Potential Hosts Number in Cuckoo Bees (Psithyrus Subgenus) In

Home , Psithyrus

research letter ISSN 1948‐6596 Potential hosts number in cuckoo bees (Psithyrus subgenus) in‐ creases towards higher elevations Jean‐Nicolas Pradervand1,*, Loïc Pellissier2, Glenn Litsios1,3, Antoine Guisan1 1Department of Ecology and Evolution, University of Lausanne, Bâtiment Biophore, 1015 Lausanne, Swit‐ zerland. 2 The Arctic Research Centre, Department of Bioscience, Aarhus University, Aarhus, Denmark. 3Swiss Institute of Bioinformatics, Quartier Sorge, 1015 Lausanne, Switzerland * jean‐[email protected]

Abstract. In severe and variable conditions, specialized resource selection strategies should be less fre‐ quent because extinction risks increase for species that depend on a single and unstable resource. Psithyrus (Bombus subgenus Psithyrus) are bumblebee parasites that usurp Bombus nests and display inter‐specific variation in the number of hosts they parasitize. Using a phylogenetic comparative frame‐ work, we show that Psithyrus species at higher elevations display a higher number of hosts species com‐ pared with species restricted to lower elevations. Species inhabiting high elevations also cover a larger temperature range, suggesting that species able to occur in colder conditions may benefit from recruit‐ ment from populations occurring in warmer conditions. Our results provide evidence for an ‘altitudinal niche breadth hypothesis’ in parasitic species, showing a decrease in the parasites’ specialization along the elevational gradient, and also suggesting that Rapoport’s rule might apply to Psithyrus. Keywords. altitude, altitudinal gradient, bee, generalist, geographic range, gradient, niche breadth, parasitism, Rapoport, specialist

Introduction reproductive females after taking over the nest Optimal resource selection is crucial for all organ‐ (Wcislo 1987). As host availability and population isms and directly impacts fitness (Parker and Stu‐ stability vary along environmental gradients art 1976). Among species, resource selection (Brinck 1974), parasites should be more host spe‐ ranges from highly specialized (i.e., small range of cialized in more productive and stable environ‐ resources consumed) to highly generalized (i.e., ments, compared with severe conditions. Latitude large range of resources consumed). Furthermore, has been traditionally associated with such an en‐ different resource selection strategies may not vironmental gradient, inspiring the ‘latitude niche generate identical costs and benefits in mild com‐ breadth hypothesis’ (MacArthur 1972). It may ex‐ pared to severe environments. Indeed, resource plain, for example, the increased host range ex‐ specialization is expected to be more advanta‐ ploited by cuckoo birds at higher latitude (Yom‐ geous in stable environments, where fluctuations Tov and Geffen 2005). By extending this rule to of resources’ availability are low, and intense the geographical range of species, Rapoport sug‐ competition favours niche partitioning (MacArthur gested that species occupying, on average, higher 1955). In contrast, more generalist species should latitudes should also display a larger latitudinal be advantaged in severe environments where re‐ range (Rapoport 1975). Yet the use of latitude sources are fluctuating, since they do not depend may not represent a stability gradient and is con‐ on a restricted range of resources that may be‐ founded by historical and biogeographical factors come temporarily unavailable (Currie 1991). (Vázquez and Stevens 2004). For instance, it re‐ Generally, parasites obtain resources from a mains unclear whether the higher frequency of host in the form of nutrients (e.g., body fluids), social parasites at higher latitude arises from eco‐ and also habitats (e.g., nest). The latter is particu‐ logical or historical constraints (Wcislo 1987). To larly widespread in social insects, because nest keep away from such biases, one would need to parasites benefit from the workforce of non‐ study a geographically restricted zone which en‐

