CryoLetters 35 (3), 247-254 (2014) © CryoLetters, [email protected]

OVERWINTERING OF THE BOREAL BUTTERFLY COLIAS PALAENO IN CENTRAL EUROPE

Pavel Vrba1, Matthias Dolek2, Oldřich Nedvěd1,3*, Helena Zahradníčková3, Cristiana Cerrato4 and Martin Konvička1,3 1 Faculty of Science, University of South Bohemia, Ceske Budejovice, Czech Republic; 2 Büro Geyer und Dolek, Wörthsee, Germany; 3 Institute of Entomology, Biology Centre, Academy of Sciences of the Czech Republic, Ceske Budejovice, Czech Republic; 4 Department of Life Science and System Biology, University of Turin, Torino, Italy. *Corresponding author email: [email protected].

Abstract

BACKGROUND: Colias palaeno (Linnaeus, 1761) (Lepidoptera: Pieridae) is a butterfly with boreal distribution with declining populations in peat bogs and subalpine habitats in Central Europe. OBJECTIVE: We investigated the cold tolerance of overwintering caterpillars from one mountain population from Czech Republic (960m a.s.l.) and one alpine population from Italy (2000m a.s.l.). METHODS: We measured supercooling point (SCP), lower lethal temperature (LLT) and content of cryoprotectants. RESULTS: The caterpillars were freeze-avoiding, with lower LLT close to their very low SCP (–25 to –27°C). The mountain population accumulated high concentrations of glycerol (5% fresh mass) and sugars (trehalose 0.8%, glucose 0.2%), while the Italian alpine population only moderate amounts of glycerol (0.3%) and sugars (trehalose 0.5%, glucose 0.3%) without effect on their cold hardiness. Larvae that overwintered at +5°C had a lower body mass than those overwintering in natural conditions, indicating a metabolic weight loss, but both groups survived equally well. CONCLUSION: We hypothesize that the high concentration of glycerol contributes to the high desiccation tolerance. Keywords: cold hardiness, cryoprotective polyols, insect conservation, mountain ecology

INTRODUCTION mountains (33, 11). Supercooling capacity and protection against chill injury are enhanced by Low temperatures affect insect survival, the synthesis and accumulation of protective particularly in extremely cold environments, substances, including glycerol, glucose, sorbitol such as mountains. Insects have to complete and trehalose (37, 31). Glycerol functions as an their development there in a limited part of the antifreeze, decreasing the freezing temperature, season (32) and have to deal with temperature as a cryoprotectant binding free water (26) and extremes during overwintering (15, 25). On the against desiccation generally. other hand, predictable snow cover in such Model species distribution and activity regions insulates ground-dwelling organisms Colias palaeno (Linnaeus, 1761) against low ambient temperatures, improving (Pieridae), is a Boreal zone butterfly, inhabiting their survival (8, 27, 28, 43). Similarly, humid a vast circumpolar range in northern Eurasia, Sphagnum tussocks in boggy habitats buffer from approximately the 50° parallel northwards. against microclimatic fluctuations (41). In addition, it forms multiple relic populations Selection of an appropriate microhabitat is situated more southerly, in higher-elevated and important for survival (22, 45). From a cooler areas of the temperate zone (42). conservation point of view, recent changes of C. palaeno occurs in a wide range of habitats, temperature and (or) precipitation regimes provided that its host plant, Vaccinium should be risky for species inhabiting limited uliginosum L., is present. In temperate Central extrazonal habitats, e. g. cold peat bogs or high

