephalad; in zelicaon this “pupil” is even larger and tends to equal or exceed the thickness of the orange area, whereas in brucei the “pupil” is usually very thin, transverse, and less than half the thickness of the orange area. This character often shows strong asymmetry, with one hind- wing having a smaller “pupil” than the other. 10. Hindwing below with the first spot of the submarginal yellow row (hsm/ in Fig. 1B) tending to be somewhat broader than in zelicaon and much narrower than in brucei; on the upper- side this spot tends to obsolescence in many Zelicaon. 11. Hindwing tail tends to be slightly shorter and slenderer than that of zelicaon; that of brucei tends to be longer and slenderer than that of gothica or zelicaon. 1968 A NEW SIBLING PAPILIO 5) The most reliable characters for distinguishing adults of mon- tane Coloradan gothica from lowland coastal Californian zelicaon appear to be numbers 2, 3, 4, and 5S. In addition to these wing characters, it should be noted that in P. gothica the abdomen has the venter usually plain black and the lateral yellow line slender; zelicaon resembles gothica, but on females the lateral yellow stripe is wider, and brucei has much more extensive yellow on the sides and often on the entire venter. A preliminary inspection of the several available preserved larvae and pupae has not revealed readily quantifiable taxonomic distinctions. However, there is an interesting gene-frequency dif- ference in larval spot color (see below, under Genetical Studies ). The chromosome complement of P. gothica, reported under temporary terminology as “Papilio ‘brucei’” by Maeki and Remington (1960), shows n= 30, without the m-chromosome reported from P. polyxenes. Two males from the supposed zelicaon population from the Cascade Mountains of Okanogan County, Washington, also had n = 30 and lacked the m-chromo- some (Remington & Maeki, unpublished). Testes of various low- land Californian zelicaon have been collected but not yet sectioned. TYPE SPECIMENS Although there is no doubt that P. gothica is widely distributed in the Rocky Mountain region and probably in some areas farther west (see below), I am limiting my type series to specimens from the Colorado mountains, as follows: HOLOTYPE MALE (Figs. 2 and 4): Gothic, 9500’, Gunnison Co., Colorado, 14 June 1956, leg. Eric E. Remington; in Peabody Museum of Natural History. REPRESENTATIVE FEMALE PARATYPE (Figs. 3 and 5): Gothic, 9500’, Gunnison Co., Colorado, 28 June 1960, leg. Eric E. Remington; in Peabody Museum of Natural History. PARATYPES (two pairs shown in Figs. 6-9 and 14-17). Forty adult males, twelve adult females, four larvae, and sixty-two pupae, all from Colorado. Gothic, Gunnison Co., males: 13 June 1956 (C. L. Remington), 14 and 27 June 1956 (E. E. Remington), Gnuly 1956°CR. W. Peaser.), 11-12 duly-1957 (CR. W. Pease 6 Postilla No. 119 Jr.), 28 June 1959 (E. E. Remington & B. Baker), 12 July 1959 GR Wes Peasendr.);. 28 June 1960 (e: E> Remimeton) “(CER testes +740), 13 July 1960 (ex pupa, bred ab ovo, CLR testes +629), two 1 July 1961 (E. E. Remington); females: 10-14 July 1957 (S.A. Ae), 30 July 1957 (CE. E. Remington), 1 Aug: 1957 (Si: A: Ac); 28 and 29 June 1960 (E. E. Remington), 26 June 1961 (B. Baker). East of Copper Lake, 10,500’, above Gothic, Gunnison Co., 4 July 1960 (E. E. Remington), female. Virginia Basin, 12,500’, above Gothic, Gunnison Co., 20 July 1956 (R. W. Pease Jr.), male. Cumberiand Pass, 12,600’, Gun- nison Co.: 28 July 1955 (C. L. Remington), two males (fathers of CLR hybrid broeds 2410 and 4:12), 12 July 1967 (C.{G: Oliver), male. Mt. Bellview, 12,500’, north of Gothic, Gunnison Co., 4 July 1956 (R. W. Pease Jr.), 6 males. Elkton town- site, 10,000’, Gunnison Co., 12 July 1959 (C. L. Remington), male. Eldora, 8800’, Boulder Co.: 16 June 1933 (@. S. & C. L. Remington), female; 28 June 1937 (P. S. & C. L. Reming- ton), male; 30 June 1937 (2. S. & C. LW. Remington),;.3 males and |. female; 3 July 1937 (P. SS: .&C. L. Remington), 3' males; 23 July 1949 (C. L. Remington), 2 males. Mt. Audubon, 11,000’, Boulder Co., 26 July 1949 (C. L. Remington), male. Arapaho Pass, 11,900’, Boulder Co., 17 July 1937 (P. S. & C. L. Remington), male. Boulder Canyon, 8000’, Boulder Co., 19 June 1933 (P.°S. & C. L. Remimeton), male: 5 mi. Ne of Eldora; 11,500’, Boulder Co., 10 July 1947 (P. S. Remington), female. Tolland, 8700’, Gilpin Co., 2 July 1937 (P. S. & C. L. Reming- ton), 2 females. 3 mi. N. of Rabbit Ears Pass, 10,000’-10,600’, Routt Co:, 15. and 16 July 1956 (Fo & Po Rindse)- Gemales: The 6 Routt Co. males are in the American Museum of Natural History, and the 1967 Cumberland Pass male is in the collection of Charles G. Oliver. The remaining 33 males and 11 females are at present in the Peabody Museum of Natural History and my own genetical collection. Also designated as paratypes are four preserved larvae taken at Gothic (three, 6 Aug. 1956, F, of wild female, leg. E. E. Remington & S. A. Ae; one, found on Umbelliferae, preserved 18 July 1955, leg. R. W. Pease Jr.) and 42 living, diapausing pupae reared from Oliver female +1 collected wild at Taylor Park, 9400’, Gunnison Co., 2 July 1967, by C. G. Oliver. 