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

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

Journal of the Lepidopterists' Society 33(1), 1979, 6-20

EXPERIMENTAL HYBRIDIZATION BETWEEN PHYCIODES THAROS AND P. BATESII (NYMPHALIDAE)

Charles G. Oliver1 R. D. 1, Box 78, Scottdale, Pennsylvania 15683

ABSTRACT. Fi hybrids and backcrosses were made between the nymphalid butterflies Phyciodes thaws and P. batesii. The two species differ in larval, pupal, and adult phenotypic appearance, ecology, and larval diapause response. Genetic incompatibility was shown by significant hybrid inviability, growth irregularities, and abnormal adult sex ratios and development times. The reciprocal Fi hybrids differed greatly in their expression of incompatibility. Hybrid inviability is attributed to breakdowns in the genetic mechanisms controlling growth and development.

The relationship of Phyciodes tharos Drury to P. batesii Reakirt has been little understood despite the fact that they both occur in comparatively densely populated areas of the northeastern United States and have been known to be distinct for well over a century. This confusion appears to be due to two causes. First, P. tharos is common to abundant over the entire range of P. batesii. Since P. batesii occurs in widely separated, small populations, it is probably often overlooked due to its superficial resemblance to P. tharos. Second, the biology of P. tharos itself actually is poorly known. Rearing and hybridization studies now in progress in my laboratory indicate that "P. tharos" in the Northeast consists of two entities differing in larval and adult phenotypic appearance and voltinism and showing significant incompatibility when hybridized in the laboratory (Oliver, in prep.). The more southern entity, hereafter referred to as "Type A," occupies the Transition and Austral Life Zones, whereas the more northern "Type B" is the "P. tharos" of northern New England, northern New York State, and southern Canada. The phenotypic differences of Types A and B have resulted in Type B passing as P. batesii in many collections, although it resembles P. batesii little more than does Type A.

Phyciodes batesii is very local in the Northeast. It appears to be restricted to dry sites, often of the barrens type. The Onondaga Co., New York, population used in these experiments occurs on dry limestone ledges. In the Appalachians of Pennsylvania and West Virginia, the localities I have investigated are for the most part shale barrens or rocky riparian slopes. One of the best known localities, along the banks of the Ottawa River near Aylmer, Quebec, was described by McDunnough (1920) as the "lower dry slopes of a small ridge." Both Types A and B of P. tharos occur in a wide variety of habitats, including those of P. batesii.

1 Adjunct Assistant Professor, Dept. of Biology, West Virginia University, Morgantown, West Virginia 26506.

Volume 33, Number 1

7

Phyciodes batesii flies in early June in West Virginia and southwestern Pennsylvania and in mid June in central New York State. In West Virginia and southern Pennsylvania this is between the rather discrete first and second broods of P. tharos Type A, which has a total of three to four broods. In Onondaga Co., New York, however, P. batesii and P. tharos Type B fly together. Thus, the observation of Forbes (1960) that broods of P. tharos and P. batesii alternate at any given locality seems to apply only to the southern portion of the range of P. batesii.

The life history of P. batesii was described and figured by McDun-nough (1920). The larva and pupa differ in a number of characters from those of P. tharos (Table 1) and are much more like those of P. campes-tris, which were figured by Comstock (1930) and compared briefly to P. tharos in another paper of mine (Oliver, 1978). First and second instar larvae of both P. campestris and P. batesii live communally within a loose web spun over the feeding area on the foodplant. P. tharos has a similar communal behavior, but no web is spun.

McDunnough was able to obtain oviposition by P. batesii on "a species of Aster with heart-shaped leaves," found wild larvae on this plant, and successfully reared them through to adults. This Aster was probably A. undulatus L., which is very common in P. batesii habitats. Wild-caught P. batesii females from Onondaga Co., New York, refused to oviposit on Aster undulatus in the laboratory, but laid readily on A. simplex (Willd.) Burgess, a common foodplant of P. tharos. Newly-hatched larvae of both Phyciodes fed readily on A. undulatus when transferred to its leaves. Larvae of both species also accepted A. laevis L.

