SEARCH

SEARCH BY CITATION

Keywords:

  • avian endocrinology;
  • climate change;
  • energetic constraint;
  • environmental stressors;
  • species distribution

Summary

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

1. Species’ geographical ranges can be limited by a variety of biotic and abiotic factors. Physiological challenge in response to unsuitable environmental conditions can establish limits to geographical ranges.

2. We studied the physiology of Song Wrens (Cyphorhinus phaeocephalus) across their geographical range on the isthmus of Panama, an area characterized by a strong rainfall gradient. Wrens are common on the Caribbean slope of the isthmus where annual rainfall is greatest, but wren abundance declines towards the south as annual rainfall declines. Song Wrens are completely absent from the driest third of the isthmus.

3. We searched for the existence of a physiologically induced distribution limit by measuring body condition (an integrative measure of energy balance), hematocrit (% packed red blood cells in a given blood sample), and corticosterone levels (CORT, a steroid hormone that regulates the availability of energy and the endocrine stress response) in males and females. We caught birds by luring them into nets when they responded to playback of conspecific song.

4. Wrens living in drier habitat near the geographical range edge were significantly more likely to have abnormally low hematocrit scores. Baseline CORT levels were negatively associated with rainfall in one of our three best-fit path models, indicating potential energetic challenge in some individuals. Maximum CORT levels during a 60-min period of restraint correlated significantly only with sex, being higher in females. Birds with the poorest body condition lived at the dry end of the gradient. Birds on the wet end of the gradient responded fastest to conspecific song.

5. Environmental conditions vary across geographical ranges and may influence the physiological conditions of organisms, thereby enforcing limits to species’ distributions. Highly specialized species, such as birds of the rain forest understory, may be especially susceptible to environmental variation associated with changing climatic conditions.


Introduction

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

Understanding why species can live in some locales but not others has long interested biologists (Allee 1926; Gaston 2009). Identifying the factors that delineate species’ range borders is challenging because they can vary from interactions with other species to a diverse set of abiotic variables and because they can occur over large spatial scales (Brown 1995; Gaston 2003, 2009). Physiological limitations of particular species, ranging from plants to insects and vertebrates, can explain some range boundaries (Brown 1995; Aerts, Cornelissen & Dorrepaal 2006; Calosi et al. 2010; Hoffmann 2010). Such limitations can be influenced by gene flow preventing local adaption or evolution, patterns of selection, and genetic variability in traits that create range edges (Kirkpatrick & Barton 1997; Futuyma 2010; Hoffmann 2010). Understanding how climatic factors influence species distribution is currently of special importance because of projected climate changes caused by anthropogenic forcing (IPCC, 2007; Gaston 2009). Tropical species may be especially sensitive to climate change (Deutsch et al. 2008; Tewksbury, Huey & Deutsch 2008; Huey et al. 2009; but see Hoffmann 2010).

Tolerance of specific ranges of temperatures or moisture conditions is known to limit bird distributions (Smith 1977; Karr 1980; Karr & Freemark 1983; Root 1988; Walther et al. 2002; Böhning-Gaese & Lemoine 2004; Williams & Middleton 2008). Even in moist tropical forests, limits to many bird species distributions appear to align with geographical gradients in annual rainfall. In the Panama Canal region, for example, a steep rainfall gradient occurs as moisture-laden air masses move from the Caribbean Sea over land, dropping the heaviest rains on the northern Caribbean slope and progressively less rain as the air mass moves south. In central Panama, species richness in a variety of taxa is positively correlated with annual rainfall (Condit et al. 2004; Robinson et al. 2004; Odegaard 2006; Rompréet al. 2007); in birds, dozens of species common in the wettest forests of the Caribbean slope decline and disappear towards the Pacific slope as annual rainfall declines (Robinson et al. 2004; Rompréet al. 2007). Among those species, understory insectivores appear to be especially sensitive to reductions in annual rainfall. Many understory insectivores are also affected by habitat changes that alter moisture and humidity in forests, such as forest fragmentation and degradation (Thiollay 1992; Stouffer & Bierregaard 1995; Canaday 1997; Stratford & Stouffer 1999; Sekercioglu et al. 2002; Williams & Middleton 2008). The response of understory insectivores to changes in moisture and humidity could be caused by a variety of mechanisms, including a decrease in abundance and availability of their invertebrate prey as a result of the prey’s sensitivity to habitat conditions, and physiological intolerance of the birds to the habitat conditions. For example, abundance of terrestrial insectivores is more strongly positively correlated with abundance of litter arthropods – their prey – than abundance of other avian foraging guilds to litter or foliage arthropods (Karr & Brawn 1990), and species in many guilds of tropical forest bird, including terrestrial insectivores, avoid xeric microhabitats (Karr & Freemark 1983).

We studied Song Wrens [Cyphorhinus phaeocephalus (Sclater 1860)], a terrestrial, understory insectivore, throughout their distribution alongside the Panama Canal. Song Wrens are widespread on the wet, Caribbean side of the Canal corridor, but are rare to absent on the drier, Pacific side (c. 20 km; Ridgely & Gwynne 1989). Thus, a prominent boundary for the species passes through the isthmus and is coincident with the occurrence of the rainfall gradient. A potential limit to the wrens’ distribution, therefore, is the change in moisture and humidity along the rainfall gradient (Morton 1978; Karr & Freemark 1983; TR Robinson, unpublished data).

We explored the potential of physiological constraints as limits to the distribution of Song Wrens by evaluating signs of physiological challenge in individuals living at the drier end of the rainfall gradient near the geographical range boundary. We measured hematocrit, body condition, and the hormone corticosterone (CORT). Hematocrit, the per cent of packed red blood cells per unit blood volume, is widely used in clinical settings to assess health status – primarily anaemia – and can be used to assess condition in non-clinical studies (Dawson & Bortolotti 1997; Piersma et al. 2000; Kasprzak, Hetmański & Kulczykowska 2006). For example, migratory Blue Tits [Cyanistes caeruleus (Linnaeus 1758)] with small fat stores have lower hematocrit levels than birds with higher fat scores (Svensson & Merilä 1996). However, the relationship between hematocrit and condition often exists only when the population sampled includes birds in extremely poor condition (Dawson & Bortolotti 1997; Piersma et al. 2000). Hematocrit can also vary with sex and life-history stage (Morton 1994; Svensson & Merilä 1996; Piersma et al. 2000; Kasprzak, Hetmański & Kulczykowska 2006; but see Dawson & Bortolotti 1997). For example, hematocrit can be influenced by both reproductive hormones and the physiological changes that accompany moult and migration (Morton 1994; Kasprzak, Hetmański & Kulczykowska 2006). Finally, hematocrit levels can be influenced by infection (Dawson & Bortolotti 1997).

