Behavioral Ecology Advance Access published online on January 17, 2007
Behavioral Ecology, doi:10.1093/beheco/arl103
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Aerial dispersal plasticity under different wind velocities in a salt marsh wolf spider
Terrestrial Ecology Unit, Department of Biology, Ghent University, K.L. Ledeganckstraat 35, B-9000 Ghent, Belgium
Address correspondence to D. Bonte. E-mail: dries.bonte{at}ugent.be.
Received 4 September 2006; revised 27 November 2006; accepted 12 December 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Dispersal can be regarded as a process operating both between and within patches of suitable habitat. For uncontrolled dispersal processes, the risk of crossing the borders of the habitat patch and arriving in the unsuitable landscape matrix will increase with decreasing patch area, in particular when the distance between isolated habitat patches is larger than the species' average dispersal capacity. Ballooning dispersal in spiders can be considered as a passive dispersal process, in which dispersed distances depend on the prevalent wind velocity. We executed a reaction norm analysis to analyze how dispersal propensity of the salt marsh wolf spider Pardosa purbeckensis depended on population characteristics (patch size) and the environment (wind velocity). Dispersal propensity was affected by the interaction between wind velocity and maternal patch size. Ballooning propensities decreased with decreasing salt marsh size. Interestingly, genotypes from large salt marshes show higher ballooning propensities under higher wind velocities, whereas those from small habitat patches show their highest dispersal propensity under low wind velocities. Crossing reaction norms and subsequently strong genotype x environment interaction variation was observed in all populations but tended to be lower in genotypes from large salt marshes. It is likely that this pattern results from differences in wind velocity–related costs of within-habitat dispersal in salt marshes of different sizes.
Key words: ballooning, behavioral plasticity, Lycosidae, Pardosa purbeckensis, reaction norm.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Theoretical and empirical studies suggest that dispersal, that is, one-way movements to breeding locations away from the place of birth (Dingle 1996
), is favored by natural selection through kin competition, inbreeding avoidance, and temporal variability in habitat quality, and counterselected by a high level of spatial environmental heterogeneity and niche specialization (reviews in Clobert et al. 2001; Bullock et al. 2002). Dispersal is often considered at the landscape level, that is, as movements between patches of suitable habitat. When these patches become so isolated that between-habitat dispersal is seriously hampered, selection against dispersal can be expected. However, in case the species occurs in aggregated, discrete microhabitats, dispersal implies also within-habitat movements for which dispersal may remain evolutionary beneficial, despite large distances between the habitat patches. The nature of the dispersal process involved, that is, active dispersal with control versus passive uncontrolled dispersal, may influence the outcome of selection processes (Bonte et al. 2006) in interaction with area and connectivity of the habitat patch. In case of controlled dispersal, costs of within-habitat dispersal are intrinsically equal for habitats of different sizes, whereas for uncontrolled dispersal processes, the risk of crossing the borders of the habitat patch and arriving in the unsuitable landscape matrix will increase with decreasing patch area or increased habitat specialization (Bonte et al. 2003).
In contrast to predator-imposed escape behavior (e.g., Plaistow and Siva-Jothy 1999; Relyea 2004; Laurila et al. 2006; Wohlfahrt et al. 2006), between-population variation in plasticity of dispersal behavior is currently hardly addressed. For spiders characterized by passive dispersal (ballooning), the length of the dispersal path is determined to a large extent by environmental factors related to landscape structure, such as the spatial configuration of "obstacles" and wind velocity (Bell et al. 2005), with higher velocities resulting in longer dispersal events (Thomas et al. 2003). Consequently, the probability of reaching suitable habitat is unpredictable at the moment an individual decides to disperse because it depends on intrinsically unpredictable wind currents. According to Thomas et al. (2003), single flight dispersal distances equal on average 978 m but may reach 3000 m and more under wind velocities of 1.75 m s–1. In contrast to smaller species, ballooning of larger wolf spiders is restricted to periods in which spiderlings have optimal mass and size, that is, during the second or third instar (Richter 1970, 1971). This period takes only 2 or 3 weeks and its starting time largely depends on the timing of egg sac production and hatching.