1968 A NEW SIBLING PAPILIO 7 NOMENCLATURE The new entity is named for Gothic, a locale at 9500’ elevation in the West Elk Range of the Colorado Rocky Mountains. Gothic was originally founded as a town during the silver mining boom of the late 1870’s and early 1880’s but was soon abandoned. In the early 1930’s the Rocky Mountain Biological Laboratory was developed on the site and has been an unexcelled center for research in the biology of montane organisms by many investigators. Much of my experimental and field work with the evolutionary genetics of Lepidoptera, including Papilio gothica, has been carried out there, hence the appropriateness of the name. The name gothica as used here is a substantive, not an adjective. It is curious that no name was available in the literature that might be applied to this very wide-ranging, handsome, and often abundant butterfly of the Rocky Mountain region, but I am cer- tain that this is so. The types of Lucas’ zelicaon (which I exam- ined in the Muséum National d’Histoire Naturelle in Paris in 1958) and Boisduval’s zolicaon were collected solely in California, by P. J. M. Lorquin. P. coloro Wright (1905), from California deserts, is probably zelicaon but may be a hybrid zelicaon rudkini. Fischer's (1908) “impunctata”, “melanotaenia’, and “formosa” were named without locality designations as aberra- tions of zelicaon; these are in fact trivial minority forms in zelicaon (and perhaps in gothica) populations, and their names are nomen- claturally unavailable at the species—subspecies level. Gunder (1928) named ““mcdunnoughi” as a “transition form”, and it, too, clearly applies to a minority form in populations and is unavailable as the name for gothica, although the holotype is from Waterton Lakes, Alberta, and may be gothica; Gunder’s paratypes of this aberration are from Wyoming, “Colorado”, and California. The names dodi McDunnough and avinoffi F. & R. Chermock apply to the hudsonianus complex (see below), not to gothica. I have proven P. nitra Edwards (the types of which I have examined at the Carnegie Museum) to be an interspecific hybrid, and this name is unavailable for any species or subspecies. Edwards’ true P. brucei, the status of which is discussed below, is not the taxon here named gothica, although the name brucei has recently been used for it with explicit reservations (Remington, 1958; Maeki & Remington, 1960; Ae, 1965). The specimen figured without 8 Postilla No. 119 locality notation by Brown (1956) as “zelicaon” was meant to represent what I now call gothica, but if it is gothica, it is highly atypical in my characters 2, 3, 4, 5, and 8. STATUS AND RELATIONSHIPS Although I am formally erecting Papilio gothica as a nomen- clatural entity with full species status, I consider this a tentative placement. It needs to be tested by thorough backcross studies of hybrid fertility and developmental viability and by analysis of interactions in localities where some sympatry exists. My assistants and I have successfully crossed gothica and zelicaon eight times, but the status of gothica has been a relatively narrow aspect of my broad experimental study of evolutionary processes in the polyxenes—machaon group, and by chance I have not yet had the ideal congruence of the appropriate livestock and seasonal breed- ing facilities that would have allowed me to complete the gothica— zelicaon tests. Furthermore, I have not been able to do any field studies in sites of sympatry. In considering 1) the evidence of normally concealed genetic differences between gothica and zelicaon, even in wing characters, that are revealed in their separate hybrids with polyxenes tester stocks, 2) the difference in sex-ratio distortion also shown in test- crosses with polyxenes, and 3) their probably profound biological differences, I have only a little hesitation in placing them as sep- arate species. In all groups of animals, most pairs of largely allopatric, closely related taxa that are routinely treated as separate species are even less well known genetically than these two Papilio. There are several other North American entities in the polyxenes— machaon complex that have the broad yellow wing-band and the yellow-and-black-striped abdomen and_ therefore superficially resemble P. gothica and P. zelicaon. Most of these are shown, along with some of presently unknown status, in Figs. 24-29 and 32-37: Papilio oregonia Edwards and P. brucei Edwards appear to me to be subspecies of P. bairdii Edwards. Ever since the brilliant work of Edwards and Bruce in the late nineteenth century, this has been a widely accepted allocation. However, the nature of natural hybridization (a rather narrow zone of polymorphism 1968 A NEW SIBLING PAPILIO 9 rather than clinal intermediacy) hints that bairdii had attained the species level prior to sympatry with brucei. All three taxa seem to be restricted to Artemisia dracunculoides (Compositae) as their larval foodplant and to be multivoltine throughout their range. All are unusually variable, perhaps due substantially to intro- gression between the three and to a lesser degree to hybridization with all the other members of the polyxenes-machaon complex with which they have some present or recent sympatry. Since bairdii does not resemble gothica, its dead-specimen characters need not be considered here. P. oregonia (Figs. 26 and 34) is easily distinguishable in having extensive yellow coloration in the forewing cell below, a tendency for a peppering of yellow scales over the dark areas of the upperside of both wings, a marked caudad displacement of the black “pupil” in the hindwing anal eyespot (rarely, the “pupil” is centered as in zelicaon). P. brucei (Figs. 27 and 35) is a useful name for the oregonia-like populations of Colorado, Utah, Arizona, and possibly the north- ern Rocky Mountain region. It has less yellow scaling in the fore- wing cell than does oregonia (sometimes no yellow at all), and the black “pupil” of its eyespot tends to be less displaced than in oregonia. Differences between brucei and gothica or zelicaon are noted in the diagnosis for gothica above. Papilio rudkini Comstock appears to be a distinct species, although it was originally named as a race of bairdii. It seems to be more closely related to P. zelicaon than to the bairdii