122 frontiers of biogeography 5.2, 2013 — © 2013 the authors; journal compilation © 2013 The International Biogeography Society Jean‐Nicolas Pradervand et al. compasses a wide range of environmental condi‐ subgenus Psithyrus. We sampled 149 bumblebee tions. communities along an elevational gradient in the Elevation clines developing from lowlands western Swiss Alps. We coupled environmental to the summits of mountain ranges show similar niche information with data from the literature on environmental gradients to that of latitude from host species and tested the following hypotheses: the tropics to the poles. High‐elevation environ‐ 1) Host species number decreases with tempera‐ ments form a series of marginal habitats often ture and increases with elevation. characterized by stronger fluctuations in physical 2) Species occupying colder conditions also span factors, more pronounced temporal (daily and a larger temperature range, in accordance with seasonal) variability, and also more stressful and Rapoport’s rule. stochastic climatic conditions, than in lowland en‐ vironments (Brinck 1974). Such conditions proba‐ Materials and methods bly affect species’ population dynamics. Indeed, at The study area is located in the western Swiss Alps high elevations, insect populations undergo heavy and covers 700 km², ranging from 1000 m to 3210 fluctuations and community structures are less m a.s.l. We selected the sampling sites using bal‐ stable (Gutiérrez and Menéndez 1998). This in‐ anced stratified random sampling, restricted to creased stochasticity may have important conse‐ open meadows and based on elevation, slope and quences for the life history of high‐elevation in‐ aspect, thus allowing for accurate estimates of sect parasites, but this is not yet known. species’ niche limits along the elevational gradi‐ Psithyrus (cuckoo‐bees) is a monophyletic ent. From June to September 2009 and 2010, 149 group within the Bombus genus, whose species sites of 50 m x 50 m were sampled (Figure 1). We parasitize nests of other Bombus species visited each site every three weeks during the (Cameron et al. 2007). The Psithyrus individual most favourable hours for bumblebee activity enters the host colony and either adopts the (10:00–17:00) and in good weather conditions odour of the invaded nest or tries to take over the (i.e., little wind, sunny, high temperatures; num‐ queen directly after the nest invasion (Goulson ber of field days = 146). At each site, we used a 2010). Then it lays eggs and uses the non‐ net to collect every Bombus and Psithyrus individ‐ reproductive female workers of the invaded col‐ ual we could during a 45‐minute search. Identifi‐ ony to raise its progeny (Goulson, 2010). Among cation of specimens took place in the laboratory. Psithyrus species the number of hosts can vary The number of hosts for each Psithyrus spe‐ (Rasmont 1988), but it is not yet known how this cies in the study area was obtained from the lit‐ variation is distributed along environmental gradi‐ erature (Amiet 1996, Williams 2008) and ranged ents and whether the number of hosts per species from one to five host species. These records con‐ depends on the environmental conditions. An‐ stitute an accurate list of all host Bombus species tonovics and Edwards (2011) showed that parasite that are known to be parasitized by each Psithyrus ‐free zones exist at the limit of the range of the species and thus constitute the ‘potential host species. The advantage provided by being general‐ range’. The ‘realized host range’, i.e., the field re‐ ist (high number of potential hosts) or specialist cord of all the hosts used by Psithyrus species at (low number of potential hosts) may depend on each location, would be more informative. How‐ the prevailing environmental conditions, varying ever, Bombus nests are cryptic and very difficult to along the altitudinal gradient. Following this ra‐ find in the field, making such data almost impossi‐ tionale, we would thus expect to find higher pro‐ ble to record at the scale of our study zone. portions of generalist Psithyrus at high elevations, As a measure of environmental severity, whereas at low elevations specialists should pre‐ degree‐days were calculated and interpolated vail. using data from meteorological stations, with a Here we studied the link between host digital elevation model at 100 m resolution (for specificity and environmental conditions using the more detail, see Zimmermann and Kienast 1999). frontiers of biogeography 5.2, 2013 — © 2013 the authors; journal compilation © 2013 The International Biogeography Society 123 Psithyrus along an elevational gradient

al. 2007). We related the species‐specific host number, the 20th percentile of degree‐days and the variance in degree‐days using phylogenetic Generalized Least Squares (pGLS), as implemented in the caper R package (Orme et al., 2011). The strength of phylogenetic dependency in our data was accounted for by estimating, using Maximum Likelihood, the λ branch transformation parame‐ ter. Other transformations assuming different evolutionary scenarios exist, but they require lar‐ ger phylogenetic trees to be correctly estimated (R. Freckleton, personal communication). Finally, to show a potential change in the number of Bom‐ bus species that might affect the availability of hosts to Psithyrus, we related the number of Bom‐ bus species to degree‐days using a Generalized Linear Model (GLM) with a quasi‐Poisson error distribution.

