Photoperiodic Responses of Phenologically Aberrant Populations of Pierid Butterflies (Lepidoptera)

Great Basin Naturalist

Volume 35 Number 3 Article 7

9-30-1975

Photoperiodic responses of phenologically aberrant populations of pierid butterflies (Lepidoptera)

Arthur M. Shapiro University of California, Davis, California

Follow this and additional works at: https://scholarsarchive.byu.edu/gbn

Recommended Citation Shapiro, Arthur M. (1975) "Photoperiodic responses of phenologically aberrant populations of pierid butterflies (Lepidoptera)," Great Basin Naturalist: Vol. 35 : No. 3 , Article 7. Available at: https://scholarsarchive.byu.edu/gbn/vol35/iss3/7

This Article is brought to you for free and open access by the Western North American Naturalist Publications at BYU ScholarsArchive. It has been accepted for inclusion in Great Basin Naturalist by an authorized editor of BYU ScholarsArchive. For more information, please contact [email protected], [email protected]. PHOTOPERIODIC RESPONSES OF PHENOLOGICALLY ABERRANT POPULATIONS OF PIERID BUTTERFLIES (LEPIDOPTERA)

Arthur M. Shapiro'

Abstr.'^CT.— T^vo local pierid populations in western Norlh Ainerit.i showing regi(jnalh' aix'rrant l)henologies were investigated in the lahoratory. Neither a parlialK hivoltine Piciis iiapi from the

Sierra Nevada foothills in El Dorado County. California i sui rnuiidi'd hy luiivoltine populations), nor a vcrnal-univoltine P. occidenlalis from a foothill outlier of llie (Colorado Front Range (below hivoltine populations) showed unusual resjwnses to controlled developmental regimes in the laboratory. Their unusual phenologies are hypothesized to be the produi t of microclimate. Failure to under- go genetic adaptation to unusual microclimates is discussed witii ])articulai- reference to the i)res- ence or absence of gene flow from neaib\' normal populations.

The timing of life-history phenomena naeus complex) or not (P. protodice Bois- in an insect population is (leterminetl by duval & LeConte, P. occidciitdUs Reakirt). physiological responses to en\ironniental Recent st tidies have shown that univol- stimnli. These proximate controls reflect tinism in both grouj)s is derivative from a genetic basis Ijelievecl to be the jiroduct multivoltinism, accompanying invasion of natural selection for seasonal cycles of a short-summer climate (P. occidcn- appropriate to the environment of the ta/is, Sha})iro, 1975a) or persistence in a population. In the western LTnited States progressively drier one (P. napi, Shapiro, topography has a dramatic impact on cli- 1975b). Such patterns are defined over mate, and great differences may occur broad geographic areas. California over short ground distances. How closely P. napi, for example, is differentiated into can insect populations ada])t to their im- a commonly biv'oltine, heavily pigmented mediate climates on a microgeograj)hic subspecies in the coastal summer-fog belt scale? Phenological adaptation is merely and a univoltine, more lightly marked one case of the more general jiroblem of subspecies in the interior, where summers population differentiation (cf. Ehrlich are clear and hot. The transition between and Raven, 1969; Ehrlich et al., 1975). the subspecies apjjears to be in the form In most organisms, at least prior to the of a stee]) cline through the central advent of electrophoretic genetics, popu- Coast Ranges (Shapiro, in preparation). lation differentiation was assessed on the Recently Lees and Archer (1974) have basis of visible phenotypic characters. reported the existence of phenological dif- Such characters, like the enzyme systems ferences among napi populations on a studied by electrophoresis, are often not mvich finer scale. They have found ap- translatable into specific selection pres- parently relict univoltine populations in sures. In markedly seasonal climates the suitable (bog-heath) habitats completely nature of selective pressures acting on surrounded by multivoltine ones in the phenology may be very apparent. Where British Isles. Their preliminary interpre- local deviations from the broad geograph- tation of this situation is that it provides ic pattern of voltinism are observed in a (n ideiice for midtiple invasions of Britain species, the potential exists for the demon- h\ napi stocks having different pheno- (or ecotypic) stration of microgeograj)hic logical characteristics and source regions. differentiation. This is the fourth paper In th(> course of recent work on pierid in a series exploring the evolution of sea- |)hen()logy and e\olution. the existence sonality in the butterfly genus Pieris in of regionally aberrant populations has western North America. b(>en i)r()ught to my attention in both the In various multi\oltine Pieridae both and protodicc-occidcntalis groups. In phenotype and diapause are under photo- napi

( ases the populations appear to be periodic control. The two sets of develop- both repeating mental options (diapause /direct develop- uniciiie, rather than forming a pattern as in British P. napi. They would ment; vernal /esti\ a 1 j)henotype) may be physiologically coupled {Pieris napi Lin- therefore seem to be good candidates for

'Departmcnt of Zoology. Univcisily n[ Cnlifoniin. D.ivi';. Ca

310 .

