• Dispersal;
  • emigration;
  • flowers;
  • habitat quality;
  • immigration;
  • movement;
  • Parnassius;
  • patch;
  • sex ratio


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Abstract 1. Nectar flower abundance was manipulated through flower removal, and sex ratio was manipulated by moving individual butterflies within a series of nine alpine meadows. The movement and abundance of the butterfly Parnassius smintheus in the meadows were monitored using mark–release–recapture methods.

2. A total of 937 butterflies, 698 males and 239 females, was captured. There were 223 observed between-meadow movements. Fifty-two per cent of males and 35% of females moved among meadows.

3. The immigration of male butterflies was related positively to nectar flowers, host plant abundance, and female butterflies. Male emigration was not affected by any of the treatments. The number of males captured was related positively to nectar flowers and host plants but not affected by sex ratio. The number of resident male butterflies was greater in meadows containing flowers and was related positively to host plant abundance, but unaffected by sex ratio.

4. Flower removal, sex ratio, and abundance of Sedum had no significant effect on the abundance, movement, or residence time for female butterflies, in part due to small sample size.

5. The fact that males immigrate to higher quality meadows suggests that male butterflies are assessing meadow quality, either by sampling meadows or potentially from a distance using olfactory cues.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The movement of organisms is a fundamental process for population biology at both local and regional scales. At a local scale, the abundance of organisms is determined by birth and immigration adding individuals and by death and emigration subtracting individuals. At larger spatial scales, dispersal affects gene flow among local populations (Wright, 1969; Saccheri et al., 1998), fluctuations in the abundance in local populations (Pulliam, 1988; Matter, 1999), and the persistence of networks of local populations (Hanski, 1994). Thus, understanding factors affecting dispersal is of vital importance for population biology (Hanski, 1998).

Examinations of dispersal have focused on both non-mechanistic and mechanistic causes. From a non-mechanistic perspective, organisms can be thought of as diffusing particles, the movement of which is governed strictly by physical processes. This approach has been successful in relating numbers of immigrants and emigrants to factors such as patch area and patch isolation (Kareiva, 1983; Turchin, 1986; Lomolino, 1990; Matter, 1996, 1997), predicting the spread of invasive species (Lewis, 1997), and as dispersal components of population models (Hanski, 1994; Lele et al., 1998). Despite the success of non-mechanistic approaches, it is also important to understand the mechanisms of dispersal. It is frequently desirable to know how specific factors influence immigration to and emigration from particular sites. From this mechanistic perspective, dispersal can be related to factors such as population density (Denno & Peterson, 1995), boundary and matrix conditions (Stamps et al., 1987; Haddad, 1999; Roland et al., 2000), and, often, local habitat quality (Kuussaari et al., 1996; Peterson, 1997; Brommer & Fred, 1999).

Despite extensive study of the spatial population dynamics of butterflies (reviewed by Hanski and Kuussaari, 1995), there have been surprisingly few experimental studies of factors affecting their dispersal (but see Haddad, 1999). Correlative research has identified nectar flowers (White & Levin, 1981; Kuussaari et al., 1996; Peterson, 1997; Brommer & Fred, 1999) and mating opportunities (Baguette et al., 1998) as important factors affecting the dispersal of butterflies. Similarly, host plant abundance, population density, patch size, and boundary conditions may influence the movement of some species (Shapiro, 1970; Murphy & White, 1984).

In the work reported here, the effects of two aspects of habitat quality, mating opportunities and nectar flower abundance, on the abundance and movement of adults of the butterfly Parnassius smintheus Doubleday were examined. Parnassius smintheus is abundant in alpine and sub-alpine meadows in the Rocky Mountains of Canada and the United States, despite several closely related species being threatened in other parts of the world (Heath, 1981; Väisänen & Somerma, 1985). Its larval host plant, Sedum lanceolatum, occurs in gravelly sites above the tree-line (Fownes, 1999; Roland et al., 2000). Parnassius smintheus is univoltine, with an adult flight period from July to late August. Male butterflies are generally more apparent than females. Females tend not to fly as readily as do males, often searching for oviposition sites by crawling. Despite the difference in the ability to observe each sex, estimated dispersal distances are similar for the two sexes (Roland et al., 2000). Most dispersal occurs through non-forested areas when possible, and there appears to be little movement across valleys (Keyghobadi et al., 1999; Roland et al., 2000). Adult butterflies nectar-feed predominantly on yellow-flowered species such as Potentilla fruticosa, Solidago multiradiata, Senecio canus, and S. lanceolatum (Fownes, 1999).