Results In the 149 inventoried sites we found eight of the nine Psithyrus species (total sampled individuals = 183) occurring in Switzerland and 20 non‐parasitic Figure 1. Map of the western Swiss Alps and their loca‐ Bombus species out of the 31 occurring in Switzer‐ tion in Switzerland. The coloured area shows the degree‐ land (total sampled individuals = 1609) (see Table days for all the areas above 1000 m elevation (the lower 1 and Figure 1, Pradervand et al. (2011) and Amiet sampling limit). The black dots represent the sampling sites. (1996) for more information). We have lumped together three closely related Bombus species (B. Degree‐days are the sum of degrees above 3°C for terrestis, B. lucorum and B. cryptarum), at the sub‐ all days in the year. Decreasing temperature along genus level (Bombus sensu stricto), because of the the elevational gradient is also associated with high probability of wrong identification for this increased precipitation and climatic variability. In group, especially with old worker bees. The bum‐ the study area, degree‐days are closely related to blebee species richness is higher at intermediate elevation (R² = 0.98, p < 0.001). The temperature values of degree‐days, corresponding to mid‐ variance increases with elevation (MeteoSwiss elevations, and reaches its lowest value in the data for 187 recording stations in Switzerland, coldest conditions at high elevation (linear: t = based on the variance of mean daily temperature, 4.14, DF = 148, p < 0.001, quadratic: t = ‐3.91, p <

R² = 0.23, p < 0.001). As a measure of colder niche 0.001; Figure 2; see Pellissier et al. 2012b for more edge, we extracted, for each species, the degree‐ information). The Psithyrus species richness shows days values for the sites where it occurred and a slight linear increase in species richness (linear: t th calculated the 20 percentile of those values. We = 2.56, DF = 74 p = 0.013 quadratic: t= ‐3.5*10‐7, p th used the 20 percentile instead of the minimum > 0.05). degree‐days value to limit the noise of outlier oc‐ Overall, Psithyrus individuals are relatively currences. We also measured the variance of each rare, and, while they were found in 49% of our species’ degree‐days values to represent its tem‐ sites, the proportion of Psithyrus among bumble‐ perature range. We accounted for species related‐ bee individuals collected was very low (on average ness by using a published phylogeny (Cameron et 10% of all observations, consistent with usual pro‐

124 frontiers of biogeography 5.2, 2013 — © 2013 the authors; journal compilation © 2013 The International Biogeography Society Jean‐Nicolas Pradervand et al. portions of healthy mountain populations; Goulson et al. 2008, Roulston and Goodell 2011). The pGLS regression showed a signifi‐ cant relationship between the species’ colder range limit and the number of hosts: Psithyrus species tended to have more hosts in cold than in mild conditions (F = 11.54; DF = 2,6; p = 0.009; adjusted R² = 0.60; λ = 6*10‐5; Figure 3a). Because of the low value of λ we tested whether an Ornstein–Uhlenbeck model of evolution would better fit the data using the Geiger R package, but the Akaike Information Criterion difference was marginal (0.23). Fi‐ nally, we also found a significant negative cor‐ relation between the species colder niche edge and the variance in temperature occu‐ pied by the species computed from occur‐ Figure 2. Maximum number of non‐parasitic (true bumblebees, in rence data (F = 7.75; DF = 2,6; p = 0.022; ad‐ grey) and parasitic (Psithyrus subgenus, in black) species in rela‐ justed R² = 0.49; Figure 3b). Psithyrus species tion to degree‐days. The dotted line represents the generalized occurring in colder conditions spanned a lar‐ linear model of the number of non‐parasitic species in relation to ger range of temperature. degree‐days. The numbers above the bars indicate the number of sampled communities.