complex. Its larvae are said to be restricted to Thamnosma montana (Ru- taceae). and it is multivoltine. It is sympatric with P. zelicaon in some desert regions of southern California, with P. oregonia in western Nevada and perhaps eastern California and southwestern Oregon, and possibly with P. brucei and even P. gothica in Utah and northern Arizona. P. rudkini (Figs. 24, 25, 32, 33) differs from gothica and zelicaon in having the second postmedian spot of the forewing below (see pm2 in Fig. 1B) with its outer edge arrowhead-shaped, the black spot in pm2 very large, the caudad yellow bar of the postmedian row (pm9 in Fig. 1A) with a promi- nent caudad inward extension especially in males, the forewing submarginal spots (sm in Fig. 1B) much more discrete and rounded, the anal-edge cell of the forewing above (an in Fig. 1A) with the basal black more restricted, the hindwing above with ie) Postilla No. 119 much less postmedian blue color in males, the tail usually longer, the forewing slenderer and forming a more acute angle between the costal and outer margins, and a tendency for the yellow ground color of the male to be more ochreous and that of the female more pale-yellow(i.e., more sexual dimorphism in ground-color). Two supposed forms of rudkini (“comstocki” and “clarki’”) do not closely resemble gothica and need not be considered here. Their genetical status is now being investigated by Fred T. Thorne. Papilio hudsonianus Clark and similar entities from the Dakota Badlands, Nebraska, northeastern Wyoming, Manitoba, Saskat- chewan, Alberta, and Montana have been associated with the European and Alaskan P. machaon Linné (e.g., Clark, 1932). There is as yet little biological or genetic evidence for this asso- ciation, and I am inclined to doubt its correctness (Remington, 1956). These populations resemble gothica and zelicaon but are, in my opinion, specifically distinct; they seem to be multivoltine and to feed on Umbelliferae. We found eggs and larvae on Zizia sp. (probably aurea) at Riding Mountain, Manitoba. All the many specimens I have seen from North Dakota, Nebraska, and Mani- toba belong to this group. Neither true gothica nor zelicaon seems to occur in those areas, although there is a small, univoltine, gothica-like Papilio in the higher elevations of the Black Hills of South Dakota. The hudsonianus-like specimens tend to have some yellow scales in the forewing cell below and to have the eyespot “pupil” displaced caudad, much as in brucei, but the forewing tends to be longer, narrower, and more acute-tipped than in brucei. In fact, these specimens resemble brucei in so many subtle charac- ters that they might be considered conspecific if they were not regularly associated with Umbelliferae. Should Artemisia-feeding larvae also be found in this northern Plains region (where A. dracunculoides is not uncommon), this would support my suspi- cion from dead-specimen analyses that brucei and the hud- sonianus type have recently become widely sympatric and are hybridizing relatively freely. This is a region in which extensive and probably recent natural hybridization is now known in many genera of plants, vertebrates, and insects (see Remington, 1968). As Clarke and Sheppard (1955) noted, this group is excep- tionally suitable for the study of processes of speciation. But it is much too complicated for grand conclusions based on scanty 1968 A NEW SIBLING PAPILIO iL | breeding experiments or on specimen.samples from a few distant, randomly-chosen localities. Not surprisingly, recent experimental and field studies have tended to support the tentative taxonomic conclusions of workers who were widely acquainted with these but- terflies in the field as well as from extensive specimen material (e.g., Edwards, 1895, and Bauer, 1955). Unfortunately, foreign and some North American authors who have attempted, without such acquaintance, to deal with the relationships and status of these and other North American members of the polyxenes— machaon complex have been wide of the mark in their conclu- sions (e.g., Rothschild & Jordan, 1906, Clark, 1936, Eller, 1936, Clarke & Sheppard, 1953 ef seq.). Throughout this paper I have used binomina without necessarily implying full species status. For example, I consider Papilio brucei to be a definite deme, probably conspecific with P. oregonia and possibly with P. bairdii. Papilio avinoffi is surely conspecific with P. hudsonianus and possibly with P. machaon. Where I do not wish at this time to specify the hierarchical status of a taxon, it is most convenient to refer to it with a binomen. GEOGRAPHY Populations with the gothica—zelicaon facies are known from New Mexico, Arizona, Colorado, Utah, Nevada, South Dakota, Wyom- ing, Montana, Idaho, Alberta, British Columbia, Washington, Oregon, California, and Baja California. As will be discussed below, each deme is at present best allocated to either gothica or zelicaon after hybridization against polyxenes tester stocks and after the assay of voltinism, of oviposition plants, and probably of flight behavior. Most of the dead-specimen records from this wide geographic range are not now placeable with certainty because they are not associated with enough data. I am confidently allocating the specimens from the higher mountains of New Mexico, Colorado, and Wyoming to P. gothica and all specimens from the Pacific Slope lowlands of California south of San Francisco Bay to P. zelicaon. P. zelicaon also occurs farther to the north and east, but I do not know how far. I suspect, from a study of wing characters, the montane populations of Montana, 12 Postilla No. 119 Idaho, Utah, Alberta, and eastern British Columbia are gothica. The two females from the Ruby Mountains of