Sept. 1975 311

local genetic tliflcreiitiatioii under al\[j lurliuni offn iiialc W. Br. Rorippa na.s-

ical microclimates turliu!ih(i(jii(ih(nni Schii :. & Thell.) gr'ows in [xTincUient stn ims (Fig. 1). Most .,r ih,> uiid ,1,1, hutterflies are ricns iiapi ii he Sierra Foothills idonlicil to lal)orat( bred Sierran

Pieris napi from interior (-alitornia are. "cdstnrid' I h'ig. 1 i as noted above, luiivoltine and moiio- On J<) Man h l<)7) niii(> in.de and three phenic in nature. Under laboratory (au- lemale lirst-hrood, vernal phenotyjie napi ditions they can lie reared without dia- were colIectcMl in one of these ravines. pause; then they produce the eslixai These included two coj)ulating j)airs in phenoty|)(» ''castoria'' Airtually unknown which the females were soft-winged, in- in the wild in the interior (Sha])iro. dicating that they had developed in the 1975b). In June 1974 Mr. William Pat- ravine^ itself. The eggs from these females terson of Sacramento, California, took were used in photoperiod experiments

se\-eral 'Vy/.s/o/vV/"" of ( ( wild hoth sexes in Table 1 ) . Rearing methods are de- the canyon of the American Ri\ er below scribed m Shaj)iro. 'l975a and 1975b.) Auburn in the Sierra Nevada foothills Hie results are entirely typical for Sier- (El Dorado County, 650 feet). The oc- ran stock and do not suggest that x-Xmer- currence of a second brood there was con i(an Ri^er material has a greater pro- firmed in 1975. P. napi is connnon in the pel isit\' to flevelop directly than do stocks canyon, producing its usual \ernal pheiid tVoju purely uniAoltiiie localities, at least type in March, llie second brood, whi( li under our laboratory regimes. However, is much scarcer, unlike the first is ex- ihis is not particularly surprising. The tremely localized ^vithin llu^ camon -al second brooci of napi in the American present being known from onh' two Ri^er gorge is nukh rarer than the first, densely shaded ravines where the intro- indicating that is is onl}' partial; its num- duced cruciferous weed watercress (Nas- bers also fluctuate from year to year.

(^5SS£S^IJ4;>rSjgx%,c-;.v|f^Vr^

Fig. 1. Locations of ravines ("B") where !)ivoItiiie Pieris tuipi Ill the American River gorge; univoltine napi are generally distriluited at low (Icnsity. l'S(;S 7. life "Aulnun"' quadrangle. 312 GREAT BASIN NATURALIST Vol. 35, No. 3

Table 1. Incidence of diapause ( D) and non- the host plant at the optimum time failed pupae in bivoltine (American diapause (ND) to turn up ;ni\ napi immatures, although River, 650') and univoltine (Placerville, 1800') ten Pieris rapac lar^'ae Pieris nopi from El Dorado Co., Calironia, reared were found.) on watercress at 27 C under two photoperiods. These circumstances suggest that the production of a second brood here is ac- Photophase: Continuous 15 hr cidental, resulting from the peculiarly Stock: Pupae: D ND D ND cool and moist conditions within the ra- American River 15 29 16 \'ines. There is no evidence that the bi- Placerville 10 23 18 Aoltine sites are in any sense isolated from adjacent univoltine ones, nor is it clear that there is successful re])roduction (In 197:5 three trips by Patterson and by the second brood in all years nor even Shapiro in season turned up only tNvo that there is genetic continuity from year males and one female. A later search of to year in the ravines; perhaps a few pu-

^P'-

Fig. 2. Wild second-brood Pieris napi from the American River gorge, collected by W. Patterson in June 1974 fmales at top; dorsal (left) and ventral (right) surfaces). The heavily marked female is atypical for an inland population.