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Experiments and mark–recapture of butterflies took place in a series of nine meadows along Powderface Ridge, in the Kananaskis region of Alberta, Canada (50°56′N, 114°52′W; Fig. 1). This site is ≈ 10 km SE of similar sites used in earlier movement studies (Roland et al., 2000). The meadows are located just above the tree-line at an elevation of ≈ 2100 m. The vegetation in the meadows is dominated by grasses, sedges, and wild flowers including S. lanceolatum. Meadows are bordered on their lower slopes by forest consisting of lodgepole pine Pinus contorta, subalpine fir Abies lasiocarpa, and Engelman spruce Piecea engelmannii.


Figure 1.  Experimental treatments and map of the study area, Powderface Ridge, Alberta, Canada. Shaded areas indicate flower removal meadows, outlined areas indicate meadows with a natural abundance of flowers. F indicates female-biased meadows; M indicates male-biased meadows. The lower case letters correspond to the meadow identity in Table 1. The inset shows the location of the study area. The province of Alberta is outlined.

Download figure to PowerPoint

Two factors related to meadow quality for adult butterflies were manipulated: the abundance of nectar flowers and mating opportunities. Nectar flower abundance was altered by the removal of all yellow flowers in four of the nine meadows. Flower removal was done by hand, removing flowers and buds from individual plants. The flower removal manipulation was initiated before the commencement of mark–recapture and was maintained throughout the study.

Mating opportunity was manipulated by altering the sex ratio through the removal and addition of butterflies in four meadows. Two meadows were made male biased (males added, females removed) and two meadows were made female biased (females added, males removed). The density of butterflies in each meadow was unaltered, as each butterfly removed was replaced by one of the opposite sex, except for one instance where one butterfly was inadvertently not released. All butterfly transfers were solely between meadows under investigation. Butterflies were moved beginning on the first day of mark–recapture and movement of butterflies continued through the duration of the study. One male-biased and one female-biased meadow were crossed with the flower removal treatment, while the three remaining meadows were unmanipulated and served as controls.

The abundance of S. lanceolatum in each meadow was also quantified. The density of Sedum was estimated by counting the number of plants in 31 randomly placed 1.69-m2 quadrats in each large meadow and 16 quadrats in the three smaller meadows. The total abundance of S. lanceolatum in each meadow was estimated by multiplying its mean density by meadow area (Table 1). The area of each meadow was estimated from aerial photographs (1:20 000 scale) taken in 1993. Photographs with the meadows centred in the photograph were used, thereby minimising distortion due to differences in angle.

Table 1.   Summary of meadow characteristics. Meadow areas were estimated from (1:20 000) aerial photographs. The density of Sedum lanceolatum was measured in 31 (1.69 m2) plots (16 plots were used for meadows C, D, and E).
MeadowArea (ha)Sedum density/1.69m2SD

Meadows were sampled for butterflies on 13 dates between 26 July and 17 August 2000. Each meadow was sampled eight to 11 times. Butterflies were captured using hand nets and each newly captured individual was given a unique three-letter mark on the hindwing using a felt-tipped pen (Roland et al., 2000). For all captures, the date, location, and the butterfly's sex and identity mark were recorded. To ensure equal sampling effort among meadows, mark–recapture continued until ≈ 66% of all butterflies captured in a meadow had been captured on the same day.