Data that associate CORT levels with environmental conditions throughout a species’ range are rare (Dunlap & Wingfield 1995). CORT is a systemic marker for physiological demands related to energy and is released in response to a diversity of stressors, such as inclement weather, food or water shortage, parasitism and psychological challenge (Cain & Lien 1985; Wingfield et al. 1998; Sapolsky, Romero & Munck 2000; Busch & Hayward 2009). For instance, CORT levels can be elevated when an animal is living in sub-optimal habitat (Marra & Holberton 1998; Suorsa et al. 2003; Busch & Hayward 2009). In the short term, high glucocorticoid titres alter physiology and behaviour in a way that helps individuals to cope with energetic challenge and psychological stress (Wingfield & Ramenofsky 1999; Sapolsky, Romero & Munck 2000; Busch & Hayward 2009). However, chronically high glucocorticoid titres can be damaging, for example depressing the immune system, causing muscle wasting and neuronal cell damage, and inhibiting reproductive capabilities (Sapolsky, Romero & Munck 2000; Korte et al. 2005; Busch & Hayward 2009). The relationship between baseline CORT levels and survival is complicated, likely being positive or neutral up to a point at which the negative effects of CORT overwhelm physiological systems and survival and CORT titres are negatively correlated (Busch & Hayward 2009).

We hypothesized that habitat at the dry end of the isthmus of Panama’s rainfall gradient could be suboptimal for Song Wrens and could present an energetic challenge to the species, thereby limiting its distribution. We predicted that, in response to the possible energetic challenges of living in potentially suboptimal environments, Song Wrens at the dry end of the rainfall gradient would have higher baseline CORT levels and higher peak CORT levels during a 60-min restraint. We also predicted that birds living at the dry end of the gradient would have lower body condition and lower hematocrit because of the energetic and physiological challenges induced by environmental conditions there. We expected that our variables of interest would not correlate with birds’ response to our capture technique – luring birds into nets with playback of conspecific song – a common assumption in avian field biology. We conducted the study mid-wet season, a period when birds engage in both moult and breeding, the two life-history stages thought to be most energetically demanding for non-migrants (Cyr, Wikelski & Romero 2008). We hypothesized that individuals in poor condition, potentially because of living in suboptimal habitat, would be more likely to show signs of physiological challenge when energetic demand is at its peak.

Materials and methods

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

Song Wrens are non-migratory, understory insectivores that occupy moist, lowland tropical forests from Honduras to western Ecuador (Ridgely & Gwynne 1989). The species forages on the forest floor, eating arthropods in the leaf litter (Skutch 1940; Stiles, Skutch & Gardner 1989; Robinson, Robinson & Edwards 2000). Pairs are socially monogamous and maintain year-round, all-purpose territories. In Panama, the species’ reproductive season coincides with the rainy season, starting in June and ending in December (Robinson, Robinson & Edwards 2000). While reproductive attempts are asynchronous during this period, the species’ gonads are developed from at least April to October (Wikelski et al. 2003) and the majority of birds have nests in June and September–November (Robinson, Robinson & Edwards 2000). Young birds stay with their parents for up to 8 months (Robinson, Robinson & Edwards 2000).

We sampled 56 Song Wrens (39 males: 20 second-year, 19 after-second-year; 17 females: two hatch-year, 9 second-year, six after-second-year) in lowland, moist forest north of the continental divide along the Panama Canal (Republic of Panama, 9°N 79°W) at locations in Camino de Cruces and Soberania National Parks, the Fort Sherman area, and forest near Achiote Road. This transect extends c. 45 km and spans elevation of c. 0–125 m above sea level. All samples were taken during the rainy season in August and September 2002. Birds were located using song playback with a Sony tape player (WM-FS473, Sony, New York, NY, USA) and Radio Shack speakers (Mini-amplifier speaker, catalogue number 277-1008C, Radio Shack, Fort Worth, TX, USA). Once we located a bird or family group, we set up a mist-net close to the responding bird(s) and used playback to draw the bird(s) into the net. Our playback recording came from songs of at least four individuals. We recorded the total time that each bird was exposed to playback before capture, which ranged from 1 to 75 min (mean = 16·2 min, SD = 18·2 min). We did not sample birds before 0700 or after 1600 to avoid variation in CORT levels due to the diel cycle (Romero & Remage-Healey 2000).

Upon capture, each individual was subjected to standardized handling, restraint and sampling, termed a ‘stress series’ (Wingfield, Vleck & Moore 1992). Within 3 min of capture, we took a small sample (40 μL) of blood from the alar vein of the wing with a heparinized hematocrit tube and used this sample to measure baseline CORT levels (Wingfield, Smith & Farner 1982). Additional 40-μL samples were taken from each bird at 5, 10, 30 and 60 min after capture. These additional samples were used to measure maximum CORT levels. Between each sampling time, we placed each bird individually in its own cloth bag. After blood sampling, we collected data on mass and tarsus, wing, and tail length. Wing and tail length were measured to the nearest 0·5 mm using a straight ruler, tarsus was measured to the nearest 0·1 mm using dial calipers, and mass was measured to the nearest 0·1 g with a Pesola spring scale. We scored age of birds by evaluating the crispness of the black bands on their tail feathers (T.R. Robinson & W.D. Robinson, unpublished data) and skull ossification. While some hatch-year birds were sampled, no young of the season were sampled. Birds were examined for moult in all regions of the body; however, birds were categorized as ‘moulting’ only when growing their primaries. All birds were in breeding condition, but, aside from females with edematous brood patches, we could not assess the exact stage of each individual’s reproductive state.

After sealing both ends of each hematocrit tube with clay, we stored blood samples on ice for 1–12 h, when we centrifuged samples for 10 min. We measured hematocrit of the 3-min blood sample with a hematocrit ruler (% packed red-blood-cell volume) and then collected plasma with a Hamilton syringe. Plasma and red-blood-cell samples were stored at a maximum of −20 °C and transported to the University of Washington on dry ice. Hormone levels were measured using the radioimmunoassay technique described in Wingfield, Vleck & Moore (1992) and the Supporting Information. Each individual was sexed genetically (see Supporting Information).

Statistical analyses

Exposure to playback and the presence of a net and researchers may induce an endocrine stress response. CORT peaked at either 30 or 60 min for all individuals, so we used t-tests to assess whether the duration of exposure to playback was different at the two peak CORT times in males and females.

We used Wilcoxon tests to assess if location along the rainfall gradient varied with the presence of an edematous brood patch in females or the presence of fledged young with either males or females. These analyses tested for relationships between position on the rainfall gradient and timing of reproduction. To increase our sample size, for these analyses we used data from individuals collected for this study and for Busch et al. (2008). The birds sampled for Busch et al. (2008) are different individuals than those sampled for this study, but were sampled during the same time frame as this study and in the two preceding months. Busch et al. (2008) do not address variation in physiology along the rainfall gradient.

