Nest microclimate predicts bill growth in the Adelaide rosella (Aves: Psittaculidae)

Nest microclimate predicts bill growth in the Adelaide rosella (Aves: Psittaculidae) Abstract Diet and foraging have traditionally been considered key drivers of bill morphology. It is now known that bills play an important thermoregulatory role, and recent studies revealed that temperature is positively associated with the size of bills relative to body size or weight in adult birds, in accordance with Allen’s rule. Studies have attributed these patterns to local adaptation or an evolutionary response to climate change, but the contribution of ontogenetic plasticity remains unclear. We tested whether temperature experienced inside the nest predicted nestling growth in bill size and weight and in a parrot, the Adelaide rosella (Platycercus elegans adelaidae). We predicted that nest microclimate may affect bill ontogeny, leading to a positive association between relative bill size and temperatures during rearing. Growth in bill surface area was greater in nests that were warmer during the day and night, but temperature variability had no effect. Higher day and night-time mean temperatures, and less variable night-time temperatures, were positively associated with nestling weight. Our findings indicate that nest microclimate influences nestling growth, including relative bill size, and that daytime heat dissipation may be a driver of bill ontogeny. Bill plasticity in response to temperature during rearing could be an important but little studied contributor to morphology, due to the role of the bill in thermoregulation. Allen’s rule, ontogeny, phenotypic plasticity, Platycercus elegans adelaidae, thermoregulation INTRODUCTION Climate is an important determinant of animal morphology. This may arise due to evolutionary responses to temperature and adaptation to climatic niches, but temperature also has important implications for ontogeny in a wide range of taxa, such as mammals (Sedgeley, 2001; Serrat, King & Lovejoy, 2008), reptiles (Brewster, Sikes & Gifford, 2013; Pike, 2014) and birds (Burness et al., 2013; Şekercioğlu, Primack & Wormworth, 2012). Nest-site microclimate has been found to affect breeding success and ontogeny in many bird (such as passeriformes, piciformes and apodiformes) and bat species, particularly those breeding in nest boxes and hollows (Ardia, Perez & Clotfelter, 2010; Dawson, Lawrie & O’Brien, 2005; Larson et al., 2015). In birds, suboptimal temperature during incubation and during offspring development can result in increased incubation periods and reduced embryonic growth, hatchling weight, nestling growth, body condition and immunity, which can have long-term effects on fitness (Ardia, Perez & Clotfelter, 2006; Ardia et al., 2010; Burton, 2007; Cunningham et al., 2013; Larson et al., 2015; Murphy, 1985; Salaberria et al., 2014). Recently, it has also been shown that zebra finch (Taeniopygia guttata) parents communicate information on ambient temperature to embryos using acoustic signals, and that these calls alone alter the reproductive success and thermal preferences of these offspring as adults (Mariette & Buchanan, 2016). Nest microclimates can vary due to a large range of factors, such as orientation, construction materials, colour and nest size (Ardia et al., 2006; Goldingay, 2015). The considerable role of the bill for thermoregulation in birds, reviewed by Tattersall, Arnaout & Symonds (2017), has been identified through the use of thermal imaging (e.g. Greenberg et al., 2012a; Hagan & Heath, 1980; Tattersall, Andrade & Abe, 2009; van de Ven et al., 2016). Bills are uninsulated, and birds are able to dissipate large amounts of heat by controlling blood flow to their highly vascularized bills, whilst minimizing water loss (Hagan & Heath, 1980; Tattersall et al., 2009). Despite this control, bill surface area is thought to be subject to a trade-off between the need for heat dissipation in warm conditions and heat retention in cold conditions (Friedman et al., 2017). Recent studies have shown that bill surface area relative to a measure of body size/weight (relative bill surface area, RBSA) shows interspecific variation in relation to geographical distribution, and within-species variation in relation to habitat (Greenberg et al., 2012a; Symonds & Tattersall, 2010), seasonal temperatures (Danner & Greenberg, 2015; Friedman et al., 2017; Greenberg et al., 2012b) and long-term climate change (Campbell-Tennant et al., 2015). As a result of these patterns and the bill’s thermoregulatory function, spatial and temporal variation in RBSA is therefore thought to follow Allen’s rule (Allen, 1877). Allen’s rule predicts selection for larger appendage size (relative to body size) in warmer environments, such as lower latitudes/elevations, to achieve more efficient thermoregulation (Allen, 1877; Symonds & Tattersall, 2010; Tattersall et al., 2017). Most studies of relationships between bill size and temperature have done so in adult, fully developed birds, and the patterns are often attributed to local adaptation or an evolutionary response to climate change (Campbell-Tennant, Gardner, Kearney, Symonds & Ladle, 2015; Friedman et al., 2017; Symonds & Tattersall, 2010). However, it has also been proposed that developmental plasticity could contribute significantly to such patterns (Symonds & Tattersall, 2010; Tattersall et al., 2017), as it does in the limbs of mammals (Serrat et al., 2008). A recent study of captive precocial Japanese quail (Coturnix japonica) demonstrated experimentally that rearing temperature (at constant 15 °C or 30 °C) influenced bill growth in line with Allen’s rule, and resulted in irreversible effects on the thermoregulatory physiology of the bills which persisted even when birds were subsequently housed at a common temperature (Burness et al., 2013). If these patterns are general, then such bill size plasticity could be extremely important for nestlings as they may have less control of heat exchange through their bill than adults (Tattersall et al., 2009), coupled with little control over their thermal environments or dietary access to water compared to adults. This may be particularly true for obligate cavity nesters, such as parrots (Psittaciformes), as the often limited and specialized nature of their nest cavities may constrain their control of nest microclimates. To our knowledge, no study so far has investigated the effect of nest microclimate on bill ontogeny in altricial nestlings, or under natural conditions. Moreover, studies that have investigated the effects of temperature on reproductive success or ontogeny have often either done so in controlled laboratory conditions where temperature is kept constant, or have focused primarily on the effects of mean, maximum or minimum temperatures in isolation (e.g. Dawson et al., 2005; Salaberria et al., 2014; Sidhu et al., 2012). The importance of temperature variability is thus often not considered, although it may be expected to alter ontogeny or adaptation in relation to temperature, and some studies have highlighted the importance of temperature variability in climate change projections and for wildlife reproduction (Larson et al., 2015; Vasseur et al., 2014). For example, our previous study found that less extreme low temperatures resulted in heavier crimson rosella (Platycercus elegans elegans) nestlings, while greater temperature variability tended to reduce fledging success (Larson et al., 2015). In this study, we tested whether nest microclimate was associated with nestling growth and RBSA in a wild population of Adelaide rosellas (Platycercus elegans adelaidae) inhabiting temperate woodlands in south-eastern Australia. We recorded internal nest box temperature hourly, and analysed the thermal conditions encountered inside each nest during nestling growth in terms of both mean temperature and temperature variability (standard deviation). To assess diurnal changes in thermoregulatory demands, we also analysed daytime and night-time temperatures separately. This allowed us to distinguish effects arising from the higher, predominately increasing temperatures encountered during the day, when a need to dissipate heat is more likely, from those of the lower, predominately decreasing temperatures during the night, when a need to retain heat is more likely and food/water is less available. We used these data to test whether nest microclimate was related to (1) growth in nestling RBSA, or (2) nestling growth overall (body weight). We predicted that greater exposure to high temperatures inside the nest (higher or greater variability in temperature), especially during the daytime, would be associated with greater RBSA growth, due to the improved heat dissipation possible with larger RBSA. We also predicted that warmer nests, particularly at night when the lowest temperatures are experienced, would