Ballooning is characterized by stereotypic behavioral adaptations, referred to as tiptoe behavior (e.g., Weyman et al. 2002; Bell et al. 2005). As any other behavioral trait, it can be expected to show phenotypic plasticity (West-Eberhard 1989; Pigliucci 2001). Adaptive behavioral plasticity (i.e., in which the behavioral plasticity is beneficial; Gotthard and Nylin 1995) is expected to evolve more readily than adaptive morphological plasticity because there are more potential cues within the environment (West-Eberhard 1989), and changes in behavior will become effective in a shorter time period. Because different attributes of behavioral plasticity (e.g., height and slope of the reaction norm) may evolve independently (e.g., Schlichting and Levin 1986), only comparative analysis of behavioral plasticity (e.g., Burghardt and Krause 1999; Van Buskirk 2002; Bell and Stamps 2004) will allow insight into microevolutionary mechanisms in spatially structured populations (Relyea 2002). For instance, an equal average phenotypic trait over all genotypes in different environments may either result from the combined effect of genetic invariability and canalization (i.e., the phenotype is independent of environmental variation—a flat reaction norm and the absence of plasticity and of genetic variation in plasticity) or from the antagonistic genetic variation in plasticity (i.e., genotypes have crossing reaction norms). Assessment of this distinction is important in the context of evolutionary conservation because absence of plasticity implies low resilience toward changing environmental conditions (Via and Lande 1987; Gillespie and Turelli 1989).
Here, we report the results of an experiment testing for patterns in individual dispersal propensity and its plasticity toward 2 wind velocities for a highly stenotypic salt marsh wolf spider from several highly isolated populations. Assuming that wind velocity–related and area-related costs differ between spiderlings of salt marshes from different size, we predict 1) that ballooning frequency will increase with increasing patch size under both low and high wind velocities, with a stronger canalization toward low dispersal propensity in small populations, 2) a more pronounced increase in dispersal propensity under high wind velocities, and 3) that higher spatial heterogeneity within larger salt marshes will result in a stronger genotype x environment interaction.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Females of the species and populations
Pardosa purbeckensis (F.O. Pickard-Cambridge 1895) is the most dominant wolf spider species in European coastal salt marshes (Baert and Maelfait 1999
; Pétillon et al. 2004). The species has an annual–biennial life cycle with reproduction in May–June. Egg sacs are attached to spinnerets of wandering females and hatch from June onward to August. Females carrying egg sacs were sampled in 7, strongly isolated populations along the coastline between Normandy (Mont Saint Michel) and the Netherlands (Zwin); see Figure 1. These salt marshes differ in age and area of suitable habitat (Table 1), which are significantly positively correlated (R = 0.83; P < 0.05). Earlier population genetic analyses through allozymes indicated low genetic differentiation among populations and no differences in genetic diversity within populations (Maelfait J.-P, Hendrickx F, unpublished data). Offspring were individually reared under standardized temperature (20 ± 2 °C) and 16:8 h day/night regimes and fed ad libidum with Sinella curviseta springtails until they reached the second postnatal instar. Because all spiderlings hatched in the laboratory, they had not experienced their natal environment.
|
|
Assessment of aerial dispersal propensity
Ballooning behavior of P. purbeckensis was quantified under standardized laboratory conditions. Groups of 5 individuals originating from the same mother were placed on a platform, of which escaping was prevented by a vertical border. On the platform, mosses were provided as shelter and ballooning was possible by climbing on vertical screws. A mesh was placed 5 cm above the platform, enabling attachment of the produced silk threads. Only ballooning spiderlings could get attached to this mesh. After 30 min testing, all attached spiderlings were collected. Two similar experimental setups were used with upward current of 0.3 and 0.8 m s–1, temperature of 30 °C, and relative aerial humidity of 40–50%, respectively. These wind velocities were chosen because preliminary tests indicated that species from all populations engage in ballooning dispersal under these velocities, whereas much lower frequencies were recorded at velocities higher than 1.2 m s–1 (Bonte D, unpublished data). Given the binomial distribution of the analyses (see further), sufficient variation (and power) is consequently guaranteed.