eastern Nevada that I have examined are like gothica in wing characters (see Figs. 23 and 31); the few from Reno are zelicaon or have indeci- sive character states. The populations of the high Sierras and of the Coast Range of northwestern California and western Oregon resemble gothica somewhat more than zelicaon. | have examined long series from low elevations of Okanogan County, Washington, and Vancouver Island, British Columbia, and these seem to be closest to zelicaon; the F, hybrids between eastern polyxenes and the Okanogan County deme definitely have the phenotype of polyxenes »< California zelicaon. (See discussion, below, of this hybrid difference.) Partial sympatry between the two is to be sought at mountain—lowland contacts in central and eastern California, western Nevada, Oregon, southern and western Idaho, western Washington, and eastern Montana. There may be a cline connecting gothica and zelicaon in this region, but my skimpy evidence suggests that there is interspecific hybridization rather than clinal intermediacy. BIOLOGY Papilio gothica in Colorado is principally a montane taxon, occurring moderately abundantly from about 6000 feet above sea level, up to tree-line, in most or all of the mountain ranges in the state. The tree-limit is at about 11,400 feet in the Front Range in Boulder County (Marr, 1961) and slightly higier in the Elk Range in Gunnison County (Langenheim, 1962), the two areas where I have worked most intensively with P. gothica. Although there is no evidence that gothica is a permanent resident above the tree-line, males are commonly found exhibiting _hill- topping behavior around summits above tree-line, even as high as 14,000 feet. Typical P. gothica probably does not normally occur as a resident out on the lowlands to the east of the Front Range or to the west between the mountain ranges. There is a pheno- typically slightly different population at the plains—mountain interface at the eastern edge of the Front Range which may prove to have major biological distinctions from gothica. Without further field and breeding study it would be premature to name PEATE "SECTION FIG. 2. Papilio gothica Remington, HOLOTYPE @. Gothic, 9500’, Gunnison Co., Colorado, 14 June 1956, leg. E. E. Remington. FIG. 3. Papilio gothica, representative PARATYPE @ . Gothic, 9500’, Gunnison Co., Colorado, 28 June 1960, leg. E. E. Remington. Dorsal surface; for venter see Figs. 4 and 5. » 1.5 FIG. 4. Papilio gothica, HOLOTYPE ¢. FIG. 5. Papilio gothica, representative PARATYPE @. Ventral surface of specimens shown in Figs. 2 and 3. FIG. 6. Papilio gothica Remington, PARATYPE @, Gothic, 9500’, Gunnison Co., Colorado, 6 July 1956, leg. R. W. Pease Jr. FIG. 7. Papilio gothica, PARATYPE @ , Gothic, 9500’, Gunnison Co., Colorado, 28 June 1960, leg. E. E. Remington. FIG. 8. Papilio gothica, PARATYPE &, Mt. Audubon, 11,000’, Boulder Co., Colorado, 26 July 1949, leg. C. L. Remington. FIG. 9. Papilio gothica, PARATYPE @ , Eldora, 8800’, Boulder Co., Colorado, 30 June 1937, leg. P.S. & C. L. Remington. FIG. 10. Papilio zelicaon Lucas, ¢ , Berkeley, California, 22 June 1932. FIG. 11. Papilio zelicaon, 2 , Berkeley, California, 25 May 1932. FIG. 12. Papilio zelicaon, ¢, La Jolla, San Diego Co., California, 27 Aug. 1943, leg. D. Starrett. FIG. 13. Papilio zelicaon, 2 , Summerland, Santa Barbara Co., Cali- fornia, 4 May 1931, leg. C. W. Kirkwood. Dorsal surface; for venters see Figs. 14-21. > .75 FIG: FIG, BIG: FIG. FIG. FIG, FIG. FIG. 14. ey 16. 7: 18. 19) 20: PAN: Papilio gothica, PARATYPE ¢@. Papilio gothica, PARATYPE @. Papilio gothica, PARATYPE 2. Papilio gothica, PARATYPE @. Papilio zelicaon, é . Papilio zelicaon, @ . Papilio zelicaon, @. Papilio zelicaon, @ . Ventral surface of specimens shown in Figs. 6-13. FIG. 22. Papilio (zelicaon?), ¢ , Pateros, Okanogan Co., Washing- ton, 1 May 1935, leg. A. Anderson. FIG. 23. Papilio (gothica?), @, Lamoille Canyon, Ruby Mts., Elko Co., Nevada, 28 June 1959, leg. T. W. Davies. FIG. 24. Papilio rudkini Comstock, @ , Mexican Well, Ivanpah Mts., San Bernardino Co., California, 1-2 Sept. 1934, leg. C. N. Rudkin. FIG. 25. Papilio rudkini, ? , Mexican Well, el. 4800’, Ivanpah Mts., San Bernardino Co., California, larva on Thamnosma montana 17 Sept, 1934, eclosed 28 Feb; 1936, leg. C. Henne. FIG. 26. Papilio oregonia Edwards, ? , Brewster, Okanogan Co., Washington, 2 Aug. 1952, leg. J. C. Hopfinger. FIG. 27. Papilio brucei Edwards, ¢ , Glenwood Springs, Garfield Co., Colorado, 18 July 1961, leg. O. R. Taylor Jr. FIG. 28. Papilio avinoffi Chermock & Chermock, ¢?, F, of C.L.R. cross #83 (mother and father reared from ova found on Zizia, Riding Mt., Manitoba, leg. C. L. Remington & R. W. Pease Jr.), eclosed 30 June — 2 July 1956. FIG. 29. Papilio hudsonianus — avinoffi group, ¢ , Killdeer Mts., Dunn Co., North Dakota, 23 May 1964, leg. J. Oberfoell. Dorsal surface; for venters see Figs. 30-37. > .72 FIG. FIG. FIG; FIG. FIG. BIG: FIG, FIG. 50): SN ks 32. Beh 34. 