For each meadow, the mean per census period for the number of captures (excluding multiple captures of the same individual in the same meadow on one day), individual butterflies, resident butterflies, and immigrations were calculated. This approach was taken to account for differences in the number of censuses per meadow. Residents were defined simply as butterflies that were captured consecutively within the same meadow. The proportion of butterflies emigrating and the duration that butterflies spent in a meadow were also examined. The proportion emigrating was defined as the total number of emigrations from a meadow divided by the total number of individuals seen in that meadow. The minimum residence time of individual butterflies was calculated for all butterflies that were recaptured. For cases of emigration, half of the length of time between captures was assigned to the meadow of emigration and half to the meadow of immigration. More sophisticated population estimation methods such as Jolly–Seber could not be used because of the low number of recaptures and the inability to generate transition probabilities among meadows. As such, these estimates should be considered as relative indices rather than as precise estimates of abundance or movement. The number of emigrants from and immigrants to each meadow was taken directly from the mark–recapture data. This method underestimates the total number of migrants but should be unbiased.

Overall, the experimental design consisted of two levels of flower removal (removal and control) crossed with three levels of sex ratio (male biased, female biased, and control). As some factorial combinations were unreplicated, the analysis was conducted using a general linear model. For all analyses, the abundance of S. lanceolatum was used as a covariate. This metric combines the effect of variation in both meadow size and Sedum density among meadows (Table 1). Analysis from non-experimental situations has shown that Sedum abundance is a better predictor of P. smintheus abundance, immigration, and emigration than is meadow area (S. F. Matter et al., unpublished). Thus, including Sedum abundance as a covariate in the model accounts for differences in butterfly abundance, immigration, and the proportion emigrating that are due to differences in population size. Male and female butterflies were analysed separately as they may respond differently to the experimental treatments.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

A total of 937 individual butterflies, 698 males and 239 females, was captured. The overall recapture rate was 34.7%. Males were recaptured more often (40.5%) than were females (18.0%). These estimates exclude multiple captures of the same individual within the same meadow during one census period. Butterflies that were moved appeared to behave normally on release, however translocated butterflies had a lower number of captures (1.23 ± 0.07 SE) than did butterflies that were not moved (1.55 ± 0.03 SE, separate variance t = 4.43, d.f. = 93.82, P < 0.01), indicating possible handling effects (Mallet et al., 1987). For this reason, and to clarify experimental results, butterflies that were moved (n = 65) were excluded from analyses (except for evaluating treatment effects, below). Thus, experimental effects are for butterflies that were in their appropriate meadow.

The movement of male and female butterflies resulted in significantly different sex ratios among the treatment groups (control vs. male-biased t = −4.57, P < 0.01; control vs. female-biased t = 3.17, P < 0.05). Over the entire study, the mean proportion of females in male-biased patches was 0.04, the mean proportion of females in female-biased patches was 0.35, and the mean proportion of females in control patches was 0.23 (Fig. 2). The data also show that over the course of the study the populations became increasingly female. No systematic quantification of flowers was conducted to evaluate the effectiveness of the flower removal treatment. An attempt was made to remove all yellow flowers within flower removal meadows, however this was not accomplished. Estimates derived from photographs of one meadow prior to and immediately after initial flower removal showed a 98.7% reduction in the number of flowers. Due to regrowth between visitations of each meadow, the reduction in the number of nectar flowers achieved over the course of the study was closer to 75%.


Figure 2.  Sex ratios of butterflies in the meadows. The proportion female is shown for each meadow for censuses in which more than five butterflies were captured. Data include butterflies that were moved and the natural population of butterflies.

Download figure to PowerPoint

There were 223 between-meadow movements, 209 by males and 14 by females. Significantly more male butterflies that were recaptured at least once (152/291) moved between meadows one or more times than did female butterflies recaptured at least once (14/40; Gadj = 4.21, d.f. = 1, P < 0.05).