A total of 1342 spiderlings, originating from 60 clutches were tested. Neither maternal size (F6,54 = 1.61; P = 0.163) nor clutch size (F6,53 = 0.19; P = 0.977) differed between the sampled populations.
Statistical analysis
The observed ballooning probabilities b of spiderlings from clutch i, location (area) j, under wind velocity k was modeled using the following binomial mixed model:
 | (1) |
ijk are variance components of genotype (clutch), genotype-by-wind (G x E), and residuals, respectively. Analyses were conducted for 7 populations, 2 wind velocities (0.3 and 0.8 m s–1), and 60 genotypes by Bayesian estimation using a Monte-Carlo Markov chain (MCMC) procedure in WinBugs v. 1.4. (Spiegelhalter et al. 2003). Within this regression model (Equation 1), the following parameter posterior distributions were estimated: intercept 
0, slope patch area area, effect wind velocity wind, slope interaction wind velocity x patch area windarea; variance due to clutch varG and variance due to clutch x wind interaction varGE. In a second model, varGE estimates (1000 random samples from MCMC permutation) for each of the 7 populations were regressed on patch area in order to assess the distribution of the estimated posterior regression slopes between genotype x environment interaction variance and patch area of the population of origin. In the latter, posterior distributions of the intercept (b0) and the regression slope (b1) were estimated. Area of the salt marshes was log10 transformed.
Flat priors for regression coefficients were drawn from a normal distribution with a mean of 0 and a standard deviation (SD) of 1000 because we did not have any a priori knowledge on parameter distributions. Priors for variance components were drawn from a positively constraint uniform distribution with a mean of 0 and SD 1. To assure accurate MCMC simulations from the prior distributions, an initial "burn in" of 10 000 iterations was performed and discarded from analysis. This was followed by 1 00 000 iterations for both analyses. After visual inspections for possible autocorrelation and assessing chain convergence (Brooks-Gelman-Rubin diagnostics; Brooks and Gelman 1998), the mean and SD of each parameter (regression coefficients and variance estimates) were calculated, as were the 2.5th and 97.5th percentiles of the sample. These were used to describe the 95% Bayesian credible interval. The mean and SD of the posterior distributions of the parameters were used to assess effects of each of the variables.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Effects of wind velocity and area of the maternal habitat
We were not able to assess directly the main effects of wind velocity and the area of the maternal salt marsh on the probability of ballooning because both show a strong interaction (Table 2). Ballooning probabilities within clutches are lower in populations from small salt marshes compared with those from the larger salt marshes (on average 10% in the smallest areas up to on average 30% in the largest ones; Figure 2). As expected, ballooning probabilities show a steeper increase with increasing marsh size at the highest wind speed.
|
|
Genotypes from the larger populations (>200 ha) show higher ballooning probabilities (up to 40%) under high wind velocity (0.8 m s–1) compared with low wind velocity (up to 25%). In contrast, genotypes from smaller salt marshes show higher dispersal probabilities under low wind velocity. Overall genotype (clutch) and genotype x environment (clutch x wind) variation are substantial and according to the Bayesian credibility intervals larger than zero (Table 2).