More than 200 unmated, laboratory-reared P. batesii adults were released shortly after eclosion during August into a western Pennsylvania old-field habitat containing abundant Aster simplex (but no A. undulatus). Several pairs were observed in copulo the next day. In early October a group of small larvae was recovered from a clump of A. simplex and reared through to normal adults the following spring, indicating that P. batesii may choose more than one species of Aster as natural foodplants.

In the northern Midwest (e.g., Michigan and Wisconsin) P. batesii has a somewhat different biology. Colonies occur in moist areas, and there is sometimes a partial second brood (Nielsen, in litt.). Populations in the Northeast seem to be strictly univoltine. Midwestern P. batesii differ slightly in appearance from those in the Northeast and may possibly represent a separate entity.

Procedure

Stock of P. batesii used in these experiments was derived from four wild-inseminated females collected 11 June 1976, in Syracuse, Onondaga

Table 1. Comparison of phenotypic appearance of P. thaws, P. batesii, and the

Character

P. tharos

P. batesii

I.  Mature larva

a.  Body ground color

b.  Tubercle color

c.  Tubercle tip color

d.  Dorsal light stripes

e.  White patches, head capsule

f.  Setal coloration

II.  Pupa

Overall appearance

III. Adult A. Dorsal

1. Black pattern expression

2.  Wing fringes color

3.  Body vestiture color

B. Ventral

1.  Antennal club color

2.  FW light ground color

3.  FW median dark spots

4.  Rel. sizes, lowest median vs. subapical black patches

5.  HW ground color

6.  Small submarginal dark spots ( S S only)

7.  Anal dark patch

Dark chocolate brown

Gray or dark brown

White

Faint, broken

Extensive dorsally and laterally

Light brown

Much more angular in outline; dorsal projections relatively pronounced

Relatively light; large open orange areas

Checkered light gray and dark

gray W. large proportion of tawny

hairs

Brown

Dark orange

Isolated, not forming band

Subapical > median

Brown w. pinkish tinge Light pinkish brown Concolorous light brown Compar. heavy, even Gr. restricted dorsally and

laterally Chocolate brown

Much more rounded in outline dorsal projections almost ab sent

Very heavy; wing area predom inantly black

Checkered white and dark gra

Almost all gray

Dark gray

Pale orange to straw yellow Tend to form continuous ban Median > subapical

Dark straw yellow                             Very pale straw yellow

Brown w. or w/o pinpoint black      Black

center

Large, often clouding silver      Usually only silver crescent

crescent                                              present

Volume 33, Number 1

9

Co., N.Y. Stock of P. thaws Type A was derived from four females taken 10 June 1976, in Acme, Westmoreland Co., Pennsylvania, and of Type B from two females taken 11 June 1976, at the Onondaga Co., N.Y. locality. Cultures were maintained at 15 to 28°C and separated into controlled photoperiod (18 h light/24 h or 24 h light/24 h) and natural photoperiod (for Fayette Co., Pa., latitude 40°N) groups. Artificial lighting after sunset was provided by a 100-watt incandescent bulb at a distance of 1 to 2 meters.

Matings were made by the hand-pairing method (Clarke, 1952). Fecundated females were provided with cut sprigs of Aster simplex for oviposition. Eggs were left in situ until hatching and the aster sprigs kept fresh in water. Larvae were reared on cut sprigs of A. simplex in water and housed in 10 X 20 cm glass cylinders.

The photoperiodically-regulated larval diapause in Phyciodes occurs at the beginning of the third instar. Diapausing larvae were removed from active cultures, placed in groups in 90 mm plastic petri dishes, sealed in airtight containers, and stored in a domestic refrigerator at 0 to 2°C until April or May of the following year. Upon removal from the diapause containers all larvae were maintained at 15 to 28 °C and 18 h light/24 h until pupation, at which time they were transferred to natural light conditions.