Figure 3. (a) Number of hosts (‘Hosts number’) for the eight Psithyrus species plotted against the colder niche edge (measured as the 20th percentile of their distribution along a gradient of degree‐days). The regression line indicates the negative relationship (R² = 0.60, p = 0.009). (b) Variance of degree‐days for each species plotted against the colder niche edge. Species with a smaller temperature range tend to occur in warmer conditions (R² = 0.49, p = 0.022). The outlier corresponds to P. campestris, which occurs over the whole gradient but is found most in colder conditions. The sizes of the points in both graphs correspond to the number of sites occupied (see Table 1). frontiers of biogeography 5.2, 2013 — © 2013 the authors; journal compilation © 2013 The International Biogeography Society 125 Psithyrus along an elevational gradient

‐

‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ L L Bom

column sylvestris

‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ L L R R indicated.

rupestris occupied’

are

‘Plots

‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ L/R The

quadricolor

percentile

th 20 species

‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ L recorded. and

norvegicus Psithyrus

‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ L L eight

minimum flavidus

the individuals

of of

‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ L R R R R Hosts clades.

variance,

campestris two

number

their the

the

‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ L and

text).

between bohemicus show

main

(DDEG)

‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ L R R (see

records 1000m.

days ‐ soroeensis

barbutellus The B.

below

degree subgenus

area.

2012b),

the th al.,

‐ ‐ ‐ 20 s.str.’ et 762.8 1656.7 1675.5 1363.3 1341.7 1045.7 1342.9 1407.1 1592.3 973.34 1834.6 1283.6 study distributions 1438.98 1778.54 1100.36 1238.42 1910.62 1256.88 1050.72 1267.24 1261.76 1515.68 1296.92 1191.36 1428.94 1738.62 1138.36 1551.54 1345.42

the species

with

but (Pellissier ‐ ‐ ‐ Bombus

‘

to 744.8 861.3 799.3 744.8 799.3 799.3 969.8 799.3 799.3 799.3 868.4 908.3 799.3

each 1656.7 1047.3 1341.7 1363.3 1363.3 1038.5 1034.4 1117.4 1250.8 1558.1 1513.2 1038.5 1068.8 1601.7 1350.9 1116.8

minDDEG area

the

For

in variables

study recorded

according ‐ ‐ ‐ ‐ ‐ the

40500 in 1991016 1288370 1693758 2208483 1077486 1768128 1141664 2384115 1635555 2300616 1077605 1787626 2485308 2513346 2220459 6645632 1899382 1029020 1819100 2792438 2045365 Climatic varDDEG 977965.6 944057.6 655907.2 840954.1 457142.7

species. clade

species the

days] amalgamated ‐ recorded by variance

‐ ‐ ‐ [°C

789.8 are 1656.7 1363.3 1465.9 1721.8 1770.9 2041.2 2027.5

1871.04 1697.26 1910.94 1859.307 2055.282 2060.982 1803.036 1334.042 1571.726 1508.973 2115.075 1772.165 1275.649 1802.367 1501.695 1693.253 1622.431 1654.075 1424.471 1758.236 1894.368 Psithyrus species

DDEG elevation

two

occupied and

are

cryptarum

plots

‐ 2 1 1 0 3 0 5 4 7 6 5 Restricted 28 39 11 76 24 23 47 22 52 57 37 36 39 68 93 12 40 22 28

B. occupied 113

Bombus R=

statistics

pomorum and Plots the

clade,

sampling

Bombus

Presence

2 2 2 0 3 0 5 4 9 9 6 35 71 13 44 27 72 27 84 66 50 67 12 67 38 38 234 101 279 135 290 1792 lucorum and