35° 36. Sues Papilio (zelicaon?), 2. Papilio (gothica?), @. Papilio rudkini, ¢ . Papilio rudkini, @ . Papilio oregonia, @ . Papilio brucei, 2 . Papilio avinoffi, 2 . Papilio hudsonianus — avinoffi group, 2 . Ventral surface of specimens shown in Figs. 22-29. FIG. 38. Papilio F, hybrid 2° polyxenes < 4 gothica (Remington brood #10), 4, eclosed 14 May 1956, bred C. L. Remington & R. W. Pease Jr.; usual phenotype for this parentage. FIG. 39. Papilio F, hybrid @ polyxenes < 8 gothica (brood #10), 6 ,eclosed 19 May 1956; chosen as showing wider hindwing median yellow band than usual for this parentage. FIG. 40. Papilio F, hybrid @ polyxenes < é zelicaon (Remington brood #115), &, eclosed 24 July 1956, bred C. L. Remington & R. W. Pease Jr.; chosen as showing narrower hindwing median yel- low band than usual for this parentage. FIG. 41. Papilio F, hybrid @ polyxenes « ¢ zelicaon (brood #115), 6, eclosed 23 July 1956; usual phenotype. FIG. 42. Papilio F, hybrid @ gothica « ¢ bairdii (Remington brood #337A), 4, eclosed 14 May 1958, bred C. L. Remington & R. W. Pease Jr. FIG. 43. Papilio F, hybrid @ gothica « ¢ bairdii (brood #337A), é , eclosed 23-29 Aug. 1957. FIG. 44. Papilio F, hybrid 2 zelicaon « 8 bairdii (Remington brood #452), 6, eclosed 5 June 1958, bred R. W. Pease Jr. FIG. 45. Papilio F; @ zelicaon « 68 bairdii (brood #452), ¢, eclosed 4 June 1958. Dorsal surface (ventral surface not illustrated). .68 1968 A NEW SIBLING PAPILIO 29 this population, but I am excluding all specimens of it from the type series of gothica. | have examined a large sample of this plains-edge deme collected at the mouth of Jarre Canyon in Douglas County, Colorado, and I have inspected other specimens from similar biotopes in Wyoming, Montana, the Dakotas, and Alberta. I have noted (Remington, 1958, 1968) that typical gothica is presently hybridizing with the eastern Papilio polyxenes. I now know that the plains-edge deme is also crossing with P. polyxenes. Surprisingly, the two kinds of hybrids are phenotypi- cally dissimilar, and this is the principal evidence that inclines me to the view that there are two gothica-like species or semispecies in Colorado. Because no consistent phenotypic differences have been found between typical gothica and the plains-edge deme, it is not at present possible to assay natural interbreeding between these two. Similarly, I cannot with certainty recogn'ze backcross or even F, hybrids between gothica and brucei, so | do not yet know how extensively these two are hybridizing at their few known contact points; Edwards’ (1895) report of the rearing of occa- sional “zelicaon’’-like specimens from brucei—bairdii broods from Glenwood Springs, Colorado, perhaps indicates hybridization there with gothica. Typical P. gothica is crisply univoltine in Colorado and _ prob- ably throughout its range. Not only do the pupae normally require chilling or long aging to break diapause, but larval development in the laboratory is significantly slower than that of the several multivoltine members of the polyxenes-machaon group that I have reared. Typical P. zelicaon has several generations per year, and fresh adults are present in parts of coastal southern California almost every month of the year (see, e.g., Comstock, 1927). P. brucei is at least bivoltine at Glenwood Springs. It is not known whether there is a second generation of the plains-edge deme at Jarre Canyon, but its presumed counterpart in Wyoming and Montana has at least two generations per year. In flight P. gothica seems to be swifter and more elusive than typical P. zelicaon. An objective quantitative assay of flight behav- ior would be difficult to make, but my subjective impression of differences comes from the three times when I have successively observed gothica and zelicaon closely in the field during the same or consecutive years. 30 Postilla No. 119 There also appears to be a substantial difference in foodplant choice by ovipositing females. In coastal California, the principal foodplant of P. zelicaon is the introduced weed Foeniculum vul- gare (see, e.g., Coolidge, 1924; Comstock, 1927), and it readily oviposits on cultivated Umbelliferae and even the rutaceous genus Citrus. P. gothica, however, seems to reject weedy and cultivated umbellifers. Several times, at the proper season, I have searched with no success for eggs and larvae on large beds of carrot (Daucus carota) and parsley (Petroselinum crispum) in kitchen gardens in Crested Butte, near Gothic; P. gothica is present at Crested Butte. Similarly, I have fruitlessly examined hundreds of Queen-Anne’s-Lace (wild D. carota) in the mountains of Boulder County, in localities where gothica is abundant and at a season when larvae were to be expected. The failure is a true indicator of gothica oviposition choice rather than any lack of experience in my finding eggs and larvae, since I have had no difficulty in finding many hundreds of wild eggs, small larvae, and mature larvae of the closely related P. polyxenes on wild and cultivated D. carota and cultivated Petroselinum and A pium, as well as other hundreds, in total, of the related P. brevicauda, P. kahli, P. zeli- caon, P. bairdii, and P. gothica itself on various native species of Umbelliferae and Artemisia dracunculoides. My assistants and I have recorded the foodplants of five wild larvae of P. gothica, all taken at Gothic; all were on Pseudocymopterus montanus (A. Gray) Coulter & Rose: early July (2) and 25 July 1956; 13 and 20 August 1961. During the course of my genetical studies of these Papilio species we have brought into the laboratory from the vicinity of the Gothic Meadow several thousand fresh plants of Ligusticum porteri Coulter & Rose and Oxypolis fendleri (A. Gray) A. Heller, and hundreds of P. montanus and Heracleum lanatum Michx., for use as larval food. All are satisfactory laboratory foods for gothica, and all are abundant where females fly, but careful examination of all of these plants has revealed larvae or eggs only on P. montanus. Pseudocymopterus is surely the preferred wild host of gothica in Colorado. Another difference in ovipositional behavior is suggested by my observations in Colorado and coastal California. All of our wild eggs and first instar larvae of P. gothica have been found on the flower umbels, although foliage was conspicuous and abundant. 