Variation in genotype x environment interactions
Median values and credibility intervals of the posterior distributions of G x E variation indicate the presence of this interaction variation within each of the 7 populations (Figure 3). G x E variation shows a tendency (slope= –0.072 ± 0.002 SD; Table 2) to be lower in very large habitats. Reaction norms are indeed proportionally more parallel in clutches from larger habitat (Figure 4). Consequently, higher observed phenotypic similarity under both wind velocities in genotypes from small areas does not originate from flat reaction norms, but rather from crossing reaction norms, as indicated by the large estimated G x E variation in these populations.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results indicate that the probability of the onset of passive dispersal (ballooning) for a salt marsh wolf spider depended on the interaction between factors related to the population (patch size) and the strength of the transporting medium (wind velocity). As expected, ballooning probabilities showed a steeper increase with increasing marsh size at the highest wind speed. This means that habitat loss negatively affects dispersal probabilities and that this effect is more pronounced under higher wind velocities. This lower dispersal probability was however not accompanied by a loss of genetic variation in plasticity, but rather related to a higher amount of crossing reaction norms. Because age and area of the several populations were positively correlated in our study, we can conclude that founder effects (Desender et al. 1998
; Hanski et al. 2004) are not responsible for the observed variation in ballooning behavior and plasticity.
Dispersal behavior and reaction norms
This study differs from other reaction norms approaches related to environmental gradients (e.g., stress: Köhler et al. [2000], prey availability: Burghardt and Krause [1999], or predator regime: Van Buskirk [2002]) in a sense that the environmental variation was not imposed during a long term on organisms during development. Instead, we studied short-term plasticity in behavior toward rapidly changing environmental factors and used reaction norm analysis to assess responses of different genotypes. This resulted in a strong among-genotype variation of dispersal behavior in relation to the prevailing wind speed. In the majority of studies on dispersal plasticity, strong context dependence dispersal behavior is detected, predominantly, because the individuals experienced differences in habitat structure or quality (Van Dyck and Baguette 2005), rather than due to genetic factors. In our setup, however, spiderlings are reared in the laboratory from wild-caught females, so that an effect of experience with the native habitat can be excluded. Moreover, the environment differed only in one factor (i.e., wind velocity). Consequently, the observed dispersal plasticity is most likely to be caused by genetic or maternal factors (genotype or population of origin) and may potentially reflect adaptive (i.e., beneficial) patterns through selection or maternal effects.
Adaptive plasticity
The observed patterns in reaction norms corresponded to what was expected with respect to differences in dispersal costs between small and large populations. This indicates that ballooning dispersal propensity as a behavioral plastic trait is adaptive, that is, beneficial under local conditions (Gotthard and Nylin 1995). This adaptive plasticity may be the result of ultimate (evolutionary) factors as well as of proximate maternal effects. Although no differences in female size (related to spiderling size; Bonte et al. 2006), clutch size, or sex-ratio distortion (with possible gender-related dispersal; Ims 1990) during controlled breeding (Bonte D, unpublished data) were observed, maternal effects cannot be completely excluded (Ims and Hjermann 2001; Bowler and Benton 2005) because, for example, hormonal differences may have caused similarity among siblings (Massot and Clobert 2000).
Ultimate selection pressures on dispersal traits but habitat area (i.e., temporal and spatial variability in habitat quality, avoidance of competition; Bowler and Benton 2005) can be assumed to be similar in all salt marshes. The underlying mechanism to explain the observed patterns in dispersal plasticity is hence most likely selection against dispersal in small populations. Despite evolutionary benefits of within-habitat dispersal in naturally disturbed environments (Southwood 1962), ballooning may be hampered by the large distances between coastal salt marshes, whereby multiple takeoffs are hardly possible due to the hostile matrix. This means that the risk of arriving in the nonsuitable matrix after within-habitat dispersal is probably too high for individuals from small salt marshes, in particular, when wind velocity is high. Ballooning propensities of genotypes from small areas are indeed observed to be lower under high wind velocity than under low wind velocity, indicating the high dispersal costs under these conditions.