247 Europe, C. palaeno occurs as a glacial relic at We used hibernating caterpillars of two main habitat types: cold peat-bogs, i.e. C. palaeno, originating from two distinct azonal oligotrophic wetlands in submountain localities in the Czech Republic and Italy. The and mountain elevations, from approximately Czech mountain population (Šumava Mts., 400m altitude onwards, and open, sometimes 49°02'N, 13°40'E, 960 m a.s.l.) inhabits a drier, heathlands at and above the timberline in mosaic of Sphagnum-bog, wet grasslands and high-elevations of the Alps (5, 6, 39). Larvae waterlogged Picea abies – Betula pubsecens are commonly reported to overwinter on dry woodlands. Here, fertilized females (n = 45) leaves attached to the stems by silk fibre or were captured in July 2010, and transported to falling with the leaf to the ground, mainly in the an outdoor rearing facility. The Italian alpine lower parts of the host plant (15, 23) or leave population (Alpi Lepontine, 46°10'N, 8°10'E, the host plant and overwinter at Sphagnum 2000 m a.s.l.) inhabits dwarf shrubland habitats covered ground (Dolek et al., unpublished). at the timberline containing the host plant Many of the southern relic populations of Vaccinium uliginosum.We directly collected 30 C. palaeno are declining at present (7, 30). This third-instar larvae from V. uliginosum leaves in particularly applies for populations inhabiting September 2011, and transported them to the Sphagnum-bogs within submontane and same rearing facility. mountain elevational zones. The species is thus The rearing facility, located in lowland extinct in Belgium as well as in northern and semi-natural conditions (at 400 m a.s.l.) some central states of Germany.Since the 1990s, consisted of wired cages (50 x 50 x 100cm) a loss of about 50% of localities occurred in this covered by nylon mesh. Each of cages contained southern Bavaria (6). Concurrently in the Czech a flower pot with a V. uliginosum plant. Republic, the species lost about 40% of its mid- From the mountain population, we released 20th century distribution. The species does not up to five females into each cage and fed them seem to be threatened in the Alps (9). Central with a 5% sucrose solution. After the larvae European populations and high altitude alpine hatched, they were let to develop without populations are different subspecies, C. p. disturbance, except for watering host plants. In europome (Esper, 1778) and C. p. europomene the first decade of November, after the larvae (Ochsenheimer, 1808), respectively. Direct stopped feeding and entered diapause, being exchanges of individuals between the prealpine rolled in dry leaves of V. uliginosum attached to bog and hill area and the Alps are unlikely. stems, we transferred pots to an air-conditioned The reasons implicated to explain the loss room with constant 5°C temperature and about of lower-elevation populations include habitat 40% r.h. The plants were regularly sprayed with degradation (drainage, successional change) (8), water to avoid desiccation. Larvae from the and climatic warming (cf. 34). No information alpine population were already showing signs of regarding the ecophysiological limits of the entering diapause when released to the cages. species’ persistence is known. Obtaining such After a few days in the rearing facility, they information may shed some light on persistence were treated identically as the mountain ones. patterns of northern species in more southerly Cold hardiness measurements latitudes, and thus on their future prospects. In January, we removed individual larvae In this paper, we investigated the lower from the host plant and measured their cold thermal survival limits of C. palaeno diapausing hardiness. The supercooling point (SCP) was larvae. We compared larvae from two contrast measured individually in sixteen individuals per southern populations, from mountain peat bogs population using a line recorder with hand-made in Czech Republic and from timberline in the thermocouples (14), which were attached to the Alps, Italy. We measured how these two body of the experimental caterpillars (10) and populations differed in their cold hardiness, cooled at rate 1 degree per minute. After the expressed as supercooling point, lower lethal exotherm appeared on the line recorder, the temperature, postdiapause survival and body larva was kept in the cooling device until its content of cryoprotectants. body temperature decreased again to the value of crystallization temperature. It was then MATERIALS AND METHODS warmed up, removed from the chamber and kept in a Petri dish at +5°C for 24 hours. Survival Rearing experimental individuals was subsequently checked at 20°C as presence

