HARVARD UNIVERSITY ‘i? Library of the Museum of Comparative Zoology I THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA The Lepidoptera Research Foundation, Inc. c/o Santa Barbara Museum of Natural History 2559 Puesta Del Sol Road Santa Barbara, California 93105 William Hovanitz Rudolf H. T. Mattoni, Editor Lorraine L. Rothman, Managing Editor Scott E. Miller, Assistant Editor Emilio Balletto, Italy Miguel R. Gomez Bustillo, Spain'!" Henri Descimon, France Thomas Emmel, U.S.A. Lawrence Gall, U.S.A. Brian 0. C. Gardiner, England Hansjuerg Geiger, Switzerland Otakar Kudrna, Germany Dennis Murphy, U.S.A. Ichiro Nakamura, U.S.A. Arthur Shapiro, U.S.A. Atuhiro Sibatani, Japan 'j'Deceased December 17, 1985 Manuscripts may be sent to the Editor at: 9620 Heather Road, Beverly Hills, CA 90210 (213) 274-1052 Notices Material may be sent to the Managing Editor. The JOURNAL is sent to all members of the FOUNDATION. CLASSES OF MEMBERSHIP: Regular (Individual) $ 15.00 year (vol.) Contributing $ 25.00 or more, year (vol.) Student/Retired — Worldwide $ 11.00 year (vol.) Subscription Rate/Institutions $ 25.00 year (vol.) Life $200.00 STATEMENT OF OWNERSHIP AND MANAGEMENT THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA is published four times a year, Spring, Summer, Autumn, and Winter, by THE LEPIDOPTERA RESEARCH FOUNDATION, INC. The office of the publication and the general business office are located at 2559 Puesta Del Sol Road, Santa Barbara, California 93105. The publisher is THE LEPIDOPTERA RESEARCH FOUNDATION, INC. The Editor is R. H. T. Mattoni at the above Beverly Hills address. The Secretary-Treasurer is Barbara Jean Hovanitz at the general business office. All matters pertaining to membership, dues, and subscriptions should be addressed to her, including inquiry concerning mailing, missing issues, and change of address. The owner is THE LEPIDOP¬ TERA RESEARCH FOUNDATION, INC. , a non-profit organization incorporated under the laws of the State of California in 1965. The President is R. H. T. Mattoni, the Vice President is John Emmel, the Secretary- Treasurer is Barbara Jean Hovanitz. The Board of Directors is comprised of Barbara Jean Hovanitz, Lorraine L. Rothman, and R. H. T. Mattoni. There are no bond holders, mortgages, or other security holders. ISSN 0022 4324 Published By: Founder: Editorial Staff: Associate Editors: The Journal of Research on the Lepidoptera 25(1):1-14, 1986 Hidden Genetic Variation in Agraulis vanillae incarnata (Nymphalidae) Thomas E. Dimock and Rudolf H. T. Mattoni 111 Stevens Circle, Ventura, California 93003, U.S.A. and 9620 Heather Road, Beverly Hills, California 90210, U.S.A. Abstract. Two culture lines of Agraulis vanillae incarnata were established from one wild-collected female. One line was mass selected for reduced black markings, the other for increased black markings. Both lines were maintained through seven generations, at which time the phenotypic differences between the lines diverged in response to selec¬ tion; a scale deformity also occurred among some individuals in the lightly marked culture. Some genetic aspects of the variation discovered are discussed. Introduction The purpose of the present work was to determine if any genes for albinic, melanic, or immaculate Agraulis vanillae incarnata (Riley) were carried by a field collected gravid female. These variants occur rarely in nature, and have been described as aberrants “hewlettii” Gunder (1930), “comstocki” Gunder (1925), “margineapertus” Gunder (1928), and “fumosus” Gunder (1927). The probability of randomly selecting a specimen carrying such gene(s) is admittedly small, and when it became obvious that simple single gene variants were not likely to be expressed in culture, the breeding program was modified to ascertain whether extreme opposite phenotypes could be produced in parallel cultures through mass selection. Since the color and pattern of A. vanillae throughout its range in southern California is quite constant, this program would provide infor¬ mation on the amount of hidden genetic variation in the taxon. In A. vanillae , which is orange with black markings on the upperside, the approach involved selecting adults toward an all-orange upperside in one culture line and an all-black upperside in the other. A final goal was to cross the selected lines in order to test whether the resulting hybrids would restore the normal phenotype, as might be expected in a complex polygenic system (Lerner, 1954, 1958). Although the original project never reached completion, the results obtained after seven generations are of sufficient interest to be presented here. 2 J. Res. Lepid. Mating and Rearing Protocol The original female was collected on 29 October 1982 in Ventura, California. She was confined for oviposition in a flight cage 51 x 51 x 122 cm with several water-potted cuttings of the larval foodplant Passiflora caerulea. The cage received afternoon sunshine plus light from a 75-watt GE Gro and Sho Spotlight after sunset. Although courtship and mating in this species occurs throughout the day in warm weather in natural conditions, in this indoor breeding program these activities were limited to the latter half of the day when the flight cage received more light. Smaller cages for mating single pairs of adults were made from card¬ board boxes measuring 23 x 32 x 32 cm. The top and sides were cut out, nylon netting glued over these sides, and a door cut out from the margin of one side. When a specific pair had been chosen and mating required confir¬ mation, the caged butterflies were left in a warm room with ambient light until noon, when they were taken out to a car. When placed on the front seat in sunlight (or occasionally on bright but overcast skies), and with an inside temperature of 24-30°C, matings almost always occurred. Opening one or both car windows to provide a slight breeze helped stimulate mating. Ovipositing females and their progeny were left in these small cages. Cut P. caerulea in water lasts up to two weeks, so new cuttings were added as the older ones began to decline or were consumed, and the larvae even¬ tually found their way onto the new plants. The large flight cage was also used as a rearing cage for broods of up to 800 larvae. During the last instars it was necessary to clean out the denuded vine stems and frass twice weekly. It was important to keep an adequate supply of foodplants readily available for the larvae, as they cannibalized pupae if these were discovered before foodplant. When most or all larvae had pupated, the cage was cleaned and twigs with prepupae or pupae were cropped to c. 8 cm and pushed into styrofoam mounted on a cardboard base. Eclosing adults were examined for characteristics desired for breed¬ ing and placed into appropriate cages. Less extreme phenotypes were saved as papered specimens. All others were liberated. Under the above conditions, the average time for one life cycle was 45 days. Thus the entire breeding schedule described here required IOV2 months. The Breeding Program A pedigree of the breeding program is shown in Figure 1 . Except for those instances where a single pair of adults was mated and their offspring reared separately, the majority of the culture lines involved several mixed pairs representing an extreme selected phenotype. Thus the term mass selection is used, as multiple individuals were involved in most crosses. The number varied in each generation, but was usually limited to the five 3 25(1) : 1-14, 1986 or ten lightest or darkest pairs. When more extreme phenotypes occurred late in the broods, their earlier less extreme counterparts and their ova were discarded or moved to a general mass rearing cage. Adults from broods 1, 2, 4 and 7 showed no significant variation from typical phenotypes. One-third of the pupae of brood 6 blackened and died, and the remainder discarded. Many adults in brood 5 were unable to fly properly, suggesting viability modifiers. They fluttered upside down on the cage floor and so were not used for breeding, with the exception of a dark female (2(5)DAD) which was mated to a slightly dark male from the mixed brood 2mix. No G3 or G4 descendants from this mating expressed the flight affliction. One female (3D/S DAD female #3) from the G3 of this line showed reduced silver on the hindwing underside and was mated to a male (3D/S D) from the dark culture (results below) . The 3D/S D male was mated to two other females. The hindwing upperside marginal chain pattern tended to break on the discal side in brood BB, a characteristic we call “broken bridges”. This line was inbred until the G3 adults were obtained (Figs. 41-42), then aban¬ doned due to lack of space. Extreme Light and Dark Lines Brood 8 was the major source of lightly marked adults used for selecting the “immaculate” phenotype. This line was maintained through the seventh generation. By the seventh generation, the upperside black spots in the forewing interspaces M3, Cul5 Cu2 and hindwing interspaces RS, M3 and at the base of RS-Ml5 were entirely absent in most specimens. The forewing marginal triangles and the hindwing marginal chain markings were also greatly reduced. However, the forewing discal cell markings did not respond to selection and remained normal in size. The gradual development of this phenotype is shown in Figures 2-16. The results sug¬ gest that different genes or sets of genes independently control these two sets of markings. A selection of dark adults from the mixed brood was the source of specimens for the dark phenotype. The remainder of specimens from the mixed brood and those from brood 3 were discarded. After four generations they did not exhibit facies as dark as the G2 dark mixed brood. From this G2 brood, the darkest specimens (3DD) were bred for one generation. Then, in the G4, offspring from the G3 female 3DD female #2 and G3 male 3D/S D were included in this brood. The dark line was then inbred until the G7 adults eclosed. The development of this phenotype is shown in Figures 17-28. (This line shares with the “immaculate” line the Pj female and Gj adults in Figures 2-4.) Variation in Undersurface Silvering The 3D/S D male (Fig. 37), which mated three times, displayed reduc¬ tion in the silver maculation on the hindwing underside. Development of 4 J. Res. Lepid. the silver markings was never a consideration in the selection of the light and dark phenotypes, but the presence of two females with similar reduc¬ tions in silver markings presented an opportunity to breed this variation. When the female 3D/S D female #1 (Fig. 38) was mated to this male, off¬ spring were as follows: 36 normal males, 52 normal females, one male with slight silver reduction, and three females with moderate silver reduction. When the same male was mated to female 3D/S DAD female #3 (Fig. 40), which also displayed reduced silver markings, the offspring (26 males and 20 females) were all silvered normally. The same male was mated to a nor¬ mally silvered dark female, 3DD female #2 (Fig. 39), and their offspring (35 males and 49 females) were also all silvered. Finally, from a mixed brood of several pairs of adults with reduced silver markings, the following offspring were obtained: 53 normal males, 58 normal females, and two males and four females with reduced silver markings. The partially unsilvered condition exhibited by several adults was usually not displayed by their offspring, thus the heritability of the character will remain in doubt until further controlled experiments can be performed. Comparison of the silver markings of the G7 light phenotypes (Figs. 29- 30) with those of the dark phenotypes (Figs. 31-32) show differences in development, especially with the “slipper-shaped” silver spot in inter¬ space RS on the hindwing. In the light phenotype the two halves of this spot nearly coalesce; in the dark phenotype the halves have become widely separated and smaller. Greasy-wing A variation having a scale deformity occurred in about 12 individuals of the G5 generation of the “immaculate” line. These variants were called “greasy-winged” (GW) because of their resemblance to specimens with wings smeared by body fluids. This scale deformity affected all scales, including those on the body. Examples of the deformity are shown together with scales from a normal specimen in the scanning electron microscope photographs in Figures 43-48. The SEM photographs show that in wild type individuals the pigmented scales differ in shape from silver scales (Fig. 42). At the highest magnifications, the pigmented scales show spaces between the ribs, while the silver scales in the inter-rib area appear solid. In GW individuals all scales are reduced in size and are narrower. Further, the ultrastructure (Figs. 45 & 48) is modified such that the inter-rib area of all scale types appears partially filled or plugged. The effect is apparently a breakdown of the diffractive properties of the scale surface, producing partial trans¬ parency. The source of GW variants was the pooled brood of 4imm adults. This brood consisted of ca. 10-15 pairs of normal adults, and their pooled G5 offspring consisted of ca. 100 normal “immaculate” phenotypes and ca. 12 GW adults (the 5GW and 6GW lines were established from these) . None of 5 25(1) :1-14, 1986 these “greasy-wings” was as extreme as those which occurred in later broods. However, extreme “greasy-wings” did result from matings be¬ tween normal “immaculate” G6 adults. The results of further crosses of the 4imm line to show inheritance of GW follow: GW x GW (mass mating) 2 0 64 66 From above, Normal x Normal G6 (pair) Normal male 19 Normal female 23 GW male 17 GW female 7 G4 offspring 100 normal: 12 GW Normal x Normal G5 (pair) Normal male 12 Normal female 10 GW male 9 GW female 7 Extreme GW examples are illustrated in Figures 33 & 36. The trans¬ parency of the wings is indicated by the striped paper placed under the wings of the specimen in Figure 34. Data from both G5 and G6 pairings indicate the character is autosomal and probably digenic, resulting from the interaction of paired non-linked complementary genes. Thus a cross of two heterozygote wild types would be expected to produce a 9+ :7 GW ratio. The pooled data give a better fit (X2 = 1.4, df = 1, p >.25) to 9:7 than to the 3:1 ratio for a single recessive (X2 = 10. 1 , df = 1 , p < .001 ) . F urther , if GW were due to a simple recessive, the phenotype should have appeared in the G2 or G3 generations. The “greasy- wing” scale deformity probably decreases fitness, and would appear to be strongly selected against in nature. Meconium dis¬ charged from eclosing adults was not repelled by individuals with the GW wings, but rather stuck to them and dried, sometimes causing wings to stick together. Other specimens had lesions on the wings which oozed body fluids at the time of wing expansion. Individuals that were able to expand their wings successfully behaved normally; but because of the non¬ repellent nature of their wings, they would most likely experience dif¬ ficulty in humid or rainy conditions. The cultures were terminated at the end of August 1983 for two reasons. First, the abandoned orchard which had become completely overgrown with Passiflora caerulea was cleared for development, eliminating the foodplant resource, and an artificial diet was not further available. Second, TED developed a bronchial irritation from constant exposure to the culture, which was maintained in his living quarters. 6 J. Res. Lepid. Discussion The exploratory work reported here clearly shows that a significant amount of potential genetic variation was present but masked in a single mated female of a phenotypically constant butterfly. The variants pro¬ duced during the course of seven generations of inbreeding and selection included individuals with substantially greater or lesser quantities of melanin than typical A. uanillae , individuals with the scale anomaly greasy-wings, and individuals with behavioral modifications. The genetic systems producing these effects range from what appears to be complexes of polygenes controlling general wing pattern density to a digenic men- delian pair of genes for greasy-wings. Little research on inbreeding and mass selection has been reported in the Lepidoptera (Robinson, 1971), although inbreeding and selection is a widely used technique to determine the amount of genetic variation in organisms (Lerner, 1954, 1958; Lewon- tin, 1974). However, note Oliver’s, 1981, work on inbreeding depression and some early work by Schrader, 1911. Oliver (1981) showed genetic variability in terms of lethal equivalents in 12 lepidopteran species. To the extent that the mated female tested is representative (a subse¬ quent abbreviated three generation experiment involving another female produced both light and dark trending individuals, as well as the “broken bridges” phenotypes) one might ask: why are such variants not observed more frequently in nature? Two named aberrants “comstocki” and “margineapertus” occasionally turn up in collections. These aberrations closely resemble the extreme dark and light individuals selected in this breeding program. Our results imply the genetic control of wing melanin is not based on a few simple mendelian genes at the simultaneous recovery of both “comstocki” and “margineapertus” types from a single individual would be quite unlikely given the rarity of these aberrants in nature. An albino aberrant “hewlettii” occurs very rarely in nature. The “greasy- wing” trait, reported above, has not been reported previously. The data reported here lead us to conclude that substantial heterozygosity for wing character variants exist in natural populations of A. vanillae. The magnitude of genetic variation we extracted from a single mated individual has implications to both conservation and systematics. In con¬ servation the increasing use of captive breeding programs cannot over emphasize the necessity of attempting to utilize a selective and breeding scheme which maintains wild type individuals, while sequestering variance. In systematics, the results are interesting as applied to butter¬ flies, since butterfly taxonomy at both species and subspecies levels is based largely on wing character states which involve minor changes. The range of variants mass selected from our single female could well represent several different subspecies, if not species, if found fixed in natural populations. 25(1) : 1-14, 1986 7 Acknowledgments. We are deeply indebted to Arthur M. Shapiro, Department of Zoology, University of California at Davis (UCD), for his advice and suggestions for the improvement of the original manuscript and for his remarks on the “greasy- winged” condition. Rick Harris, UCD, generously provided the SEM photographs. Our thanks to Sue Mills for proofreading the manuscript. Literature Cited COMSTOCK, J. A., 1927. Butterflies of California, publ. by author. Los Angeles. 334 pp. GUNDER, J. D., 1925. Several new varieties of and aberrant Lepidoptera from California. Ento. News 36:5-6. _ , 1927. New transition form or “abs” and their classification. Ento. News 38:129-138. _ , 1928. Additionl transition forms. Can. Ento. 60:162-168. _ , 1928b. New butterflies and sundry notes. Bull. Brooklyn Ento. Soc. 24:327-328. LERNER, I. M., 1954. Genetic homeostasis. Wiley, New York. 134 pp. _ , 1958. The genetic basis of selection. Wiley, New York. 298 pp. LEWONTIN, R. C., 1974. The genetic basis of evolutionary change. Columbia University Press, New York. 346 pp. OLIVER, C. G., 1981. A preliminary investigation of embryonic inbreeding depres¬ sion in twelve species of Lepidoptera. Jr. Lep. Soc. 35:51-60. ROBINSON, R., 1971. Lepidoptera genetics. Pergamon, Oxford. 687 pp. SCHRADER, w., 1911. Inbreeding of Junonia coenia under high temperatures through twenty-two successive generations. Pomona Jr. Zool. & Ento. 4: 673-677. I Pedigree of mass selection breeding project using Agraulis vanillae incarnata. Single sex symbols indicate one pair of adults were mated to obtain progeny; double sex symbols indicate two or more pairs of adults. Numbers indicate filial generation. Abbreviations : BB Broken Bridges; D Dark; DAD Dark and Disabled; DD Darkest Darks; D/S Diminished Silver; GW Greasy-winged; IMM immaculate phenotype; MIX Large number of randomly mated individuals; ND Next Darkest. 25(1):1-14. 1986 9 10 J. Res. Lepid. 25(1):1-14, 1986 11 12 J. Res. Lepid. 48 14 J. Res. Lepid. Selectively bred adults of Agraulis vanillae incarnata. Males on left, females on right, except where noted otherwise. Fig. 2. wild P1 female. Figs. 3, 4. Gr Adults bred for reduced black markings (“immaculate” phenotype): Figs. 5, 6: G2; Figs. 7, 8: G3; Figs. 9, 10: G4; Figs. 11,12: G5; Figs. 13, 14: G6; Figs. 15, 16: G7. Adults bred for increased black markings (dark phenotype): Figs. 17, 18: G2; Figs. 19, 20: G2; Figs. 21,22: G4; Figs. 23, 24: G5; Figs. 25, 26: G6; Figs. 27, 28: Gr Figs. 29, 30: Undersides of “immaculate” phenotypes in Figs. 15 and 16. Figs. 31, 32: Undersides of dark phenotypes in Figs. 27 and 28. Figs. 33, 34: “Greasy-winged” scale deformity. Male F6, female F?. Figs. 35, 36: Undersides of Figs. 33 and 34. Fig. 37. G3 male 3D/S. Left side ventral. Fig. 38. G3 female 3D/S D£#1. Left side ventral. Fig. 39. G3 female 3DD$#2. Left side ventral. Fig. 40. G3 female 3D/S D£#3. Right side ventral. Figs. 41, 42. G3 “broken bridges” phenotype. Fig. 43. Scanning electron microscope photograph of the wing underside of a normal female, brood 7imm. Magnification 160 X. Dark brown scales at upper left, silver scales at lower right. Fig. 44. Same as Fig. 43, magnification 640 X, dark brown scales. Fig. 45. Same as Fig. 43, magnification 2500 X. Fig. 46. Wing underside of a “greasy-winged” female, brood 7immGW. Magni¬ fication 160 X. Dark brown scales. Fig. 47. Same as Fig. 46, magnification 640 X. Fig. 48. Same as Fig. 46, magnification 2500 X. Note: The photographs of the adult butterflies were all shot at f5.6 at 1 /250th of a second exposure. Varying developing exposures by the automated commer¬ cial processing equipment has resulted in photographs in which the orange ground color of the butterflies appears to vary in brightness, which in reality is not the case. Both males and females of the “immaculate” phenotype are consis¬ tently bright orange. Dark phenotype males are slightly deeper orange, and their females are auburn-orange. The Journal of Research on the Lepidoptera 25(l):15-24, 1986 Electrophoretic Evidence for Speciation within the Nominal Species Artthocharis sara Lucas (Pieridae) Hansjurg Geiger and Arthur M. Shapiro Department of Zoology, University of California, Davis, California 95616 Abstract. The taxa Anthocharis sara and A. Stella in northern California are shown to be differentiated at the species level, using electrophoretic genetics of both allopatric and parapatric populations. Both are also strongly differentiated from a sample of Colorado A. julia. Introduction Taxonomists confronted with sets of apparently closely-related, allopat¬ ric entities are usually forced to decide on purely morphological grounds whether to call them species or subspecies. Occasionally their judgment can be put to test when genetic information becomes available on the entities in question. Since the discovery of sibling speciation, it has been generally recognized that there is no a priori correlation of morphological differentiation and barriers to gene flow. The outcome of such genetic tests, thus, is frequently surprising. Anthocharis sara was described by Lucas in 1852, presumably from somewhere near San Francisco, California. Its “subspecies” of current usage, Stella W. H. Edwards, 1879 and julia W. H. Edwards, 1872, were described from Nevada (type locality restricted to Marlette Peak, Carson Range, Washoe Co., by F. M. Brown, 1973) and Colorado (type locality restricted by Brown, loc. cit., to Beaver Creek, Park Co.). The present study of the A. sara complex was undertaken when one of us (AMS) ob¬ served an unusual pattern of interaction in the geographic distributions of the northern California taxa — a pattern which suggested that sara sara and sara “stella” might in fact be full species. Anthocharis sara sara is distributed in the Central and North Coast Ranges, the Yolla Bollys, the Siskiyou Mountains (including the Trinity Alps), the Cascades north of Mount Shasta, the Sierra Nevada foothills and lower montane zone on the west slope, and in Sierra Valley on the east slope at 1500m, 40 km N of Truckee. In northern California outside the Sierras, it reaches at least 2000m. On the Sierran west slope, AMS has done regular sampling at a series of stations in the South Yuba river coun¬ try since 1972. At the lowest of these, Washington (803m), only sara sara 16 J. Res. Lepid. has been seen. At Lang Crossing (1500m) neither sara nor Stella appears to be a permanent resident, but both have been taken with about equal fre¬ quency and no sign of intergradation. At Donner Pass (2100m), Stella is a permanent resident and sara has been recorded three times; at Castle Peak (2750m) sara was seen twice. At Truckee (ca. 1800m), on the east slope, only Stella occurs. That sara occasionally intrudes at Donner Pass was noted by Emmel and Emmel (1962, p.30), who wrote that “males identical to typical white reakirtii were occasionally taken in fresh condi¬ tion” (“reakirtii” Edwards being a spring form of sara). The suspicious components of this distribution are: i) the replacement of Stella by nominate sara at high altitudes outside the Sierra; ii) the fluctuating altitudinal range at Sierran mid-elevations, without apparent intergrada¬ tion (Table 1); and iii) the close juxtaposition of Stella with nominate sara north of Truckee, in an apparent Great Basin habitat (juniper woodland and meadows with a characteristic Basin butterfly fauna). We therefore decided to seek electrophoretic evidence bearing on the probability of gene flow and the degree of genetic differentiation among accessible pop¬ ulations. Colorado A. “sara” julia was brought into the study as an independent geographic comparison because a sample was available; we had no predictions concerning its status. Materials and Methods Samples were collected as listed in Table 2; California localities are shown in Fig. 1. All animals were transported alive and immediately stored at -70° C until elec¬ trophoresis. Only 1984 and 1985 catches were used. The head and thorax of each individual were homogenized in 4 volumes of Tris- HC1 buffer (0.05 M, pH 8.0). Horizontal starch gel electrophoresis was used, following slightly modified standard procedures (Ayala et. al., 1972; Geiger, 1981). Twenty enzymes were scored: adenylate kinase (loci AK-1 and AK-2), aldolase (ALD), arginine kinase (APK), fumarase (FUM), glutamate -oxaloacetate trans¬ aminase (GOT-1, GOT-2), glutamate-pyruvate transaminase (GPT), glyceral- dehyde-phosphate dehydrogenase (GAPDH), oc-glycerophosphate dehydrogenase (dx-GPDH), indophenol oxidase (IPO), isocitrate dehydrogenase (IDH-1, IDH-2), malate dehydrogenase (MDH-1, MDH-2), malic enzyme (ME-1), phosphoglu- comutase (PGM), 6-phospho-gluconate dehydrogenase (6-PGD), phosphoglucose isomerase (PGI), and pyruvate kinase (PK). The genetic interpretation of the zymograms is based on the analysis of the pro¬ geny of parents with various phenotypes at each polymorphic locus in Pieris brassicae (L.) (Geiger, 1982). No deviation from the pattern observed in P. brassicae has been found in any of the three taxa investigated here. However, there is some evidence for sex-linked inheritance of the very weakly polymorphic 6-PGD in Stella (no polymorphism has been detected in female sara or julia). As this is quite speculative, it has been neglected in the calculations of allelic frequencies; this treatment does not affect any of the conclusions of this paper. The designation of the alleles indicates the difference in the mobility of the enzyme relative to the most frequent electromorph found in P. brassicae (index 100). An allele 95, then, codes for an enzyme that migrates 5 mm less than the P. 25(l):15-24, 1986 17 Fig. 1. Localities of Anthocharis samples studied. Abbreviations as in Table 2. brassicae variant. The allelic frequencies (Tables 3 and 4) have been used to calculate the statistic I (Nei, 1972). These values have then been used to construct a dendrogram (Fig. 2) by cluster analysis (UPGMA method, see Ferguson, 1980). Results The same electromorphs (treated as alleles) occur in all individuals of all three taxa at nine of the 20 loci investigated (AK-1, AK-2, ALD, APK, FUM, GPT, GADPH, IPO, IDH-2). At four other loci (GOT-2, oc-GPDH, 6-PGD, PK) very infrequent polymorphism is observed (frequency of the common allele >95%, with the exception of the Donner Pass sample ( Stella ) at the 6-PGD locus, fcommon aiieie = 85%). All samples of all three taxa share the same common allele for these loci. Variation within and/or 18 I-values 10 -i 0-9 - 0 8 - 07 Stella / ' / _a>

o O a o "a> GO >> Q) £ a) CO o -o o o o o ttt— 1 99i , ; 99 , .00 i •97 i •91 83 Fig. 2. Dendrogram representing degree of relationship among Anthocharis populations for which large samples are available. between the three taxa was found at seven loci (GOT-1, IDH-1, MDH-1, MDH-2, ME-1, PGM, PGI). The allelic frequencies at these loci are pre¬ sented in Tables 3 and 4 for all samples with at least five individuals and for pooled samples of the three taxa. At three loci (GOT-1, MDH-1, PGI) most alleles detected in sara with frequencies >10% are also found in Stella (Table 3) . The two taxa show only small differences in the allelic fre¬ quencies at these three loci. This is also true for the observed variation within the two taxa, with the exception of the Sierra Valley sample of sara. In this sample the allele 98 is the common allele at GOT-1, with a fre¬ quency of 67% (Table 3) . Only a very low level of polymorphism is recorded in our julia sample at these three loci. The common alleles reach very high frequencies but appear identical with the common alleles in sara and Stella. The situation is different at four other loci (IDH-1, MDH-2, ME-1, PGM) (Table 4). Statistically significant differences occur at all four loci among the three taxa (P<1%). The IDH-1 allele 72 is found at 97% in sara and 100% in julia but only 3% in Stella. The common allele in Stella at the IDH-1 locus is the allele 82 that is found at 3% in sara but not in julia. At the MDH-2 locus the allele 91 is monomorphic in all sara and Stella sam- 19 25(1) :15-24, 1986 pies, but an allele 94 is monomorphic in julia. Sara and Stella share the same polymorphism at the ME-1 locus and in both taxa, allele 100 is the common allele. The allele 103 that reaches 19% in sara and 9% in Stella is the common allele in julia , with a frequency of 100%. At the PGM locus, alleles 97, 103 and 111 are observed with frequencies >5% in sara. Only allele 97 occurs in julia , and only at very low frequency. The three most common alleles in Stella (90, 105, 113) are not recorded in sara and julia at all. The common allele in julia (88) is found at low frequency in sara, and not at all in Stella. These data show a low degree of differentiation within the taxa, even over substantial distances and in different climatic regimes (sara), but a much higher degree between taxa. The quantified data are presented as I- values in a dendogram (Fig. 2). Overall genetic differences within Stella are small (I- values ^_0. 99). A very similar degree of divergence occurs be¬ tween the sara samples, despite their wider geographic separation. Within sara, near-coast samples are more similar to one another than to Sierran ones (Skelton Canyon, west slope; Sierra Valley, east), as would be pre¬ dicted. All the within- taxa comparisons are similar to values obtained within other Pierid taxa at morphospecies level (Geiger, 1981; Geiger and Scholl, 1982a, 1982b, 1985). The genetic differences between the taxa are much more pronounced, and similar to those observed between morpho¬ species of Pieridae (references as above) . The degree of heterozygosity is remarkably low in julia (Hobs =0.028, Hexp =0.019). The values for sara (Hobs =0.091, Hexp =0.117) and stella (Hobs =0.107, Hexp =0.120) are clearly higher. Discussion Low genetic differences among local populations within sara and stella are good indicators of either contemporary or recent gene flow. The situa¬ tion is very different when these two taxa are compared, even over short geographic distances. The Sierra Valley population of sara, which is 40 km north of the Truckee stella population (and only about 14 km from the nearest known stella, at Yuba Pass), is somewhat different from other sara samples but not in any way that suggests any gene exchange with stella’, to the contrary. At two loci (IDH-1, PGM; Table 4) the two taxa only very infrequently have the same alleles in common, and at PGM the com¬ monest allele in each taxon is completely unknown in the other. These are unambiguous indicators of a lack of gene flow between the taxa. As Table 1 shows, the opportunity for contact exists at least in the South Yuba River country and probably elsewhere. We have never, however, found any specimen intermediate between sara and stella either in the wild or in collections, nor do we know of any permanent population (as contrasted with the Lang Crossing case) in which both coexist. Are sara and stella distinct species, then? In the absence of breeding- compatibility data such a claim may seem premature, but their level of 20 J. Res. Lepid. genetic differentiation is quite normal for Pierid morphospecies; to put it another way, the decision to rank them as subspecies rather than species has been based on a perceived low level of morphological differentiation, which may not be commensurate with genomic differentiation. They are kept apart by a narrow elevational band at mid-elevations on the Sierran west slope in which both may colonize but neither appears capable of per¬ manent establishment. That this band is not “simply” a consequence of habitat selection is shown by the fact that sara replaces stella in very similar habitats and plant associations at high elevations in the Trinity Alps (Shapiro, Palm, and Wcislo, 1981) and the Cascades north of Mount Shasta (Ball Mountain). The nature of the exclusion from mid-elevations on the west slope needs further study. It is duplicated with remarkable precision in at least two other difficult groups: Phyciodes pratensis Behr/ montana Behr (Nymphalidae) and Polites sabuleti Bdv. /tecumseh Grin¬ ned (Hesperiidae) . The genetic differences are even more pronounced between sara/stella and Colorado julia. This julia population possesses an MDH-2 allele so far unknown in the other taxa; at the PGM locus it shares a common polymorphism with sara but with a different common allele. Given the wide range of the taxon julia (Wyoming to New Mexico) and the complex variability of the sara complex in the Rocky Mountains and Great Basin, it is certainly premature to say too much — except that, on the face of things, julia looks genetically like a well-defined morphospecies. The average heterozygosity for sara and stella is typical for Pierid species (Geiger, unpublished data) and only a little lower than for inver¬ tebrate species in general (H =0.134; Ayala, 1984). Julia is extraordinarily homozygous, however. This could be due to sampling error (n=9), although this value seems not to be affected by similar or even smaller numbers in our sara and stella samples (e.g., sara, Big Bar, n=5, Hobs 0.124; stella, Castle Peak, n=ll, Hobs =0.102). If the low value (H0bs. =0.028) is not a sampling artifact, it could be due to (i) recent origin of the species, (ii) a recent bottleneck for either the species or the local pop¬ ulations, (iii) founder effect, (iv) low effective population size, (v) strong selection, or some combination of these and other factors. These matters cannot be resolved until more information is obtained on the genetic struc¬ ture of julia populations from different parts of its range. This, in turn, is a prerequisite for determining its precise taxonomic standing vis-a-vis not only sara and stella but the six other named entities of the sara complex. At the same time, re-examination of the morphological characters in the complex and the criteria for weighting seems in order, as do compatibility experiments and a careful comparison of both the standard and micro¬ morphology of the early stages. Acknowledgments. We thank Francisco J. Ayala for permitting the use of his facilities, and Oakley Shields, Adam Porter, Jane Hayes, and Steve Courtney for 21 25(l):15-24, 1986 contributing material. HJG’s work at Davis was supported by National Science Foundation grant BSR-8306922 (Systematic Biology Program) to AMS. This paper forms part of California Agricultural Experiment Station project CA-D*- AZO-3994-H, “Climatic Range Limitation of Phytophagous Lepidopterans,” AMS, Principal Investigator. Literature Cited AYALA, F. J., J. R. POWELL, M. L. TRACEY, C. A. MOURAO & S. PEREZ-SALAS, 1972. Enzyme variability in the Drosophila willistonii group. IV. Genic variation in natural populations of Drosophila willistonii. Genetics 70:113-139. BROWN, F. M., 1973. The types of the butterflies described by William Henry Edwards: Pieridae. Trans. Amer. Ent. Soc. 99:41, 44. EMMEL, T. C. & J. F. EMMEL, 1962. Ecological studies of Rhopalocera in a high Sierran community — Donner Pass, California. I. Butterfly associations and distributional factors. J. Lepid. Soc. 16:23-44. FERGUSON, A., 1980. Biochemical Systematics and Evolution. Blackie, Glasgow and London. GEIGER, H. J., 1981. Enzyme electrophoretic studies on the genetic relationships of Pierid butterflies. I. European taxa. J. Res. Lepid. 19:181-195. _ , 1982. Biochemisch-genetische Untersuchungen zur Systematik und Evolution von Weisslingen des europaischen Faunengebietes. Ph.D. thesis, University of Bern. GEIGER, H. J. & A. SCHOLL, 1982. Enzyme electrophoretic approach to the systema¬ tics and evolution of the butterfly Euchloe ausonia. Experientia 38:927-928. _ , 1982b. Pontia daplidice in Siideuropa — eine Gruppe von Zwei Arten. Mitt. Schw. ent. Ges. 55:107-114. _ , 1985. Systematics and evolution of holarctic Pierinae: an enzyme electrophoretic approach. Experientia 41:24-29. NEI, M., 1972. Genetic distance between populations. Am. Nat. 106:283-292. SHAPIRO, A. M., C. A. PALM & K. L. WCISLO, 1981. The ecology and biogeography of the butterflies of the Trinity Alps and Mount Eddy, northern California. J. Res. Lepid. 18:69-152. Table 1. Records of Anthocharis sara sara and A. “sara” Stella in the South Yuba River country, northern Sierra Nevada, 1972- 1985. Washington, Nevada Co., 803 m: sara sara only, uncommon. Lang Crossing, Nevada Co., 1500 m: sara sara: 29.iv.74, 15.vi.74, 18. V.75, 15.vi.78; “sara” Stella: 2.vi.74, 9.vi.75, 17.iv.77, 6-8.V.84, 19. V.84. Donner Pass, Placer-Nevada Cos., 2100 m: sara sara: 2.vii.75, 15.vi.77, 13.vii.77; “sara” Stella abundant all years. Castle Peak, Nevada Co., 2750 m: sara sara: 30.vi.72, 8.vii.77; “sara” Stella all years, scarce to abundant. 22 J. Res. Lepid. Table 2. Samples of the Anthocharis sara complex used in this study. Abbreviations are as in Fig. 1. California sara sara: Trinity-Siskiyou Mountains: Trinity County, Big Bar (BB), Hwy. 299, 37 km W Weaverville, 475 m, 5.V.1985 (n=5). North Coast Ranges: Napa County: Turtle Rock (TR), Hwy. 128 near Lake Berryessa, serpentine, 160 m, 17.iii.1984 (n=l). Solano County: Gates Canyon (GC), Vaca Hills above Vacaville, 250-500 m, 20.iii.1984 (n=9), 4.iv.l985 (n=3). Cascade Range: Siskiyou County: Little Shasta Meadow (LM), jet. USFS roads 47N03 and 40N09, Ball Mountain, 2000 m, 12. vi. 1985 (n=3). East Slope Sierra Nevada: Sierra County: Sierra Valley (SV), Hwy. 49, 4 km NE Sierraville, 8. v. 1984 (n=6). West Slope Sierra Nevada: Mariposa County: Skelton Canyon (SK), 1200 m, 9.V.1984 (n=6). Eldorado County: 7 km S Coloma (CO), 300 m, 11. v. 1984 (n=l). California “sara” Stella: West Slope Sierra Nevada: Nevada County: vie. Lang Crossing (LC), USFS road 18N18at South Yuba River, 1500 m, 8.V.1984 (n=2). Nevada + Placer Counties: Donner Pass (DP), Hwy. 40, 2100 m, 27.V.1984 (n=3), 6.vi.l985 (n=10). Crest, Sierra Nevada: Nevada County: Castle Peak (CP), 2700 m, 6.vi.l984 (n=10), 25.vii.1985 (n=l). Eldorado County: Red Lake Moun¬ tain (RM), Carson Pass, 3000 m, 29.vi.1985 (n=l). East Slope Sierra Nevada: Nevada County: Truckee (TE), 1700 m, 8. v. 1984 (n = 17). Colorado“sara” julia: Grand County: Willow Creek Cyn., 3.vii.l984 (n=9). number of animals investigated 25(l):15-24, 1986 23 G X g G G G £ a> -Q G CO G 1g CO 00 CM pp Ph PP Cl P O G E G E co G O G Pi G G G PQ CO u ■u CO G i—l CO a; Pi t — i G 4J U 00 ■u CD G r-H C CO G rH •H G •H i—( O G Pi r— 1 PQ O CO CO < Q CJ H < G i—l G CO i—l •H U a) i—l CO p» 3 CO co *r-> number of animals investigated 24 J. Res. Lepid. CO x cC <3 o> . e> o m nO NO ph •— • • * * co o nO 00 m co r- o > NT co co o • on • ■ on O O o • r-H * rH r— ' C in CNJ NO nO NT CO NT ON rH CO rH rH NT a a) o rH CO CO r^ CO >> r—l 0) CO cO 0) c CO T— 1 CO 0) i — 1 cO > Or PH Ph cx Vj u c E OJ E cO cO o CO u 0) -i 0) rH cO 4J 3 CO CO •«”» The Journal of Research on the Lepidoptera 25(l):25-29, 1986 Genetic Differentiation Between Subspecies of Euphydryas phaeton (Nymphalidae: Nymphalinae) A. Thomas Vawter1 and Janet Wright2 department of Biology, Wells College, Aurora, New York 13026 2Section of Ecology and Systematics, Cornell University, Ithaca, New York 14853 Introduction The checkerspot butterfly Euphydryas phaeton inhabits eastern North America from the maritime provinces of Canada south to Georgia and west to Missouri (Masters, 1968; Bauer, 1975). It is the only species of the genus that occurs in this region, and thus, represents a biogeographic pat¬ tern different from that of its congeners in the west, which have ranges that are generally overlapping and in some cases of limited extent. Although E. phaeton is clearly distinct from the western species and does not show the extreme phenotypic variation that some of them do, two subspecies have been described. Euphydryas phaeton phaeton (Drury) occurs in the northern portion of the species’ range where it typically inhabits marshy meadows and similar moist habitats favored by its larval foodplant Chelone glabra (Scrophulariaceae); E. p. ozarkae (Masters) occurs to the south and southwest and favors drier upland forested habitats where it reportedly feeds on Gerardia (= Aureolaria: Scrophulariaceae) (Masters, 1968). Bauer (1975) reports that E. p. ozarkae feeds on Lonicera and that larvae from eggs deposited on Lonicera die when transferred to Chelone , and those from Chelone die when placed on Lonicera . He suggests that this larval foodplant intolerance be used as a basis for dividing the taxa. D. Bowers (personal communication) feels Bauer (1975) is in error; she reports that E. p. ozarkae feeds naturally on Gerardia spp., although both it and E. p. phaeton will accept Lonicera and survive on it. Furthermore, E. p. phaeton can be reared equally well on Gerardia or Chelone , but E. p. ozarkae does significantly better on Gerardia. Gerardia- feeding pop¬ ulations apparently also occur in upland habitats in some areas of New York state (Shapiro, 1975). Although these two recognizable groups of populations are most often treated as subspecies, the marked ecological differences between them and the apparent overlap in their geographic ranges suggests the possibility that they may be sibling species. Here we report the results of our study of genetic differentiation between 26 J. Res. Lepid. E. p. ozarkae from Missouri and E. p. phaeton from central New York. Materials and Methods Samples of Euphydryas phaeton were collected in the summer of 1982 from three areas in central New York and a single area in eastern Missouri. The New York collections were made near Slaterville Springs, Tompkins Co. (N=33); at the Oneonta Airport, Otsego Co. (N=30); and near Milford, Otsego Co. (N=26). The Missouri collection (N=28) was made at Merramec State Park, Franklin Co. The New York populations inhabited wet meadows; the Missouri population inhabited mesic woodland. All butterflies collected were stored in liquid nitrogen prior to electro¬ phoretic analysis. Allozyme variation was assayed at 25 presumptive gene loci, following the methods of May et. al. (1979) . Details of electrophoretic methods and a table of electromorph frequencies are available from ATV on request. Electromorphic frequencies were calculated from direct counts of the elec¬ trophoretic phenotypes. Nei’s (1972) measure of genetic similarity was used to quantify genetic differentiation between populations. Results There are very few differences in electromorph frequencies among the 3 New York and 1 Missouri populations of E. phaeton we examined. The average heterozygosity per locus is 0.116 + 0.019 (mean + S.E.) and the proportion of polymorphic loci is 0.80. Log-likelihood tests for hetero¬ geneity in electromorph frequencies at each of the 25 loci (Sokal and Rohlf, 1981) illustrate the fundamental genetic similarity among the four pop¬ ulations. At only one locus (MPI) is there a heterogeneity significant at the p=0.05 level, and one expects to find such heterogeneity at the 0.05 level incorrectly in one in 20 such tests. The genetic identities (Nei, 1972) further illustrate the similarities among the populations (Table 1). The three New York populations attributed to E. p. phaeton are somewhat more similar to each other (ave. 1=0.989) than any of them is to the Missouri population attributed to E. p. ozarkae (ave. 1=0.967), although all four populations are quite similar. The average genetic identity between E. p. phaeton and E. p. ozarkae that we report here is slightly less than that reported by Brussard et al. (1985), although their value (ave. 1=0.991) was determined by electrophoresis of some of the same specimens. The discrepancy is due to a number of fac¬ tors. We examined more specimens, especially of E. p. phaeton, but we used only 25 loci rather than the 28 they used. We felt on our further analysis that we could not score all loci with confidence. We also made some minor changes in scoring some of the loci we retained. All of these changes are minor, and none alters the conclusions made in the earlier work. 27 25(l):25-29, 1986 Table 1. Nei (1972) genetic identities and their standard errors (in parentheses) between three populations of E. p. phaeton from New York and one population of E . p. ozarkae from Missouri. Nei’s index has a value of 1.0 for two populations that share all alleles at the same frequency, and a value of 0.0 for two pop¬ ulations that have no alleles in common. Abbreviations for the localities are as follows: MO = Merramec State Park, MO; NY1 = Slaterville Springs, NY; NY2 = Oneonta, NY; NY3 = Milford, NY. NY1 NY2 NY3 MO 0.977(0.018) 0.968(0.024) 0.956(0.033) NY1 — 0.990(0.004) 0.989(0.006) NY2 — — 0.988(0.006) Discussion Lack of differentiation at allozyme loci does not preclude the possibility that the populations in question are reproductively isolated and therefore “good” species; in the absence of other evidence that isolation exists, however, it seems very unlikely that populations that are genetically so similar represent separate species. Sibling species in Lepidoptera for which data are available are clearly more different than these populations of E. phaeton. Angevine and Brussard (1979) analyzed differentiation at allozyme loci in populations of the satyrine butterflies Lethe eurydice and L. appalachia that fly in dissimilar but adjacent habitats within a few meters of each other. Although these Lethe species are morphologically nearly indistinguishable, the genetic similarity between them was 1=0.865. Furthermore, although there were no diagnostic loci (i.e. one population fixed for an electromorph that does not occur in the other pop¬ ulation), there were significant differences in electromorph frequencies at 5 of the 8 loci examined, and 4 of these were highly significant. Within the genus Euphydryas , sibling species are also genetically more distant from each other than are E. p. phaeton and E. p. ozarkae. The average genetic identity between E. editha and its two sibling species E. chalcedona and E. anicia is reported by Brussard et. al. (1985) to be 1=0.837, and Euphyd¬ ryas chalcedona and E. anicia, considered by those authors to be semi¬ species, have a genetic identity of 1=0.858. (Here we are following the conservative nomenclature of Bauer (1975) rather than that of Miller and Brown (1981), since there are no justifiable reasons to separate North American Euphydryas into three separate genera (see Brussard et. al., 1985) ) . Non-sibling species of butterflies are even more distinct: within the genus Euphydryas average between-species identity is only 1=0.674 (Brussard et. al., 1985); and among European pierids it is 1=0.728 (Geiger, 1980). 28 J. Res. Lepid. Butterfly subspecies are on the average much more similar to each other than are sibling species. Table 2 shows genetic identities between sub¬ species in 3 genera of butterflies. All are high, most above 1=0.950; and some (e.g., napi-bryoniae complex in Pieris) are probably not meaning¬ fully different from unity. These subspecies, therefore, though recogniz¬ able on morphological or ecological grounds, and perhaps geographically distant from conspecific populations, are often genetically as similar as local populations. Brittnacher et. al. (1978) suggested that the availability of many visually discernible characters in Lepidoptera makes it easy to find morphological differences among local populations and to elevate some of these to races or subspecies. This may account for the low level of genetic differentiation detected among butterfly subspecies compared to that detected in Drosophila. There are a number of visible phenetic or morphological differences be¬ tween E. p. phaeton and E. p. ozarkae. The latter is larger and has reduced orange marginal markings on the ventral side of the wings. There are also the pronounced ecological differences in habitat and foodplant choice. Nonetheless, our analysis of allozymes reveals very little genetic difference among the populations we have examined, even though they are more than 1000 km apart. The lack of concordance between the ecological and morphological traits on the one hand and the electrophoretic traits on the other is not surprising. Singer (1982, 1983) has described variation in host plant preference among and within populations of E. editha , and has sug¬ gested how shifts in host plant use may evolve. Under strong selection, this evolution may occur relatively quickly. The comparatively slight allozyme differences, however, may have resulted from much weaker selection or none at all, and may indicate that the two lineages have been separate for only a short time. Such would be the case if, as a growing body of evidence now suggests (Wilson et. al., 1977; Thorpe, 1982), allozyme differences accumulate at a stochastically constant rate and thus may serve as a molecular evolutionary clock. In summary, our results do not provide a definitive answer to the ques- Table 2. Average genetic identities (Nei, 1972) between subspecies of butterflies. Species Pieris napi-bryoniae Speyeria callipe S. coronis S. zerene Euphydryas editha E. anicia F. chalcedona E. phaeton 0.983 0.929 0.982 0.970 0.964 0.922 0.967 I 0.992 Reference Geiger, 1980 Brittnacher et. al., 1978 Brittnacher et. al., 1978 Brittnacher et. al., 1978 Brussard et. al., 1985 Brussard et. al., 1985 Brussard et. al., 1985 This study 29 25(l):25-29, 1986 tion of the appropriate status of E. p. phaeton and E. p. ozarkae. Overall, there appears to have been little genetic differentiation between the two; however, the striking behavioral and ecological differences remain. Additional evidence from the field on the geographic distribution of the two types of populations and laboratory studies of degrees of interfertility would help to resolve this question. Acknowledgments . Phil Koenig provided much useful information on E. p. ozarkae and assisted in collecting the specimens. Robert Lacy collected the New York samples. Deane Bowers and an anonymous reviewer offered many useful sug¬ gestions in the preparation of the manuscript. The electrophoresis was performed at Cornell University in the laboratory of Peter F. Brussard and supported by a grant, DEB 8116332, to him from the National Science Foundation. The adminis¬ tration of Merramec State Park, Missouri, permitted us to collect within the park; we thank them for their cooperation. Literature Cited angevine, M. w. & P. F. BRUSSARD, 1979. Population structure and gene frequency analysis of sibling species of Lethe. J. Lepidopt. Soc. 33(l):29-36. BAUER, D. L., 1975. Tribe Melitaeini. In: Howe, W. H. The Butterflies of North America. Doubleday and Co., Garden City, New York. pp. 139-195. BRITTNACHER, J. G., S. R. SIMS & F. J. AYALA, 1978. Genetic differentiation between species of the genus Speyeria (Lepidoptera: Nymphalidae). Evolution 32: 199-210. BRUSSARD, P. F., P. R. EHRLICH, D. D. MURPHY, B. A. WILCOX & J. E. WRIGHT, 1985. Genetic distances and the taxonomy of checkerspot butterflies (Nymphalidae: Nymphalinae). J. Kans. Ent. Soc. 58:403-412. GEIGER, H. J., 1980. Enzyme electrophoretic studies on the genetic relationships of Pierid butterflies (Lepidoptera: Pieridae): I. European taxa. J. Res. Lepid. 19:181-195. MASTERS, J. H., 1968. Euphydryas phaeton in the Ozarks (Lepidoptera: Nym¬ phalidae). Ent. News 79(4):85-91. MAY, B., J. E. WRIGHT & M. STONEKING, 1979. Joint segregation of biochemical loci in Salmonidae: results from experiments with Saluelinus and review of the literature of other species. J. Fish. Res. Bd. of Canada. 36(9): 1114-1128. NEI, M., 1972. Genetic distance between populations. Amer. Natur. 106:283-292. SHAPIRO, A. M., 1975. Butterflies of New York state. Search (New York State Coll. Agr. publ.) 4:1-60. SINGER, M. C., 1982. Quantification of host preferences by manipulation of oviposi- tion behavior in the butterfly Euphydryas editha. Oecologia 52:224-229. _ , 1983. Determinants of multiple host use by a phytophagous insect popu¬ lation. Evolution 37(2): 389-403. SOKAL, R. & F. J. ROHLF, 1981. Biometry, 2nd ed. W. H. Freeman and Co., San Francisco. 776 pp. THORPE, J. P., 1982. The molecular clock hypothesis: Biochemical evolution, genetic differentiation, and systematics. Ann. Rev. Ecol. Syst. 13:139-168. WILSON, A. C., S. S. CARLSON & T. J. WHITE, 1977. Biochemical evolution. Ann. Rev. Biochem. 