For this reason, dispersal patterns may be shaped by an increased cost of within-habitat movements, rather than by costs of interhabitat dispersal (Bonte et al. 2006). This implies that a decreased probability of dispersal behavior is related to the reduction in area of suitable habitat, as has been shown in wind-dispersed plants of inshore islands (Cody and Overton 1996). Additionally, lower ballooning probabilities under low wind velocities for genotypes from larger salt marshes suggest low benefits of short aerial trajectories. As recently shown for a related wolf spider, similar short distances can be covered by hitchhiking dispersal (i.e., transportation by the mother; Bonte et al. forthcoming), with lower risks due to trajectory control abilities (Morse 2002; Bonte et al. 2004; Krauss and Morse 2005).
Loss of plasticity at the population-level but not its genetic variation
Variation in reaction norms is a prerequisite to maintain genetic variation in spatially or temporally heterogeneous environments with overlapping generations (Via and Lande 1987; Gillespie and Turelli 1989) and assures sufficient resilience toward changing environmental conditions. We found that ballooning plasticity in all investigated populations was accompanied by variation in plasticity (G x E interactions), indicating the presence of genetic variation (Via et al. 1995). Moreover, populations from small salt marshes seemed to show increased variation due to antagonistic crossings of reaction norms. As evidenced by van Kleunen and Fisher (2001), genetic variation in plasticity can be maintained on small spatial scales of only a few meters (Stratton 1994; van Kleunen and Fisher 2001). Temporal variation in weather conditions, including wind velocity, can here be assumed to maintain dispersal plasticity because spiderlings from different genotypes differ in the timing of dispersal events due to variation in the time window of hatching and juvenile development. Therefore, maladaptive reaction norms in small salt marshes (i.e., genotypes promoting dispersal under high wind velocities) may only be counterselected during periods of high wind velocities but not during periods of low wind velocities. In contrast, these behavioral types can be expected to be favored in populations from large areas in disadvantage of the genotypes not promoting dispersal under high wind velocities. This may result in more uniform reaction norms, which was confirmed by the lower observed variation in plasticity.
|
ACKNOWLEDGEMENTS |
|---|
The first author holds a postdoc fellowship at the Fund for Scientific Research—Flanders (FWO). Kris Struyf (conservator of the Zwin reserve) and ir. Jean-Louis Herrier (leading manager Administration Nature and Landscape Nature division; Baai van Heist, Ijzermonding) kindly provided permission for sampling in the respective salt marshes. This study was partially funded by the FWO grant G.0202.06.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Baert L and Maelfait J-P. (1999) The spider fauna of the Belgian salt marshes. Bull Inst R Sci Nat Belg 69:5–18.
Bell AM and Stamps JA. (2004) Development of behavioural differences between individuals and populations of sticklebacks, Gasterosteus aculeatus. Anim Behav 68:1339–1348.[CrossRef]
Bell JR, Bohan DA, Shaw EM, Weyman GS. (2005) Ballooning dispersal using silk: world fauna, phylogenies, genetics and models. Bull Entomol Res 95:69–114.[CrossRef][Web of Science][Medline]
Bonte D, Lens L, Maelfait J-P. (2004) Lack of homeward orientation and increased mobility result in high emigration rates from low-quality fragments in a dune wolf spider. J Anim Ecol 73:643–650.[CrossRef]
Bonte D, Van Belle S, Maelfait J-P. Forthcoming. Maternal-care and reproductive-state dependent mobility determine initial natal dispersal in a wolf spider. Anim Behav.
Bonte D, Vanden Borre J, Lens L, Maelfait J-P. (2006) Geographic variation in wolfspider dispersal behaviour is related to landscape structure. Anim Behav 73:655–662.[CrossRef]
Bonte D, Vandenbroecke N, Lens L, Maelfait J-P. (2003) Low propensity for aerial dispersal in specialist spiders from fragmented landscapes. Proc R Soc Lond B Biol Sci 270:1601–1607.[Medline]
Bowler DE and Benton TG. (2005) Causes and consequences of animal dispersal strategies: relating individual behaviour to spatial dynamics. Biol Rev 80:205–225.[Medline]
Brooks S and Gelman A. (1998) General methods for monitoring convergence of iterative simulations. J Comput Graph Stat 7:434–457.[CrossRef]
Dispersal ecology. (2002) (British Ecological Society, Blackwell PublishingIn Bullock JM, Kenward RE, Hails RS (Eds.). , Cambridge).