Records

variance

number

Prevalence

sylvarum

elevation

s.str. the

terrestris, Large

jonellus lapidarius ruderarius sicheli soroeensis subterraneus sylvarum* barbutellus bohemicus campestris flavidus norvegicus quadricolor rupestris sylvestris alpinus gerstaeckeri hortorum humilis hypnorum mendax mesomelas monticola mucidus pascuorum pomorum* pratorum pyrenaeus veteranus wurfleini

Bombus =

B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. Bombus P. P. P. P. P. P. P. P. Total L * Table bus shows

126 frontiers of biogeography 5.2, 2013 — © 2013 the authors; journal compilation © 2013 The International Biogeography Society Jean‐Nicolas Pradervand et al. Discussion needed. In contrast, specialist species that mimic Host specialization is known to vary between the odours of a particular host species probably Psithyrus species (Rasmont 1988), but it was un‐ show a double limitation to their distribution. known whether, as for latitudinal gradients, a spe‐ First, they are associated with host species that cialist‐to‐generalist progression could be found are mostly absent in cold environments along an elevation cline. In this study we com‐ (Pradervand et al. 2011), and second, environ‐ bined environmental data on Psithyrus species mental constraints of unstable environments may with their evolutionary history and found that, prevent the persistence of populations. indeed, species occurring at higher elevation are Generalist Psithyrus species occurring in the more generalist than their low‐elevation counter‐ coldest environmental conditions also occupy parts. The trend was observed despite the de‐ warmer sites. Among their potential hosts they crease of bumblebee species richness at high ele‐ thus have a range of different species occurring at vation (Figure 2). Bumblebee species richness fol‐ different elevations. This result parallels the much lows a hump‐shaped pattern typical of a mid‐ ‐discussed Rapoport’s rule, which suggests that altitudinal diversity peak (Pellissier et al. 2012b). latitudinal ranges of plants and animals are gener‐ Also, we provide evidence of increasing niche ally smaller at lower than higher latitudes breadth with altitude, which parallels the results (Rapoport 1975, Stevens 1992). We found that found for latitude (Krasnov et al. 2008). The fact generalist species occupy a wider temperature that generalist Psithyrus occurring in the coldest range than more specialized species and are environment also occupy a larger temperature therefore likely to encounter more potential host range suggests that Rapoport’s rule might also be species to parasitize. Paralleling our results, Sand‐ valid for altitude (Stevens 1992). ers (2002) found that the elevational range sizes Generalist species may be more prevalent of ants from Colorado, Nevada and Utah widen in colder environments than specialists because with increasing elevation. Such patterns following they can compensate for the variation in host Rapoport’s rule may be explained by a rescue ef‐ populations by parasitizing more species and thus fect: unstable populations occurring at high eleva‐ increase the chance of finding a host. Cold and tions might benefit from constant recruitment unstable conditions may disadvantage species from stable populations occurring at lower eleva‐ that are bound to a particular host species. In‐ tions (Stevens 1992). deed, it is known that host populations fluctuate In this study, we sampled and analysed the more in colder environments (Brinck 1974), mak‐ distribution of adult specimens under the assump‐ ing the resources less reliable for specialist para‐ tion that they were caught close to the nest they sites. In addition, the length of the vegetation parasitize. To remove this potential source of bias, growing season is reduced at high elevations, one would need to sample the Psithyrus directly in which constrains the foraging period and thus the the parasitized nest, which is very difficult be‐ time to complete the life cycle (Goulson et al. cause they are cryptic and buried deep in the soil. 2005). This forces Psithyrus species emerging at Moreover, invasion can only be confirmed when high elevations to quickly find a host to raise their the offspring leaves the nest because parasites progeny before the end of the nesting season. sometimes use the nest as periodic shelter Being able to colonize any nest encountered is (Rasmont 1988). thus likely to be an advantage. Interestingly, gen‐ In conclusion, we showed that Psithyrus eralist species occurring in cold conditions, such as species occurring at high elevation tend to parasi‐ P. rupestris and P. campestris, lack the specific tize more host species (Figure 3a), suggesting an cuticular carbohydrates that are used as recogni‐ environmental filtering effect that allows only tion chemicals (Martin et al. 2010), leaving open generalist species to survive in those conditions. the possibility of invading a larger number of host Our results not only provide direct support consis‐ species’ nests and reducing the search time tent with the altitudinal niche breath hypothesis, frontiers of biogeography 5.2, 2013 — © 2013 the authors; journal compilation © 2013 The International Biogeography Society 127 Psithyrus along an elevational gradient