We defined body condition as the residuals from a regression of log10 bird mass against body size (Hood, Boersma & Wingfield 1998; Busch et al. 2010; Supporting Information). Body size was defined as the first principal component (PC1) from an analysis of tarsus, wing and tail length (n = 144, r2 = 0·27, P = 0·0001). PC1 explained 53% of the variation in the data set. The loading of PC1 was: tarsus = 0·49, wing = 0·66 and tail = 0·57.

We used data from the birds sampled for Busch et al. (2008) in two other ways. First, we used data from females with edematous brood patches collected for both this study and Busch et al. (2008) to assess, with a Wilcoxon test, if hematocrit varies with the expression of an edematous brood patch. Second, we used a data set on sex hormone titres [dihydroepiandosterone (DHEA), luteinizing hormone (LH), progesterone and testosterone] presented in Busch et al. (2008) to test with a linear regression if hematocrit and sex steroid levels correlate in Song Wrens. Hematocrit data were not presented in Busch et al. (2008), so we consider these analyses new.

The hematocrit score of birds is quite variable among species (ranging from 30 to 70 in healthy individuals of different species), thus making it difficult to define abnormally low hematocrit scores in species that are not well studied. We defined the abnormally low hematocrit state as values <1 standard deviation below the mean of all hematocrit scores of Song Wrens (mean = 46, SD = 4, low hematocrit state <42). We used a Wilcoxon test to determine if birds with abnormally low hematocrit scores are found more often at the wet or dry end of the rainfall gradient.

The Panama Canal Authority (Panama Canal Authority, Meteorology and Hydrology Branch, Republic of Panama) collects precipitation data in the region surrounding the Panama Canal. Although measurements are made at more than two dozen locations across the isthmus, those data do not include all locations at which birds were sampled for this study. Distance from the continental divide and annual precipitation are tightly correlated (R2 = 0·91, F1,5 = 58·79, P < 0·001), so we used interpolation to estimate rainfall where birds were sampled. Other studies in the region have taken a similar approach (Robinson et al. 2004; Rompréet al. 2007; Rompré, Robinson & Desrochers 2008). The exact values for precipitation at each site are of little importance to this study, as our results depend more on how sampled birds are arranged along the rainfall gradient.

Path models

We conducted path analyses to test for the direct and indirect effects of factors that influence hematocrit and CORT levels. We define direct effects as those caused by interaction of a factor with the measured variable of interest and indirect effects as those caused by interaction of a factor with other factors that directly or indirectly influence the measured variable of interest. The hematocrit models included rainfall, body condition and sex as factors. We used separate models for baseline and maximum CORT levels. For each of these models, the factors we included were rainfall, body condition, the amount of time the birds were exposed to playback prior to capture, and sex. For all analyses, we built models with all possible interactions among all variables. We then took away relationships between variables and variables themselves until we found the most significant model(s) to describe the data. Additional analyses exploring the sensitivity of relationships significantly correlated with the rainfall gradient (body condition, baseline CORT and hematocrit) to sample size are presented in the Supporting Information.

Analyses were conducted using JMP 5.1, R, or AMOS 5 (University of Texas at Austin, 2002; SAS Institute, 2005; R Development Core Team, 2009). Normality was tested with the Shapiro-Wilk W test. Non-normal data sets were log transformed to achieve normality. Nonparametric statistics were used when log transformation did not achieve normality. Significance level was set at α ≤ 0·05. Results are presented as mean ± SD.

Results

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

Hematocrit

The distribution of hematocrit scores in females with and without edematous brood patches and both males and females with and without young were similar (Wilcoxon test, females with/without edematous brood patch: Z = 0·05, d.f. = 16, P = 0·96; females with/without young: Z = −1·48, d.f. = 16, P = 0·14; males with/without young: Z = −0·34, d.f. = 38, P = 0·73). In females, hematocrit levels correlated positively with ln progesterone (R2 = 0·27, F1,10 = 5·21, P = 0·05), but no other sex hormone (DHEA R2 = 0·07, F1,10 = 1·84, P = 0·21; ln testosterone R2 = 0·15, F1,10 = 3·85, P = 0·12; ln LH R2 = 0·00, F1,26 = 0·98, P = 0·33). In males, hematocrit levels correlated positively with DHEA (R2 = 0·15, F1,23 = 5·19, P = 0·03), but no other sex hormone (ln progesterone R2 = 0·03, F1,23 = 1·73, P = 0·20; ln testosterone R2 = −0·04, F1,23 = 0·09, P = 0·77; ln LH R2 = −0·02, F1,63 = 0·01, P = 0·91). Birds with abnormally low hematocrit scores (n = 7) were significantly more likely to live at the dry end of the rainfall gradient than birds with normal hematocrit scores (Fig. 1a, Wilcoxon test: Z = −2·16, d.f. = 54, P = 0·03). Males (n = 4, all moulting) and females (n = 3, 2 moulting) both occurred in the low hematocrit score category.

image

Figure 1.  The threshold for abnormally low hematocrit (a, indicated by dashed line) and significant linear regression relationships from best-fit path models (b–e). Analyses for c–e were performed with ln-transformed data, but untransformed data are depicted here.

Download figure to PowerPoint

There were two best-fit path models for hematocrit (Supporting Information Fig. S1). In both models, the only significant relationship was between rainfall and body condition (standardized regression R2 = 0·43, Fig. 1b). No model factors correlated significantly with hematocrit. The best-fit models explained a small amount of the variation in each data set: 19% for body condition and 18% for hematocrit.

CORT and body condition

Corticosterone levels in most birds peaked 60 min after capture (14 of 16 females, 31 of 38 males). The duration of exposure to playback did not affect the timing of maximum CORT levels in males (t = −1·12, d.f. = 6, P = 0·31). Because all but two females had peak CORT levels at the 60-min sample, a statistical comparison of the relationship between CORT-peak timing and exposure to playback was uninformative.

There were three best-fit path models for ln baseline CORT (Supporting Information Fig. S2). All of these models included a significant relationship between rainfall and body condition (standardized regression R2 = 0·41), body condition and ln exposure to conspecific playback (standardized regression R2 = −0·27 to −0·34), and ln exposure to conspecific playback and ln baseline CORT (standardized regression R2 = 0·33–0·38, Fig. 1, Supporting Information Fig. S2). The direction of the relationship between ln exposure to conspecific playback and ln baseline CORT was uncertain, as significant models contained the relationship in either direction (ln exposure to conspecific playback to ln baseline CORT; ln baseline CORT to ln exposure to conspecific playback). One best-fit model supported a significant relationship between rainfall and ln baseline CORT, although the amount of variance in ln baseline CORT explained by this model was low (0·07, Fig. 1e, Supporting Information Fig. S2b). No best-fit model included sex as a factor. The best-fit models explained a small amount of the variation in each data set: 17% for body condition, 11–21% for ln exposure to conspecific playback and 7–17% for ln baseline CORT.