Burghardt GM and Krause MA. (1999) Plasticity of foraging behavior in garter snakes (Thamnophis sirtalis) reared on different diets. J Comp Psychol 113:277–285.[CrossRef]
Dispersal. (2001) (Oxford University PressIn Clobert J, Danchin E, Dhondt AA, Nichols JD (Eds.). , Oxford).
Cody ML and Overton JM. (1996) Short-term evolution of reduced dispersal in island plant populations. J Ecol 84:53–61.[CrossRef]
Desender K, Backeljau T, Delahaye K, De Meester L. (1998) Age and size of European saltmarshes and the population genetic consequences for ground beetles. Oecologia 114:503–513.[CrossRef]
Dingle H. (1996) Migration. The biology of life on the move. (Oxford University Press, Oxford).
Ellison AM. (2004) Bayesian inference in ecology. Ecol Lett 7:509–520.[CrossRef]
Falconer DS. (1989) Introduction to quantitative genetics. (Longman, New York).
Gillespie JH and Turelli M. (1989) Genotype-environment interactions and the maintenance of polygenic variation. Genetics 121:129–138.
Gotthard K and Nylin S. (1995) Adaptive plasticity and plasticity as an adaptation: a selective review of plasticity in animal morphology and life history. Oikos 74:3–17.[CrossRef][Web of Science]
Hanski I, Erälahti C, Kankare M, Ovaskainen O, Siren H. (2004) Variation in migration propensity among individuals maintained by landscape structure. Ecol Lett 7:958–966.[CrossRef]
Ims RA. (1990) Determinants of natal dispersal and space use in grey-sided voles, Clethrionomys rufocanus: a combined field and laboratory experiment. Oikos 57:106–113.[Medline]
Ims RA and Hjermann DO. (2001) Condition-dependent dispersal. In Clobert J, Danchin E, Dhondt AA, Nichols JD (Eds.). Dispersal(Oxford University Press, Oxford) pp. 203–216.
Köhler HR, Zanger M, Eckwert H, Einfeldt I. (2000) Selection favours low hsp70-levels in chronically metal-stressed soil arthropods. J Evol Biol 13:569–582.[CrossRef]
Krauss JM and Morse DH. (2005) Seasonal habitat shifts in an intertidal wolf spider: proximal cues associated with migration and substrate preference. J Arachnol 33:110–123.[CrossRef]
Laurila A, Pakkasmaa S, Merila J. (2006) Population divergence in growth rate and antipredator defences in Rana arvalis. Oecologia 147:585–595.[CrossRef][Web of Science][Medline]
Massot M and Clobert J. (2000) Processes at the origin of similarities in dispersal behaviour among siblings. J Evol Biol 13:707–719.[CrossRef]
Merila J, Soderman F, O'Hara R, Rasanen K, Laurila A. (2004) Local adaptation and genetics of acid-stress tolerance in the moor frog, Rana arvalis. Conserv Genet 12:513–527.[CrossRef]
Morse DH. (2002) Orientation and movement of wolf spiders Pardosa lapidicina (Araneae, Lycosidae) in the intertidal zone. J Arachnol 30:601–609.[CrossRef]
Pétillon J, Ysenl F, Le Gleut S, Lefeuvre J-C, Canard A. (2004) Responses of spider communities to salinity and flooding in a tidal salt marsh (Mont St.-Michel Bay, France). Arthropoda Selecta Spec Issue 1:235–248.
Pigliucci M. (2001) Phenotypic plasticity: beyond nature and nurture. (John Hopkins University Press, Baltimore (MD)).