248 of spontaneous movement or reaction to twigs with young leaves. Their body mass was mechanical stimuli. measured after the transfer and then after one To measure lower lethal temperature and two weeks. Pupation and adult eclosion (LLT), groups of ten larvae were put on dry were recorded, sex of the adults was filter paper in a Petri dish, and placed in determined. Differences between the two incubator providing slow cooling to the set up treatments were analyzed by t-test with temperature measured inside the dish by resistor independent samples (Statistica 10). thermometer. Series of temperatures close to and above SCP were used. Larvae were RESULTS removed after 24 hours and treated as above to check their survival. LLT50 was calculated using Cold hardiness the logistic regression method in Statistica 10 Overwintering larvae of Colias palaeno did (38) package. We also measured lethal time (Lt) not survive SCP measurements with freezing of at selected temperature above SCP using the their body fluids, and thus exhibited the freeze- same method as for LLT, with exposure times avoiding strategy. Mean SCP (±SD) were – from one to seven days. Again, Lt50 was 24.8±3.9°C (mountain population) and – calculated using logistic regression. 26.8±2.7°C (alpine population); the difference was not significant (t-test: t = 1.72, P = 0.10). Analysis of cryoprotectants 30 The lower lethal temperature, estimated for the Eighteen individuals from the mountain mountain population only (–26°C), was close to population and seven individuals from the the SCP. Survival of overwintering larvae at – alpine population were analyzed individually. 26°C decreased with exposure duration Diapausing larvae were frozen at –80°C, then according to logistic regression: melted, homogenized in 70% ethanol, samples centrifuged, supernatant derivatized with O- e1.56− 0.167. ED S= 1.56− 0.167. ED . methylhydroxylamine (oximation) and 1+e trimethylsilylimidazol (silylation). The concentrations of low-molecular mass sugars Lt50 was 9.4 days. It follows that hibernating and polyols were determined by gas larvae of C. palaeno are highly cold-resistant, chromatography coupled to mass spectrometry able to survive extreme winter conditions. (19, 21). Concentrations were expressed as Polyols µg/mg fresh body mass. Differences between The Czech mountain population larvae two populations were analyzed using t-test with exhibited much higher total concentrations of independent samples (Statistica 10). Standard polyols and sugars, mainly glycerol and probability value of 0.05 for significance test trehalose, than the alpine Italian population. On was lowered in this series of several parallel the other hand, the alpine population larvae had tests (Table 1) using the Sidak correction (1). a higher concentration of glucose (Table 1). We also used the redundancy analysis Glycerol formed by far most of the osmotically (RDA), a constrained ordination method that active body substances in the mountain compares the entire composition of samples population, whereas trehalose, followed by with respect to external predictors; in this case, glucose and glycerol in comparable entire polyol profiles with respect to the origin concentrations, were most prominent in the of the individuals. We used the CANOCO alpine population. The other compounds were programme, statistically testing for the present in concentrations too low to contribute difference between populations using the to cold hardiness and they did not differ Monte-Carlo permutation test with 999 between the two populations. RDA analysis repetitions (24). (Figure 1) also pointed to a highly significant Postdiapause growth difference between the two populations in the Ten caterpillars from the mountain composition of cryoprotective polyols. population that overwintered outdoors and ten Postdiapause growth that overwintered at constant 5°C were The initial body mass of caterpillars after transferred in April to laboratory rearing overwintering in natural conditions was slightly conditions: 14L:10D photoperiod and 20°C. higher (10.8 ± 2.1mg) than after overwintering They were placed individually into Petri dishes at constant 5°C (8.6 ± 2.2mg; t16 = 2.18, P = (15 cm diameter) and fed with V. uliginosum 0.044). The difference after one week (5 4 ±

249 14mg vs. 37 ± 20mg; t14 = 1.86, P = 0.083) was Lepidoptera with both positive (Chilo