be associated with increased growth in body weight due to reduced energetic costs of thermoregulation. MATERIAL AND METHODS Study species and sites We studied an abundant hollow-breeding parrot, the Adelaide rosella. The general breeding biology, behaviour and sensory ecology of P. elegans has been previously documented (Krebs, 1998, Krebs, 1999, Krebs, 2001; Krebs & Magrath, 2000; Ribot et al., 2012, 2013; Mihailova et al., 2014, 2018; Larson et al., 2015; Knott et al., 2017). Platycercus elegans is a medium-sized (c. 35 cm long) parrot which occupies temperate, mesic habitats in south-eastern Australia, where it is an obligate cavity-nester (Krebs, 1998; Krebs & Magrath, 2000). The breeding season spans September to January, with most pairs laying 3–8 eggs (Krebs, 1998; Larson et al., 2015); incubation is 16–28 days (Krebs, 1998), with fledging occurring at 28–40 days post-hatching (Krebs, 1998; Larson et al., 2015). Adult P. e. adelaidae weigh 127.0 g ± 13.5 SD (females) to 130.4 g ± 8.0 SD (males), and tarsi lengths are 19.9 mm ± 1.0 SD (females) to 20.6 mm ± 0.7 SD (males) (our unpubl. data; no bill size data available). The time taken for nestlings to attain full size in this species varies depending on the trait being considered, and bills continue to grow linearly throughout the nestling period (Fig. 1). Figure 1. View largeDownload slide A, Nestling bill surface area (mm2) increases with nestling age (days) in Platycercus elegans adelaidae (N = 482 measuring events of 133 nestlings in 27 broods); growth in bill surface area was linear throughout the nestling stage. B, Nestling body weight (g) increases with nestling age (days) in P. e. adelaidae (N = 484 measuring events of 133 nestlings in 27 broods); an age of 22 days (vertical line) was considered the end of the linear growth phase for nestling weight. C, Nestling tarsus length (mm) increases with nestling age (days) in P. e. adelaidae (N = 484 measuring events of 133 nestlings in 27 broods); an age of 14 days (vertical line) was considered the end of growth for nestling tarsus length. Males are indicated by crosses and solid lines, females by circles and dashed lines; fit lines were derived from locally weighted scatterplot smoothing (LOESS). Figure 1. View largeDownload slide A, Nestling bill surface area (mm2) increases with nestling age (days) in Platycercus elegans adelaidae (N = 482 measuring events of 133 nestlings in 27 broods); growth in bill surface area was linear throughout the nestling stage. B, Nestling body weight (g) increases with nestling age (days) in P. e. adelaidae (N = 484 measuring events of 133 nestlings in 27 broods); an age of 22 days (vertical line) was considered the end of the linear growth phase for nestling weight. C, Nestling tarsus length (mm) increases with nestling age (days) in P. e. adelaidae (N = 484 measuring events of 133 nestlings in 27 broods); an age of 14 days (vertical line) was considered the end of growth for nestling tarsus length. Males are indicated by crosses and solid lines, females by circles and dashed lines; fit lines were derived from locally weighted scatterplot smoothing (LOESS). In this study, P. e. adelaidae nests were studied in dry sclerophyll eucalypt forests at Carey Gully, South Australia (34.97°S, 138.77°E). Sixty-one nest boxes were observed from September to December 2013. Of these nest boxes, 34 had been erected in 2004 as part of previous research projects (Joseph et al., 2008; Ribot et al., 2009, 2011; Berg & Bennett, 2010; Knott et al., 2013; Eastwood et al., 2014, 2015, 2017), while 27 were new boxes erected in July 2013 for the purpose of this study. Nest boxes were spaced at least 50 m apart and standardized for height (4-5 m above ground level), aspect (all faced south-east) and tree type (Eucalyptus spp., almost exclusively E. obliqua), as described by Larson et al. (2015). Nest boxes were constructed of 19-mm treated pine and were 24 cm wide × 28 cm deep × 42 cm high, with an entrance hole 7.5 cm in diameter and a sliding side door to allow observers access to nests. A c. 10-cm layer of Eucalyptus spp. wood chips had been placed in each nest box when erected as per Larson et al. (2015). Nestling measurements Nest boxes were checked every 2–3 days from 4 September 2013 until clutches were complete, and then weekly until 20 December 2013. Platycercus elegans lay eggs typically 2 days apart and have a high level of hatching asynchrony because incubation begins prior to clutch completion (Krebs, 1998, 1999). Nest box checks were more frequent (every 1–2 days) around the predicted hatching date to accurately determine hatching date. During this study, 283 eggs were laid, of which 149 hatched; reasons for failure to hatch included abandonment and predation/damage by possums. Thirty-three nestlings died (including those predated and 16 which died before any measurements were taken), leaving 116 to fledge successfully. Nestlings (N = 133 in 27 broods) were weighed using a spring scale (to the nearest 0.25 g; Pesola AG, Schindellegi, Switzerland) approximately every 7 days (mean 3.9 ± 1.1 SD visits per brood). At each weighing event (with the exception of two occasions), tarsus length, and bill length (BL), width (BW) and depth (BD) were also measured, by the same person (E.R.L.) each time, with callipers, to the nearest 0.1 mm. This allowed us to calculate bill surface area using eqn. 1, following previous studies on P. elegans and other species (Greenberg et al., 2012a, b; Greenberg & Danner, 2012; Campbell-Tennant et al., 2015). A limitation of this method is that it does not incorporate curvature of the bill; however, linear bill measurements show a high correlation (r = 0.98) with bill area in Australian parrots (Symonds & Tattersall, 2010). Nestlings were marked for individual identification, initially by trimming nails and down in specific combinations until approximately 10 days of age, when they were ringed with an individually numbered metal band (Australian Bird and Bat Banding Scheme). A small blood sample (< 100 µL) was taken from nestlings and used to determine sex, as described by Eastwood et al. (2015).  Bill surface area=(BD+BW4)×BL  Nest box temperature Seventy HOBO pendant temperature data loggers (Onset Computer Corporation, Bourne, MA, USA) were deployed in nest boxes on 24 June 2013. Loggers were placed in the top left-hand (i.e. upper north-west, hence out of direct sun exposure) corner of the nest boxes, and set to record temperature, date and time every hour following Larson et al. (2015). Logger accuracy drift over time was previously found to be negligible (Larson et al., 2015). Temperature variables To test whether nest temperature predicted nestling growth, we chose two temperature variables to represent the distribution of temperatures within nest boxes: mean temperature (Tmean), and variability in temperature (Tvariability; standard deviation). We used standard deviation as our measure of variability, as we were interested in the effects of absolute variability in temperature (i.e. not relative to mean temperature). We calculated Tmean and Tvariability separately for day and night periods; separating these periods for analysis allowed us to distinguish effects arising from the higher, mainly increasing temperatures encountered during the day from those of the lower, mainly decreasing temperatures during the night when a need for retaining heat is more likely. Day and night periods were defined by the mean sunrise and sunset times for the approximately 1-week periods between the weighing/measurement event at each nest box, and were therefore different for each brood as the timing of broods was not synchronous across the season. For every nest box, these temperature variables (daytime Tmean and Tvariability, and night-time Tmean and Tvariability) were calculated over each of the approximately 1-week periods between the weighing/measurement events, which allowed us to analyse the association between growth and temperature for each nestling since it was last weighed/measured. Thus, we had four different temperature variables to use as predictors for each time a brood was weighed and measured. To provide additional context for our study conditions we also provide, for daytime and night-time periods, the Tmean, mean minimum and mean maximum temperatures, averaged across all days in the nestling period for each brood. Statistical analysis We used linear mixed models (GLM; proc. MIXED with REML estimation) to test which temperature variables were associated with nestling growth in terms of RBSA and body weight. Bill growth was linear throughout the nestling period (Fig. 1A), but growth in weight slows after 22 days (Fig. 1B). Therefore, analyses presented below were restricted to nestlings from 1 to 22 days old (following Krebs, 1998; Larson et al., 2015), although the relationships between bill size and temperature were