but also provide echoing evidence consistent with Cameron, S.A., Hines, H.M. & Williams, P.H. (2007) A compre‐ Rapoport’s rule as applied to an altitudinal gradi‐ hensive phylogeny of the bumble bees (Bombus). Biological Journal of the Linnean Society 91, 161–188. ent. Interestingly, generalist species able to occur Currie, D.J. (1991) Energy and large‐scale patterns of animal‐ at higher elevation have a more basal position in and plant‐species richness. The American Naturalist, the phylogeny, suggesting that the specialized 137, 27–49. Goulson, D. (2010) Bumblebees: behaviour, ecology, and species may have evolved from generalist ances‐ conservation. Oxford University Press. tors. However, this requires further research using Goulson, D., Hanley, M.E., Darvill, B., Ellis, J.S. & Knight, M.E. a larger Psithyrus species sample because incom‐ (2005) Causes of rarity in bumblebees. Biological Con‐ servation, 122, 1–8. plete taxon sampling could bias ancestral state Goulson, D., Lye, G.C. & Darvill, B. (2008) Diet breadth, coex‐ reconstruction (Litsios & Salamin 2012). Recent istence and rarity in bumblebees. Biodiversity and studies are starting to reveal a common pattern Conservation, 17, 3269–3288. for the prevalence of generalist species in more Gutiérrez, D. & Menéndez, R. (1998) Phenology of butterflies along an altitudinal gradient in northern Spain. Jour‐ severe conditions at high elevations in trophic and nal of Zoology, 244, 249–264. mutualistic interactions, in insects and plants Krasnov, B.R., Shenbrot, G.I., Khokhlova, I.S., Mouillot, D. & (Pellissier et al. 2012a). Further investigations of Poulin, R. (2008) Latitudinal gradients in niche breadth: empirical evidence from haematophagous other species’ interactions and niche breadths ectoparasites. Journal of Biogeography, 35, 592–601. along altitudinal gradients are thus required to Litsios, G. & Salamin, N. (2012) Effects of phylogenetic signal assess the universality across biotic interactions of on ancestral state reconstruction. Systematic Biology, 61, 533–8. the altitudinal niche breadth hypothesis. MacArthur, R.H. (1955) Fluctuations of animal populations and a measure of community stability. Ecology, 36, Acknowledgments 533–536. MacArthur, R.H. (1972) Geographical ecology. New York: We thank all the people who helped with the field Harper & Row. work. Special thanks to F. Amiet for his help with Martin, S.J., Carruthers, J.M., Williams, P.H. & Drijfhout, F. P. identifications and S. Litsios for English correc‐ (2010) Host specific social parasites (Psithyrus) indi‐ cate chemical recognition system in bumblebees. tions. We are thankful to J. Beck, R. Field and the Journal of Chemical Ecology, 36, 855–63. two anonymous reviewers for their useful com‐ Orme, C.D.L., Freckleton, R.P., Thomas, G.H., Pedzoldt, T. & ments and remarks. This study was funded by The Fritz, S. (2011) Caper: comparative analyses of phy‐ logenetics and evolution in R. European Commission (ECOCHANGE project, con‐ Parker, G. & Stuart, R. (1976) Animal behavior as a strategy tract nr. FP6 2006 GOCE 036866) and NSF grant optimizer: evolution of resource assessment strate‐ no. 31003A‐125145 (BIOASSEMBLE project). GL is gies and optimal emigration thresholds. The American Naturalist, 110, 1055–1076. supported by NSF grant CRS113‐125240. LP was Pellissier, L., Litsios, G., Fiedler, K., Pottier, J., Dubuis, A., Prad‐ supported by The Danish Council for Independent ervand, J.‐N., Salamin, N. & Guisan, A. (2012a) Loss of Research grant no. 12‐126430. The computations interactions with ants under cold climate in a regional myrmecophilous butterfly fauna. Journal of Biogeog‐ were performed at the Vital‐IT Center for high‐ raphy, 39, 1782–1790. 1 performance computing of the Swiss Institute of Pellissier, L., Pradervand, J.‐N., Williams, P.H., Litsios, G., Bioinformatics. Cherix, D. & Guisan, A. (2012b) Phylogenetic related‐ ness and proboscis length contribute to structuring bumblebee communities in the extremes of abiotic References and biotic gradients. Global Ecology and Biogeogra‐ phy, 22, 577–585. Amiet, F. (1996) Hymenoptera Apidae, 1. Teil. Allgemeiner Pradervand, J.‐N., Pellissier, L., Rossier, L., Dubuis, A., Guisan, Teil, Gattungsschlüssel, die Gattungen Apis, Bombus A. & Cherix, D. (2011) Les Alpes vaudoises, un havre und Psithyrus. Insecta Helvetica (Fauna) 12, 98pp. de diversité pour les bourdons (Bombus Latreille, Antonovics, J. & Edwards, M. (2011) Spatio‐temporal dynam‐ Apidae). Mitteilungen der Schweizerischen ics of bumblebee nest parasites (Bombus subgenus entomologischen Gesellschaft – Bulletin de la Société Psythirus ssp.) and their hosts (Bombus spp.). Journal Entomologique Suisse, 84, 46–66. of Animal Ecology, 80, 999–1011. Rapoport, E.H. (1975) Areografia: estrategias geograficas de Brinck, P. (1974). Strategy and dynamics of high altitude fau‐ las especies. Mexico City: Fondo de Cultura Economi‐ nas. Arctic and Alpine Research, 6, 107–116. ca.