Table 1. Contents of cryoprotectants in overwintering caterpillars of Colias palaeno expressed in µg per mg fresh mass. Differences between the Czech mountain and Italian alpine populations significant on the adjusted probability level are expressed in bold. Czech mountain Italian alpine Protectants t P Mean SD Mean SD Glycerol 47.72 21.53 2.901 2.155 5.43 0.000016 Arabinitol 0.167 0.032 0.191 0.073 1.17 0.25 Ribitol 0.031 0.013 0.031 0.019 0.01 0.99 Fructose 0.360 0.068 0.442 0.129 2.09 0.048 Glucose 1.994 0.417 2.971 0.958 3.62 0.0014 Mannitol 0.159 0.027 0.210 0.071 2.61 0.016 Sorbitol 0.248 0.041 0.304 0.109 1.91 0.068 Myo-inositol 0 0 0.036 0.094 1.66 0.11 Saccharose 1.163 0.391 0.841 0.271 1.99 0.059 Trehalose 8.397 1.860 5.237 2.509 3.46 0.0021 Threitol 0.041 0.028 0.055 0.018 1.26 0.22 Erythritol 0.078 0.067 0.127 0.044 1.78 0.088 Ribose 0.099 0.058 0.150 0.046 2.05 0.052 Σ 60.46 21.54 13.50 3.947 5.66 0.000009 not significant due to a high variability in the suppressalis (17)) and no clear relationship 5°C treatment. After two weeks, differences in (Ostrinia nubilalis (3), Malacosoma disstria body mass between the two treatments were not (40), Sesamia nonagrioides (4)) between significant (238 ± 79 vs. 212 ± 115 mg; t12 = glycerol content and cold hardiness. 0.49, P = 0.640), although the individual The higher concentration of glucose in the differences within 5°C treatment were still alpine population can be explained as retaining apparent. Pupation was delayed in the smaller the initial intrinsic higher content of the source individuals; the post-diapause larval for synthesis of other cryoprotectants, while in development lasted 19 days in the fastest the mountain population, the source was mostly individual and 31 days in the slowest one. utilized for this purpose. Such inverse pattern However, the differences in body mass and was found in the forest tent caterpillar, duration of development might be influenced by Malacosoma disstria (Lepidoptera: the bias in sex of individuals, because there Lasiocampidae) (40). were more females (6F:1M) in the outdoor The high glycerol concentration in treatment and more males (2F:5M) in the 5°C C. palaeno may rather function as a protection treatment. However, the mass after two weeks against desiccation since glycerol is did not differ between the two sexes (females: hygroscopic. The caterpillars survived several 247 ± 90, males: 261 ± 32 mg; t10 = 0.27, P = months of exposure at constant 5°C and low 0.691). Postdiapause survival was 7 out of 8 and humidity with no apparent changes in body 7 out of 10 individuals in the two treatments. structure. Under snow, in the Alps, humidity is 100% from November until April. Snow cover DISCUSSION may be temporarily low in the Bavarian and Czech mountains, and twigs with caterpillars Cryoprotectants exposed. Low temperatures in field (Figure 2) Interestingly, the two populations differed interact with low humidity on twigs. The substantially in the content and composition of relationship between drought and glycerol cryoprotectants. Although the samples differed content during winter was studied in larvae of in their collecting history and year of sampling, the rice stem borer Chilo suppressalis: glycerol the long acclimation procedure was equal for concentration was significantly higher (48% of them. It seems that the high content of glycerol, haemolymph) and SCP lower (–10°C) at low found in the mountain but not in the alpine soil moisture (25%) than at higher moisture population, is not necessary for efficient cold (35%, –7°C) (16). hardiness. There are examples of other

250 Survival are never experienced by larvae overwintering Overwintering caterpillars of C. palaeno in the moss layer and insulated by snow. They were freeze-avoiding, with the lower lethal can be, however, experienced by larvae 6

. Glycerol

0 Saccharose

Trehalose Ribitol Threitol Ribose Glucose Mannitol Cz Myo-inositol It Erythritol Fructose Arabinitol Sorbitol 4 . 0 - -1.0 1.0 Figure 1. Profiles of polyols in the Czech mountain (Cz; n=18) and Italian alpine (It; n=7) overwintering larvae of Colias palaeno. RDA ordination diagram (1st and 2nd ordination axes). The origin of samples (Cz vs. It) explained 68% of variation in the polyol profiles; the proportional variation attributable to individual ordination axes was 0.678 (1st), 0.270 (2nd), 0.040 (3rd) and 0.006 (4th). Monte-Carlo test: F = 48.5, P<0.001. temperature close to the supercooling point. spending winters at host plant twigs. Even at the Their cold hardiness was low in comparison to twigs, temperatures below -20°C occurred for numerous insect species occurring in temperate only brief periods of time, while several days of Central Europe (e.g., SCP in ladybirds continuous exposure to such frosts would be Coccinella septempunctata: -15°C, necessary for substantial mortality. Ceratomegilla undecimnotata: –19°C (29); bugs The C. palaeno larval overwintering at Pyrrhocoris apterus: –15°C (18); Graphosoma constant temperature, which was relatively lineatum: –17°C to –18°C (36); five gerrid higher than the temperatures experienced semi-aquatic bugs: SCP: –10.5 to –20.5°C, outdoors, was probably accompanied with a LLT: –8 to –16°C (12)). The supercooling point higher metabolic rate, and resulted in higher was similar in magnitude to the mountain bark mass loss (depletion of energetic reserves). Our beetle Ips typographus: –20 to –22°C (20), and caterpillars well survived the unnatural the alpine butterfly Erebia epiphron: –22°C conditions of constant 5°C, compensating for (43), which inhabit similar habitats as the C. their mass loss after the onset of feeding and palaeno mountain and alpine populations, successfully developing into adults. The respectively.. In three other species of Colias unusually cold and desiccation tolerance butterflies, both lowland and alpine, SCP was contribute to C. palaeno’s wide distribution much higher (–18 to –14°C) (44). from northern tundra and boreal woodlands to The values found for C. palaeno document high altitudes and continental Asia. a safe survival of low temperatures by its The reasons of the C. palaeno low latitude overwintering larvae. Furthermore, the two populations’ recent decline remain unexplained, C. palaeno populations did not differ in cold the overwintering abilities of the larval stage hardiness, although the Czech mountain safely rule out overwintering larval mortality as population likely experiences occasional snow- the responsible factor. Neither relatively higher melts, while the Italian alpine population is average winter temperatures that appear as a likely insulated by snow for most of the winter. consequence of climate warming nor very low This is well illustrated using temperature temperatures nor low humidity that may occur measurements from the Italian alpine in the overwintering microhabitat due to population, and from a population in Bavaria, decreased snow cover would cause a substantial inhabiting conditions similar to the Czech larval mortality. Causes of declines should be mountain population studied here (Breitmoos sought for in other factors, such as (possibly peat bog, 47°45'N, 12°46'E, altitude 400 m) climate-driven) changes in host plant quality (Fig. 2). Evidently, low subzero temperatures (2), possible phenological mismatches between