1968 A NEW SIBLING PAPILIO 31 Further, without identifying the specific releasers, we have con- sistently gotten much larger egg production from confined females given umbels than those given leaves as an oviposition substrate. Thus, it appears that gothica prefers to oviposit on the flower head. In my studies of zelicaon at Santa Barbara and on Santa Cruz Island, in California, I have found large numbers of eggs and first instar larvae on foliage of Foeniculum, usually on very young shoots; far fewer have been on umbels, except when the plants are mature and tall and have few or no young leaf shoots. Thus, it appears that zelicaon chooses to oviposit on both the foliage and the flower head, perhaps preferring the leaves. Among related species, my rather extensive observations show that P. kahli in Manitoba chooses the flower heads of Zizia, P. brevicauda (or its hybrid with polyxenes) in New Brunswick chooses flower heads of Ligusticum, and P. polyxenes in Connecticut and Missouri chooses flower heads of wild Daucus; in the absence of flowers, 1 have found that wild females of polyxenes oviposit readily on foliage of Petroselinum, Apium, and garden Daucus. GENETICAL STUDIES The wing characters of Papilio gothica and P. zelicaon being nearly identical, it was not surprising that F, hybrids between them do not show distinctive wing characters of their own. Unex- pectedly, however, when gothica and zelicaon were crossed with a single tester species, P. polyxenes, the F, hybrids of polyxenes x gothica proved to be easily distinguishable from polvxenes >< zelicaon. A similar set of crosses using P. baird i as the tester like- wise revealed phenotypic differences between F, of gothica bairdii and zelicaon >< bairdii. These and other tester crosses also showed that gothica parentage produces a hybrid sex ratio significantly different from that from zelicaon parentage. Thus, it is clear that the extreme similarity in dead-specimen characters con- ceals important genetic divergence between P. gothica and P. zelicaon. Some details follow. F, hybrids of P. polyxenes 2? < P. gothica é (Figs. 38 and 39) have all the yellow markings more reduced on the upperside of the wings and on the abdominal sides than do the F, hybrids of P. polyxenes 2 > P. zelicaon 4 (Figs. 40 and 41). Also, these yellow wing markings are paler with gothica parentage and more ochreous 32 Postilla No. 119 from zelicaon parentage. For a brief account of the genetics of the broad-banded wing and striped abdomen of gothica and the dark wing and spotted abdomen of polyxenes, see my earlier paper (Remington, 1958). When bairdii is the tester, both combinations show wider yel- low postmedian bands than with polyxenes, but again the band is narrower in F, of gothica? < bairdii 6 (Figs. 42 and 43) than in F, of zelicaon@? bairdii 6 (Figs. 44 and 45), and the lateral spotting of the abdomen is greatly reduced in the hybrids with gothica parentage. Sex ratios of broods from combinations of three of these species are shown in Table |. Note that gothica parentage is nearly totally lethal for one sex, but that a substantial percentage of the deficient sex survives in hybrids with zelicaon mothers. Thus, the probability is greater than 95% that, in crosses with polyxenes males, the sex ratios of the two species are truly different. Table 2 shows in sections a and b the fertility and hatchability of eggs of F, crosses between P. gothica and P. zelicaon. Section c shows fertility, as well as developmental viability of eggs, of these F, hybrids mated with 3 kinds of non-hybrid relatives. Control data from pure gothica and pure (but probably inbred) zelicaon appear in sections d and e. Egg fertility and hatchability of several other hybrid combinations involving gothica or zelicaon are given for comparison in sections f through k. At present this large body of data may appear more confusing than illuminating. But a few observations are appropriate. First, note that in general the various F, hybrid adults tended to exhibit high fertility, but their offspring showed low embryonic viability. The small brood 207A suggests that this tendency applies to F, gothica zelicaon, and it will be of great interest to see whether it is confirmed in future back- crosses from such F, hybrids. Second, note that when non-hybrid females are mated to males of quite separate species, fertilizability and hatchability of eggs are commonly very high. Obviously, in studying hybrid fertility and viability it is essential to test the hybrids themselves, preferably by backcrossing. Of several hundred larvae of P. gothica which my associates and I have reared, all were plain yellow in their subdorsal spotting; less than ten of these were from field-collected eggs or small larvae, the rest from about twenty confined wild-caught females. The 1968 A NEW SIBLING PAPILIO 33 TABLE 1. Sex Ratios in Papilio F, Hybrids A. Brood data. Brood Parentage 1B 2 Fié é J 2Q #B-26(Ae) 2 gothica & & polyxenes 2) 0 1.00 #411 2 zelicaon < & polyxenes 101 11 90 #426 Q zelicaon 6 polyxenes 8 6 S57) #10 2 polyxenes 6 gothica 1 24 .04 #P-1-28(Ae) @ polyxenes * é gothica 0) 23 .00 eo i Be) 2 polyxenes & zelicaon 0 U 00 #417 @ polyxenes & 6 zelicaon 0 3 .00 B. Ratio comparisons (29 : 6). @ gothica < & polyxenes Di 0 2 zelicaon * & polyxenes NOON a7 2 polyxenes < 6 gothica 1a: Ay, Q polyxenes é zelicaon OF 10 * §. A. Ae obtained one intersex but no ? @ in this brood. smaller number of P. zelicaon studied (perhaps forty from wild larvae and eggs and two or three hundred from about six captive females) showed polymorphism for this character, every sample of several larvae including yellow-spotted, pale-orange-spotted, and red-orange-spotted individuals. These are probably the phenotypes, respectively, of the yellow homozygote, the heterozygote, and the red-orange homozygote of two alleles at a single locus (see Clarke & Sheppard, 