There were two best-fit path models for maximum CORT (Fig. 1, Supporting Information Fig. S3). Both models included a significant relationship between rainfall and body condition (standardized regression R2 = 0·41), body condition and ln exposure to conspecific playback (standardized regression R2 = −0·34 to −0·35), and sex and maximum CORT (standardized regression R2 = −0·29 to −0·30). No best-fit models included relationships between maximum CORT and factors other than sex. The best-fit models explained a small amount of the variation in each data set: 17% for body condition, 11–15% for ln exposure to conspecific playback and 9–12% for maximum CORT.

Discussion

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

We predicted that, because of energetic challenge, Song Wrens living at the drier end of a rainfall gradient and closer to their geographical range limit would have lower body condition and hematocrit and higher baseline and maximum CORT levels. Consistent with our predictions, we found that birds living in drier habitat closer to their range limit had poorer body condition and were more likely to have abnormally low hematocrit scores. We found significant relationships among rainfall, body condition, time to capture (exposure to playback) and baseline CORT, although the strength of each individual relationship was sometimes relatively small. Collectively, however, our data indicate that Song Wrens living closer to the limit of their geographical range in central Panama were in worse condition and exhibited more signs of being energetically challenged. Whether this result is because of direct physiological challenge of the habitat gradient on the birds themselves or indirect influences mediated by the physiological response of the birds’ prey would need to be teased out through additional study on Song Wrens and their prey, including manipulative experiments.

The path models explained a relatively small amount of the variation of each variable (7–21%). We recognize that the explanatory factors included in these models are, thus, limited, but believe that this limited explanatory ability does not decrease the importance of the results. The results suggest that birds living on the dry side of the rainfall gradient experience energetic challenges, which could be mediated by variation in prey availability/quality and/or by other environmental and physiological variables that have been hypothesized to influence bird distributions (Karr & Freemark 1983; Root 1988; Sekercioglu et al. 2002; Böhning-Gaese & Lemoine 2004; Williams & Middleton 2008). The idea that physiological limits may influence the geographical distribution of species, while long-standing, is receiving new attention, especially in ectotherms, given the growing interest in effects of climate change (Janzen 1967; Böhning-Gaese & Lemoine 2004; Chown, Gaston & Robinson 2004; Ghalambor et al. 2006; Williams & Middleton 2008; Gaston 2009; Gaston et al. 2009; Calosi et al. 2010; Sinervo et al. 2010). However, other factors known to affect species distribution (e.g. limited availability of food or reproductive sites, predator and competitor abundance, and pathogens) may have a larger impact than the physiological variables we chose to study (Robinson & Terborgh 1995; Corten 2001; Woodworth et al. 2005; Diekötter et al. 2006; Donahue et al. 2009; Gaston 2009) and our results could be caused by physiological limitations on the birds’ arthropod prey alone if such factors indirectly influence ability of Song Wrens to acquire appropriate quantity or quality of food.

Although we did not find a significant linear relationship between hematocrit and annual rainfall at sampling locations, we did find that mean hematocrit levels were lower at the dry side of the rainfall gradient and that birds living on the dry side of the rainfall gradient were significantly more likely to have an abnormally low hematocrit score. Birds at the dry edge of the range could have low hematocrit levels for a number of reasons, including a higher incidence of the following factors: resource- or condition-induced anaemia, infection with diseases or parasites that decrease hematocrit, presence of moult, or laying in females (Morton 1994; Dawson & Bortolotti 1997; Piersma et al. 2000; Kasprzak, Hetmański & Kulczykowska 2006). Given that more than half of the birds with abnormally low hematocrit were males, it is unlikely that egg-laying females drove this relationship. Also, it is unlikely that variation in reproductive state drove the observed abnormal hematocrit-rainfall relationship, because hematocrit levels did not correlate with sex hormones known to vary significantly with reproduction in this species (Busch et al. 2008). Finally, because the majority of birds we sampled for this study were moulting (92% of males, 65% of females), it is unlikely that the relationship between rainfall and abnormally low hematocrit was driven by the presence of the moult life-history stage.

None of the best-fit path models included a significant relationship between hematocrit and body condition or hematocrit and rainfall. It is possible that hematocrit is not a good indicator of health and condition status (e.g. Dawson & Bortolotti 1997). However, it is also possible that there is little correlation between hematocrit and body condition in our data because few individuals were in poor-enough condition to be anaemic or because body condition, as we measured it, does not correlate with anaemia (Svensson & Merilä 1996; Dawson & Bortolotti 1997; Piersma et al. 2000). Finding and capturing anaemic birds would be extremely challenging, given the low probability of their survival and that they likely would not respond to our capture protocol, which mandated territorial defense. The null relationship between hematocrit and rainfall could be expected given results from other studies: data on quail and fence lizards have shown that prior experience with water-restriction and arid environments, respectively, had no effect on hematocrit (Dunlap 1995; Goldstein 1995).

The most frequently re-occurring result present in path models was the positive correlation of body condition with rainfall. Birds living at wetter sites, farther from the range limit, were typically in better condition. Direct effects of humidity on physiological functioning could have caused this relationship, but the mechanisms for such an effect are not immediately apparent. The effect of rainfall on body condition is more likely mediated by changes in prey biomass (Strong & Sherry 2000; Williams & Middleton 2008). Four studies of leaf-litter arthropods, the primary food of Song Wrens, in the Panama Canal region have shown that, in general, arthropod abundance is lower in the dry season and higher in the wet season (Willis 1976; Gradwohl & Greenberg 1982; Levings & Windsor 1982; Robinson 2000), suggesting a link between food abundance and moisture availability. However, none of those studies sampled arthropods across our entire study area. Soil and leaf-litter humidity can have a large effect on the survivorship of the arthropod community living within it: litter arthropods are at risk of desiccation and a small reduction in average relative humidity can cause death (review in Levings & Windsor 1982). In addition, rainfall allows leaf-litter decomposition to proceed more quickly, releasing nutrients that support populations of leaf-litter detritivores (Cusack et al. 2009). Given current knowledge, the hypothesis that arthropod abundance is lower in the leaf-litter towards the geographical range limit of Song Wrens is a reasonable explanation for the lower body condition of Song Wrens. Low body condition as a result of low resource abundance could reflect current food availability or carry-over effects divorced from current conditions. Additional sampling of, and perhaps experimentation with, moisture and leaf-litter arthropod abundance is needed, especially sampling that targets the microhabitat Song Wrens use in each season.