Fi progeny of wild-collected females was used for the hybrid pairings and as parental-type stock for backcrosses; no stock used was inbred. Observations were made on parental population and Fi hybrid phenotypic appearance, interspecific courtship behavior, development periods and adult eclosion patterns, fertility, adult sex ratios, embryonic, pupal, and adult viability, and on backcross embryonic viability. Controls were reared concurrently at all times for comparison with experimental broods.

Data on egg fertility, viability, and sex ratios were treated statistically using the Wilcoxon Two-sample Test. Adult fertility was measured by a count of the number of visibly developing eggs divided by the total number of eggs laid after a single mating. Development periods from hatching of the egg to eclosion of the adult were estimated by calculating the 99% confidence intervals for the medians of the distributions (Owen, 1962). Distributions of development times within broods or series of broods were represented by adult eclosion curves, graphs of the number of adults eclosing from pupae each day.

Results

Interspecific courtship behavior

Courtship in both P. thaws and P. batesii is apparently dependent on a variety of stimuli. The presence of a butterfly of appropriate coloration

10

Journal of the Lepidopterists' Society

Volume 33, Number 1

11

and size elicits approach by males. If the approached individual does not leave or, in the case of conspecific males, show aggressive behavior, as attempt at copulation will be made. Females which do not wish to copulate avoid the male's probing genitalia by dorsal or lateral movement of the abdomen. Males of both P. thaws and P. batesii showed the courtship approach response when presented with females of the other species. After an initial response, however, males of neither species attempted to copulate. Some stimulus, perhaps olfactory, appeared to terminate courtship.

Phenotypic appearance

Differences in phenotypic appearance of the fifth instar larvae, pupae, and adults of the parental species and Fi hybrids are summarized in Table 1. Typical specimens of adults are shown in Fig. 1. There was wide variation in adult phenotypic appearance among the Fi hybrid broods.

The artificial illumination levels used to extend daily photophase in these experiments eliminated facultative diapause in P. thaws (see below), but the induction of the naturally-occurring adult seasonal forms was little affected. The photoperiodic regulation of polyphenism in P. thaws has been described in a previous paper (Oliver, 1976). Univoltine P. batesii do not, of course, show natural seasonal polyphenism. Under the artificial long-day laboratory conditions described above, however, non-diapausing larvae produced late summer and early fall adults which had significantly heavier expression of the dark wing pattern elements on both the ventral and dorsal sides. This artificially induced form did not differ from the naturally occurring phenotype as the seasonal forms of P. thaws differ from each other. Fi hybrid adults emerging in September and October showed some expression of the short photophase phenotype but not to as great an extent as did the P. thaws controls.

Phyciodes thaws and P. batesii differ in the length of time required for full embryonic development. Eggs of P. thaws kept at natural outdoor temperatures during early August hatched after 6 days. Those of P. batesii required an additional day or day and a half. The hatching times of the Fi hybrids varied within broods from 6 to 71/£ days.

Fertility, viability, and sex ratio

Egg fertility and embryonic viability of control broods was very high (Table 2). Fertilizability of eggs was slightly reduced in the Fi hybrid

Fig. 1. Parental-type and Fi hybrid adults: Row A—P. thaws; B—P. batesii; C—Fi hybrids P. batesii $ X P. thaws $ ; D—Fi hybrids P. thaws $ X P. batesii $ . Specimens show, left to right, male dorsal, female dorsal, male ventral, female ventral.

Table 2. Mean egg fertility (fertile/laid) and embryonic viability (hatched/fertile) of P. tharo d backcross broods. Second decimal value indicates standard deviation. Tests of significance refer ls.

Mating

No. of broods

No. of eggs

Fertile/laid

Acme, Pa.

Syracuse, N.Y. : Syracuse, N.Y.