1956). I have found the same polymorphism in samples of wild larvae of P. polyxenes in Connecticut and Mis- souri. In P. bairdii in the San Bernardino Mountains of California and P. machaon aliaska Scudder in the Brooks Range of Alaska I have found only the red-srange phenotype. Thus, P. gothica is unusual in having yellow fixed in the known populations. Postilla TABLE 2: No. 119 Egg Fertility and Embryonic Viability in Crosses of Papilio gothica, P. zelicaon, and Various Controls and Comparisons i Hatched Brood Eggs laid Colored (i.e., fertile) a. Papilio F, 2° gothica * é& zelicaon 3723 22 i | +90 20 19 18 totals: 42 26 19 colored / laid — .619 hatched /laid — .452 hatched/colored — .731 b. Papilio F, 2 zelicaon * & gothica #721 431 428 384 +719 382 379 364 #744 306 287 254 +746 157 140 112 #743 122 108 85 HT45A L10 LO9 97 totals: 1S08 1451 1296 colored / laid — .962 hatched/laid — .859 hatched / colored — .893 c. Crosses including F, gothica < zelicaon Papilio 2 zelicaon X 6 (F: 2 gothica X é zelicaon) 207A 25 P| 9 Papilio Q machaon melitensis X & (F: 9 gothica X é zelicaon) 224A 34 34 25 Papilio 2 (F: 9 gothica X é zelicaon) X é kahli #220A 39 xy 36 d. Papilio gothica — wild QQ already fecundated #705 450 411 39] #704 293 274 256 +7T09A PING) 197 186 +707 127 19 1S + 699,701,708 131 122 107 totals: 1218 1023 955 colored / laid — .840 hatched / laid — .784 hatched / colored — .934 1968 A NEW SIBLING PAPILIO TABLE 2 (continued) a5 Brood Eggs laid Colored (i.e., fertile) Hatched e. Papilio zelicaon — hand-paired (possibly sibs ) #749 396 365 348 #750A 360 112 32 #713 257/51 331 207 totals: 1131 808 587 colored /laid — .714 hatched/laid — .519 —hatched/colored — .726 f. Papilio F, ° polyxenes >< ¢ gothica #726 623 623 605 #700 280 241 234 #709B 61 59 56 totals: 964 923 895 colored /laid — .957 hatched/laid — .928 — hatched/colored — .970 g. Papilio F,; 2 polyxenes > 4 zelicaon #736A 479 477 438 #735A 472 471 459 #734A 298 291 275 totals: 1244 1239 72 colored/laid — .996 hatched/laid— .942 hatched/colored — .946 h. Papilio Fy, 2 machaon > é gothica #732 452 437 396 #722 406 404 390 #730 358 356 329 #733 340 331 299 #731 126 122 ILS) totals: 1682 1650 1529 colored /laid — .981 hatched /laid — .909 hatched/colored — .927 36 Postilla No. 119 TABLE 2 (continued ) Brood Eggs laid Colored (i.e., fertile ) Hatched i. Papilio F, 9 machaon « 6 zelicaon SET S7 398 39] 384 #720 136 86 80 758A 80 79 67 totals: 614 556 531 colored /laid — .906 hatched /laid — .865 hatched/colored — .955 j. Crosses including F; gothica « polyxenes Papilio F, 2 gothica ¥ & (Fi @ polyxenes X ¢ gothica) #87 5 5 5 Papilio F, 2 kahli * @ (Fi 9 polyxenes * ¢& gothica) #74 10 10 5 Papilio F: 2 polyxenes X & (Fi 2 polyxenes * & gothica) 82 15 14 14 Papilio? F: #82 X 6 gothica Baa) lg/p Pa 13 10 Papilio 9 zelicaon * & (Fi 9 polyxenes * 4 gothica) #102 11 11 0 eae LILY) 12 11 5 Papilio 2 machaon melitensis X & (Fi Q polyxenes * 6 gothica) #81 15 15 1S) k. Crosses including F, zelicaon « polyxenes Papilio 2 (F: 9 zelicaon X & polyxenes) * & polyxenes #413 70 67 55) Papilio 2 (F, 2 zelicaon * 6 polyxenes) X 6 zelicaon #418 93 90 uy Papilio 2 polyxenes X & (Fi 9 polyxenes X 6 zelicaon) #240 62 ? v7] Papilio 2 (Fi 2 zelicaon * & polyxenes) X 4 sib #416 198 184 ? #420 4] 4] ) 1968 A NEW SIBLING PAPILIO 27/ PHENOTYPIC STABILITY WITH PHYLETIC DIVERGENCE It has been shown that extreme phenotypic similarity in wing and body characters has been retained in Papilio gothica and zelicaon while genotypic differences evolved in allopatry. Analogous find- ings have been reported for a wide scattering of animals and plants in which different demes look alike in one or more characters but hybridization reveals differences in the genotypic control of these characters. A recently reviewed example is that of certain poeciliid fishes, Xiphophorus hellerii and X. montezumae, in which the two parental species have the usual greenish wild-type Swordtail coloration, but the F, hybrids and offspring of repeated backcros- ses to hellerii have the body bright red, sometimes with the caudal fin and its peduncle black (Kallman & Atz, 1966). This instance differs in detail from the gothica—zelicaon case in that F, hybrids between the two species of Papilio do not show major differences. Various authors have discussed the adaptive significance of maintaining a constant phenotype despite substantial genotypic change. One mechanism for this homeostatic regulation has been called “canalization” by C. H. Waddington, a concept recently extended experimentally by Rendel (e.g., 1968) and others. The extreme phenotypic similarity of many genetically well- differentiated pairs of species must be due to the maximal fitness associated with a stable phenotype. I consider it probable that a substantial percentage of “single species” of animals and plants presently well represented in taxonomic collections will prove to be a phenotypically stable group of two or more sibling species. Helped by the refinements of their taxonomic procedures and the relative ease of culturing their breeding stocks, the drosophi- lists have long explored their species at this level of taxonomic sophistication. No other group has been so elegantly analyzed, although some advances in this direction have been made by chromosomal studies (e.g., Erebia and plebejine butterflies by Z. Lorkovic and H. de Lesse, grasshoppers by M. J. D. White, coccinellid beetles by S. G. Smith, and simuliid flies by K. H. Rothfels ). In some of these phenotypically stable clusters, whose siblings are not yet sympatric, the museum worker’s “character displace- ment” may evolve when they eventually become partly sympatric. But I consider “displacement” unlikely in the hypothetically 