Plaistow S and Siva-Jothy MT. (1999) The ontogenetic switch between odonate life history stages: effects on fitness when time and food are limited. Anim Behav 58:659–667.[CrossRef][Web of Science][Medline]
Relyea RA. (2002) Costs of phenotypic plasticity. Am Nat 159:272–282.[CrossRef]
Relyea RA. (2004) Fine-tuned phenotypes: tadpole plasticity under 16 combinations of predators and competitors. Ecology 85:172–179.
Richter CJ. (1970) Aerial dispersal in relation to habitat in eight wolf spider species (Pardosa, Araneae, Lycosidae). Oecologia 5:200–214.[CrossRef]
Richter CJ. (1971) Some aspects of aerial dispersal in different populations of wolf spiders, with particular reference to Pardosa amentata (Araneae, Lycosidae). Misc Pap Landbouwhogesch Wagening 8:77–88.
Schlichting CD and Levin DA. (1986) Phenotypic plasticity: an evolving plant character. Biol J Linn Soc 29:37–47.
Southwood TRE. (1962) Migration of terrestrial arthropods in relation to habitat. Biol Rev 37:171–214.[Medline]
Spiegelhalter D, Thopas A, Best N, Lunn D. (2003) Winbugs user manual version 1.4. (MRC Biostatistics Unit, Cambridge (UK)).
Stratton DA. (1994) Genotype-by-environment interactions for fitness of Erigeron annuus show fine-scale selective heterogeneity. Evolution 48:1607–1618.[CrossRef]
Thomas CFG, Brain P, Jepson PC. (2003) Aerial activity of linyphiid spiders: modelling dispersal distances from meteorology and behaviour. J Appl Ecol 40:912–927.[CrossRef]
Van Buskirk J. (2002) A comparative test of the adaptive plasticity hypothesis: relationships between habitat and phenotype in anuran larvae. Am Nat 160:87–102.[CrossRef]
Van Dyck H and Baguette M. (2005) Dispersal behaviour in fragmented landscapes: Routine or special movements? . Basic and Appl Ecol 53:5–545.
van Kleunen M and Fisher M. (2004) Adaptive evolution of plastic foraging in a clonal plant. Ecology 82:3309–3319.[CrossRef]
Via S, Gowulkiewicz R, de Jong G, Scheiner SM, Schlichting CD, Tienderen PHV. (1995) Adaptive phenotypic plasticity: concensus and controversy. Trends Ecol Evol 10:212–216.[CrossRef]
Via S and Lande R. (1987) Evolution of genetic variability in a spatially heterogeneous environment: effects of genotype-environment interaction. Genet Res 49:147–156.[Web of Science][Medline]
Vugts HF and van Wingerden WKRE. (1976) Meteorological aspects of aeronautic behaviour of spiders. Oikos 27:433–444.[CrossRef]
Wade PR. (2000) Bayesian methods in conservation biology. Conserv Biol 14:1308–1316.[CrossRef]
West-Eberhard MJ. (1989) Phenotypic plasticity and the origin of diversity. Annu Rev Ecol Syst 20:249–278.[Medline]
Weyman GS. (1993) A review of the possible causative factors and significance of ballooning in spiders. Ethol Ecol Evol 5:279–291.
Weyman GS, Sunderland KD, Jepson PC. (2002) A review of the evolution and mechanisms of ballooning by spiders inhabiting arable farmland. Ethol Ecol Evol 14:307–326.
Wohlfahrt B, Mikolajewski DJ, Joop G, Suhling F. (2006) Are behavioural traits in prey sensitive to the risk imposed by predatory fish? Freshw Biol 51:76–84.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Bonte, J. M. J. Travis, N. De Clercq, I. Zwertvaegher, and L. Lens Thermal conditions during juvenile development affect adult dispersal in a spider PNAS, November 4, 2008; 105(44): 17000 - 17005. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||