251 Figure 2. Comparison of minimum and maximum daily temperatures likely experienced by overwintering larvae of the butterfly Colias palaeno. a) A mountain population (Breitmoos peat bog, southern Bavaria), measurements taken next to twigs of the Vaccinium uliginosum host plant (plant) and on the ground (moss layer). b) An alpine population (Alpi Lepontine, Italy), measurements taken at 1.5 m and near the ground (10 cm, where larvae start overwintering diapause). Note that two measurements cover different periods of time.

larval spring emergence and host plant leaves Acknowledgements: We are grateful to A. flushing (35), or perhaps increased infestation Bartonova, A. Pavlikova, P. Vlasanek and M. by pathogens or parasitoids (13) or micro Zapletal for help with rearing caterpillars. The climatic conditions during summer study was supported by the Czech Science development. In any case, results of this study Foundation (P505/10/1630, P505/10/2167), the highlight the efficiency of low temperature University of South Bohemia (144/2010/100) adaptations of cold-adapted insects, and point to and the Bavarian Academy of Nature a possibly high diversity of physiological Conservation and Landscape management mechanisms to achieve their impressive cold (ANL). The study is part of the ANL research hardiness. project “Development of management strategies for habitats and species of the annexes of the Habitats Directive: Analysis of the reasons for the large-scale decline of Colias palaeno”.