Y th 2 X ha $

2 X NY th $ 2 X Pa th $

Y  th2 X (NY th 2 X ha $ ) $ a 2 X NY th $)2 X NY th $

2 X (NY th 2 Xha $)$ NY th 2 Xha $)2 XNY th $

Y  th 2 X (ha 2 X NY th $ ) $

Volume 33, Number 1

13

crosses. Embryonic viability was heavily depressed in the P. hatesii 2 X P. thaws S Fi hybrid series, but not demonstrably affected in the reciprocal cross. Embryonic viability of the P. hatesii 2 X N.Y. P. thaws $ Fi hybrid series may differ from that of the P. hatesii 2 X Pa. P. thaws S series (P = .10). Backcross embryonic viability was drastically reduced. Embryonic viability of both Fi hybrids and backcrosses showed great variability among broods (Table 2).

Post-larval viability (i.e. during pupation and eclosion and as pupae) was greatly reduced in the P. hatesii 2 X P. thaws S Fi hybrids and to a lesser extent in the reciprocal cross. The decrease in viability was almost entirely during the pupal stage (Table 3).

Adult sex ratios of the P. hatesii 2 X P. thaws $ Fi hybrid series did not differ significantly from those of the parental broods (Table 3). In the two P. thaws 2 X P. hatesii $ broods, however, female adults were entirely absent (Brood 76-43) or greatly reduced in numbers (to 16.07% in Brood 76-42).

Structural abnormalities

Structural abnormalities involving segmental irregularities were relatively common in the P. hatesii 2 X P. thaws $ hybrid broods, though absent in the parental controls and reciprocal crosses. Between 0 and 15% (no exact counts made) of the larvae of each hybrid brood showed a lack of development of one side of an abdominal segment. The affected segment half was both narrower and shorter than the corresponding half, often lacked a tubercle, and resulted in the larval abdomen abruptly bending to one side. Two larvae were segregated and observed throughout development. In these, the semental irregularity persisted through the pupa and into the adult (Fig. 2). Both larvae produced apparently otherwise normal male adults.

Fi hybrid males from the cross P. hatesii 2 X P. thaws $ had disproportionately large, flaccid abdomens (Fig. 1). The genitalia of many individuals were apparently permanently extruded, and hand-pairings using any males were very difficult. No males would mate naturally in cages.

Voltinism The P. thaws Type A population culture from Westmoreland Co., Pa., showed no incidence of larval diapause when reared on a photoperiod regime of 18 h light/24 h or under natural photoperiod conditions during June and July. Broods reared under natural photoperiod during August and September showed a significant incidence of diapause (Table 4). The P. thaws Type B population culture from Onondaga Co., N.Y., on the

Table 3. Incidence (percentages) of inviability during and after pupation of New York hybrids and backcrosses. Tests of significance refer to comparisons of hybrids with con

Volume 33, Number 1

15

Fig. 2. P. batesii $ X P. thaws $ Fi hybrid male adult showing asymmetrical development of abdominal segment (arrow).

16

Journal of the Lepidopterists' Society

Table 5. Development times in days from hatching of egg until eclosion of adult for non-diapausing P. tharos and P. batesii control broods, Fi hybrids, and backcrosses. Medians with 99% confidence. See text for rearing conditions.

other hand, showed significant incidence of larval diapause when reared during June and July on natural photoperiod. This diapause response was entirely facultative, since there was a complete absence of diapause in larvae reared on 18 h light/24 h.

Phyciodes batesii reared on natural photoperiod during June and July showed a 100% incidence of larval diapause. Many of those reared on 24 h

Volume 33, Number 1

17

116 134

DAYS UNTIL ECLOSION

Fig. 3. Distributions of times required for development of New York State P. thaws, P. batesii, and Fi hybrid broods from hatching of eggs until eclosion of adults. A—P. thaws hatching in late June, natural photophase, B—P. batesii hatching in late June, 24 h light/24 h; C—Fi hybrid P. batesii 9 X P. thaws $ hatching in early August, 18 h light/24 h; D—Fi hybrid P. thaws $ X P. batesii $ hatching in early August, 18 h light/24 h.

light/24 h, however, developed without diapause, indicating that at least part of the culture was composed of facultatively diapausing individuals. There was no incidence of diapause in Fi hybrid larvae from the cross P. batesii 2 X P. thaws S. The reciprocal hybrid, however, had an incidence of diapause intermediate between those of the parental species (Table 4). Survival of the Fi hybrid larvae during diapause storage was normal compared with that of the parental species.