Sex ratio, the abundance of nectar flowers, and the abundance of Sedum within meadows had significant effects on male P. smintheus. More male butterflies were captured in, immigrated to, and were residents in meadows containing a natural abundance of flowers compared with meadows from which flowers had been removed (Table 2, Figs 3–5). The removal of flowers had no effect on the proportion of male butterflies emigrating or on the length of time that males spent in meadows. Significantly more males immigrated to female-biased meadows than to control meadows. Sex ratio had no significant effect on the number of male captures, number of male individuals, number of male residents, proportion of males emigrating, or on the length of time that males remained in meadows. Male captures, immigrants, individuals, and residents all increased with the abundance of Sedum. Sedum abundance had no effect on male residence time or on the proportion of males emigrating. Flower removal, sex ratio, and Sedum abundance had no significant effect on any metric for female butterflies, in part due to small sample size.

Table 2.   Summary of results of general linear models.
Type III SSd.f.FPType III SSd.f.FP
Mean number of captures
Corrected model1254.0869.550.09864.0061.730.411
Sex ratio304.2326.950.12627.2922.210.311
Sedum abundance673.36130.780.03123.1213.750.192
Flowers sex ratio91.7422.100.3234.2520.340.744
Error43.762  12.332  
Total3500.729  185.599  
Corrected total1297.848  76.338  
Mean number of individuals
Corrected model580.3865.760.15547.2461.610.431
Sex ratio130.3323.880.20519.0021.950.339
Sedum abundance334.90119.950.04719.5414.000.183
Flowers sex ratio25.4020.760.5692.6620.270.786
Error33.582  9.772  
Total1789.199  139.989  
Corrected total613.968  57.008  
Mean number of immigrants
Corrected model53.16623.910.0410.3063.160.260
Sex ratio17.65223.810.0400.0020.080.923
Sedum abundance20.94156.500.0170.1117.080.117
Flowers sex ratio5.7527.760.1140.1023.040.248
Error0.742  0.032  
Total128.099  0.659  
Corrected total53.918  0.338  
Proportion emigrating
Corrected model0.08760.950.5950.021161.960.376
Sex ratio0.04221.390.4190.014824.090.196
Sedum abundance0.01410.940.4350.006113.400.207
Flowers sex ratio0.04221.380.4200.000320.080.927
Error0.0312  0.00362  
Total0.7129  0.04099  
Corrected total0.1178  0.02488  
Mean number of residents
Corrected model44.64613.720.0691.1162.760.290
Sex ratio8.7028.020.1110.4123.040.247
Sedum abundance22.48141.460.0230.2013.030.224
Flowers sex ratio4.8824.500.1820.1521.090.478
Error1.082  0.132  
Total107.069  2.669  
Corrected total45.738  1.248  
Residence time
Corrected model6.4464.830.18228.2961.390.476
Sex ratio4.3329.740.0938.2121.210.453
Sedum abundance3.23114.510.0631.8210.530.541
Flowers sex ratio1.2722.840.2607.0321.030.492
Error0.452  6.812  
Total120.679  117.009  
Corrected total6.898  35.108  

Figure 3.  The effects of sex ratio and nectar flowers on the mean number of Parnassius smintheus captures and individuals per census period. Estimated marginal means shown are adjusted for the covariate, Sedum abundance. Error bars = 1 SE.

Download figure to PowerPoint


Figure 4.  The effects of sex ratio and nectar flowers on the mean number of Parnassius smintheus immigrating per census and the proportion of individuals emigrating. Estimated marginal means shown are adjusted for the covariate, Sedum abundance. Error = 1 SE.

Download figure to PowerPoint


Figure 5.  The effects of sex ratio and nectar flowers on the number of resident Parnassius smintheus and residence time. Estimated marginal means shown are adjusted for the covariate, Sedum abundance. Error bars = 1 SE.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Meadow quality, in terms of mating opportunity, the abundance of nectar flowers, and host plant abundance, has a significant effect on the movement and abundance of male P. smintheus. The results indicate that more males immigrate to meadows that contain an abundance of flowers and female butterflies. A greater number of male butterflies stay within patches that contain flowers, however they do not tend to remain for a significantly longer duration in meadows containing flowers. Interestingly, these aspects of meadow quality do not appear to affect male emigration from meadows.