46:573-639. The Journal of Research on the Lepidoptera 25(l):30-38, 1986 On the Monophyly of the Macrolepidoptera, Including a Reassessment of their Relationship to Cossoidea and Castnioidea, and a Reassignment of Mimallonidae to Pyraloidea James A. Scott 60 Estes Street, Lakewood, Colorado 80226 There have been persistent reports that the closest relatives of various Macrolepidoptera are the Cossoidea or Castnioidea. Thus Brock (1971) claimed that butterflies evolved from Castnioidea, Bombycoidea (includ¬ ing Sphingoidea) evolved from Cossoidea, and Noctuoidea-Geometroidea evolved from Pyraloidea. Brock’s paper is a worthwhile contribution to certain aspects of morphology of adult Lepidoptera, but he failed to place exact character changes on the branches of his tree, so his tree cannot be considered either phylogenetic in any sense, or phenetic, but rather intui¬ tive (of course, every author claims that his tree represents the one and only true phylogeny, but other workers have the right to demand proof in terms of actual characters). However, a detailed examination of Lepidopteran anatomy of all life stages reveals that a very large number of characters separate the Cossoidea and Castnioidea from the Macrolepidoptera, and that the Macrolepidoptera form a monophyletic group. The traits are listed below and numbered, and the numbers placed on the phylogenetic tree (Fig. 1) where they changed in the manner described in the text. For larval traits, see Fracker, 1915; Petersen, 1965; Forbes, 1923-1960, and Common and Edwards, 1981. For pupae, see Mosher, 1916; Common, 1974. No doubt there are dissenting views, and the author has no great per¬ sonal experience with moth anatomy; others should publish their phylogenies, provided that they are supported by actual character changes and their exact positions on the lineage, so that objective judgments may be made about them. Shared Derived Traits of Pyraloidea + Macrolepidoptera (1) On the larva, the postnatal (“subprimary”) seta L3 was lost on the prothorax, leaving only LI and L2. Nearly all other moths have LI, L2, and L3. (2) On the larva, only one L seta is on abdomen segment 9 (other moths have several). This trait is variable in Pyraloidea, in which some Pyralidae subfamilies have two L setae on A9, and Pterophoridae have many secondary setae, but Thyrididae, Carposinidae, Alucitidae, Mimal- 25(1) :30-38, 1986 31 Hesperioidea Papilionoidea Fig. 1 . Phylogeny of Ditrysian Lepidoptera. The numbers refer to gains, losses, or other alterations of the characters numbered and described in text (character 51 is in Table 1). X, possible origin of Bombycoidea- Sphingoidea, see text. lonidae, and most Pyralidae subfamilies have only one L seta, indicating that one is the primitive state in the Pyraloidea. (3) On the pupal abdomen, only segments 5-6 (joints 4-5, 5-6, 6-7) are movable (in other Ditrysia, generally segments 3-7 move in males and 3-6 in females). (4) On the pupal abdomen, the segments lost their special spines and the pupa no longer protrudes from the larval burrow or cocoon (Tortricoidea, Sesioidea, Zygaenoidea, Castnioidea, and Cossoidea have two rows of backward-directed spines per abdomen segment used to wriggle out of the pupation site before adult emergence). The setose pupa of many Pterophoridae seems to be a later derivation; their long spines must have 32 J. Res. Lepid. another purpose entirely, as they lack a cocoon. (5) Wing vein M is ves¬ tigial in the discal cell (it is present, even branched, in most other moths). (6) Tympana evolved on the abdomen base. Shared Derived Traits of Macrolepidoptera (7) On the larval abdomen, setae LI and L2 became far apart; they are close together in other moths. (8) On the pupa, maxillary palpi were lost. (9) The adult maxillary palpi shrank to minute size (they are 3-4 segmented in Pyraloidea and earlier moths). (10) The jugal fold was lost on the forewing base (Sharplin, 1964). (11) The CuP wing vein became rudimentary, rather than a distinct functional vein in earlier moths. (12) Inside the adult mesothorax, the discrimen (of Ehrlich, 1958) became large (it is small in other moths, though moderate in size in Cos- soidea). (13) In the adult thorax, the third metatergopleural muscle assumed an advanced state (Sharplin, 1964). (14) The postmedian wing lever (median wing process of Sharplin, 1964) became large (it is usually small in other moths). In addition, all Macrolepidoptera have the heart looped to the top of the thorax, which may be another shared derived trait, though some microlepidoptera also have a looped heart (Hessel, 1969). Shared Derived Traits of Noctuoidea + Bombycoidea + Sphingoidea + Hesperioidea + Papilionoidea (15) The tympana moved to the metathorax. The lack of additional shared derived traits allows for the possibility that the Geometroidea is polyphyletic, but I will leave this possibility to other workers. Shared Derived Traits of Bombycoidea + Sphingoidea + Hesperioidea + Papilionoidea (16) Secondary larval setae became abundant on older larvae. (17) The larval crochets diversified into two or three lengths (only one length in most other moths). (18) The two adult ocelli were lost. (19) On the adult mesothorax wall, the upper sector of the paracoxal sulcus (“precoxal suture” of Brock) was lost (Brock’s “precoxal suture” in skippers actually is the secondary sternopleural sulcus). (20) The tympanum was lost. Shared Derived Traits of Sphingoidea + Hesperioidea + Papilionoidea (21) The cocoon was lost. (22) The adult antennae are distally enlarged (antennae vary in more primitive moths, but filamentous antennae occur in nearly all groups) . (23) On the adult mesothorax wall, the parepisternal rift was lost (Brock, 1971). 25(l):30-38, 1986 Shared Derived Traits of Hesperioidea + Papilionoidea 33 (24) Eggs are upright. This is a rare condition, also possessed by Noc- tuoidea, and a few members within other moths (some Geometroidea, Choreutidae, Heliodinidae). Cossoidea and Castnioidea eggs have been stated to be upright, but actually both taxa have flat eggs (I. Com¬ mon pers. comm.; Common and Edwards, 1981). (25) The larva has a ventral neck gland used for defense, as in Noctuoidea. (26) On the pupa, the foreleg femur is no longer visible as it is in nearly all moths. (27) The forewing lacks an areole, and vein branches from R basad of R: in the pupal wing (Zeuner, 1943). This areole occurs in most moths and in moths vein R45 branches distad of Rv (28) On the adult mesothorax wall, the anapleural cleft is fused together and undetectable (Brock, 1971). (29) Inside the adult metathorax the furcal arms are mesally fused (Brock, 1971). (30) The adult heart is chambered where it loops to the top of the thorax (Hessel, 1969) . The heart is looped in some moths, but only some Cossidae have a chambered heart (other Cossidae have only a ventral un¬ chambered heart, indicating that the chamber of some Cossidae is just convergence). (31) On the adult abdomen, the anterodorsal apodemes on sternum 2 became minute (Brock, 1971). They are large in nearly all moths. (32) The adult wings lost the ability to be roofed over the abdomen. I have not attempted to decipher the details of the phylogeny of the Dit- rysians more primitive than Pyraloidea, except to determine that none of them are phylogenetically close to Macrolepidoptera. The most primitive Ditrysians, the Tineoid superfamilies, are distinguished from other Dit- rysia by their (33) dual-rod coupling of abdomen sternum 2 with the thorax (Brock, 1971; Heppner, 1977). In addition, the Tineoid superfamilies (34) generally have only one row of backward-directed abdomen spines per seg¬ ment (used to wriggle out of the cocoon or burrow), whereas Cossoidea, Castnioidea, Zygaenoidea, Sesioidea, and Tortricoidea have two rows per segment (see character 4) . The latter five superfamilies are rather similar. The Sesioidea apparently branched from the Ditrysian trunk after the Cossoidea-Castnioidea-Zygaenoidea, after two wing base traits changed (Sharplin, 1964: (35) the metabasalare lost its connection to the epister- num or prescutum; (36) the insertion of the third metatergopleural mus¬ cle changed to an advanced condition) . Tortricoidea apparently appeared still later after the Ditrysian trunk evolved (37) a true pointed and crocheted cremaster (present in Tortricoidea, Pyraloidea, and Macro¬ lepidoptera), setting the stage for the appearance of Pyraloidea. The persistent suggestions that various Macrolepidoptera evolved independently from various primitive Ditrysia (Brock, 1971, argued that butterflies evolved from Castnioidea, and Bombycoidea from Cossoidea) seem wrong on both cladistic and phenetic grounds, as detailed below. Butterflies show numerous differences fromCastnioidea and Cossoidea 34 J. Res. Lepid. (see in particular Common, 1974), including the previous characters 1, 2, 3, 4, 5, 7, 8 (see Common and Edwards, 1981), 9, 10 (see Common and Edwards, 1981), 11, 12, 13, 16 (secondary setae absent or rare in Cossoidea- Castnioidea), 17, 18, 19, 21, 22 (antenna somewhat clubbed but plumose- tipped in Castnioidea, simple to bipectinate in Cossoidea), 23, 24, 25, 27, 28, 29, 30, 31, 32, 35, 36, 37. In addition, the following traits differ between butterflies and Cossoidea-Castnioidea: (38) the larval crochets are in a circle or mesoseries in butterflies, in two transverse bands in Castnioidea and many Cossoidea; (39-40) the larval head is prognathous and strongly notched middorsally in Cossoidea-Castnioidea but not in butterflies; (41- 43) the olfactory pits on the larval head are unusual in position in Cossoidea-Castnioidea (pit Pb is beside VI, La is far behind LI, Aa is near the P setae, Common and Edwards, 1981), normal in butterflies; (44) on the pupa, mandible remnants are definite bumps in Cossoidea-Castnioidea, but are weakly developed in butterflies (the “pilifers” of Mosher, 1916); (45) on the pupa a clypeolabral sulcus occurs in Cossoidea- Castnioidea but not in butterflies; (46) Cossoidea lack a proboscis, pre¬ sent in Castnioidea and butterflies; (47) chaetosema are absent in Cossoidea-Castnioidea, present in butterflies; (48) the mesepimeron on the adult thorax has a membranous division in most Cossoidea, lacking in Castnioidea and butterflies (Brock, 1971). Obviously, these 41 traits demonstrate a vast gap separating Cossoidea- Castnioidea from butterflies. In fact, Cossoidea-Castnioidea are primitive members of the suborder Ditrysia, only slightly advanced from the Tineoidea. And the peculiar positions of the three olfactory pits (char¬ acters 41-43) on the larval head of Cossoidea-Castnioidea, (49) the lateral position of seta AF2 on the larval head (noted by Common and Edwards, 1981 and Hinton, 1946; my Zygaenidae larvae (first instar Zygaena trifolii) have these traits as well, except the position of pit Aa is normal), the absence of a proboscis, and the membranous epimeron cleft of Cossoidea surely indicate that the Cossoidea-Castnioidea-Zygaenoidea is a derived offshoot of the moth line which could not possibly have produced the butterflies or any other Macrolepidoptera. Evidently the superficial butterfly-like appearance, clubbed antennae, and day-flying habits of Castniidae have swayed the intuitive phylogenists, despite the vast morphological gap. Nevertheless, at least 16 families of moths have day¬ flying species with colorful wings, and the microscopic details of the anten¬ nae of Castniidae and Hesperiidae are very different (Jacqueline Miller, pers. comm.) despite their similar overall shape. Some Cossoidea have a chambered dorsal heart as in most butterflies (character 30), but other Cossids have the primitive ventral non-chambered heart (Hessel, 1969), so this must be convergence. The story regarding the relationship between Sphingidae-Bombycoidea and Cossoidea-Castnioidea is much the same, though they are similar in these traits: the eggs of Bomby coidea are also flat (character 24), larvae 25(1) :30-38, 1986 35 lack the neck gland (25), a cocoon is present (21), chaetosema are absent (47), antennae are bipectinate in Bomby coidea as in some Cossoidea (22), the anapleural cleft is a rift (28), a parepisternal rift occurs in Bom¬ by coidea (23), the metafurcal arms are more similar (29), and the sternal apodemes are longer (31). But there still remain some 34 traits separating Sphingoidea from Cossoidea-Castnioidea, and 32 separating Bom¬ by coidea from them. Evidently certain superficial similarities between Bombycoidea and Cossoidea (bipectinate antennae, loss of proboscis, and the presence of secondary setae in Limacodidae (including Megalopyginae) and Bombycoidea, similar adult appearance of Megalopyginae and Lasiocampidae) led intuitive phylogenists to claim a relationship, but obviously the relationship is not genealogical. The relationship between Cossoidea-Castnioidea and Geometroidea- Noctuoidea shows the same wide gap, of course. In addition, Noctuoidea have: (50) a unique MD2 seta present on T3 and A1 (present in Notodon- tinae and other Noctuidae, Hinton, 1946) ; and Geometroidea-Noctuoidea have tympana (characters 6, 15) . It seems probable that their tympana are descended from that of Pyralidae, because the Geometroid tympanum is on the first abdomen segment as in Pyralidae, and the Noctuoid tym¬ panum, which moved to the metathorax, retains a hood on the first abdomen segment and commonly has a ventral abdominal pouch that may have once possessed a tympanum. The Noctuoid tympanum shows sufficient variation as to allow for the possiblility that it is descended from the abdominal type. The internal phylogeny of Macrolepidoptera seems straightforward ex¬ cept for the placement of Bombycoidea and Sphingoidea (see Table 1). The Geometroidea and Noctuoidea seem the most primitive Mac¬ rolepidoptera because their larvae generally lack secondary setae and retain one-length (uniordinal) crochets, their pupae retain the temporal cleavage line and the visible prothorax femur, their adults retain ocelli, tympana, and the upper sector of the paracoxal sulcus, and with Bom¬ bycoidea their adults retain the parepisternal rift and an areole. The Geometroidea with its flat eggs, abdominal tympana (as in Pyraloidea), and merely pectinate (not bipectinate) antenna is the more primitive of the two. The most advanced group of Macrolepidoptera, butterflies, shares several derived traits with Noctuoidea: upright eggs, and a ventral larval neck gland used for chemical defense. While the latter gland may be con¬ vergent, or lost in other Macrolepidoptera, the upright eggs of butterflies- Noctuoidea are nearly unique (except in Heliodinidae, Choreutidae, and some Geometridae; the Cossidae, including Cossinae, and Castniidae always have flat eggs, I. Common, pers. comm, and Common and Edwards, 1981). If the upright egg is genuinely co-ancestral then the Bombycoidea-Sphingoidea branched off at point X of Figure 1. However, using the characters and weights of Table 1, the tree of Figure 1 is the most 36 J. Res. Lepid. Table 1. Characters of the Macrolepidoptera superfamilies. F, flat; U, upright; +, present; absent; M, mesoseries (medial crescent); 0, oval; B, biordinal (two lengths); U, uniordinal; T, triordinal; S, simple or filamentous; P, pectinate (two projections from each antenna segment); B, bipectinate (four projections); C, clubbed. In addition, traits 28-31 are derived traits of butterflies (Hesperioidea-Papilionoidea), and 50 is a derived trait of Noctuoidea. a o T3 as as V 3 a 0> a 1 a 3 O u i Si £ *3 a 4* o ’3 a 12 a #g *3 *2 a> ft a _o p. ■s •pN Trait W 3 5 Z o P5 rm ft 00 » E as Ph o £ 16 secondary setae - - + + + + + + + + 1 (+ prolegs) (+ rarely) 17 chrochet length B(U) U(B) B B T(B) T(B) V4 18 ocelli ±. - - - - Vi 19 upper sector of paracoxal sulcus ±- - - - - y2 20 tympana ±. ±- - - - - Vi 21 cocoon + + + - - - 1 (-) 22 antenna shape S,P B,P,C,S B,P C C C Vi (P rare, short) 23 parepisternal rift JL ±. _±_ - - - Vi 24 egg F U F F u u 1 25 ventral neck gland on larva - - - ±- J±_ 1 26 foreleg femur on pupa visible ±. ±. - JL - - Vi 27 areole ±. ±. j±_ ±. - - 1 32 wings roofed over abdomen ±. - ±. - - Vi 38 crochet arrangement M 47 chaetosema ±_ M M M 0 M old Vi 0 young + + Vi 51 temporal cleavage line of pupa ±_ ±_ ±_ - + 1 37 25(l):30-38, 1986 parsimonious, requiring the fewest character changes of any of the possible trees. This is partly because the Bombycoidea-Sphingoidea-butterfiles share certain traits (crochets always bi- or triordinal, secondary setae abundant, tympana and ocelli lost, and the upper sector of the paracoxal sulcus lost. Because three of these traits represent losses, there is some doubt about this parsimonious scheme, and first-instar butterflies have primary setae, whereas first-instar Bomby coidea- Sphingoidea apparently do not. Hopefully current and future research will add more characters to the table to resolve this question. At the present time Figure 1 seems most probable, which suggests that the ancestor of Bombycoidea-Sphingoidea- butterflies was a dayflier, resulting in the loss of tympana and ocelli, and the development of colorful wings. Sphingoidea and butterflies do share the loss of a cocoon and a roughly similar antenna. Eye morphology may provide relevant characters within Macrolepidop- tera (Horridge, 1975), and demonstrates similarities between skippers and other Macrolepidoptera. Many large nocturnal moths and skippers have a clear zone in the eye, and skippers are similar to Bombycoidea in having retinula cell extensions across the clear zone to the lens system (but skip¬ pers differ from Bombycoidea and others in lacking any anatomical wave guides) and skippers resemble Agaristidae in lacking pigment in the clear zone in daylight. Skippers and some night-adapted Macrolepidoptera have a well-focused eye, unlike Papilionoidea (one spot on the retina receives light focused from many ommatidia besides its own) . It should be noted that Mimallonidae (=Lacosomidae=Perophoridae), which have secondary setae only on the prolegs (Forbes, 1923, gives a setal map), have been placed in Bombycoidea and Geometroidea, but various traits place them in the Pyraloidea: abdominal setae LI and L2 adjacent; sometimes two (or one) L setae on abdominal segment 9 (Fred Stehr, pers. comm.); only two postnatal prothorax L setae; crochets in a circle; a well- developed CuP vein. Acknowledgments. I thank John B. Heppner and Ian Common for providing some information, though their views do not necessarily correspond with Figure 1. Clas M. Naumann kindly provided first instar Zygaenidae larvae. Literature Cited BROCK, J., 1971. A contribution toward an understanding of the morphology and phylogeny of ditrysian Lepidoptera. J. Nat. Hist. 5:29-102. COMMON, I. F. B., 1974. Lepidoptera. Chapter 36 in Insects of Australia. CSIRO. Melbourne Univ. Press. COMMON, I. & E. EDWARDS, 1981. The life history and early stages of Synemon magnifica (Castniidae). J. Austral. Ent. Soc. 20:295-302. EHRLICH, P. R., 1958. The comparative morphology, phylogeny, and higher classi¬ fication of the butterflies (Lepidoptera: Papilionoidea). Univ. Kans. Sci. Bull. 39:305-370. FORBES, W. T. M., 1923-1960. Lepidoptera of New York and Neighboring States. 38 J. Res. Lepid. Parts I-IV. Memoirs 68, 274, 329, 371 of Cornell Univ. Agric. Exp. Station. Ithaca, N.Y. FRACKER, S., 1915. The classification of lepidopterous larvae. Contr. Ent. Labs Univ. Ill. #43:1-169. HEPPNER, J. B., 1977. The status of the Glyphipterigidae and a reassessment of relationships in Yponomeutoid families and Ditrysian superfamilies. J. Lepid. Soc. 31:124-134. HESSEL, J. H., 1969. The comparative morphology of the dorsal vessel and acces¬ sory structures of the Lepidoptera and its phylogenetic implications. Ann. Ent. Soc. Amer. 62:353-370. HINTON, H. E., 1946. On the homology and nomenclature of the setae of Lepidop¬ terous larvae, with some notes on the phylogeny of the Lepidoptera. Trans. Roy. Ent. Soc. (London) 97:1-37. HORRIDGE, G. A., ed., 1975. The compound eye and vision of insects. Clarendon Press, Oxford, England. MOSHER, E., 1916. Lepidopterous pupae. A classification of the lepidoptera based on characters of the pupa. Bull. Ill. State Lab. Nat. Hist. 12:17-159. PETERSEN, A., 1965. Larvae of insects. Columbus, Ohio. Published by author. SHARPLIN, J., 1964. Wing base structure in Lepidoptera. HI. Taxonomic characters. Can. Ent. 96:943-949. ZEUNER, F., 1943. On the venation and tracheation of the Lepidopterous fore wing. Ann. and Mag. of Nat. Hist. 10:289-304. The Journal of Research on the Lepidoptera 25(l):39-47, 1986 Electrophoretic Confirmation of the Species Status of Pontia protodice and P. occidentalis (Pieridae) Arthur M. Shapiro and Hansjurg Geiger Department of Zoology, University of California, Davis, California 95616 Abstract. Electrophoretic study of sympatric and allopatric pop¬ ulations of the taxa Pontia protodice and P. occidentalis demonstrates unequivocally that they represent closely related but independent gene pools. Each is genetically very homogeneous over its geographic range, strongly suggesting high levels of migration, colonization, and/or gene flow. P. protodice is less like European P. callidice than is Californian occidentalis , suggesting a possible phylogeny which agrees with previous inferences from morphology and biogeography. Introduction The taxa Pontia (or Pieris or Synchloe) protodice Bdv. & LeC. and P. occidentalis Reak. have posed an ongoing problem for Lepidopterists; though Chang (1963) demonstrated morphological differences between them and Shapiro (1976) summarized the by then copious biological and distributional information on hand — all of which tended to support their status as separate species — many workers, including some professionals, have remained unconvinced and profess to be unable to classify many specimens to one species or the other. The present study was undertaken in the hope of further clarifying their status by comparing population sam¬ ples of both from areas of sympatry and allopatry , using electrophoresis to quantify genomic similarities and differences. An ancillary objective was to test the prediction that both species would show very little inter- populational differentiation, due to their apparent pattern of colonization and their epigamic behavior. Materials and Methods The sources of our samples are listed in Table 1. All animals were transported alive and immediately stored at -70°C until electrophoresis. Only 1984 and 1985 catches were used. All animals were determined as protodice or occidentalis by AMS, using wing phenotype, and all wings were saved for post-electrophoresis verification. Only one possibly ambiguous specimen was used in the study. The head and thorax of each 40 J. Res. Lepid. Fig. 1 . Localities for California samples. Abbreviations, numbers and makeup by species as in Table 1. butterfly were homogenized in 4 volumes of Tris-HCl buffer (0.05 M, pH 8.0) . We used horizontal starch gel electrophoresis, following slightly mod¬ ified standard procedures (Ayala et al., 1972; Geiger, 1981). Twenty- three enzymes were scored: acid phosphatase (locus ACPH), adenylate kinase (AK-1, AK-2), aldolase (ALD), arginine kinase (APK), fumarase (FUM), glutamate-oxaloacetate transaminase (GOT-1, GOT-2), glutamate- pyruvate transaminase (GPT), glyceraldehyde-phosphate dehydrogenase (GAPDH), cc-glycerophasphate dehydrogenase (cd-GPDH), hexokinase (HK), indophenol oxidase (IPO), isocitrate dehydrogenase (IDH-1, IDH- 2), malate dehydrogenase (MDH-1, MDH-2), malic enzyme (ME-1, ME- 2), phosphoglucomutase (PGM), 6-phospho-gluconate dehydrogenase (6-PGD), phosphoglucose isomerase (PGI), and pyruvate kinase (PK). Analysis of the progeny of parents with different phenotypes in Pieris brassicae L. (Geiger, 1982) was the basis for interpreting zymograms of polymorphic loci. No deviation from the pattern observed in P. brassicae has been found in any individual investigated in this study. The dis¬ tributions of alleles are also in good accord with Hardy- Weinberg expectations. The most frequent allele (“common allele”) inP. brassicae was used as a standard. This allele is designated with the index 100. Electromorphs with different mobilities are designated in relation to this standard; an allele 25(1) :39-47, 1986 41 Table 1. Localities for samples used in this study, with notes on sympatry. California: Lassen County: 2.5 km S Adin, 1500 m, ll.viii.1985 (n=24)(AD), occidentalis abundant, protodice unrecorded but possible infrequent immigrant. Siskiyou County: Ball Mountain, 2175 m, 10.viii.1985 (n=28)(BM), occidentalis only, very abundant. Sierra County: Sierra Valley, 4 km NE Sierraville, 1500 m, 25-30. vii. 1985 (nprot =55, nocc =46)(SV), both abundant and permanently sympatric. Placer and Nevada Counties: Donner Pass, 2100 m, 15.viii.1985 (nprot =6, nocc =4)(DP), occidentalis common resident, protodice frequent im¬ migrant, overwintering once in 14 yr. Nevada County: Castle Peak, 2750 m, 6.vii.l985 (n=26)(CP), occidentalis only (14 yrs. of observation). Alpine County: Leviathan Peak, 2800 m, 25.vii.1984 (nprot =2, nocc =17) (LP), occidentalis common resident, protodice immigrant. Mono County: nr. Mono Lake, 1800 m, 2. vii. 1985 (n ^ =17, nocc =2)(ML), pro¬ todice common, occidentalis infrequent (but commoner at higher elevations). Kern County: Lake Isabella, 780 m, 16.viii.1985 (n=19) (LI), protodice only. San Bernardino County: Route 38 N Onyx Sum¬ mit, elevations not available, 15.vi.1985 (n=3)(OS), protodice only. Nevada: Churchill County: vie. Fallon, 1250 m, 13.viii.1984 (n=30), protodice abundant, occidentalis very rare (none taken). Florida: Broward County: vie. Davie, metropolitan Miami, 25. iv. 1984 (n=9), protodice only. Mexico: Distrito Federal: Xochimilco-San Gregorio, 27.vi-2.vii.1984 (n=12), protodice only. 105, then, codes for an enzyme that migrates 5 mm faster than the P brassicae variant. The statistic I (Nei, 1972) was used to estimate the genetic similarity between the samples over all loci. The calculated I- values for the pooled samples of the two North American taxa plus P. callidice Hiibner have been used to construct a dendrogram (Fig. 2) by cluster analysis (UPGMA method, Ferguson, 1980). The Lvalues for the pooled samples are based on only 22 loci (without ACPH) to make the data comparable to an earlier study (Geiger and Scholl, 1985). Results At 16 of the 23 loci compared, protodice and occidentalis show only very low polymorphism (fcommon allele^1 0-98), and share the same common allele (ALD, AK-1, AK-2, APK, FUM, GADPH, GOT-2, cc-GPDH, IDH-1, IDH-2, IPO, MDH-2, ME-1, ME-2, 6-PGD, PK) . At four other loci (GOT- 42 J. Res. Lepid. I-value a> a '■g "o o 10, the frequencies of all alleles are remarkably similar and show no statis¬ tically significant interpopulational differences. The situation is different for the three remaining loci ( ACPH, HK, GPT) (Table 3) . There are two alleles at the GPT locus which are found in both taxa but at very different frequencies: the common allele in protodice (GPT 86) is found at very high frequency (f >0.96) in all samples of that taxon but at much lower frequencies (f _<_0.25) in the occidentalis samples. The allele 97 is the com¬ mon GPT allele in occidentalis (f >_0.75) and is very rarely recorded in protodice (f ^_0.04). The genetic differences between the taxa are even more pronounced at the HK locus, where only the allele 93 is found in pro¬ todice (f = 1.00); this allele occurs in occidentalis only at frequencies ^_0.02. At the ACPH locus each taxon is monomorphic for a different allele. It is important to underscore the fact that there were no heterozygous individuals for the ACPH locus at those localities where the taxa are fre¬ quently to permanently sympatric (Donner, Sierra Valley — see Table 1). The one possibly ambiguous individual, a female from Sierra Valley, was electrophoretically “pure protodice” which taxon it most closely resembled in wing phenotype. The frequencies of heterozygotes for HK and GPT did not vary significantly between sympatric and allopatric samples. The very similar allelic frequencies at all loci among population samples result_in very high I- values for the within-taxon comparisons (l0CCldentaZis = 1.00, Iprotodice >0.99) (Table 4). When the two taxa are compared the I- value is (as expected) lower, x(I) = 0.88 ±_ 0.01. Because HJG had already studied European P. callidice, and there has been considerable speculation concerning its relationship to the American taxa (Higgins and Riley, 1970; Shapiro, 1980), we compared it to our results using the 22 loci (omitting ACPH) studied for all three taxa. 25(1) :39-47, 1986 43 Table 2. Allelic frequencies at polymorphic loci. CTJ B •H GOT- 1 MDH- ■1 PGM PGI CTJ c 110 115 120 125 89 100 90 95 103 107 109 87 97 107 115 125 Sierra Valley 55 .18 .81 .02 .97 .03 .07 .90 .03 .02 .06 .84 .08 Lake Isabella 19 .05 .95 .84 .16 .03 .10 .84 .03 .03 .05 .89 .03 Mono Lake 17 .12 .85 .03 .88 .12 .03 .12 .85 .06 .91 .03 Donner Pass 6 .17 .67 .17 .92 .08 .08 .75 .17 .25 .75 cu a Fallon 30 .17 .82 .02 1.0 .02 .08 .80 .10 .03 .92 .05 'd o u Florida 9 .28 .72 .95 .05 1 .0 1.0 o u Ch Mexico 12 .13 .87 .96 .04 .04 .04 .88 .04 .08 .88 .04 all samples 153 .15 .83 .02 .95 .05 .01 .08 .86 .04 .01 .01 .06 .88 .04 .01 Ball Mtn 28 .07 .79 .14 1 .0 .07 .75 .16 .02 .04 .93 .04 CO Ad in 24 .02 .10 .71 .17 1.0 .08 .75 .15 .02 .06 .85 .08 •U Sierra Valley 46 .07 .79 .14 .97 .03 .01 .03 .76 .20 .01 .03 .90 .05 c cu T3 Castle Peak 26 .02 .81 .17 .90 .10 .02 .71 .23 .04 .04 .89 .08 O Leviathan 17 .03 .74 .23 1.0 .09 .74 .18 .12 .82 .06 all samples 147 .01 .06 .77 .17 .97 .03 .01 .05 .75 .18 .01 .01 .05 .88 .06 Table 3. Allelic frequencies at loci with high variability between the taxa. n animals ACPH 88 95 GPT 86 97 HK 93 96 Sierra Valley 55 1.0 .99 .01 1.0 Lake Isabella 19 1.0 1 .0 1.0 Mono Lake 17 1.0 .97 .03 1.0 Donner Pass 6 1.0 1 .0 1.0 at Fallon 30 1.0 1.0 1.0 •H XI o 4-1 Florida 9 1.0 1.0 1.0 o o. Mexico 12 1.0 .96 .04 1 .0 all samples 153 1.0 .99 .01 1.0 Ball Mtn 28 1.0 .23 .77 .02 .98 Adin 24 1.0 .23 .77 .02 .98 •H Sierra Valley 46 1.0 .13 .87 .02 .98 c XI Castle Peak 26 1 .0 .25 .75 1.0 u o o Leviathan 17 1.0 .06 .94 1.0 all samples 147 1.0 .18 .82 .01 .99 44 J. Res. Lepid. Occidentalis and callidice cluster at a slightly higher level than protodice. The relationships with other species of the genus Pontia remain unchanged. Discussion We had predicted a high level of genetic similarity over large distances in these species because both are highly vagile, colonizing or “weedy” species and because P. occidentalis is a facultative “hilltopper,” a mating strategy which would tend to promote gene flow and prevent local ecotypic differentiation. (For discussion of the population dynamics of P. pro¬ todice , see Shapiro, 1979; for dispersal ability of P. protodice, Shapiro, 1982 and of P. occidentalis, Shapiro, 1977; an explicit prediction was made in Shapiro, 1984, p. 181). We were nonetheless surprised at the extreme homogeneity of protodice over a continent-wide range (Florida, Mexico, Nevada, California). We know that these populations are not so homogeneous for such adaptive traits as the photoperiodic thresholds for induction of pupal diapause, the programming and control of diapause, hostplant adaptation and disease resistance (AMS, unpublished data). Mexican protodice also lay smaller eggs than other populations, even under standardized rearing conditions and on a standard diet (Shapiro, in press). All protodice populations tested to date have been fully reproduc- tively compatible with one another, even in such wide crosses as New York Table 4. I-values for the comparison of all populations samples (n >6) based on the data of 23 enzyme loci. protodice occidentalis Lake Isabella Mono Lake Donner Pass Fallon Florida Mexico Ball Mtn Ad in Sierra Valley Castle Peak Leviathan Sierra Valley 1.00 1.00 1.00 1.00 1.00 1.00 .89 .89 .88 .89 .87 Lake Isabella 1.00 1.00 1.00 1.00 1.00 .89 .88 .88 .89 .87 Mono Lake 1.00 1.00 1.00 1.00 .89 .89 .88 .89 .87 Donner Pass 1.00 .99 1.00 .88 .89 .88 .88 .87 Fallon 1.00 1.00 .89 .89 .88 .89 .87 Florida 1.00 .89 .89 .88 .88 .87 Mexico .89 .89 .88 .89 .87 Ball Mtn 1.00 1.00 1.00 1.00 Ad in 1.00 1.00 1.00 Sierra Valley 1.00 1.00 Castle Peak 1.00 25(1) :39-47, 1986 45 X California or Texas, or Mexico X California; but the control of diapause is routinely disrupted in such wide crosses, usually resulting in failure to diapause or very rapid spontaneous termination, and much less often in extended, lethal diapause. We have less extensive experience crossing occidentalis populations but have found complete compatibility among California and Colorado ones and between California and the Alaskan subspecies nelsoni Edwards (Shapiro, 1975) which is highly incompatible with European callidice (Shapiro, 1980). Diapause is largely unstudied in these cases. Vawter and Brussard (1984) found similar uniformity in the weedy, introduced species Pieris rapae L. in eastern North America, but more genetic diversity in the west. Populations of P. rapae in the west are dis¬ continuous, separated (except in the Central Valley of California) by broad expanses of inhospitable terrain. The ability of P. rapae , as an obligatorily multivoltine species, to accommodate to western climates by altitudinal migration seems very limited in comparison to P. protodice ; indeed, rapae is largely confined to local “mesic” pockets created by irrigation within arid or semiarid regions, while protodice is able to colonize throughout. This is most dramatically illustrated in central Mex¬ ico: protodice, a native species, is quite generally distributed, but the introduced rapae is ecologically “trapped” in the floating gardens of Xochimilco, near Mexico City, where continuous breeding is possible. It is hardly surprising that the homogenizing effects of gene flow are more evi¬ dent in P. protodice than in P. rapae. Vawter and Brussard argue that gene flow should be countered in colonizing or fugitive species by genetic drift and founder effect, which would tend to cause stochastic differences among populations. We are examining the genetic structure of truly ephemeral populations of P. pro¬ todice (presumably resulting from colonizations by single females) in the hope of addressing this question. The clear genetic differences between protodice and occidentalis at three loci are in sharp contrast to the low variation within the taxa. The most important samples are those from Sierra Valley, where both taxa are very abundant and apparently in stable coexistence (over 10 years, AMS observations) and where occasional (<3%) ambiguous phenotypes are encountered. There is absolutely no evidence for gene flow in the sympat- ric Sierra Valley samples; the lack of ACPH heterozygotes shows that there were no hybrids in our collections. There may indeed be occasional, very rare spontaneous hybridization (AMS has collected one mixed pair in copula in Donner Pass), but the electrophoretic data provide clear confirmation that protodice and occidentalis represent separate gene pools, corresponding to biological species. There is no evidence of introgression (that is, the two taxa are not more similar genetically in sym- patry than in allopatry). It is exceedingly difficult to hybridize these two taxa spontaneously, and 46 J. Res. Lepid. pairings can normally only be secured with a pre-excited male and a sub¬ stituted teneral allospecific female. Hand-pairings are easily achieved, but to date the level of developmental incompatibility has been high, resulting in dwarfing, high mortality, malformation, deficiency of the heterogametic sex (female), and hybrid sterility. Further information on experimental hybridization will be published elsewhere; it suffices to note that it is fully in accord with the electrophoretic results. Genetic differences within the group of three taxa {protodice, occiden¬ tal is, callidice) are relatively low but within the range previously reported for closely-related species in Pieridae (Geiger, 1981; Geiger and Scholl, 1985) . This observation can be interpreted as evidence for the recency of speciation in that group. Shapiro (1980) interpreted the group as derived by fragmentation of the range of a widespread circumglacial steppe- tundra entity more or less resembling the climatic adaptation of some con¬ temporary occidentalis populations. The phenotypic characteristics of protodice are clearly derivative reductions from the full occidentalis pat¬ tern, which in the western United States is polyphenic and shows some reduction from the nelsoni- callidice pattern. Larval and pupal characters vary concordantly (Shapiro, unpublished data). The dendrogram thus further supports the proposed phylogeny, which would derive occidentalis from circumpolar proto -callidice and protodice in turn from occidentalis, without specifying time scales. Certain central Asian taxa assigned as sub¬ species to callidice (orientalis Alph., kalora Moore, etc.) are extremely close pheno typically to western North American occidentalis. This may represent parallel evolution in similar climates — but then again, it may not. Nominate callidice from the Alps and Pyrenees seems to represent the extreme end of a long cline, physiologically as well as geographically. We note in closing that the extreme genetic homogeneity shown by pro¬ todice and occidentalis over a large range suggests the reality of a “general purpose genotype” associated with weediness and physiological adapt¬ ability (Baker, 1965) and the utility of these common animals as vehicles to get a closer look at its structure. Acknowledgments. We thank Francisco J. Ayala for permitting the use of his facilities, and Adam Porter and Marc Minno for supplying specimens. HJG’s work at Davis was supported by National Science Foundation grant BSR-8306922 (Sys¬ tematic Biology Program) to AMS. Field assistance was provided at various times by Adam Porter, Doug Eby, and Cecile La Forge. This paper forms part of Cali¬ fornia Agricultural Experiment Station project CA-D*-AZO-3994-H, “Climatic Range Limitation of Phytophagous Lepidopterans,” AMS, Principal Investigator. The Xochimilco sample was collected with aid from a UC-MEXUS grant to AMS and Jorge Llorente B., Universidad Nacional Autonoma de Mexico. Literature Cited AYALA, F. J., J. R. POWELL, M. L. TRACEY, C. A. MOURAO & S. PEREZ-SALAS, 1972. Enzyme variability in the Drosophila willistonii group. IV. Genic variation 47 25(1) :39-47, 1986 in natural populations of Drosophila willistonii. Genetics 70:113-139. BAKER, H. G., 1965. Characteristics and mode of origin of weeds, in H. G. Baker and G. L. Stebbins, eds., The Genetics of Colonizing Species. Academic Press, N.Y., pp. 147-172. CHANG, V. C. S., 1963. Quantitative analysis of certain wing and genitalia char¬ acters of Pieris in western North America. J. Res. Lepid. 2:97-125. FERGUSON, A., 1980. Biochemical Systematics and Evolution. Blackie, Glasgow and London. GEIGER, H. J., 1981. Enzyme electrophoretic studies on the genetic relationships of Pierid butterflies. I. European taxa. J. Res. Lepid. 19:181-195. _ , 1982. Biochemisch-genetische Untersuchungen zur Systematik und Evolution von Weisslingen des europaischen Faunengebietes. Ph.D. thesis, University of Bern. GEIGER, H. J. & A. SCHOLL, 1985. Systematics and evolution of holarctic Pierinae: an enzyme electrophoretic approach. Experientia 41:24-29. HIGGINS, L. G. & N. D. RILEY, 1970. A Field Guide to the Butterflies of Britain and Europe. Houghton Mifflin, Boston, p. 50. NEI, M., 1972. Genetic distance between populations. Am. Nat. 106:283-292. SHAPIRO, A. M., 1975. The genetics of subspecific phenotype differences in Pieris occidentalis Reakirt and of variation in P. o. nelsoni W. H. Edwards (Pieridae). J. Res. Lepid. 14:61-83. _ , 1976. The biological status of Nearctic taxa in the Pieris protodice- occidentalis group (Pieridae). J. Lep. Soc. 30:289-300. _ , 1977. Apparent long-distance dispersal by Pieris occidentalis (Pieridae). J. Lep. Soc. 31:202-203. _ , 1979. Weather and the lability of breeding populations of the check¬ ered white, Pieris protodice Bdv. & LeC. (Pieridae). J. Res. Lepid. 17:1-23. _ , 1980. Genetic incompatibility between Pieris callidice and P. occiden¬ talis nelsoni : differentiation within a periglacial relict complex. Can. Ent. 112:463-468. _ , 1982. A new elevational record for Pirns protodice in California (Lepi- doptera: Pieridae). Pan-Pac. Ent. 58:162. _ , 1984. Polyphenism, phyletic evolution, and the structure of the Pierid genome. J. Res. Lepid. 23:177-195. _ , in press, r and K selection at various taxonomic levels in the pierine butterflies of North and South America, in F. Taylor and R. Karban, eds., Evolution of Insect Life Histories. Springer-Verlag, Berlin, pp. VAWTER, A. T. & P. F. BRUSSARD, 1984. Allozyme variation in a colonizing species: the cabbage butterfly Pieris rapae (Pieridae). J. Res. Lepid. 22:204-216. The Journal of Research on the Lepidoptera 25(1):48-51, 1986 Susceptibility of Eggs and First-Instar Larvae of Callosamia promethea and Antheraea polyphemus to Malathion1 Thomas A. Miller, William J. Cooper2 and Jerry W. Highfill3 US Army Medical Bioengineering R&D Laboratory, Fort Detrick, Maryland 21701 Abstract. Susceptibility levels to malathion water emulsions are established for Callosamia promethea (Drury) and Antheraea poly¬ phemus (Cramer): C. promethea eggs, LC50 = 0.1 mg/ml, probit = 3.3 + 0.9 (log cone); C. promethea lst-instar larvae, LC50 = 0.01 mg/ml, probit = 11.3 + 2.9 (log cone); A. polyphemus eggs, LC50 between 15.6 and 31.2 mg/ml; A. polyphemus lst-instar larvae, LC50 = 0.06 mg/ml; probit = 9.9 + 4.1 (log cone); C. promethea embryos, LC50 = 248 mg/ml, probit = -2.3 + 3.0 (log cone). Introduction Giant silkworm moths, such as Callosamia promethea (Drury) and Antheraea polyphemus (Cramer), used for research purposes, are reared on foliage obtained in agricultural or domestic situations where organo- phosphate insecticides may be applied or stored. Information is not avail¬ able on the susceptibility of the early stages of these giant silkworm moths to foliar applications or organophosophate insecticides. To estimate the susceptibility of these insects to natural, potentially contaminated foliage, we determined the laboratory susceptibility of eggs and lst-instar larvae of C. promethea and A. polyphemus to malathion, 0,0-dimethyl phosphorodithioate of diethyl mercaptosuccinate, a representative, commonly-used organosphophate material. These estimates of suscepti¬ bility are based on laboratory testing procedures, and do not account for field conditions (e.g., pH, temperature, humidity, residue age) that might alter the effective toxicity of the malathion. Materials and Methods Test insects were obtained from colonies of C. promethea and A. polyphemus reared in sleeve cages on wild cherry ( Prunus serotina ) or 1Lepidoptera: Saturniidae. This research was not supported by public funds. The opinions contained herein are those of the authors and should not be construed as official or reflecting the views of the Department of the Army. 2Present address: Drinking Water Research Center, Florida International University, Miami, FL 33199. 3Present address: USEPA Health Effects Research Laboratory, Research Triangle Park, NC 27711. 25(1):48-51, 1986 49 various maples (. Acer spp.) for 10-12 generations. Ortho Malathion 50 Insect SprayR (Chevron Chemical Company Ortho Division, San Fran¬ cisco, CA 94119, EPA Reg. No. 239-739-AC) an emulsifiable concentrate containing 50 percent actual malathion by weight (521 mg/ml), was used for all testing. The appropriate quantity of emulsifiable concentrate was added to water with a volumetric pipet to make a stock emulsion. The stock emulsion was diluted with water, with constant stirring of stock and dilutions, to produce the test concentrations. For larvicide tests, individual hostplant leaves were dipped in test emulsions for 30 seconds and allowed to air dry. Control leaves were dried separately after being dipped only in water. Unfed, lst-instar larvae (<24 hours old) were placed in random sequence on the treated and control leaves and held in plastic cups and covered with facial tissues for 48 hours before recording mor¬ tality. For ovicide tests, egg masses (<48 hours old) were dipped in ran¬ dom sequence in test emulsions for 30 seconds, then allowed to air dry. Control egg masses were dipped in water only, then allowed to air dry separately. Mortality of treated eggs was determined as those that failed to hatch after all control eggs hatched. Embryos of C. promethea were tested at high concentrations of malathion (up to 500 mg/ml) to determine the existence of a concentration- mortality relationship. The percentage of embryo mortality rather than the hatch failure was the measured criterion. Mortality data on embryos, eggs, and larvae were subjected to probit analysis using a computer program package described by Barr, et al. (1976). Results and Discussion The label recommended concentration for spray application of mala¬ thion water emulsion to fruit and ornamental foliage is 12.0 mg Al/ml. Table 1 shows that both eggs and lst-instar larvae of C. promethea and lst-instar larvae of A. polyphemus were highly susceptible to recommended doses. The concentration-mortality relationships for C. promethea eggs and lst-instar larvae are shown in Figures 1 and 2. Eggs of A. polyphemus were not susceptible at label concentrations; those exposed to concen¬ trations of up to 15.6 mg/ml showed no ovicidal effects, while those exposed to concentrations of 31.2 mg/ml and above showed no hatch. No concentration- mortality relationship was established for A. polyphemus eggs. The concentration-mortality relationship applicable to A. poly¬ phemus lst-instar larvae is shown in Figure 3. Embryos of C. promethea were not susceptible to label concentration of malathion water emulsion (Table 1). However, toxicity to embryos was observed at the higher con¬ centrations used in initial range finding. In some of these tests, fully- formed larvae were present in the eggs, but did not hatch. In C. promethea the developing dark larva imparts a gray hue to the otherwise white egg. The presence of larvae in the “gray” eggs was confirmed by dissection. 50 J. Res. Lepid. Table 1. Malathion susceptibility of eggs and larvae of Callosamia pro- methea and Antheraea polyphemus.1 Stage Tested Number Tested LC50 LC90 Relative Susceptibility2 at LC50 Callosamia promethea Embryos 450 248 659 0.05 Eggs 650 0.1 0.36 120 lst-Instar Larvae 380 0.01 0.02 1200 Antheraea polyphemus Eggs3 120 >15.6 >15.6 >0.77 lst-Instar Larvae 280 0.06 0.13 200 1all values in mg/ml 2label concentration/lethal concentration (LC50) 3no ovicidal effects observed at 15.6 mg/ml This condition was not observed in ovicide tests with A. polyphemus because the light yellow larvae are not visible through the tan egg shells and no dissections were performed. Tests conducted with high concen¬ trations of malathion (Figure 4) demonstrated that a concentration- mortality relationship existed. Mortality preceding eclosion of eggs has been observed in many lepidop- terans in connection with exposure to ovicides, but the phenomenon in general is poorly understood (Smith & Salkeld, 1966). Potter, et al. (1957) reported that high concentrations of TEPP (tetraethyl pyrophosphate) caused mortality in eggs of Pieris hrassicae Linnaeus; older eggs being considerably more susceptible than younger ones. They concluded that toxicity due to organophosphate exposure appears to involve cholines¬ terase inhibition at some stage of embryonic development. The C. pro- methea eggs we tested were less than 48 hours old. For those exposed to malathion water emulsion at 0.01 mg/ml all of the embryos developed, but only 50 percent hatched; for those exposed to 248 mg/ml only 50 percent of the embryos developed and none hatched. These studies establish baseline susceptibility of immature stages of C. promethea and A. polyphemus to malathion water emulsions, and demonstrate that the immature stages are susceptible at recommended dosages. The particular hazard presented by malathion, or other more toxic or more persistent organosphosphate materials, depends on other variables, such as pH, temperature, humidity, and residue age, not included in these studies. PERCENT MORTALITY PERCENT MORTALITY 25(0:48-51, 1986 51 MALATHION (mg/ml) MALAT H IO N(mg/ ml) Figs. 1-4. Susceptibility of Giant Silkworm Moth Eggs and Larvae to Malathion Water Emulsion: 1. Callosamia promethea Eggs; 2. Callosamia pro- methea Ist-lnstar Larvae; 3. Antheraea polyphemus Ist-lnstar Lar¬ vae; 4. Callosamia promethea embryos. Literature Cited BARR, A. J., J. H. GOODNIGHT, J. P. SAUL & J. T. HELWIG, 1976. A User's Guide to SAS 76. SAS Institute, Inc., Raleigh, NC pp. 206-211. POTTER, C., K. A. LORD, J. KENTON, E. H. SALKELD & D. V. HOLBROOK, 1957. Embryonic Development and Esterase Activity of Eggs of Pieris brassicae in Relation to TEPP Poisoning. Ann. Appl. Biol. 45:361-375. SMITH, E. H. & E. H. SALKELD, 1966. The Use and Action of Ovicides. Ann. Rev. Entomol. 11:331-368. PROBITS PROBITS The Journal of Research on the Lepidoptera 25(l):52-62, 1986 Pupal Mortality in the Bay Checkerspot Butterfly (Lepidoptera: Nymphalidae) Raymond R. White 788 Mayview Avenue, Palo Alto, California 94303 Abstract. Mortality for pupae of Euphydryas editha bayensis (Lepi¬ doptera: Nymphalidae) placed in the field ranged from 53 to 89%. Preda¬ tion and cold weather during the period of pupation were the major mortality factors. Mortality during this stage is high enough to affect total numbers of adults and other life stages and variable enough to affect the population dynamics of these butterflies. Studies of these and other holometabolous insect species should include estimates of pupal mortality. Introduction Few complete life tables have been published for natural populations of butterflies (see Dempster, 1983). This is partly because at least one life stage of these holometabolous insects is difficult or impossible to observe in the field. For example, Euphydryas editha bayensis Sternitzky (1937) (the Bay Checkerspot butterfly) is among the most thoroughly studied insects, but only its adult stage is easily observable. Eggs and prediapause larvae have only recently been found in numbers, and diapausing larvae remain essentially a “black box” to us. Many post-diapause larval sam¬ ples have been collected and some data on parasitoid rates have been published (Ehrlich, 1965; White, 1973 and Stamp, 1984). Pupae are almost never seen. Prior to this study the only information on pupal mortality in Euphydryas editha was Singer’s observation that several out of 20 pupae placed out at Jasper Ridge were eaten and the wooden tongue depressors used to mark them had been chewed on by rodents (Singer, 1971). Life table data for butterfly populations that have been published show pupal mortalities ranging from 0 to 100%, but averaging around 60% (Table 1). Most of the pupal mortality identified was due to predation. With this background I did an experiment designed to quantify pupal mortality in the Bay Checkerspot butterfly. Materials and Methods Large post-diapause larvae were collected in late February and early March from field sites at Edgewood Park (EW) in 1982 and 1983 and Morgan Hill (MH) in 1984. Both sites are serpentine grasslands (Krucke- 53 25(l):52-62, 1986 Table 1. Available data on lepidopteran pupal mortality. Pupal Major Species Mortality Factor n Source Pieris rapae .31 parasitoids large Harcourt 1966 .38 virus 42 Dempster 1967 .08 virus 27 Dempster 1967 .05 virus 65 Dempster 1967 Papilio machaon .59 predation 150 Wiklund 1975 .90 predation 158 Wiklund 1975 Papilio xuthus .83 parasitoids 12 Watanabe 1976 .12 predation 25 Watanabe 1976 Artopoetes pryeri .45 predation 42 Watanabe & Omata 1978 Papilio glaucus 1.00 predation 112 West & Hazel 1982 .80 predation 109 .88 predation 128 .55 predation 127 Battus philenor .91 predation 140 West & Hazel 1982 .94 predation 139 .77 predation 80 .96 predation 80 Battus philenor .14 predation 64 Sims & Shapiro 1983 .67 predation 109 Agraulis vanillae .08 predation 364 I.L. Brown pers. comm. berg, 1984; Sommers, 1984; Crittenden and Grundmann, 1984) where adverse soil conditions favor the native plants on which the butterflies depend. Edgewood Park is in San Mateo County at 37° 27' 50" latitude, 122° 17' 10" longitude, and 660' (200m) elevation. Morgan Hill is in Santa Clara County at 37° 11' 28" latitude, 121° 40' longitude, and 1000' (300m) elevation. For comparison, Jasper Ridge is in San Mateo County at 37° 25' latitude, 122° 19' longitude, and 550' (170m) elevation. Rainy weather in 1982 and 1983 and a large population in 1984 (at MH) allowed longer collection periods than normal. Larvae were kept in groups of about four in plastic petri dishes (37mm in height, 150mm diameter) and fed daily until they pupated, on average about a week. They were fed primarily the Eura¬ sian weed Plantago lanceolata L., which they seem to prefer in the laboratory but which is rarely used in the field (Tilden, 1958). Supplemen¬ tary feeding with the normal foodplants ( Plantago erecta Morris and Orthocarpus spp.) was done when possible. As soon as pupae hardened enough to permit handling they were placed in the field. Transects were laid out in areas from which larvae had been collected (areas of relatively high larval densities). Pupae were placed directly on the soil or foliage every 25cm (my span plus 2cm) along the 54 J. Res. Lepid. transects (Fig. 1). Edgewood Park is open to the public and I wanted my transects to be inconspicuous to people as well as to potential predators, so I marked each pupa with a tiny (7 x 4mm) paper flag mounted on an insect pin. These I could easily relocate. A typed number on the flag identified each pupa. An acrylic spray (Krylon Crystal Clear 1301) applied to the page before cutting the flags out made the numbers proof against rain. Pupae were checked every three to seven days, depending on weather conditions, and their fates were recorded as follows: (1) Parasitized — two kinds of parasitoids emerged from pupae. One was a tachinid fly (Siphost urmia melitaeae Coquillet, determined by Paul Arnaud, Calif. Academy of Sciences) the larva of which bored out the side of the pupa and then itself pupated, sometimes near enough to be found. The exit hole was larger than that made by the piercing predators. The other parasitoid was a large ichneumonid which caused the pupae to change to an orangish hue. In emerging from an infected pupa, this wasp cut a circular cap off the top of the pupa. This cut (Fig. 2) was entirely dif¬ ferent from the typical lines of fracture resulting from butterfly eclosion (Fig. 3). Butterflies that successfully eclosed left behind a case fractured along typical lines and very much thinner than that left by even the most thorough predator. (2) Stepped on — pupae crushed. The evidence often included signs of trampling, showing the outline of a footprint, usually of cattle. (3) Died intact — pupae remaining, apparently unmolested, throughout the study. They eventually either shrank and were found to be empty, or they turned black and contained a foul black liquid (probably due to a virus) . (4) Vanished — pupae not relocated, although their marking flags were. None of the traces mentioned below were found. (5) Predated — pupae clearly damaged by one predator or another. One predator left behind lA to V2 of the pupal case, the inside of which was well cleaned out. Another made rough gashes (Fig. 4) and ate most of the con¬ tents, leaving the inside of the case coated with gore. Another predator or suite of predators pierced the pupal case and sucked out some or all of the contents. The damage in the two latter cases was consistent with “tasting but not eating”. Related species are known to be unpalatable as adults and to a lesser extent as pupae (Bowers, 1980, 1981). Degree Days (F.) were calculated according to Rahn (1971): [(daily max <86) + (daily min >50)]/2 -50. Results Total pupal mortality ranged from 53 to 89% (Fig. 5). The major mor¬ tality factors, in order of increasing importance, were the following: Parasitism was a minor factor, taking 1-10% of the pupae. The tachinid (Siphosturmia melitaeae ) is endemic to virtually all E. editha bayensis populations, but its average infection rate is only 7.8% (45 samples from 25(1) :52-62, 1986 55 Fig. 2. Remains of E. editha pupa placed in the field at Edgewood Park in 1983. Note the precise circular break made by a parasitoid as it emerged. 