Development periods and eclosion patterns

The median development periods from hatching of the egg to eclosion of the adult were significantly longer for non-diapausing P. batesii than for concurrently reared non-diapausing P. thaws from New York State or Pennsylvania (Table 5). In addition, emergences were very scattered, producing a very different eclosion pattern from that of either population of P. thaws (Figs. 3A and B).

18

Journal of the Lepidopterists' Society

Table 6. Post-diapause development periods of P. tharos, P. batesii, and their Fi hybrid P. tharos $ X P- batesii $. Medians with 99% confidence. All larvae removed from storage on same day (4 April 1977).

Fi hybrids of the cross P. batesii 9 X P. tharos $ showed eclosion patterns intermediate between P. tharos and P. batesii, but male development periods about the same as those of P. batesii. Females of this cross had shorter development periods and tended to emerge before the males, rather than the normal reverse (Fig. 3C).

Male development times of the cross P. tharos 9 X P. batesii S were significantly shorter, though not as short as those of concurrently reared P. tharos. The spread of eclosion times for male adults was much more like that of P. tharos than P. batesii (Fig. 3D). All P. tharos $ X P. batesii S Fi adults emerging without diapause were males; all those emerging after diapause, females. Resumption of feeding after diapause of these larvae was much delayed after removal from cold storage, and growth was much slower than that of the parental controls (Table 6).

Discussion

It is clear from the results that P. tharos and P. batesii are well-differentiated species. There are marked differences in phenotypic appearance, voltinism, development rate, and ecology. The adults show strong behavioral isolating mechanisms during courtship in the laboratory. The heavy reduction in Fi hybrid viability and fertility and in backcross embryonic viability indicates a high degree of genetic incompatibility between the species.

Phyciodes tharos and P. batesii appear to have achieved rather similar phenotypes (compared to P. tharos and P. campestris, for example) by somewhat different genetic means. This is attested to by the wide range of phenotypic variation in the Fi hybrid adults and is especially marked in Fi females from the cross P. batesii $ X P. tharos S. Expression of the

Volume 33, Number 1

19

dorsal dark pattern elements ranges from as dark or darker than P. batesii to almost as light as P. tharos (Fig. 1). This indicates that the two species look more alike than they actually are.

I have discussed at length in another paper (Oliver, in press) the genetic basis of incompatibility effects involved in hybrid breakdown and surveyed the literature on viability of butterfly hybrids. The hybrid incompatibility shown between P. tharos and P. batesii may involve differences in the genetic control of hormones that direct growth and development. Disruption of normal hormonal control leads to formation of inviable embryos, abnormal tissue differentiation patterns, lessened fertility, lowered ability to pass from one life cycle stage to another, and abnormal development rates.

In general the genetic incompatibility between P. tharos and P. batesii is fairly similar to that shown between P. tharos and P. campestris mon-tana Behr (Oliver, 1978). Both sets of Fi hybrids have similar adult eclosion patterns. However, embryonic viability is reduced much more in the series P. tharos X P. batesii than in P. tharos X P. campestris. On the other hand, there is a much greater deficiency of Fi hybrid females in the series P. tharos X P. campestris than in P. tharos X P. batesii. These differences and those in ecology, voltinism, and so on indicate that P. batesii and P. c. montana are physiologically quite distinct, and P. batesii probably should not be regarded as an eastern representative of P. campestris. P. batesii is more specialized than either of the other species. It appears to have evolved from multivoltine stock by lowering of the threshold of diapause induction to include all naturally encountered photoperiod conditions. Although capable of feeding on at least several species of Aster, it has become closely associated with (though perhaps not restricted to) a single species, A. undulatus, and seems in the Northeast to be found only in the rather narrow habitat range of this aster. Other butterflies have followed similar courses of evolution. Pieris virginiensis Edw., for example, has evolved univoltinism in apparently the same way (Shapiro, 1971; author's unpub. data) and has become restricted to the narrow habitat of its single foodplant, Dentaria diphylla Michx.