qualitatively unchanged when analysed over the entire nestling period. This resulted in 354 weighing events (mean = 2.02 ± 0.94 SD per nestling) from 133 nestlings in 27 broods for analysis. To analyse growth in bill size relative to overall size, we tested bill size controlled for body weight (Symonds & Tattersall, 2010). We used bill surface area as the dependent variable, with nestling weight and age (in days) as fixed covariates. We also repeated these models with weight excluded to test effects on growth of absolute bill surface area, and using tarsus length (modelled as a quadratic effect) as an alternative measure of nestling size instead of body weight. Tarsus was not our preferred measure of body size to calculate RBSA, as tarsus growth in this species plateaus very early in the nestling period (Fig. 1C), and because the uninsulated legs of birds are appendages which may themselves play a role in thermoregulation (Symonds & Tattersall, 2010; Burness et al., 2013). To analyse nestling growth in terms of body weight, we used nestling weight as the dependent variable, with age as a fixed covariate. To test the effects of temperature, Tmean and Tvariability were included as fixed covariates, and daytime and night-time temperatures were tested in separate models. Because brood size may affect growth due to sibling competition for food, brood size was included in all models We used current brood size (i.e. at the time of each weighing/measuring event) as we expected this would provide the most robust indication of competition, but also report the outcome of using initial brood size (i.e. brood size at the time of hatching) instead. For the analysis of nestling weight, brood size was included as a fixed factor because parameter estimates suggested that the relationship between brood size and nestling weight was non-linear. Parameter estimates suggested an approximately linear relationship between bill size and brood size, so brood size was included as a covariate for RBSA growth models; however, using brood size as a fixed factor had no qualitative effect on our results (not shown). Hatching date was included as a fixed covariate in all models to control for seasonal effects on growth. Because adult male P. elegans have larger bills than females (Krebs, 1999), we hypothesized that there may be sex differences in growth, particularly of the bill, and therefore included sex as a fixed factor in all models also. No interactions between sex and temperature variables were significant predictors for growth in either RBSA (P > 0.067) or weight (P > 0.084), or qualitatively changed the effects of other parameters, so these interactions were not included in the final models. To account for the repeated measures from nestlings and for nestlings being clustered within nest boxes, both nestling identity and nest box identity were included as random intercepts. Platycercus elegans generally have a single breeding attempt each season (Krebs, 1998; Larson et al., 2015), and in our study most breeding attempts overlapped temporally, so we assumed that each breeding pair was included in the analysis only once. All analyses were performed using SPSS Statistics for Windows version 24 (IBM Corp., 2016, Armonk, NY, USA). Conformity to assumptions including normality and homoscedasticity was confirmed following Quinn & Keough (2002). Variance inflation factors were < 1.4, indicating no problematic collinearity among predictors (Tmean, Tvariability, brood size, hatching date, age). Dates were converted into Julian date for all analyses. Degrees of freedom from mixed models were rounded to whole numbers. Means and estimates are presented with standard error unless otherwise noted, and the significance level was set a priori at α = 0.05. RESULTS The Tmean experienced inside nest boxes during our study, averaged across broods (N = 27), was 17.5 °C ± 1.7 SD (range 14.4–22.7 °C) during the day and 13.9 °C ± 1.5 SD (11.3–18.0 °C) during the night. Mean maximum daytime temperature was 21.3 °C ± 1.9 SD (18.6–26.8 °C) and mean minimum night-time temperature was 11.8 °C ± 1.6 SD (9.4–15.8 °C). Both daytime and night-time mean temperature had positive associations with nestling RBSA growth (Table 1, Supporting Information Fig. S1), with each 1 °C increase in mean temperature estimated to increase RBSA by 0.37 and 0.43 mm2 for daytime and night-time temperatures, respectively. Temperature variability was not associated with RBSA growth (Table 1, Fig. S1). When not controlling for nestling weight, daytime and night-time mean temperatures still had significant positive effects on bill surface area growth (daytime mean: estimate 0.487 °C ± 0.148 SE, P = 0.001; night-time mean: 0.605 °C ± 0.178 SE, P = 0.001), but temperature variability did not (daytime: −0.526 °C ± 0.498 SE, P = 0.292; night-time: −0.215 °C ± 0.482 SE, P = 0.656). The relationships between bill surface area and temperature were qualitatively the same when using tarsus as an alternative measure of nestling size (effect of Tmean: P < 0.041; results not shown). Table 1. Effects of internal nest box temperature, hatching date, brood size and sex on relative bill surface area growth (bill surface area controlling for age and weight) of Platycercus elegans adelaidae nestlings Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 246  0.371  0.126  8.705  0.003  0.123  0.618    Tvariability  1, 224  −0.619  0.415  2.223  0.137  −1.437  0.199    Hatching date  1, 33  −0.050  0.047  1.126  0.296  −0.146  0.046  Brood size  1, 61  0.979  0.296  10.912  0.002  0.386  1.571  Sex  1, 90  −3.469  0.610  32.378  < 0.001  −4.680  −2.258    Age  1, 323  1.123  0.127  77.802  < 0.001  0.872  1.373    Weight  1, 316  0.383  0.022  293.565  < 0.001  0.339  0.427  Night  Tmean  1, 250  0.431  0.150  8.260  0.004  0.136  0.727    Tvariability  1, 254  0.177  0.403  0.194  0.660  −0.616  0.971    Hatching date  1, 33  −0.056  0.047  1.385  0.248  −0.152  0.041  Brood size  1, 65  0.859  0.302  8.085  0.006  0.256  1.463  Sex  1, 90  −3.421  0.613  31.156  < 0.001  −4.638  −2.203    Age  1, 326  1.129  0.125  81.175  < 0.001  0.882  1.375    Weight  1, 320  0.383  0.022  298.924  < 0.001  0.339  0.426  Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 246  0.371  0.126  8.705  0.003  0.123  0.618    Tvariability  1, 224  −0.619  0.415  2.223  0.137  −1.437  0.199    Hatching date  1, 33  −0.050  0.047  1.126  0.296  −0.146  0.046  Brood size  1, 61  0.979  0.296  10.912  0.002  0.386  1.571  Sex  1, 90  −3.469  0.610  32.378  < 0.001  −4.680  −2.258    Age  1, 323  1.123  0.127  77.802  < 0.001  0.872  1.373    Weight  1, 316  0.383  0.022  293.565  < 0.001  0.339  0.427  Night  Tmean  1, 250  0.431  0.150  8.260  0.004  0.136  0.727    Tvariability  1, 254  0.177  0.403  0.194  0.660  −0.616  0.971    Hatching date  1, 33  −0.056  0.047  1.385  0.248  −0.152  0.041  Brood size  1, 65  0.859  0.302  8.085  0.006  0.256  1.463  Sex  1, 90  −3.421  0.613  31.156  < 0.001  −4.638  −2.203    Age  1, 326  1.129  0.125  81.175  < 0.001  0.882  1.375    Weight  1, 320  0.383  0.022  298.924  < 0.001  0.339  0.426  Significant terms are shown in bold. View Large Table 1. Effects of internal nest box temperature, hatching date, brood size and sex on relative bill surface area growth (bill surface area controlling for age and weight) of Platycercus elegans adelaidae nestlings Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 246  0.371  0.126  8.705  0.003  0.123  0.618    Tvariability  1, 224  −0.619  0.415  2.223  0.137  −1.437  0.199    Hatching date  1, 33  −0.050  0.047  1.126  0.296  −0.146  0.046  Brood size  1, 61  0.979  0.296  10.912  0.002  0.386  1.571  Sex  1, 90  −3.469  0.610  32.378  < 0.001  −4.680  −2.258    Age  1, 323  1.123  0.127  77.802  < 0.001  0.872  1.373    Weight  1, 316  0.383  0.022  293.565  < 0.001  0.339  0.427  Night  Tmean  1, 250  0.431  0.150  8.260  0.004  0.136  0.727    Tvariability  1, 254  0.177  0.403  0.194  0.660  −0.616  0.971    Hatching date  1, 33  −0.056  0.047  1.385  0.248  −0.152  0.041  Brood size  1, 65  0.859  0.302  8.085  0.006  0.256  1.463  Sex  1, 90  −3.421  0.613  31.156  < 0.001  −4.638  −2.203    Age  1, 326  1.129  0.125  81.175  < 0.001  0.882  1.375    Weight  1, 320  0.383  0.022  298.924  < 0.001  0.339  0.426  Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 246  0.371  0.126  8.705  0.003  0.123  0.618    Tvariability  1, 224  −0.619  0.415  2.223  0.137  −1.437  0.199    Hatching date  1, 33  −0.050  0.047  1.126  0.296  −0.146  0.046  Brood size  1, 61  0.979  0.296  10.912  0.002  0.386  1.571  Sex  1, 90  −3.469  0.610  32.378  < 0.001  −4.680  −2.258    Age  1, 323  1.123  0.127  77.802  < 0.001  0.872  1.373    Weight  1, 316  0.383  0.022  293.565  < 0.001  0.339  0.427  Night  Tmean  1, 250  0.431  0.150  8.260  0.004  0.136  0.727    Tvariability  1, 254  0.177  0.403  0.194  0.660  −0.616  0.971    Hatching date  1, 33  −0.056  0.047  1.385  0.248  −0.152  0.041  Brood size  1, 65  0.859  0.302  8.085  0.006  0.256  1.463  Sex  1, 90  −3.421  0.613  31.156  < 0.001  −4.638  −2.203    Age  1, 326  1.129  0.125  81.175  < 0.001  0.882  1.375    Weight  1, 320  0.383  0.022  298.924  < 0.001  0.339  0.426  Significant terms are shown in bold. View Large Higher mean daytime and night-time temperatures were both positively associated with nestling growth in terms of body weight (Table 2), with each 1 °C increase in daytime mean temperature estimated to increase nestling weight by 0.61 g and each 1 °C increase in night-time mean temperature estimated to increase mean nestling weight by 0.88 g (Table 2). Temperature variability at night was negatively associated with growth (Table 2), with each 1 °C increase in temperature standard deviation estimated to reduce mean nestling weight by 2.2 g. Table 2. Effects of internal nest box temperature, hatching date, brood size and sex on nestling growth (weight controlling for age) of Platycercus elegans adelaidae nestlings Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 236  0.608  0.299  4.134  0.043  0.019  1.120    Tvariability  1, 215  −1.068  1.000  1.139  0.287  −3.039  0.904    Hatching date  1, 28  −0.182  0.125  2.129  0.156  −0.437  0.074  Brood size  7, 92      3.008  0.007      Sex  1, 99  −2.756  1.730  2.539  0.114  −6.189  0.676    Age  1, 241  5.380  0.087  3853.167  < 0.001  5.209  5.550  Night  Tmean  1, 241  0.880  0.360  5.992  0.015  0.172  1.589    Tvariability  1, 236  −2.218  0.963  5.301  0.022  −4.116  −0.320    Hatching date  1, 27  −0.168  0.126  1.772  0.194  −0.426  0.091  Brood size  7, 90      3.101  0.006      Sex  1, 98  −3.052  1.725  3.131  0.080  −6.475  0.371    Age  1, 236  5.343  0.087  3807.018  < 0.001  5.173  5.514  Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 236  0.608  0.299  4.134  0.043  0.019  1.120    Tvariability  1, 215  −1.068  1.000  1.139  0.287  −3.039  0.904    Hatching date  1, 28  −0.182  0.125  2.129  0.156  −0.437  0.074  Brood size  7, 92      3.008  0.007      Sex  1, 99  −2.756  1.730  2.539  0.114  −6.189  0.676    Age  1, 241  5.380  0.087  3853.167  < 0.001  5.209  5.550  Night  Tmean  1, 241  0.880  0.360  5.992  0.015  0.172  1.589    Tvariability  1, 236  −2.218  0.963  5.301  0.022  −4.116  −0.320    Hatching date  1, 27  −0.168  0.126  1.772  0.194  −0.426  0.091  Brood size  7, 90      3.101  0.006      Sex  1, 98  −3.052  1.725  3.131  0.080  −6.475  0.371    Age  1, 236  5.343  0.087  3807.018  < 0.001  5.173  5.514  Significant predictors are shown in bold. Brood size was treated as a categorical predictor. View Large Table 2. Effects of internal nest box temperature, hatching date, brood size and sex on nestling growth (weight controlling for age) of Platycercus elegans adelaidae nestlings Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 236  0.608  0.299  4.134  0.043  0.019  1.120    Tvariability  1, 215  −1.068  1.000  1.139  0.287  −3.039  0.904    Hatching date  1, 28  −0.182  0.125  2.129  0.156  −0.437  0.074  Brood size  7, 92      3.008  0.007      Sex  1, 99  −2.756  1.730  2.539  0.114  −6.189  0.676    Age  1, 241  5.380  0.087  3853.167  < 0.001  5.209  5.550  Night  Tmean  1, 241  0.880  0.360  5.992  0.015  0.172  1.589    Tvariability  1, 236  −2.218  0.963  5.301  0.022  −4.116  −0.320    Hatching date  1, 27  −0.168  0.126  1.772  0.194  −0.426  0.091  Brood size  7, 90      3.101  0.006      Sex  1, 98  −3.052  1.725  3.131  0.080  −6.475  0.371    Age  1, 236  5.343  0.087  3807.018  < 0.001  5.173  5.514  Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 236  0.608  0.299  4.134  0.043  0.019  1.120    Tvariability  1, 215  −1.068  1.000  1.139  0.287  −3.039  0.904    Hatching date  1, 28  −0.182  0.125  2.129  0.156  −0.437  0.074  Brood size  7, 92      3.008  0.007      Sex  1, 99  −2.756  1.730  2.539  0.114  −6.189  0.676    Age  1, 241  5.380  0.087  3853.167  < 0.001  5.209  5.550  Night  Tmean  1, 241  0.880  0.360  5.992  0.015  0.172  1.589    Tvariability  1, 236  −2.218  0.963  5.301  0.022  −4.116  −0.320    Hatching date  1, 27  −0.168  0.126  1.772  0.194  −0.426  0.091  Brood size  7, 90      3.101  0.006      Sex  1, 98  −3.052  1.725  3.131  0.080  −6.475  0.371    Age  1, 236  5.343  0.087  3807.018  < 0.001  5.173  5.514  Significant predictors are shown in bold. Brood size was treated as a categorical predictor. View Large Nestling sex was a significant predictor of RBSA growth (Table 1) but not growth in weight (Table 2), with males having higher RBSA growth. No effects of hatching date on growth in nestling RBSA (Table 1) or weight (Table 2) were found. Brood size (at the time of weighing/measuring) significantly predicted nestling growth in terms of both RBSA (Table 1) and weight (Table 2); broods with two to six nestlings had lower growth than smaller and larger brood sizes. In contrast, initial brood size (at hatching) did not significantly affect growth in RBSA (P > 0.157) or weight (P > 0.073), nor did replacing current brood size with initial brood size qualitatively change the conclusions about any other variables (results not shown). As expected, because nestling bills and weights were measured during a period of rapid growth, age always had a significant effect on both RBSA (Table 1) and weight (Table 2). After controlling for age, RBSA was still positively predicted by weight (Table 1) or tarsus length (P < 0.045). DISCUSSION Bills are important thermoregulatory structures in both adult and developing birds, with bill size and vascularity affecting the radiative capacity of the bill (Burness et al., 2013; Tattersall et al., 2009, 2017). Recent studies of RBSA in adult birds have shown interspecific variation in relation to geographical distribution, and within-species variation in relation to habitat (Symonds & Tattersall, 2010; Greenberg et al., 2012a), seasonal temperatures (Greenberg et al., 2012b; Danner & Greenberg, 2015; Friedman et al., 2017) and long-term climate change (Campbell-Tennant et al., 2015). However, the contribution of temperature-dependent development to such patterns has been little reported (Burness et al., 2013). Our study revealed that, in nestling Adelaide rosellas, growth in bill size and RBSA was positively related to mean temperature inside the nest, and this was true for both daytime and night-time temperatures. Temperature variability was not related to RBSA, but nestling growth in terms of body weight was negatively related to night-time temperature variability. Our study was conducted in temperate conditions in south-eastern Australia, and the temperatures that nestlings were exposed to inside nest boxes during our study, particularly during the day, were within the range where radiant heat dissipation through the bill is possibly effective and related to bill size (Tattersall et al., 2009, 2017; Ryeland, Weston & Symonds, 2017; but see van de Ven et al., 2016), and also where effects of temperature on bill growth have been previously demonstrated in the laboratory (Burness et al., 2013). Temperatures were also generally below critical thresholds above which nestling growth or survival was impaired in other species (e.g. Cunningham et al., 2013). To our knowledge, our study is the first to investigate the effects of temperature during rearing on bill ontogeny in situ in free-living birds, and to compare the effects of day and night temperatures on bill size for adult or juvenile birds. This may be important, as historical data and climate change projections indicate night-time temperatures are increasing faster than daytime temperatures globally (Davy et al., 2017). Growth in relative bill size Larger RBSA is expected to permit better heat dissipation by increasing the radiative capacity of the bill, and may result from a developmental response to temperature (Burness et al., 2013; Tattersall et al., 2017; Tattersall, Chaves & Danner, 2018). Therefore, we predicted that greater exposure to high temperatures inside the nest, especially during the day, would be associated with greater RBSA growth. Our results were consistent with this prediction, as we found that higher daytime and night-time mean temperature was positively associated with growth in RBSA (controlled for nestling weight), although we found no effects of temperature variability. This pattern corresponds to Allen’s rule, which states that in warmer environments animals possess larger appendages in order to eliminate excess heat (Allen, 1877). Our findings suggest that relative bill size may at least partially reflect an ontogenetic response to variation in rearing temperature resulting from nest microclimates. If developmental differences in bill size or thermoregulatory function persist into adulthood, as shown by Burness et al. (2013), then such developmental plasticity