56 J. Res. Lepid. Fig. 3. Remains of E. editha pupa from which an adult butterfly successfully emerged. Note the thinness of the cast shell and fracture lines typical of normal emergence. 1963-1984, 407 tachinids/5212 larvae) and was only 1-2% in these three samples. Presumably the tachinid infects prediapause larvae, but death of the host does not occur until the pupal stage. Infected pupae can often be identified by their low weights. Healthy female pupae average about 380mg and males about 280mg. Tachinid parasitized pupae weigh under 200mg. A large ichneumonid was found to oviposit in pupae in the field, a phenomenon previously undetected. The first observation was actually of a female (probably parthenogenetic) wasp palping a pupa in the field. This predatory species is probably generally unimportant, having taken 10/239 pupae in 1982, 3/160 in 1983, and 0/260 at MH in 1984 (nor did it turn up in a larger sample at MH in 1985) . Since it is necessary to collect or observe pupae in order to detect it, it is not surprising that this predator is known to date only from EW. Crushing generally was found to be a minor factor, but the large number of cattle grazing at MH raised it to 10% in the 1984 study. There are no cat¬ tle at EW and horses are supposed to be restricted to trails. Cattle were evicted from Jasper Ridge in 1960 (P. R. Ehrlich pers. comm.). The proportion of pupae that died intact varied from 9 to 34% and apparently changed with weather patterns. The higher mortality that occurred in 1982 was undoubtedly a result of the very unusual cold and rainy weather. The number of Degree Days measured at Jasper Ridge from January 1 to March 31 in 1982 was 263, 1983 it was 353, and in 1984 it was 570. 1 expect that this pattern of high mortality occurs whenever late win- 57 25(l):52-62, 1986 ter weather is cold. Pupae that vanished without a trace before any others in their age class had eclosed were “taken” by something, presumably a predator. Pupae disappearing while others in their age class were eclosing might have suc¬ cessfully eclosed and their cast cases might have blown away or been otherwise removed. This possibility could not be distinguished from removal by a predator. Here I estimated the proportion of the missing pupae to have eclosed by taking the proportion of same age class of pupae which did leave evidence of having eclosed. The remaining proportion I considered to have been eaten. The effect of this estimate is probably to underestimate predation (the accuracy of this estimate is important only in the 1983 sample). Weather-delayed pupae lasted much longer than nor¬ mal in 1983; 42% of them disappeared. In this unusually late year (Fig. 6) an opportunistic predator (perhaps a bird or rodent) took larger propor- Fig. 4. Pupa of E. editha placed in the field at Edgewood Park, 1983, showing evidence of predation. The damage is consistent with “tasting but not eating” as might occur when a naive predator attacks an unpalatable subject. 58 J. Res. Lepid. tions of pupae later in the season. In the other two years this form of mor¬ tality was very low (Fig. 5). This temperature dependent pattern parallels that observed by Pollard (1979) for Ladoga Camilla (Nymphalidae). Predators that left physical remains took 23 to 32% of the pupae, making such predation the least variable factor over the three year study. One habitat difference at MH allowed a refinement of the experimental technique used. As at any serpentine grassland site there were small areas of a fraction to several square meters in which the foliage was extremely sparse, especially due to lack of the common bunch grasses. These bare areas at MH alternated with areas of denser foliage so that my transects regularly passed in and out of them. I recorded whether pupae were placed in areas of denser foliage, bare areas, or in-between sorts of areas. Analysis of the data for MH in 1984 showed that pupal mortality varied significant¬ ly with microhabitat (G = 21.41, df = 8, P <.01; Table 2). Being crushed was more likely in barer spots (G = 8.07, df = 2, P < .025) . Dying intact was less frequent in barer spots (G = 7.79, df = 2, P <.025). Neither the “eaten” group nor the “vanished” group varied significantly with micro¬ habitat, but one might add these together as presumed predation. In that case, predation was less frequent in spots with more foliage (G = 5.992, df = 2, P = .05). Successful eclosion was not significantly better, but was nearly so, in spots with more foliage (G = 4.73, df = 2, P <.10). Table 2. Fates of pupae placed in field at MH in 1984, according to ground cover of spot where pupae were put. Bare Mixed Dense Foliage n Eclosed successfully .430 .412 .565 122 Died in place .035 .078 .141 21 Stepped on .158 .078 .043 26 Eaten .237 .314 .174 59 Vanished .140 .118 .076 29 Totals 114 51 92 257 Discussion The weather of any given study is unusual and this study merely repre¬ sents an extreme of that situation (Kerr, 1985). Both 1982 and 1983 were very cool, wet, and therefore late years. They differed significantly in that there were some normally sunny days early in 1982 so that development to pupation was probably normal. Then the cold set in and pupae became subject to attack by fungi and viruses. In 1983 there was an extensive period of cold, but when that ended temperatures were warm enough to allow normal pupation. On the other hand, 1984, was an extremely dry year. The rains ended very early and normal temperatures followed. Flight began and ended early (Fig. 6). 25(l):52-62, 1986 59 Fig. 5. Successful emergence and mortality rates by cause in three samples of Euphydryas editha pupae which were place in the field. Fig. 6. Flight seasons of E. editha at Edgewood Park, from first to last adult seen. Shaded areas represent peak flight. 60 J. Res. Lepid. Pupation in the field took longer than expected. Laboratory eclosion is common in 10-11 days and even possible in 7 days, I had expected (in spite of Tilden’s (1958) estimate of three weeks) normal field times to be about 14 days. In 1983 and 1984 field pupation periods averaged about 18 days (Table 3) . The average was 27 days in the inclement weather of 1982 and many pupae (34%) died undisturbed. We have wondered for some ten years why larvae of Euphydryas editha bayensis do not break diapause earlier in the winter in order to get through the requisite life stages and enter diapause before the inevitable spring senescence of their annual foodplants (Ehrlich et al., 1975). It may be that earlier pupation would too often lead to longer, often fatal, pupation periods during cooler, rainier weather of January in the Mediterranean climate of the Bay Area. The proportion of pupae crushed by cows at MH was great enough to sug¬ gest that this might be an important mortality factor for other life stages of the butterfly. The animal is probably not significantly exposed to this fac¬ tor when diapausing or when in the adult stage. The observed fifteen day exposure of pupae resulted in a 10% mortality rate (90% survival rate), which is equivalent to .993 survival per day. Euphydryas editha probably spends about 65 days total exposed to crushing as eggs, prediapause and postdiapause larvae, and as pupae. Therefore I estimate that on the order of 35% (l-(.993)65) of the total population could be lost to crushing each generation in colonies where heavy grazing occurs. Iwasa et al. (1983) have pointed out that pre-emergence patterns of mor¬ tality are critical in analyses of phenomena such as protandry. But the implications of the data published here (and those collected in Table 1) are of more general importance. Successful eclosion varied from 11 to 47% of the pupae placed in the field. Given that estimated adult numbers at Jas¬ per Ridge (H and C) changed from one year to the next year by factors of 0.20 (80% decrease) to 5.00 (400% increase) in Ehrlich’s twenty-five year study, this four-fold range in pupal mortality makes it clear that mortality during this stage must be estimated if we are to understand the dynamics of these populations. Leaving this as a “black box” may make any other efforts ineffective or inaccurate in explaining observed fluctuations in numbers. Table 3. Length of pupation period in the field for Euphydryas editha bayensis. Site and Year n X s 95% Cl Range EW 1982 47 27.0 7.02 24.9 -29.1 14-43 days EW 1983 15 17.5 7.01 13.6 -21.4 10-26 days MH 1984 males 52 19.9 4.38 18.7 -21.1 12-27 days females 69 16.6 4.04 15.7 -17.6 12.23 days 61 25(l):52-62, 1986 Summary 1. Pupal mortality in the field was high enough in all three years to be a major factor in determining the sizes of checkerspot butterfly popu¬ lations. 2. The pattern of pupal mortality was variable enough over time to play an important part in controlling the population dynamics of these animals; the proportion of pupae successfully eclosing ranged from .11 to .47. 3. Predation by predators leaving remains was the most constant portion of pupal mortality from year to year. 4. Other mortality factors (predation by predators that left no traces, being stepped on, and dying intact) varied greatly from one year to the next. 5. An ichneumonid parasitoid was found which oviposits in and emerges from pupae of the Bay Checkerspot butterfly. 6. Pupal mortality varies with the amount of foliage around the pupa, with more foliage resulting in less mortality from predation and crushing, but more from mold and viruses. More foliage results in a net improve¬ ment in survival rate. 7. Pupation in the field took 18 days under relatively favorable thermal conditions. Under colder conditions it took as long as 27 days and develop¬ mental failure was common. Acknowledgments. I gratefully acknowledge my debt to Dennis D. Murphy, without whose cooperation this work would have been much more onerous. Paul R. Ehrlich provided access to his group’s accumulated data. Irene L. Brown, Jane L. Hayes, Dennis Murphy and two anonymous reviewers provided manuscript critiques. Literature Cited BOWERS, M. D., 1980. Unpalatability as a defense strategy of Euphydryas phaeton. Evolution 34:586-600. _ , 1981. Unpalatability as a defense strategy of Western checkerspot butterflies. Evolution 35:367-375. DEMPSTER, J. P., 1967. The control of Pieris rapae with DDT. I. The natural mortality of the young stages. J. Applied Ecol. 4:485-500. _ , 1983. The natural control of populations of butterflies and moths. Biol. Rev. 58:461-481. CRITTENDEN, M. & A. GRUNDMANN, 1984. Jasper Ridge. Fremontia 12 (April):20- 21. EHRLICH, P. R., 1965. The population biology of the butterfly, Euphydryas editha.U. The structure of the Jasper Ridge colony. Evolution 19:327-336. EHRLICH, P. R., R. R. WHITE, M. C. SINGER, S. W. McKECHNIE & L. E. GILBERT, 1975. Checkerspot butterflies: An historical perspective. Science 188:221-228. EHRLICH, P. R. & COLLEAGUES. Unpublished mark-release-recapture data in Ehrlich’s Files at Stanford University, Stanford, CA. Collected by Paul and 62 J. Res. Lepid. his many students 1960-1984. HARCOURT, D. G., 1966. Major factors in survival of the immature stages of Pieris rapae (L). Can. Ent. 98:653-662. IWASA, Y., F. J. ODENDAAL, D. D. MURPHY, P. R. EHRLICH & A. E. LAUNER, 1983. Emergence patterns in male butterflies: An hypothesis and a test. Theo¬ retical Population Biology 23:363-379. KERR, R. A., 1985. Wild string of winters confirmed. Science 227:506. KRUCKEBERG, A. R., 1984a. California’s serpentine. Fremontia 11 (January):ll- 17. _ , 1984b. The flora of California’s serpentine. Fremontia 12 (April) :3- 10. POLLARD, E., 1979. Population ecology and change in range of the white admiral butterfly Ladoga Camilla L. in England. Ecol. Ent. 4:61-74. RAHN, J. J., 1971. Growing Degree Days of the 1971 growing season. Weekly Weather and Crop Bulletin March 29, page 11. SIMS, S. R. & A. M. SHAPIRO, 1983. Pupal color dimorphism in California Battus philenor (L.) (Papilionidae): Mortality factors and selective advantage. J. Lepid. Soc. 37:236-243. SINGER, M. C., 1971. Ecological Studies on the butterfly, Euphydryas editha. Ph.D. dissertation. Stanford University, Stanford, California. SOMMERS, S., 1984. Edgewood Park. Fremontia 12 (April): 19-20. STAMP, N. E., 1984. Interactions of parasitoids and checkerspot caterpillars Euphydryas spp. J. Res. Lepid. 23:2-18. STERNITZKY, R. F., 1937. A race of Euphydryas editha Bdv. Can. Entomol. 69: 203-205. TILDEN, J. W., 1958. Notes on the life history of Euphydryas editha hayensis. The Lepidopterists’ News 12:33-36. WATANABE, M., 1976. A preliminary study on population dynamics of the swallow¬ tail butterfly, Papilio xuthus L. in a deforested area. Researches on Popula¬ tion Ecology 17:200-210. WATANABLE, M. & K. OMATA, 1978. On the mortality factors of the lycaenid butter¬ fly, Artopoetes pryeri M. (Lepidoptera, Lycaenidae). Jap. J. Ecol. 28:367- 370. WEST, D. A. & W. N. HAZEL, 1982. An experimental test of natural selection for pupa¬ tion site in swallowtail butterflies. Evolution 36:152-159. WHITE, R. R., 1973. Community relationships of the butterfly, Euphydryas editha. Ph.D. dissertation, Stanford University, Stanford, California. WIKLUND, C., 1975. Pupal color polymorphism in Papilio machaon L. and the survival in the field of cryptic versus non-cryptic pupae. Trans. R. Entomol. Soc. Lond. 127:73-84. The Journal of Research on the Lepidoptera 25(l):63-66, 1986 Chromosome Aberrations in the Holocentric Chromosomes of Philosamia ricini (Saturnidae) Kunja Bihari Padhy Department of Zoology, Bonaigarh College, Sundergarh, Orissa, India Abstract. The nature of chromosome aberrations was studied in Fx male progeny of irradiated male parents. Translocation rings (0.34%), chains (3.7%) and fragments (15.9%) were found. Translocation chains outnum¬ bered the frequency of rings and appear to be produced from the latter by dissociation of chiasma. Dissociation of more than one chiasmata pro¬ duces bivalents and monovalents indistinguishable from the parental ones. Fragments were transmitted to the Fx offsprings stabily. Inversions were rare. Introduction Chromosome aberrations in the holocentric chromosomes usually behave in a different pattern from those in monocentrics: fragments are frequent and are stabily transmitted through several generations (Tempel- aar, 1979). The nature of observed rearrangements of holocentric chromosomes in structural hybrids are, however, still doubtful (White, 1973) and to date reports on such aberrations in Lepidoptera are almost lacking. This perhaps is due to the isodiametric, numerous, much smaller holocentric chromosomes. In view of the above parameters the present study was undertaken with Philosamia ricini using 60Co gamma ray source as the inducing agent. Methods and Material Adult males of P. ricini, were irradiated with an acute dose of 60Co gamma ray (dose rate, 165.5 R/min.). They were held for 24 hours and then mated to virgin females. The Fx male offspring were examined for meiotic chromosomal rearrangements. Cytological preparations were stained by an improved Orcein-Giemsa (OG) technique. Slides were observed and photographed using high power light microscopy. Results Spontaneous chromosome aberrations have been rarely observed in this species. However, preparations of chromosomes of the Fx male progeny obtained from the crosses of the irradiated male parent revealed the trans¬ location rings, chains and fragments during the first spermatocytic phase 64 J. Res. Lepid. (Padhy, 1983). Chromosomal translocations of chains (Fig. 2) and rings (Fig. 3) included reciprocal exchanges of segments between non- homologous chromosomes of the irradiated parent. These formed synapses and chiasmata, usually terminally, characteristic in the Lepidoptera (Fig. 1). Such reciprocal translocations have also been reported in the mite Tet- ranichus utricae (Tempelaar, 1979). Terminalisation of chiasmata was also seen in the translocated chromosomes. Terminalization of one of the chiasmata of a tetravalent ring (Fig. 3) could result in a straight chain of four (two exchanged) chromosomes with three chiaqsmata intervened among them (Fig. 2). This structure is frequently noticed (Fig. 4). In the meiotic spermatocytes bearing the reciprocal translocations of the Fv the number of bivalents was reduced to 12 (n=14, Fig. 4) or less, excluding the translocation tetravalent. The translocation chains, however, cannot result from either fragmentation or differential condensation of chromatin material because such events were not observed in the gamma irradiated meiotic preparations of the male parent studied after 12, 24, 48 hour inter¬ vals (Padhy, 1983). As is clear from Table 1, translocation chains (3.7%) were more frequent than the translocation rings (0.34%) and are thus about ten times more fre¬ quent. This result indicates a rapid terminalisation of one of the chiasmata of the tetravalent before metaphase I (Fig. 5A) . Rapid terminalization of two chiasmata could give rise to two bivalents each attached by a single chiasma in between (Fig. 5B). These cannot morphologically be differen¬ tiated from normal bivalents and therefore pass undetected. The third type (Fig. 5C) indicates a chain of three chromosomes attached by two chiasma which come across in Fx meiosis. The other complement appears Figs. 1 and 4. Normal diakinesis and metaphase I in spermatocytes. Fig. 2. Translocation chain in a diakinesis spermatocyte of the F1 male from the gamma irradiated male parent. Two exchanged and two nonexchanged chromosomes remain associated by the inter¬ vening chiasmata. Fig. 3. Metaphase I spermatocyte of the F1 male progeny of the gamma irradiated male parent. Arrow indicates the translocation quad¬ rivalent with a chromosome complement of 13 bivalents. 25(1) :63-66, 1986 65 b) o dj f 'WWW + /VVAAA' ,+ (VWWV /WW^ ; Fig. 5. Mating protocol of the irradiated male P. ricini crossed to a normal female. F1 male meiosis indicates a translocation ring during diakinesis. This could possibly give rise to four types of chromosome associations by thechiasma: a) chains of four, b) chains of two, c) chains of three and one, and d) four univalents. + = chiasma points, / = irradiated, TD = translocation during diakinesis Table 1. Frequency of aberrations in the Fx male progeny of the gamma irradiated male parents of P. ricini. Stage Cells Observed Translocation Rings N Translocation Chains N Total Translocations (%) Fragments N diplotene- diakinesis 140 2 16 12.8 24 metaphase I 740 1 17 2.4 44 Total 880 3 33 _ 68 Control 1000 (0.34%) (3.7%) (15.9%) 66 J. Res. Lepid. as a fragment or univalent. An association of four univalents (Fig. 5D) is also possible. The occasional appearance of univalents were noticed in many of the translocated spermatocytes, but these could not be differentiated from normal monovalents relative to the dissociation of a single bivalent. The percentage of translocations was more frequent in late prophase spermatocytes than in metaphase. The reasons for this are given in Figure 5B which shows that translocations cannot be differentiated morpho¬ logically from the normal complements during metaphase I. This abnor¬ mality in the formation of chiasmata might partly be due to intragenic alterations and partly due to the absence of a centromere in lepidopteran chromosomes. The latter explanation is supported by the work of Bauer (1967) in the Lepidoptera, Murakami and Imai (1974) in Bombyx mori and Cooper (1972) in the mite Siteropsis graminum. Fragments were, however, transmitted to the F: structural hybrid in 15.9% of the spermatocytes (Padhy, 1983). Inversions were rare. Acknowledgments. B. Nayak, Khallikote College, Berhampur, generously supervised this work. Literature Cited BAUER, H., 1967. Die Kinetische Organisation der Lepidopteran-Chromosomen, Chromosoma, 22:102-125. COOPER, R. S., 1972. Experimental demonstration of holokinetic chromosomes and of differential radiosensitivity during oogenesis in the grass mite Siteropsis gramium (reuter). J. Exptl. Zool. 182:69-72. MURAKAMI, A. & H. T. IMAI, 1974. Cytological evidence of holocentric chromosomes of the silkworm Bombyx mori and Bombyx mandarina (Bombycidae, Lepi¬ doptera), Chromosoma, 47:167-178. PADHY, K. B., 1983. Ph.D. thesis, Utkal University, India. TEMPELAAR, M. J., 1979. Aberrations of holocentric chromosomes and associated lethality after X- irradiation of meiotic stages in Tetranichus utricae Koch. (Acari, Tetranychidae) Mut. Res., 61:259-274. WHITE, M. J. D., 1973. Animal Cytology and Evolution, 496 pp. The Journal of Research on the Lepidoptera 25(l):67-70, 1986 Opinion. Opinion is intended to promote communication between lepidopterists resulting from the content of speculative papers. Com¬ ments, viewpoints and suggestions on any issues of lepidopterology may be included. Contributions should be as concise as possible and may include data. Reference should be limited to work basic to the topic. Rebuttal to Murphy on Factors to the Distribution of But terfly Color and Behavior Patterns — Selected Aspects Benjamin H. Landing 4513 Deanwood Drive, Woodland Hills, California 91364 In reply to Murphy’s critique of my work (Jr. Res. Lep., 1985, 24:4). Since I think the scientific system ultimately decides the validity of pro¬ positions by scientific criteria, without too much attention to who holds positions pro or con, and doubt that anyone benefits from further airing of our differences, however strongly felt, on matters which I don’t think have much to do with the scientific content of the book, such as why the material was published in this form rather than some other, whether the title is misleading, or why the cover is black and white, I am concerned that the most generally useful response may be none at all. (The butterfly on the cover, parenthetically, is an Ideopsis juuenta , an oriental region danaid.) However, since Murphy proposes that he and I have basic dif¬ ferences about various scientific facts and scientific procedures, I will attempt to make clear my position on several points. 1. Murphy objects to the use of models or general schemes to explain observations as involving circular reasoning. I disagree, and think this is exactly the general function of models or general schemes in scientific reasoning — to bring order to uncoordinated observations or to offer an explanation of unexplained ones. It is true that making predictions is in effect proposing explanations for observations not yet made, but this is normally done in the circumstance that the proposed explanation appears to explain an observation similar to the one predicted, namely, when that model has already fulfilled its function. I cannot think of a way of explain¬ ing observations which does not require knowing what at least some are. 2. I think the definition of ecological niche given by Murphy is inade¬ quate for butterflies because it lacks the qualifier, “. . for each stage of the life cycle.” I think that a definition which says that adult butterflies and their caterpillars have exactly the same ecological niches overlooks too much, and do not see the objection to the concept of niches, or subsets of 68 J. Res. Lepid. niches, for adult butterflies. The calculations I gave did not specify either the terms of such niches, nor the number of different possible values of each term, but simply illustrated what one got if one did make certain assumptions about these. Murphy and I agree that not all loci contain all possible niches, which was the point, although (see below) we disagree on why this is so, at least in part, if the interrelated features of color pattern and preferred height of flight in the vegetation are part of the definition of the “niche” of an adult butterfly. 3. Although the book contains a variety of conclusions and propositions on a variety of matters, I think the single most important part is that (Chapters 1-5 and parts of others) dealing with the proposed general scheme, which relates the color patterns of the butterfly species found at a locus to the height of vegetation at that locus and the preferred height of flight in the vegetation (from the top down) of butterflies with each specific category of color patterns. The scheme addresses the fact that, as one goes north from the tropics in this hemisphere, for example (or up a mountain in the tropics), specific color patterns disappear in a sequence, with, as one goes north, transparent patterns dropping out in Mexico, tiger-stripe (orange with transverse dark stripes) patterns at about the Texas border, and black with red patterns and black with blue patterns progressively farther north, so that at about the arctic circle (or above tree line) the only species truly resident at the locus have as color patterns only relatively “pure color” white, yellow, orange, blue or lighter brown, or intergrades of these. Murphy proposes that this is due to the reduction in total number of species resident at any locus which occurs as one goes north, but I believe this is not an adequate explanation because it does not explain the systematic shift in the proportions of species resident at any locus which have specific color patterns as one does this. The question the scheme is addressing is not, “why are there fewer transparent or tiger stripe species in the United States than there are in the tropics?,” for example, but, “why are there none?” 4. The sequence given above coincides with that given by Papageorgis in her description of the layering of flight levels of butterflies with various color patterns in amazonian forests. Murphy says her paper on this is “controversial,” but not that her description of the layering is incorrect, and my own field observations in five countries in the American tropics convince me that it is correct. It is presented as fact (although without specific attribution) and illustrated by Sbordoni and Forestiero (pages 212-213), for instance. 5. My proposition is that the identity of these two sequences is not an accident, but reflects the workings of a specific underlying mechanism, and I supported the proposition that the sequences are what they are because selection has “geared” color pattern to height of flight in the vegetation because each pattern is most effectively cryptic at the level in the vegetation (again, from the top down) at which species with that pat- 25(1) :67-70, 1986 69 tern regularly fly. Murphy says that I did not adequately consider the roles of “oviposition host selection and breadth, the role of nectar as a limiting resource, the use of alternative sources of carbohydrates and amino acids, thermal constraints on butterfly activities, how resources are partitioned, how butterfly diversity and plant diversity correlate and so on.” I do not see that any of these necessarily make specific butterfly color patterns occur or not occur at specific loci, and do not think his list contains the mechanism. We know for instance, both that closely related species (e.g., the viceroy and the red-spotted purple) can develop both color patterns in different color classes and the appropriately different flight levels, and that males and females of sexually dimorphic species (e.g., eastern tiger swallowtail, Diana fritillary) can have patterns in different color classes. I also do not think that differences in heat-collecting capacity of different wing colors are the explanation because, for example: 1) the color pattern group most specifically associated with the deepest part of amazonian forests is the transparent one, not one of the darker ones, and; 2) the color patterns persisting in the far north are the lighter “pure color” white, yellow, orange, blue or brown ones, not the darker ones. 6. I think the next most important section of the book (Chapters 7, 8) deals with the points that a number of still stated criteria of mimicry sys¬ tems are unnecessary, and in many specific instances not correct. These include the idea that in Batesian mimicry systems models must be more abundant than mimics in all loci, and that in Batesian systems models and mimics, and in Muellerian systems co-models (or co-mimics), must have the same ranges, because the rules overlooked the point that many birds migrate. Murphy happens not to criticize this portion of the book. 7. Murphy sees the data tables as a “smoke screen.” Again, I disagree, because I do not think one can expect people to evaluate scientific con¬ clusions or propositions without access to the data on which they were based. Most of the data are not mine originally, as is made clear throughout, but are derived from the publications of others, and are assuredly not generally wrong, so I think drawing conclusions from them or making propositions based on them is not scientifically inappropriate. The largest data set in the book which is strictly my own is that in the chapter on interference color patterns, which chapter Murphy happens not to criticize. 8. The book discusses a variety of other facts which “stick out” of the data, and offers conclusions or propositions based on them, including: a) there is, overwhelmingly, a systematic relation between the color patterns of males and females of sexually dimorphic species, and the differences follow the pattern of the classes in the general scheme. If the whole thing is chance, why should this be? b) the proportions of pierid and lycaenid species which have mistletoe¬ feeding larvae decline disproportionally as one goes north from the 70 J. Res. Lepid. tropics. c) toxic/protected papilionids are less likely than Papilio species in the same regions to show sexual dimorphism with the sexes in different color classes. d) toxic/protected species are more likely than others to have similar color patterns on both upper and lower wing surfaces. As regards these latter two, since intraspecific Batesian mimicry is accepted for the monarch, for instance, I don’t think that what amounts to propositions that intraspecific Muellerian mimicry and, in fact, intra¬ individual Muellerian mimicry, also occur are particularly radical ideas, but I have never heard either one presented before. To me these again illus¬ trate the importance of access to the data. (A possible volume two, perhaps unfortunately already over 300 pages long, contains a proposed explanation of the point on mistletoes, among many other things.) The Journal of Research on the Lepidoptera Notes 71 Field Notes on Clossiana improba harryi Ferris (Lepidoptera: Nymphalidae) This species was described in 1984 (Ferris, C. D., Bull. Allyn Mus. 89:1-7) from specimens collected in 1982 by Jack L. Harry of Salt Lake City, Utah. Field collect¬ ing by Ferris in 1984 and in 1985 by Lisa Snyder from the Audubon Ecology Camp of the West (University of Wyoming Trail Lake Ranch), near Dubois, Wyoming, has increased our knowledge of this species with respect to its behavior and geographic distribution. This butterfly is a denizen of remote, high-alpine areas (above 11,000' (3355 m)) as shown in the type locality photograph (Fig. 1). It flies in early August, and was known originally only from the vicinity of Mt. Chauvenet in the Wind River Range of central -western Wyoming in Fremont Co. The type locality is situated in the Popo Agie Primitive Area of the Shoshone National Forest. C. i. harryi was des¬ cribed originally as occurring in eleven colonies extending for approximately 4.5 miles along the Bears Ears Trail. In 1984, I found that the distribution in this region is not discrete, but rather continuous from west of Adams Pass to west of Mt. Chauvenet. In 1985, Snyder discovered two additional colonies of harryi in the Fitzpatrick Wilderness Area at Goat Flat and Ram Flat. These localities are re¬ spectively 40 and 45 air miles NW of the type locality, also in the Shoshone National Forest in Fremont Co. Figure 2 is a map of this butterfly’s range, as currently known. The habitat of this species is in relatively level, somewhat xeric, areas of granitic Fig. 1. Type locality (looking to the West) and typical habitat of C. improba harryi. 72 J. Res. Lepid. Fig. 2. Map showing the distribution (cross-hatched circles) of C. improba harryi. Only the larger lakes are shown. The dotted lines are hiking trails. gravel on which mats of the larval hostplant ( Salix arctica Pall) grow abundantly. This plant is widespread throughout alpine areas of Wyoming, but the butterfly is very local. Adults of harryi dorsally bask on gravel patches and on the pale granite boulders distributed over their habitat. From the rather dark aspect of museum specimens of this species, one would think that these butterflies would be very con¬ spicuous against the pale background of the gravel and boulders. This is not the case, however, in the field. The pale central areas of the wings (dorsally) produce a cryptic pattern which blends very well with rocky substrates and renders the but¬ terflies difficult to detect. To date, this species has been found only on the east slope of the Wind River Range in Fremont Co., Wyoming. The eastern slope of the Range is considerably drier than the western slope which supports many butterfly fauna. It will be sur¬ prising if harryi is not eventually discovered in neighboring Sublette Co., at appropriate elevation, on the western slope of the Wind River Range. Access to suitable habitat areas, however, is only by foot or horseback over 20 miles or more of rugged terrain. This butterfly is abundant once a colony has been located, and is in no sense endangered, as may possibly be the case for its sibling species in Colorado C. acrocnema (Gall & Sperling). Clifford D. Ferris, P. 0. Box 3351, University Station, Laramie, Wyoming 82071 - 3351 INSTRUCTIONS TO AUTHORS Manuscript Format: Two copies must be submitted (xeroxed or carbon papered), double-spaced, typed, on 8V2 x 11 inch paper with wide margins. Number all pages consecutively and put author’s name at top right corner of each page. If your typewriter does not have italic type, underline all words where italics are intended. Footnotes, although discouraged, must be typed on a separate sheet. Do not hyphenate words at the right margin. All measurements must be metric, with the exception of altitudes and distances which should include metric equivalents in parenthesis. Time must be cited on a 24-hour basis, standard time. Abbreviations must follow common usage. Dates should be cited as example: 4. IV. 1979 (day-arabic numberal; month-Roman numeral; year- arabic numeral). Numerals must be used before measurements (5mm) or otherwise up to number ten e.g. (nine butterflies, 12 moths). Title Page: All papers must have the title , author’s name, author’s address, and any titular reference and institutional approval reference, all on a separate title page. A family citation must be given in parenthesis (Lepidoptera: Hesperiidae) for referencing. Abstracts and Short Papers: All papers exceeding two typed pages must be ac¬ companied by an abstract of no more than 300 words. An additional summary is not required. Name Citations and Systematic Works: The first mention of any organism should include the full scientific name with author (not abbreviated) andyear of description. New descriptions should conform to the format: male: female, type data, diagnosis, distribu¬ tion, discussion. There must be conformity to the current International Code of Zoological Nomenclature. We strongly urge deposition of types in major museums, all type depositions must be cited. References: All citations in the text must be alphabetically listed under Literature Cited in the format given in recent issues. Abbrevations must conform to the World List of Scientific Periodicals. Do not underline periodicals. If four or less references are cited, please cite in body of text not in Literature Cited. Tables: Tables should be minimized. Where used, they should be formulated to a size which will reduce to 4 x 6^2 inches. Each table should be prepared as a line drawing or typed with heading and explanation on top and footnotes below. Number with Arabic numerals. Both horizontal and vertical rules may be indicated. Complex tables may be reproduced from typescript. Illustrations: Color must be submitted as a transparency (i.e., slide) ONLY, the quality of which is critical. On request, the editor will supply separate detailed instructions for making the most suitable photographic ilustrations. Black and white photographs should be submitted on glossy paper, and, as with line drawings, must be mounted on stiff white cardboard. Authors must plan on illustrations for reduction to the 4 x SV2" page. Allowance should be made for legends beneath, unless many consecutive pages are used. Drawings should be in India ink at least twice the final size. Include a metric scale or calculate and state the actual magnification of each illustration as printed. Each figure should be cited and explained as such. The term “plate” should not be used. Each illustration should be identified as to author and title on the back, and should indicate whether the illustration be returned. Legends should be separately typed on pages entitled “Explanation of Figures”. Number legends consecutively with separate paragraph for each page of illustrations. Do not attach to illustrations. Retain original illustrations until paper finally accepted. Review: All papers will be read by the editor(s) & submitted for formal review to two referees. Authors are welcome to suggest reviewers, and if received, submit name & comments of reviewers. THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA Volume 25 Number 1 Spring 1986 IN THIS ISSUE Date of Publication: October 1, 1986 Hidden Genetic Variation in Agraulis vanillae incarnata (Nymphalidae) Thomas E. Dimock & Rudolf H. T. Mattoni 1 Electrophoretic Evidence for Speciation within the Nominal Species Anthocharis sara Lucas (Pieridae) Hansjurg Geiger & Arthur M. Shapiro 15 Genetic Differentiation Between Subspecies of Euphydryas phaeton (Nymphalidae: Nymphalinae) A. Thomas Vawter & Janet Wright 25 On the Monophyly of the Macrolepidoptera, Including a Reassessment of their Relationship to Cossoidea and Castnioidea, and a Reassignment of Mimallonidae to Pyraloidea James A. Scott 30 Electrophoretic Confirmation of the Species Status of Pontia protodice and P. occidentalis (Pieridae) Arthur M. Shapiro & Hansjurg Geiger 39 Susceptibility of Eggs and First-Instar Larvae of Callosamia promethea and Antheraea polyphemus to Malathion Thomas A. Miller, William J. Cooper & Jerry W. Highfill 48 Pupal Mortality in the Bay Checkerspot Butterfly (Lepidoptera: Nymphaqlidae) Raymond R. White 52 Chromosome Aberrations in the Holocentric Chromosomes of Philosamia ricini (Saturnidae) Kunja Bihari Padhy 63 Opinion: Rebuttal to Murphy on Factors to the Distribution of Butterfly Color and Behavior Patterns — Selected Aspects Benj amine H. Landing 67 Notes 71 COVER ILLUSTRATION: Selectively bred adults of Agraulis vanillae incarnata, see Dimock and Mattoni, pages 1-14. THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA K SR SD a. UJ to gi <1 x§ THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA The Lepidoptera Research Foundation, Inc. c/o Santa Barbara Museum of Natural History 2559 Puesta Del Sol Road Santa Barbara, California 93105 William Hovanitz Rudolf H. T. Mattoni, Editor Lorraine L. Rothman, Managing Editor Scott E. Miller, Assistant Editor Emilio Balletto, Italy Miguel R. Gomez Bustillo, Spain')' Henri Descimon, France Thomas Emmel, U.S.A. Lawrence Gall, U.S.A. Brian 0. C. Gardiner, England Hansjuerg Geiger, Switzerland Otakar Kudrna, Germany Dennis Murphy, U.S.A. Ichiro Nakamura, U.S.A. Arthur Shapiro, U.S.A. Atuhiro Sibatani, Japan '('Deceased December 17, 1985 Manuscripts may be sent to the Editor at: 9620 Heather Road, Beverly Hills, CA 90210 (213) 274-1052 Notices Material may be sent to the Managing Editor. The JOURNAL is sent to all members of the FOUNDATION. CLASSES OF MEMBERSHIP: Regular (Individual) $ 15.00 year (vol.) Contributing $ 25.00 or more, year (vol.) Student/Retired — Worldwide $ 11.00 year (vol.) Subscription Rate/Institutions $ 25.00 year (vol.) Life $200.00 STATEMENT OF OWNERSHIP AND MANAGEMENT THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA is published four times a year, Spring, Summer, Autumn, and Winter, by THE LEPIDOPTERA RESEARCH FOUNDATION, INC. The office of the publication and the general business office are located at 2559 Puesta Del Sol Road, Santa Barbara, California 93105. The publisher is THE LEPIDOPTERA RESEARCH FOUNDATION, INC. The Editor is R. H. T. Mattoni at the above Beverly Hills address. The Secretary-Treasurer is Barbara Jean Hovanitz at the general business office. All matters pertaining to membership, dues, and subscriptions should be addressed to her, including inquiry concerning mailing, missing issues, and change of address. The owner is THE LEPIDOP¬ TERA RESEARCH FOUNDATION, INC., a non-profit organization incorporated under the laws of the State of California in 1965. The President is R. H. T. Mattoni, the Vice President is John Emmel, the Secretary- Treasurer is Barbara Jean Hovanitz. The Board of Directors is comprised of Barbara Jean Hovanitz, Lorraine L. Rothman, and R. H. T. Mattoni. There are no bond holders, mortgages, or other security holders. ISSN 0022 4324 Published By: Founder: Editorial Staff: Associate Editors: Journal of Research on the Lepidoptera 25(2): 73-82, 1986(87) A New Species of Calisto from Hispaniola with a Review of the Female Genitalia of Hispaniolan Congeners (Satyridae) by Kurt Johnson Department of Entomology, American Museum of Natural History, Central Park West at 79th St., New York, New York 10024, Eric L. Quinter 37-06 72nd Street, Jackson Heights, New York 11376 and David Matusik Department of Entomology, Field Museum of Natural History, Roosevelt Road, Chicago, Illinois, 60605 Abstract. Calisto ainigma , new species, is described from a unique holotype female collected at Jarabacoa, Dominican Republic in 1985. Female genitalia, not previously studied in Calisto , are compared for twenty-two Hispaniolan congeners. Of all congeners, wing patterning in C. ainigma only slightly resembles C. elelea Bates, a species of limited Haitian distribution. Female genitalia suggest the two may be distant sister species. Introduction The diversity of the satyrid genus Calisto is remarkable. With the recent work of Schwartz (1983a, 1983b, 1987 [in press]) Schwartz and Gali (1984) and Gali (1985), twenty-five Calisto species are recognized as occurring on Hispaniola. Eleven of these result from work of Schwartz and Gali in seldom collected areas of Hispaniola and repre¬ sent endemic taxa with extremely limited known geographic ranges. This latter characteristic of Calisto led Schwartz and Gali (1984, p. 10) to suggest the discovery of further endemic Calisto from Hispaniola as inevitable. Genericly, Calisto are endemic to the Antilles and characte¬ rized by each of the islands exhibiting various endemic species (Mun- roe, 1950). Given the increasing diverity of Calisto taxa recognized as occurring on Hispaniola (11: Bates, 1935, 1939; 12: Michener, 1943; 14: Clench, 1943a, 1943b; Riley, 1975; 20: Schwartz and Gali [see p. 1]; 25: Gali, 1985) it may be anticipated that when adequately sampled, Cuba may also yield a diversity of Calisto taxa. With the appearance of 74 J. Res. Lepid. Schwartz’s (1983a; 1987 [in press]) treatments of Hispaniola butter¬ flies, an ample amount of data concerning the taxonomy and distribu¬ tions of Hispaniolan Calisto has now been accumulated. Characteristic of the results of Schwartz and Gali’s work has been documentation that among the few cosmopolitan Hispaniolan Calisto (like confusa Lathy and obscura Michener) there occurs a number of other endemic species characterized by (a), marked wing pattern differ¬ ences from the previously known congeners and slight, if any, sexual dimorphism and (b). extremely limited local known distributions often marked by restriction to a single known locality or limited habitat. The former results from the lack of comprehensive collecting prior to the work of Schwartz and Gali; the latter reflects the often extreme fragmentation of the native habitats of the island, remnants of which are now often only found in very limited undisturbed areas of inaccessi¬ ble topography. This latter factor also seems to explain the lack of any recent association of specimens with the names C. montana and C. micheneri Clench (1943a, b). The holotypes of these species have been illustrated by Riley (1975) but both are from localized and remote localities from which no further specimens have been taken in recent years. Hitherto, all studies of Calisto have examined characters solely of the wing and male genitalia. Given the recent accumulation of studies of Calisto cited above, an examination of female genitalia of the group is requisite and timely. Further, such an examination has been required by the collection in the Central Cordillera in 1985 of a female specimen of Calisto (hereinafter in introduction referred to as “the Jarabacoa female”) with wing pattern quite unlike any previously known taxon of the genus (Albert Schwartz, pers. comm). Matusik captured the speci¬ men while he and Johnson were collecting along a stream near Jaraba¬ coa, La Vega Province, Dominican Republic, June 26, 1985. A perfectly fresh specimen, it had attracted attention because amongst extremely common C. obscura and C. confusa which “flash” brown and submar¬ ginal white when flying, this specimen was markedly yellowish. Upon capture, the several unique traits noted in the following diagnosis were obvious and further heightened interest in the specimen. Unfortunate¬ ly, due to pre-arranged itinerary the collectors had to leave the area that day; they returned with additional local collectors a week later but concerted Calisto collecting yielded no further examples. Fig. 1. Female genitalia of Flispaniolan Calisto. Format, each entry, above: papillae anales, lateral view; below: genital plate, ductus bursae and corpus bursae, ventral view. A. C. elelea (AMNH), Sierra de Baoruco, 12 km. from Las Abejas on Las Abejas highway, Dominican Republic [D. R.], 400 m., May, 1984, D. Matusik; B. C. ainigma, holotype (AMNH); C. C. obscura, paratype (AMNFI), Puerta Plata, D. R., 7-8 May 1915; D. C. confusa (AMNH), Trujillo City, D. R., 1946, A. L. Stillman; E. C. debarriera (AMNH), 10 km. SE Constanza, D. R., 1270 m., D. Matusik; F. C. batesi, same data as A.; G. C. lyceia Bates (MCZ), Isla Saone, D. R.; H. C. tragia (ASC), 1-4 km. WNW 25(2): 73-82, 1986(87) 75 Fig. 1. (Cont). Scierie, Sud-Est, Haiti [H], 2000 m., 4 September 1984, A. Schwartz; I. C. micrommata (ASC), 2 km. NE Puesto Piramide 204, La Estrelleta, D. R., 1700 m., 16 July 1983, A. Schwartz; J. C. sommeri (AMNH), 38 km. marker, 2 km right turn to Nursery, highway to Las Abejas, D. R., 1600 m., May, 1984, D. Matusik; K. C. hysia Godart (AMN), Paradis, D. R., 600 m., 15 August 1932; L. C. grannus (ASC), 21 km. SE Constanza, La Vega, D. R., 2500 m., 10 July 1980, A. Schwartz. 76 J. Res. Lepid. The Jarabacoa female was donated to the AMNH and has since been studied in more detail, along with examination of the relevant litera¬ ture, specimens from the collections of the junior author, AMNH, Allyn Museum of Entomology, Museum of Comparative Zoology (Harvard) (MCZ), Albert Schwartz (ASC) and Frank Gali, and female genitalia of Hispaniolan congeners represented in these collections (Figs. 1 and 2). There has been considerable discussion amongst students of Hispa¬ niolan butterflies concerning the status of the Jarabacoa female con¬ sidering its extremely unique wing markings and occurrence at one of the most frequently collected localities on Hispaniola. Schwartz (pers. comm.) advised that even though its markings did not seem closely comparable with any known Calisto, the specimen must be suspected as a possible aberration of either of the common local congeners, C. confusa or C. obscura. Dissection of the unique Jarabacoa female has revealed a genitalic configuration differing radically from both C. confusa (Fig. 1, D) and C. obscura (Fig. 1, C) as represented by topotypical, paratypical and syntopic/synchronic examples. In addition, the genitalia of the Jarabacoa female do resemble those of another known Calisto (Fig. 1, A) and further examination of this latter taxon has indicated certain wing pattern similarities (Fig. 3). As a result, these data suggested three alternative treatments concerning the Jarabacoa female: 1. Conclude from the wing pattern and genitalic characteristics that it represents an undescribed species of Calisto whose taxonomic posi¬ tion in the genus is concordant ( sensu Murphy and Ehrlich, 1984, p. 27) with an overall view of morphological and biogeographic characteristics of the group. 2. Conclude by speculation that it is an aberration of some previously described species of Calisto , though the latter cannot be designated because of the divergent wing morph of the former. 3. Accord no published recognition to the unique specimen pending further sampling. We believe that genitalic and wing character evidence assembled in this study (Figs. 1—3) along with the highly insular nature of many Calisto distributions warrants the first kind of treatment. We would have accepted the second treatment if the genitalia of the Jarabacoa female had resembled any geographically proximate congener. We Fig. 2. Female genitalia of Flispaniola Calisto, continued. Format as in Fig. 1. A. C. areas (ASC), 14 km. SE Constanza, La Vega, D. R., 2100 m., 20 July 1985, A. Schwartz (small letters referenced in text); B. C. crypta Gali (AMNH), Monte Christi, D. R., 13 March 1931, A. L. Stillman; C. C. franciscoi (ASC), 8 km. ESE Canoa, Barahona, D. R., 28 July 1985, A. Schwarts; D. C. hender- soni (ASC), 4 km. E El Limon, Independencia, D. R., 2 April 1984, A. Sch¬ wartz; E. C. schwa rtzi (AMNH), same data as Fig. 1, J; F. C. clydoniata (ASC), 2 km. NE Puesto Piramid 204, La Estrelleta, D. R., 1400 m., 13 August 1983, A. Schwartz; G. C. galii (ASC), 10 km. SE Constanza, D. R., 1800 m., 25(2): 73-82, 1986(87) 77 Fig. 2. (Cont). 9 July 1980, A. Schwartz; H. C. clenchi Schwartz and Gali (ASC), 5 km, Ne Los Arroyos, Pedernales, D. R., 1800 m., 30 June 1983, A. Schwartz; I. C. chrysaoros Bates (ASC), 5 km. NE Los Arroyos, Pedernales, D. R., 1800 m., 4 October 1983, A. Schwartz; J. C. neiba Schwartz & Gali (ASC), 15 km. S. Elias Pina, La Estrelleta, D. R., 1100 m., 26 July 1981, A. Schwartz. 78 J. Res. Lepid. Fig. 3 Eight under surface similarities between C. ainigma (left) and C. elelea (right). 