Roland et al. (2000) examined the dispersal patterns of P. smintheus on a nearby ridge top. By comparing movements between pairs of meadows, they found that butterflies tended to leave meadows with few butterflies and immigrate to meadows containing an abundance of butterflies, and suggested that this pattern may result from butterfly abundance simply being a surrogate for meadow quality. The results of the present study indicate that male butterflies do indeed aggregate to higher quality meadows, in terms of nectar flower and host plant abundance, and that female butterflies themselves may be a resource for males. Thus, the patterns seen by Roland et al. (2000), which were dominated by male butterflies, probably represent a combination of conspecific (Odendaal et al., 1988) and resource-based attraction.

Most studies examining the effects of habitat quality on movement patterns have found greater emigration from low-quality habitats rather than greater immigration to high-quality habitats. High rates of emigration for butterflies are often associated with a lack of nectar resources (Gilbert & Singer, 1973; Murphy & White, 1984; Kuussaari et al., 1996; Brommer & Fred, 1999) or with a segregation of larval and adult resources (Gilbert & Singer, 1973; Brommer & Fred, 1999). In a notable exception, Kuussaari et al. (1996) found that immigration of male Melitea cinxia L. increased with increasing nectar flower abundance within patches. There has been less research into the role of mating opportunity as a component of quality. For female butterflies, high rates of emigration are often associated with male harassment (Shapiro, 1970). Baguette et al. (1998) found that Proclossiana eunomia females emigrate at high male density and that males emigrate at low female density. They attributed female emigration to male harassment and male emigration to their searching for females. In an experimental study of the red milkweed beetle Tetraopes tetraophthalmus (Forster), Lawrence (1987) found that male emigration increased as patches became increasingly male biased. Herzig and Root (1996) found that male goldenrod beetles Trirhabda virgata LeConte immigrated preferentially to goldenrod patches containing females than to patches without beetles. It is unclear, however, whether the response was related to mating opportunity or was simply a response to conspecifics, given that males also immigrated preferentially to patches containing both sexes over empty patches.

The fact that more male P. smintheus immigrate to high-quality meadows implies that butterflies may be assessing meadow quality. Males may sample two or more meadows, gaining information on their relative quality, and immigrate accordingly. Alternatively, greater immigration to meadows containing flowers and females may indicate that males can judge meadow quality from a distance. Female butterflies and/or flowers may emit olfactory cues to which males respond. Two lines of evidence are consistent with a sampling hypothesis but less so with a remote sensing hypothesis. (1) There is a great deal of movement among meadows. Over half of all males recaptured were captured in at least two different meadows, thus males have the opportunity to sample meadow quality. Even assessing quality in only two meadows would produce the patterns seen here. (2) The proportion of males emigrating does not vary among meadows on the basis of quality. If butterflies are sampling meadows remotely and moving along a gradient of meadow quality, directed movement from low-quality to high-quality meadows and a tendency to remain in peak-quality meadows would be expected, and hence greater emigration from lower quality meadows. In contrast, relatively equal amounts of emigration, as seen in this experiment, would be expected if butterflies need to leave meadows to sample other meadows and assess quality. If male butterflies are sampling meadows to judge quality, their response to quality is dependent on being able to access other meadows. In situations where access to other meadows is more restricted, different movement patterns with respect to meadow quality may be expected.

Overall, for females there were no effects of flowers, host plant abundance, or sex ratio on abundance or on movement. Kuussaari et al. (1996) found that while the immigration of male Melitea cinxia increased with flower abundance, female immigration did not. They attributed this difference to differences in energetic demands. Because males disperse further and more frequently than the more sedentary females, males may be more responsive to flowers to meet their energetic needs. This may also be the case for P. smintheus, females of which are relatively sedentary even though the net displacement of females is equal to that of males (Roland et al., 2000). In the present study, the general trend was for females to show a positive response to flowers and a negative response to males. Because females that were moved were excluded from analyses, the number of females observed was quite low: there were only 40 recaptures of females and 14 observations of female movement. Drawing conclusions regarding female responses to meadow quality from these limited data would be tenuous at best. Given the low recapture rate of female P. smintheus, augmented approaches to mark–release–recapture, such as harmonic radar (Roland et al., 1996), may be needed to collect sufficient data on their movement.