Acknowledgments I greatly appreciate the help of Mr. Edward Jennejohn, Manlius, New York, in obtaining laboratory stock of Phyciodes batesii. Mr. Joseph O. Brenneman provided invaluable photographic assistance.

Literature Cited

Clarke, C. A. 1952. Hand pairing of Papilio machaon in February. Entomol. Rec. 64: 98-100.

20

Journal of the Lepidopterists' Society

Comstock, J. A. 1930. Egg, larva, and pupa of Phyciodes campestris Behr. Bull.

So. Calif. Acad. Sci. 29: 136. Forbes, W. T. M. 1960. Lepidoptera of New York and neighboring states. Part IV.

Memoir 371, Cornell U. Agric. Exper. Sta. McDunnough, J. 1920. Notes on the life history of Phyciodes batesi Reak. (Lepid.).

Canad. Entomol. 52: 56-59. Oliver, C. G. 1976. Photoperiodic regulation of seasonal polyphenism in Phyciodes

thaws (Nymphalidae). J. Lepid. Soc. 30: 260-263. ----------. 1978. Experimental hybridization between the nymphalid butterflies

Phyciodes tharos and P. campestris montana. Evolution 32: 594-601. ----------. 1979. Genetic differentiation and hybrid viability within and between

some Lepidoptera species. Am. Nat. (in press). Owen, D. B. 1962. Handbook of statistical tables. Addison-Wesley Publ. Co.,

Reading, Mass. Shapiro, A. M. 1971. Occurrence of a latent polyphenism in Pieris virginiensis

(Lepidoptera: Pieridae). Entomol. News 82: 13-16.

Journal of the Lepidopterists' Society 33(1), 1979, 20

TEMPORARY RANGE EXTENSION AND LARVAL FOODPLANT OF DYNAMINE DYONIS (NYMPHALIDAE) IN TEXAS

The northern normal limit of Dynamine dyonis Geyer in Texas is Cameron and Hidalgo counties. The occurrence of D. dyonis north of its usual range was noted on 27 July 1966, when I collected a tattered female on the Salado Creek, three miles southeast of the Northeast Preserve, a city park in San Antonio, Texas. The inundations of Hurricane Beulah in September 1967 produced lush vegetation in south Texas and may have caused the invasion of D. dyonis to extend as far north as Collin Co. (18 Sept. 1968, 2 S S, leg. Edward Reid). Further evidence of this movement are records by the following collectors in other counties in 1968: Gonzales Co., Hidalgo Co. (both M. A. Rickard); Bell Co., San Patricio Co. (both R. O. and C. A. Kendall); Travis Co. (C. J. Durden); Brazos Co. (J. E. Hafernik). The last known record in 1968 was Bexar Co., 23 Nov., 1 $, leg. J. F. Doyle. The total number of D. dyonis taken by collectors in Texas from 27 July 1966 through 23 Nov. 1968 was 134 (68 $ $, 66 £ £ ). To my knowledge no populations remain in central or northern Texas.

On 5 May 1968, I observed a female D. dyonis as it fluttered about a trailing plant in a dry creek bed in the Northeast Preserve. The plant, Tragia ramosa Torrey (Eu-phorbiaceae), and the butterfly were caged and placed outdoors at my home in San Antonio, Texas. Twenty-four eggs were deposited that same day. The first larva emerged on 9 May. Only 6 larvae remained on 19 May because of cannibalism. Adults which emerged were: (31 May) 3 $ $, 1 $; (1 June) 1 #, 1 $.

Larvae were collected at the Northeast Preserve site in 1968 and reared on T. ramosa. These larvae were collected on 19 May and pupated 24 May. One adult ( $ ) emerged on 5 June. Larvae also were collected on 14 July and pupated between 17 and 18 July. Adults (1 $, 1 £ ) emerged on 23 and 25 July.

Joseph F. Doyle III, 11839 Monticeto Lane, Stafford, Texas 77477.