1 http://www.vital‐it.ch

Rasmont, P. (1988) Monographie écologique et zoogéogra‐ Vázquez, D.P. & Stevens, R.D. (2004) The latitudinal gradient phique des Bourdons de France et de Belgique in niche breadth: concepts and evidence. The Ameri‐ (Hymenoptera, Apidae, Bombinae). PhD thesis, Fa‐ can Naturalist, 164, 1–19. culté des Sciences agronomiques de l’Etat, Gembloux, Wcislo, W.T. (1987) Parasitism rates in relation to nest site in 309+LXIIpp. bees and wasps (Hymenoptera: Apoidea). Journal of Roulston, T.H. & Goodell, K. (2011) The role of resources and Insect Behavior, 9, 643–656. risks in regulating wild bee populations. Annual Re‐ Williams, P. (2008) Do the parasitic Psithyrus resemble their view of Entomology, 56, 293–312. host bumblebees in colour pattern? Apidologie, 39, Sanders, N.J. (2002) Elevational gradients in ant species rich‐ 637–649. ness: area, geometry, and Rapoport’s rule. Ecography, Yom‐Tov, Y. & Geffen, E. (2005) Host specialization and lati‐ 25, 25–32. tude among cuckoos. Journal of Avian Biology, 36, Stevens, G.C. (1992) The elevational gradient in altitudinal 465–470. range: an extension of Rapoport’s latitudinal rule to Zimmermann, N. & Kienast, F. (1999) Predictive mapping of altitude. The American Naturalist, 140, 893–911. alpine grasslands in Switzerland: species versus com‐ Stireman, J.O., Dyer, L.A., Janzen, D.H. et al. (2005) Climatic munity approach. Journal of Vegetation Science, 10, unpredictability and parasitism of caterpillars: impli‐ 469–482. cations of global warming. Proceedings of the Na‐ tional Academy of Sciences of the United States of America, 102, 17384–17387. Edited by Jan Beck