may be an underappreciated contributor to widely observed patterns of relative bill size, which have often been attributed to local adaptation or an evolutionary response to climate change (Symonds & Tattersall, 2010; Campbell-Tennant et al., 2015; Friedman et al., 2017; Tattersall et al., 2017, 2018). Within- and between-species comparisons have observed strong positive relationships between higher temperatures and larger bill size in several taxa, including parrots, galliforms, penguins and gulls (Symonds & Tattersall, 2010; Campbell-Tennant et al., 2015), Meliphagoidea (Gardner et al., 2016; Friedman et al., 2017), salt-marsh sparrows (Ammodramus caudacutus) (Greenberg et al., 2012b), song sparrows (Melospiza melodia) (Danner & Greenberg, 2015), hornbills (Tockus leucomelas) (van de Ven et al., 2016) and starlings (Sturnus vulgaris) (Cardilini et al., 2016). However, all of these studies were based on observations of adult birds, whereas we studied nestlings during development. Our findings complement a previous laboratory experiment on Japanese quail, a precocial species not confined to a nest site (Burness et al., 2013). That study reported that juvenile quail which had been raised in captivity at 15 °C developed shorter bills than those raised at 30 °C. One strength of that study is that it also followed birds into maturity and measured bill surface temperatures, revealing that although previously cold-reared birds exhibited catch-up growth when subsequently housed at a common temperature, the difference in thermoregulatory function of the bills in cold- and warm-reared birds persisted, possibly due to differences in vascularity of the bill. Thus, the results suggested that differences in bill development related to rearing temperature may result in persistent, possibly lifelong, physiological changes (Burness et al., 2013). As we could not follow birds after fledging and it appears the bills were still growing (linearly) at fledging (Fig. 1A), it remains unknown whether bill differences (in size or thermoregulatory function) which develop during the nestling stage are maintained after fledging under natural conditions, when birds are free to choose microclimates and diets in maturity. To resolve this, future studies should follow juveniles under natural conditions. If differences are maintained, then bill size plasticity in nestlings could have implications for the resilience of birds to climate change (Campbell-Tennant et al., 2015), and a mismatch between temperatures experienced during development and adulthood may result in heat loss or heat deficits through the bill in adult birds. Moreover, diet, foraging strategies and song characteristics can all be affected by bill size (Christensen, Kleindorfer & Robertson, 2006). Hence, it is possible that ontogenetic variation in bills, if maintained in adulthood, could have considerable fitness effects throughout life. Our study suggests that natural differences in microclimate between nest sites, which are much smaller than the experimental temperature differences used by Burness et al. (2013), may also impact on bill size development. Unlike Burness et al. (2013), our study was correlational, so we are unable to fully exclude other factors which may be correlated with nest microclimates, such as brood size (although this was controlled for statistically in our models) or behaviour. Humidity was not measured in the present study, but may also be of considerable importance to avian thermoregulation, and thus bill development, because it mediates evaporative water loss (e.g. Gerson et al., 2014). Bill size, which contributes to non-evaporative heat dissipation, may become more important at high humidity (van de Ven et al., 2016). Experimental manipulations of nest temperature (e.g. using insulation or heating pads), possibly combined with brood swaps to control for genetic influences, would be a useful next step to overcome such limitations. A further limitation of our study, and most studies on bill size, is that the calculation of bill surface area is based on linear measurements and does not account for variation in the curvature of the bill. More accurate methods for measuring surface area (e.g. from photos) may yield more robust results. We found that larger brood size at the time of measuring events was associated with greater RBSA (and weight) growth, which we speculate may occur due to the increased heat generated by larger broods and limitations on evaporative cooling imposed by more crowded nest cavities. Interestingly, no effects of brood size at hatching were found, reinforcing the view that variation in brood size was directly affecting short-term growth rather than reflecting indirect effects of nest or parental quality or investment. Nestling sex was also a strong predictor of RBSA, which concurs with the larger head–bill size observed in adult males (41.3 mm ± 0.9 SD, N = 35) compared to females (38.5 mm ± 0.9 SD, N = 47) which we have previously observed in this study population (t80 = 14.114, P < 0.001; our unpubl. data). A recent comparative study of shorebirds has shown that behavioural thermoregulation is mediated by bill size at the species level, with placement of the bill within the plumage for insulation while resting being more common in species with larger bills relative to their body weight (Ryeland et al., 2017). Therefore, it would also be interesting for future studies to investigate the possible behavioural implications of temperature-related ontogenetic variation in bills. Similarly, future research could also assess whether behavioural responses to temperature by parents, such as time spent in foraging or parental care activities (Cunningham et al., 2013; Du Plessis et al., 2012), are associated with the bill development of nestlings. Such effects on development may even begin in the incubation stage: in a recent study of zebra finches, Mariette & Buchanan (2016) found that higher nest temperatures were associated with higher weight throughout the nestling stage, but this pattern was reversed when embryos were experimentally exposed to conspecific incubation calls prior to hatching. Growth in body weight We found that growth in body weight (weight controlled for age) increased with higher mean temperatures in the nest during growth, which is consistent with previous research in other species (e.g. Siikamaki, 1996; Eeva et al., 2002; Dawson et al., 2005; Pipoly et al., 2013; Chausson et al., 2014; Mariette & Buchanan, 2016; Martin et al., 2017). However, the opposite pattern has been found in some studies. For example, in their experimental study of Japanese quail, Burness et al. (2013) found that birds raised at 15 °C gained more body weight than those raised at 30 °C, and retained a small (approximately 1%) but significant lead when housed subsequently at a common temperature. In that experiment, the birds had ad libitum access to food, which may allow the cold-reared birds to compensate with a larger food intake. Another explanation for relationships between growth and temperature is that temperature-mediated changes in parental foraging activity or efficiency could lead to reduced provisioning to offspring and resultant decrease in nestling growth rates (Ricklefs & Hainsworth, 1968; Murphy, 1985; Du Plessis et al., 2012; Cunningham et al., 2013). These patterns may therefore be expected to differ between altricial and precocial species, and to depend on the correlation between nest microclimates and the ambient temperatures experienced by parents while foraging outside the nest. Relationships between temperature and growth are also dependent on the magnitude of temperatures experienced, and a positive relationship between temperature and growth arising from lower energy demands for thermoregulation may reverse once ambient temperature surpasses a critical threshold (Cunningham et al., 2013). In our study, estimated effects of mean temperature were slightly larger for the night, and only night-time variability in temperature affected growth, suggesting that temperatures at night are particularly influential on nestling weight. This may reflect increased energetic costs of thermoregulation for nestlings which experience low temperatures at night (Dawson et al., 2005; Cunningham et al., 2013). Our current results concur with the findings of our previous study on P. e. elegans, which found that minimum temperature experienced at night during the nestling period, followed by mean temperature, were the most important predictors of nestling growth (Larson et al., 2015). A relationship between growth and temperature variability was also identified by Larson (2015), although that study is not comparable in this context as it did not separate the daytime and night-time periods, and thus greater temperature standard deviation could arise either from more extreme daytime temperatures or lower night-time