Body condition was negatively correlated with time to capture (exposure to playback) in all best-fit path models: the better an individual’s body condition, the more quickly we captured it. Birds in better condition may be more likely to engage in territory defense, and, thus, more likely to be caught in our nets when engaging in defensive behaviour against the ‘intruder.’ Furthermore, a bird in good condition may be more likely to engage in a physical altercation with a challenging intruder, flying within close proximity of the ‘intruder’. Although the relationship between body condition and aggression seems intuitive, few studies have evaluated the idea (Sapolsky 1987; Renison, Boersma & Martella 2002; Owen-Ashley et al. 2006). Nevertheless, if birds in good condition were more likely to be caught and sampled because they were more aggressive to playback, then we may have under-sampled birds in poor condition.

Given our assumption that habitat at the drier edge of the distribution is more likely to be marginal and energetically challenging to Song Wrens living there, we predicted that rainfall and baseline CORT levels would be negatively correlated. A significant negative relationship was found in just one of the three best-fit path models. The lack of significance for this relationship in all best-fit models probably indicates that the relationship is weak. Some individuals closer to the distribution edge may not be affected by marginal habitat or not affected in ways that we measured, the habitat may not be marginal, or Song Wrens may occupy patches of higher-quality habitat even near the range edge.

Because our data were correlative, we cannot be certain why baseline CORT levels may vary with rainfall. Alternative explanations for variation in baseline CORT are also likely: the forest in which we sampled birds could be more disturbed on the drier side of the distribution, leading to increases in baseline CORT because of variables other than rainfall [e.g. altered vegetation structure or increased predator abundance (Busch & Hayward 2009)]. Given that Song Wrens breed in closed canopy forest and that such forests are available from the Caribbean Sea to Pacific Ocean in central Panama (Condit et al. 2001), we doubt this explanation is a reasonable one. Another alternative explanation is that the phenology of breeding varied across the isthmus. Using data on the presence of female brood patches and of young-of-the-year, we found no evidence for this phenomenon. The energetic demands of engaging in both breeding and moult may have elevated baseline CORT levels of birds across the isthmus, dampening the potential signal of the rainfall gradient. The density of Song Wrens may correlate positively with rainfall (e.g. high population density in wetter habitat), making the change in physiology we observed across space potentially because of variation in intraspecific competition for resources and aggression. However, the patterns in body condition, hematocrit and baseline CORT data are opposite of what we would expect if the birds were in fact responding to variation in population density: high population density – expected at high rainfall sites – is typically associated with high baseline CORT levels and low body condition and hematocrit – phenomena we observed in low rainfall sites. Finally, energetic challenge in Song Wrens, as reflected by their baseline CORT titres, may be more directly correlated with the physiological performance of Song Wrens’ arthropod prey, which might track environmental or ecological conditions that are not linearly correlated with rainfall.

Our data suggest that birds living in drier habitat manifest signs of energetic challenge even in the wet season when food resources are likely most abundant. The dry season may represent a resource bottleneck that drives avian abundance, limiting population size in dry areas (Williams & Middleton 2008). There are two ways that this energy bottleneck may manifest into low abundance: mortality during the dry season or poor reproductive performance in the wet season. We found evidence of energetic challenge in Song Wrens at the dry end of the rainfall gradient during the period of reproductive activity. Further study is needed to explore whether reproductive success varies along the rainfall gradient and correlates with the variables measured here. Dry season mortality may not impact Song Wren populations because individuals shift habitat use within their territories. Song Wrens typically avoid xeric microhabitats like ridges, siting territories instead on flat terrain that often contains mesic habitat (Karr & Freemark 1983). One of us (WDR) has noticed that Song Wrens are most often found in these mesic areas during the dry season. Data from Costa Rica suggests that arthropod abundance in the dry season is higher in wetter areas (Janzen & Schoener 1968). Thus, Song Wrens living on territories with preferred habitat may not experience dramatically different environmental or prey conditions during the dry and wet seasons.

Our data are consistent with the hypothesis that physiological measures, such as baseline CORT, body condition, and hematocrit, degrade as a function of proximity to a geographical range edge. We found that body condition was lower and the incidence of abnormally low hematocrit scores was higher in Song Wrens living at the drier edge of their range in central Panama. Baseline CORT levels might be directly affected by the rainfall gradient, but associations were not strong. In addition, evidence revealed that the rainfall gradient correlates with baseline CORT levels indirectly through body condition and response to conspecific playback. Further research is needed to understand the drivers of these changes in physiology.

Changes in rainfall patterns are an important consequence of climate change for tropical species (Williams & Middleton 2008; Hoffmann 2010). If moisture and humidity do influence, directly or indirectly, the condition and distributional limits of Song Wrens and the current climatic drying trend in Panama continues as expected (Condit 1998; Neelin et al. 2006; Rauscher et al. 2008), the species’ range limit may retract northward in the future. Recent evidence suggests that climate change may have already caused a reduction in the diversity and activity of birds across the South American continent (Nores 2009), and modelling exercises predict that increased seasonality associated with warming and drying will cause changes in avian abundance, in some cases inducing extinction (Sekercioglu et al. 2008; Williams & Middleton 2008; Gasner et al. 2010). These effects of climate change on tropical bird populations could be direct – as they seem to be in tropical forest lizards (Huey et al. 2009; Sinervo et al. 2010) – or indirect – caused by the response of birds to changes in food resources (Deutsch et al. 2008; Calosi et al. 2010).