Parnassius species in many parts of the world, especially in Europe, are threatened or endangered. Many of these species share a population structure similar to P. smintheus due to the patchy distribution of their resources (e.g. Brommer & Fred, 1999; Meglecz et al., 1999). Thus, the results presented here are probably relevant to the conservation and protection of these species. If Parnassius emigrate at a high rate regardless of quality, as found here for P. smintheus and found by Brommer and Fred (1999) for P. apollo, conservation efforts should be directed towards promoting connectivity among habitat patches. For P. smintheus, forested areas act as a barrier to dispersal relative to open habitat (Roland et al., 2000). Thus, connectivity may be maintained by preventing forest encroachment and maintaining unforested corridors among patches. Increasing habitat quality through augmenting nectar flowers and host plant abundance may result in increased butterfly abundance and immigration but may not prevent substantial emigration.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank G. Dearborne, C. Gibbs, E. Lang, and K. Sabourin for assisting with the fieldwork, and two anonymous reviewers for comments on an earlier version of this paper. This research was supported by an Alberta Conservation Association Challenge Grant in Biodiversity to S. F. Matter and a NSERC Operating Grant to J. Roland. S. F. Matter was supported by a Killam Postdoctoral Fellowship.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Baguette, M., Vansteenwegen, C., Convi, I. & Neve, G. (1998) Sex-biased density dependent migration in a metapopulation of the butterfly Proclossiana eunomia. Acta-Oecologia, 19, 1724.
  • Brommer, J.E. & Fred, M.S. (1999) Movement of the Apollo butterfly Parnassius apollo related to host plant and nectar plant patches. Ecological Entomology, 24, 125131.
  • Denno, R. & Peterson, M. (1995) Density-dependent dispersal and its consequences for population dynamics. Population Dynamics (ed. by N.Cappuccino and P.Price), pp. 113130. Academic Press, New York.
  • Fownes, S. (1999) Habitat of the butterfly Parnassius smintheus. MSc thesis, Department of Biological Sciences, University of Alberta, Edmonton, Canada.
  • Gilbert, L.E. & Singer, M.C. (1973) Dispersal and gene flow in a butterfly species. American Naturalist, 107, 5872.DOI: 10.1086/282817
  • Haddad, N.M. (1999) Corridor and distance effects on interpatch movements: a landscape experiment with butterflies. Ecological Applications, 9, 612622.
  • Hanski, I. (1994) A practical model of metapopulation dynamics. Journal of Animal Ecology, 63, 151162.
  • Hanski, I. (1998) Metapopulation dynamics. Nature, 396, 4149.
  • Hanski, I. & Kuussaari, M. (1995) Butterfly metapopulation dynamics. Population Dynamics (ed. by N.Cappuccino and P.Price), pp. 149171. Academic Press, New York.
  • Heath, J. (1981) Threatened Rhopalocera (Butterflies) in Europe. Council of Europe, Nature and Environmental Service, Strasbourg, France.
  • Herzig, A.L. & Root, R.B. (1996) Colonization of host patches following long-distance dispersal by a goldenrod beetle, Trirhabda virgata. Ecological Entomology, 21, 344351.
  • Kareiva, P.M. (1983) Local movement in herbivorous insects: applying a passive diffusion model to mark–recapture field experiments. Oecologia, 57, 322327.
  • Keyghobadi, N., Roland, J. & Strobeck, C. (1999) Influence of landscape on the population genetic structure of the alpine butterfly Parnassius smintheus (Papilionidae). Molecular Ecology, 8, 14811495.DOI: 10.1046/j.1365-294X.1999.00726.x