temperatures. Many studies that have investigated the relationship between nest-site temperature and nestling growth have been undertaken in the northern hemisphere at higher latitudes than our study, but our results indicate that even under mild temperate conditions (mean day and night temperatures in nest boxes in our study were 17.5 and 13.9 °C, respectively) nestling growth may be positively related to higher mean temperatures. However, such benefits may be tempered by an increasing frequency or magnitude of temperature extremes under a warming climate (Cunningham et al., 2013). CONCLUSIONS We found that nest microclimate predicts nestling growth in terms of both body weight, and bill surface area relative to body weight, in an Australian parrot. Higher mean temperature was positively related to growth in weight and RBSA, and this was true for both daytime and night-time temperatures. In contrast, greater night-time temperature variability negatively affected weight of nestlings. Recent studies have shown that bills play an important role in thermoregulation for birds (Tattersall et al., 2009; Burness et al., 2013; van de Ven et al., 2016; Ryeland et al., 2017). Differences in RBSA of adult birds have been reported within and between species in relation to temporal and geographical variation in temperature, in line with Allen’s rule (Tattersall et al., 2017). Our results support the hypothesis that plasticity in bill development has the potential to contribute to such patterns due to variation in nest microclimate. Future work could build on these results through experimental manipulations of nest microclimate, and by following into adulthood the thermoregulatory capacity of bills and related fitness consequences. Further study of the developmental plasticity of bills, the potential for temperature during rearing and nest microclimate to influence bill morphology, and the implications for thermoregulation in adult birds, will facilitate a better understanding of avian responses to increasing temperatures and climate variability due to climate change. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s web-site. Figure S1. Relationships between temperature variables and predicted bill surface area (mm2) of nestling P. elegans adelaidae derived from the models shown in Table 1. Symbols indicate measurement events at each brood (solid circles: first measurement; open circles: second measurement; crosses: third measurement; diamonds: fourth measurement; triangle: fifth measurement). Overall fit lines are for illustrative purposes, and do not account for repeated measures of nestlings. ACKNOWLEDGEMENTS We are grateful to the landholders who allowed us to work on their properties or assisted with the research, particularly Brian and Maggie Colton, Robert and Jennie Henzell, Andrew Stanley and Chic Stirling Phillips, and George and Alice Avramidis. We also thank Martin and Virginia Gare, Raoul Ribot, Kate Buchanan, and Bill Buttemer for assistance, and the many field volunteers who helped collect data used in this study. Juan Amat, Ross Goldingay, Peter Korsten, anonymous reviewers and the editor provided valuable comments on earlier drafts of the manuscript. This work was funded by a Deakin University postgraduate scholarship to E.R.L., an Australian Research Council Linkage grant (LP140100691) to A.T.D.B. and M.L.B., Birdlife Australia, Holsworth Research Endowment and Deakin University. A.T.D.B. and M.L.B. were also supported by Australian Research Council Discovery grant DP180103494. We are grateful to the Australian Bird and Bat Banding Scheme for the supply of bands and support. All applicable institutional and/or national guidelines for the care and use of animals were followed. The research reported in this study conformed to the laws of Australia and the state of South Australia (Deakin University Animal Ethics Committee approval A33-2008 and State research permit 10004759-3). The authors declare no conflicts of interest. REFERENCES Allen JA. 1877. The influence of physical conditions in the genesis of species. Radical Review  1: 108– 140. Ardia DR, Perez JH, Clotfelter ED. 2006. Nest box orientation affects internal temperature and nest site selection by Tree Swallows. Journal of Field Ornithology  77: 339– 344. Google Scholar CrossRef Search ADS   Ardia DR, Perez JH, Clotfelter ED. 2010. Experimental cooling during incubation leads to reduced innate immunity and body condition in nestling tree swallows. Proceedings of the Royal Society of London, Series B: Biological Sciences  277: 1881– 1888. Google Scholar CrossRef Search ADS   Berg ML, Bennett ATD. 2010. The evolution of plumage colouration in parrots: a review. Emu  110: 10– 20. Google Scholar CrossRef Search ADS   Brewster CL, Sikes RS, Gifford ME. 2013. Quantifying the cost of thermoregulation: thermal and energetic constraints on growth rates in hatchling lizards. Functional Ecology  27: 490– 497. Google Scholar CrossRef Search ADS   Burness G, Huard JR, Malcolm E, Tattersall GJ. 2013. Post-hatch heat warms adult beaks: irreversible physiological plasticity in Japanese quail. Proceedings of the Royal Society of London, Series B: Biological Sciences  280: 20131436. Google Scholar CrossRef Search ADS   Burton NHK. 2007. Intraspecific latitudinal variation in nest orientation among ground-nesting passerines: a study using published data. Condor  109: 441– 446. Google Scholar CrossRef Search ADS   Campbell-Tennant DJE, Gardner JL, Kearney MR, Symonds MRE, Ladle R. 2015. Climate-related spatial and temporal variation in bill morphology over the past century in Australian parrots. Journal of Biogeography  42: 1163– 1175. Google Scholar CrossRef Search ADS   Cardilini AP, Buchanan KL, Sherman CD, Cassey P, Symonds MRE. 2016. Tests of ecogeographical relationships in a non-native species: what rules avian morphology? Oecologia  181: 783– 793. Google Scholar CrossRef Search ADS   Chausson A, Henry I, Almasi B, Roulin A. 2014. Barn Owl (Tyto alba) breeding biology in relation to breeding season climate. Journal of Ornithology  155: 273– 281. Google Scholar CrossRef Search ADS   Christensen R, Kleindorfer S, Robertson J. 2006. Song is a reliable signal of bill morphology in Darwin’s small tree finch Camarhynchus parvulus, and vocal performance predicts male pairing success. Journal of Avian Biology  37: 617– 624. Google Scholar CrossRef Search ADS   Cunningham SJ, Martin RO, Hojem CL, Hockey PA. 2013. Temperatures in excess of critical thresholds threaten nestling growth and survival in a rapidly-warming arid savanna: a study of common fiscals. PLoS ONE  8: e74613. Google Scholar CrossRef Search ADS   Danner RM, Greenberg R. 2015. A critical season approach to Allen’s rule: bill size declines with winter temperature in a cold temperate environment. Journal of Biogeography  42: 114– 120. Google Scholar CrossRef Search ADS   Davy R, Esau I, Chernokulsky A, Outten S, Zilitinkevich S. 2017. Diurnal asymmetry to the observed global warming. International Journal of Climatology  37: 79– 93. Google Scholar CrossRef Search ADS   Dawson RD, Lawrie CC, O’Brien EL. 2005. The importance of microclimate variation in determining size, growth and survival of avian offspring: experimental evidence from a cavity nesting passerine. Oecologia  144: 499– 507. Google Scholar CrossRef Search ADS   du Plessis KL, Martin RO, Hockey PAR, Cunningham SJ, Ridley AR. 2012. The costs of keeping cool in a warming world: implications of high temperatures for foraging, thermoregulation and body condition of an arid-zone bird. Global Change Biology  18: 3063– 3070. Google Scholar CrossRef Search ADS   Eastwood JR, Berg ML, Ribot RFH, Raidal SR, Buchanan KL, Walder KR, Bennett ATD. 2014. Phylogenetic analysis of beak and feather disease virus across a host ring-species complex. Proceedings of the National Academy of Sciences USA  111: 14153– 14158. Google Scholar CrossRef Search ADS   Eastwood JR, Berg ML, Spolding B, Buchanan KL, Bennett ATD, Walder K. 2015. Prevalence of beak and feather disease virus in wild Platycercus elegans: comparison of three tissue types using a probe-based real-time qPCR test. Australian Journal of Zoology  63: 1– 8. Google Scholar CrossRef Search ADS   Eastwood JR, Ribot RFH, Rollins LA, Buchanan KL, Walder K, Bennett ATD, Berg ML. 2017. Host heterozygosity and genotype rarity affect viral dynamics in an avian subspecies complex. Scientific Reports  7: 13310. Google Scholar CrossRef Search ADS   Eeva T, Lehikoinen E, Ronka M, Lummaa V, Currie D. 2002. Different responses to cold weather in two pied flycatcher populations. Ecography  25: 705– 713. Google Scholar CrossRef Search ADS   Friedman NR, Harmáčková L, Economo EP, Remeš V. 2017. Smaller beaks for colder winters: thermoregulation drives beak size evolution in Australasian songbirds. Evolution  71: 2120– 2129. Google Scholar CrossRef Search ADS   Gardner JL, Symonds