Acknowledgements

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

This study was made possible by logistical support of the Smithsonian Tropical Research Institute (STRI), especially with help from M. Leone, O. Arosemena, and R. Urriola. We thank the Autoridad Nacional del Ambiente for permission to work in the Republic of Panama. I. Ochoa assisted in the field and L. Erckmann assisted in the laboratory. Thanks to M. Hau, M. Wikelski, I. Moore, N. Perfito, S. O’Brien, S. Kitaysky, M. Ramenofsky, B. Semmens and the STRI community for help and input and to two anonymous reviewers for comments that improved the manuscript. We were supported by NSF IBN-9905679 to JCW, NSF IBN-0212587 to WDR, and a NSF Graduate Research Fellowship to DSB. This research was conducted with approval from the University of Washington’s Animal Care and Use Committee (permit number 2212-38) and the Republic of Panama’s Ministry of the Environment (permit number SE/A 050-02).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Aerts, R., Cornelissen, J.H.C. & Dorrepaal, E. (2006) Plant performance in a warmer world: general responses of plants from cold, northern biomes and the importance of winter and spring events. Plant Ecology, 182, 6577.
  • Allee, W.C. (1926) Distribution of animals in a tropical rain-forest with relation to environmental factors. Ecology, 7, 445468.
  • Böhning-Gaese, K. & Lemoine, N. (2004) Importance of climate change for the ranges, communities and conservation of birds. Birds and Climate Change. Advances in Ecological Research (eds A.P.Møller, W.Fiedler & P.Berthold), 35, p. 211236. Elsevier Academic Press, Amsterdam, Netherlands.
  • Boonstra, R., Hik, D., Singleton, G.R. & Tinnikov, A. (1998) The impact of predator-induced stress on the snowshoe hare cycle. Ecological Monographs, 68, 371394.
  • Brown, J. (1995) Macroecology. University of Chicago Press, Chicago, IL, USA.
  • Busch, D.S. & Hayward, L.S. (2009) Stress in a conservation context: a discussion of glucocorticoid actions and how levels change with conservation-relevant variables. Biological Conservation, 142, 28442853.
  • Busch, D.S., Robinson, T.R., Hahn, T.P. & Wingfield, J.C. (2008) Sex hormones in the song wren: variation with time of year, molt, gonadotropin releasing hormone, and social challenge. The Condor, 110, 125133.
  • Busch, D.S., Addis, E.A., Clark, A.D. & Wingfield, J.C. (2010) Teasing apart the effects of environment and life-history stage on corticosterone modulation in Costa Rican rufous-collared sparrows, Zonotrichia capensis costaricensis. Physiological and Biochemical Zoology, 83, 8796.
  • Cain, J.R. & Lien, R.J. (1985) A model for drought inhibition of bobwhite quail (Colinus virginianus) reproductive systems. Comparative Biochemistry and Physiology, 82A, 925930.
  • Calosi, P., Bilton, D.T., Spicer, J.I., Votier, S.C. & Atfield, A. (2010) What determines a species’ geographical range? Thermal biology and latitudinal range size relationships in European diving beetles (Coleoptera: Dytiscidae). Journal of Animal Ecology, 79, 194204.
  • Canaday, C. (1997) Loss of insectivorous birds along a gradient of human impact in Amazonia. Biological Conservation, 77, 6377.
  • Chown, S.L., Gaston, K.J. & Robinson, D. (2004) Macrophysiology: large-scale patterns in physiological traits and their ecological implications. Functional Ecology, 18, 159167.
  • Condit, R. (1998) Ecological implications of changes in drought patterns: shifts in forest composition in Panama. Climatic Change, 39, 413427.
  • Condit, R., Aguilar, S., Hernandez, A., Perez, R., Lao, S., Angehr, G., Hubbell, S.P. & Foster, R.B. (2004) Tropical forest dynamics across a rainfall gradient and the impact of an El Niño dry season. Journal of Tropical Ecology, 20, 5172.
  • Condit, R., Robinson, W.D., Ibanez, R., Aguilar, S., Sanjur, A., Martinez, R., et al. (2001) The status of the Panama Canal Watershed and is biodiversity at the beginning of the 21st Century. Bioscience, 51, 389398.
  • Corten, A. (2001) Northern distribution of North Sea herring as a response to high water temperature and/or low food abundance. Fisheries Research, 50, 189204.
  • Cusack, D.F., Chou, W.W., Yang, W.H., Harman, M.E., Silver, W.L. & Team, T.L. (2009) Controls on long-term root and leaf litter decomposition in neotropical forests. Global Change Biology, 15, 13391355.
  • Cyr, N.E., Wikelski, M. & Romero, L.M. (2008) Increased energy expenditure but decreased stress responsiveness during molt. Physiological and Biochemical Zoology, 81, 452462.
  • Dawson, R.D. & Bortolotti, G.R. (1997) Are avian hematocrits indicative of condition? American kestrels as a model. The Journal of Wildlife Management, 61, 12971306.
  • Deutsch, C.A., Tewksbury, J.J., Huey, R.B., Sheldon, K.S., Ghalambor, C.K., Haak, D.C. & Martin, P.R. (2008) Impacts of climate warming on terrestrial ectotherms across latitude. Proceedings of the National Academy of Sciences, 105, 66686672.
  • Diekötter, T., Walther-Hellwig, K., Conradi, M., Suter, M. & Frankl, R. (2006) Effects of landscape elements on the distribution of the rare bumblebee species Bombus muscorum in an agricultural landscape. Biodiversity and Conservation, 15, 5768.
  • Donahue, M., Nichols, A., Santamaria, C.A., League-Pike, P.E., Krediet, C.J., Perez, K.O. & Shulman, M.J. (2009) Predation risk, prey abundance, and the vertical distribution of three Brachyuran crabs on Gulf of Maine shores. Journal of Crustacean Biology, 29, 523531.
  • Dunlap, K. (1995) External and internal influences on indices of physiological stress: II. Seasonal and size-related variations in blood composition in free-living lizards, Sceloporus occidentalis. The Journal of Experimental Zoology, 272, 8594.
  • Dunlap, K. & Wingfield, J.C. (1995) External and internal influence on indices of physiological stress. I. Seasonal and adrenocortical secretion of free-living lizards, Sceloporus occidentalis. The Journal of Experimental Zoology, 271, 3646.
  • Futuyma, D.J. (2010) Evolutionary constraint and ecological consequences. Evolution, 64, 18651884.
  • Gasner, M.R., Jankowski, J.E., Ciecka, A.L., Kyle, K.O. & Rabenold, K.N. (2010) Projecting the local impacts of climate change on a Central American montane avian community. Biological Conservation, 143, 12501258.
  • Gaston, K.L. (2003) The Structure and Dynamics of Geographic Ranges. Oxford University Press, New York.