38 Postilla No. 119 numerous instances in which the adaptiveness of phenotypic stability is not outweighed by the adaptive advantage of alteration of the visual phenotype to achieve the anti-hybridization and anti- competition sequelae to sympatry (see Remington, 1968, for formal discussion of the sequelae). For several years I have had under study no fewer than three abundant “species” of butterflies in Connecticut and four in Colorado, each of which is probably a pair of widely sympatric and fully speciated entities. As with Papilio gothica, I have delayed formal naming of these species in the hope of finding recognition characters useful for determining museum specimens. ACKNOWLEDGEMENTS The day-to-day maintenance of the many laboratory broods required for this study has been carried out largely by several research assistants, Roger W. Pease Jr. during several seasons, and Orley R. Taylor Jr., Bruce Baker, and Eric E. Remington for shorter periods. These individuals, and most extensively Eric Remington, collected many of the wild specimens of P. gothica. James Scott provided the large sample from Jarre Canyon of the plains-edge deme. The late John C. Hopfinger, Richard Guppy, and Frank P. Sala sent breeding stocks of crucial Papilio from north-central Washington, Vancouver Island, and southern Cali- fornia. Christopher Henne, Charles N. Rudkin, and Fred T. Thorne kindly donated essential specimens of P. rudkini, and James Oberfoell went to special pains to collect for me a large series of the fascinating polyxenes-machaon taxa resident in the Dakotas. Dr. S. A. Ae kindly gave me relevant hybrid sex ratio data. Dr. Frederick H. Rindge of the American Museum of Natural History and Mr. Lloyd M. Martin of the Los Angeles County Museum have kindly guided me to the fine holdings of the polyxenes-machaon complex in their institutions. I have also studied important specimens in the collections of J. Donald Eff and G. R. DeFoliart, and especially the collection of P. Sheldon Remington now in the Peabody Museum. Thomas Brown prepared the excellent photographs. Essential support for various parts of this research was provided by National Science Foundation grants 1968 A NEW SIBLING PAPILIO 39 G1739, G3830, and G23781; the field work at Riding Mountain was financed by grant no. 1884 from the American Philosophical Society. To all I am grateful. LITERATURE CITED Ae, S. A., 1965. A study of classification by interspecific hybridization in Papilio. Academia (Nanzan Univ.) 45-46: 221-237. Bauer, D. L., 1955. Notes on the Papilio machaon complex in Arizona. Lepid. News 9: 7-10. Brown, F. M., 1956. Colorado butterflies. Part IV. Pieridae, the whites and the sulphurs. Papilionidae, the swallowtails and parnassians. Denver Mus. Nat. Hist., Proc., 6: 177-236. Clark, A. H., 1932. The forms of the common Old World swallowtail but- terfly (Papilio machaon) in North America, with descriptions of two new subspecies. U. S. Nat. Mus., Proc., 81(11): 15 p., 8 pls. 1936. The swallowtail butterflies. Smithsonian Rept. 1935: 383-408. Clarke, C. A., and P. M. Sheppard, 1953. Further observations on hybrid swallowtails. Ent. Record Suppl. 65(9): 12 p. 1955. A preliminary report on the genetics of the machaon group of swallowtail butterflies. Evolution 9: 182-201. 1956. A further report on the genetics of the machaon group of swallowtail butterflies. Evolution 10: 66-73. Comstock, J. A., 1927. Butterflies of California. Comstock, Los Angeles. 334 p., 63 pls. Coolidge, K. R., 1924. California butterfly notes —II. Brooklyn Ent. Soc. Bull. 19: 44-47. Edwards, W. H., 1895. Notes on collecting butterflies in western Colorado, with a particular account of certain papilios. Canad. Ent. 27: 229-242. Eller, K., 1936. Die Rassen von Papilio machaon L. Bayer. Akad. Wiss. math.-nat. Abt., Abh. n.F. 36: 96 p., 16 pls. Fischer, E., 1908. Neue Tagfalterformen meiner Sammlung. Societas Ent. 23: 129-130. Gunder, J. D., 1928. Additional transitional forms. Canad. Ent. 60: 162- 168, 2 pls. Kallman, K. D., and J. D. Atz, 1966. Gene and chromosome homology in fishes of the genus Xiphophorus. Zoologica 51: 107-135, 6 pls. Langenheim, J. H., 1962. Vegetation and environmental patterns in the Crested Butte area, Gunnison County, Colorado. Ecol. Monogr. 32: 249-285. Maeki, K., and C. L. Remington, 1960. Studies of the chromosomes of North American Rhopalocera. L. Papilionidae. Lepid. Soc. Journ. 13: 193-203. Marr, J. W., 1961. Ecosystems of the east slope of the Front Range in Colorado. Univ. Colorado Studies (Biol.) 8: 134 p. 40 Postilla No. 119 Remington, C. L., 1956. Interspecific relationships of two rare swallowtail butterflies, Papilio nitra and Papilio hudsonianus, to other members of [the] Papilio machaon complex. Amer. Phil. Soc. Year Book 1955: 142-146. 1958. Genetics of populations of Lepidoptera. Xth Int. Con- gress Ent. Proc. 2: 787-805. 1968. Suture-zones of hybrid interaction between recently joined biotas. Evol. Biol. 2: 321-428. Rendel, J. M., 1968. The control of developmental processes, p. 341-349. In E. Drake [ed.] Evolution and environment: a symposium presented on the occasion of the centennial of the foundation of the Peabody Museum of Natural History at Yale University. Yale Univ. Press, New Haven. Ridgway, R., 1912 [1913]. Color standards and color nomenclature. Ridg- way, Washington. 44 p., 53 pls. Rothschild, W., and K. Jordan, 1906. A revision of the American papilios. Nov. Zool. 13: 411-752. Villalobos, C. and J., 1°47. Atlas de los colores. El Ateneo, Buenos Aires. 88 p., 38 pls. Wright, W. G., 1905. Butterflies of the West Coast. Whittaker and Ray, San Francisco. 257 p., 32 pls. HHA 3 2044 066 BOUND DEC 1972 Sas sak fag wal Pie FPPC EE Sa dl iil te Mine tal PF eee pans