1. completely red-orange discal cell; 2. lightened apical ground color; 3. outline of marginal band ( elelea ) (fully black in ainigma ); 4. light post-basal line; 5. darkening costad in medial area; 6. darkening costad in limbal area; 7. light mesial line extending from anal angle basad hindwing ocellus; 8. darkened spot at anal angle. consider the third action inappropriate because (a), there has been a paucity of study of female genitalia in Calisto hitherto (see Remarks), (b). lack of such study has left a number of variant females cited in the literature as undetermined and (c). not recognizing the wing and genitalic features of the Jarabacoa female would result in loss of their potential taxonomic and biogeographic information as regards ongoing studies of Calisto. We therefore propose the following: Calisto ainigma, Johnson, Quinter & Matusik, new species Figs IB, 3A, 4 Diagnosis. Distinguishable from all other known Calisto by the following marked characters: (1). undersurface ground color distinctly yellow to ochre [yellower than in photo (Fig. 4), such hues caused by differential spacing of deep brown scales amongst bright yellow scales], not brown or grey as on congeners; (2). both wing undersurfaces with wide (1 mm.) olive-black margin¬ al band, not occurring on any congener; (3). aside from unique marginal band, hindwing lacking any bands (congeners variously have postbasal medial, postmedial and/or submarginal bands of various colors and/or a dark basal disc with its distad margin bandlike [C. montana, C. micheneri ]). Rather, C. 25(2): 73-82, 1986(87) 79 ainigma has a yellow ground color appearing as blackish-grizzled from the wing base distad to an indistinct medial juncture with purer yellowish ground color distad in the postmedian areas to margin. Intense blackish-grizzling centered costad along this medial juncture, along with invasion basad of the marginal line in cells Mi and M2, adds further oddity to the pattern; latter suffusions resemble C. elelea Bates (Fig. 3) which is otherwise banded; (4). as only in C. montana, C. micheneri, and C. tragia Bates, hindwing with single ocellus [cell CUi] devoid of any obvious surrounding patterning [not with [a], single ocellus surrounded by various maculation expansive distad (C. confusa, C. obscura, C. hysia Bates, C. elelea, C. clydoniata Schwartz and Gali) or [b]. two ocelli, one at each end of the limbal area (C. grannus Bates, C. micrommata Schwartz and Gali, and C. sommeri Schwartz and Gali)]; C. ainigma, like C. elelea, has distinctly lighter ground color based cell CUx ocellus, in latter a band; (5). hindwing undersurface with two white dots in cells M2 and M3 (not with three in cells M2 to Rs as in C. galii Schwartz [see Remarks for further significance of this feature]. Description. Male. Unknown. Female. Uppersurface of the Wings: Ground color ochre-tinted olive brown, especially distad, with wings darker olive basad. Otherwise no distinctive markings. Undersurface of the Wings: Forewing ground color ochre- tinted olive with prominent subapical ocellus [diameter 2.8 mm.], black centrad, ringed yellow and with two blue-white dots within. Surrounding subapical area and adjacent postmedian area sheened lighter yellowish olive. Prominent 1 mm. olive black marginal band. Hindwing ground color yellowish-ochre; except for 1 mm. wide olive-black marginal band, with¬ out any other bands. Rather, blackish-grizzling proceeds from wing base to variously distinct medial juncture with yellow-ochre ground color distad on remainder of wing. Blackish grizzling concentrates costad along this medial juncture; marginal band intrudes basad in cells M2 and Mi. Distad medial juncture of black grizzling, yellowish ground color broken only by two white dots in cells M2 and M3 and small but prominent ocellus [diameter 1.0 mm.] in cell CU2, black centrad, ringed yellow and with white dot within. Forewing length: 16 mm. Male Genitalia. Unknown. Female Genitalia. Fig. IB. Of congeners studied, sharing with C. elelea (a), thickened ring of genital plate (see Remarks), ring heavily “wrapped” with membranous folds obscuring widened under-lying sclerotized ring which in other ringed taxa (see Remarks) is thinner and not heavily membranous, and (b). dorsad configuration of the ring comprised of two bilaterally symmetrical widened areas, extremely thick¬ ened and bulbous relative to congeners and which on C. ainigma shows a tapered, dorsad pointing extension. Corpus bursae markedly shorter on C. ainigma than C. elelea and with signa of former located far cephalad the juncture of this bursae with the membranous ductus bursae. Type. Holotype, female, deposited AMNH, La Vega Province, Dominican Republic, 930 m. in central portion of Cordillera Central, June 26, 1985, by David Matusik at site characterized as follows: along a small (1.5— 2.5 m. wide) stream currently running between the Hotel Pinar Dorado’s group of “caba¬ nas” and the highway that proceeds from the immediate entrance to the hotel grounds about 4 km. northwest to central Jarabacoa (which is expanding its outer perimeter by active outlying home development). Stream crosses a fenced cattle grazing break between the stands of Australian Pine which border it west along the highway and east east of the cabanas. Specimen taken in grass 80 J. Res. Lepid. along this stream about 300 meters north of the hotel and its entrance to the highway (e.g. ca. 4 km. southeast of Jarabacoa). Remarks. Schwartz (1983a, fig. I, J) and Schwartz and Gali (1984) mention variant females which they either associate as aberrants with known Calisto taxa or which show facies leading them to conclude “another species of Calisto presumably occurs in the Cordillera Central” (Schwartz and Gali, 1984, p. 10). Concerning these, and undescribed taxa currently being described by Schwartz or Schwartz and his colleagues, Schwartz (pers. comm.) has assured us that none is similar enough to the facies of C. ainigma to warrant discussion here. The genitalic survey conducted during this study warrants the following general remarks. Characters of the female genitalia apparently provide a far more useful reference for Calisto than those of males. Male genitalia of Calisto , which have been reviewed to some extent by nearly all authors cited herein, are mostly alike. Minor but consistent differences have been cited, particularly by Michen- er and Schwartz et al., to distinguish various taxa which also have distinctive wing pattern characters. Within Calisto, as presently defined, the only radical¬ ly divergent male genitalia amongst Hispaniolan taxa occur in C. elelea, C. pulchella Lathy, C. areas and C. raburni Gali. As can be seen in Figs. 1 and 2, such divergence is reflected in the female genitalia of both C. elelea and C. areas, though the former is more like other Calisto. C. pulchella is not figured becaused its female genitalia are so divergent as to suggest lack of recognizable homology with other taxa presently placed Calisto, a matter presently under study. C. raburni is recently described and its female unknown. Two general genital plate configurations are apparent in Calisto studied, those with two obvious components and those with only one. Other taxa are intermediate between these extremes. C. areas (Fig. 2A) best exemplifies a two component structure: a sclerotized ring (Fig. 2A, a) with a sculptured dorsal crown (Fig. 2A, b) and a sclerotized ductal tube (Fig. 2A, c) with dorsad “horns” (Fig. 2A, d). In C. franciscoi Gali and C. hendersoni Gali (Fig. 2C, D) apparent remants of these horns appear within a configuration otherwise characterized by distinct separation of the ring and crown. C. schwartzi Gali (Fig. 2E) exhibits remants of the horns closely allied with the ring and crown combination. In the remaining Calisto (Fig. 1, 2F, F-J) the ring, closely combined with the crown, forms a generalized configuration. However, within this group some taxa exhibit a sclerotized loop within the ring (Fig. 1C, F-L), or without a loop, variously developed cephalad pointing prongs (Fig. 1, A, B, D-J). The particu¬ lar structure characterizing C. elelea and C. ainigma has been described in the above description. Within Calisto there are also apparent differences in the configurations of the papillae anales, ductus bursae and corpus bursae with its associated signa. It is likely that these characters will prove very useful in examining the taxonomic and biogeographic relations of the Calisto endemic to various Antillean islands. Such a study is in progress. At present it is important to note that female genitalic characters corroborate the species statuses accorded the numerous presently recognized species in Hispaniola, and particu¬ larly of interest those named very recently by Schwartz and Schwartz and Gali [see p. 1]. The only exception might be C. confusa and C. debarriera Clench which , though considered full species on biological grounds (Schwartz, pers. comm.) are very similar compared to other congeners. As regards the often 25(2): 73-82, 1986(87) 81 debated species status of C. hysius ssp. batesi Michener (Clench, 1943b; Schwartz, 1983a; Riley, 1975), female genitalia appear to provide a moderately strong argument supporting C. batesi’ s specificity. The similarity in female genitalic facies of C. ainigma and C. elelea was unanticipated. The latter species has a highly insular distribution limited to montane areas surrounding Port-au-Prince, Haiti. Subsequently noted similar- Fig. 4. Holotype female, Calisto ainigma , new species. Left, upper surface of the wings; Right, under surface of the wings. ities in certain aspects of the wings patterns of C. elelea and C. ainigma (Fig. 3) are likewise suggestive and have invited the conclusion that the facies of C. ainigma is not so extraordinary as originally presumed by us and other workers familiar with Hispaniolan Calisto (e.g. Schwartz, pers. comm.). Male genitalia of C. elelea are distinctive such that among Calisto Brown and Heineman (1972, based on Michener, 1943) placed this species within a monotypic species group. It will be of extreme interest whether the male of C. ainigma, once discovered, further corroborates C. ainigma’s placement with C. elelea as a sister taxon. Etymology. The name is Greek for “enigma”, referring to the curious wing pattern, occurrence at the often collected Jarabocoa area, and unanticipated suggested sister species relationship to C. elelea. Upon the suggestion of Schwartz and Gali (1984) and Gali (1985) species names in this paper have been made to conform to the feminine gender of the name Calisto. A single exception is C. grannus, the origin of which name Schwartz states is indeter¬ minate. Acknowledgements Albert Schwartz (Miami Dade County Community Col¬ lege, Miami, Florida) kindly reviewed drafts of the manuscript and supplied various female Calisto for dissection. Lee D. Miller (Allyn Museum of Entomol¬ ogy of the Florida State Museum, Sarasota, Florida) and Frederick H. Rindge (AMNH) also reviewed the manuscript. Luis Marion (Santo Domingo, Domini¬ can Republic) and Robert R. Postelnek (Skokie, Illinois) kindly facilitated aspects of this work. 82 J. Res. Lepid. Literature Cited BATES, M. 1935. The satyrid genus Calisto. Occ. Papers Boston Soc. Nat. Hist. 8: 229-248. BROWN, F. M. & B. HEINEMAN. 1972. Jamaica and its Butterflies. E. W. Classey, Ltd., London, sv + 478 pp. CLENCH, H. K. 1943a. Some new Calisto from Hispaniola and Cuba (Lepidoptera: Saryridae). Psyche 50: 23-29. CLENCH, H. K. 1943b. Supplementary notes on Calisto (Lepidoptera: Satyridae). Psyche 50, unnumbered page. GALI, F. 1985. Five new species of Calisto (Lepidoptera: Satyridae) from Hispa¬ niola. Milwaukee Public Museum Cont. to Biol, and Geol. 63: 16 pp. MICHENER, C. D. 1943. A review of the genus Calisto. Am. Mus. Novt. 1236: 1-6. MUNROE, E. G. 1950. The systematics of Calisto. (Lepidoptera, Satyrinae), with remarks on evolutionary and zoogeographic significance of the genus. J. New York Ent. Soc. 58: 211-240. MURPHY, D. D. & P. R. EHRLICH. 1984. On butterfly taxonomy. J. Res. Lepid. 23: 19-34. RILEY, N. D. 1975. A Field Guide to the Butterflies of the West Indies. New York Times Book Co., New York, 224 p. SCHWARTZ, A. 1983a. A new Hispaniolan Calisto (Satyridae). Bull. Allyn Mus. 80: 1-10. SCHWARTZ, A. 1983b. Haitian Butterflies. Mus. Nac. Hist. Nat., Santo Domingo, 69 p. SCHWARTZ, A. 1987, in press. The Butterflies of Hispaniola. Mus. Nac. Hist. Nat., Santo Domingo. Journal of Research on the Lepidoptera 25(2): 83-109, 1986(87) Records of Prolonged Diapause in Lepidoptera Jerry A. Powell Department of Entomological Sciences, University of California, Berkeley Abstract. Previously unpublished records of diapause and adult em¬ ergence one or more years beyond that of other individuals in the species are reported for 19 species of moths in 8 superfamilies. Records of prolonged diapause are summarized, representing 90 species in 10 superfamilies. Prodoxidae, Saturniidae, Pieridae, and Papilionidae predominate, but other taxa may be disproportionately underrepre¬ sented owing to lack of study. In Lepidoptera, extended diapause occurs in prepupal larvae or pupae and is most often observed in species that live in areas of seasonal drought and in cone- and seed-feeding species that depend upon crops of erratic abundance. We do not have convincing evidence for a genetically fixed polyphenic expression, wherein a small number of individuals carryover irrespec¬ tive of environmental conditions. Prolonged diapause is the maintenance of the dormant state in insects for one or more years beyond the period of emergence by most individuals in the population. There have been many records of the phenomenon in Lepidoptera, particularly in butterflies and Saturniidae, most often originating from pupae held indoors or in climates distant from the natural ones. In the past such records were regarded as aberrant, even astonishing occurrences that had no particular biological significance. Few researchers were sufficiently interested to carry out controlled experimental research on the relationships between the underlying genetic variability and en¬ vironmental factors that might demonstrate causes and possible adaptive values of prolonged diapause. In recent years, however, a number of reports suggest that in many insects multiannual delay of development is neither anomalous nor even exceptional and that it may have important adaptive significance (e.g., Danks, 1983; Hedlin et al., 1982; Nakamura & Ae, 1977; Shapiro, 1981; Sunose, 1978; Takahashi, 1977; Tauber et al., 1986). A selective advantage of facultative carryover seems to be especially true in cone- and seed-feeding species that depend upon hosts that produce seed crops of erratic abundance (e.g. Hedlin, 1967; Hedlin et al., 1982; Nesin, 1984; Sunose, 1978) and in desert insects, both phytophagous and predaceous (e.g. Ferris, 1919; Comstock & Dammers, 1939; Linsley & MacSwain, 1945, 1946; Nakamura & Ae, 1977; Powell, 1974, 1975, 1984b, present data). 84 J. Res. Lepid. Twelve years ago I summarized some examples of prolonged diapause in various insects (Powell, 1974), and that paper has been cited several times as though it was a review of the subject, but it is not. Recently two more comprehensive reviews have appeared (Sunose, 1983; Usha- tinskaya, 1984). Sunose reviewed my records as well as others and tabulated 64 insect species in which the dormancy has been reported to extend more than a year. Ushatinskaya, evidently unaware of the Sunose compilation, listed a similar number, many of which had not been noted by Sunose. These include eggs of grasshoppers, first instar larvae of parasitic Hymenopera and of tachinid flies that live within sawfly or moth larvae which undergo prolonged diapause, first or last instar larvae of gall gnats, mature larvae of bees, sawflies and meloid beetles, and adults of chrysomelid beetles. In Lepidoptera multiannual dormancy is known only in prepupal larvae or pupae, although in many species diapause occurs in eggs, first or second instar larvae, or adults. Sunose (1983) listed records of prolonged diapause in 20 species of Lepidoptera, and Ushatinskaya (1984) tabulated 14, of which 10 are additions to Sunose’s total. There are a great many more instances known. Probably any lepidopterist who has reared many Papilionidae or Saturniidae is familiar with carryover pupae and emergences of the adults in later years. I have assembled a list of records representing about 90 species, including those reported here (Table 1). These have been reported in more than 60 bibliographic references and several unpublished personal communications. Even excluding the yucca moths (Prodoxidae), which are restricted to North America and for which I have scores of delayed dormancy rearings, about 65% of the records are for Nearctic species. This implies that search of Old World literature has been cursory, and that the phenomenon is known in many more species than I have compiled. In fact, it would be impossible to collect a complete list of references to prolonged diapause because often its records are buried in life history studies, reports on insects of economic concern, or in taxonomic works. My purposes here are to record previously unpublished instances of delayed emergence in a diversity of moth taxa and to call attention to the likelihood that prolonged diapause is much more prevalent in Lepidoptera than previously supposed. For example, four of the occurr¬ ences listed below are species of Pyralidae, Geometridae and Noc- tuidae. These are families for which I have done only incidental rearing, and therefore one might expect records of extended dormancy to be commonplace in these taxa, yet I have seen few published. This suggests that diapause may be prolonged commonly in these large families, but students have not had sufficient patience to continue surveillance of pupae that do not develop in the first season and to test them in various artificial overwintering regimes. Diapause develop¬ ment is a dynamic process that takes place over weeks or months in North Temperate Zone insects, and the physiological responses to 25(2): 83-109, 1986(87) 85 Table 1. Taxonomic and geographical distribution of some Lepidoptera in which prolonged diapause is recorded MONOTRYSIA Prodoxidae DITRYSIA Tineidae Coleophoridae Gelechiidae Ethmiidae Tortricidae (Olethreutinae) Cochylidae Pyralidae Geometridae Lasiocampidae Saturniidae Sphingidae Notodontidae (including Thaumetopoeinae) Noctuidae (including Agaristinae) Pieridae Papilionidae No. of Species Nea retie Pa lea retie Other Duration (y-'s) 12 1 2 1 7 8 1 3 5 1 18 1 5 12 1 6 6 1 2 3 15 1 2 2-17 1 2 2-3 I1 1.5-2 1 1.5-4 22 2-3 1 I3 2-3 1 I4 2-6 I5 3 1.25-7 2 3 1.5-9 3 2 1 2-4 17 7 10 2-6 642 2-6 1 Pectinophora gossypiella (Saunders), diapause recorded in Egypt (Gough, 1916) and Hawaii (Busck, 1917). The species is believed to have originated from the Indo- Australian Region. 2 Includes Cydia pomonel/a (L.), which probably is introduced from the Palearctic (observed by several Nearctic workers and in Yugoslavia). 3 Loxostege frustralis Zeller (Pyraustinae), recorded in South Africa by Broodryk (1969). 4 Adults of the Australian species Arhodia (l)retractaria Wlk. (Ennominae) were reared after 21-23 months in diapause (McFarland, in litt.) 5 Family but no species mentioned by Danilevski (1951) in Russia. environmental changes are genetically variable. A stimulus that elicits successful development in one species, such as constant chilling for a certain period, may not be effective in another species or another population of the same species from a differing climatic zone or eleva¬ tion, or even among all individuals within a population. Typically, prolonged diapause involves some individuals that wait one or more full years beyond emergence of their sibs, in populations in which all individuals enter dormancy, for one of three life cycle pat¬ terns: a) vernal feeding followed by 9 or 10 months dormancy; b) vernal feeding followed by a few months aestivation and autumnal flight, as in Hemileuca (Comstock & Dammers, 1937, 1939; Ferguson, 1971), or c) facultatively double-brooded populations such as in Ethmia semilugens (Z.) (Powell 1974) and Ant hoc haris (dos Passos & Klots, 1969; Shapiro, 86 J. Res. Lepid. 1981), so that either a few weeks or nearly a full year in diapause elapses. I also include examples in Tineidae and Cochylidae in which individuals may wait one year even though sibs have emerged within a few days, apparently without undergoing any diapause. The potential for such species to wait more than one season seems likely. Rearing Methods Foliage-feeding larvae usually were held in transparent polyethylene bags lined with folded paper toweling to absorb moisture and provide a substrate for cocoon construction. If the host plant material was excep¬ tionally susceptible to excess moisture and decay problems, or the moth species were suspected to use soil for pupation, the lots were placed in translucent plastic boxes or one-gallon tubs with a few cm of sterile sand. Thus natural photoperiod normally was available. Prodoxids were housed in subdued light, in sealed cardboard boxes with a 32— mm emergence aperture at one end. During 1964—1970 most of the initial rearing was conducted in a temperature controlled lab (20— 25°C) with variable humidity (RH 38—48% in dry weather, 52—78% during rainy periods). Since 1971, the active larval lots have been handled in a mobile trailer lab on the University of California, Berkeley, campus. Here minimum temperature was controlled (usually 15-16°C) but not maximum, and humidity varied with outside air conditions. Temperature and relative humidity were recorded continuously by Bendix-Friese hygrothermographs placed on the lab shelving or in temperature cabinets with the collections or in a weather shelter located near outdoor cages. During the emergence season, moths were harvested daily or at 2— 3 day intervals. Prodoxids that failed to remain in emergence vials and died inside boxes were harvested at irregular intervals and at the end of each season. Rearing lot numbers. - A number-letter designation was assigned to each collection of one or more larvae. It reflects the year and month in which the collection was made (e.g., JAP 70C8 refers to the eighth lot recorded in March 1970). The number accompanies all associated material, including reared moths and parasitoids, preserved larvae and other artifacts such as pupal shells, and the data in notebooks. The habitat, hostplant, behavioral, emergence, and preservation data are summarized in a d-Base II program. Voucher specimens and associated data are deposited in the Essig Museum of Entomology, University of California, Berkeley. Overwintering regimes. - At the end of each season, usually in October or November, lots known or suspected to contain carryover larvae or pupae were exposed to one or a combination of two, storage methods used to manipulate winter temperature conditions: 1. Laboratory: A constant temperature (20° ± 1°C), low humidity (40—60% RH) room on the U.C. campus, was used for control sublots in 25(2): 83- 109, 1986(87) 87 studies of prodoxids. Other overwintering lots sometimes were left in the mobile trailer lab, which was unheated for 6 weeks in midwinter during 1976-1979. 2. Berkeley cage : Many collections were exposed to natural winter temperature and humidity in outdoor screen cages at the Oxford Tract, U.C. Berkeley. Cages were provided with a roof, but in windy storms the containers received direct moisture. Temperatures are moderate at this coastal station, and did not fall below 0°C during several winters monitored. Weekly means of daily maximum and minimum tempera¬ tures remained above 10°C during most of the winter. RH fluctuated daily and seasonally, generally between 50—80% in dry weather, 65—95% during storms. 3. Refrigerator. A kitchen refrigerator without precise temperature monitor (4° ± 1.5°C) was used for chilling during part of the winter in a few instances. 4. Russell insectary : An unheated, fully ventilated lab at the U.C. Russell Reserve near Lafayette, CA, was used to expose prepupal larvae to uncontrolled winter temperatures and humidity. The site is situated ca 10 airline km inland from San Francisco Bay, in the Briones Hills at ca 250 m elevation. Temperatures frequently dropped below freezing and weekly means of daily maxima and minima ranged ca +6°C to 11°C in mild winters, -4°C to +7°C in colder winters. These are much colder conditions than at Berkeley. For example, average monthly mean temperatures at Russell in 1971-73 ranged from 4.5°C lower in October to 8.5° and 7.4° lower in December and January than the 20 -year average at Berkeley. PRODOXIDAE Prolonged diapause is documented in most yucca moths (Koebele, 1894; Powell, 1984a, 1984b; Powell & Mackie, 1966; Riley, 1892). I have recorded emergences of adults following multiannual dormancy in the prepupal larvae of Parategeticula pollenifera Davis, and in nearly all the species of Prodoxus and Agavenema. Larvae of Tegeticula have been observed to survive more than one season; Riley (1892:117) noted that a large percentage fail to complete development in the first year, with some of the moths “not issuing until the second, third of fourth year,” but he did not give specific data or report conditions of over¬ wintering. I carried out extensive tests with 4 Prodoxus species associ¬ ated with Y ucca whipplei in California and Y. schottii in Arizona over a 20— year period. The larvae of these species commonly remain in diapause 4—8 years in artificial conditions even though neighbors in the same plant complete development in a prior year. Mass emergence of a whole colony may wait 6 years, if exposed to constant temperature, but mortality was significantly higher as compared to year IV (Powell, 1984a); and in one instance mass emergence occurred after 16 and 17 years in diapause (Powell, 1985, unpubl. data). 88 J. Res. Lepid. Diapause development in Prodoxus aenescens Riley and P. cinereus Riley is a complex and dynamic process, responding to gradually changing temperatures, probably coupled with moisture factors. Lar¬ vae held in constant temperature (± 20°C) and natural photoperiod throughout winter, or exposed to constant temperature chilling (0° to 9°C) for 50 days, remain in diapause, while refrigeration in constant darkness in gradually decreased (6 weeks), then gradually increased (7 weeks) temperatures at means of 3° to 10°C induced varying propor¬ tions of individuals to develop (Powell, unpubl. data). The following records are for species that have not been extensively studied and originate from localities distant from Berkeley, characte¬ rized by extremely different seasonal climates from those the larvae were exposed to in rearing. Prodoxus quinquepunctellus (Chambers) This species is widespread, from Arizona eastward, in association with an array of yuccas in the Sections Sarcocarpa and Chaenocarpa (Davis, 1967). Larvae were reported by Riley (1892) to sometimes remain in the dry floral stalks 2, 3, or 4 years, although apparently he did not observe successful development of carryover individuals. I obtained delayed emergences of P. quinquepunctellus from three collections taken in Arizona and New Mexico, the latter over 4-5 year periods. In contrast to California species of Prodoxus , some individuals developed even when held in constant temperature. The first material, consisting of stalks thought to be two species of Yucca , possibly intermedia and g/az/ca, was collected in late September, 1963, 5 km W of Albuquerque, Bernalillo Co. by J. A. Chemsak (JAP 63J1-J2). A sample of 24 larvae was removed for preservation. The remainder were held in constant temperature through the following two seasons, and diapause development occurred in 7 individuals, one in 1964, 6 in 1965. In November, 1965, half the stalks were transferred to the Russell insectary, where winter III elicited emergence of 5 P. quinquepunctellus in 1966. During the same season, the remaining stalks in the lab produced 6 moths; one more emerged from them in 1967. Thus, development of one or more individuals took place each year in the lab, 73% of those that emerged (fig. 1). Moths eclosed in April, 1964, and March to early May in 1965, approximately coincident with the flight period in the Albuquerque area (Davis, 1967). A second New Mexico collection was made near Portales, Roosevelt Co., in late October, 1973, by N. M. Jorgenson (JAP 73K1), and consisted of current year stalks of Yucca glauca. These were held in my lab until December 1, then at the Russell insectary over winter, and 40 P. quinquepunctellus responded in diapause development in 1974. In midwinter, 1974-75, a sample of 12 carryover larvae was removed for preservation, and the rest of the lot was moved to the outdoor cage at 25(2): 83-109, 1986(87) 89 63JI- J2 20 - L L R R R L L L/R R/B B B B B 70C8 , Cl I '69 *70 ‘71 ‘72 ‘73 *74 ‘75 *76 '77 '78 FFLRRRR/BBBB '70 '71 '72 '73 F L R R Fig. 1 (upper): Successive annual numbers of Prodoxus quinquepunctellus (Chambers) that emerged from two collections of yucca inflorescence stalks (see text for data). (lower): Successive annual numbers of Prodoxus co/oradensis Riley that emerged from four collections of Yucca schidigera inflorescence stalks from the Mojave Desert (see text for data). Overwintering sites: F, in field; B, Berkeley, outdoor cage; R, Russell Reserve, unheated insectary; L, laboratory at 20 ± 2°C. 90 J. Res. Lepid. Berkeley, where it was stored for 5 years. Three additional moths emerged, 2 in 1976 and one in 1978, after 3 and 5 years in diapause (fig. 1). One additional stalk was collected from Yucca angustissima , 1 km W of Cottonwood, Yavapai Co., AZ, 30 July 1970, by R. E. Dietz and P. A. Rude (JAP 70G35). It was retained in the lab for one year, during which no moths emerged, then transferred to the Russell insectary. There 26 P. quinquepunctellus successfully completed development, 30 May to 20 June 1972, following the second winter. Only one flaccid- appearing larva was discovered by splitting the stalk at the end of 1972. Prodoxus coloradensis Riley This species feeds in stalks of Yucca schidigera and Y. baccata in California and in other yuccas of the Section Sarcocarpa in the western U.S. (Davis, 1967). Collections made in the Mojave Desert, 31 March — 2 April 1970 (Dietz & Powell), indicated that prolonged diapause in natural populations is commonplace in that habitat. At a site 8.5 km north of Cottonwood Springs, Joshua Tree National Monument, Riverside Co., CA, Y. schidigera was in full to late bloom on March 31, and none was seen with newly emerging inflorescences. P. coloradensis adults were numerous, yet there appeared to be no 1969 stalks. Dry stalks in the vicinity appeared to originate from 1968; each had emergence holes along with few to many carryover larvae (JAP 70C7-8). Because adults ofP. coloradensis had already emerged, it could not be inferred with certainty that the observed stalks were older than one year. It was evident, however, that either a substantial portion of larvae had carried over from 1968 or a previous season, or that many 1969 larvae had not completed diapause development for the 1970 season. At Belle Campground, 32 km to the northwest and 300 m higher than Cottonwood Springs, Y. schidigera showed no signs of inflorescence development. Flowering and yucca moths had been observed at this site in mid- April, 1963, and it was evident that activity would not begin before mid to late April in 1970. Again, there appeared to be no 1969 stalks, while those from a previous year contained old emergence holes along with carryover Prodoxus larvae (JAP 70C11), which tended to confirm our estimate of stalk age at Cottonwood Springs. On the basis of weathered appearance of the stalks, subsequent vegetative growth, and lack of prior emergences, Y. schidigera stalks at Ryan Mountain, Joshua Tree Natl. Mon. (JAP 70C14) and Cedar Canyon, 27 airline km NE of Kelso, San Bernardino Co., CA, (JAP 70D1), also were judged to be more than one year old. The 1970 collections were housed at the Russell insectary beginning in early April, and no P. coloradensis emerged in 1970, nor in 1971 25(2): 83-109, 1986(87) 91 after transfer to the constant temperature lab over winter 1970—71. Most moths (82%, n=78) completed development in 1972, following their first overwintering at the Russell insectary. Additional larvae were discovered in 70C11 in February, 1974; the lot was retained and exposed to winters in the cage at Berkeley. Only 2 adults completed development in the succeeding 2 years, then, inexplicably, 11 P. col- oradensis (38% of the Belle Campground total, n=29) emerged in April 1977, following a second full winter of drought conditions in Berkeley, 9 years after their larval feeding in 1968 stalks (fig. 1). It is possible that the 70C14 and 70D1 collections, which originated from higher or more northern sites than Belle Campgound, were made sufficiently early (e.g., 4 or 5 weeks ahead of flowering) that develop¬ ment was interrupted; but negative results of all other 1970 collections including those of P. sordidus Riley (see below) and P. y-inversus Riley (Powell, 1985) suggested that the 1969—70 winter was one that failed to elicit diapause development in Mojave Prodoxus generally. Evidence of 1969 failure in the Y. schidigera inflorescence crop was dramatic at a site on Black Canyon Road, south of Cedar Canyon. Here an extensive stand of tree-like Y. schidigera possessed 1969 stalks that were dry, hard, and black, atrophied before they were fully grown, as though they were killed by a late freeze. None had any signs of lepidopterous larval feeding. A number of 1968 stalks were found containing small numbers of larvae; some had emergence holes left by sibs that must have faced catastrophe in 1969. A collection (JAP 70D4) of the pre-1969 stalks produced no adults in 1970 and only one P. coloradensis in 1972 after overwintering at Russell. Dissections of the stalks in early 1973, however, revealed a few larvae still in diapause. Prodoxus sordidus Riley Moths treated under this name are believed to represent two species (D. S. Frack, in litt.) associated with Joshua Tree, Yucca brevifolia. It appears that a species with tan forewings feeds in the inflorescence scapes, while a whitish moth originates from larvae in pods. However, some of my stalk-inhabiting larvae produced the pale morph, and more information is needed to confirm separation of larval feeding niches of the two. My rearing data are pooled. Collections at Antelope Valley, 33 km E Gorman, Los Angeles Co., CA, March 11, 1963 (Chemsak & Powell, JAP 63C8) just ahead of the current season flowering, confirmed that prolonged diapause occurs — most emergences took place in late March and early April, 1963, but 2 P. sordidus completed development in February, 1964 after storage in the lab. A midwinter collection from 5 km SE Pinon Hills, L. A. Co., 23 December 1969 by P. A. Opler (JAP 70A1) produced only 6 adults after housing in the lab for the remainder of the season. None emerged in 1971 after winter II in constant temperature, but in 1972, 90 moths 92 J. Res. Lepid. responded to winter III at the Russell insect ary. Three more individuals waited until 1973, following a second winter at Russell, and one P. sordidus emerged in the 6th season, after the material was transferred to Berkeley in midwinter 1973-74. The same emergence pattern was shown by population samples made 31 March - 2 April 1970 in the central Mojave Desert, despite the fact that they were collected after a full winter in the field. One of these (JAP 70C12) at Belle Campground, Joshua Tree Natl. Mon., Riverside Co., CA, was made while Y. brevifolia was in full bloom and Prodoxus was active in low numbers. It was not possible to judge stalk age precisely, but it appeared at least two seasons growth contained larvae in diapause; none of these matured in 1970. Three more collections were made at higher, more northern sites, ahead of 1970 Y. brevifolia flowering (70D2: Cedar Canyon, 27 airline km NE Kelso, San Bernar¬ dino Co., CA; 70D10: 6 km S of Barnwell, 44 airline km NE Kelso; 70D24: Kyle Canyon, 15 km W Highway 95, 30 airline km WNW Las Vegas, Clark Co., NV). In each sample, no P. sordidus matured in 1970, nor in 1971 following a year in the constant temperature lab. Most of the emergence (84%, n=140) occurred in year III following overwintering at the Russell insectary in 1971-72, while 6 individuals eclosed in each of the next two years after another winter at Russell and at Berkeley from midwinter 1973-74. Only one of the collections (70D10) was retained beyond the 5th year, and 10 P. sordidus de¬ veloped successfully in 1975 after storage over winter in the outdoor cage at Berkeley, but none matured in the 7th season (fig. 2). Prolonged diapause would seem less critical to continuity of popula¬ tions in P. sordidus than in other Prodoxus because Yucca brevifolia blooms more consistently. Nonetheless, combined with observations of P. coloradensis , the data indicate that certain winters are suboptimal to diapause termination in Prodoxus in widespread areas of the Mojave Desert. Weather records from Twenty nine Palms (15 airline km N of Belle Campground, at 610 m) and Mountain Pass (35 air km NNW of Cedar Canyon, at 1450 m) for the preceding 4 and 3 years, respectively, showed that the 1969-70 winter was unusual, if not exceptional in a long-term sense. At both stations the rainfall total for October through March was lower than in any of the preceding few years, 19 mm contrasted to a 5— year average of 45 mm at Twentynine Palms, and 50 mm vs. a 4— year average of 92 mm at Mountain Pass. Possibly of greater significance to the moths, the 1969-70 winter was characte¬ rized by unusually mild temperatures. In the 13— week midwinter period, December through February, monthly means averaged 11.1°C at Twentynine Palms in 1969—70, contrasted to a range of 8.5— 10.9° (avg. 9.9°) during the preceding 4 winters; while at the higher station monthly means averaged 7.0°C in 1969—70, but ranged 3.1 to 6.7° and averaged only 4.8° during the preceding 3 seasons. 25(2): 83-109, 1986(87) 93 30 70CI2 70D2 20 10 ° '70 '71 '72 ‘73 '74 80 60 40 20 F L R R R/B 70DI0 -i 70 7 1 72 73 74 F L R R R/B 70D24 I 7o 7 1 72 73 74 75 76 F L R R R/B B B 70 7 1 72 73 74 F L R R R/B Fig. 2: Successive annual numbers of Prodoxus sordidus Riley that emerged from four collections of Yucca brevifolia inflorescence stalks from the Mojave Desert (see text for data). Overwintering site designations explained under figure 1. 94 J. Res. Lepid. Agavenema barberella (Busck) Agavenema species feed on Agave and display larval habits similar to those of stalk-inhabiting Prodoxus. The adults are not found in the flowers of the larval host, as are Prodoxus , and larvae usually occur in the main scape well below the inflorescence. Collections were made from various species of Agave in Arizona in 1968—1970. Stalks were housed at the Russell insectary in 1968—69 and 1969—70 winters, in the lab 1970—71, returned to Russell in 1971 — 72 and not retained beyond the fourth season. Development occurred irrespective of the kind of artificial winter conditions. Lots collected at Madera Canyon, Santa Rita Mts. in June 1968, prior to the current season flight, from Agave palmeri (JAP 68F43-44) (Opler & Powell), produced moths primarily the same season (83% of those reared, n=130). Emergences occurred from 15 July to 13 Octo¬ ber, and 20 larvae were harvested in late September. Only 7 indi¬ viduals completed development in 1969, after exposure to winter II at the Russell insectary, followed by none in 1970 and 1971, yet 15 A. barberella emerged in the 4th season after storage at Russell. Larvae were collected in Agave schottii at Molino Basin, Santa Catali¬ na Mts., in September, 1969, after the summer activity period (JAP 69J10). Small numbers of moths issued, 5 in 1970, 3 in 1971 (following a lab winter), and 7 in 1972, after winter III at Russell (47% of the total, n=15). Lastly, 7 collections of one-year old or current season stalks were made from 3 Agave species in central and northern Arizona,1 29 July to 2 August, 1970. All yielded similar results: one or no moths in 1970, modest numbers in 1971 (19% of the total reared, n=134), despite having overwintered in the lab, followed by most of the emergence (78%) in the 3rd season after exposure to winter at Russell in 1971-72. The normal flight period of A. barberella is poorly known, with scattered records from March to September in southern Arizona (Davis, 1967). The flowering phenology of most agaves probably en¬ ables a more protracted activity season by Agavenema within popula¬ tions than is characteristic of yucca moths. Agavenema pallida Davis This species, which is closely related to the preceding one and may be a geographical race, occurs in the deserts of California and Baja California Norte, Mexico (Davis, 1967; UCB specimens). Its seasonality is more restricted than that of A. barberella , similar to the California *JAP 70G30: 4 km E of Peach Springs, Mohave Co., 1969 stalks of Agave utahensis ; 70G31: same data, 1970 stalks; 70G32: 5 km SW Mingus Mtn. summit, Yavapai Co., 1969 stalks of A. parryi\ 70G33: same date, 1970 stalks; 70G36: Molino Basin, Sta. Catalina Mts., Pima Co., 1969 stalks A. schottii ; 70H6: same data, 1970 stalks; 70H5: 1.5 km NE of Colossal Caves, Pima Co., 1970 stalks A. schottii (R. E. Dietz, J. Powell and P. A. Rude). 25(2): 83-109, 1986(87) 95 species of Prodoxus. Collections from Agave deserti in March, 8 km E Jacumba, San Diego Co., CA (JAP 63C31) and April, 1963, Pinyon Flat, 29 airline km SE of Idyllwild, Riverside Co., CA (JAP 63D10) revealed prolonged diapause. Carryover larvae remained at the end of the season, and in 63D10 a few adults emerged in 1964 and 1965 despite housing in the lab during the preceding winters. Two collections made in Baja California in March, 1972 and 1973, yielded A. pallida adults in diminishing numbers in years I, II, and III. A total of 25 moths emerged in year I following overwintering in the field (51%, n=49), 17 in year II (35%), and 7 in year III (14%), and 2 carryover larvae were discovered in winter IV (JAP 72C12: 8 km E of El Rosario, 1971 stalk of Agave shawii; 73C5: 5 km S of Rancho Santa Ynez, 1972 stalks of A. cerulata spp. lnelsoni\ Doyen & Powell). TINE ID AE: SCARDIINAE “Scardia” cryptophori (Clarke) This large gray tineid is widespread in montane western North America. It was transferred from Morophaga H.-S. to the genus Scar¬ dia Tr. by Davis (1983), who treated Morophaga and its Palearctic type species as a synonym of Scardia . However, S. cryptophori is structural¬ ly and biologically dissimilar from Nearctic members of Scardia. In contrast to other Scardiinae and other fungus-feeding Tineidae, this species is host-specific, feeding in the sporophores of Polyporus ( Cryp - tophorus) volvatus, which grows on recently dead conifers (Lawrence & Powell, 1969). This fungus is available throughout the season but appears to be in a fresh state preferred by S. cryptophori primarily in spring following winter precipitation and snow melt. Among numerous collections from the Sierra Nevada and Cascade Ranges in California, two from Trinity County produced some larvae that proceeded to maturity without dormancy, as most do, and others that entered diapause for one year. This suggests that adverse condi¬ tions, particularly desiccation, may prolong diapause. Four collections of sporophores were made in the vicinity of Hayfork, Trinity Co. in late May, 1973. Moths (n=36) emerged from all 4 lots 15—25 June, but in two groups (JAP 73E46, E47) a number of larvae spun cocoons in dry spots remote from the fungus, and most of these did not emerge. A few cocoons were cut open in September, revealing carryover prepupal larvae. The lots were stored in boxes at the Russell insectary. After removal to Berkeley in early June, 1974, 10 M. cryptophori emerged in late June, one year later than their sibs. ETHMIIDAE Ethmia plagiobothrae Powell This species flies in early spring, primarily in March, in the foothills of the Coast Ranges and Sierra Nevada in California. The larvae feed on 96 J. Res. Lepid. flowers of Plagiobothrys (Boraginaceae) a spring annual, and pupae remain in diapause from the beginning of the dry season until early spring the following season. I have made numerous collections of larvae, but they are highly susceptible to disease in rearing, and the few adults obtained emerged after one year (Powell, 1971). In one instance (JAP 69D58, Powell, 1971) 10 E. plagiobothrae were reared after one winter (100% of the emergence) even though the collection data and rearing conditions were essentially the same as for a collection of the closely related species, E. scylla Powell, that produced moths 2, 3 and 4 years after pupation (Powell, 1974). In another collection (13 km SE Three Rivers, Tulare Co., 4 May 1979, JAP 79E24) all 4 adults emerged the next year, between Feb. 14—27. On two occasions, howev¬ er, pupae were still healthy appearing during the second winter, after 19 and 21 months, but development did not take place (Powell, 1971). Thus it was not surprising that a collection in 1980 produced one E. plagiobothrae that delayed until the second spring before emerging. We found larvae on Plagiobothrys nothofulvus, 6 km S of Rough and Ready, Nevada Co., CA, on 4 May 1980 (M. Buegler, J. DeBenedictis & Powell) (JAP 80E17). About a dozen were placed in pill boxes, 2 to a container with inflorescences and folded tissue paper. After cocoon construction the boxes were left open inside a translucent plastic box, stored in the mobile trailer lab at Berkeley through the summer. In October they were moved to the outdoor cage, but no development resulted after the first winter. In 1981—82 the material was again exposed to winter in the cage, and one male emerged between 24 February and 24 March, 1982. Other individuals were unsuccessful in pupation, succumbing either to disease or desiccation prior to or after cocoon construction. Ethmia epileuca Powell This species was described from the Panamint Range, CA, in the northern Mojave in 1959 and subsequently has been recorded in the low deserts of southern California, Baja California, and Arizona (Powell, 1973; UCB specimens). The foodplant, an annual Phacelia , was discovered in 1977, and the one specimen reared spent two years in diapause. A few Ethmia larvae were found feeding externally on Phacelia crenulata (Hydrophyllaceae) at Zzyzx Springs, near Baker, San Ber¬ nardino Co., CA, 20 April 1977 (JAP 77E94). They were provided with soft paper toweling, but only one E. epileuca successfully constructed its cocoon. It was stored in a transparent plastic bag in the mobile trailer lab through summer and fall, then in an outdoor cage at Berkeley during late winter and spring, 1978, but diapause develop¬ ment did not occur. The cocoon was again placed in the cage for the 1978—79 winter until mid-February, then in the mobile lab. A female 25(2): 83-109, 1986(87) 97 E. epileuca emerged on 2 April 1979 after nearly 24 months in di¬ apause. The 1978-79 winter was appreciably colder than the preced¬ ing year at Berkeley, 190 heating day degrees greater (based on 18.3°C), during October through February. Interestingly, a related species, E. semilugens (Z.), which has the capability of developing with a short diapause of a few weeks, of one year, or of several years (Powell, 1974), also feeds on Phacelia crenulata (Powell, 1971). Collection records for E. epileuca , however, indicate that this species has a univoltine cycle with flight in early spring. Ethmia (Macelhosiella Group) n. spp. A, B Larvae that proved to be two undescribed species which are structur¬ ally and biologically similar to Ethmia geranella (B. & Bsk.) were discovered in western Fresno County, CA in March 1975 and collected again in April 1978. Both species feed in spring, estivate as pupae, and fly in late fall, as do other members of the Macelhosiella Group. Eggs hibernate, presumably in diapause or a temperature-dependent quiesc¬ ence (Powell, 1973). In both Fresno County collections, some indi¬ viduals carried over to a second or subsequent autumn; however the pattern was quite different between the two species, even though they were reared and held over winter in identical conditions. Species A: The larva is green with faint longitudinal, gray in- tegumental shading. The adult is a whitish moth with faint ochreous along the discal cell, and it differs from other western species of the group by lacking a hind wing costal penicillus in the male. P. A. Rude and I collected larvae on Phacelia tanacetifolia (Hydrophyllaceae) at the Ciervo Hills, 29 airline km SE of Mendota, Fresno Co., CA, 16—18 March 1975 (JAP 75C1) and at Jocalitos Cyn., 9 airline km S of Coalinga, Fresno Co., 17 March 1975 (JAP 75C2). Larvae were placed in translucent plastic vials and cardboard pill boxes, 1 or 2 per contain¬ er, with folded tissue and small blocks of yucca scape pith, into which they burrowed for cocoon construction. Others were confined in trans¬ lucent plastic bags with paper toweling and hostplant material. They were stored in the mobile trailer lab at Berkeley for the remainder of spring and early summer; in late July about half the containers were moved to the Russell insectary, while the remainder were placed in the outdoor cage at Berkeley in September. Most of the moths (n=25) emerged in October, 1975. The carryover cocoons were exposed to winter conditions at Russell during January — April, 1976, and subsequently were retained in the outdoor cage at Berkeley. Two adults completed development in October, 1976, and 13 others developed but were trapped in the cocoons and unable to distend the wings, probably in the first season. No more emerged in 1977 or 1978. In early April, 1978, following the 1975 — 77 drought, Jocalitos Canyon 98 J. Res. Lepid. was again a sea of wildflowers, and we made additional collections of the two Ethmia. Larvae of Species A (78D12) became diseased and most were preserved for taxonomic study; 7 were retained in translu¬ cent plastic boxes with paper toweling and yucca blocks in the mobile trailer lab. Four of these emerged as adults 25-30 October 1978. The remaining 3 carried over and eclosed 10-27 Oct. 1979, after storage over winter in the outdoor cage at Berkeley. Thus 29 of 34 successful emergences (85%) of Species A occurred during the first fall after larval feeding, while the remainder developed one year later. Ethmia Sp. B: The larva is irregularly mottled, predominately gray, with longitudinal bands of orange dorsally and laterally, resembling that of E. charybdis Powell (Powell, 1971). The adult is dark gray; the forewing has a trace of ochreous and a variable black line along the discal crease and a whitish dot at the end of the discal cell, and the male hind wing has a costal penicillus. Thus the adult is structurally quite similar to that of E. timberlakei Powell, which feeds on Phacelia ramosissima in southern California, but the larva of the latter is green with a yellow dorsal stripe (Powell 1971). Larvae of Species B were mixed with those of Species A in the field at Jocalitos Canyon in March, 1975, but were outnumbered ca. 10:1. I could not detect a spatial differentiation on the plants; both fed within the scorpioid floral spikes. In the lab, larvae were segregated by color and handled in the same rearing conditions as outlined for Species A. In marked contrast to the preceding species, no individuals of Species B completed development in 1975. Instead, all 4 moths that meta¬ morphosed did so after carrying over, two in early November, 1976, and one each in November 1978 and 1979. The same diapause behavior obtained in the April 1978 collection at Jocalitos Canyon: about a dozen larvae were confined, of which none developed in 1978, 2 moths emerged in October 1979, none the following year, and 2 more in October, 1981, after 42 months in diapause. I visited the Jocalitos Canyon site on 21 March 1977, following two drought years in California, and found the spring vegetation dry; there had been essentially no germination of annuals. Hence, a large larval eclosion of Phacelia- feeding Ethmia would have been mostly doomed, and if prolonged dispause in the pupal stage acts as a buffer against such disasters, the strategy must be keyed to maintenance of diapause through autumn preceding a dry spring. Adults of Species B were taken at blacklights at Jocalitos Canyon on 10 November 1977 (Powell & Rude). There had been no heavy rains in the region that fall, only ca 6 mm in the Fresno area, less than 1/4 the normal, by early November; the drought did not end until late Novem¬ ber and December, 1977. This leaves unanswered the question of how prepupal larvae appraise the potential for spring growth of annual Phacelia and either maintain diapause or undergo development in October. 25(2): 83-109, 1986(87) 99 TORTRICIDAE : OLETHREUTINAE Grapholita vitrana (Walsingham) Grapholita vitrana was originally described from northern Oregon, and it is commonly associated with Astragalus (Fabaceae) in sandy situations along coastal areas of California, on San Miguel and Santa Catalina Islands, and in interior and coastal areas of Baja California, Mexico, including Isla Cedros (SDNHM, UCB Specimens). The larvae feed on green seed in the bladder-like pods of locoweed. A collection of pods containing larvae of G. vitrana and Everes amyn- tula (Bdv.) (Lycaenidae) was made from an Astragalus growing on riverine dunes along the Salinas River near King City, Monterey Co. on May 3, 1974 (JAP 74E20). The butterflies emerged within a month, while the tortricids spun tough, papyrus-like cocoons in which prepupal larvae entered diapause. After drying, the lot was stored in the mobile trailer lab on the Berkeley campus. Seven G. vitrana emerged between 28 March and 28 April, 1975. The remaining cocoons were left undis¬ turbed and in the same overwintering conditions except without heat control for 5 weeks in December and January, 1975-76, so were exposed to temperatures of 2° to 4°C on several occasions. Diapause development occurred in 4 individuals (2d, 29), and moths eclosed between 23 March and 11 May 1976. Similar housing of remaining carryover larvae during the third winter, 1976—77, resulted in 5 more G, vitrana (2d, 39) emerging, between 26 February and 17 April, 1977, after 33-35 months in diapause. Diapause extending to a second or third year has been recorded in several other seed-feeding species of Grapholita and the closely related genus Cydia (e.g., Dickson, 1949; Hedlin, 1967; Nesin, 1984; Tripp, 1954). COCHYLIDAE Cochylis yuccatana (Busck) This distinctive species is widespread in the southwestern deserts, from Texas (TL: Nuecestown) to southern California and Baja Califor¬ nia Norte, Mexico (UCB Specimens). The type series was reared from Yucca haccata\ we have found the larvae in flowers of Yucca brevifolia in the Mojave Desert, and Agave shawii in coastal Baja California Norte. There are no records of multiannual diapause, but the species is capable of either completing development without delay or remaining dormant until the following year, which suggests a potential for pro¬ longed diapause exceeding one year. A single larva taken 5 km W of Palmdale, Los Angeles Co., CA, from Yucca brevifolia in late March, 1968, pupated and produced an adult 15 days after collection (JAP 68C62). Several caterpillars feeding on Agave near San Telmo, Baja Calif., in mid-March, 1972, did not mature that spring, but two moths eclosed in late April, 1973, after 13 months in diapause (JAP 72C6); while another collection from the same host 100 J. Res. Lepid. near El Rosario, Baja Calif., in late March, 1973, produced larvae of both types. Five C. yuccatana emerged between 24 April and 8 June, 1973, and one later that summer, and one prepupal larva carried over to complete development in May or June, 1974 (JAP 73C4). Both carryover lots were housed at the Russell insectary. PYRALIDAE: CHRYSAUGINAE Satole ligniperdalis Dyar This curious polymorphic and sexually dimorphic species is wide¬ spread in the southwestern Nearctic, from western Texas to the west¬ ern edge of the deserts in California (TL = Portal, AZ). Adults have been reared from seed pods of Chilopsis linearis (Bignoniaceae) in southern Arizona and southern California (USNM specimens). A collection of the linear fruits of Chilopsis was made by J. T. Doyen on the Kelso Dunes, ca 12 airline km SW Kelso, San Bernardino Co., CA, on July 14, 1974 (JAP 74G9). Ten adults of S. ligniperdalis emerged between 18—27 July, and two more later that fall. Although several larvae abandoned the pods by late July and the plant material became badly covered by sooty mould, an apparently healthy larva was revealed by dissecting its cocoon in February, 1975. The lot was therefore placed in a cardboard box and stored in the outdoor cage at Berkeley. Two females emerged in early September, 1975, one on June 25, 1976, and a final one in the fourth season, between October and December, 1977, after 38—40 months in diapause. Evidently larvae were fully fed at the time of collection and spent diapause as prepupal larvae in cocoons. Because they were held in subdued light and the emergence dates varied by more than two months after late June, it appears photoperiod was not a primary factor influencing diapause development. PYRALIDAE: CRAMBINAE Loxocrambus sp. near mojaviellus Forbes A somewhat heterogeneous assemblage of specimens from the low deserts of California has been designated as a new species with a manuscript name by A. B. Klots (AMNH). We accumulated large series of this species in a survey of active dune systems of the Colorado and Mojave deserts (Powell, 1978). Included was one specimen that re¬ mained unfed, presumably as a prepupal larva in diapause, for 28 months. The larva was sifted from active dunes south of Rice, Riverside Co., CA, Jan. 30, 1977, by J. T. Doyen and P. A. Rude (JAP 77A19). In early February I opened its sand-covered silken tube to find a large pyralid larva. The larva evidently was fully fed and did not accept plants in the lab. It was retained in a dry container with sand in the mobile trailer 25(2): 83-109, 1986(87) 101 lab (unheated 6 weeks in midwinter) until February, 1978, then in the outdoor cage at Berkeley for one year. Rainfall resulted in wetting the sand once or twice during the 1978—79 winter. After storage in the mobile trailer lab again for four months, a somewhat dwarfed male emerged 1 June, 1979. GEOMETRIDAE Eupithecia dichroma - johnstoni group A single specimen of Eupithecia was reared after spending two win¬ ters in diapause, and although the species identification is question¬ able, the record is noteworthy because prolonged diapause has been recorded for few geometrids. The moth failed to expand its wings fully upon emergence, but it seems to match the description of E. johnstoni McDunnough with only minor differences. The latter species, which was known only from the type male from Inyo County, CA, at the time of McDunnough’s revision in 1949, has the whitish ground color on the forewings more contrasting with the red-brown subbasal and subter¬ minal bands than seems to be true of the deformed reared specimen from Modoc County, CA. The male genitalia, particularly the unique aedeagus, confirm a close association with johnstoni and E. rindgei McDunnough, a paler species described from Plumas County, CA. The cornutus is lightly sclerotized basally in the Modoc example, a feature not shown in McDunnough’s (1949) illustrations, but otherwise the three species are quite similar in this character. The geometrid caterpillar was collected incidentally along with larvae of Pyramidobela quinquecristata Braun (Ethmidae) and a polyphagous tortricid, Sparganothis tunicana (Walsingham), which were webbing and feeding on inflorescences of Penstemon laetus spp. roezlii (Scrophu- lariaceae) at Rock Creek, 15 km NE of Adin, Modoc Co., 12 June 1974 (JAP 74F21). Several adults of the Pyramidobela and Sparganothis were reared within a month; neither of these species diapauses as a prepupal larva or pupa. In late July I discovered a geometrid pupa in the material and placed it in a plastic vial with tissue paper. This was retained in the mobile trailer lab at Berkeley until mid-December, then transferred to the insectary at Russell for 60 days. Examination indi¬ cated the pupa to be still alive in August, 1975, and it was again moved to the Russell insectary for overwintering. It was transferred back to Berkeley on 14 March 1976, and the moth emerged 14 days later, after ca 21 months in diapause. SATURNIIDAE Hemileuca electra (Wright) This fall-flying, diurnal species occurs in southern California, where the larvae feed on Eriogonum fasciculatum (Polygonaceae) in spring. 