252 REFERENCES 19. Košťál V & Šimek P (1995) J. Insect Physiol. 41, 627–634. 1. Abdi H (2007) In Encyclopedia of 20. Košťál V, Doležal P, Rozsypal J, Measurement and Statistics. (ed) Salkind Moravcová M, Zahradníčková H & Šimek P NJ, Thousand Oaks (CA), Sage, pp. 103– (2011) J. Insect Physiol. 57, 1136–1146. 107. 21. Košťál V, Zahradníčková H, Šimek P & 2. Ammunet T, Klemola T & Saikkonen K Zelený J (2007) J. Insect Physiol. 53, 580– (2011) Ecography 34, 848–855. 586. 3. Andreadis SS, Vryzas Z, Papadopoulou- 22. Layne JR, Edgar CL & Medwith RE (1999) Mourkidou E & Savopoulou-Soultani M Am. Midl. Nat. 141, 293–304. (2008) Physiol. Entomol. 33, 365–371. 23. Lepidopterologen Arbeitsgruppe 1987: 4. Andreadis SS, Vryzas Z, Papadopoulou- Tagfalter und ihre Lebensräume. Schweiz Mourkidou E & Savopoulou-Soultani M und angrenzende Gebiete. Arten, (2011) CryoLetters 32, 297–307. Gefährdung, Schutz. Schweizerischer Bund 5. Anwander H (2010) Merkblatt Artenschutz für Naturschutz, Basel. 516 pp. 33, Bayerisches Landesamt für Umwelt, 24. Lepš J& Šmilauer P 2003: Multivariate Augsburg. analysis of ecological data using CANOCO. http://www.lfu.bayern.de/natur/artenhilfspro Cambridge University Press, Cambridge. gramme_zoologie/ merkblaetter/doc/33.pdf 282 pp. 6. Anwander H, Dolek M & Scherzinger C. 25. Li NG (2012) CryoLetters 33, 134–142. (2013) In: Bräu, M., Bolz, R., Kolbeck, H., 26. Lovelock JE (1953) Biochim. Biophys. Acta Nunner, A., Voith, J. & Wolf, W.: Tagfalter 11, 28–36. in Bayern. – Stuttgart, Verlag Eugen Ulmer. 27. Marshall KE & Sinclair BJ (2011a) J. Exp. pp. 164–167. Biol. 214, 1205–1212. 7. Beneš J, Konvička M, Dvořák J, Fric Z, 28. Marshall KE & Sinclair BJ (2011b) Havelda Z, Pavlíčko A, Vrabec V & Naturwissenschaften 99, 33–41. Weidenhoffer Z (eds) (2002) Motýli České 29. Nedvěd O (1993) Eur. J. Entomol. 90, 465– republiky: Rozšíření a ochrana [Butterflies 470. of the Czech Republic: Distribution and 30. Nilsson SG, Franzén M & Jönsson E (2008) Conservation] I, II. SOM, Praha, 857 pp. Insect Conserv. Diver. 1, 197–207. 8. Boggs CL & Inouye DW (2012) Ecol. Lett. 31. Ramlov H (2000) Hum. Reprod. 15, 26–46. 15, 502–508. 32. Roff D (1980) Oecologia 45, 202–208. 9. Bonelli S, Cerrato C, Loglisci N & Balletto E 33. Schmitt T, Čížek O & Konvička M (2005) (2011) J. Insect. Conserv. 15, 879–890. Biol. Conserv. 123, 11–18. 10. Brunnhofer V Nedvěd O & Hodková M 34. Settele J, Kudrna O, Harpke A, Kühn I, van (1991) Acta Entomol. Bohemos. 88, 349– Swaay C, Verovnik R, Warren M, Wiemers 350. M, Hanspach J, Hickler T, Kühn E, van 11. De Groot M, Rebeusek F, Grobelnik V, Halder I, Veling K, Vliegenthart A, Govedic M Salamun A & Verovnik R Wynhoff I & Schweiger O (2008) Climatic (2009) Eur. J. Entomol. 106, 77–84. Risk Atlas of European Butterflies. Pensoft, 12. Ditrich T & Košťál V (2011) Physiol. Sofia-Moscow, 712 pp. Entomol. 36, 261–270. 35. Singer MC & Parmesan C (2010) Philos. T. 13. Gripenberg S, Hamer N, Brereton T, Roy R. Soc. B 365, 3161–3176. DB & Lewis OT (2011) Ecol. Entomol. 36, 36. Šlachta M, Vambera J, Zahradníčková H & 271–281. Košťál V (2002) J. Insect Physiol. 48: 14. Hanson AA & Venette RC (2013) 1031–1039. CryoLetters 34, 261–267. 37. Somme L (1982) Comp. Biochem. Physiol. 15. Havel L (1970) Sbor. Jihoč. Muz.Č. Bud. A 73, 519–543. 10, 80–84 [In Czech]. 38. StatSoft Inc. (2011) STATISTICA (data 16. Hou M, Han Y & Lin W (2009b) Entomol. analysis software system), version 10. Sci. 12, 155–161. www.statsoft.com. 17. Hou M, Lin W & Han Y (2009a) Environ. 39. Tolman T & Lewington R (2008) Collins Entomol. 38, 1182–1188. Butterfly Guide. HarperCollins Publishers, 18. Kalushkov P & Nedvěd O (2000) Eur. J. London. 384 pp. Entomol. 97, 149–153.

253 40. Trudeau M, Mauffette Y, Rochefort S, Han 263 pp. (December 2000), Vol. 2 (plates), E & Bauce E (2010) Environ. Entomol. 39, pp. 264–308, 183 pls. (January 2001). 498–504. 43. Vrba P, Nedvěd O & Konvicka M (2013) J. 41. Turlure C, Choutt J, Baguette M & vanDyck Entomol. Sci., in press H (2010) Glob. Change Biol. 16, 1883– 44. Vrba P, Konvička M & Nedvěd O (2012) 1893. CryoLetters 33, 251–258. 42. Verhulst J (2000–2001) Les Colias du 45. Wagner D, Doak P, Sformo T, Steiner PM Globe. Monograph of the Genus Colias. & Carlson B (2012) Environ. Entomol. 41, Goecke & Evers, Germany, Vol. 1 (text), 180–187

254