  • Kuussaari, M., Nieminen, M. & Hanski, I. (1996) An experimental study of migration in the Glanville fritillary butterfly Melitaea cinxia. Journal of Animal Ecology, 65, 791801.
  • Lawrence, W.S. (1987) Effects of sex ratio on milkweed beetle emigration from host plant patches. Ecology, 68, 539546.
  • Lele, S., Taper, M.L. & Gage, S. (1998) Statistical analysis of population dynamics in space and time using estimating functions. Ecology, 79, 14891502.
  • Lewis, M. (1997) Variability, patchiness, and jump dispersal in the spread of an invading population. Spatial Ecology (ed. by D.Tilman and P.Kareiva), pp. 4669. Princeton University Press, Princeton, New Jersey.
  • Lomolino, M.V. (1990) The target area hypothesis: the influence of island area on immigration rates of non-volant mammals. Oikos, 57, 297300.
  • Mallet, J., Longino, J.T., Murawski, D., Murawski, A. & Simpson de Gamboa, A. (1987) Handling effects in Heliconius: where do all the butterflies go? Journal of Animal Ecology, 56, 377386.
  • Matter, S.F. (1996) Interpatch movement of the red milkweed beetle, Tetraopes tetraophthalmus: individual responses to patch size and isolation. Oecologia, 105, 447453.
  • Matter, S.F. (1997) Population density and area: the role of within and between patch processes. Oecologia, 110, 533538.
  • Matter, S.F. (1999) Population density and area: the role of within and between generation processes over time. Ecological Modelling, 118, 261275.
  • Meglecz, E., Neve, G., Pecsenye, K. & Varga, Z. (1999) Genetic variations in space and time in Parnassius mnemosyne (L.) (Lepidoptera) populations in north-east Hungary: implications for conservation. Biological Conservation, 89, 251259.DOI: 10.1016/S0006-3207(99)00006-3
  • Murphy, D. & White, R. (1984) Rainfall, resources, and dispersal in southern populations of Euphydryas editha (Lepidoptera: Nymphalidae). Pan-Pacific Entomologist, 60, 350354.
  • Odendaal, F.J., Turchin, P. & Stermitz, F.R. (1988) An incidental-effect hypothesis explaining aggregation of males in a population of Euphydryas anicia. American Naturalist, 132, 735749.DOI: 10.1086/284885
  • Peterson, M. (1997) Host plant phenology and butterfly dispersal. Ecology, 78, 167180.
  • Pulliam, H. (1988) Sources, sinks, and population regulation. American Naturalist, 132, 652661.DOI: 10.1086/284880
  • Roland, J., Keyghobadi, N. & Fownes, S. (2000) Alpine Parnassius butterfly dispersal: effects of landscape and population size. Ecology, 81, 16421653.
  • Roland, J., McKinnon, G., Backhouse, C. & Taylor, P.D. (1996) Even smaller radar tags on insects. Nature, 381, 120.
  • Saccheri, I., Kuussaari, M., Kankare, M., Vikman, P., Fortelius, W. & Hanski, I. (1998) Inbreeding and extinction in a butterfly metapopulation. Nature, 392, 491494.
  • Shapiro, A. (1970) The role of sexual behavior in density related dispersal of pierid butterflies. American Naturalist, 104, 367372.DOI: 10.1086/282670
  • Stamps, J., Buchner, M. & Krishnen, V. (1987) The effects of edge permeability and habitat geometry on emigration from habitat patches. American Naturalist, 129, 533552.DOI: 10.1086/284656
  • Turchin, P. (1986) Modeling the effect of patch host size on Mexican bean beetle emigration. Ecology, 67, 124132.
  • Väisänen, R. & Somerma, P. (1985) The status of Parnassius mnemosyne (Lepidoptera, Papilionidae) in Finland. Notulae Entomologicae, 65, 109118.
  • White, R. & Levin, M. (1981) Temporal variation in vagility: implications for evolutionary studies. American Midland Naturalist, 105, 348357.
  • Wright, S. (1969) Evolution and Genetics of Populations, Vol. 2. The Theory of Gene Frequencies. University of Chicago Press, Chicago, Illinois.

Accepted 1 September 2001