102 J. Res. Lepid. Summer is passed by pupae in diapause. As noted by Comstock & Dammers (1939), in captivity the larvae are susceptible to disease and are difficult to rear. I collected seven mature larvae at Mission Gorge, San Diego Co., CA, [ca 8 km W of Santee] on E. fasciculatum , 19 March 1950. None successfully developed that season, but one pupa held indoors carried over, and a female eclosed 27 November 1951, after 20 months di¬ apause. The capability of emerging following one summer of dormancy or delaying until the succeeding or a later autumn may be characteris¬ tic for all species of Hemileuca. It has been recorded for H. maia (Drury) (Ferguson 1971; W. D. Winter in litt.), H. burnsi Watson (Comstock and Dammers, 1937), H.juno Packard (Comstock & Dammers, 1939), and H. eglanterina (Boisduval) (Winter, in litt.) and inferred for others by Tuskes (1985). Saturnia mendocino Behrens This diurnal saturniid occurs in the North Coast Ranges and Sierra Nevada of California (Ferguson, 1972), south at least to El Dorado Co. (UCB specimen). The larvae have been recorded by several authors to feed on Arbutus and Arctostaphylos (Ericaceae); Ferguson suggests that S. mendocino also feeds on shrubs of other families, apparently based on circumstantial associations for the closely related species, S. walterorum Sala & Hogue, in southern California. David Wagner collected an ovipositing female and egg cluster of S. mendocino on Arctostaphylos pungens var. montana near Alpine Lake, Marin Co., CA, during a field trip with our immature insects class, on April 13, 1979. The female continued to produce eggs for several days; larvae were reared, May 4 to July 3, 1979, on Arctostaphylos from the U. C. campus (DLW lot L10— 14— 79). The cocoons were held in an outdoor cage in a plastic bag with damp moss during the 1979-80 winter. Four adults emerged in May, 1980. The remaining pupae were left in a drawer at room temperatures, yet produced two more moths in 1981, none the following year, and one S. mendocino finally emerged in May, 1983, after nearly 4 years in diapause. Third and fourth year emergence also is recorded in the Palearctic species, Saturnia pyri (Schiffermiiller), in Maryland (Bryant, 1980). NOTODONTIDAE Pheosia rimosa Packard I discovered two larvae of this widespread species at Rock Creek, ca 2 km SW of Tom’s Place, Mono Co., CA, 26 August 1983, on Populus trichocarpa (Salicaceae) (JAP 83H122). The caterpillar, which I mis¬ took for Sphingidae owing to the short caudal horn, was described by Dyar (1891) and others and illustrated in Packard’s monograph, but its remarkable crypsis seems not to have been mentioned. The larvae are 25(2): 83-109, 1986(87) 103 peculiar compared to most Notodontidae, being naked, gray, with a greasy or pearly sheen, prominent spiracles and exaggerated interseg- mental constrictions. They perch during the day, hanging downward, on the stems of poplar, back of the distal leaves. There they resemble the older stems, which develop rings of enlarged nodal growth that are matched exactly in color by the larvae. The collection was temporarily housed in a plastic bag and trans¬ ported in a field ice box during a trip, and one of the larvae pupated loose in the bag by August 31. The larva, pupa and foliage were transferred to a plastic box with sandy soil September 1, but the lot was allowed to become moldy while stored in the mobile trailer lab at Berkeley, and the second larva died. In February, 1984, the material was still damp and the loose pupa on the soil surface was noted to have a thin bloom of mold on its surface and was presumed dead. It was placed in a refrigerator (± 4°C) for 5 weeks but did not metamorphose in that season. The pupa was left in situ on the soil surface and was refrigerated during the 1984-85 winter, from October until February. A large, normally developed female of P. rimosa , which is of the pale morph characteristic of populations east of the Sierra Nevada in California, emerged during 16—23 March 1985, after 18 months in diapause. That is, nearly one year later than presumably would be normal for this bivoltine species. NOCTUIDAE Egira crucialis (Grote) There appear to be few records of prolonged diapause in Noctuidae, although many overwinter as pupae. Thus carryover records of Egira crucialis , for which we do not have accurate emergence dates, seem worthy of recording, to call attention to the potential for extended dormancy in noctuids. Species of Egira Duponchel ( =Xylomyges Gn. and Xylomania Hamp.) are univoltine and fly in early spring, often at quite cold temperatures. In central California they are active from late December to May, the particular flight period varying with the species and elevation. Larvae feed during spring foliation, and pupae in diapause estivate and hiber¬ nate until midwinter or spring. We have reared only a few of them, but two E. crucialis remained in diapause beyond the normal spring flight period. Evidently Egira crucialis is a general feeder; we found young larvae on new foliage of Pseudotsuga menziesii (Pinaceae), while previous records are from hardwoods. Crumb (1956) listed collections from Alnus (Betulaceae) and Quercus (Fagaceae) in Washington State, and Prentice (1962) recorded those plant genera as well as Arbutus (Eri¬ caceae) and Salix (Salicaceae) from Vancouver Island, British Col- 104 J. Res. Lepid. umbia. Our material was taken ca 2 km west of Angwin, Napa Co., CA, 15 May and 1 June, 1979 (DeBenedictis & Powell — JAP 79E62, 79E73). Larvae were reared in polyethylene bags, then transferred to translucent plastic boxes with sterile soil after ca 3 weeks. Pupae in soil-encrusted cells were transferred to the outdoor cage in December and held there until April, 1980. First-year adults should have emerged by this time because in this area E. crucialis flies from mid-February to early April. The collections were not given close surveillance after spring, 1980; they were stored in the mobile trailer lab, without heating for 6 weeks in midwinter. One dead female E. crucialis was found in the 79E73 lot in early April, 1981, and a dead male in 79E62 in June, 1982. The circumstances indicated that both emerged during the 1980-81 winter (i.e., after 18-22 months in diapause), while I was away on sabbatic leave, although the male may have held over an additional year. Discussion Previously unpublished instances of diapause extending one or more years beyond that believed to be normal in the population are reported for 19 species of moths, representing 8 superfamilies. Including these, Table 1 lists taxa for which I have seen records of prolonged diapause in 90 species in 10 superfamilies. This summary is incomplete but reflects the state of knowledge about the taxonomic distribution of the phe¬ nomenon in Lepidoptera. The preponderance of records in a few fami¬ lies, Prodoxidae, Saturniidae, Pieridae, and Papilionidae, at least in part indicates rearing efforts, while some taxa such as Geometridae and Noctuidae may be disproportionately underrepresented owing to the failure of lepidopterists to look for viable carryover individuals. It is no coincidence that most of the microlepidoptera listed in Table 1 are Prodoxidae and Ethmiidae, the two families with which I have worked most intensively. While it would be premature to attempt a detailed summary of the pattern of occurrence of prolonged diapause in Lepidoptera, a few generalizations seem apparent: a) Dormancy persists beyond the nor¬ mal flight season in prepupal larvae and pupae; it is rare or undetected in adults, eggs, and early instar larvae, b) It has been observed most often in cone- and seed-feeding species that depend upon fruit crops of erratic abundance and in Lepidoptera that live in areas of seasonal drought, c) The ability to carryover appears to be more prevalent among certain butterflies and larger moths than in smaller moths. The generalization that prolonged diapause is more common in Mac- rolepidoptera may be a picture painted with too broad a brush; more likely the phenomenon is characteristic of certain taxa, and is rare in others, within most Ditrysian superfamilies. For example, in the Tor- tricidae, I have seen records of delayed emergence in 8 species, reported 25(2): 83-109, 1986(87) 105 in 17 references by prior researchers, in addition to the one given here. All of these are seed-feeding Eucosmini (1 species) and Grapholitini (7 spp.) (Olethreutinae). Although biologies of a large number of Tortrici- nae have been studied, apparently prolonged diapause has been re¬ ported in none of them. Tortricinae generally and members of the dominant, worldwide tribe Archipini in particular, tend to be species that are indiscriminate in feeding preferences and life cycle pattern, often homodynamic with no fixed dormancy stage. Species that under¬ go diapause do so as eggs or first instar larvae or in adults; it is very rare in full grown larvae or pupae (Powell, 1964: 17). By contrast, most Olethreutinae are host specific (6% polyphagous vs. 24% in Tortricinae, Powell 1980) with a fixed life cycle, and dormancy commonly occurs in prepupal larvae. Not coincidentally, olethreutines, especially Eucosmi¬ ni, reach their greatest diversity, while tortricines are depauperate, in desert areas. On the basis of literature reports and the taxonomic diversity of prolonged diapause among my rather few rearings of desert species, I speculate that most oligophagous Lepidoptera in areas of seasonal drought estivate as prepupal larvae or pupae and that most if not all are capable of producing a facultative second flight and/or carrying over to a subsequent season. Because 2—3 year diapause can occur successfully in tiny moths, such as Coleophora in the Turkistan desert (Falkovitch, 1973), we may expect that in groups such as Gelechiidae, which are characteristically diverse in arid and semiarid regions, the capability of prolonged diapause is not rare. For such insects winter temperatures may be important mitigators of diapause development, as in yucca moths, but rainfall has been implicated as critical in some butterflies (e.g., Emmel, 1975:144, and unpubl. in litt.; Nakamura & Ae, 1977), as has been documented for various other insects. Various aspects of diapause and its importance in insect seasonality have been extensively studied, but the physiological mechanisms of prolonged diapause are poorly understood (e.g., Tauber et al., 1986). Presumably particular token stimuli needed to promote the late phases of diapause maintenance and diapause termination are not received. Hence, when thermal or other thresholds are reached that would have resulted in postdiapause development, the diapause maintenance period continues. The degree of individual variation poses interesting, as yet unanswered questions; often some individuals metamorphose, while others exposed to the same stimuli do not. Usually this occurs in environmental conditions that are abnormal, but such variation indi¬ cates there are differing genetic factors in diapause potential within colonies, or even among sibs of one egg clutch. Tauber et al. (1986: 53, 188, 198, 274) have reviewed the role of prolonged diapause in the evolution of seasonality, life histories and speciation. In their discussion there is an assumption which has been made by several authors that extended dormancy regularly occurs in a 106 J. Res. Lepid. certain proportion of the population as an evolutionary bet-hedging tactic. Tauber et al. credit me with recognizing that there are two kinds of prolonged diapause, either a response by whole populations to adverse conditions by carrying over, or a normal, genetically deter¬ mined occurrence in a certain proportion of the individuals (Powell, 1974). However, I also pointed out that we do not have experimental evidence to demonstrate that there is a fixed polyphenic expression of the genotype, wherein a small number of individuals carry over irrespective of environmental conditions as a kind of buffer against extraordinary climatic extremes. This is still true; in Lepidoptera we do not have data to document that populations of any species express this phenomenon. In yucca moths ( Prodoxus ) I have convincing evidence that such genetic predisposition is not the case; under optimum winter environ¬ ments all or nearly all larvae undergo development, while in adverse conditions all or nearly all maintain diapause (Powell, 1984a, b, 1985). Multiannual emergence patterns such as reported by Carolin (1971) for Coloradia (Saturniidae) appear to represent a fixed polyphenism, but that kind of genetic variability may be manifested only in response to suboptimal climatic situations. Hence, there may not be two discrete classes of prolonged diapause. Rather, populations adapted to erratical¬ ly variable seasonal and biotic environments may be composed of genetically differing individuals such that none, few, many or all maintain diapause depending upon the degree of fitness to optimum seasonal conditions. With Prodoxus it is easy to obtain 100% carryover but almost impossible to promote 100% diapause development under experimental circumstances in the first year. Later, after several years in diapause, individuals will respond to environmental cues that were not sufficient in the first year and proceed through development. This kind of response, rather than a genetically predisposed calendar of events that occurs irrespective of external stimuli, may be producing successive year emergence observed in other insects. Lepidopterists are urged to record observations such as those given here, particularly the environmental conditions to which dormant stages are subjected, as a necessary step toward more detailed analysis of prolonged diapause. Acknowledgements. I thank the following, who made many of the larval collections which led to data presented here: J. A. Chemsak, J. T. Doyen, N. M. Jorgensen, R. E. Dietz IV, P. A. Opler, and P. A. Rude. The last three were supported in part by N. S. F. Grant GB-6813X during 1967 — 1970, which funded field travel during those years. Several people responded to inquiries by offering unpublished records of carryover pupae and subsequent emergences; these included R. S. Bryant, Baltimore, MD; J. A. DeBenedictis, Berkeley, CA; T. C. Emmel, Gainesville, FL; N. McFarland, Sierra Vista, AZ; D. L. Wagner, Berkeley, CA; and W. D. Winter, Dedham, Mass. Authorities of the Depart- 25(2): 83- 109, 1986(87) 107 ment of Forestry, U. C. Berkeley, and particularly A1 Bianchi, caretaker, provided facilities at the Russell Reserve, where collections were held for overwintering during 1965-1976. Identifications of agaves were provided by H. S. Gentry, Phoenix, AZ and of many other flowering plants by J. L. Strother, U. C. Berkeley Herbarium. The cooperation of all these and other students and colleagues is greatly appreciated. Literature Cited BROODRYK, S. W. 1969. The biology of Chelonus ( Microchelonus ) curvimaculatus Cameron. J. Ent. Soc. So. Afr., 32: 169-189. BRYANT, R. S. 1983. Prolonged pupal diapause of Alypia octomaculata (Agaristi- dae). J. Lepid. Soc., 36(3): 237. BUSCK, A. 1917. The pink bollworm, Pectinophora gossypiella. J. Agr. Res., 9: 343-370. CAROLIN, v. M. 1971. Extended diapause in Coloradia pandora Blake (Lepidop- tera: Saturniidae). Pan-Pacific Entomol., 47(1): 19-23. COMSTOCK, J. A. & c. M. dammers. 1937. Notes on the early stages of three Californian moths. Bull So. Calif. Acad. Sci., 36: 68-78. COMSTOCK, J. A. & C. M. DAMMERS. 1939. Studies on the metamorphoses of six California moths. Bull So. Calif. Acad. Sci., 37: 105—128 [19381. CRUMB, S. E. 1956. The larvae of the Phalaenidae. U. S. Dept. Agric., Tech. Bull. 1135; 356 pp. DANILEVSKY, A. S. 1951. On conditions favouring a several year diapause in Lepidoptera [in Russian] [no abstract]. Ent. Obozr., 31: 386-392. DANKS, H. V. 1983. Extreme individuals in natural populations. Bull. Entomol. Soc. Amer., 29(1): 41-45. DAVIS, D. R. 1967. A revision of the moths of the subfamily Prodoxinae. U. S. Natl. Mus., Bull. 255; 170 pp. DAVIS, D. R. 1983. Tineidae. in: Hodges, R. W. (ed.) Check List of the Lepidoptera of America north of Mexico, :5-7. E. W. Classey, Ltd. and Wedge Entomol. Res. Found.; London. DICKSON, R. C. 1949. Factors governing the induction of diapause in the Oriental fruit moth. Ann Entomol. Soc. Amer., 42: 511-537. DOS PASSOS, C. F. & A. B. KLOTS. 1969. The systematics of Anthocharis midea Htibner (Lepidoptera: Pieridae). Entomol. Amer., 45: 1-34. DYAR, H. G. 1891. Preparatory stages of Pheosia dimidiata H. S. Psyche, 6: 194-196. EMMEL, T. C. 1975. Butterflies. Their World, Their Life Cycle, Their Behavior. A. Knopf; New York; 260 pp. FALKOVITCH, M. I. 1973. To the study of Coleophoridae (Lepidoptera) in the Kizil-Kum desert. Trudy Vsesoyuz. entomolog. Obschestra, 56: 199-233 [in Russian]. FERGUSON, D. C. 1971 — 72. The Moths of America North of Mexico. Fasc. 20. 2A, 20. 2B Bombycoidea, Saturniidae. E. W. Classey Ltd. & R. B. D. Publ.; London; 295 pp. FERRIS, G. F. 1919. A remarkable case of longevity in insects (Hem., Horn.) Entomol. News, 30: 27-28. GOUGH, L. H. 1916. The life history of Gelechia gossypiella from the time of the 108 J. Res. Lepid. cotton harvest to the time of cotton sowing. Egypt, Dept. Agr., Tech. & Sci. Surv. Bull. 4: 1-10. HEDLIN, A. F. 1967. The pine seedworm, Laspeyresia piperana (Lepidoptera: Olethreutidae) in cones of ponderosa pine. Canad. Entomol., 99: 264-267. HEDLIN, A. F., G. E. MILLER & D. S. RUTH. 1982. Induction of prolonged diapause in Barbara colfaxiana (Lepidoptera: Olethreutidae): correlations with cone crops and weather. Canad. Entomol. 114(6): 465-472. KOEBELE, A. 1894. A striking instance of retarded development. Insect Life, 6: 336. LAWRENCE, J. F. & J. A. POWELL. 1969. Host relationships in North American fungus-feeding moths (Oecophoridae, Oinophilidae, Tineidae). Bull. Mus. Comp. Zool., 138(2): 29-51. LINSLEY, E. G. & J. W. MACSWAIN. 1945. Longevity of fifth instar larvae of Hornia boharti Linsley. Pan-Pacific Entomol., 21: 88 LINSLEY, E. G. & J. W. MACSWAIN. 1946. Longevity of Trichodes and Pelonium larvae. Pan-Pacific Entomol., 22: 18. MCDUNNOUGH, J. 1949. Revision of the North American species of the genus Eupithecia (Lepidoptera, Geometridae). Bull. Amer. Mus Nat. Hist., 93(8): 537-728. NAKAMURA, I. & S. A. AE. 1977. Prolonged pupal diapause of Papilio alexanor. arid zone adaptation directed by larval host plant. Ann Entomol. Soc. Am., 70(4): 481-484. NESIN, A. P. 1984. Studying diapause in some pests of cones and seeds of conifers [in Russian; abstr. English]. Entomol. Obozr., 63(2): 225-230. POWELL, J. A. 1964. Biological and taxonomic studies on tortricine moths, with reference to the species in California (Lepidoptera: Tortricidae). U. Calif. Publ. Entomol., 32; 317 pp. POWELL, J. A. 1971. Biological studies on moths of the genus Ethmia in Califor¬ nia (Gelechioidea). J. Lepid. Soc., 25, Suppl. 3; 67 pp. POWELL, J. A. 1973. A systematic monograph of New World ethmiid moths (Lepidoptera: Gelechioidea). Smithson. Contr. Zool., 120; 302 pp. POWELL, J. A. 1974. Occurrence of prolonged diapause in ethmiid moths (Lepi¬ doptera: Gelechioidea). Pan-Pacific Entomol., 50(3): 220-225. POWELL, J. A. 1975. Prolonged diapause in Enoclerus zonatus (Cleridae). Coleop. Bull., 29(1): 44. POWELL, J. A. 1978. Survey of Lepidoptera inhabiting three dune systems in the California desert. U. S. Dept. Interior, Bur. Land Mgmt., Res. Contract C A-060 - CT7 - 2827 , Final Rept.; 17 pp. POWELL, J. A. 1980. Evolution of larval food preferences in Microlepidoptera. Ann. Rev. Entomol., 25: 133 — 159. POWELL, J. A. 1984a. Prolonged diapause in yucca moths. XVII Int. Congr. Entomol., Abstr. Vol.: 307. POWELL, J. A. 1984b. Biological interrelationships of moths and Yucca schottii. U. Calif. Publ. Entomol., 100: 1-93. POWELL, J. A. 1985. Synchronized, mass-emergence of a yucca moth, Prodoxus y-inversus, after 16 years in diapause. 36th Ann. Meet., Lepid. Soc., U. Illinois, Urbana; 19 July 1985 (abstr.). POWELL, J. A. & R. A. MACKIE. 1966. Biological interrelationships of moths and Yucca whipplei (Lepidoptera: Incurvariidae, Blastobasidae, Pyralidae). U. Calif. Publ. Entomol., 42; 46 pp. 25(2): 83- 109, 1986(87) 109 PRENTICE, R. M. (COMPILER). 1962. Forest Lepidoptera of Canada. Vol. 2; Nycteoli- dae, Notodontidae, Noctuidae, Liparidae. Canad. Dept. Forestry, Bull. 128: 77-281. RILEY, C. V. 1892. The yucca moth and yucca pollination. 3rd Ann. Rept. Mo. Bot. Gard., 3: 99-159. SHAPIRO, A. S. 1981. Egg-load assessment and carryover diapause in Anthochar- is (Pieridae). J. Lepid. Soc., 34(3): 307-315 [“1980”]. SUNOSE, T. 1978. Studies on extended diapause in Hasegawaia sasacola Monzen (Diptera, Cecidomyiidae) and its parasites. Kontyu, 46: 400-415. SUNOSE, T. 1983. Prolonged diapause in insects and its ecological significance. Kotaigun Seitai Gakkai, Kaiho, 37: 35-48. TAKAHASHI, F. 1977. Generation carryover of a fraction of population members as an animal adaptation to unstable environmental conditions. Res. Popul. Ecol., 18: 232-235. TAUBER, M. J., c. A. TAUBER & s. masaki. 1986. Seasonal Adaptations of Insects. Oxford U. Press, xv + 411 pp. TUSKES, P. M. 1985. The biology and distribution of California Hemileucinae (Saturniidae). J. Lepid. Soc., 38(4): 281-309 [“1984”]. TRIPP. H. A. 1954. Description and habits of the spruce seed worm, Laspeyresia youngana (Kft.) (Lepidoptera: Olethreutidae). Canad. Ent., 86: 385-402. USHATINSKAYA, R. S. 1984. A critical review of the superdiapause in insects. Ann. Zool., 21: 3-30. Journal of Research on the Lepidoptera 25(2): 110-120, 1986(87) An exceptional case of paternal transmission of the dark form female trait in the tiger swallowtail butterfly, Papilio glaucus (Lepidoptera: Papilionidae) by J. Mark Scriber1 and Mark H. Evans Department of Entomology, University of Wisconsin, Madison, WI 53706 Abstract. The melanic dark and yellow forms of the tiger swallowtail butterfly, Papilio glaucus glaucus L., are believed to be controlled by a locus on the Y (W) chromosome. Since the female is the heterogametic sex (XY) in Lepidoptera, dark females should (and generally do) produce only dark daughters while yellow females produce only yellow daughters. Exceptional broods have been reported in which some yellow females arise from dark, and more rarely some dark females arise from yellow mothers. Scriber et al (1986) have shown that these results (as well as both colors of females arising from either colored mother) can be obtained experimentally by hybridizing and backcros- sing with the northern subspecies Papilio glaucus canadensis R & J. The purpose of this communication is to describe the results of a highly unusual case in which the locus for the dark gene controlling melanism from a dark female P. glaucus was transmitted by a male in two separate pairings. This observation has never before been re¬ ported and is significant that it suggests that the locus for black color is not necessarily totally lost when it (rarely) dissociates from its normal (Y) chromosome. Since chiasmata at oogenesis in female Lepidoptera are generally believed to be non-existent, crossing-over is believed not to occur in female Lepidoptera. While pur results do not permit us to distinguish between a cross-over event and a non¬ disjunction of the sex chromosome, we nonetheless have observed results of a rare event, especially for Lepidoptera. Introduction: The melanic dark and yellow forms of the tiger swallowtail butterfly, Papilio glaucus glaucus L., are thought to be controlled by a locus on the Y (W) chromosome (Clarke and Sheppard, 1959, 1962). In fact, this locus controlling dark morph expression in female P. glaucus is one of 1 Current address (and reprint requests to): Dept, of Entomology, Michigan State University, East Lansing MI. 48824 25(2): 110-120, 1986(87) 111 but a few sex-linked marker genes in butterflies (Robinson, 1971; Soumalainen, 1973; R. Hagen, 1986, and pers. comm.). The female is the heterogametic sex in Lepidoptera, and dark females should (and generally do) produce only dark daughters, and yellow females produce only yellow daughters. Exceptional broods have been reported (Ed¬ wards, 1884; Clarke and Sheppard, 1959, 1962; Scriber et al., 1986) in which some yellow females arise from dark, and more rarely some dark females arise from yellow mothers. Scriber et al. (1986) have shown that these results (as well as both colors of females arising from either colored mother) can be obtained experimentally by using hybrids with the northern subspecies Papilio glaucus canadensis R & J. Intermedi¬ ate colored females with a “peppered” or “sooty” color over the yellow tiger-striped background are also observed in nature (see Edwards, 1884; Clark and Clark, 1951), and have been experimentally produced by hybridization or backcrossing with P. rutulus or P. g. canadensis (Clarke and Willig, 1977; Clarke and Clarke, 1983; Scriber et al, 1986). In addition to the partial or complete suppression of the Y-linked melanism in female P. glaucus when paired to P. rutulus and P. g. canadensis males (Scriber et al., 1986), it has also been suggested that, if the Y chromosome bearing the locus for the dark gene is occasionally lost during meiosis of dark females, yellow daughters would be pro¬ duced (Clarke and Sheppard, 1962; Clarke et al., 1976). Scriber et al. (1986) describe such a case in which loss of the locus for dark color in F2 hybrid females is likely to have occurred. However these authors have also observed cases of yellow hybrid daughters of dark mothers which retain the locus for black color. Depending on the male used in subsequent matings, all yellow, all black, or both colors can be obtained from these yellow hybrid or yellow backcross females (Scriber, 1985; Scriber et al., 1986). Here we describe the results of a highly unusual case in which the locus for the dark gene controlling melanism in female P. glaucus seems to have been transmitted by a male to two different hand- pairings. This situation has never before been reported and is signi¬ ficant in that it suggests that the locus for black color is not necessarily totally lost when it (rarely) dissociates from its normal (Y) chromosome. Since chiasmata at oogenesis in female Lepidoptera are generally believed to be absent, crossing-over is believed not to occur in female Lepidoptera (Haldane, 1922; Robinson, 1971; Clarke and Sheppard, 1973; Soumalainen et al., 1973; Turner and Sheppard, 1975). However a suspected crossover in the supergene controlling female polymorph¬ ism in Papilio memnon L. has been reported (Clarke and Sheppard, 1977). Our recent studies of the genetic basis of dark morph expression in Papilio glaucus have involved hand-pairings and mass-rearing of thousands of individuals derived from various geographic locations across North America (Scriber and Evans, 1986a). All of these and the 112 J. Res. Lepid. following specimens are maintained in the Papilio research collection of J.M.S. at the Department of Entomology at Michigan State University. Results. A near-normal but slightly melanic or “sooty” (Fig. 1) yellow male adult butterfly was obtained from a normal-appearing dark morph mother (#674) which was field-captured in Adams County, Ohio by M. H. Evans and W. W. Warfield on July 1983. This male eclosed in 1984 and was hand-paired on May 14, 1984 (#1129) to a virgin yellow morph female (P. g. glaucus) which was lab-reared from a yellow morph Ohio female (#631) field-captured on 14 May, 1983. On the following day (May 15, 1984) a second pairing (#1132) using the same male was made to a virgin P. g. canadensis female (which was lab-reared in 1983 from a 25 June, 1983 field-captured yellow female from Barron County, Wisconsin). All subsequent larvae were reared through to pupation under identical controlled environment conditions (16/18 photo/ scotophase, corresponding thermoperiod of 23 V2°C/19 V2°C). Pupae were weighed and adults were permitted to eclose in cylindrical screen cages. To our surprise, we observed dark as well as the expected (yellow) females in the progeny of both crosses (Table 1). According to all understanding to date, dark females were not expected to occur in offspring of either of these pairings. We have never before observed dark daughters in the lab-reared offspring of more than 600 different P. g. canadensis mothers. While it is possible that Papilio glaucus larvae could be accidentally introduced with foodplant leaves in our laboratory mass rearing procedures at Madison, this is unlikely and careful precautions are continually made to prevent this possibility. We Table 1. Special case in which an aberrent-colored male mated to two different yellow females resulted in dark female daughters (Madison, Wl, 1984). Females Pupae Total eclosed remaining pupae Males - Mating reared eclosed Yellow Dark Dead Alive Parentage* number (n) (1984) (1985) (1984/1985) (1984/1985) (n) (1986) OH(Y) x OH(D)* 1129 69 32 2 9 3 16 1 2 4 PgcxOH(D)* 1132 82 40 4 12 0 21 0 3 2 * Female parent listed first. The OH(D) parent represents an aberrent-colored male (see Fig. la and 1b) reared in 1983 from a normal appearing dark morph mother (#674) captured in Adams County, Ohio on 8 July, 1983. This male was mated to a yellow daughter of an Ohio P. g. glaucus yellow female #632 on 14 May, 1984 (mating #1129); and to a daughter of P. g. canadensis female #614 on 15 May, 1984 (mating #1132). 25(2): 110-120, 1986(87) 113 have not observed any such occurences in the last 5 years with nearly 120,000 ova in our lab. Such errors cannot possibly account for the 38 dark females produced from these two pairings. Discussion. We interpret the results as evidence of male transmission of the gene controlling black color (which is found on the Y chromosome of the heterogametic female). The sons of pairings 1129 and 1132 were all normal in appearance (i.e. they were not black or dark colored as the female morph can be). Approximately 2/3 of the daughters were dark morph and 1/3 yellow morph, and none of the daughters exhibited partial color or mosaic patterns (e.g. dark with irregular blotches/ patches of yellow background showing; see Scriber et al, 1986). This suggests that all cells of dark daughters contain the gene for black color, and favors the idea of non-disjunction or a cross-over of this locus, rather than a particulate cytoplasmic explanation (see Clarke and Sheppard, 1959, 1962). FOLLOWING PAGE CAPTION: Fig. 1. Offspring of dark female #674 from Adams Co., Ohio, 1983: a) dorsal and b) ventral of "slightly aberrant" male (wt. 1.1642); c) dorsal and d) ventral of a "normal" sibling (wt. 1.0459). This first (aberrant) is the male parent in crosses 1129 and 1132 (see Table 1), and is our suspected "carrier" of the female melanism locus. Fig. 2. Ft hybrid offspring of pairing #1132 (a virgin daughter of a 1983 Barron Co., Wisconsin P. g. canadensis female x the aberrant male, wt. 1.1642, of Fig. la & b). a) dorsal and b) ventral of a "slightly aberrant" male (wt. .9386) and c) dorsal and d) ventral of a normal sibling male (wt. .8522). Fig. 3. Offspring of an F2 pairing (#1695; see Table 2) of a dark daughter and a slightly aberrant male (shown in Fig. 2a, 2b) both derived from pairing #1132 (Table 1). a) dorsal and b) ventral of an aberrant male (wt. .9410), and c) dorsal and d) ventral of a normal male sib (wt. 1.0416). Fig. 4. Female offspring of a cross between a yellow morph P. g. glaucus x the "aberrant" male (cross #1129): a) dorsal and b) ventral of a typical dark morph, (wt. 1.4010) and of a typical yellow morph c) dorsal d) ventral (wt. 1.3700) sibling (see Table 1). Fig. 5. Female offspring of Fn hybrid cross of a P. g. canadensis x "aber¬ rant" male P. g. (cross #1132). a) dorsal and b) ventral of a typical dark morph (wt. 0.6964), and c) dorsal and d) ventral of a typical yellow morph (wt. 0.6755) sibling (see Table 1). Fig. 6. A wild collected "aberrant" male from Dane County, Wisconsin (collected 10 August 1983). 114 J. Res. Lepid. illllllilllHIli 25(2): 110-120, 1986(87) 115 liiimmum 116 J. Res. Lepid. Under the hypothesis of a non-disjunction as a causal mechanism, we could expect our male to be of the genotype X (XYD), where the “Y” represents the Y chromosome carrying the gene for dark color. The P. g. canadensis and yellow morph P. g. glaucus females would both be of the genotype XY, and offspring (1129 and 1132; Table 1) would be expected to be the following: XX and X (XYD) males, XY yellow females, and Y (XYD) dark females. This explanation would account for the occurrence of both dark and yellow female daughters; however so would the hypothesis of a cross-over event. In a cross-over of the locus for dark color in this species we would expect the male parent of 1129 and 1132 to be of the genotype XDX. When paired to the P. g. canadensis female and the yellow morph P. g. glaucus female (both presumably XY genotype) we would expect the following: XXD and XX males, XY yellow females, and XDY dark females. Both the cross-over hypothesis and the non-disjunction hypothesis provide explanations for observed yellow and dark daughters. Howev¬ er, neither explains the observed deviation from an expected 50:50 ratio of female morphs. Similarly, the reason for the melanism being restricted to only (some) daughters and none of the sons of this male carrier (not expressed in himself or his sons) is unresolved for both hypotheses. We did, however, notice a slight “sootiness” or semi¬ melanism in the generally normal tiger-striped yellow background proximally on the dorsal surface of the wings in this original male parent (Fig. 1) and in one of his 44 sons (Fig. 2). We had hoped that this could prove to be a phenotypic marker for the male black locus carriers reflecting the X (XYD) or the XXD genotype (from either a non¬ disjunction or a crossover, respectively). Subsequent pairings with offspring of pairings 1129 and 1132 (Table 1) have yielded poor results. Nonetheless, when the aberrant male son (shown in Fig. 2) was mated to one of his sisters (pairing 1695; Table 2) one of the resulting 5 male F2 hybrid sons was markedly melanic in the proximal 1/3 of the wings (Fig. 3a, b). Female offspring resulting from pairings #1129 (P. g. glaucus x P. g. glaucus ) and #1132 (P. g. canadensis x P. g. glaucus) are typical dark or typical yellow in color pattern (Figs. 4 and 5) with one exception, where one daughter is a “dark intermediate’. It should be noted that the female progeny of cross 1129 are larger than those of 1132, reflecting the genetic differences in size between P. g. canadensis and P. g. glaucus. It is also noteworthy that the dark females of cross #1132 represent the only known case of melanism being expressed in Fx hybrids from a P. g. canadensis mother (Fig. 5). In an attempt to follow up the genetic explanation of our unique results in pairings 1129 and 1132, (Table 1), we hand-paired male, yellow female, and dark female offspring of both crosses. Twenty-one different yellow and dark females from cross 1132 were hand-paired 25(2): 110-120, 1986(87) 117 Table 2. A 1984 F2 hybrid pairing of a dark female and slight aberrant male (both from pairing 1132; Table 1). Females eclosed Total Males - Pupae Mating pupae eclosed Yellow Dark alive Parentage number reared (1984) (1985) (1984) (1985) (1984) (1985) (1985 October) (1 132*)2 1695 10 2 3 1 0 1 1 2 * Male with aberrant color; dark morph female Table 3. Pairings of a yellow female Ft hybrid and two of her dark daughters (Madison, Wl; 1985). Females eclosed Total Males - Pupae Mating pupae eclosed Yellow Dark alive Parentage number reared (1984) (1985) (1984) (1985) (1984) (1985) (1985 October) 1 129(Y) x Pgg* *wild Wl male 2343 (see below) 204 — 37 — 6 — 90 71 2343(DK) x Pgg* *wild OH male 2957 88 — 35 — 0 — 37 16 2343(DK) x Pgg* *wild OH male 2974 81 — 35 — 0 — 36 10 (with copulations of 30 minute durations or more). These pairings resulted in only 8 females which produced eggs, only two of which produced any larvae (#1695 produced 34 larvae, #1542 produced 1 larva). The single most useful cross of these attempts was #1695 — an F2 hybrid of a dark female from 1132 x her aberrent male sibling; see Figs. 2a, 2b. This cross generated both yellow and dark daughters as expected under the crossover/non-disjuction hypotheses (Table 2). Since none of his normal-type siblings (see Figs. 2c, 2d) produced female daughters from fourteen mating attempts, we are unable to evaluate whether this aberrent color in males is indicative of possession of the female melanism gene (i.e. a “carrier” criterion). Seven different females from cross #1129 were also hand-paired, of which only 3 produced eggs and only one (pairing #2343 in 1985) produced any larvae. This backcross of a yellow morph female (from 1129) to a wild Wisconsin P. g. glaucus male resulted in 310 larvae, which produced 204 pupae. Unfortunately, instead of resolving the genetic explanation of the paternal transmission of the melanism 118 J. Res. Lepid. capacity, cross #2343 has become an enigma. This cross involving a yellow mother produced 90 dark daughters, 6 yellow daughters and 37 sons (Table 3). The existence of dark daughters was totally unexpected under our hypotheses of crossover and/or non-disjunction because this female parent (XY) should have produced only yellow daughters. Two subsequent pairings of her dark daughters (#2957 and #2974; also in 1985) to wild male P. g. glaucus from Ohio yielded the expected all dark female offspring and an equal sex ratio (Table 3; cf Scriber et al, 1986). We had hoped that the matings in the 1985 season would clarify our suspected crossover/non-disjunction hypotheses, but this was not the case. At present, we have no explanation for the appearance of dark daughters in pairing 2343 (Table 3). The yellow mother from cross 1129 (Table 2) would presumably have been dark if she possessed the gene for melanism, since any autosomal suppressor in P. g. canadensis would not be involved in any pure P. g. glaucus lineage (Scriber et al., 1986) . However, we are not absolutely certain that the Adams County (Ohio) population is free of P. g. canadensis genes from the Appa¬ lachian Mountain region (e.g. Ritland and Scriber, 1985; Scriber and Hainze, 1986). Conclusions. We must emphasize that although we cannot prove a crossover or non-disjunction event, we nonetheless have observed the transmission by a male of the dark morph trait to his daughters (from a mating with a Wisconsin P. g. canadensis yellow female, and from a mating with an Ohio P. g. glaucus yellow female). We do not feel that this phenomenon (appearance of dark daughters from yellow mothers of two different subspecies) is likely to be explained by autosomal melanism suppressor effects from P. g. canadensis introgression. This would require that both the yellow Ohio female and the yellow northern Wisconsin female (the female parents in Table 1) were the result of P. g. canadensis introgression into an ancestrally dark stock. This possibility may not be as farfetched as initially assumed (see Scriber and Evans, 1986a and 1986b). Another remote explanation is that the wild Wisconsin male used in pairing #2343 was simply another independent example of a crossover/non-disjunction, which would also explain dark daughters from a yellow mother presumed to lack the gene for melanism. In this regard it is interesting that partially melanic males (e.g. Fig. 6; and compare Figs. 1, 2, and 3) have been captured from the same popula¬ tion in Wisconsin as the mated male in cross 2343. None have been tested for the dark gene transmission potential however. Should we be correct in assuming that our results reflect some form of crossover in female Lepidoptera, then there should be special precau¬ tions taken by systematists who employ maternal DNA (maternal inheritance of DNA) techniques in evaluating phylogenies, and assume a clear record of the maternal lineage (see Avise and Lansman, 1983 25(2): 110-120, 1986(87) 119 for further discussion). The adaptive significance of achiasmatic meiosis (and the assumption that this is accompanied by the absence of crossing-over) are not entirely clear, but it has evolved repeatedly in at least 10 major groups of animals (White, 1973). Sexual mosaics, color mosaics, and bilateral gynandromorphs of Papilio glaucus may be more common than generally believed, especially near suspected hybrid zones (Clarke and Clarke, 1983; Scriber and Evans, 1986b). Perhaps such chromosomal/developmental abnormalities will provide us with other additional opportunities to evaluate our crossover/non¬ disjunction hypotheses in the future. Acknowledgments. This research was supported in part by grants from the National Science Foundation (DEB #7921749, BSR #8306060, BSR #8503464), the USDA (#85CRCR-1-1598), the Graduate School and College of Agriculture and Life Sciences (Hatch 5134) of the University of Wisconsin, Madison. We especially thank William Warfield for his assistance in the field and Sir Cyril Clarke, Robert Hagen, and Sarah Rockey for their comments on the manuscript. Literature Cited AVISE, J. C. & R. A. LANSMAN. 1983. Polymorphism of mitochondrial DNA in populations of higher animals, pp. 147-164 In (M. Nei and R. K. Koehn, eds.) Evolution of Genes and Proteins. Sinauer, Sunderland, Mass. CLARK, A H. & CLARK, L. F„ 1951, The butterflies of Virginia. Smithsonian Inst. Misc. Coll. 116 (no. 7): 124-144. CLARKE, C. A. & F. M. M. CLARKE, 1983, Abnormalities of wing pattern in the Eastern Tiger Swallowtail butterfly, Papilio glaucus. Syst. Entomol. 8:25-28. CLARKE, C. A. & SHEPPARD, P. M., 1959, The genetics of some mimetic forms of Papilio dardanus, Brown, and Papilio glaucus , Linn. J. Genetics 56:236-260. CLARKE, C. A. & SHEPPARD, P. M., 1962, The genetics of the mimetic butterfly, Papilio glaucus. Ecology, 43:159-161. CLARKE, C. A. & P. M. SHEPPARD, 1973, The gentics of four new forms of the mimetic butterfly, Papilio memnon L. Proc. Royal Society of London, B, 184:1-14. CLARKE, C. A. & SHEPPARD, P. M., 1977, A new tailed female form of Papilio memnon L. and its probable genetic control. Systematic Entomology 2:17-19. CLARKE, C. A., SHEPPARD, P. M„ & MITTWOCH, U., 1976, Heterochromatin poly¬ morphism and colour pattern in the tiger swallowtail butterfly, Papilio glaucus L. Nature, 263:585-587. CLARKE, C. A. & WILLIG, A., 1977, The use of a-ecdysone to break permanent diapause of female hybrids between Papilio glaucus L. Female and Papilio rutulus Lucas male. J. Res. Lepid. 16:245—248. EDWARDS, W. H., 1884, The butterflies of North America (Vol. II). Houghton Mifflin Co. HAGEN, R. H. 1986. The evolution of host plant use by the tiger swallowtail butterfly, Papilio glaucus. Ph. D. Thesis. Cornell Univ., Ithaca, N. Y. p. 297. 120 J. Res. Lepid. HALDANE, J. B. S., 1922, Sex ratio and unisexual sterility in hybrid animals. J. Genetics 12:101-109. RITLAND, D. B. & J. M. SCRIBER., 1985. Larval developmental rates of three putative subspecies of tiger swallowtail butterflies, Papilio glaucus , and their hybrids in relation to temperature. Oecologia 65:185-193. ROBINSON, R., 1971, Lepidoptera genetics. Pergamon, Oxford. SCRIBER, J. M., 1985, The ecological and genetic factors determining geographic limits to the dark morph polymorphism in Papilio glaucus. Bulletin of the Ecological Society of America 66:267. SCRIBER, J. M. & M. H. EVANS., 1986a. The genetic control and ecological factors affecting the North American distribution of melanic (dark morph) poly¬ morphism in female tiger swallowtail butterflies (Papilionidae: Lepidop¬ tera). Ecology (in prep.). SCRIBER, J. M. & M. H. EVANS., 1986b. Bilateral gynandromorphs, sexual mosaics, and other aberrants in the tiger swallowtail butterfly, Papilio glaucus (Papilionidae: Lepidoptera). J. Research Lepid. (submitted). SCRIBER, J. M„ M. H. EVANS, & D. RITLAND, 1986, Hybridization as a causal mechanism of mixed color broods and unusual color morphs of female offspring in the eastern tiger swallowtail, Papilio glaucus. In: (M. Huettel, ed.), Evolutionary Genetics of Invertebrate Behavior. The USDA and Univ. of Florida, Gainesville (in press). SCRIBER, J. M. & J. HAINZE. 1986. Geographic variation in host utilization and the development of insect outbreaks. Chapter 20 In (P. Barbosa and J. C. Schultz, eds.) Insect Outbreaks: Ecological and Evolutionary Processes (in press). SUOMALAINEN, E., L. M. COOK, & J. R. G. TURNER, 1973, Achiasmatic oogenesis in the Heliconiine butterflies. Hereditas, 74:302-304. TURNER, J. R. G. & SHEPPARD, P. M., 1975, Absence of crossing-over in female butterflies ( Heliconius ). Heredity, 34:265-269. WHITE, M. J. D., 1983, Animal Cytology and Evolution. 3rd Edition, Cambridge University Press. London. 961 pp. Journal of Research on the Lepidoptera 25(2): 121-135, 1986(87) The Phenetics and Comparative Biology of Euphilotes enoptes (Boisduval) (Lycaenidae) from the San Bernardino Mountains Gordon F. Pratt & Greg. R. Ballmer Entomology Department, University of California, Riverside, CA 92521 Abstract. Euphilotes enoptes larvae in the San Bernardino moun¬ tains utilize both perennial and annual Eriogonum species. Many San Bernardino mountain locations have the same Eriogonum species; despite this their utilization as hosts varies amongst populations. Seasonal flight periods which correspond to the initiation of the major host’s bloom were not only variable amongst populations, but from year to year. One spring emerging population did not fly during 1984 and 1985 and another had shortened flight periods. Despite differ¬ ences in hostplants and flight periods, these populations appear to be more closely related in larval setation to each other than to six other described subspecies. Introduction Populations of Euphilotes enoptes (Boisduval) are widely distributed in western North America. They can be found in a variety of habitats from sea level to over 11,000 feet, and from moist cool climates in the Sierra Nevada mountains to the hot dry desert around Palm Springs. The nine described subspecies of this small blue (Miller and Brown, 1981) are often better defined by geographic distribution, flight period, and host plant selection than by adult morphological characters. Dis¬ tribution of certain subspecies can be quite large as in E. enoptes ancilla (Barnes and McDunnough) covering the seven states, (Califor¬ nia, Colorado, Idaho, Montana, Nevada, Oregon, and Wyoming) or extremely small as in E. enoptes smithi (Mattoni) which is found only along the coast of Monterey Co., California (Shields, 1977). The larvae oiE. enoptes feed exclusively on blossoms of various Eriogonum species. Most subspecies are known to utilize a single host plant species in a given location and all are believed to be univoltine with the flight season coinciding with the onset of the host flowering period. Various populations fly in every month from March to October. In southern California two subspecies, E. enoptes dammersi (Com¬ stock & Henne) and E. enoptes mojave (Watson & Comstock) are recognized. The former flies in late summer and fall in the mountains and foothills of the Colorado and eastern Mojave deserts; its larval hosts are Eriogonum davidsonii Greene, E. elongatum Benth., E. wrightii nodosum (Small) Reveal, and E. w. wrightii (Torr.) S. Stokes. 122 J. Res. Lepid. Euphilotes e. mojaue flies in the spring in the Mojave Desert and western fringe of the Colorado Desert; its larval hosts are the annuals, E. pusillum Torr. and E. reniforme Torr. & Frem. Shields (1975) speculated on the basis of similarities in distribution and male genitalia that these subspecies are closely related to each other. The life histories of both subspecies have been published (Comstock & Henne, 1965; Comstock, 1966) but the larval descriptions lack sufficient detail to differentiate them from each other or even other lycaenid species. The present study is an effort to better define the ecological and evolution¬ ary relationships of these 2 taxa and to compare them with other named and unnamed montane populations of E. enoptes. The San Bernardino Mts. are an extremely complex and interesting geological area. Here the Mojave and Colorado deserts, the coastal chaparral, and the cooler, moister higher elevations of the mountains all meet. Along the northern and northeastern slopes occur spring flying populations of E. enoptes. The northwestern slopes have late summer populations. In the high elevations there are populations that fly in early summer and, depending on rainfall patterns, can be found through mid October. Along the eastern slopes there are populations that fly exclusively in the fall. Materials and Methods Studies of E. enoptes in the San Bernardino Mountains entailed numerous field observations at four colony sites (figure 1) to determine seasonal activity, host range, and larval behavior. Doble (DB), el. 6700’, (2,000 meters) located at the northeastern end of Baldwin Lake, is an open gently sloped flat of rocky clay soil. The major vegetation consists of short ground cover perennials with scattered Pinus monophylla Torr. & Frem. and Artemisia tridentata Nutt. A second locality (AC) about ten miles (16 km) south-southeast of DB is situated along the steep slope east of Arrastre Creek (AC), el. 7100’ (2,200 meters). This site is open with scattered Juniperus occidentalis Hook, P. monophylla , Ceanothus cordulatus Kell, Cercocarpus ledifolius Nutt., and a diverse but sparse community of smaller annuals and perennials; the soil is rocky and porous. A third locality about 12 miles west of DB at Big Pines Flat (BP), el. 6800’ (2,100 meters), has uneven terrain with P. monophylla and Pinus ponderosa Dougl. ex P. & C. Lawson forming open stands interspersed with scattered low perennials and annuals. The fourth locality, Mojave River Forks (MR), el. 3100’ (950 meters), is 25 miles (40 km) west of DB at the northwestern corner of the San Bernardino Mountains. This site is warmer than the other sites. It is a gently sloped alluvium cut by numerous shallow washes and intermit¬ tent creeks; vegetation is diverse, containing elements of Mojave De¬ sert, montane forest, and coastal chaparral communities. Major vegetation includes Juniperus californica Carr., Artemisia tridentata , 25(2): 121-135, 1986(87) 123