Fig. 3. Phenotypes of representative lab-reared nondiapause Pirris napi from the Ameiican Riv( stock; 27 C. continuous light; dorsal (left) and ventral (right). Sept. 1975 SHAPIRO: BUTTERFLIES 313 pae will develop directly there whenever any female napi happens to colonize them. Experienced California collectors (R. L. Langston, B. Walsh) agree that even near the coast some localities produce second-brood napi every year and others only rarely or sporadically. Exj)eriments have shown both developmental and phenotypic differences between coastal and inland stocks but not among the coastal stocks themselves.

Watercress is known to be host of P. napi in various Sierran sites up to about •5,000 feet (Shapiro, 1975c). The only other record of a Sierran ^'castoria" known to me is a fresh male taken flying

Table 2. Incidence of diapause (D) and nondiapause (ND) pupae in two split broods of a Barbarea verna-ieeAin^ univoltine Pieris napi (Gates Canyon, Inner Coast Ranges, Solano Co., 750') reared at 27 C on continuous light. None of the differences was significant.

Brood Host Pupae: ND

Brassica kabet-^ 7 Nasturtium officinale^' 5

Brassica kaber''^ 6

Lepidiuni latifolium^'-'' 5 314 GREAT BASIN NATURALIST Vol. 35, No. 3

Fig. 5. Wild Pieris occidentalism vernal phenotype C'calycc''). from Haystack Mountain, collected by R. E. Stanford. Males at left; dors;d (left) and ventral (right) surfaces.

response to competition from P. protodice by testing the developmental responses (although no such phenomenon is known of multivoltine P. occidentalis stock from at other localities where the two are sym- elsewhere reared at Haystack Mountain. patric). Haystack Mountain is probably We hope to carry out such experiments the lowest elevational record for P. oc- within the next couple of seasons. cidentalis in Colorado. In the Rockies Phenological differences are known be- proper it is bi\oltine at middle elevations tween plains and lower montane popu- (perhaps locally trivoltine) and univol- lations of a inunber of Colorado butter- tine in the Alpine zone (cf. Brown, Eff, flies (.I.A. Scott, R.E. Stanford, pers. and Rotger, 1957) and has two seasonal comm.), but they may go in a direction phenotypes (Shapiro, 1975d). o])posite to those observed in Pieris oc- A laboratory stock was established cidentalis. Two species (Colias olexandra from ova laid by five females collected by Edwards, Pieridae; Plebeius icarioides Stanford on 6 April 1975. Under labora- complex, Lycaenidae, both Legume feed- tory conditions their developmental and ers) are bivoltine on the plains and uni- phenotypic responses (Table 3 and Fig. voltine in the mountains. The basis for 6) were identical to both Sierran multi- these differences is uninvestigated. voltine and Colorado Alpine stocks (Sha- With no evidence for genetic differen- piro 1974, 1975d). Once again we have tiation of Haystack Mountain occidentalis. no experimental evidence for the evo- the attractive hypothesis of competitive lution of a phenological ecotype and are therefore forced to look for micro- Tahi.e 1 Incidence of diapause iD) and non- climatic explanations. Since Haystack diapause (ND) pupae in veiiial-univoltine (Hay stack Mountain, Colordao, o.oSO') and bivoltine is effectively in Mountain the Great (Donnor Pass, Colorado, 7,000') Pieris occident- Plains climatic regime, which is hotter alis reared on Brassica kaher at 27 C. and drier than the usual regime of P. occidentalis, it may not be surprising Photophase: Continuous 15 hr that conditions there would be associated Stock: Pupae: D ND D ND wdth summer dormancy. This (juestion Haystack Mountain. 1975 22 settled can be only b}' laboratory duplica- Donner Pass. 1973 16 tion of Haystack Mountain conditions or Sept. 1975 SHAPIRO: BUTTERFLIES 315

^ ^- '. -^_f -1

Fig. 6. Phenotypes of representatives lal)-reared nondiapause Picris occidcntalis from the Haystack Mountain stock; 27 C. continuous light. Estival phenotvi)es characteristic of nnitlivoltine. Males at left; dorsal (left) and venti-al (right) surfaces.