  • Gaston, K.J. (2009) Geographic range limits of species. Proceedings of the Royal Society B: Biological Sciences, 276, 13911393.
  • Gaston, K.J., Chown, S.L., Calosi, P., Bernardo, J., Bilton, D.T., Clarke, A., Clusella-Trullas, S., Ghalambor, C.K., Konarzewski, M., Peck, L.S., Porter, W.P., Pörtner, H.O., Rezende, E.L., Schulte, P.M., Spicer, J.I., Stillman, J.H., Terblanche, J.S. & van Kleunen, M. (2009) Macrophysiology: a conceptual reunification. The American Naturalist, 174, 595612.
  • Ghalambor, C.K., Huey, R.B., Martin, P.R., Tewksbury, J.J. & Wang, G. (2006) Are mountain passes higher in the tropics? Janzen’s hypothesis revisited. Integrative and Comparative Biology, 46, 517.
  • Goldstein, D.L. (1995) Effects of water restriction during growth and adulthood on renal function of bobwhite quail, Colinus virginianus. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology, 164, 663670.
  • Gradwohl, J. & Greenberg, R. (1982) The breeding season of antwrens on Barro Colorado Island. The Ecology of a Tropical Forest: Seasonal Rhythms and Long-Term Changes (eds E.G.Leigh, A.S.Rand & D.M.Windsor), pp. 345351. Smithsonian Institution Press, Washington, DC.
  • Hoffmann, A.A. (2010) Physiological climatic limits in Drosophila: patterns and implications. Journal of Experimental Biology, 213, 870880.
  • Hood, L.C., Boersma, P.D. & Wingfield, J.C. (1998) The adrenocortical response to stress in incubating Magellanic penguins (Spheniscus magellanicus). The Auk, 115, 7684.
  • Huey, R.B., Deutsch, C.A., Tewksbury, J.J., Vitt, L.J., Hertz, P.E., Àlvarez Pérez, H.J. & Garland, T. (2009) Why tropical forest lizards are vulnerable to climate warming. Proceedings of the Royal Society B: Biological Sciences, 276, 19391948.
  • IPCC (2007) Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, 2007. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
  • Janzen, D.H. (1967) Why mountain passes are higher in the tropics. The American Naturalist, 101, 233249.
  • Janzen, D.H. & Schoener, T.W. (1968) Differences in insect abundance and diversity between wetter and drier sites during a tropical dry season. Ecology, 49, 96110.
  • Karr, J.R. (1980) Geographic variation in the avifaunas of tropical forest undergrowth. The Auk, 97, 283298.
  • Karr, J.R. & Brawn, J.D. (1990) Food resources of understory birds in Central Panama: quantification and effects of avian populations. Studies in Avian Biology, 13, 5864.
  • Karr, J.R. & Freemark, K.E. (1983) Habitat selection and environmental gradients: dynamics in the stable tropics. Ecology, 64, 14811494.
  • Kasprzak, M., Hetmański, T. & Kulczykowska, E. (2006) Changes in hematological parameters in free-living pigeons (Columbus livia f. urbana) during the breeding cycle. Journal of Ornithology, 147, 599604.
  • Kirkpatrick, M. & Barton, N.H. (1997) Evolution of a species’ range. The American Naturalist, 150, 123.
  • Korte, S.M., Koolhaas, J.M., Wingfield, J.C. & McEwen, B.S. (2005) The Darwinian concept of stress: benefits of allostasis and costs of allostatic load and the trade-offs in health and disease. Neuroscience and Biobehavioral Reviews, 29, 338.
  • Levings, S.C. & Windsor, D.M. (1982) Season and annual variation in litter arthropod populations. The Ecology of a Tropical Forest: Seasonal Rhythms and Long-Term Changes (eds E.G.Leigh, A.S.Rand & D.M.Windsor), pp. 355387. Smithsonian Institution Press, Washington, DC.
  • Marra, P.P. & Holberton, R.L. (1998) Corticosterone levels as indicators of habitat quality: effects of habitat segregation in a migratory bird during the non-breeding season. Oecologia, 116, 284292.
  • Morton, E.S. (ed.) (1978) Reintroducing Recently Extirpated Birds into a Tropical Forest Preserve. University of Wisconsin, Madison, WI, USA.
  • Morton, M.L. (1994) Hematocrits in montane sparrows in relation to reproductive schedule. The Condor, 96, 119126.
  • Neelin, J.D., Munnich, M., Su, H., Meyerson, J.E. & Holloway, C.E. (2006) Tropical drying trends in global warming models and observations. Proceedings of the National Academy of Sciences, 103, 61106115.
  • Nores, M. (2009) Are bird populations in tropical and subtropical forests of South America affected by climate change? Climatic Change, 97, 543551.
  • Odegaard, F. (2006) Host specificity, alpha- and beta-diversity of phytophagous beetles in two tropical forests in Panama. Biodiversity and Conservation, 15, 83105.
  • Owen-Ashley, N.T., Turner, M., Hahn, T.P. & Wingfield, J.C. (2006) Hormonal, behavioral, and thermoregulatory responses to bacterial lipopolysaccharide in captive and free-living white-crowned sparrows (Zonotrichia leucophrys gambelli). Hormones and Behavior, 49, 1529.
  • Piersma, T., Koolhaas, A., Dekinga, A. & Gwinner, E. (2000) Red blood cell and white blood cell counts in sandpipers (Philomachus pugnax, Calidris canutus): effects of captivity, season, nutritional status, and frequent bleedings. Canadian Journal of Zoology, 78, 13491355.
  • R Development Core Team (2009) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.
  • Raouf, S.A., Smith, L.C., Brown, M.B., Wingfield, J.C. & Brown, C.R. (2006) Glucocorticoid hormone levels increase with group size and parasite load in cliff swallows. Animal Behaviour, 71, 3948.
  • Rauscher, S.A., Giorgi, F., Diffenbaugh, N.S. & Seth, A. (2008) Extension and intensification of the meso-American mid-summer drought in the twenty-first century. Climate Dynamics. 31, 551571.
  • Renison, D., Boersma, P.D. & Martella, M.B. (2002) Winning and losing: causes for variability in outcome of fights in male Magellanic penguins (Spheniscus magellanicus). Behavioral Ecology, 13, 462466.
  • Ridgely, R.S. & Gwynne, J.A. Jr (1989) A Guide to the Birds of Panama, 2nd edn. Princeton University Press, Princeton, NJ, USA.
  • Robinson, T.R. (2000) Factors affecting natal dispersal by Song Wrens (Cyphorhinus phaeocephalus): Ecological constraints and demography. PhD thesis, University of Illinois, Urbana, IL, USA.
  • Robinson, T.R., Robinson, W.D. & Edwards, E.C. (2000) Breeding ecology and nest-site selection of song wrens in central Panama. The Auk, 117, 345354.
  • Robinson, S.K. & Terborgh, J. (1995) Interspecific aggression and habitat selection by Amazonian birds. Journal of Animal Ecology, 64, 111.
  • Robinson, W.D., Angehr, G.R., Robinson, T.R., Petit, L.J., Petit, D.R. & Brawn, J.D. (2004) Distribution of bird diversity in a vulnerable neotropical landscape. Conservation Biology, 18, 510518.