Victorville San Jacinto Mountains Figure 1 . Map of the study sites in the San Bernardino Mts, abbreviations as in the Materials and Methods. Line shows 5,000 ft. elevation of mountains. and Quercus wislizenii A. Dc. with scattered thickets of Adenostoma fasciculatum H. & A., Ceanothus spp., and Cercocarpus betuloides Nutt, ex T. & G. Each of these sites except BP was visited several times during the years 1983 to 1985. Larvae of E. enoptes were also acquired from the following 18 sites (see fig. 2) for comparison of setal characters: (BG) Bob’s Gap, N. base San Gabriel Mts., Los Angeles Co., Ca., el. 4000’ (1200 meters), 22. V. 83., on E. pusillum GRB; {E. e. mojave ) (CC) Chino Canyon, San Jacinto Mountains, Riverside Co., Ca., el. 2600’ (800 meters), 27. IX. 83., on E. davidsonii and E. w. nodosum , GRB & GFP; (Type locality for E. e. dammersi ) (CS) Upper Centennial Spring, Coso Range, Inyo Co., el 6100’ (1900 meters), 1. VIII. 83., on Eriogonum nudum Dougl. ex Benth., J. F. Emmel; (subspecies undefined) (EP) El Paso Mountains, vie. Last Chance Canyon, Kern Co. Ca., el. 2500’ (760 meters), 19. V. 83., on E. nudum , GRB & GFP; (undefined subspecies near E. e. mojave) (LH) Landels Hill Big Creek Reserve, Monterey Co., Ca., el. < 100’ (30 meters), 17, VIII. 84., on Eriogonum parvifolium Sm. in Reese, GFP; (E. e. smithi) 124 J. Res. Lepid. Figure 2. Map of the collection sites; abbreviations as in the Materials and Methods. (MA) Marina, Monterey Co., Ca., el. < 100’ (30 meters), 22. VII. 83. on Eriogonum latifolium Sm. in Reese, GRB; (E. e. smithi) (MCA) Big Morongo Canyon, Riverside and San Bernardino Cos., Ca., el. ca 2000” (600 meters), 17. IV. 84., on E. pusillum, GFP; ( E . e. mojave ) (MCS) same data as above except 15. IX. 84., on Eriogonum elonga- tum Benth., GPP; (E. e. dammersi ) (MY) Mayer, Yavapai Co., AZ., el. 4500’ (1400 meters), 15. X. 82., on E. w. wrightii, GRB; (E. e. dammersi ) (PM) Pyramid Mountain, San Jacinto Mountains, Riverside Co., Ca., el. 6000’ (1800 meters), 17. VI. 82., 6 VI. 83., 1. VII. 83., 26. V. 84., on E. davidsonii, GRB & GFP; (subspecies undefined) (PR) Point Richmond, Contra Costa Co., Ca., el. < 100’ (30 meters), 23. VII. 83., on Eriogonum nudum auriculatum (Benth.) Tracy ex Jeps., GRB; (E. e. bayensis ) 25(2): 121-135, 1986(87) 125 (PV) 28 mi. E. of Pine Valley on HWy. 8, San Diego Co., Ca., el. 3500’ (1100 meters), 26. VII. 84., on E. elongation, GRB; ( E . e. dammersi ) (RN) Randsburg, Kern Co., Ca., el. 3500’ (1100 meters), 19. V. 83., on E. pusillum, GRB & GFP; (E. e. mojave ) (SP) Santa Paula, Ventura Co., Ca., el. 1000’ (300 meters), 20. VI. 84., on E. parvifolium, GFP; ( E . e. tildeni ) (ST) Stanton, Yavapai Co., Az., el. 3500’ (1100 meters), 15. X. 82., on E. w. wrightii, GRB; (E. e. dammersi ) (WA) Warren Canyon, near Tioga Pass, Mono Co., Ca., el. 9000’ (2700 meters), 17. VII. 83., on E. nudum, GRB & GFP; (E. e. enoptes ) (WC) Wildhorse Canyon, Mid Hills, eastern Mojave Desert, San Ber¬ nardino Co., CA., el. 4000’ (1200 meters), 2. X. 82., on E. w. wrightii, GRB & GFP; (E. e. dammersi) (WW) Wrightwood, 1 mi. W., Los Angeles Co., Ca., el. 6000’ (1800 meters), 7. IX. 82. and 13. VIII. 83., on E. nudum saxicola (Heller) S. Stokes, GRB & GFP; (Shields, 1977, places populations from this area in the nominate subspecies but they may be closer to E. e. tildeni) Larvae were obtained by beating host plant inflorescences, searching for floral shelters, or by rearing from ova. Ova and larvae were often found with other lycaenid species including Celastrina argiolus (Lin¬ naeus), Hemiargus ceraunus gyas (W. H. Edwards), Icaricia acmon (Westwood & Hewitson), Icaricia neurona (Skinner), and Strymon melinus Hubner. Ova of E. enoptes were easily distinguished by their poorly defined chorionic ridges, and larvae were separated by setal outlines. Although color can be variable, larvae of E. enoptes are usually yellow or white (never green) with pink or red chevron mark¬ ings while larvae of the other species are often green. Samples of larvae from all localities were injected with Kahle’s fluid, fixed in hot water, and stored in 80% ethanol. Often larvae were reared on host plants from their collection sites; occasionally, other hosts were substituted. Since E. enoptes larvae are cannibalistic they were reared individually in screened vials with flower stalks placed in water to maintain freshness; flowers were frequently changed to avoid mold. Most larvae were permitted to pupate in soil or the rearing container. Pupae were kept under a variety of conditions, as shown in Table 1. Eclosion dates were re¬ corded. As with most Lycaenids, mature (fourth instar) larvae (Langston and Comstock (1966) and Arnold (1983) state that E. enoptes hayensis and E. e. smithi respectively have 5 instars, yet in hundreds of rearings of various E. enoptes populations we have found only four instars of E. enoptes) are covered with short secondary setae and possess a variable number of more prominent (longer and erect) setae grouped in loca¬ tions where primary setae should occur (sensu Hinton, 1946) (Fig. 3). These locations are dorsal (just lateral to the midline), subdorsal (slightly dorsal to the line of spiracles), and lateral (along the lateral fold). Also, on the prothorax, a few prominent setae occur on the shield 126 J. Res. Lepid. Table 1. Pupae initially kept at 27-35°c were refrigerated (5°c) for at least two months ending in December; afterwards they were kept at 22-27°c. Pupae initially kept at 22-27°c were refrigerated from Dec. 22, 1983 to April 3, 1984 and returned to 22-27°c. Those pupae not refrigerated first year were refrigerated with other pupae their second year for 2-3 months. Those pupae not refri¬ gerated were subjected to moderated daily fluctuations in temperature. The number of pupae from each site is shown in parenthesis. Pupae kept at 27 — 35°c Pupae kept at 22-27°c Pupae not refrigerated first year Pupae not refrigerated AC (4) BG (5) MY (1) BP (5) EP (7) BP (9) ST (1) MR (3) LH (25) CC (3) WC (3) DB (54) McA (15) DB (8) SP (3) McS (65) EP (8) MA (1) PM (7) MR (5) PV (2) PM (1) PR (9) PV (6) WA (2) WW (3) ABD 1 *7 Figure 3. Diagram of an E. enoptes mojave larva showing the positions of the setae counted for the 11 characters. Those positions are (D) Dorsal, (SD) Subdorsal, (L) Lateral, (ABD 1-7) Abdominal Segments 1 to 7, and (Shield) Prothoracic Shield. 25(2): 121-135, 1986(87) 127 and many more are located in front of the shield and ventrolateral to the shield in poorly defined groups. Elsewhere, prominent setae occur in specified locations and are most abundant on the mesothorax. No apparent difference in number of prominent setae was found among abdominal segments 1-6 but an increase in number of lateral pro¬ minent setae was often noted on the remaining segments. No promin¬ ent dorsal setae occur on the seventh abdominal segment in the region of Newcomer’s organ (honey gland); also none occur on the more posterior segments. Both the number and size of prominent setae vary. For comparative purposes the total prominent setae in each location (both sides of the larva) were summed for each segment. Prominent setae were given a value of one if they were at least twice (>0.2 mm) as long as surround¬ ing secondary setae and one-half if they were 1.5—2 times (0.15—0.2 mm) as long as the secondary setae. Prominent setae in eleven locations were quantified and subjected to statistical analysis using Duncan’s Multiple Range Test. These loca¬ tions were: (1) prothoracic shield, (2) dorsally on the mesothorax, (3) dorsally on the metathorax, (4) dorsally on abdominal segments 1-6, (5) subdorsally on the mesothorax, (6) subdorsally on the metathorax, (7) subdorsally on abdominal segments 1—6, (8) laterally on the mesothorax, (9) laterally on the metathorax, (10) laterally on abdomin¬ al segments 1—6, and (11) laterally on abdominal segment 7. Promin¬ ent subdorsal setae on abdominal segment 7 and prominent lateral setae on abdominal segments 8-10 may offer characters for statistical analysis but were not included. A variable number of larvae were used to represent each population; the minimum number was 6 the maximum 30. Consecutive genera¬ tions of larvae were sampled at DB in June (DB1) and July (DB2) 1983. Samples were taken from MR in September 1982 (MR1) and October 1983 (MR2). These populations were compared statistically to ascertain the stability of mean character states. For the populations PM and WW, larvae from consecutive generations were pooled. Tables 2-5 present the results of statistical analysis. Populations are listed according to specified abbreviations followed by the number of larvae (n), character mean, standard error, and results of Duncan’s Test at the 1% error level. Results Larval setation analysis separates the E. enoptes populations studied herein into 4 basic groups. The populations of E. e. mojave (MCA, BG, RN) have the largest mean number of prominent setae. The number of prominent setae for these populations is significantly higher than for all other populations in dorsal, subdorsal, and lateral positions. With population EP they also have a significantly higher mean number of prominent setae on the prothoracic shield. 128 J. Res. Lepid. Population EP ranks next in mean number of prominent setae in the same locations. It differs significantly from all other populations in prominent setae dorsally on all segments (Table 2) and laterally on abdominal segments 1 — 7 (Tables 3 and 5). EP and PM together differ significantly in prominent subdorsal setae on the metathorax (Table 2). Populations CS, PM, and WA often rank together with means higher than all other populations except those above. They are not significant¬ ly different from each other in mean number of prominent dorsal setae on all segments and prominent subdorsal setae on abdominal segments 1—6. PM and WA differ significantly from other populations in promin¬ ent lateral setae on abdominal segments 1—6; they differ significantly from each other, but not from CS, in prominent subdorsal setae on the mesothorax and prominent lateral setae on the metathorax. PM and CS differ significantly from each other but not from WA in lateral prominent setae on abdominal segment seven. There is little overall difference in mean number of prominent setae among the populations AC, DB1, DB2, BP, CC, LH, MA, MR1, MR2, MY, PR, PV, SP, ST, WC, and WW. For all setal characters they either do not differ significantly from each other or form a series of overlap¬ ping nonsignificant subsets. Populations DB1 and DB2, which repre¬ sent consecutive generations in June and July, respectively, differ slightly, but not significantly, for all means, except subdorsal setae on abdominal segments 1-6; these are identical. Populations MR1 and MR2, which represent consecutive generations at MR in 1982 and 1983, respectively, differ slightly but not significantly for all means except prominent dorsal setae on the metathorax and dorsal and subdorsal setae on abdominal segments 1-6; these are identical. Character means for the San Bernardino Mountains populations, (AC, DB1, DB2, BP, MR1, MR2), generally do not differ significantly. However, the mean number of prominent setae on the prothoracic shield is significantly higher for AC than for DB2 and MR2. Also, DB1 differs significantly from MR1 and BP in mean number of prominent dorsal setae on the mesothorax; it also differs significantly from MR1 in mean number of prominent lateral setae on the metathorax and from both MR1 and MR2 in mean number of prominent lateral setae on abdominal segment seven. According to field observations (Table 6) there are at least three separable populations of E. enoptes in the San Bernardino Mts. The one at AC is single brooded and can be found only in the spring. Another population (BP and MR) occurs as adults during late summer and early fall. At DB E. enoptes appears in early spring, but can be found, depending on rainfall, into early fall overlapping the flight periods of the two other populations. The rainfall patterns also affected AC and MR over the three years. The spring of 1983 was wet, whereas both 1984 and 1985 were seasonably dry. This may account for adults 25(2): 121-135, 1986(87) 129 emerging up to 2 weeks earlier at DB and MR, and both larvae and adults at AC absent during 1984 and 1985. Weekly visits to MR during 1982 and 1983 revealed no E. enoptes adults or larvae prior to August 21 except three larvae on E. pusillum (29. V. 1982) which had similar setation to E. e. mojaue larvae from BG, MCA, and RN. Although E. elongatum and E. wrightii trachygonum normally do not bloom until August, E. dauidsonii is abundant at MR and blooms from spring to summer. However, the only lycaenid larvae found on E. dauidsonii at MR were I. acmon. The eclosion dates for pupae from the four San Bernardino Mountain sites correspond to field observations. All pupae from DB failed to diapause. Of four AC pupae, initially kept at 27— 35°C, one failed to diapause, while the other 3 eclosed within four weeks after removal from refrigeration in January 1984. Three of five pupae from BP did not diapause at 27-35°C. The remainder were kept at unheated Riverside temperatures from December 1983 until they eclosed in July and August 1984. Nine other BP pupae were kept at 22-27°C, and refrigerated from December 28, 1983 to April 3, 1984. They eclosed from May 30, 1984 to July 27, 1984. Five MR pupae were refrigerated and incubated with those from BP, and eclosed July 9, 1984 to Sept. 2, 1984. Another three MR pupae were not refrigerated but kept at 27°C, as with those from BP, and eclosed mid July to mid August 1984. All three pupae from MY, one from ST, and five from WC eclosed during September and October 1983. In 1984 (after refrigeration treat¬ ment) two more from WC eclosed in July and August; one pupa each from ST and WC still remained in diapause. A variable number of pupae from most locations eclosed within four weeks when initially kept at 27-35°C. These include those from the San Bernardino Mountains, as noted above, DB (50), EP (2), LH (21), MCA (3), MCS (4), PM (1), SP (3), WA (3), and WC (2). Of the pupae initially kept at 22— 27°C only those from PV (6), WA (2), and DB (8), failed to diapause. Many pupae eclosed within 4—5 weeks after removal from refrigera¬ tion in December. These include EP (5), MCA (12), and PM (5); one pupa from PM did not eclose in the winter of 1983 but, after a second season including refrigeration again, eclosed in January 1984. Larvae of E. enoptes were found on five species of Eriogonum in the San Bernardino Mountains. Eriogonum dauidsonii is an annual which begins to bloom in spring and may continue into summer and fall, depending on soil moisture. It is the only apparent host oiE. enoptes at AC and may be the preferred host at BP, since about twice as many larvae were found on it as on E. wrightii subscaposum. This plant is absent from DB. At MR it blooms primarily in spring. Eriogonum kennedyi, which occurs in a few isolated sites in the San Bernardino Mountains, begins to bloom in May or June; at DB it bloomed during May and June to September during 1983. In 1984 at DB it bloomed 130 J. Res. Lepid. Table 2. The mean total prominent dorsal setae on the mesothorax, meta¬ thorax, and abdominal segments 1 through 6; means followed by the same letter are not significantly different at the 1% level. Locality n meso¬ thorax mean S.E. 1* meta¬ thorax mean S.E. 1* A 1-6 mean S.E. 1* MCA 23 8.04 0.54 A 4.41 0.53 A 19.37 1.37 A EC 17 7.18 0.51 A8 3.88 0.44 A 19.76 1.20 A RN 19 6. 58 0. 43 8 4.21 0.38 A 21.11 1.41 A EP 16 4.72 0.44 C 1.78 0.35 6 10.33 1.59 6 itffi 12 3. 42 0.29 D 0.00 0.00 c 0.00 0.00 C PM 17 3. 24 0.39 D 0.42 0.19 c 2.44 0.88 C CS 14 3.07 0.21 D 0.00 0.00 c 0.00 0.00 C BL1 23 1.08 0.23 EF 0.09 0.06 c 0. 13 0.13 C SP 6 1.17 0. 48 EF 0.00 0.00 c 0.00 0.00 c Uk 16 1.08 0.27 EF 0.00 0.00 c 0.00 0.00 C PR 26 0.98 0.20 EF 0.00 0.00 c 0.00 0.00 c BL2 17 0.91 0.26 EF 0.00 0.00 c 0.00 0.00 c MRS 22 0.57 0.21 EF 0.00 0.00 c 0.00 0.00 c AC 8 0.21 0.21 F 0.00 0.00 c 0.00 0.00 c LH 30 0.37 0.13 F 0.00 0.00 c 0.00 0.00 c CC 7 0.29 0.29 F 0.00 0.00 c 0.00 0.00 c 8PF 14 0.28 0.15 F 0.00 0.00 c 0.00 0.00 c MY 9 0.22 0.22 F 0.00 0.00 c 0.11 0.11 c ICS 27 0.20 0.10 F 0.00 0.00 c 0.00 0.00 c ST 13 0.19 0.05 F 0.00 0.00 c 0.05 0.03 c WC 21 0.05 0.03 F 0.00 0.00 c 0.00 0.00 c PV 15 0.03 0.03 F 0.00 0.00 c 0.00 0.00 c MR1 17 0.03 0.03 F 0.00 0.00 c 0.00 0.00 c MA 10 0.00 0.00 F 0.00 0.00 c 0.00 0.00 c Table 3. The mean total prominent sub-dorsal setae on the mesothorax, metathorax, and abdominal segments 1 through 6; means follow¬ ed by the same letter are not significantly different at the 1 % level. Meso¬ meta- abd. thorax thorax seg. Locality n mean S. £. 1* mean S.E. 1* 1-6 S.E. 1* MCA 23 7. 80 0.64 A 4.28 0.28 A 12.80 1.69 A RN 19 7.21 0.41 A 4.74 0.30 A 10.63 1.39 A 86 17 6. 76 0.65 A 4.41 0.41 A 11.47 1.29 A EP 18 4.67 0.29 B 2.22 0.11 B 1.33 0.62 B PM 17 4. 24 0.32 B 1.86 0.30 B 1.06 0.69 B CS 14 3.50 0.39 BC 0.75 0.20 C 0.00 0.00 B WA 12 2.50 0.15 CD 0.08 0.06 c 0.00 0.00 B DBi 23 2.48 0.24 CD 0.43 0.15 c 0.00 0.00 B D62 17 2.29 0.28 CDE 0.32 0.18 C 0.00 0.00 B AC 8 2.13 0.30 CDE 0.56 0.30 c 0.00 0.00 B MR2 22 1.80 0.26 DE 0.09 0.09 c 0.00 0.00 B CC 7 1.79 0.46 DE 0.00 0.00 c 0. 00 0.00 B MY 9 1.78 0.32 DE 0.00 0.00 c 0.00 0.00 B PV 15 1.77 0.19 DE 0.07 0.07 c 0.00 0.00 B ST 13 1.69 0.29 DE 0.00 0.00 c 0.00 0.00 B MR1 17 1.56 0. 31 DE 0.00 0.00 c 0. 12 0.07 B LH 30 1.52 0.20 DE 0.00 0.00 c 0.00 0.00 B MCS 27 1.44 0.21 DE 0.00 0.00 c 0.00 0.00 B PR 26 1.44 0.14 DE 0.00 0.00 c 0.00 0.00 B WW 18 1.31 0.19 DE 0.03 0.03 c 0.00 0.00 B MA 10 1.20 0.29 DE 0.00 0.00 c 0.00 0.00 B WC 21 1.17 0.19 DE 0.00 0.00 c 0.02 0.02 B SP 6 1.17 0.65 DE 0.00 0.00 c 0.00 0.00 B BP 14 0.89 0.27 E 0.00 0.00 c 0.00 0.00 B 25(2): 121-135, 1986(87) 131 Table 4. The mean total prominent lateral setae on the mesothorax, meta¬ thorax, and abdominal segments 1 through 6; means followed by the same letter are not significantly different at the 1% level. Locality n ffleso- thorax Mean S. E. lU meta- thorax mean S.E. lit abd. seo. 1-6 S.E. lit wa IS 10.83 0.37 A 6.75 0.37 AB 12.29 1.26 C LH 30 10.05 0.AS AB 3.88 0.A2 CDEF 0.95 0.32 D MCA S3 9.91 0.6A AB 8. A3 0.5A A A3. 28 2.95 A cs 1A 9. BA 0.39 ABC 5.1A 0.31 BC 5.36 1.38 D £P 16 9.33 0.63 ABCB 6.78 0.A6 AB 30.06 2.65 B RN 19 9.26 0.21 ABCD 7.A7 0.39 A A6.53 1.76 A SP 6 9.00 0.26 ABCDE A. 25 0.60 CDE 0.00 0.00 D BG 17 8.76 0.28 BCDE 7.35 0.23 A A3.2A 1.A9 A MA 10 8.50 0.5A BCDE A. 10 0.55 CDE 0.10 0. 10 D PR SB 8.0A 0.30 BCDEF A. 58 0.A1 CD 0.73 0.38 D BP 1A 7.79 0.58 CDEF 3.00 0.50 DEFGH 1.A6 0.5A D Ml 18 7.50 0.AA DEFG 3.06 0.A5 DEFG 0.6A 0.30 D PM 17 7.00 0.33 EFG 5.00 0.33 C 16.71 1.69 C DBS 17 7.00 0.23 EFG 2.A1 0.38 EFGHI 0.88 0.A6 D DB1 S3 6.% 0.29 EFG 3.07 0.27 DEFG 2. 17 0. 79 D MRS SS 6.A5 0.A2 FGH 1.18 0.A2 HIJ 0.05 0.05 D MR1 17 6.18 0.5A FSH 2. 12 0.51 FGHIJ 0.00 0.00 D AC 8 6.13 0.35 FGH 2.13 0. A0 F6HIJ 0.63 0.30 D MY 9 6.00 0.87 FGH 0.67 0.A7 IJ 0.00 0.00 D CC 7 5.71 0.57 GH 0.29 0.18 J 0.00 0.00 D PV 15 5.67 0.31 GH 0.90 0.32 IJ 0.07 0.07 D MCS S7 5.65 0.26 GH 1.69 0.26 GHIJ 0.00 0.00 D WC 21 A. 86 0.A8 H 0.A3 0.16 J 0.00 0.00 D ST 13 A.5A 0.A9 H 0.69 0.26 IJ 0.00 0.00 D Table 5. The mean total prominent setae on the prothoracic shield and laterally on abdominal segment 7; means followed by the same letter are not significantly different at the 1% level. setae setae on shield abd.seg. 7 Locality n mean S.E. 1* mean S.E. 1* RN 19 3.A7 0. 19 A 7.58 0.50 A BG 17 3.12 0.23 AB 7.A1 0.37 A E? 18 2.72 0. 2A AB i J. 3*5 0.5A B MCA 23 2. A8 0. 15 B 7.98 0.3A A PM 17 1.71 0.2A C 3.A7 0. A2 C WA 12 1.67 0. 22 C 2.58 0.19 CD WW 16 1.39 0.20 CD 0.67 0. 23 EFG AC 8 1.38 0.38 CD 0.63 0.32 FG PR 26 1.23 0.17 CDE 1.06 0. 17 EFG CS 1A 0.93 0.22 CDEF 1.86 0.23 DE DB1 23 0.91 0.21 CDEF 1. A3 0.18 EF MA 10 0.70 0.21 DEFG 0.50 0.27 FG MR1 17 0.68 0.21 DEFG 0. 12 0.12 G BP 1A 0.57 0.20 DEFG 0.A6 0.19 FG LH 30 0.53 0.1A EFG 0.97 0.19 EFG DBS 23 0.53 0.17 EFG 1.06 0.22 EFG SP 6 0.33 0. 33 FG 0.33 0.21 FG WC 21 0.2A 0.10 FG 0.00 0.00 G MR2 22 0.18 0.08 FG 0.09 0.09 G CC 7 0.1A 0. 1A FG 0.29 0.16 FG MCS 27 0.11 0.06 FG 0.19 0.08 6 PV 15 0.07 0.07 G 0.17 0.09 G ST 13 0.00 0.00 6 0.00 0.00 G MY 9 0.00 0.00 6 0.11 0.11 G 132 J. Res. Lepid. Table 6. The weeks of each month on which L — larvae or A — adults were observed at the 4 different San Bernardino Mt. sites. MAY JUNE JULY AUGUST SEPTEMBER OCT. 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 AC: 1983 AA L L L BP: 1983 L 1985 L DB: 1983 AL A AL AL A 1984 AL AL 1985 A AL MR: 1983 A A AL AL L L 1984 L L 1985 A A L during May and June and again in September after summer rains. Eriogonum wrightii subscaposum, which is common and widespread above 5000’ in the San Bernardino Mountains, blooms from August to October. This plant is utilized by E. enoptes at BP and DB. Eriogonum wrightii trachygonum , which is common mostly below 5000’, also blooms from August to October. This plant is utilized by E. enoptes at MR. Eriogonum elongatum , a common species below 5000’, especially along the southern slopes of the San Bernardino Mountains, blooms primarily from August to October. It is the major host of E. enoptes at MR. The presence of the aforementioned hosts does not always correspond to the presence of E. enoptes. At MR no E. enoptes larvae were found on E. davidsonii ; nor were any found on E. wrightii at AC. Both of these plants are common at many sites in the San Bernardino Mountains where E. enoptes has not been found. Eriogonum elongatum and E. nudum , which is utilized by E. enoptes in the adjacent San Gabriel Mountains, are widespread and abundant at many sites along the southern slopes of the San Bernardino Mountains yet no populations of E. enoptes are known to utilize them there. Eriogonum umbellatum Torr., another common species above 6000’, is a preferred host for some populations of both Euphilotes battoides (Behr) and E. enoptes , but is utilized by neither in the San Bernardino Mountains. In fact, E. enoptes larvae from MR die when fed the local E. umbellatum munzii (Reveal) as do larvae of E. battoides glaucon (Edwards) (J. F. Emmel, personal communication), which utilizes another subspecies of E. umbellatum in the Sierra Nevada. 25(2): 121-135, 1986(87) 133 Observations of larval behavior were noted for several populations of E. enoptes. In the field, larvae from LH, MR, SP, WW, AC, and PM, often tie together dry and partially consumed flowers to create loose shelters within the host inflorescence. At Morongo Canyon (MCS) mostly first and second instar larvae, rather than later instars, as expected, were found on the host E. elongatum from September 15 to November 24, 1984. Under laboratory conditions, these larvae ma¬ tured to third and fourth instars. The larvae fed nocturnally and remained concealed at the base of host plants by day; they made no floral shelters. Field evidence (the lack of mature larvae on blossoms) suggests that larvae from CC and PV may have a similar behavior. Discussion Adult eclosion has two determining factors: conditions which termin¬ ate diapause, and thermal summation for subsequent development. Pupae which break diapause simultaneously may eclose at different times in the field due to different temperature regimes (in their environments). Euphilotes enoptes pupae from some populations break diapause in response to warming after cold treatment, while others may break diapause in response to other conditions, perhaps indepen¬ dent of cold treatment. When reared under the same conditions, early- flying populations eclose soon after the end of refrigeration, indepen¬ dent of the time of year, while late-flying populations do not eclose until several weeks or months later. Both types of diapause occur in the San Bernardino mountains and one population is facultatively multi voltine. A high temperature regime (27— 35°C) during development is more conducive to breaking diapause (or inhibiting its induction) in E. enoptes than is a lower temperature regime (22-27°C). This has been shown in other insects as well (Chapman, 1971). Other E. enoptes populations (PV and WA), in addition to the Doble population, appear to be at least bivoltine, as indicated by their pupae failing to diapause when kept at 22— 27°C. Conditions which Induce diapause in the multivoltine DB population are not known, but probably are related to host plant condition and/or moisture stress. Some E. enoptes pupae can also diapause for more than one year. Termination of diapause in these populations may be affected by rainfall patterns, temperature, and/or photoperiod. Various populations of E. enoptes utilize several species of Eriogonum in the subgenera Eucycla and Ganysma (Reveal, 1969). In the San Bernardino Mountains E. enoptes utilizes at least four Eriogonum species belonging to both subgenera and often more than one in a given locality. However, not all available hosts are utilized nor are the acceptable hosts utilized wherever they occur. Thus, the distribution of E. enoptes in this area is largely independent of availability of hosts. Euphilotes e. mojave may have the most restricted diet of the E. 134 J. Res. Lepid. enoptes subspecies. So far it has been found only on E. pusillum and E. reniforme even at sites where other hosts occur, as at BG where E. davidsonii grows along side E. pusillum. Larvae of E. e. mojave from MCA collected on E. pusillum , which would switch to E. reniforme in the lab would not feed on E. davidsonii or E. nudum. Yet larvae of E. enoptes from MR2 collected on E. elongatum easily switched to E. davidsonii, E. pusillum and E. microthecum. First and second instar E. enoptes remain within host plant infloresc¬ ences. Third and fourth instars, from some populations, often create shelters by tying blossoms together with silk, where they remain until mature or until food is depleted. Older larvae of E. e. dammersi at Morongo Canyon do not make floral shelters but probably conceal themselves at the plant base by day and feed on blossoms nocturnally, or crepuscularly. Similar behavior may also occur in some other popula¬ tions of E. e. dammersi. Setation patterns of mature larvae vary among populations of E. enoptes. These patterns are relatively constant from generation to generation and offer reliable characters for comparing different popula¬ tions. Many populations (E. e. bayensis, E. e. dammersi, E. e. enoptes, E. e. smithi, and E. e. Tildeni) have very few prominent setae. E. e. enoptes larvae have few prominent setae dorsally and dorso-laterally, but a relatively large number laterally on all segments. Larvae of E. e. mojave have far more prominent setae than the other subspecies in nearly all body regions. This permits them to be readily distinguished from the others. Larvae of populations of E. enoptes in the San Bernardino Mountains more closely resemble setal patterns of E. e. dammersi than E. e. mojave , both of which occur nearby. At sites where both E. e. mojave and another subspecies of E. enoptes occur, as at Mojave River Forks and Morongo Canyon, there is no apparent dilution of larval characters in either. Therefore, it seems unlikely that any gene mixing occurs. Of course, in both cases their flight seasons are widely separate. The similarity in larval setation of the San Bernardino Mountain populations suggests that these are closely related. San Bernardino Mountains populations of E. enoptes are more-or-less intermediate in setal characters between the E. e. dammersi populations to the east and populations of E. e. bayensis, E. e. smithi, and E. e. tildeni to the west. General Conclusions Larval hostplant and setation characters can be utilized to consistent¬ ly separate certain populations of subspecies of E. enoptes from others. Among the other subspecies (at least E. e. dammersi, E. e. smithi, E. e. tildeni, and the San Bernardino Mountains populations) host plant specificity and seasonal flight period are variable from location to 25(2): 121-135, 1986(87) 135 location. The plasticity of these characters may render them unreliable as indicators of subspecific relationships. Acknowledgments. The authors wish to express their gratitude to John F. Emmel for supplying E. enoptes from Upper Centennial Spring and his know¬ ledge of Euphilotes populations. Particular thanks also to Andrew C. Sanders for plant identifications. Thanks to David Wright for careful reading of the manuscript and helpful suggestions. Rudolf H. T. Mattoni also provided helpful suggestions and guidance to the Santa Paula site. Literature Cited ARNOLD, R. A. 1983. Conservation and management of the Endangered Smith’s Blue Butterfly, Euphilotes enoptes smithi (Lepidoptera: Lycaenidae), J. Res. Lepid. 22: 135-153. CHAPMAN, R. F. 1971. The Insects Structure and Function, American Elsevier Publishing Company, Inc. N. Y., 819 pages. COMSTOCK, J. A. 1966. Life History of Philotes mojave (Lepidoptera: Lycaenidae). Trans. San Diego Soc. Nat. Hist. 14: 133-136. COMSTOCK, J. A. & C. HENNE. 1965. Notes on the Life History of Philotes enoptes dammersi. Bull. So. Calif. Acad. Sci. 64: 153 — 156. HINTON, H. E. 1946. On the Homology and Nomenclature of Lepidopterous Larvae, with Some Notes on the Phylogeny of the Lepidoptera. Trans. Roy. Ent. Soc. London 97: 1 — 37. LANGSTON, R. L. & J. A. COMSTOCK. 1966. Life History of Philotes enoptes hayensis (Lepidoptera: Lycaenidae). Pan Pac. Ent. 42: 102-108. MILLER, L. D. & F. M. BROWN, 1981. A Catalogue/Checklist of the Butterflies of America north of Mexico. Lepid. Soc. Mem. no. 2. REVEAL, J.L. 1969. A Revision of the Genus Eriogonum (Polygonaceae). Ph. D. dissertation, Brigham Young Univ., Univ. Microfilms, Inc., Ann Arbor, Michigan, 546 pages. SHIELDS, 0. 1975. Studies on North American Philotes (Lycaenidae) IV. Taxono¬ mic and Biological Notes, and New Subspecies. Bull. Allyn Mus. 28: 1-30. - , 1977. Studies on North American Philotes (Lycaenidae) V. Taxono¬ mic and Biological Notes, Continued. J. Res. Lepid. 16: 1-67. Journal of Research on the Lepidoptera 25(2): 136-145, 1986(87) A New Genus and Species from the Southwestern United States (Noctuidae: Acontiinae) Richard M. Brown 323 Calvert Ct., Antioch, California 94509 Abstract. The species albiciliata Smith (1903) is removed from the genus Cobubatha Walker 1863, and made the type of a new genus, Allerastria. The genitalia of albiciliata are described, apparently for the first time. Three new taxa are described in the new genus, two subspecies of albiciliata ( paula , from the San Joaquin Valley, Califor¬ nia, and chacoensis, from the Chaco Canyon National Monument, New Mexico) and a new species ( annae from southern California). All species and subspecies are figured and diagnosed. Introduction In 1977 I took a long series of Cobubatha albiciliata (Smith) at the western mouth of Titus Canyon, on the valley floor of Death Valley National Monument. Mixed with this series was a short series of moths that could not be assigned to albiciliata and is described as new. With the borrowing of additional material and the investigation of the other species of Cobubatha a number of characteristics were found that separated albiciliata and the new species from Cobubatha. From the time albiciliata was described in 1903 by Smith, it has had uncertain placement. Smith (1903) stated “the species is not really an Yrias , but it resembles that genus in general form and may remain here until further material makes a better reference possible.” Barnes and McDunnough (1912) in their description of the synonym bifasciata were not certain of the generic assignment when they placed “the species for the present in Eustrotia” McDunnough described Nerastria in 1937, and moved albiciliata to that genus in 1938. Most recently Franclemont and Todd, in R. W. Hodges et al (1983) placed Nerastria as a synonym of Cobubatha Walker (1863). Based on the charaters described below, I feel that albiciliata and the new species should be assigned to a separate genus. Allerastria R. M. Brown, new genus Type species: Yrias albiciliatus Smith, 1903 Adult. Head with eyes of both sexes large, round, greater in diameter than width of front; front (fig. 2, 4) with a rounded projection, scaling giving front a squared appearance when viewed laterally; labial palpi upturned, second segment nearly straight, paralleling front, third segment short, conical to middle of eye; antennae serrate in both sexes, males with ventral setae much 25(2): 136-145, 1986(87) 137 longer than in female, nearly equal to diameter of antennal shaft; male antennae with 50-57 segments, female 48-53 segments. Thorax robust, fore tibia with epiphysis arising approximately one-third distance from basal end of tibia in both sexes; epiphysis shorter in females; metathoracic tibia of males slightly swollen with long hair scales on the inner surface forming a vestigial hair pencil, both pair of spurs present. Abdomen slender in males, more robust in females, extending beyond hind wings, abdomen without dorsal tufts. Fore wings longer than wide, apex angulate, outer margin rounded; Sc free, ending seven-tenths from base; Ri from discal cell; R3 anastomising with R4 forming an accessory cell; R2 from top of accessory cell; R5 from apex of accessory cell; Mx from bottom of accessory cell widely separated from M2 and M3; end of discal cell open; M2 and M3 from lower angle of discal cell, M3 closer to M2 than to CuAL; CuA2 arising from beyond middle of cell; 1A straight and free. Hind wing full and without angulation; Sc and R confluent for one-fourth of length of cell; R and Mx separate from upper angle of cell; M2 and M3 from lower angle of cell, M3 closer to CuAi than to M2; M3 and CuAx occasionally stalked; CuA2 arising from middle of cell; cell open; 1A and 2A straight and free. Male genitalia (fig. 1, 3). Valvae simple long, slender with parallel sides, length 6—7 times width; inner surface of valvae moderately setose; uncus long, tubular, down hooked, with lateral setae; scaphium long, slender, stalk with a widely bifurcated tip, area between tips roundly concave, scaphium length .55— .65 mm; juxta with basal margin deeply excavated; saccus, variable; aedaeagus 1.0- 1.3 mm long, .25 -.33 mm wide. Female genitalia (figs. 5, 6). Corpus bursae round to oval, membranous, without signum; ductus bursae membranous, short; ostium with sclerotized collar; posterior apophyses shorter than anterior apophyses, length 0.4-0. 5 mm to 0.75-0.8 mm; ovipositor lobes well developed, 0.7 -0.8 mm in length, densely covered with setae. Diagnosis. The species of Allerastria can be separated from those of Cobu- batha Walker (1863) by a number of characters. In both sexes of Allerastria the front is projecting greatly beyond the eyes, but less so than in the genus Amiana Dyar (1904). The third segment of the labial palpi in Allerastria are short and conical, slightly longer than their diameter, where as in Cobubatha the length of the third segment is at least twice the diameter. The male antenna has the ventral setae much longer than in Cobubatha. The species of Allerastria, similar in color and maculation, have the median areas of the fore wings predominantly cream- white and differ from much of the Cobubatha species which have a dark brown median band on the fore wing. The male genitalia of Allerastria have the valvae long, narrow with parallel sides, in Cobubatha these structures broaden toward the apex and are not quite so long. Distribution. Allerastria flies in the deserts of the southwestern United States, and the San Joaquin Valley, California. The majority of specimens used in this study are from southern California. The moths are on the wing from April through September. Etymology. Allerastria is to read as another (alios) erastria and is feminine. 138 J. Res. Lepid. Figs. 1-2, Male genitalia. Allerastria albiciliata. la main body of genitalia; 1b aedaeagus; 1c degrees of uncus downflex. 2a head, left lateral; 2b front. Figs. 3-4, Male genitalia. Allerastria annae. 3a main body of genitalia; 3b aedaeagus; 3c degrees of uncus downflex. 4a head, left lateral; 4b front. Figs. 5-6, Female genitalia. Fig. 5 Allerastria albiciliata. Fig. 6 A. annae. 25(2): 136-145, 1986(87) 139 Fig. 7-14, Adults. Fig. 7 Allerastria albiciliata albiciliata, male. Fig. 8 A. a. albiciliata, female. Fig. 9/4. a. pau/a, male, Holotype. Fig. 10/4. a. paula, female. Allotype. Fig. 11 A. a. chacoensis, male, Holotype. Fig. 12/4. a. chacoensis, female, Allotype. Fig. 13/4. annae, male, Holotype. Fig. 14 A. annae, female, Allotype. Illustration 2 x natural size. 140 J. Res. Lepid. Key to Species, BASED ON MACULATION la) Underside wings prominently bicolored, basal half cream-white to white, distal half lead-gray to tan . 2a. lb) Underside wings not as in la, uniformly colored cream-white . . annae n. sp. 2a) Small, fore wing length 9.0—10.0 mm, heavly suffused with brown scaling, from the San Joaquin Valley (Tulare Co.), California . albiciliata paula n. subsp. 2b) Larger, fore wing length 9.0-13.0 mm, from the southwestern (Arizona, Southern California, Nevada, and New Mexico) United States. Upper side fore wing white, tan or pink . 3a. 3a) Upper fore wing (length 9.0—11.0 mm) may or may not have a pink flush. Upper fore wing crossed with basal sub-terminal lines lead-gray. Median area varies from white to pink. Southern California, western Arizona and Nevada . . albiciliata albiciliata. 3b) Upper fore wing (length 11.0-13.0 mm) cream-white to tan with light brown scales scattered over wing. In well marked specimens brown scaling forms indistinct cross lines. No lead-gray color or contrasting median area as in 3a. Northwest New Mexico . . albiciliata chacoensis n. subsp. BASED ON MALE GENITALIA la) Uncus down flexed approximately 163° (fig. lc) with heavy setae laterally arranged; juxta basal margin deeply excavated to near caudal margin (fig. la) . albiciliata. lb) Uncus down flexed approximately 50° (fig. 3c) with fine setae laterally arranged; juxta with basal margine deeply excavated to caudal margin (fig. 3a) . annae n. sp. BASED ON FEMALE GENITALIA la) Ductus bursae nearly twice as long as wide; corpus bursae regularly oval; ductus seminalis rapidly narrows to straight tube leading to the right . albiciliata. lb) Ductus bursae short, wider than long; corpus bursae irregularly round with right caudal quadrent roundly projecting, ductus seminalis broadly based and tapers to a spiraled tube leading to the right . annae. Allerastria albiciliata albiciliata (Smith) new combination (figs. 1, 2, 5, 7, 8) Yrias albiciliatus Smith, 1903, Trans. Amer. Entomol. Soc. 29:215-216. (TL: Yuma Co., Arizona) 25(2): 136-145, 1986(87) 141 Eustrotia bifasciata Barnes & McDunnough, 1912, Canad. Entomol., 44:218. (TL: La Puerta Valley, San Diego Co., California) Nerastria albiciliatus, McDunnough, 1938, Check list of Lepid. of Canada and U. S. Amer. Part 1, Macrolepidoptera, Mem. So. Calif. Acad Sci., 108. Cobubatha albiciliata, Franclemont & Todd, in R. W. Hodges et al., 1983, Check List of the Lepidoptera of America North of Mexico, p. 132. The description of albiciliata by John B. Smith and the description of Eustro¬ tia bifasciata by Barnes and McDunnough are sufficient to make further descriptions of maculation unnecessary. However, the genitalia of either sex has not been described. Male genitalia (fig. 1). Valvae long, narrow, with slight constriction midway between base and apex, costa with long setae, apex rounded and clothed with long setae, sacculus and median ridge with short setae; saccus tapering to a blunt point; uncus (fig. lc) sharply down-flexed with apical half slightly swollen, terminating in a sharp spine, long prominent setae laterally arranged; scaphium with apex widely bifercated, slightly concave between tips, narrow¬ ing to a long thin shaft to point of attachment; juxta with sides incurved, basal margin deeply excavated nearly dividing juxta in two, caudal margin straight; aedaeagus (fig. lb) robust, armed with flat, narrow chitinous structure. Female genitalia (fig. 5). Corpus bursae oval, without signum or other structures; ductus bursae membranous; ostium with narrow, lightly chitinized band at caudal opening of ductus bursae; bursae seminalis arising ventrally from moderately broad base, narrowing rapidly to a straight tube; posterior apophyses 0.5 mm long, anterior apophyses 1.0 mm long; ovipositor lobes well developed and densely covered with setae. Distribution. Arizona: Coconino Co.; Maricopa Co.; Pima Co.; Santa Cruz Co.; Yuma Co. California: Imperial Co.; Inyo Co.; Kern Co.; Mohave Co.; Riverside Co.; San Bernardino Co.; San Diego Co. Nevada: Clark Co. Remarks. Two hundred forty seven specimens (78 males and 169 females), 6 genitalic and 3 wing slides have been studied. Allerastria albiciliata paula R. M. Brown new subspecies (figs. 9, 10) Male. Head cream-white with scattered gray-tan scales; antennae tan and gray checked. Thorax ventrally cream, dorsally cream with scattered gray scales; prothoracic tibia gray with middorsal cream-white spots, scales over epiphysis cream- white; tarsi gray with cream- white at joints; mesothoracic leg with longest spur 2.5 times longer than shorter; tibia dorsally gray with tufted middorsal cream- white scales; tarsi gray with cream-white at joints; metathor- acic leg with two pair of spurs, longest upper spur 2.0— 2.5 times longer than shorter; tibia cream- white with very little gray; tarsi light gray, cream- white at joints. Abdomen ventrally cream-white, dorsally gray with cream-white bands. Fore wing above with basal area cream- white with heavy overlay of gray scales; t.a. line sinuous, tan with basal border heavy gray, distal border less defined by gray scales; median area light tan, median band represented by a gray square on costa; t. p. line light gray, lightly outlined by dark gray scales, with white scales on the veins; subterminal area light tan; fringe with spatu- late scales of various length, concolorous with subterminal area; fore wing 142 J. Res. Lepid. below shiny with inner half cream- white, outer half gray- white. Hind wing above, shiny, with inner half cream- white, outer half gray; below marked as above but with more color contrast. Fore wing length 9.0-10.0 mm. Female. Slightly larger than male, with maculation the same. Genitalia. As in nominate subspecies. Types. Holotype, male, California (Tulare County), Exeter, 7 -VI-1924, (R. M. B. slide #309). Allotype, female, California, (Tulare County), Exeter, 15-VI-? (R. M. B. slide #324). Paratypes, 3 males, same locality as types, data as follows; 5-IX-1924, R. Dodge; 18-VII-1924, R. Dodge; 15-VI-1940, E. A. Dodge. The types and paratypes are in the collection of the California Academy of Sciences, San Francisco, California. Distribution. This moth is known only from the type locality and is on the wing from June to early September. Remarks. Five specimens (4 males and 1 female), 4 genitalic slides and 1 slide of the wings have been studied. Subsequent to the photograph (fig. 10) of the allotype, an Anthrenus sp. (Coleoptera: Dermestidae) devoured the thorax leaving only the head, legs and wings. These parts have been attached to paper supports and are still useable for identification. The abdomen is mounted on a slide as noted above. No difficulty should be encountered in recognizing this subspecies; paula is the smallest and darkest taxon in this genus. The heavy brown overlay gives this moth a distinctive appearance. Etymology. I am calling this moth paula because of its small size. The name is feminine. Allerastria albiciliata chacoensis R. M. Brown new subspecies (figs. 11, 12) Male. Head cream- white with tan scales; antennae checked with tan; thorax cream- white, dor sally scattered with tan scales; prothoracic leg, tibia dor sally gray with cream-white band midlaterally, scales over epiphysis cream-white; tarsi gray with cream-white at joints; mesothoracic leg with longer spur 2.0 times longer than shorter spur; tibia dorsally scattered with gray scales, with middorsal cream-white tufted scales; tarsi gray with cream-white at joints; metathoracic leg with two pair spurs, longest spur 2.3 -2.5 times longer than shorter spur; tibia cream- white with very little gray; tarsi dorsally gray with cream-white at joints; abdomen ventrally cream- white, dorsally with faint tan bands. Fore wing above, all lines weakly represented, ground color cream- white with mixture of light and dark tan scales; costa with seven variable dark-tan checks; basal and t. a. lines variably present; subterminal area with heaviest concentration of dark-tan scales; subterminal line present and scal¬ loped. Fringe cream-white with long spatulate scales, dark-tan checks at end of veins. Fore wing below, shiny with inner half cream- white, outer half tan. Hind wing below marked as fore wing below. Fore wing length 11 — 13 mm (Holotype, 11 mm). Female. Simular in size and color with maculation less distinct than in male. Genitalia. As in nominate subspecies. Types. Holotype, male, New Mexico, San Juan County, Chaco Canyon 25(2): 136-145, 1986(87) 143 National Monument, 2-VIII-1962, S. F. Wood. Allotype, female, New Mexico, San Juan County, Chaco Canyon National Monument, 3-VIII-1962, S. F. Wood, Paratypes, 3 males and 5 females. Same locality and collector as types, data as follows; lcf 8-VII-1962; lcf 1? 17-VII-1962; lcf 2$ 2-VII-1962; 1$ 3 -VIII- 1962; 1$ 6 -VIII- 1962. The types and paratypes are in the collection of the Los Angeles County Museum of Natural History, Los Angeles, California. Distribution. This moth is only known from the type locality. It is on the wing in July and August. Remarks. Ten specimens (4 males and 6 females), and two genitalic slides have been examined. This moth can be seperated from the other subspecies by the general distribu¬ tion of tan scales and extremely weak markings. The maculation of the fore wing is not divided into easly recognizable areas, although the tan scaling forms weak striations. Entymology. I have named this moth after the Chaco Canyon National Monument to honor and point out the valuable role the national park system plays in preservation of the wildlife resource. For with out this great system and the many dedicated people the rare and unusual would have been lost long ago. Allerastria annae R. M. Brown new species (Figs. 3, 4, 6, 13, 14) Male. Head (fig. 4) dirty cream-white; labial palpi upturned to above middle of eye with scattered lead-gray scales; antennae, lead-gray with white checks. Thorax dirty white with light pink tinge; collar and tegulae with most pink; prothoracic leg cream-white, tarsus lead-gray with white at joints, epiphysis three-fourths fore tibial length; mesothoracic leg marked as prothoracic leg; metathoracic leg marked as previous legs. Above fore wing with basal space bicolored, inner area concolorous with thorax, outer lead-gray; costal area above discal cell with two diffused white spots; t. a. line sinuous, lead-gray, basally edged with a few white scales, distally by rust-red; t. p. line lead-gray, basally shaded red, distally with lighter gray, t. a. line begins on costa, crosses to vein CuA1? accompanied by scattering of white scales forming a faint line, thence basad to median shade, turning then to inner margin; median area light gray-tan, divided in costal area by median shade. Subterminal line represented by white scales on veins, subterminal area tan with darker lunuals at vein ends. Fringe gray. Hind wing ground color concolorous to thorax, distal half with light gray shading. Fringe concolorous with thorax. Wings below un¬ marked, concolorous with thorax. Fringe on fore wing slightly darker than on hind wing. Female, similar to male in maculation, the markings less contrasting. Male genitalia (fig. 3). Valvae long, narrow with parallel sides; apex rounded, heavily clothed with setae on distal half, basally naked except for a small cluster of setae on low median ridge; saccus base square and one-third saccal width; uncus (fig. 3c) tubular, long, pointed, down-flexed, short fine setae laterally arranged; scaphium with apex widely bifurcated, deeply concave, narrowing to a long narrow shaft; juxta bifid, deeply excavated basally and apically appearing as two triangular units narrowly united; aedaeagus (fig. 3b) 144 J. Res. Lepid. robust, less than combined length of tegumen and saccus, without internal armature or spicules, posterior end produced into a shelf-like projection, dorsal surface heavily chitinized with short stout setae. Length three times diameter. Female genitalia (fig. 6). Corpus bursae irregularly oval with a membranous projection on right caudal quadrent; ductus bursae short and asymmetrically placed; ostium with narrow chitenized band separated from caudal opening of ductus bursae; ductus seminalis arising ventrally from a very broad base and narrows to a spiraling tube. Apophyses short, not reaching ostium; ovipositor lobes well developed and densely covered with setae. Fore wing length in holotype, 13.5 mm; allotype 13.0 mm; paratypes, 12-13 mm. Types. Holotype, male and Allotype, female, California, Inyo County, Death Valley National Monument, western mouth Titus Canyon, elevation 1000 ft. (305 M), 6-IV-1977, Richard M. and Paula J. Brown. The genitalia of the holotype is mounted on R. M. B. slide #302, and the allotype is on R. M. B. slide #318. Paratypes, 2 males, 7 females, same locality and data as holotype. California: Inyo County; 1$ Furnace Creek Death Valley, 10-IV-1931, G. Willett; 1$ Triangle Springs Death Valley, 11-12-IV-1942, G. Willett; 1$ Mesquite Springs Death Valley, 19-22— IV-1943, G. Willett; Riverside County; lcf Palm Springs, 21-IV-1920; San Bernardino County; 1§ Baldy Mesa, 9-IV-1932, J. A. Comstock; 12 near Barstow, 10-V-1940, C. Ingham; lcf Yermo, 28-VI-1938; lcf Yermo, 7-IV-1939. The National History Museum of Los Angeles County, California will receive eight paratypes, one pair of paratypes to the National Museum of Natural History, Washington, D.C. The balance of the type material will be in the collection of the California Academy of Sciences, San Francisco. Distribution. The desert regions of southern California. As more specimens are taken, it probably will be found to fly sympatrically with A. a. albiciliata. On the wing from April through June. Remarks. Twenty two specimens (6 males and 16 females) and 8 slides of the genitalia were studied. This is an extremly variable species in maculation. It varies from nearly immaculate individuals to those like the well marked holotype. The first color to be lost is the rust red, the lead-gray band then fades but never completely disappears. Allerastria annae is very close to albiciliata in maculation but lacks the sharp definition of pattern found in albiciliata. Allerastria annae also has much less contrast between the light and dark areas of the hind wing than is found in albiciliata. The male genitalia of annae can be told from those of albiciliata by the uncus being down flexed approximately 50° (fig. 3c) . In albiciliata the scaphium has a long shaft with parallel sides, whereas in annae the sides gradually diverge to a widely bifurcated apex. Also annae can be separated from albiciliata by the deep basal excavation of the juxta found in annae which gives an appearance of two triangular units losely united. Etymology. This moth is named after my wife, Ann, who has given so much in support and understanding. The name is feminine. Acknowledgments. Special thanks to Robert W. Poole, National Museum of Natural History, for reviewing my specimens and the original manuscript, his confirmation of my conclusions and critical comments are greatly appreciated. I 25(2): 136-145, 1986(87) 145 am also indebted to Julian P. Donahue, National History Museum of Los Angeles, and to Paul H. Arnaud, Jr., The California Academy of Sciences, for without their generous loans of specimens this paper would have been much less comprehensive. Thanks to my wife, Ann, who produced the photographs used in this paper, and also to two anonymous reviewers who caused a great deal of work. Literature Cited BARNES, WM. & J. MCDUNNOUGH 1912. New Noctuid Species. Canadian Entomol. 44:216-218. DYAR, HARRISON G. 1904. Additions to the List of North American Lepidoptera, No. 2. Proc. Entomol. Soc. Wash. 6:103—117. (Orig. descrip of Amiana on p. 104) FRANCLEMONT, JOHN G. & E. L. TODD IN RONALD W. HODGES, ETAL. 1983. Check List of the Lepidoptera of America North of Mexico, p. 132. MCDUNNOUGH, J. 1937. Notes on North American Noctuid Genera. Canad. Entomol. 69:40-47, 58-66. (Orig. descrip, of Nerastria on p. 65.) - 1938. Check List of the Lepidoptera of Canada and the United States of America, Part 1, Macrolepidoptera. Mem. So. Calif. Acad. Sci., 1:108. SMITH, JOHN B. 1903. New Noctuids for 1903, No. 4, with notes on certain described species. Trans. Amer. Entomol. Soc. 29:191-224, Plate III. Journal of Research on the Lepidoptera 25(2): 146-148, 1986(87) Observations on Problems, bulenta George O. Krizek 2111 Bancroft Place, N.W., Washington, D.C. 20008 Paul A. Opler 5100 Greenview Ct, Fort Collins, CO 80525 The rare skipper, Problema bulenta (Boisduval and LeConte), is uncommonly observed and has never been photographed in nature. Here we report behavioral and flower use observations made in July, 1984 at Blackwater National Wildlife Refuge, Dorchester County, Maryland. Previous nectar utilization has been reported by Jones (1926), who observed the species visiting pickerelweed (. Pontederia cordata ) in North Carolina and by Covell and Straley (1973), who reported bulenta visiting swamp milkweed ( Asclepias incarnata) in Virginia. At the Maryland locality P. bulenta was fairly abundant and was observed by several persons, and as a result, more flower visiting observations were possible. The primary nectar source at Blackwater NWR was button- bush ( Cephalanthus occidentalis) . Secondary nectar sources observed were the two previously reported, swamp milkweed and pickerelweed, as well as red clover ( Trifolium pratense ) and dogbane ( Apocynum cannabinum) (J. Fales, W. Grooms, R. Smith, pers. comm.). In 1984, adults were seen from June 20 to July 14 — being most common later in the flight period. Our observations and those of J. Fales indicate that females are seldom seen at flowers and vary from 4 to 10 males seen for every female. Females may spend more time in the immediate vicinity of their host — suspected to be a large grass. Adults fly very close to the water at all times; most were seen within 30 cm. The highest above water was one seen 1.5 m. The flight is rapid, strong and noisy. Individuals seem to return again and again to the same area of a nectar plant. When visiting buttonbush these skipper usually visit low flowers preferring to rest on the under surface of the inflorescence, a site that is often in shadow. At the Maryland locality the wind blows almost constantly in variable gusts. The butterflies at flowers are constantly turning and moving from flower to flower. Flower visitation is from 10.00 to 15.00 hr; after that time the butterflies are no longer to be found. Other butterflies sharing the buttonbush flowers with P. bulenta were Epargyreus clarus, Erynnis horatius, Ancyloxipha numitor, Wallengre- nia egeremet, Poanes viator (abundant), Phyciodes tharos, and Vanessa virginiensis. 25(2): 146-148, 1986(87) 147 This skipper is very difficult to photograph. One must go into the river’s water or stand at its muddy edge. This together with the almost constant wind, and the low, nervous flight of the insect makes such attempts trying at best. The habitat is similar to that found along the Chickahominy River in Virginia (Coveil and Straley, 1973). It seems likely the Maryland colony is univoltine as suggested by Opler and Krizek (1984). Further investigation is necessary to reveal the host plant and reproductive biology of this uncommon insect. Acknowledgements. We thank John H. Fales, Calvert, Maryland; Richard Smith, Baltimore; and William C. Grooms, Tysons Corner, Virginia, for sharing their field notes on this butterfly. Literature Cited COVELL, C. V., JR. & G. B. STRALEY. 1973. Notes on Virginia butterflies, with two new state records. J. Lepid. Soc. 27: 144-154. JONES, F. M. 1926. The rediscovery of Hesperia bulenta Bdv. and LeC., with notes on other species (Lepidoptera: Hesperiidae). Ent. News 37: 194-198. OPLER, P. A. & KRIZEK, G. O. 1984. Butterflies east of the Great Plains: an illustrated natural history. Johns Hopkins University Press, Baltimore. xvii+294 pp. 148 J. Res. Lepid. Figures 1-2. Adult Problems bu/enta nectaring at buttonbush {Cephalan- thus occidenta/is) at Blackwater National Wildlife Refuge, Dorchester County, Maryland, on July 14, 1984. 1. Male, 2. Male. J. Res. Lepid. 