seasonal displacement with protodicc simple manipulation of constant rearing must be set aside. The host plants of both temperatures and unchanging day- species on Haystack Mountain are un- lengths, have gi^en an oversimplified identified. Both prefer species of pepper- picture of the developmental versatility grass, Lepidium, throughout their ranges. of pierids in the field. If microclimate On the plains most crucifers are vernal determines aberrant voltinism in these species, as in lowland California. stocks, it is very likely that humidity, for exam])le, may interact with photoperiod temperature in controlling de- DiSCEISSION and \'elopment in natural populations. The Many instances are on record of eco- same genetic information may allow typic differentiation on a microgeogjraphic Picris occidcntalis to respond ap])ropriate- scale, particularly in plants, which have ly to regimes as diverse as those at Hay- more versatihty in developing isolating stack Mountain (5,589 feet) and Love- mechanisms than do animals (Jain and land Pass (12,400 feet). Bradshaw, 1966). Given such jilasticity, we may wonder As noted above, the lack of a genetic whether the Baldwin effect (Simpson, basis for biovoltinism in Sierran Picris 1953) might not come into play in popu- napi is not very surprising, granted the lations in extreme environments. Briefly, extremely restricted habitat and the ex- the Baldwin effect postulates the buildup tensive distribution of univoltine butter- by selection of a genetically obligate basis flies, with ample opportunity for gene for the ada})tations produced via develo])- flow. The failure of the Haystack Moun- menlai plasticity. In an atypical hut pre- tain P. occidcntalis to differentiate is dictable climate like Haystack Mountain, more intriguing. It is, of course, possible might not the developmental flexibility that it has differentiated and that the lab chara( t(>ristic of montane poj)ulations be rearing regimes were too crude or inap- lost? (^Alaskan Picris occidcntalis nclsoni

propriately selected to show it. It is cer- seem to be evohing in this direction; tain that experiments to date, involving Shai)ir(>, 1975a.) One important counter- 316 GREAT BASIN NATURALIST Vol. 35, No 3. vailing force would be gene flow, which LiTER.-^TURE Cited is almost certainly operating on high- Brown. F. M., D. Eff, .\nd B. Rotger. 1957. elevation univoltine occidcnUdis in Colo- Colorado butterflies. Denver Museum of rado (Shapiro, 1975d). Haystack Moun- Natural History. Denver. 368 pp. Ehrlich. p. R.. and p. tain is about 30 air miles from timberline H. Rwen. 1969. Dif- ferentiation of populations. Science 165: and much closer than that to the montane 1228-1233. zone, but how isolated it actually is is EiiRLicH. P. R.. R. R. White. M. C. Singer. quite unknown. Nor is there any infor- S. W. McKechnie. and L. E. Gilbert. mation bearing on how long occidentalis 1975. Checkerspot butterflies: a historical perspective. Science 188:221-228. has been there—whether it is a Pleisto- Jain. S. K.. and A. D. Bradshaw. 1966. Evo- cene relict or a recent colonization. There lutionary divergence among adjacent pop- are much more isolated, certainly relict ulations. I. The evidence and its theoretical occidentalis populations in other localities analysis. Heredity 21:407-441. Lees, E.. and D. M. Archer. 1974. Ecology east of the Front Range the Black Hills — of Pieris napi (L.) (Lepidoptera. Pieridae) of South Dakota and perhaps the Pine m Britain. Ent. Gazette 25:231-237. Ridge of northwestern Nebraska—which Shapiro. A. M. 1974. Photoperiodic control of deserve study in this regard. seasonal polyphenism in Pieris occidentalis Reakirt (Lepidoptera: Pieridae). Wasmann J. Biol. 31:291-299.

Acknowledgments . 1975a. Photoperiodic control of de- velopment and phenotype in a subarctic Of the collectors who provided vital in- population of Pieris occidenlalis (Lepidoptera: Pieridae). Canad. formation and who have been credited Ent. . 1975b. Developmental and phenotypic in the text, special thanks are due Mr. responses to photoperiod in uni- and bivoltine William Patterson and Dr. Ray Stanford. Pieris napi (Lepidoptera: Pieridae) in Cali- without whose help these experiments fornia. Trans. Rov. Ent. Soc. London 127: would have been impossible. Mr. Mark 65-71. . 1975c. The role of watercress, Nas- field Kauzer assisted in w^ork and Mrs. lurtiuni officinale, as a host of native and Adrienne R. Shapiro in rearing. This re- introduced pierid butterflies in California. search is part of a larger study of colo- J. Res. Lepid.

. 1975d. variation in montane nizing ability and the evolution of season- Ecotvpic butterflies. Wasmann J. Biol. 32:267-280. ality in Pieris funded by the Committee Simpson. G. G. 1953. The Baldwin Effect. on Research, UCD, under grant D-804. Evolution 7:110-117.