MRE, Joseph L, Ikin K, Stein J, Kruuk LEB. 2016. Spatial variation in avian bill size is associated with humidity in summer among Australian passerines. Climate Change Responses  3: 11. Google Scholar CrossRef Search ADS   Gerson AR, Smith EK, Smit B, McKechnie AE, Wolf BO. 2014. The impact of humidity on evaporative cooling in small desert birds exposed to high air temperatures. Physiological and Biochemical Zoology  87: 782– 795. Google Scholar CrossRef Search ADS   Goldingay RL. 2015. Temperature variation in nest boxes in eastern Australia. Australian Mammalogy  37: 225– 233. Google Scholar CrossRef Search ADS   Greenberg R, Cadena V, Danner RM, Tattersall GJ, Tattersall G. 2012a. Heat loss may explain bill size differences between birds occupying different habitats. PLoS ONE  7: e40933. Google Scholar CrossRef Search ADS   Greenberg R, Danner R, Olsen B, Luther D. 2012b. High summer temperature explains bill size variation in salt marsh sparrows. Ecography  35: 146– 152. Google Scholar CrossRef Search ADS   Greenberg R, Danner RM. 2012. The influence of the California marine layer on bill size in a generalist songbird. Evolution  66: 3825– 3835. Google Scholar CrossRef Search ADS   Hagan AA, Heath JE. 1980. Regulation of heat loss in the duck by vasomotion in the bill. Journal of Thermal Biology  5: 95– 101. Google Scholar CrossRef Search ADS   Joseph L, Dolman G, Donnellan S, Saint KM, Berg ML, Bennett ATD. 2008. Where and when does a ring start and end? Testing the ring-species hypothesis in a species complex of Australian parrots. Proceedings of the Royal Society of London B: Biological Sciences  275: 2431– 2440. Google Scholar CrossRef Search ADS   Knott B, Berg ML, Ribot RFH, Endler JA, Bennett ATD. 2017. Intraspecific geographic variation in rod and cone visual pigment sensitivity of a parrot, Platycercus elegans. Scientific Reports  7: 41445. Google Scholar CrossRef Search ADS   Knott B, Davies WI, Carvalho LS, Berg ML, Buchanan KL, Bowmaker JK, Bennett ATD, Hunt DM. 2013. How parrots see their colours: novelty in the visual pigments of Platycercus elegans. Journal of Experimental Biology  216: 4454– 4461. Google Scholar CrossRef Search ADS   Krebs EA. 1998. Breeding biology of crimson rosellas (Platycercus elegans) on Black Mountain, Australian Capital Territory. Australian Journal of Zoology  46: 119– 136. Google Scholar CrossRef Search ADS   Krebs EA. 1999. Last but not least: nestling growth and survival in asynchronously hatching crimson rosellas. Journal of Animal Ecology  68: 266– 281. Google Scholar CrossRef Search ADS   Krebs EA. 2001. Begging and food distribution in crimson rosella (Platycercus elegans) broods: why don’t hungry chicks beg more? Behavioral Ecology and Sociobiology  50: 20– 30. Google Scholar CrossRef Search ADS   Krebs EA, Magrath RD. 2000. Food allocation in crimson rosella broods: parents differ in their responses to chick hunger. Animal Behaviour  59: 739– 751. Google Scholar CrossRef Search ADS   Larson ER, Eastwood JR, Buchanan KL, Bennett ATD, Berg ML. 2015. How does nest box temperature affect nestling growth rate and breeding success in a parrot? Emu Austral Ornithology  115: 247– 255. Google Scholar CrossRef Search ADS   Mariette MM, Buchanan KL. 2016. Prenatal acoustic communication programs offspring for high posthatching temperatures in a songbird. Science  353: 812– 814. Google Scholar CrossRef Search ADS   Martin TE, Boyce AJ, Fierro-Calderón K, Mitchell AE, Armstad CE, Mouton JC, Bin Soudi EE. 2017. Enclosed nests may provide greater thermal than nest predation benefits compared with open nests across latitudes. Functional Ecology  31: 1231– 1240. Google Scholar CrossRef Search ADS   Mihailova M, Berg ML, Buchanan KL, Bennett ATD. 2014. Odour-based discrimination of subspecies, species and sexes in an avian species complex, the crimson rosella. Animal Behaviour  95: 155– 164. Google Scholar CrossRef Search ADS   Mihailova M, Berg ML, Buchanan KL, Bennett ATD. 2018. Olfactory eavesdropping: The odor of feathers is detectable to mammalian predators and competitors. Ethology  124: 14– 24. Google Scholar CrossRef Search ADS   Murphy MT. 1985. Nestling eastern kingbird growth: effects of initial size and ambient temperature. Ecology  66: 162– 170. Google Scholar CrossRef Search ADS   Pike DA. 2014. Forecasting the viability of sea turtle eggs in a warming world. Global Change Biology  20: 7– 15. Google Scholar CrossRef Search ADS   Pipoly I, Bókony V, Seress G, Szabó K, Liker A. 2013. Effects of extreme weather on reproductive success in a temperate-breeding songbird. PLoS ONE  8: e80033. Google Scholar CrossRef Search ADS   Quinn GP, Keough MJ. 2002. Experimental design and data analysis for biologists . Cambridge: Cambridge University Press. Google Scholar CrossRef Search ADS   Ribot RFH, Berg ML, Buchanan KL, Bennett ATD. 2011. Fruitful use of bioacoustic alarm stimuli as a deterrent for Crimson Rosellas (Platycercus elegans). Emu  111: 360– 367. Google Scholar CrossRef Search ADS   Ribot RFH, Berg ML, Buchanan KL, Bennett ATD. 2013. Is there variation in the response to contact call playbacks across the hybrid zone of the parrot Platycercus elegans? Journal of Avian Biology  44: 399– 407. Google Scholar CrossRef Search ADS   Ribot RFH, Berg ML, Buchanan KL, Komdeur J, Joseph L, Bennett ATD. 2009. Does the ring species concept predict vocal variation in the crimson rosella, Platycercus elegans, complex? Animal Behaviour  77: 581– 593. Google Scholar CrossRef Search ADS   Ribot RFH, Buchanan KL, Endler JA, Joseph L, Bennett ATD, Berg ML. 2012. Learned vocal variation is associated with abrupt cryptic genetic change in a parrot species complex. PLoS ONE  7: e50484. Google Scholar CrossRef Search ADS   Ricklefs RE, Hainsworth FR. 1968. Temperature dependent behavior of the cactus wren. Ecology  49: 227– 233. Google Scholar CrossRef Search ADS   Ryeland J, Weston MA, Symonds MRE. 2017. Bill size mediates behavioural thermoregulation in birds. Functional Ecology  31: 885– 893. Google Scholar CrossRef Search ADS   Salaberria C, Celis P, López-Rull I, Gil D. 2014. Effects of temperature and nest heat exposure on nestling growth, dehydration and survival in a Mediterranean hole-nesting passerine. Ibis  156: 265– 275. Google Scholar CrossRef Search ADS   Sedgeley JA. 2001. Quality of cavity microclimate as a factor influencing selection of maternity roosts by a tree-dwelling bat, Chalinolobus tuberculatus, in New Zealand. Journal of Applied Ecology  38: 425– 438. Google Scholar CrossRef Search ADS   Şekercioğlu ÇH, Primack RB, Wormworth J. 2012. The effects of climate change on tropical birds. Biological Conservation  148: 1– 18. Google Scholar CrossRef Search ADS   Serrat MA, King D, Lovejoy CO. 2008. Temperature regulates limb length in homeotherms by directly modulating cartilage growth. Proceedings of the National Academy of Sciences, USA  105: 19348– 19353. Google Scholar CrossRef Search ADS   Sidhu LA, Dann P, Chambers L, Catchpole EA. 2012. Seasonal ocean temperature and the survival of first-year little penguins Eudyptula minor in south-eastern Australia. Marine Ecology Progress Series  454: 263– 272. Google Scholar CrossRef Search ADS   Siikamaki P. 1996. Nestling growth and mortality of Pied Flycatchers Ficedula hypoleuca in relation to weather and breeding effort. Ibis  138: 471– 478. Google Scholar CrossRef Search ADS   Symonds MRE, Tattersall GJ. 2010. Geographical variation in bill size across bird species provides evidence for Allen’s rule. The American Naturalist  176: 188– 197. Google Scholar CrossRef Search ADS   Tattersall GJ, Andrade DV, Abe AS. 2009. Heat exchange from the toucan bill reveals a controllable vascular thermal radiator. Science  325: 468– 470. Google Scholar CrossRef Search ADS   Tattersall GJ, Arnaout B, Symonds MRE. 2017. The evolution of the avian bill as a thermoregulatory organ. Biological Reviews  92: 1630– 1656. Google Scholar CrossRef Search ADS   Tattersall GJ, Chaves JA, Danner RM. 2018. Thermoregulatory windows in Darwin’s Finches. Functional Ecology  32: 358– 368. Google Scholar CrossRef Search ADS   van de Ven TMFN, Martin RO, Vink TJF, McKechnie AE, Cunningham SJ. 2016. Regulation of heat exchange across the Hornbill beak: functional similarities with toucans? PLoS ONE  11: e0154768. Google Scholar CrossRef Search ADS   Vasseur DA, DeLong JP, Gilbert B, Greig HS, Harley CDG, McCann KS, Savage V, Tunney TD, O’Connor MI. 2014. Increased temperature variation poses a greater risk to species than climate warming. Proceedings of the Royal Society B: Biological Sciences  281: 1– 8. © 2018 The Linnean Society of London, Biological Journal of the Linnean Society This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Biological Journal of the Linnean Society Oxford University Press

Nest microclimate predicts bill growth in the Adelaide rosella (Aves: Psittaculidae)

Loading next page...