149 Book Reviews THE BIOLOGY OF BUTTERFLIES. Symposium of the Royal Entomological Society of London, Number 11 (R. I. Vane-Wright and P. A. Ackery, Eds.). Academic Press, London, 1984. Purchase the book if you take your lepidopterology at all seriously. That addresses the overall picture quite sufficiently. Of course, the thing is most certainly not without its peculiar flaws, although in fairness this review often deals with shortcomings inevitable in any such tome covering comparably broad terrain. The butterfly symposium itself took place in September 1981 at the British Museum of Natural History, coincident with the eightieth year of Prof. E. B. Ford, in whose honor the event was dedicated. A star-studded cast of 44 signed off on the 33 articles ultimately shepherded to press by the capable hands of Vane-Wright and Ackery. The articles are organized into eight major areas of research on butterfly biology, which here seem most sensibly grappled with in the order in which they appear. I. Systematics. Since this is a volume on butterfly biology, it is not surprising that butterfly taxonomy is allotted only minimal page space. Indeed, Ackery sequesters such considerations in but the first 13 pages of the total 429, with a healthy chunk of that devoted to recounting and judging faunistic works of the world. All in all, we hear relatively little about butterfly faunistics as compared to butterfly taxonomy (and, of course, bloated and imbalanced work on limited faunas seems to be the relentless vogue in the latter). Not so with Ackery’s concise and bibliographic faunistic summaries — hats off to him for skipping the pedestrian, and pointing the way into an important yet often neglected literature. II. Populations and Communities. This section contains two lengthy and two very brief articles. An Ehrlichian overview of population structure sensu strictu is both expected and appropriate in a volume of this stature, since the Stanford studies are among the forerunners. I’ve read essentially everything the Ehrlich group has put out over the years, and this is one of the most readable and widely appealing of their available reviews. More is known about population structure in Euphydryas than in most other species, and it is sobering indeed to hear Ehrlich berate his favorite creatures, and call for studies on a less biased taxonomic sampler of butterflies. Gilbert then expands the focus to entire butterfly communities; given Ehrlich’s caution, it is no surprise that generalizations are even fewer here. In short, the number of possible explanations for observed patterns increases explosively as one moves from single to multiple-species studies (infrataxono- mic differences even notwithstanding). As a synopsis of this still nascent field Gilbert’s article is fine. A gem within it is the paraphrasing of Munroe, who in his thesis had written the kernel of island biogeography well before it was popularized by MacArthur & Wilson’s subsequent monograph. However, Gil¬ bert quotes Munroe to establish a sad point, namely that “the slow progress of butterfly ecology [is because! ... it has often been an afterthought of systema¬ tic or genetic studies.” 150 J. Res. Lepid. The final two short articles in this section do not belong. Pollard’s simply doesn’t begin to do justice to the important field he has helped to engineer. Read his journal articles on relative abundance instead, and the terminal paper in the symposium volume. Morton’s asks how the process of marking butterflies influences their subsequent activity. While there has never been much doubt that mark effects occur commonly, there are also few published studies addres¬ sing the issue. Morton’s paper is generally helpful in this latter regard, but his data are useful only insofar as one tolerates the failure to measure catchability differences, and other factors central to analysis of recapture probabilities (see below). III. The Food of Butterflies. This third section mirrors its predecessor in having two long and two short articles. Chew & Robbins lead off with one of the long ones, on egg-laying in butterflies. Their article embraces a large literature, hitting subtopics as diverse as oogenesis, selecting oviposition sites, and the evolution of oviposition specificity. Give them an ‘E’ for effort. It is partly the magnitude of their selected topic (too broad for one article), but mostly their choppy prose and superabundant citations which run the article aground. Large segments of the text are choked with 2-3 times as many references as necessary, and accordingly this is one of the more difficult to read among the symposium articles. Scan through it for the goodies to your liking. Their final section is probably the most provocative, and ushers in the notion of ‘large-scale evolutionary jumps,’ a subject taken up independently in other contexts by other authors in the volume. Singer, whose writing is vastly clearer, addresses a restricted but closely related array of topics. He first reviews host discrimination by females, and then turns to the consequences of female choice upon (the essentially ship¬ wrecked) larvae. A trademark of Singer’s is careful scrutiny of intervening variables — what might loosely be described as the dozen or so factors you couldn’t measure, but which your critics seize upon with glee. His trademark is evident throughout the latter half of this article (e.g., pages 85-87). It makes for tempered discussion, and, consequently, good reading. At the end of this chapter are two more brief articles. Courtney’s is scarcely a page, is merely a listing of homilies about habitat and footplant selection, and again does not belong in the volume. My advice is the same as for Pollard’s effort in Chapter II: go read Courtney’s fine original research papers instead. Edgar’s short data paper marshalls believable evidence that plants in the family Parsonsiae represent ancestral foods for danaines and heliconiines. IV. Predation, Parasitization, and Defence. This section contains a smatter¬ ing of articles dealing with threats to butterflies — who eats them, where, when, and why, and the consequences in evolutionary time. Dempster asks the damaging question: what in fact do we know of the natural enemies of Lepidoptera themselves (cf. the abundant indirect evidence of their effects)? The essential lesson from his lead article is straightforward, and can’t be emphasized enough — we know depressingly little about the influences of natural enemies on lepidopteran populations in the field. Lane’s tantalizing short second article, on ectoparasitic midges on butterflies, only reaffirms Dempster’s point (an excellent parallel treatise to Lane’s is Treat’s book on moth mites). In sum, our understanding of predators and parasites remains in ‘seek and describe’ mode. J. Res. Lejrid. 151 These tentative articles give ground to Brower’s methodical and exacting dissection of lepidopteran chemical defense. The longest of the Symposium articles, it is also among the best, a basic and refreshing subplot within it being re-categorization of the myriad terms applied in the literature on chemical defense and mimicry. Brower first establishes these theoretical constructs, and then marches into the fray and sorts through the booty of published, often fragmentary information. His differentiation between Class I (noxious) and Class II (innocuous) defensive chemicals becomes central to arranging the mess, and understanding the roles played by diverse assemblages of chemicals in the overall picture of chemical defense. Brower’s attention to spatial and temporal diversity in predator behavior is similarly welcome. As with Singer, Brower has the keen eye for how to deal with observed variation in a systema¬ tic fashion. Marsh et al.’s short article has the unenviable distinction of following Brow¬ er’s and preceding Turner’s. Their idea is laudable: test the anti-tumor action of various lepidopterous extracts. But the data are few (though interesting), and the recitations smack a little much of the narrow approach against which Brower just finished campaigning so successfully. Turner opts for the moderator’s stance, balancing opposing arguments while dismantling traditional dichotomies between Batesian and Mullerian mimicry. This he caps off with a lengthy and poignant discourse on saltational genesis of mimicry complexes. Indeed, by the end of the article, Turner has roamed fully into a general treatment of neo-Goldschmidtian punctuationalism (his tongue- in-cheek “evolution by jerks”). Throughout he draws upon the exemplar tropical heliconiine-ithomiine mimicry rings to bolster specific arguments. Turner’s temperance helps to unravel the various concepts, and his article can certainly stake its claim as an educated precis on mimicry. The reader must again endure two plus pages of the suboptimal after a masterpiece. In their introduction to the symposium, Vane-Wright et al. indicate that Gibson’s automimicry article “generated much discussion ... at the meeting.” Within the walls of said meeting is where this off-the-cuff model should have stayed to ripen a bit. Field workers with an accompanying feel for modeling will have little difficulty flagging the several tenuous assumptions and their scant support from data. For automimicry, start with Brower et al. (1967), et seq., and work yourself forward through the literature from there. V. Genetic Variation and Speciation. Leading off the second half of this volume is a variegated assembly of papers dealing with microevolutionary matters. Brakefield’s is the principal article, a classical British ecological genetic investigation of spotting pattern in satyrines. He devotes the first ten pages to detailed and data-intense elaboration on the classification and herita- bility of these demure undersurface spots, and then traverses a shopping list of selective pressures potentially responsible for the geographic and populational variation in spotting. The “boundary phenomonon” is certainly among the funkier discordant morphological patterns thrown by Maniola. The undersurface spotting regime of this butterfly shifts abruptly along a front only dozens of meters wide in southwestern England, and the front itself moves about in both time and space. There is still no overpowering explanation for this pattern, despite several decades of research. Brakefield somewhat belabors the ambivalent results with 152 J. Res. Lepid. this and other aspects of the Maniola story, but gets on track with his own thing — the spots as anti-predator devices, fluctuating selective pressures, and a healthy plea for populational work on the immatures. Brakefield’s plate of 119 ‘pressed’ Maniola on pages 174-175 is a welcome sight. (It reminds me of the cabinets full of quite prostrate Gerould Colias who have cohabited over the years with me in my niche in the museum). More importantly, of course, it is just about the only visible affirmation in the symposium that properly executed morphological work has always been and will always continue to be central to good evolutionary study. This too often gets billed as an antiquated tenet, in this heyday of gelled and pureed creatures (and narrowly defined biochemical jobs in evolutionary biology, and divested museum holdings). Kitching picks up on the subject of chopped butterflies in the second article, but only offers a breezy two cents’ worth on his electrophoretic work on danaids, and the possible concordance between his data and the morpholgical cladistic treatments of Ackery & Vane- Wright. Of what value is that? Granted, there are concerns other than review articles when one is completing a doctoral dissertation, but what an apparently lost opportunity for a coming lepidopterist to publish some hot-off-the-press research in a major tome. So, why? Gordon follows with a brief yet stimulating notion that mimicry (and possibly speciation) in each African Acraea is linked to dispersal, which in turn is linked to patterns of local extinction. Though he errs in the same manner as others throughout the symposium by equating differences in recapture probability entirely to one of its several confounding components (in this case, to dispersal), his continued pursuit of the subject should uncover some treasures. Pierce wraps up with another short piece, but one which strikes an appropri¬ ate balance between the data presented and the conclusions drawn. The suggestion that lycaenids speciate more rapidly because of low deme sizes and selection by females for both foodplants and ‘ant’ plants is most plausible. We can also now add lycaenid larvae to the growing list of bizarre entomological edibles. VI. Sex and Communication. This is the most mature chapter overall in the volume; and Silberglied’s is easily the best paper in this chapter, being both provocative and scholarly in content and well written. Smith’s is a close second, with the differences in approach and opinion between he and Silberglied appearing to be in large part semantic (or reflecting ‘taxonomic scale,’ cf. the Introduction). Silberglied treats us first to Darwin’s view on lepidopteran coloration and an accounting of its diversity, and then settles in on visual signals important in male and female communication, respectively. His take-home message is that female butterflies choose not on the basis of visible male colors, but rather on the basis of UV signatures and smells; he leaves us thinking along intrasexual lines for explanations of male butterfly colors. Smith analyzes mate selection in Danaus and Hypolimnas, offering one of the better blends of data and discussion in the volume. A main thrust of his is distinguishing between random preferential mating, and assortative preferen¬ tial mating. It remains to be seen whether Smith’s complex findings are generalizable throughout Lepidoptera. However, Smith takes high marks among the 44 authors for his frequent admonishments about the inadvisable lumping of heterogeneous sub-classes of data, and the certainty of subsequent errors in interpretation. J. Res. Lepid. 153 Three shorties follow. Platt et al. offer a short data paper conclusively showing lack of differential mate selection in tiger swallowtail morphs. Vane-Wright notes, in particular, how male narcissism might be a unifying force for apparently equivocal and/or puzzling results in butterfly ethology. Finally, Clarke talks a little about sex-ratio distortions in gypsy moth and Hypolimnus broods. The terminal paper, by Boppre, addresses the chemical aspects of communica¬ tion among butterflies. Admittedly, much of our general knowledge of this subject comes from experiments with moths — with butterflies, it has been largely anecdotes, some major works notwithstanding. Boppre dutifully covers his material (androconia, pheromones, associated behavior, etc.) but in at least twice the number of words required. VII. Migration and Seasonal Variation. Baker’s is the primary paper in this Chapter of only vaguely related articles, offering a glimpse into what governs the movement of butterflies. His temporal frame of reference is substantially longer (lifespan) than that typical of published work on butterfly movement (days or so), and this imparts to Baker a different and healthy perspective. Indeed, he treats topics such as direction ratios that often never surface in more conventional mark-recapture studies, and it is encouraging to see such initial advances in a curiously neglected field (after all, butterflies fly, and so why don’t we know more details about their travels than we do?). From flight we jump inexplicably into study of seasonal polyphenism, cast in the light of genetic assimilation. Shapiro reworks a theme he has been pub¬ lishing on vigorously for a decade, though in this paper he treats us to data from some new taxa. I agree with the editors that Shapiro’s effort is heroic despite equivocal results; he is also a bigger man than most to confess at the end that “if . . . [so], one need only invoke ordinary Darwinian selection to evolve polyphenism, and neither genetic assimilation nor anything more arcane is necessairly required.” See his Figure 27.5 if you have doubts as to the genetic (cf. environmental) basis of polyphenism. Chapter VII continues its schizophrenia by shifting to an illuminating short piece by Porter on larval basking, and its probable link with efficient digestive activity. Two more brief, descriptive polyphenism papers follow: McLeod on Precis', and Yata et al. on Pieris (the latter being of some intrigue since it treats polyphenism in the immature stages). Vni. Conservation. Pyle is the acknowledged popularizer of lepidopteran conservation worldwide, and an article from him is obligatory. Here he focuses on the recent eruptions of Mt. St. Helens in western North America, and the influence this literally earth-shaking event had on butterflies in the area. Glean the more general of Pyle’s points, since the data are necessarily scanty and inconclusive (the appropriate comparative pre-eruption lepidopteran re¬ search sadly doesn’t exist). The second article by Parsons examines the distribution, biology, and con¬ servation problems faced by the world’s largest birdwing butterfly. While Parsons talks about habitat loss (e.g., encroaching oil palm plantations) and factors affecting foodplant distribution, he unfortunately didn’t give air time to an intriguing, tested, and successful option — ‘butterfly farming.’ This novel technique simultaneously eases commerical demand for specimens without impacting wild populations, puts cash into the local economy, and (probably most importantly) fosters local interest and commitment to the conservation 154 J. Res. Lepid. ethic. Parsons’ repeated citing of internal agency documents on the matter of butterfly farming only makes one yearn further for an expose in the more accessible, true public record. This brings us to the ultimate paper in the volume. And it is the denoument — a masterly review by Thomas of lepidopteran conservation efforts in temperate countries. I can’t praise it enough. In fact, it is pointless for me to waste your time recapping it here, except to say that it shows pithy insight on all aspects of complex conservation issues, including: the acquisition and analysis of data on population changes, pinpointing the factors responsible for the observed changes, the associated political and sociological backdrops, and how these three avenues of inquiry are (or aren’t) translated effectively into day-to-day conservation practice. He certainly doesn’t shy away from flagging the dismal failures among the gamut of conservation attempts. Thomas really does have a handle on the ‘big picture,’ and I strongly urge that his paper be read carefully, with an eye toward integrating the lessons of the other 32 papers into the unifying framework offered in the 33rd. It is a fitting wrap-up indeed for this symposium — lepidopteran conservation efforts have been gaining momentum during their formative period of the past two decades, and stand to mature in their own right during the remainder of this century. Commentary. As you have gathered, the symposium articles fall broadly into two size (and content) classes — very brief reports of narrowly defined studies, and long review articles. The short reports are of inferior quality, and detract from the impact of the symposium volume as a whole. Why juxtapose notes of passing interest alongside more permanent, scholarly reviews? After all, we have journals for the express purpose of communicating such short notes (and journal referees to reject the bad ones). Obviously, I don’t feel these short notes at all served the editors’ stated intention (page 1) of amplifying or highlighting accepted dogma or difficulties. Nevertheless, there are other factors which editors must weigh (such as affording equal air time to all participants in joint ventures). While one may dislike the schizophrenia imparted by the short papers, Vane-Wright and Ackery can’t be held wholly accountable for problems inevitable when concate¬ nating as many as 33 papers. Choppiness is one such unavoidable problem. Leaving style aside, one large matter of substance glares at me through these several hundred pages of otherwise excellent lepidopterology. Why is it that mark-release-recapture takes it on the chin in this volume? I see much innuendo on supposed ‘problems with MRR,’ especially the business of marking itself, but little concrete offered in the way of justification, let alone alternative methodology. It is telling that authors in this symposium make essentially no mention of Tabashnik’s research on sulphur butterfly population structure, insofar as it applies to the theory and practice of MRR (nor do they speak of Begon’s 1979 book). Tabashnik’s 1980 paper, published in Oecologia, is the seminal work in recent years dealing with the partitioning of recapture probability into its biologically distinct components. Not one author attempted to break down recapture probability here, yet each tried to interpret recapture probabilities. There is little excuse for continued unthinking analysis of recapture probability as if it were a unified whole. Catchability and residence are different, they J. Res. Lepid. 155 combine to form recapture probability, and the distinction is paramount. Age structure is also easy to monitor (via wing wear), and it too is central, but again few authors bothered to report it. These are unsettling omissions. I get the impression that this pooh-poohing of MRR is traceable in large part to the intermittent reports detailing detrimental effects of marking (as championed in part by Morton here, and others else¬ where). Really, though, so what if marking effects exist? They’re present by definition. This begs for tempered investigation of their impact on populational parameters, not thoughts of rejecting MRR as the basis for measuring popula¬ tion size (the fact is that few have cared to ask critical questions in this area). Non-marking techniques certainly have their place, but they can’t yet supplant MRR, and doubtfully ever will. Lawrence F. Gall, Entomology Division, Peabody Museum of Natural History, Yale University, New Haven, CT 06511 USA BUTTERFLIES OF EUROPE, Volume 1, Concise Bibliography of European Butterflies, Otakar Kudrna, Editor, 1985. AULA-Verlag, Postflach 1366, D- 6200, Luisenplatz 2, Wiesbaden 1, West Germany. 447 pp. DM 248 (series subscription price DM 216). The Concise Bibliography is the first volume to be issued of the eight-volume BUTTERFLIES OF EUROPE series edited by O. Kudrna. Volumes 3-6 will discuss butterfly families, while vols. 2, 7 and 8 pertain to lepidopterology, ecology, and conservation of European butterflies. This initial volume consists of approximately 6000 bibliographic entries in alphabetical order by author and relating to various aspects of European butterflies. The entries are numbered sequentially with a few alphanumeric citations. The volume begins with a five-page Preface and a five-page Introduc¬ tion; an eleven-page Subject Index concludes the book. The citations included date from 1901 to 1983. References are provided to earlier bibliographies that cover publications prior to 1900. The editor states in the Introduction that this volume is designed “ . . . to serve the needs of all students of butterflies of Europe ...” regardless of their professional status. This work is not intended to be comprehensive. The citations listed were selected from a database of over 10,000 references com¬ piled during the preparation of the series as a whole. Major taxonomic papers are included along with citations to treatises about ecology, distribution, conservation, etc. Full citations are provided with the use of standard abbrevia¬ tions for journal titles. The Subject Index allows the user to locate references on the basis of family, geographic region, genetics, anatomy, and many other classifications. It does not, however, permit the user to locate citations by genus or species. This is perhaps the only shortcoming of the book, and to include an index to genera and species would have increased the size of this volume considerably. This book is well made with clear type and English text. It should be a valuable addition to the library of anyone interested in European butterflies. Clifford D. Ferris, Bioengineering Program, University of Wyoming, P. O. Box 3295 University Station, Laramie, Wyoming 82071. 156 J. Res. Lepid. THE BUTTERFLY GARDEN: Turning your Garden Window Box or Backyard into a Beautiful Home for Butterflies. Mathew Tekulsky. 1985. Harvard Common Press, Boston, ISBN 0916782- 69-7. Price $8.95 paperback. This charming and broadly informative book is an excellent piece for serious lepidopterists to give their inquiring friends. Tekulsky, a professional writer, has written the book well, and, as a consequence of research conducted in the course of writing, has in fact now become an amateur lepidopterist. Works like this represent a new generation of popular entomology through emphasis on observation and data-keeping, as opposed to the collecting mania and deadend museum ideology of earlier days. In terms of public awareness, such Weltanschauung should be cultivated as one’s social responsibility, in addition to the joys of a scientific hobby. Butterflies are increasingly recognized as indicators of a world environment that is going to hell. The factual material is general, as it must be, since the means of augmenting butterfly densities by gardening practise obviously differ between Los Angeles and Brooklyn. Nevertheless, a wide and thorough set of topics is covered from classification/life cycles, life zones, to courtship, migration, foodplants and nectar sources, and conservation. An emphasis on notetaking is a good point, and the bibliography and citation of resources are excellent. Bob Pyle wrote the well done introduction. ENTOMOLOGY OF THE CALIFORNIA CHANNEL ISLANDS: Proceedings of the First Symposium. Menke, A. S. and D. R. Miller. 1985. Santa Barbara Museum of Natural History, Santa Barbara, CA 93105. 178 pp. + 8 separate maps. Price $20. paperback. For anyone with an interest in island biogeography, as the concept is strictly applied to islands, the California Channel Islands are perhaps the best surveyed such areas in the world. This volume is the latest and most complete work to date. Of the animals censused, the Lepidoptera are the second best known group (after Orthoptera). Jerry Powell authored the principal paper on Lepidoptera. Although he defines the paper as a preliminary overview, it bears reading by all interested in patterns of distribution, citing problem areas as well as general findings plus a thorough bibliography. Larry Gall provides a neat paper on the initial recorded incursion of Strymon melinus onto Santa Catalina Island. The sole habitat of its close relation (sister species?) S. avalona. The paper provides a nice but too-brief lesson in morpho¬ logical character analysis by numerical techniques of intraspecific variation and identification of potential phenetic hybrids. Scott Miller gives the intro¬ ductory perspective. Five other papers cover Orthoptera, bees, Sphecids, mea¬ ly bugs, beetles and tiger beetles. Excellent detailed finescale maps of all islands are given in a separate envelope. My sole criticism lies in the reproduc¬ tion of typescript. Even though very well done in this case, there is something psychologically ephemeral about non-typeset work. Rudolf H . T. Mattoni, 9620 Heather Road, Beverly Hills, CA 90210, USA INSTRUCTIONS TO AUTHORS Manuscript Format: Two copies must be submitted (xeroxed or carbon papered), double-spaced, typed, on 8V2 x 11 inch paper with wide margins. Number all pages consecutively and put author’s name at top right corner of each page. If your typewriter does not have italic type, underline all words where italics are intended. Footnotes, although discouraged, must be typed on a separate sheet. Do not hyphenate words at the right margin. All measurements must be metric, with the exception of altitudes and distances which should include metric equivalents in parenthesis. Time must be cited on a 24-hour basis, standard time. Abbreviations must follow common usage. Dates should be cited as example: 4. IV. 1979 (day-arabic numberal; month-Roman numeral; year- arabic numeral). Numerals must be used before measurements (5mm) or otherwise up to number ten e.g. (nine butterflies, 12 moths). Title Page: All papers must have the title , author’s name, author’s address, and any titular reference and institutional approval reference, all on a separate title page. A family citation must be given in parenthesis (Lepidoptera: Hesperiidae) for referencing. Abstracts and Short Papers: All papers exceeding two typed pages must be ac¬ companied by an abstract of no more than 300 words. An additional summary is not required. Name Citations and Systematic Works: The first mention of any organism should include the full scientific name with author (not abbreviated) andyear of description. New descriptions should conform to the format: male: female, type data, diagnosis, distribu¬ tion, discussion. There must be conformity to the current International Code of Zoological Nomenclature. We strongly urge deposition of types in major museums, all type depositions must be cited. References: All citations in the text must be alphabetically listed under Literature Cited in the format given in recent issues. Abbrevations must conform to the World List of Scientific Periodicals. Do not underline periodicals. If four or less references are cited, please cite in body of text not in Literature Cited. Tables: Tables should be minimized. Where used, they should be formulated to a size which will reduce to 4 x 6V2 inches. Each table should be prepared as a line drawing or typed with heading and explanation on top and footnotes below. Number with Arabic numerals. Both horizontal and vertical rules may be indicated. Complex tables may be reproduced from typescript. Illustrations: Color must be submitted as a transparency (i.e., slide) ONLY, the quality of which is critical. On request, the editor will supply separate detailed instructions for making the most suitable photographic ilustrations. Black and white photographs should be submitted on glossy paper, and, as with line drawings, must be mounted on stiff white cardboard. Authors must plan on illustrations for reduction to the 4 x 8V2" page. Allowance should be made for legends beneath, unless many consecutive pages are used. Drawings should be in India ink at least twice the final size. Include a metric scale or calculate and state the actual magnification of each illustration as printed. Each figure should be cited and explained as such. The term “plate” should not be used. Each illustration should be identified as to author and title on the back, and should indicate whether the illustration be returned. Legends should be separately typed on pages entitled “Explanation of Figures”. Number legends consecutively with separate paragraph for each page of illustrations. Do not attach to illustrations. Retain original illustrations until paper finally accepted. Review: All papers will be read by the editor(s) & submitted for formal review to two referees. Authors are welcome to suggest reviewers, and if received, submit name & comments of reviewers. THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA Volume 25 Number 2 Summer 1986(1987) IN THIS ISSUE Date of Publication: 21 1987 A New Species of Calisto from Hispaniola with a Review of the Female Genitalia of Hispaniolan Congeners (Satyridae) Kurt Johnson, Eric Quinter & David Matusik 73 Records of Prolonged Diapause in Lepidoptera Jerry A. Powell 83 An exceptional case of paternal transmission of the dark form female trait in the tiger swallowtail butterfly, Papilio glaucus (Lepidoptera: Papilionidae) J. Mark Scriber & Mark H. Evans 110 The Phene tics and Comparative Biology of Euphilotes enoptes (Boisduval) (Lycaenidae) from the San Bernardino Mountains Gordon F. Pratt & Greg. R. Ballmer 121 A New Genus and Species from the Southwestern United States (Noctuidae: Acontiinae) Richard M. Brown 136 Observations on Problema bulenta George O. Krizek & Paul A. Opler 146 Book Reviews 149 COVER ILLUSTRATION: Reproduction of watercolor by Gordon Pratt of last (4th) instar larva, pupa, and adult of Euphilotes enoptes mojave THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA 1 WmMk . ca s ^ ► THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA ISSN 0022 4324 Published By: The Lepidoptera Research Foundation, Inc. c/o Santa Barbara Museum of Natural History 2559 Puesta Del Sol Road Santa Barbara, California 93105 Founder: William Hovanitz Editorial Staff: Rudolf H. T. Mattoni, Editor Lorraine L. Rothman, Managing Editor Scott E. Miller, Assistant Editor Associate Editors: Emilio Balletto, Italy Miguel R. Gomez Bustillo, Spain "I" Henri Descimon, France Thomas Emmel, U.S.A. Lawrence Gall, U.S.A. Brian 0. C. Gardiner, England Hansjuerg Geiger, Switzerland Otakar Kudrna, Germany Dennis Murphy, U.S.A. Ichiro Nakamura, U.S.A. Arthur Shapiro, U.S.A. Atuhiro Sibatani, Japan j“Deceased December 17, 1985 Manuscripts may be sent to the Editor at: 9620 Heather Road, Beverly Hills, CA 90210 (213) 274-1052 Notices Material may be sent to the Managing Editor. The JOURNAL is sent to all members of the FOUNDATION. Classes OF Membership: Regular (Individual) $ 15.00 year (vol.) Contributing $ 25.00 or more, year (vol.) Student/Retired — Worldwide $ 11.00 year (vol.) Subscription Rate/Institutions $ 25.00 year (vol.) Life $200.00 STATEMENT OF OWNERSHIP AND MANAGEMENT THE JOURNAL OF RESEARCH ON THE LEPIDOPTERA is published four times a year, Spring, Summer, Autumn, and Winter, by THE LEPIDOPTERA RESEARCH FOUNDATION, INC. The office of the publication and the general business office are located at 2559 Puesta Del Sol Road, Santa Barbara, California 93105. The publisher is THE LEPIDOPTERA RESEARCH FOUNDATION, INC. The Editor is R. H. T. Mattoni at the above Beverly Hills address. The Secretary-Treasurer is Barbara Jean Hovanitz at the general business office. All matters pertaining to membership, dues, and subscriptions should be addressed to her, including inquiry concerning mailing, missing issues, and change of address. The owner is THE LEPIDOP¬ TERA RESEARCH FOUNDATION, INC., a non-profit organization incorporated under the laws of the State of California in 1965. The President is R. H. T. Mattoni, the Vice President is John Emmel, the Secretary- Treasurer is Barbara Jean Hovanitz. The Board of Directors is comprised of Barbara Jean Hovanitz, Lorraine L. Rothman, and R. H. T. Mattoni. There are no bond holders, mortgages, or other security holders. Journal of Research on the Lepidoptera 25(3): 157-178, 1986(87) Distribution and Abundance of Butterflies in the Urbanization Zones of Porto Alegre, Brazil Alexandre Ruszczyk Av. Azenha N° 330 ap. 11, Porto Alegre 90.060, RS, Brasil Abstract The distribution of butterflies in the urban area of Porto Alegre was analysed by means of transects of avenues and data collected over a grid of 111 observation points. Maps were drawn showing the urbanization zones of the city, percent of vegetation cover as well as the distribution of 29 butterfly species. Three zones with relative uniformity can be identified along the urbanization gradient: B (buildings higher than four stories), vegetation cover below 20%; HB (houses and buildings of less than four stories), vegetation cover between 20 and 40% and H (houses, also including open areas within the city), vegetation cover above 40%. The distribution of butterflies in the city showed a life zone pattern very well correlated and oriented with the urbanization gradient. The border between zones H and HB represented a barrier for several species strongly associated with woods or natural fields, representing the most important transition area in the city fauna. The increase in the urbanization and pollution was accompanied by a decrease in the number of species and indi¬ viduals registered as well as by a homogenization in butterfly distribu¬ tion. In terms of abundance and distribution of its individual elements, the butterfly community of Porto Alegre is consistently structured in accord with the urbanization gradient, represented as distance from the center of the city. The predominance of this parameter is probably due to the fact that this distance is the main conditioner of many variables which are important for butterflies (such as urban climate, percent vegetation cover, air pollution and human density). Species of open areas, with high vagility, nectar feeders and with larvae feeding on exotic cultivated plants are dominant in the city. Introduction Among the more esthetically pleasing animals which inhabit urban ecosystems along with man, birds and butterflies have high ranking. Few authors have attempted to investigate the determinants of but¬ terfly occurrence and non-occurrence in man-made environments; most publications about butterflies in urban areas simply report a list of species found in a given city. In a more in-depth study, Shapiro & Shapiro (1973) studied the Staten Island (USA) butterfly community and called attention to its homogeneity. The butterflies found in aban¬ doned lots, always the same, were increasing in number and distribu¬ tion while the native and specialized forms were declining. The first 158 J.Res.Lepid. i group included vagile colonizers with a high reproductive rate, feeding on weeds and probably tolerant of air pollution. Yamamoto (1977) studied the butterflies of Sapporo (Japan) and found that most of the individuals belonged to a small number of species; a decline in the butterfly fauna paralleled the increase of urbanization. Species of open areas, which hibernate during the pupal stage and reproduce three or more generations per year, were those more resistant to urbanization. His results showed the substitution of forest species by open area species. Singer & Gilbert (1978) offered some general theoretical con¬ siderations about butterfly ecology in urban environments. Iq this work the entire urban area of Porto Alegre (Rio Grande do Sul, Brazil) was sampled for butterflies. The main objectives were to investi¬ gate butterfly distribution over the urbanization gradient and the influence of habitat variables on butterfly abundance. Study Area The city of Porto Alegre is located in southern Brazil (30°02' S, 51°14' W), with a population of over a million inhabitants. The altitude varies from 4 to 300 m above sea level (mean about 80—100 m). The region has a temperate-subtropical climate with high humidity and moderate¬ ly high temperatures in the summer. The annual mean temperature is 13.8°C and the average rainfall 1322 mm. The city is surrounded by agrarian ecosystems to the north, south and east; to the west are found the aquatic ecosystems of Guaiba River (Figure la). Within the city are found only remnants of woods in hard-to-reach places in the southern sector, where urbanization has been partially stopped. A field vegetation, either managed or aban¬ doned, presently predominates on the periphery of the city. Over a 1:20.000 city map was laid a 5-cm grid (equal to 1 km2 real size). Using the geometric center of each square, 111 circles 3 cm in diameter (300 m radius or 0.283 km2 real size) were defined. The circles corresponded to sampling subunits called observation points (OP). The distance of each OP from OP E5 (Figure 2) in the center of the tall building zone was considered “distance from the center of the city”. The mean altitude of each OP was estimated as the arithmetic mean between its highest and lowest points. Over a 1:8.000 photograph of each OP was laid a 6 x 6 cm square of millimetered paper (0.2304 km2 actual area). The parts covered by plants, including native vegetation as well as lawns, back yards, vacant lots and street trees, were shaded. The calculated percentage of vegetation cover of each OP was extrapo¬ lated to the area of 1 km2 (Figure lb). By examination of 1:20.000 aerial photographs of the city with the aid of a stereoscope, three zones of different intensities of urbanization could be drawn over a political map at the same scale: high (buildings zone or zone B; vegetation cover below 20%; zero to 2 km distant from 25(3): 157-178, 1986(87) 159 the center of the city), medium (houses and buildings zone or zone HB; vegetation cover between 20 and 40%; 2 to 7 km distant from the center of the city) and low (houses zone or zone H; vegetation cover above 40%; 4 to 12 km distant from the center of the city). The borders of the zones were adjusted by examination in loco. The final map (Figure lc) was simplified to polygons, by drawing tangential lines to the borders of the different zones of urbanization (Figure Id). The radial arrangement of the main avenues of Porto Alegre has determined an urbanization gradient also radial and relatively similar in all directions from the center of the city. Over the urbanization gradient is found a complementary gradient of vegetation covering, also with a radial aspect and similar preferential orientation (northeast-southeast) (Figure lb). Zone B and industrial and shopping areas generally show less than 20% plant cover. Zone HB has under 40% plant cover, showing a close spatial relationship with the 15-30% class in Figure lb. Zone H has in general plant cover value over 40%, reaching a maximum of 78%. The mean value was 39.3% (s = 17.2). The minimum value for this variable was 7.2% in OP E4. Methods The distribution of butterflies was investigated using two methods: transects, and data recording in observation points. Transects Four transect routes were used (AB, CD, EF and GH), along the main avenues out from the center of the city. The censuses began at 10 a.m. at the inner end of each route, and consisted of a round trip to the outer end and back. Ten such censuses were done along each route. All butterflies seen by naked eye were registered, whether flying or sitting up to 10 meters back from the street side of the buildings. The location of each individual was determined in relation to the nearest cross street. Data Recording in Observation Points The whole set of the OPs was explored during three sampling periods (November-December 1980, March-April 1981 and June-July 1981). In each of these periods the OPs were visited sequentially, five per day. First the OPs of row 7 were visited followed by rows 8, 6, 9, 5 and thus successively (Figure 2). In each OP, a 45-minute period was spent constantly walking the streets and recording the number of individuals of different butterfly species seen. The field data were transferred to computer cards and all calculated values were obtained through the use of SPSS (Nie et al., 1975) programs. 160 J.Res.Lepid. i Results and Discussion Distribution of Butterflies along the Transect Routes Table 1 includes the total number of individuals of different species along the four transect routes. The recordings for the final 1.6 kilo¬ meters of route AB (8 km in total length) are not included in this table, since they represent extra-urban data, not comparable to those obtained for other routes. Figures 3, 4, 5 and 6 show graphic repre¬ sentations of the butterfly groups along the four routes. From AB to GH there occurs a levelling of topography, an increase of urbanization intensity and a decrease in the number of individuals recorded per km of transect (Table 1). Dryas iulia, Ascia monuste orseis and Phoebis philea were the most numerous butterflies, totalling 30% of the recorded insects along each route. The predominance of these species is due, among other factors, to their great abundance in the region (including within the city), speed and mobility, and the vivid colors of their wings which allows easy spotting. Table 1. Butterflies observed along 20 transect (center-suburbs-center) on four acess routes. Explanation in the text. SPECIES AB (17.3)* N % CD (14.7) N % ROUTE EF (11.9) N % GH (10.5) N % TOTAL N % Dryas iulia (Fabricius, 1775) 88 17.0 63 14.3 46 12.9 26 9.9 223 14.1 Ascia monuste orseis (Latreille, 1819) 65 12.5 55 12.4 49 13.8 32 12.2 201 12.7 Phoebis philea (Johansson, 1763) 44 8.5 45 10.2 47 13.2 21 8.0 157 9.9 Anartia amathea (Eschscholtz, 1821) 23 4.4 24 5.4 22 6.2 49 18.7 118 7.5 PapiUo scamander Boisduval, 1836 22 4.2 27 6.1 25 7.0 15 5.7 89 5.6 Phoebis spp. 21 4.0 22 5.0 23 6.5 17 6.5 83 5.3 Junonia evarete (Cramer, 1779) 25 4.8 19 4.3 16 4.5 21 8.0 81 5.1 Colias lesbia pyrrhothea (Hubner, 1823) 29 5.6 34 7.7 1 0.3 — - 64 4.1 Tatochila autodice (Huebner, 1818) 30 5.8 14 3.2 9 2.5 1 0.4 54 3.4 PapiUo anchisiades capys Huebner, 1809 30 5.8 15 3.4 8 2.2 - - 53 3.4 Urbanus spp. 16 3.1 13 2.9 13 3.6 8 3.1 50 3.2 Actinote spp. 6 1.2 18 4.1 7 2.0 13 5.0 44 2.8 Euptychia spp. 6 1.2 4 0.9 14 3.9 16 6.1 40 2.5 Anosia gilippus (Cramer, 1775) 22 4.2 6 1.4 8 2.2 2 0.8 38 2.4 Agraulis vanillae maculosa (Stichel, 1907) 3 0.6 8 1.8 13 3.6 8 3.1 32 2.0 Eurema spp. 9 1.7 6 1.4 10 2.8 1 0.4 26 1.6 Phyciodes spp. 2 0.4 5 1.1 9 2.5 8 3.1 23 1.5 Battus polydamas (Linnaeus, 1758) 12 2.3 3 0.7 2 0.6 — - 17 1.1 Eunica margarita (Godart, 1822) 9 1.7 5 1.1 1 0.3 2 0.8 17 1.1 Dione juno (Cramer, 1779) 2 0.4 8 1.8 4 1.1 - - 14 0.9 Methona themisto Huebner, 1818 8 1.5 3 0.7 1 0.3 1 0.4 13 0.8 PapiUo hectorides Esper, 1794 3 0.6 5 1.1 2 0.6 - — 10 0.6 Biblis hyperia (Cramer, 1779) 3 0.6 - - 4 1.1 2 0.8 9 0.6 PapiUo thoas brasiliensis Rothschild & Jordan, 1906 5 1.0 2 0.5 - — 1 0.4 8 0.5 Placidula euryanassa (Felder, 1860) - — 4 0.9 1 0.3 3 1.1 8 0.5 Adel p ha spp. - - 4 0.9 3 0.8 1 0.4 8 0.5 Heliconius erato phyllis (Fabricius, 1775) 2 0.4 3 0.7 1 0.3 1 0.4 7 0.4 Dryadula phaetusa (Linnaeus, 1758) 1 0.2 - - 2 0.6 4 1.5 7 0.4 Siproeta stelenes (Linnaeus, 1758) 1 0.2 1 0.2 3 0.8 1 0.4 6 0.4 PapiUo astyalus Latreille, 1819 4 0.8 1 0.2 - - — - 5 0.3 Diaethria spp. 1 0.2 3 0.7 1 0.3 - - 5 0.3 Doxocopa laurentia (Godart, 1821) 3 0.6 1 0.2 - — - - 4 0.3 Hamadryas amphinome (Fruhstorfer, 1916) 2 0.4 2 0.5 - - - - 4 0.3 Anaea itys (Gmelin, 1791) 1 0.2 1 0.2 1 0.3 1 0.4 4 0.3 Dismorphia spp. 2 0.4 — - 2 0.6 - - 4 0.3 Hamadryas spp. 2 0.4 1 0.2 - - - - 3 0.2 Epiphile huebneri Hewitson, 1861 - - 1 0.2 2 0.6 - - 3 0.2 Heliopetes omrina (Butler, 1870) - - - - 1 0.3 2 0.8 3 0.2 Eurema deva deva (Doubleday, 1847) 1 0.2 1 0.2 - - 1 0.4 3 0.2 Opsiphanes invirae Stichel, 1901 1 0.2 2 0.5 - - — - 3 0.2 Dynamine myrrhina (Doubleday, 1849) 1 0.2 - - - - 1 0.4 2 0.1 Riodina lysistrata (Berg, 1896) - 1 0.2 - - 1 0.4 2 0.1 Praepedaliodes phanias (Hewitson, 1861) 1 0.2 - - 1 0.3 - - 2 0.1 Mechanics lysimnia (Fabricius, 1793) 2 0.4 - — - - - - 2 0.1 Euryades corethrus (Boisduval, 1836) 1 0.2 - - - - - — 1 0.1 Pyrgus oileus (Stoll, 1790) 1 0.2 - - - - — - 1 0.1 Pyrgus communis (Giacomelli, 1928) - — - — - - 1 0.4 1 0.1 Doxocopa kallina (Staudinger, 1886) 1 0.2 — - - - - — 1 0.1 Vanessa braziliensis (Moore, 1883) 1 0.2 - - - - - - 1 0.1 Marpesia petreus (Cramer, 1776) - — 1 0.2 - — - - 1 0.1 Eypanartia bella (Fabricius, 1793) - - — - 1 0.3 - — 1 0.1 Philoros rubriceps opaca (Boisduval, 1870) - - - - — - 1 0.4 1 0.1 Others 7 1.3 11 2.5 3 0.8 - 21 1.3 Total 519 100.0 442 100.0 356 100.0 262 100.0 1579 100.0 (N/Km/TRANSECT) 25(3): 157-178, 1986(87) 161 The papilionids (Figure 3) show a reduction in the number of indi¬ viduals from AB to GH; in the latter transect, except for one individual of Papilio thoas brasiliensis, all those recorded belonged to the species P. scamander scamander. This monotony is in accord with the small number of species of this family observed in the OPs located in this area of the city. The sap and fruit eating nymphalids were most common along AB, also decreasing in the direction of GH (Figure 6). All these species are native of subtropical woods on the city periphery, showing their greatest numbers on the distal end of AB, which crosses areas with remnants of this habitat. Along the routes AB, CD and EF the different families and subfami¬ lies showed a sharp reduction of the species number and individuals inside the limits of zone B; only one or two species were recorded for each group of butterflies. On GH, however, there was a greater homogeneity in the distribution of these groups, represented all along the route by the species that on other routes were well represented in zone B. This emphasizes the environmental stress of this region. The homogeneity in the distribution of the different groups of butterflies along GH is probably due in part to the spatial uniformity of this portion of the city. This area has a very regular disposition of streets, similar to a chessboard, and is extremely flat with elevations below 5 meters, which represents a low diversification of habitats. In aerial photographs it shows great similarity among its different sites. The scarcity of vegetation on the margins of route GH tends also to increase the homogeneity in the distribution of butterflies, since it eliminates a factor of concentration of these insects. Farrapos Avenue, the greatest part of route GH, is surely the avenue with the greatest air pollution in the city due to particles and industrial gases as well as from vehicles. Pollution is a factor of homogenization of environmental conditions consequently decreasing the complexity of animal and plant communi¬ ties belonging to a certain biotope. Thus, the smaller species number and homogeneity of distribution found along GH may also be explained by the air pollution in this area. The Urban Community of Butterflies The data in Table 2 show the large number of individuals of a small number of butterfly species in the urban area. The data provide evidence that the butterfly communities of Porto Alegre are organized with a consistent structure. This can be seen from the results of the two methods used. For example, the more abundant species in the transects and OPs hold the top positions in the abundance ranking in the majority of routes and regions of the city (Tables 1 and 2). The majority of the genera and species which represent less than 2% of the records in the transects maintain this low proportion also in the OPs. It will be 162 J. Res. Lepid. Table 2. Total number of butterflies observed in 11 regions of the city of Porto Alegre. REGIONS IV V VI VII VIII IX X XI Ascia monuste orseis (Latreille. 1819) Dryas iulia (Fabricius, 1775) Junonia evarete (Cramer, 1779) Urbanus spp. Tatochila autodice (Huebner. 1818) Phoebis philea (Johansson, 1763) Papilio scamander Boisduval. 1836 Papilio anchisiades capys Huebner, 1809 Actinote spp. Agraulis vanillae maculesa (Stichel. 1907) Anosia gilippus (Cramer, 1775) Phoebis spp. Eurema spp. Battus poiydamas (Linnaeus, 1758) Euptychia spp. Anartia amathea (Eschscholtz, 1821) Papilio astyalus Latreille. 1819 Heliopetes omrina (Butler, 1870) Leptotes cassius (Cramer, 1775) Papilio thoas brasiliensis Rothschild & Jordan, 1906 Methona themisto Huebner, 1818 Vanessa braziliensis (Moore, 1883) Phyciodes daudina (Eschscholtz, 1821) Euryades corethrus (Boisduval, 1836) Papilio hectorides Esper, 1794 Heliopetes alana (Reakirt, 1868) Eurema deva (Doubleday, 1847) Dione juno (Cramer, 1 779) Fergus oileus (Stoll, 1780) Eunica margarita (Godart, 1822) Heliconius erato phyllis (Fabricius, 1775) Parides perrhebus (Boisduval, 1836) Pyrgus communis (Giacomelli, 1928) Colias lesbia pyrrhothea (Hiibner, 1823) Euptoieta hortensia (Blanchard. 1852) Hamadryas spp. Dryadula phaetusa (Linnaeus, 1758) Placidula euryanassa (Felder, 1860) Dynamine myrrhina (Doubleday, 1849) Anaea itys (Gmelin, 1791) Parides agavus (Drury, 1782) Adelpha spp. 'Philoros rubriceps opaca (Boisduval, 1870) Biblis hyperia (Cramer, 1779) Eurytides lysithous (Huebner, 1821) Praepedaliodes phanias (Hewitson, 1861) Diaethria spp. Phyciodes ithra (Kirby, 1900) Riodina spp. Dismorphia spp. Doxocopa laurentia (Godart, 1821) Battus polystictus (Butler, 1874) Siproeta stelenes (Linnaeus, 1758) * Josia angulosa (Walker, 1854) Hypanartia bell a (Fabricius, 1793) Hamadryas amphinome (Fruhstorfer, 1916) Doxocopa kallina (Staudinger, 1886) Opsiphanes invirae Stichel, 1901 Parides anchises nephalion (Godart, 1819) * Phaloe cruenta (Huebner, 1823) • Utetheisa ornatrix (Linnaeus, 1758) Prittvyitzia hymenaea (Prittwitz, 1865) Phyciodes lansdorfi (Latreille, 1820) Siproeta trayja (Hiibner, 1823) Marpesia petreus (Cramer, 1776) Morpho catenarius Perry, 1811 Anartia jatrophae (Johansson, 1763) Philaethria wernicket (Rober, 1906) 'Macrocneme chrysitis (Guerin, 1843) Epiphile huebneri Hewitson, 1861 Others 117 152 44 49 63 76 23 45 47 21 62 29 42 64 24 17 18 15 26 7 24 19 5 12 12 8 29 33 2 3 18 6 10 3 21 9 9 6 6 7 3 2 3 10 2 7 2 2 5 2 1 1 3 2 40 100 111 138 102 145 96 1 54 120 46 122 62 83 116 117 69 74 84 55 57 74 86 60 57 43 48 94 13 34 73 84 48 64 67 93 28 70 67 54 47 59 34 46 45 30 61 51 37 38 36 45 36 38 26 48 50 41 31 28 15 25 21 33 10 7 22 31 33 22 21 21 12 19 24 20 36 19 18 15 13 18 16 11 14 13 5 49 6 12 18 24 19 28 8 50 17 12 7 25 9 26 10 25 15 23 12 12 11 12 10 7 11 14 13 13 12 20 9 9 10 5 4 4 12 15 6 4 4 13 8 12 9 14 2 3 4 1 7 21 2 8 6 1 6 10 18 6 1 - 8 3 9 9 4 2 3 3 5 8 11 - - - 3 4 111 2 3 4 3 13 1- 3 14 4 6 112 2 2 1- 2 111 1-31 13 2 2 4 - 2 - 2 1-3 2 - - 2 2 1 1 2 80 56 70 64 55 29 29 23 16 59 23 27 35 19 27 46 10 30 20 5 21 24 14 28 6 17 7 2 5 3 2 9 3 2 90 109 99 80 47 67 62 39 69 21 44 37 24 36 22 13 15 17 25 22 10 17 7 6 17 9 16 5 8 10 9 2 3 4 2 5 3 2 3 80 58 62 37 64 110 61 59 43 45 62 12 51 72 30 37 25 7 34 11 62 40 40 12 41 21 16 12 17 19 56 11 12 25 28 11 9 10 13 15 16 19 12 7 3 3 13 1 12 1 16 5 7 9 1 3 3 2 7 2 3 38 5 1 2 2 2 2 2 2 2 1 2 1 1 3 70 33 39 43 53 26 12 17 49 14 15 7 3 10 13 18 16 3 1 6 10 3 2 65 17 48 30 30 22 41 15 2 6 7 2 12 2 4 6 10 j 2 5 2 5 2 33 34 38 40 39 29 30 35 7 11 138 130 126 104 103 93 86 66 63 61 56 51 45 42 29 26 24 21 21 20 20 16 15 14 14 10 9 8 7 7 6 6 5 4 3 3 2 2 2 2 336 8.71 8.39 6.82 6.48 5.10 4.80 4.48 4.35 3.74 3.63 3.49 3.32 2.68 2.58 2.22 2.14 1.69 1.58 1.56 1,41 1.34 1.33 1.20 1.19 1.12 1.09 1.02 0.90 0.89 0.80 0.74 0.57 0.54 0.48 0.44 0.39 0.36 0.25 0.22 0.21 0.18 0.18 0.17 0.17 0.16 0.14 0.13 0.12 0.12 0.09 0.08 0.07 0.06 0.05 0.05 0.05 0.04 0.03 0.03 0.03 0.03 0.03 0.02 0.02 0 02 0.02 2.89 Total 1392 1369 1568 1198 1291 950 1128 970 767 583 394 11610 100.00 * moths shown below that this consistent organization may also be applied to the distribution of the members of this fauna. Figures 7 — 10 show the distribution of the different groups of butter¬ flies in the urban zones. These maps may be seen as an estimate of the distribution areas of different species in the urban area of Porto Alegre for the period of 1980—81; this is certainly suffering gradual modifica¬ tions, considering the velocity of vertical and horizontal urbanization. Within each subfamily or genus of butterflies there are species spread out over all zones of urbanization and others found in semi-circular bands progressively narrower and farther from the zones B and HB having as virtual center zone B. This fact is related to the radial character of the urbanization gradient and vegetation covering of the city. The majority of the species show a continuous distribution over the city, decreasing in amplitude towards more intensively urbanized zones. This, along with the high degree of vagility of the dominant 25(3): 157-178, 1986(87) 163 species (and the majority of others) discourages the use of the express¬ ion mosaic distribution (often applied to soil insects) for the butterfly fauna of Porto Alegre. The expression life zones introduced by Merriam (1894) to designate the changes of plant communities due to altitude and latitude better characterizes the zonation of the butterfly distribu¬ tion on the urban gradient. Each species shows a more or less similar distribution pattern in the three samples, though some members of the subfamily Nymphalinae reveal seasonal variations in their distribution. In each case the pat¬ tern of distribution verified in the transect routes was in general similar to the one found for the OPs. The species that showed a rather wide distribution in the OPs (such as P. scamander, A. m. orseis and D. iulia ) also showed a wide distribution along the routes of transects. The species with a more restricted distribution in the OPs such as P. a. astyalus, P. hectorides and H. e. phyllis were found to be more frequent on the outer portions of the routes. The species observed only on the border of urban area ( B . polystictus, P. a. nephalion and P. agauus ), within areas not reached by the majority of the routes were very infrequent along the transects. Neverthless, they were found on the distal end of route AB which reaches the city periphery. These facts emphasize the zonation of distribution areas of butterflies in the urban area of Porto Alegre. The species that were rare in the urban area (less than 1%) in general are stenotopic in the sense of the adult being typical of field or woods. They feed in the larval stage on native plants which are infrequent or non-existent in the city. Their distribution was restricted to peripheral portions of zone H, especially in the southern sector which is richer in remnants of subtropical woods and is nearer the granitic hills of the city periphery, where still denser woods are located. In the adult stage fruit and sap feeding is predominant. In the central areas of the city, species of open areas predominate, in accord with the results of Yamamoto (1977) on the butterflies of Sapporo (northern Japan). Species typical of natural fields behaved in Porto Alegre much like the woods species, even though they were more numerous; their distribution was concentrated in zone H. The drying and warming of the urban environment makes the habi¬ tats of green areas similar to the xerothermic ones (Schweiger, 1953; Trojan, 1981) favoring species which tolerate low humidity and sub¬ light (Kouch & Sollmann, 1977; Pisarski & Czechowski, 1978). Many forest species show a preference for rather low temperatures, high humidity and shade. On the contrary the field species prefer high temperatures, low humidity and sunlight (Tischler, 1965). Naturally, the ecology of a butterfly species in the city and its success in adapting to this new environment are directly related to its ecology in natural conditions. Thus the lepidopteran species typical of fields would be better pre-adapted to urban life than forest species. 