  • Rogovin, K.A., Randall, J.A., Kolosova, I.E. & Moshkin, M.P. (2008) Long-term dynamics of fecal corticosteroids in male great gerbils (Rhombomys opimus Licht.): effects of environment and social demography. Physiological and Biochemical Zoology, 81, 612626.
  • Romero, L.M. & Remage-Healey, L. (2000) Daily and seasonal variation in response to stress in captive starlings (Sturnus vulgaris): corticosterone. General and Comparative Endocrinology, 119, 5259.
  • Rompré, G., Robinson, W.D. & Desrochers, A. (2008) Causes of habitat loss in a Neotropical landscape: the Panama Canal corridor. Landscape and Urban Planning, 87, 129139.
  • Rompré, G., Robinson, W.D., Desrochers, A. & Angehr, G. (2007) Environmental correlates of avian diversity in lowland Panama rain forests. Journal of Biogeography, 34, 802815.
  • Root, T. (1988) Environmental factors associated with avian distributional boundaries. Journal of Biogeography, 15, 489505.
  • Sapolsky, R.M. (1987) Stress, social status, and reproductive physiology in free-living baboons. Psychobiology of Reproductive Behavior: An Evolutionary Approach (ed. D.Crews), pp. 291322. Prentice-Hill Inc, Englewood Cliffs, NJ, USA.
  • Sapolsky, R.M., Romero, L.M. & Munck, A.U. (2000) How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrine Reviews, 21, 5589.
  • SAS Institute (2005) JMP. SAS Institute Inc., Cary, NC, USA.
  • Sekercioglu, C.H., Ehrlich, P.R., Daily, G.C., Aygen, D., Goehring, D. & Sandi, R.F. (2002) Disappearance of insectivorous birds from tropical forest fragments. Proceedings of the National Academy of Sciences, 99, 263267.
  • Sekercioglu, C.H., Schneider, S.H., Fay, J.P. & Loarie, S.R. (2008) Climate change, elevation range shifts, and bird extinctions. Conservation Biology, 22, 140150.
  • Sinervo, B., Mendez-de-la-Cruz, F., Miles, D.B., Heulin, B., Bastiaans, E., Villagran-Santa Cruz, M., Lara-Resendiz, R., Martinez-Mendez, N., Calderon-Espinosa, M.L., Meza-Lazaro, R.N., Gadsden, H., Avila, L.J., Morando, M., De la Riva, I.J., Sepulveda, P.V., Rocha, C.F.D., Ibarguengoytia, N., Puntriano, C.A., Massot, M., Lepetz, V., Oksanen, T.A., Chapple, D.G., Bauer, A.M., Branch, W.R., Clobert, J. & Sites, J.W. Jr (2010) Erosion of lizard diversity by climate change and altered thermal niches. Science, 328, 894899.
  • Skutch, A.F. (1940) Social and sleeping habits of Central American wrens. The Auk, 57, 293312.
  • Smith, K.G. (1977) Distribution of summer birds along a forest moisture gradient in an Ozark watershed. Ecology, 58, 810819.
  • Stiles, F.G., Skutch, A.F. & Gardner, D. (1989) A Guide to the Birds of Costa Rica. Cornell University Press, Ithaca, NY, USA.
  • Stouffer, P.C. & Bierregaard, R.O. Jr (1995) Use of Amazonian forest fragments by understory insectivorous birds. Ecology, 76, 24292445.
  • Stratford, J.A. & Stouffer, P.C. (1999) Local extinctions of terrestrial insectivorous birds in a fragmented landscape near Manaus, Brazil. Conservation Biology, 13, 14161423.
  • Strong, A.M. & Sherry, T.W. (2000) Habitat-specific effects of food abundance on the condition of overbirds wintering in Jamaica. Journal of Animal Ecology, 69, 883895.
  • Suorsa, P., Huhta, E., Nikula, A., Nikinmaa, M., Jantti, A., Helle, H. & Hakkarainen, H. (2003) Forest management is associated with physiological stress in an old-growth forest passerine. Proceedings of the Royal Society B, 270, 963969.
  • Svensson, E. & Merilä, J. (1996) Molt and migratory condition in blue tits: a serological study. The Condor, 98, 825831.
  • Tewksbury, J.J., Huey, R.B. & Deutsch, C.A. (2008) Putting the heat on tropical animals. Science, 320, 12961297.
  • Thiollay, J.M. (1992) Influence of selective logging on bird species diversity in a Guianan rain forest. Conservation Biology, 6, 4762.
  • Walther, G.-R., Post, E., Convey, P., Menzel, A., Parmesan, C., Beebee, T.J.C., Fromentin, J.-M., Hoegh-Guldberg, O. & Bairlein, F. (2002) Ecological responses to recent climate change. Nature, 416, 389395.
  • Wikelski, M., Hau, M., Robinson, W.D. & Wingfield, J.C. (2003) Reproductive seasonality in seven neo-tropical passerine species. The Condor, 105, 683695.
  • Williams, S.E. & Middleton, J. (2008) Climatic seasonality, resource bottlenecks, and abundance of rainforest birds: implications for global climate change. Diversity and Distributions, 14, 6977.
  • Willis, E.O. (1976) Seasonal changes in the invertebrate litter fauna on Barro Colorado Island, Panama. Revista Brasileira de Biologia, 36, 643657.
  • Wingfield, J.C. & Ramenofsky, M. (1999) Hormones and the behavioral ecology of stress. Stress Physiology in Animals (ed. P.H.M.Balm), pp. 151. Sheffield Academic Press, Sheffield, UK.
  • Wingfield, J.C., Smith, J.P. & Farner, D.S. (1982) Endocrine responses of white-crowned sparrows to environmental stress. Condor, 84, 399409.
  • Wingfield, J.C., Vleck, C.M. & Moore, M.C. (1992) Seasonal changes of the adrenocortical response to stress in birds of the Sonoran desert. The Journal of Experimental Zoology, 264, 419428.
  • Wingfield, J.C., Maney, D.L., Breuner, C.W., Jacobs, J.D., Lynn, S., Ramenofsky, M. & Richardson, R.D. (1998) Ecological bases of hormone-behavior interactions: the “emergency life history stage”. American Zoologist, 38, 191206.
  • Woodworth, B.L., Atkinson, C.T., LaPointe, D.A., Hart, P.J., Spiegel, C.S., Tweed, E.J., Henneman, C., LeBrun, J., Denette, T., DeMots, R., Kozar, K.L., Triglia, D., Lease, D., Gregor, A., Smith, T. & Duffy, D. (2005) Host population persistence in the face of introduced vector-borne diseases: Hawaii amakihi and avian malaria. Proceedings of the National Academy of Sciences of the United States of America, 102, 15311536.

Supporting Information

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

Fig. S1. Best-fit path models showing relationships among estimated annual rainfall at location of capture, body condition, hematocrit, and sex.

Fig. S2. Best fit path models showing relationships among estimated annual rainfall at location of capture, ln time of exposure to conspecific playback, body condition, and ln baseline corticosterone levels.

Fig. S3. Best-fit path models showing relationships among estimated annual rainfall at location of capture, ln time of exposure to conspecific playback, body condition, sex, and maximum corticosterone levels.

Data S1. Methods.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
JANE_1791_sm_FigS1-3.ppt233KSupporting info item
JANE_1791_sm_DataS1.doc52KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.