164 J.Res.Lepid. If cities are considered as well illuminated open areas, warm and with low humidity, it would be reasonable that field species would be dominant in central areas of Porto Alegre; instead, they are lacking there, since natural fields are not present. The predominant physiog¬ nomy of urban habitat is closer to a savanna, with open areas where a low vegetation can grow (in general subjected to some form of manage¬ ment), consisting of shrubs and trees interspersed by built-up areas. From this fundamental character of the urban habitat probably comes the predominance in urban Porto Alegre of species that are not typical either of fields or woods but prefer open areas. They are eurytopic in the sense that the adults may be observed either in grasslands or in woods or mixed areas. In the larval stage they utilize native and exotic plants widely spread over the town. The available adult and larval food apparently is the main biotic ecological factor that explain the great abundance of the dominant butterflies in the urban area of Porto Alegre (Ruszczyk, 1986). Typical of these species is their degree of vagility, which certainly has contributed to their wide distribution in the city. The Abundance of Butterflies in the Urbanization Zones Figure 11 shows the number of butterflies recorded in the three samples of the urban area of Porto Alegre. All samples reveal a progressive reduction of the number of individuals in the direction of zone B. The mean number found for zone B was about 40 individuals, compared with about 64 in zone HB and about 130 in zone H. There is thus an increase of 60% from zone H to zone HB and more than 100% going from zone HB to H. This last increase already appeared in the OPs of zone HB located on the border of zone H (Figure lid). The border between zone H and zone HB acts as a barrier for several butterfly species, especially those characteristic of field and wood environments. This border is the main transition area of the butterfly fauna in going out from the central area of the city. Its presence was obvious on the maps showing the number of individuals sampled in summer, winter and total seen as well as on diversity maps (in prep.). This border is also important for some bird species that are sensitive to urbanization (Ruszczyk et al., 1987). The relative influence of the variable plant cover, distance from the city center and mean altitude of the OPs was analyzed for the total number of butterflies recorded, through simple correlation and multi¬ ple regression methods. The three variables showed positive correla¬ tions with the total number of butterflies, with the respective coef¬ ficients being 0.714, 0.710 and 0.456 (all significant to the 1% level). Standardization of variables gave a standard regression coefficient of 0.326 for plant cover, 0.433 for distance from the city center and 0.154 for mean altitude, all significant to the 1% level. This indicates that the 25(3): 157-178, 1986(87) 165 distance from the city center has a greater influence on the number of individuals than the plant cover or mean altitude. These three vari¬ ables together were responsible for 61% of the explained variance of the total number of recorded butterflies for each OP. Decomposing this proportion shows the contribution of each variable: Proportion Increment Increment Increment Not attributed of variance due to the due to due to to either explained by distance plant average Xx, X2 or X3 all three from the cover altitude alone variables center of the city (In km) (arc sine V%) (m) (R2) (XJ (X2) (X3) 0.610 = 0.092 + 0.041 +0.017 +0.460 Three quarters of the explained variance in the recorded number of butterflies is due to secondary effects between variables. The predomi¬ nance of the single variable distance over plant cover and altitude is probably related to the large number of other variables which are directly related to it and are important to the butterflies. Variables such as temperature of the urban area, percent plant cover, degree of habitat disturbance (movement of vehicles and human beings), human population density, air pollution and intensity of urbanization are all organized as predominantly radial gradients due to the fundamental radial character of Porto Alegre’s urbanization. In this way the intensi¬ ty of action of these and other variables (which may be called all together anthropogenic pressure (Trojan, 1981)) on the lepidopterans depends in great portion on their position relative to the center of the city. This suggests a predominance of effects of physical factors on the distribution of these insects in urban areas (but see Ruszczyk, 1986 for a discussion of biotic factors in one common species, Papilio scaman- der). Acknowledgments. I wish to acknowledge with thanks the financial support provided by Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq). I would like to thank Aldo Mellender de Araujo, Keith S. Brown, Jr and Miriam Becker for helpful criticisms and suggestions. Celso Paulo Jaeger kindly prepared the first English version of the paper. Literature Cited KOUCH, K. & A. SOLLMANN, 1977. Durch Umwelteinfliisse bedingte Verander- ungen der Kaferfauna eines Waldgebietes in Meerbusch bei Diisseldorf. Decheniana, 20: 36-74. MERRIAM, C. H., 1894. Laws of temperature control of the geographic distribution of terrestrial animals and plants. Nat. Geogr. Mag., 6: 229-230. 166 J.Res.Lepid. NIE, N. H., HULL, C. H., JENKINS, J. G., STEINBRENNER, K. & D. H. BENT, 1975. Statistic¬ al Package for the Social Sciences. McGraw-Hill, USA. PISARSKI, B. & W. CZECHOWSKI, 1978. Influence de la pression urbaine sur la myrmecofaune. Memorab. zool., 29: 109-128. RUSZCZYK, A., 1986. Mortality of Papilio scamander scamander (Lepidoptera: Papilionidae) pupae in four districts of Porto Alegre (S. Brazil) and the causes of superabundance of some butterflies in urban areas. Revta. bras. Biol. 46(3): 567-579. RUSZCZYK, A., RODRIGUES, J. J. S„ ROBERTS, T. M. T„ BENDATI, M. M. A., MELO, M. T. Q„ MARQUES, J. C. V. & R. S. DEL PINO, 1987. Distribution patterns of eight bird species in the urbanization gradient of Porto Alegre, Brazil. Cienc. Cult. 39(1) (in press) SCHWEIGER, H., 1953. Versuch einer zoogeographischen Gliederung der rezeten Fauna der Wiener Stadgebietes. Ost. zool. Z., 4 (4/5): 556-586. SHAPIRO, A. M. & A. R. SHAPIRO, 1973. The ecological associations of the butterflies of Staten Island. J. Res. Lepid., 12(2): 65-128. SINGER, M. C. & L. E. GILBERT, 1978. Ecology of butterflies in the urbs and suburbs. In: FRANKIE, G. W. & C. S. KOEHLER, eds. Perspectives in Urban Entomology. New York, Academic. Press, p. 1-12. TISCHLER, W., 1965. Agrarokologie. Jena, G. Fischer. TROJAN, P., 1981. Urbanfauna: faunistic, zoogeographical and ecological prob¬ lems. Memorab. zool., 34: 3—12. YAMAMOTO, M., 1977. A comparison of butterflies assemblages in and near Sapporo city, northern Japan. J. Fac. Sci. Hokkaido Univ. Serie 6, Zoology, 20(4): 621-646. 25(3): 157-178, 1986(87) 167 Figure 1. a) Schematic map of ecosystems within and around the city of Porto Alegre. 1. urban area; 2. agriculture; 3. agriculture and livestock; 4. agriculture and second growth; 5. marshes; 6. subtropical forest. b) Map of percent plant cover of PA. c) Map of urbanization zones of the city (1978). d) Simplification of map "c" . 168 J. Res. Lepid. ABCDEFGH I JKLMN 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Figure 2. Location of the 1 1 1 observation points of the butterfly fauna of the city of Porto Alegre. Each observation point has a diameter of 600 m and its area was sampled three times for butterflies. The dashed line corresponds to the simplified limit of the urban area. Solid lines demarcate 11 regions into which the observation points were grouped for analysis. 25(3): 157-178, 1986(87) 169 0*1 •8*00*a O 1 K M S . . o. . o •080 ^ o* 08 • o y 9 O 8 c& o o o & •-P A. CAPYS 5m O- P T. BRASILfENSIS ] H m 9- P A. ASTYALUS ▼ - P HECTOR IDES A-BATTUS polydamas O-Parides PERRHEBUS □- P AGAVUS Figure 3. Family Papilionidae — Distribution in the urbanization zones of Porto Alegre. Five transects center-periphery-center (1, 2, 3, 4, 5) were made on each route in April and May 1980, February, April and May 1981, respectively. The line under the symbols is the topographic profile of the routes. The urbanization zones crossed by the routes (see map at lower right) are indicated under the topographic profile. 1. buildings zone; 2. houses and buildings zone; 3. houses zone; 4. houses zone with remnants of subtropical forest. 170 J.Res.Lepid. • - A S C I A M. ORSEIS O-PHOEBIS PHI LEA 9-EUREMA SPP. 1km Figure 4. Family Pieridae — Distribution in the urbanization zones of Porto Alegre. See legend of Figure 3. z’ z 25(3): 157-178, 1986(87) 171 1 o 2 K7 3 4 O O O o o 5 o o co o o o o o 8b o O o CD • o o o o &>o ooA> o o o 008 o o o o 8 o cBa o o o 8o§§» oo oft) oc§ c05o A tel lllllllIlL B O-DRYAS JULIA •-AGRAULIS V. MACULOSA A- DlONE JUNO <7- Helicon ius e. phyllis ©- Dryadula phaetusa 1 K M □ Figure 5. Family Nymphalidae (Heliconiini) — Distribution in the urbaniza¬ tion zones of Porto Alegre. See legend of Figure 3. 172 J.Res.Lepid. o. . <£ •• 2 2 © 3 OOOOOO *0*0 o oo8fio 4 o oo o oo • o» o • 5oOO ^7 00 Q • o o oo G |inn7;=i;;=;;;;.:i=:;;:-::i:i-:i:i:l5.;=::|||||H|||||H|||l||H|m|[H||H|U||H|||l||HlllllllllllllHIIII I H 1 K M O - A NARTIA A M ATM EA • - JUNONIA EVARETE S7- BlBLIS HYPER IA a - Doxocopa spp. ©- Adelpha spp. O-Diaethria spp. □ - A N A E A ITYS A- Hamadryas SPP. Figure 6. Family Nymphalidae (Miscellanea) — Distribution in the urbaniza¬ tion zones of Porto Alegre. See legend of Figure 3. 25(3): 157-178, 1986(87) 173 lJ®t 0©0©'Ab'@®® j&4 ]®- b©$x3©©@©L@ ©Oy$®A © ® ©— s qeoo^A ® © W©J ©o'®b@© • o; ob a ©® /©> 0©AA©-\/)4km ©A©/" Mill ©©O Papi'io scamander a7 n-ApQT sota© or©o o^od© A-©© o © Q— S md Woo®b \ AO/ @oo(o b©0 A A © a) © a©©©©» -0000@@^0 0060.0^ °®°g Papilio a. astyalus - vn°®T ©"A A A A o AAA _ A /A.© /Y/fl © a3 aO/a a[®T 'A A V©©©®©©,— ' \ l O 7© © a8a © @ /a AJ q O'© © ©/A a © ©( ©©M©© A A A a) /©^ A © © A © © ©-AS ©®A ®©^\b /©©©©/ /© AO7 1©a8 Papilio ^ a capvs /o- A © A £©. A-1 O /©MO /L© 'A V©Q © b A © A Os> rOOOA A A^O 70© AO ^©©o\ 0 ©gop Papilio t. brasiliensis 1.0 2. A ©r '\©/ A A a© o oJ®| ^ y bo ©a® _o 0©©-a “obo •!©) /©ooa© "Av/@ ©> '© ©"AS bO • A A^\J ©©O' Papilio hectorides 3 ©oo© ©A A © A(C)' ''©-A © © ©) '®@@oo@oo> rwW©@ A©^/ /©©©go" O °o 6@ooo O^jqJ Battus 'aq polydam as Figure 7. Family Papilionidae — Distribution in the urbanization zones of Porto Alegre. The butterfly fauna of 1 1 1 observation points of the urban area was sampled three times, in November-December 1980, March-April 1981 and June-July 1981, respectively 1, 2 and 3. 174 J. Res. Lepid. I 25(3): 157-178, 1986(87) 175 /©- a/ Av&tA Af A © A @ @ A A A © ® © C©XA ^J® A A7A © A (® A g^A A ©/ A A A( ©-A~A-A • A A A) /®^> A © A © © ©'©> @@A©@@^xJ4km A A ® A A^ Mill © ©M^ A , Asc ia m. orse i s f@^/^^A\®© Af ©0©>;<:®©/ © A p O ^am®/©a]4- /O © © @© ©"© A O O x®^aa@©aV LO/7 r d© @ ©/© a © © © ^ © a A7. 0© 70 •/& A ^CTaO AA@(2j) 70 a© A©©q©> ^® © © o ® ©^\q ® o©©,©x y ©©©/ L ® © ® O y © ® © C> A g r a u 1 1 s v. maculosa c ©'A7 ®© w — . O q D i o n e -o'f * J u n o 1.0 2. a Figure 9. Family Nymphalidae (Heliconiini) — Distribution in the urbaniza¬ tion zones of Porto Alegre. See legend of Figure 7. 25(3): 157-178, 1986(87) 177 /©- a/ ^a\©k A A A(>A.A^ /AAAA-A AJ<$@' aaa a a?a a'©Ta a"' AAA n(VA A ® ® O A— aAAAAA7AAA; ' A'A A A/A A A A-aSTA-© A A@A; A A A A © A @A> a a A^^j4km A®®A A7 I-.. I l | 1 ®Af/ ®°° Junonia e va rete /©v, A A O' Figure 10. Family Nymphalidae (Nymphalinae) — Distribution in the urba¬ nization zones of Porto Alegre. See legend of Figure 7. 178 J. Res. Lepid. Figure 11. Number of butterflies registered during three samples in 111 observation points of the urban area of Porto Alegre. In each observation point a period of 45 min was spent walking an urban area of 600 m in diameter and recording the butterflies seen. 1, 0-16 individuals; 2, 17-40; 3, 41-70; 4, 71-100; 5, 101-135. Journal of Research on the Lepidoptera 25(3): 179-187, 1986(87) The Effect of Temperature on Expression of the Dark Phenotype in Female Papilio glaucus (Papilionidae) David B. Ritland Department of Zoology, University of Florida, Gainesville, Florida 32611 Abstract. Experimental broods of Papilio glaucus produced unusual dark morph females when reared at high temperatures. Exposure to temperatures of 25-28 C during the larval and pupal stages produced adult females which were phenotypically intermediate between the normal yellow and dark morphs of the butterfly, i.e., with a dusting of yellow scales in the dark background. Naturally-occurring females with this intermediate coloration have been recorded from throughout the eastern United States, but are generally infrequent. The dark morph of Papilio glaucus appears to be canalized (buffered) against environmental modification under natural conditions. It is proposed that canalization of the dark morph is adaptive because it protects the mimetic resemblance of dark females to the unpalatable Battus philenor, and that canalization is strongest in populations of P. glaucus from areas where B. philenor is an abundant model. Introduction Phenotypic plasticity and polyphenism (Shapiro, 1976) in butterflies are presumably adaptive responses to heterogeneous or seasonal en¬ vironments. However, developmental canalization (inflexibility of the normal phenotype over a range of environmental conditions due to the action of the epigenetic system; Waddington, 1957) is also adaptive (Shapiro, 1981; Hoffman, 1982). Phenotypic stability may be particu¬ larly advantageous in mimetic species, because environmental mod¬ ification of the wing pattern would decrease the mimetic resemblance. This paper summarizes an investigation into phenotypic plasticity and canalization in the mimetic eastern tiger swallowtail, Papilio glaucus L. Two subspecies of the tiger swallowtail, P. g. glaucus L. and P. g. australis Maynard, exhibit a female sex-limited wing color dimorphism (The dimorphism does not occur in P. g. canadensis Rothschild & Jordan). One female form resembles the male in having the typical pattern of a black-banded yellow background; the other female form is heavily melanized, with the banding pattern virtually obscured by dark scales. The dark female morph is thought to mimic the unpalatable Battus philenor (L.) ( e.g ., Brower, 1958), and while both dark and yellow female morphs occur throughout the eastern U. S., the dark form is more frequent where Battus philenor is abundant (Brower and Brower, 1962). 180 J.Res.Lepid. Clarke and Sheppard (1957, 1959, 1962) and Clarke et. al. (1976) have provided compelling evidence that melanism in Papilio glaucus is controlled by a female-limited gene, presumably associated with the Y (W) chromosome. This is supported by the absence of the melanic form in males, and the fact that in virtually all cases, females produce daughters of the same color morph as themselves. Rare exceptions do occur in which both yellow and dark female progeny arise from a single mother ( e.g ., Edwards, 1884; Weed, 1917; Clarke and Sheppard, 1959). Classically, these mixed broods have been regarded as the result of abnormalities in chromosome architecture or meiotic processes, but a novel explanation for certain cases was suggested by Scriber and Ritland (in press). These authors described a genetic component in the monomorphic subspecies P. glaucus canadensis that completely sup¬ presses phenotypic expression of the dark morph in hybrid offspring from laboratory crosses between male P. g. canadensis and dark morph female P. g. glaucus. Scriber and Ritland argued that in some cases, anomalous dark morph inheritance patterns may be the result of natural hybridization between P. g. glaucus and P. g. canadensis. The rare occurrence of analagous mutant alleles in P. g. glaucus and P. g. australis may explain other cases of unusual inheritance. Occasionally, Papilio glaucus females exhibit wing patterns in¬ termediate between the normal yellow and dark morphs (i.e., with a dusting of yellow scales in the dark background). The occurrence of yellow-dark intermediate individuals is an entirely separate phe¬ nomenon from the mixed broods described above. The intermediate female phenotypes are poor mimics of Battus philenor; Clarke and Sheppard (1959) postulated the presence of an efficient genetic “switch mechanism” in P. glaucus (presumably a single gene controlling mela- nization) which prevents the occurrence of these nonmimetic in¬ termediates. Intermediate females of P. glaucus with significant yellow suffusion of the dark background are uncommon, but have been recorded from many areas in the eastern United States: New York (Edwards, 1884; Shapiro and Shapiro, 1973); New Jersey (Clarke and Clarke, 1983); Ohio (M. H. Evans, pers. comm.); West Virginia (Ed¬ wards, 1884); Virginia (Clark and Clark, 1951); Maryland (Clark and Clark, 1932); Pennsylvania (Shapiro, 1966; Ehle, 1981); Wisconsin (pers. obs); Mississippi (B. Mather, pers. comm.); Kentucky (pers. obs.); Georgia (Harris, 1972); and Florida (pers. obs.). Dark morph females with at least a slight suffusion of yellow scales probably occur in low frequency throughout the eastern United States. The general rarity of intermediate females in wild populations sug¬ gests that the dark mimetic phenotype of P. glaucus is strongly canalized (buffered) under normal environmental conditions. In¬ termediate females may arise because of either genetic shock (e.g., mutant alleles or incomplete penetrance/expressivity of normal alleles controlling melanization) or environmental shock (disruption of the 25(3): 179-187, 1986(87) 181 canalized developmental pathway by unusual environmental condi¬ tions). The present study investigates phenotypic plasticity in dark morph Papilio glaucus females as a function of one environmental variable, temperature. Phenotypic plasticity and canalization of the dark morph are discussed in relation to mimicry in this butterfly. I hypothesize that canalization of the dark morph is adaptive because it stabilizes the mimetic resemblance to Battus philenor, and that phe¬ notypic stability may be more strongly selected for in areas where B. philenor is an abundant model. Methods Experiments conducted in 1981, 1983 and 1984 investigated the effect of rearing temperature on wing coloration in samples of Papilio glaucus from eight geographic areas: Dane County, WI; Dauphin County, PA; Adams County, OH; Mercer County, WV; Bell County, KY; Jefferson County, AL; Oconee County, GA; and Alachua County, FL. Laboratory cultures were established and ova for the study were obtained from dark morph females which had been mated to male siblings by the hand-pairing method of Clarke and Sheppard (1956). Females ovipo¬ sited on foodplant leaves in plastic shoeboxes warmed by incandescent lights. Newly-eclosed larvae were transferred to environmental chambers and reared at one of three constant temperatures: 22, 25, or 28 C. Temperature readings taken at different locations within each cham¬ ber indicated fluctuations of less than 0.5 C. All treatments were maintained at a photoperiod of 16L:8D to inhibit diapause and to remove photoperiodic variability as a relevant factor. The larvae were fed leaves on excised twigs of Black Cherry, Prunus serotina Ehrh. Foodplant turgidity was maintained by placing the twigs in Aquapics. Pupae were kept in individual screen cages at the larval rearing temperature. All female progeny from this experiment were expected to exhibit the normal dark morph phenotype. To describe deviation from the normal dark pattern, the dorsal background color of each reared female was scored relative to a group of five reference specimens. These reference specimens represent five points on a continuum ranging from a normal dark morph female (assigned a rating of ‘O’) to an intermediate yellow-dark phenotype (rating = 4) which has a heavy suffusion of yellow dusting in the dark background, giving the butterfly a ‘sooty’ appearance (Figure 1). Reared females were compared to this reference group and assigned an appropriate score. The rating scale ranged by half steps from 0 to 4. The modification of the dark morph pattern at different rearing temperatures was investigated statistically via the Kruskal- Wallis one-way ANOVA for ordinal data (Siegel, 1956). This procedure com- 182 J. Res. Lepid. Figure 1. Reference specimens of Papi/io g/aucus showing grading scale used to quantify dorsal wing color of experimental specimens. pared median color rating among the three rearing temperatures within each geographic sample. Results A total of 281 dark morph females from the eight geographic samples were scored for dorsal wing background color. Table 1 presents the median color ratings and range of individual scores for each geographic sample at three rearing temperatures and the associated statistics. These data indicate that higher color ratings (greater suffusion of yellow scales in the dark background) occurred under high rearing temperature regimes; i.e., there was significant modification of dark morph expression at 28 C relative to the two lower temperatures. In addition, the eight geographic samples differ significantly from one another in the degree of phenotypic modification at 28 C (Kruskal- Wallis ANOVA, H = 9.4, p < .01). Intrasample variability is relatively high at 28 C: most samples reared at this temperature contained individuals ranging over at least two full steps on the color scale (Table 1). Such individual variation in suscepti¬ bility to environmental modification (or canalization of the normal dark color pattern) may represent individual differences in the suite of modifier genes which protects the normal phenotype (Waddington, 1961). Wing pattern elements other than the melanic background ( e.g , 25(3): 179-187, 1986(87) 183 Table 1. Median color scores and range of individual values for eight samples of dark female PapiHo g/aucus reared at three constant temperatures. Kruskal-Wallis test statistic (H) and significance level for differences in color rating among the three temperatures are indicated for each sample. 22 C 25 C 28 C Sample median (range) median (range) median (range) H P Wl 0.0 (0.0-0.0) 0.0 (0.0-0.5) 3.0 (1. 0-4.0) 16.5 .001 OH 0.0 (0.0-0.0) 0.0 (0.0-2. 0) 0.5 (0.0-3.5) 8.9 .05 PA 0.0 (0.0-0.0) 0.0 (0.0-0.0) 1.8 (0.0-2.5) 21.4 .001 AL 0.0 (0.0-0.0) 0.0 (0.0-0.0) 0.8 (0.0-1.0) 11.5 .01 WV 0.0 (0.0-0.0) 0.0 (0.0-0.0) 0.3 (0.0-2.5) 4.4 .20 (N.S.) KY 0.0 (0.0-0.0) 0.5 (0.0-2.0) 2.0 (1.0-2. 5) 8.8 .05 GA 0.0 (0.0-0.0) 0.3 (0.0-0. 5) 0.5 (0.0-3. 5) 12.5 .01 FL 0.0 (0.0-0.0) 0.0 (0.0-1.0) 0.0 (0.0-3.0) 12.1 .01 the “tiger” stripes and wing margin borders) were virtually unaffected by temperature. The yellow fore wing discal spot present in some females (see Figure 1) becomes more pronounced at higher rearing temperatures (Ritland, 1983), but varies independently of melanic background color in individual butterflies. Discussion Constant rearing temperatures of 25 and 28 C destabilized the dark morph phenotype of Papilio g. glaucus and P. g. australis. A previous experiment (Ritland, 1983) suggested that pattern development is susceptible to temperature modification only during the pupal stage; this is consistent with the suggestion (Clarke and Clarke, 1983) that the melanic background pattern develops just before adult eclosion. The physiological basis of aberrant intermediate pattern development is not known, but many processes involved in wing pattern develop¬ ment (including pigment synthesis, wing scale maturation, and hor¬ monal control systems) are subject to modification by temperature (Goldschmidt, 1938; Hintze-Podufal, 1977; Nijhout, 1980). The temper¬ ature sensitivity of tyrosinase-mediated melanization processes in par¬ ticular is well known (Waddington, 1961; Fuzeau-Braesch, 1972; Ma- jerus, 1981), and high rearing temperatures may also disrupt the pteridine pigment system involved in P. glaucus pattern development (Oldroyd, 1971). Aberrant intermediate phenotypes were expressed only in individuals reared at 25 C and above, suggesting the existence of a temperature threshold above which canalization of the normal dark phenotype breaks down. Developmental pathways are protected by such genetically-determined thresholds (Waddington, 1961), thereby cana¬ lizing the normal phenotype over a wide range of natural conditions. This experiment did not investigate photoperiodic effects on pattern modification in Papilio glaucus , but photoperiod is potentially relevant 184 J . Res. Lepid. i in the field. Long and short photoperiods induce different seasonal forms and aberrations in many butterfly species (e.g., Ae, 1957; Pease, 1962; Fukuda and Endo, 1966; Shapiro, 1976; but cf. McLeod, 1968 and Lewis, 1985 re species which are insensitive to photoperiodic manipula¬ tion). The genetic capability to produce the intermediate phenotype repre¬ sents a component of the P. glaucus genome which is not normally expressed, probably due to a combination of the genetic switch mechan¬ ism proposed by Clarke and Sheppard (1959) and developmental cana¬ lization. While the experimental conditions of this study (24 hr thermo¬ period -I- 16:8 photoperiod) do not represent natural conditions, the range of rearing temperatures certainly lies within natural limits. This experiment is therefore qualitatively different from “shock” studies, in which newly-formed pupae are exposed to extreme heat or cold. Such shock treatments can produce striking pattern modifications, but also kill or cripple the majority of individuals, suggesting that critical developmental pathways are disrupted. Changes in wing pattern in¬ duced by such radical conditions may be of questionable ecological relevance. In sharp contrast to shock studies, the relatively mild conditions of the present investigation produced aberrant wing pat¬ terns but did not significantly reduce survival or adult viability (no significant difference in viability among the three temperature regim¬ es; chi-square p < .01). It is significant that such moderate ex¬ perimental conditions could produce such extreme phenotypic modifica¬ tion, given the fact that intermediates are so uncommon in the wild. This intriguing situation is similar to that described by McLeod (1968), who found that the African nymphalid Precis octavia, which exhibits discrete seasonal forms in nature, produced a wide variety of in¬ termediate forms in his laboratory temperature studies. Environmental modification of wing pattern may disrupt mimicry in dark morph Papilio glaucus females; the intermediate phenotypes produced at 25 and 28 C appear to be very poor mimics of Battus philenor. The eight geographic samples in this study differed signi¬ ficantly in expression of the intermediate phenotype at 28 C (Table 1). Both the proportion of aberrant individuals and the degree of phenoty¬ pic alteration varied between samples. Samples from the periphery of the dark morph range, where Battus philenor is uncommon {e.g., Wisconsin and Pennsylvania) were relatively susceptible to tempera¬ ture modification (as indicated by the high median color ratings at 28 C). In contrast, samples from areas where B. philenor is abundant (West Virginia, Georgia, Alabama, north Florida) seemed to be more strongly canalized (buffered) against environmental modification. The West Virginia sample, in fact, showed no evidence of phenotypic modification by temperature. These results are consistent with the hypothesis that canalization of the dark morph is adaptive because it stabilizes the mimetic color pattern, and that the dark phenotype is most strongly canalized in 25(3): 179-187, 1986(87) 185 areas where it confers the greatest mimetic advantage, i.e., where Battus philenor is abundant as a model. In regions where B. philenor is rare and is therefore not an effective model, the selective advantage of the dark morph relative to the yellow morph is decreased; selection for genetic modifiers which canalize the dark morph developmental path¬ way should also be reduced. It is significant that many of the records for wild intermediates occur near the periphery of the dark morph range, where B. philenor is rare. The occasional occurrence of wild intermediates of P. glaucus may be due to either environmental influences (environmental shock) or direct genetic control (genetic shock). Microhabitat selection by pupating larvae (e.g., West and Hazel, 1979) may occasionally result in exposure to high temperatures which disrupt the normal dark morph develop¬ mental pathway and cause expression of the intermediate phenotype. Alternatively, mutant alleles may alter the canalization threshold of the normal dark morph (i.e., change the developmental pathway), such that the intermediate phenotype is expressed under normal environ¬ mental conditions. Such alleles might be related to the gene(s) in P. glaucus canadensis that inhibit expression of the normal dark morph (Scriber and Ritland, in press). Similar inhibitory genes have been described in Papilio rutulus ; hybrid crosses between male P. rutulus and dark morph female P. glaucus produce intermediate daughters (Clarke and Willig, 1977) that resemble the environmentally-produced intermediates (phenocopies) described in this study. The interaction of genetic and environmental factors affecting pattern development in Papilio glaucus may significantly alter the resembl¬ ance to Battus philenor. The data presented in this paper support the hypothesis that canalization of the dark female morph stabilizes the mimetic color pattern under normal environmental conditions, and that geographic variation in the degree of phenotypic canalization is correlated with the abundance of Battus philenor. Acknowledgements. The author gratefully acknowledges the insightful con¬ tributions of J. M. Scriber, Walter Goodman, and Robin L. Ritland. I am especially grateful to Lincoln P. Brower for extensive comments, and to Arthur M. Shapiro and Thomas C. Emmel for very helpful perspectives. Mark Evans, Robin Ritland, and Jane Schrimpf provided invaluable assistance in the labora¬ tory. This study was funded in part by grants from the National Science Foundation (DEB 7921749, BSR 8306060 to J. M. Scriber) and the University of Wisconsin, Madison (Hatch Project 5134). Literature Cited AE, S. A. 1957. Effects of photoperiod on Colias eurytheme. Lepid. News 11: 207-214. BROWER, J. V. Z. 1958. Experimental studies of mimicry in some North American butterflies. II. Battus philenor and Papilio troilus, P. polyxenes, and P. glaucus. Evolution 12: 123-136. BROWER, L. P. & J. V. Z. BROWER. 1962. The relative abundance of model and mimic 186 J.Res.Lepid. ! butterflies in natural populations of the Battus philenor mimicry complex. Ecology 43: 154-158. CLARK, A. H. 1932. The butterflies of the District of Columbia and vicinity. Bull. U. S. N. M. 157: 172-199. CLARK, A. H. & L. F. CLARK. 1951. The butterflies of Virginia. Smithsonian Inst. Misc. Coll. 116 (7). 239 pp. CLARKE, C. A. & F. M. M. CLARKE. 1983. Abormalities of wing pattern in the eastern tiger swallowtail butterfly, Papilio glaucus. Syst. Ent. 8: 25-28. CLARKE, C. A. & P. M. SHEPPARD. 1956. Hand pairing of butterflies. Lepid. News 10: 47-53. CLARKE, C. A. & P. M. SHEPPARD. 1957. The breeding in captivity of the hybrid Papilio glaucus female x Papilio eurymedon male. Lepid. News 11: 201-205. CLARKE, C. A. & P. M. SHEPPARD. 1959. The genetics of some mimetic forms of Papilio dardanus and Papilio glaucus. J. Genetics 56: 236-260. CLARKE, C. A. & P. M. SHEPPARD. 1962. The genetics of the mimetic butterfly Papilio glaucus. Ecology 43: 159—161. CLARKE, C. A., & P. M. SHEPPARD, & U. MITTWOCH. 1976. Heterochromatin poly¬ morphism and colour pattern in the tiger swallowtail butterfly, Papilio glaucus L. Nature 263: 585—586. CLARKE, C. A. & A. WILLIG. 1977. The use of a-ecdysone to break permanent diapause of a hybrid between Papilio glaucus L. female and Papilio rutulus Lucas male. J. Res. Lep. 16: 245-248. EDWARDS, W. H. 1884. The Butterflies of North America, vol. 2. Houghton Mifflin Co., Boston, 357 pp. EHLE, G. 1981. Letter to the editor. NEWS of the Lepidopterists Society (Sept.- Oct. 1981): p. 63. FUKUDA, S. & ENDO. 1966. Hormonal control of the development of seasonal forms in the butterfly Polygonia c-aureum L. Proc. Japan. Acad. 42: 1082-1087. FUZEAU-BRAESCH, S. 1972. Pigments and color changes. Ann. Rev. Entom. 17: 403-424. GOLDSCHMIDT, R. 1983. Physiological Genetics. McGraw-Hill Book Co.; New York HARRIS, L. JR. 1972. The Butterflies of Georgia. University of Oklahoma Press. Norman, Oklahoma. HINTZE-PODUFAL, C. 1977. The larval melanin pattern in the moth Eudia pavonia and its initiating factors. J. Insect Physiol. 23: 731—737. HOFFMANN, A. 1982. Punctuated versus gradual mode of evolution. Evol. Biol. 15: 411-436. LEWIS, J. E. 1985. Temperature induced seasonal melanism in the wings of Copaeodes minima (Lepidoptera: Hesperiidae). Fla. Entom. 68: 667-671. MAJERUS, M. E. N. 1981. The inheritance and maintenance of the melanic form nirescens ofPachycnemia hippocastanaria (Lepidoptera: Ennominae). Ecol. Ent. 6: 417-422. MCLEOD, L. 1968. Controlled environment experiments with Precis octavia Cram. (Nymphalidae). J. Res. Lep. 7: 1-18. NIJHOUT, H. F. 1980. Ontogeny of the color patterns on the wings of Precis coenia (Lepidoptera: Nymphalidae). Devel. Biology 80: 275—288. OLDROYD, S. N. 1971 Biochemical investigations of various forms of some Papilio species. Entomologist 104: 111 — 123. 25(3): 179-187, 1986(87) 187 PEASE, R. W. 1962. Factors causing seasonal forms in Ascia monuste (Lepid.). Science 137: 987—988. RITLAND, D. B. 1983. The effect of temperature on larval growth and adult phenotype in subspecies of Papilio glaucus (Lepidoptera: Papilionidae) and their hybrids. M. S. Thesis, University of Wisconsin; Madison, Wisconsin. SCRIBER, J. M. & D. B. RITLAND. in press. Hybridization as a causal mechanism of mixed color broods and unusual color morphs in the eastern tiger swallow¬ tail, Papilio glaucus. In M. Huettel (ed.), Evolutionary Genetics of Inver¬ tebrate Behavior. SHAPIRO, A. M. 1966. Butterflies of the Delaware Valley. Am. Ent. Soc. SSpecial Publ. 79 pp. SHAPIRO, A. M. 1976. Seasonal polyphenism. Evol. Biol. 9: 259-333. SHAPIRO, A. M. 1981. Phenotypic plasticity in temperate and subarctic Nymphalis antiopa (Nymphalidae): Evidence for adaptive canalization. J. Lepid. Soc. 35: 124-131. SHAPIRO, A. M. & A. R. SHAPIRO. 1973. The ecological associations of the butterflies of Staten Island (Richmond County, New York). J. Res. Lep. 12: 65-128. SIEGEL, S. 1956. Nonparametric Statistics for the Behavioral Sciences. McGraw- Hill Book Co.; New York. SONDHI, K. C. 1963. The biological foundations of animal patterns. Quart. Rev. Biol. 39: 289-327. WADDINGTON, C. H. 1957. The Strategy of the Genes. Allen and Unwin, London. WADDINGTON, C. H. 1961. Genetic assimilation. Adv. Genet. 10: 257-294. WEED, C. M. 1917. Butterflies Worth Knowing. Doubleday and Co.; New York. WEST, D. A. & W. N. HAZEL. 1979. Natural pupation sites of swallowtail butterflies (Lepidoptera: Papilionidae): Papilio polyxenes Fabr., P. glaucus L. and Battus philenor (L.). Ecol. Entom. 4: 387-392. Journal of Research on the Lepidoptera 25(3): 188-201,1986(87) Chromatic Polymorphism in Callophrys mossii bayensis Larvae (Lycaenidae): Spectral Characterization, Short- Term Color Shifts, and Natural Morph Frequencies Larry Orsak1 and Douglas W. Whitman2 Dept. Entomological Sciences, University of California, Berkeley, CA 94720 Abstract. A tristimulus colorimeter and UV-VIS spectrophotometer supplemented visual assessments of color polymorphism in wild fourth instar larvae of the endangered butterfly Callophrys mossii bayensis. Wild larvae are of many color hues; this contrasts with the distinct morphs reported from laboratory rearings. Larval color changed over short time periods when fed yellow flowers or red bracts. The precise¬ ness of visual color matching between larvae and plant substrates is higher for red than for yellow larvae. This crypsis does not extend to any precise mimicry of spectral reflectance. Genetic color-determining mechanisms seem to be supplemented by an environment-derived factor in producing the broad range in color hues found in wild larvae. The color-assessment techniques described here could be used to better understand the role of color pattern in thermoregulation, sexual selection and predation- avoidance. Introduction Body color is a universal life attribute that influences intraspecific communication, predator avoidance, and/or thermoregulation. Syste- matists use color patterns to characterize species and subspecies, especially in avian and lepidopteran taxa. Despite these important roles, color patterns are usually qualitatively described, not quantitatively characterized. Partly, this is due to the difficulty in quantifying and standardizing color description. Color standard texts (e.g., Munsell, 1963) are useful, but not widely accessible. Each text uses different descriptors and their value is limited mainly to mono-colored organisms. An added complexity is the variation in color pattern within popula¬ tions. This is particularly apparent in the Lepidoptera, with color polymorphism occurring in LARVAE (e.g., Poulton, 1888; Bell & Scott, 1937; Pinhey, 1960; Clarke, Dickson & Sheppard, 1963; Curio, 1965, institute of Ecology, University of Georgia, Athens, GA 30602 2Dept. Entomology, University of Georgia, Athens, GA 30602 25(3): 188-201,1986(87) 189 1970a, 1970b, 1970c; Boer, 1971; Emmel & Emmel, 1973; Common & Waterhouse, 1981; see also Edmunds, 1974), PUPAE (Poulton, 1890; Sims & Shapiro, 1983, and citations therein), and ADULTS (Ford, 1955; Kettle well, 1961; Clarke & Sheppard, 1963, 1972; Owen & Chanter, 1969; and citations in Wickler, 1968 and Rettenmeyer, 1970). Larvae of the federally endangered (U.S. Fish & Wildlife Service, 1976) San Bruno elfin butterfly, Callophrys mossii bayensis R. W. Brown exhibit a striking color polymorphism, with chromatic variability of both larvae and foodplant substrates. This provides an ideal situation for comparing color assessment techniques. Callophyrs mossii bayensis Color Morphs Brown (1969) first described color morphs in third and fourth instar C. m. bayensis. He felt that greenish, fresh-hatched larvae acquired the same color as the Sedum spathulifolium Hooker foodplant part they ingested. Sedum exhibits diverse colors in late spring when larvae are near maturity: Basal leaf rosettes range from deep green to rosy red. Flowering stalk stems and bracts are initially green, becoming pale to rosy, or deep red; petals are yellow. Brown’s assessment of color determination was disputed by Emmel and Ferris (1972), who described three distinct color morphs from laboratory-reared fourth instars fed only green Sedum rosettes: yellow, pale orange, and cherry red. Arnold (1978, 1983), in turn, disputed the concept of three distinct morphs: “Newly eclosed larvae were colored either red or yellow. They remained one color throughout their larval life,” and “larvae possess two distinct color morphs, red and yellow, plus an intermediate light orange.” Lumping light orange and yellow larvae, Arnold proposed a simple 1:3 allelic expression of yellow:red forms, and equated laboratory and field expression of larval color. Finally, our repeated field observations of an array of color forms conflicts with all previous reports of two or three distinct morphs in nature. Clearly, there are discrepancies regarding the expression of color, its stability, and its derivation in C. m. bayensis. This paper seeks to resolve some of them. Materials and Methods C. m. bayensis and Sedum spathulifolium samples were obtained on north-facing slopes of San Bruno Mountain (San Mateo County) Califor¬ nia between Brisbane and Colma Canyon. Larvae occur from about mid-March to very early June. About mid-May, third and fourth instar larvae ascend to budding Sedum flower stalks (Emmel & Ferris, 1972; Arnold, 1983). We took food-plant and fourth (penultimate) instar samples after ascent. COLOR CLASSIFICATION SCHEME: Following a preliminary 1977 field examination, a scheme was developed to quickly color-sort wild larvae: Seven larval “standards” (Fig. 1) divided the visual color range of 190 J. Res. Lepid. i wild larvae. These were sequentially numbered; the higher the number, the more red (or less yellow) the larva. These were photographed with Kodachrome 25 film, using two Sunpak 411 flashes. Subsequent Kodak color prints facilitated rapid color classification in the field. Although film color reproduction is inexact, we found no problem placing wild larvae into one of seven color categories. A “color category” represents a range between two points defined by the larval standards, except for Category 7 which had only one larval standard “anchor.” A larva whose general color fell anywhere between the discrete point of Standard 1 up to, but not matching Standard 2, was a Category 1 larva and so on. COLORIMETER ANALYSIS: Live larval and foodplant samples were color-analyzed using a Hunterlab Tristimulus Colorimeter Model D25M-9. This employs a source-photodetector-filter combination to simulate the colormatching response functions of a “normal” human observer. Quantifiable, repeatable results are in the form of the Lx ax b1 (henceforth, LAB) system (Hunter, 1975). “L” measures brightness (L = 100 for pure white, 0 for pure black). “A” and “B” are chromaticity dimensions. The value of “A” indicates redness ( + value), gray (0 value), and green (— value). “B” measures yellowness (+ value), gray (0 value) and blue (— value). Measurements were made by holding similar-sized samples of Sedum flowers and adjacent bracts, secondary bracts, green rosettes, or C. m. bayensis penultimate instar larvae against the 1/2-inch diameter port. SPECTRAL ANALYSIS: A Cary UV-VIS spectrophotometer with spec¬ tral capacity of 187 — 875 nanometers (nm), and equipped with a diffuse reflectance sphere, was used. Larval and foodplant samples were affixed in similar orientation on coal black cards with double-stick tape. Each sample was scanned at 1 nm/second, with a spectral band width of 3.5 nm, allowing resolution of narrow reflectance peaks. To reduce sample orientation effects, all samples were positioned similarly. After scan¬ ning, larvae were released unharmed by wetting the double-stick tape. LARVAL COLOR CHANGES: To explore short-term color changes, fourth instar larvae with previous access to all Sedum plant parts were segregated into color categories using the seven standards. Free access to all foodplant parts was maintained under low intensity fluorescent lighting. Forty-eight hours later, the larvae were color-reclassified. Only tachinid parasitoid-free larvae (assayed at pupation) were used in the data analysis. We investigated Brown’s (1969) statement that larval and ingested food colors converged: Larvae that had ingested only green rosettes for two days were grouped into pairs of identically colored larvae and color-classified. For the following 48 hours, one member of each pair was provided only yellow Sedum flowers; the other was given only very red flower stalk bracts. All experienced the same fluorescent light exposure. Pairs then were reunited and color-compared, using the larval stan¬ dards. Only parasitoid-free larvae were used in the data analysis. 25(3): 188-201, 1986(87) 191 Fig. 1. Seven larval "standards" used to characterize color polymorphism in wild Callophrys mossii bayensis. Standards are labeled sequentially, starting with the most yellow (Top row, 1-4; bottom row, 5-7). Fig. 7. (LEFT BELOW) Larval color shift from light to dark over a 48-hour period. LEFT: Flower-fed larva now in color Category 3, formerly Category 2; RIGHT: Red bract-fed larva, unchanged in Category 2. Fig. 8. (RIGHT BELOW) Larval color shift from dark to light over a 48-hour period. LEFT: Flower-fed larva now in color Category 5, formerly Category 6; RIGHT: Red bract-fed larva unchanged in Category 6. 192 J. Res. Lepid. i Results QUALITATIVE DESCRIPTION OF LARVAL COLOR: Nearly 500 wild larvae were color-classified. Virtually none were lighter yellow than Standard 1; some had less pronounced “chevrons,” the paired, dorso¬ lateral curved markings occurring on many body segments. Category 7 proved exceptionally restrictive, since few Category 7 larvae were redder than Standard 7. Larvae within each color category can be generalized as follows: 1 = Yellow, no peach tint; chevron markings generally faint 2 = Yellowish with faint orange tint; distinct chevrons. 3 = Distinctly light orange with slightly darker rosy suffusions; chevrons usually with pale outlines. 4 = Orange with darker peach-colored suffusions on much of the body; chevron outlines and dorsal midline generally pale. 5 = Orange with brownish tinge; dark chevrons and less distinct pale outlines. 6 = Rosy red, with less distinct but noticeable pale chevron outlines. Larvae in this category may be lighter colored than the previous category, but are distinctly redder. 7 = Cherry red throughout; chevrons generally faint. While color category designation was based upon general background color, ignoring fine-scale pattern differences, we also noted that chevron markings did not intensify in direct relation to increasingly red back¬ ground coloration (see Emmel & Ferris, 1972). COLOR DISTRIBUTION IN NATURE: Fig. 2 shows the color distribu¬ tion of 433 wild larvae, comparing the results to distributions obtained by Emmel and Ferris (1972) for wild larvae, and Arnold (1978) for laboratory-rearings. This alignment easily satisfied the “morph” de¬ scriptors provided by each author for his respective sample. Our categories 2 and 3 are the only ones that would fit the definition of “light orange” ( sensu Arnold, 1978). Our sample indicates larvae to be broadly distributed across color categories, at these frequencies: 1 = 6.0%; 2 = 10.6%; 3 = 14.6%; 4 = 13.9%; 5 = 12.0%; 6 = 24.7%; 7 = 18.3%. Further, our sample yielded lower frequencies of “pure” yellow (Category 1) and red (Category 7 and possibly 6) larvae than reported for laboratory rearings. Conversely, greater frequencies of “intermediate” colors were found. Even when restricting “intermediates” to larvae of categories 2-3, our combined frequencies of yellow plus “light orange” larvae (over 30%) exceed that of the laboratory-reared sample (24.4%; Arnold, 1978). Unfortunately, Arnold did not segregate frequencies for yellow and light orange larvae. Yet, he states that only “a few individuals are light orange” (Arnold, 1983), which tends to corroborate our observations of rosette-reared larvae, and by deduction, confirms the much rarer occurrence of “pure yellow” larvae in nature. 25(3): 188-201, 1986(87) 193 EEC ] Fig. 2. Color category distribution of C. m. bayensis larvae: Comparisons of wild- and laboratory-derived samples, from (A) Arnold, 1978; (B) Emmel & Ferris, 1972; (C) this study. Alignments are based upon each author's color descriptions of cited morphs. Frequency figures within each bar pertain to that study. Above-barfrequencies in (B) and (C) are provided to facilitate cross-sample comparisons. COLORIMETER VALUE COMPARISONS: Table 1 lists LAB values for C. m. bayensis larval standards and foodplant samples. Predictably, “A” values are higher for the “redder” larval standards, although the most yellow standard reflects some red. The “redness” increase, mea¬ sured by “A,” changes little for standards 2-5, increasing for 6 and 7. In contrast, “yellowness,” measured by “B,” drops in significant increments through standard 5. Thus, while larval standards broadly cover the yellow-to-red spectrum, they do not represent evenly spaced color points for either chromaticity dimension. This is clear in the composite express¬ ion of color (subtracting “A” from “B” for each sample; Table 1). These findings, however, again confirm a graded expression of color in wild larvae within their spectral range. The range in larval LAB values approximates that of foodplant parts (Table 1). However, pale or yellowish flowering stalk substrates (flowers + associated racemes) exhibit significant green colorimeter values (i.e., negative “A”) that are not duplicated in any larval standards or in Category 1 larvae. In contrast, there is close matching of “A” and “B” values for the reddest Sedum samples and larvae. SPECTRAL VALUE COMPARISONS: Sample orientation (Fig. 3), brightness, and other features of the sample can affect absolute reflect¬ ance values. However, basic reflectance peaks and dips are relatively 194 J. Res. Lepid. Table 1 . LAB colorimeter values for Cal/ophryr mossii bayensis larval stan¬ dards and Sedum spathulifolium foodplant samples. Variation of each sample did not exceed 10% between readings as long as orientation positions were kept constant. Sample Color Values L A B i (B-A) Larva: Color Standard 1 37.52 2.81 12.17 9.4 Color Standard 2 36.51 4.01 10.21 6.2 Color Standard 3 35.52 4.53 8.95 4.4 Color Standard 4 35.02 4.68 6.65 2.0 Color Standard 5 34.08 4.91 4.37 -0.5 Color Standard 6 34.30 6.83 3.98 -2.9 Color Standard 7 32.94 7.72 3.76 -4.0 Sedum Flowers: Sample 1 35.27 1.80 24.13 Sample 2 40.59 3.45 15.36 Sample 3 45.22 -2.05 19.70 Sample 4 40.55 .70 17.33 Sample 5 36.73 -1.74 18.22 Sedum Leaves: Sample 1 33.17 -6.03 13.01 (green) Sample 2 40.88 -6.82 12.42 Sample 3 47.21 -5.90 13.60 Sample 4 38.74 -2.85 5.59 Sedum Bracts: Sample 1 33.14 9.08 3.56 (reddish) Sample 2 28.23 6.27 3.55 Sample 3 35.41 4.33 4.97 Sample 4 38.10 7.44 3.34 Sample 5 44.20 4.97 2.19 Cumulative Spectral Ranges: L A B Larvae . 32-37 2.8- -7.7 3.8 -12.2 Foodplant Rosettes . 28-47 -6.0- -9.1 2.2 -24.1 Flowering Stalk . 28-45 -2.1- -9.1 2.2 -24.1 nanometers Fig. 3. Reflectance spectra of a Sedum spathulifolium leaf sample placed at different angles in the spectrophotometer diffuse reflectance sphere sample port. % reflectance 25(3): 188-201, 1986(87) 195 constant. For this reason, we limited our spectrograph comparisons to reflectance curves (slopes). In the ultraviolet and infrared (Fig. 3), no distinguishable intersample differences were found. All samples reflected strongly in the red and near-infrared range but weakly in the ultraviolet. Figure 4 A shows reflectance curves for the larval color standards. Predictably, Standard 1 (yellow) reflects highly between 600-630 nm, while Standard 7 (red) exhibits a sharp reflectance drop at wave-lengths under 640 nm. Reflectance differences are less pronounced for standards that are closer in visual color. In Figure 4B, the curves are aligned at 550 nm to show relative similarities. Standards 4—7 reflect almost identical¬ ly below 550 nm. Standards 1—3 show similar reflectance patterns but greater variation, especially Standard 1. Regardless of visual appearance, S. spathulifolium samples (Fig. 5) usually exhibited a broad absorbance peak (reflectance dip) at 670- 680 nm. Dry flower stalk stems (Fig. 5: ST) were the only exception. Larvae that matched standards 1 and 7 were run sequentially with like-colored Sedum parts (yellow blossoms or deep red flower head bracts (Fig. 6). Visual colors are not backed by fine-scale spectral reflectance mimicry. Most conspicuously, the foodplant reflectance dip at 670—680 nm is absent. As seen in the colorimeter data, larval Standard 1 does not show the strong yellow-green reflectance peak of the Sedum flower sample. Moreover, while the slope of Standard 7 and its red Sedum counterpart are very similar between 600—630 nm, at shorter wavelengths larvae have relatively higher reflectance values. Possibly, these plant-caterpil¬ lar differences are partly derived from structural disparities. COLOR SHIFTS IN INDIVIDUALS: Over one-third of larvae given free access to foodplant parts changed color over a two-day period (Table 2); some changed within the first day. Color shifts occurred in larvae of all color categories examined (1-5), with Category 4 a possible excep¬ tion. All but one (a shift from 5 to 3) of the 16 color shifts spanned one category. As defined here, a larva “color shifts” by crossing at least one color anchor (defined by a larval standard). However, twro larvae experiencing the same “one category” change, in reality, may have shifted significantly different amounts. Table 3 summarizes color shifts of 34 color-matched pairs, after being fed different-colored Sedum parts. Fifteen out of 68 larvae colordiverged from their pair mates; two examples, in different spectral directions, are illustrated in figures 7 and 8. As in the previous experiment, a Fig. 4. Reflectance spectra of C. m. bayensis larvae. A: Scans from "most yellow" (Standard 1 ) to "most red" (Standard 7) larvae in the 400-700 nm range; samples were of like size, so major differences in absolute reflectance values between samples are valid. B: Realignment of the seven larval standard scans at 550 nm show relative similarities of curves. 400 nanometers 700L ^ \X VjnJ % reflectance -*■ to