 
/lp/ou_press/nest-microclimate-predicts-bill-growth-in-the-adelaide-rosella-aves-REhRRxNw7v
Publisher
The Linnean Society of London
Copyright
© 2018 The Linnean Society of London, Biological Journal of the Linnean Society
ISSN
0024-4066
eISSN
1095-8312
D.O.I.
10.1093/biolinnean/bly058
Publisher site
See Article on Publisher Site

Abstract

Abstract Diet and foraging have traditionally been considered key drivers of bill morphology. It is now known that bills play an important thermoregulatory role, and recent studies revealed that temperature is positively associated with the size of bills relative to body size or weight in adult birds, in accordance with Allen’s rule. Studies have attributed these patterns to local adaptation or an evolutionary response to climate change, but the contribution of ontogenetic plasticity remains unclear. We tested whether temperature experienced inside the nest predicted nestling growth in bill size and weight and in a parrot, the Adelaide rosella (Platycercus elegans adelaidae). We predicted that nest microclimate may affect bill ontogeny, leading to a positive association between relative bill size and temperatures during rearing. Growth in bill surface area was greater in nests that were warmer during the day and night, but temperature variability had no effect. Higher day and night-time mean temperatures, and less variable night-time temperatures, were positively associated with nestling weight. Our findings indicate that nest microclimate influences nestling growth, including relative bill size, and that daytime heat dissipation may be a driver of bill ontogeny. Bill plasticity in response to temperature during rearing could be an important but little studied contributor to morphology, due to the role of the bill in thermoregulation. Allen’s rule, ontogeny, phenotypic plasticity, Platycercus elegans adelaidae, thermoregulation INTRODUCTION Climate is an important determinant of animal morphology. This may arise due to evolutionary responses to temperature and adaptation to climatic niches, but temperature also has important implications for ontogeny in a wide range of taxa, such as mammals (Sedgeley, 2001; Serrat, King & Lovejoy, 2008), reptiles (Brewster, Sikes & Gifford, 2013; Pike, 2014) and birds (Burness et al., 2013; Şekercioğlu, Primack & Wormworth, 2012). Nest-site microclimate has been found to affect breeding success and ontogeny in many bird (such as passeriformes, piciformes and apodiformes) and bat species, particularly those breeding in nest boxes and hollows (Ardia, Perez & Clotfelter, 2010; Dawson, Lawrie & O’Brien, 2005; Larson et al., 2015). In birds, suboptimal temperature during incubation and during offspring development can result in increased incubation periods and reduced embryonic growth, hatchling weight, nestling growth, body condition and immunity, which can have long-term effects on fitness (Ardia, Perez & Clotfelter, 2006; Ardia et al., 2010; Burton, 2007; Cunningham et al., 2013; Larson et al., 2015; Murphy, 1985; Salaberria et al., 2014). Recently, it has also been shown that zebra finch (Taeniopygia guttata) parents communicate information on ambient temperature to embryos using acoustic signals, and that these calls alone alter the reproductive success and thermal preferences of these offspring as adults (Mariette & Buchanan, 2016). Nest microclimates can vary due to a large range of factors, such as orientation, construction materials, colour and nest size (Ardia et al., 2006; Goldingay, 2015). The considerable role of the bill for thermoregulation in birds, reviewed by Tattersall, Arnaout & Symonds (2017), has been identified through the use of thermal imaging (e.g. Greenberg et al., 2012a; Hagan & Heath, 1980; Tattersall, Andrade & Abe, 2009; van de Ven et al., 2016). Bills are uninsulated, and birds are able to dissipate large amounts of heat by controlling blood flow to their highly vascularized bills, whilst minimizing water loss (Hagan & Heath, 1980; Tattersall et al., 2009). Despite this control, bill surface area is thought to be subject to a trade-off between the need for heat dissipation in warm conditions and heat retention in cold conditions (Friedman et al., 2017). Recent studies have shown that bill surface area relative to a measure of body size/weight (relative bill surface area, RBSA) shows interspecific variation in relation to geographical distribution, and within-species variation in relation to habitat (Greenberg et al., 2012a; Symonds & Tattersall, 2010), seasonal temperatures (Danner & Greenberg, 2015; Friedman et al., 2017; Greenberg et al., 2012b) and long-term climate change (Campbell-Tennant et al., 2015). As a result of these patterns and the bill’s thermoregulatory function, spatial and temporal variation in RBSA is therefore thought to follow Allen’s rule (Allen, 1877). Allen’s rule predicts selection for larger appendage size (relative to body size) in warmer environments, such as lower latitudes/elevations, to achieve more efficient thermoregulation (Allen, 1877; Symonds & Tattersall, 2010; Tattersall et al., 2017). Most studies of relationships between bill size and temperature have done so in adult, fully developed birds, and the patterns are often attributed to local adaptation or an evolutionary response to climate change (Campbell-Tennant, Gardner, Kearney, Symonds & Ladle, 2015; Friedman et al., 2017; Symonds & Tattersall, 2010). However, it has also been proposed that developmental plasticity could contribute significantly to such patterns (Symonds & Tattersall, 2010; Tattersall et al., 2017), as it does in the limbs of mammals (Serrat et al., 2008). A recent study of captive precocial Japanese quail (Coturnix japonica) demonstrated experimentally that rearing temperature (at constant 15 °C or 30 °C) influenced bill growth in line with Allen’s rule, and resulted in irreversible effects on the thermoregulatory physiology of the bills which persisted even when birds were subsequently housed at a common temperature (Burness et al., 2013). If these patterns are general, then such bill size plasticity could be extremely important for nestlings as they may have less control of heat exchange through their bill than adults (Tattersall et al., 2009), coupled with little control over their thermal environments or dietary access to water compared to adults. This may be particularly true for obligate cavity nesters, such as parrots (Psittaciformes), as the often limited and specialized nature of their nest cavities may constrain their control of nest microclimates. To our knowledge, no study so far has investigated the effect of nest microclimate on bill ontogeny in altricial nestlings, or under natural conditions. Moreover, studies that have investigated the effects of temperature on reproductive success or ontogeny have often either done so in controlled laboratory conditions where temperature is kept constant, or have focused primarily on the effects of mean, maximum or minimum temperatures in isolation (e.g. Dawson et al., 2005; Salaberria et al., 2014; Sidhu et al., 2012). The importance of temperature variability is thus often not considered, although it may be expected to alter ontogeny or adaptation in relation to temperature, and some studies have highlighted the importance of temperature variability in climate change projections and for wildlife reproduction (Larson et al., 2015; Vasseur et al., 2014). For example, our previous study found that less extreme low temperatures resulted in heavier crimson rosella (Platycercus elegans elegans) nestlings, while greater temperature variability tended to reduce fledging success (Larson et al., 2015). In this study, we tested whether nest microclimate was associated with nestling growth and RBSA in a wild population of Adelaide rosellas (Platycercus elegans adelaidae) inhabiting temperate woodlands in south-eastern Australia. We recorded internal nest box temperature hourly, and analysed the thermal conditions encountered inside each nest during nestling growth in terms of both mean temperature and temperature variability (standard deviation). To assess diurnal changes in thermoregulatory demands, we also analysed daytime and night-time temperatures separately. This allowed us to distinguish effects arising from the higher, predominately increasing temperatures encountered during the day, when a need to dissipate heat is more likely, from those of the lower, predominately decreasing temperatures during the night, when a need to retain heat is more likely and food/water is less available. We used these data to test whether nest microclimate was related to (1) growth in nestling RBSA, or (2) nestling growth overall (body weight). We predicted that greater exposure to high temperatures inside the nest (higher or greater variability in temperature), especially during the daytime, would be associated with greater RBSA growth, due to the improved heat dissipation possible with larger RBSA. We also predicted that warmer nests, particularly at night when the lowest temperatures are experienced, would be associated with increased growth in body weight due to reduced energetic costs of thermoregulation. MATERIAL AND METHODS Study species and sites We studied an abundant hollow-breeding parrot, the Adelaide rosella. The general breeding biology, behaviour and sensory ecology of P. elegans has been previously documented (Krebs, 1998, Krebs, 1999, Krebs, 2001; Krebs & Magrath, 2000; Ribot et al., 2012, 2013; Mihailova et al., 2014, 2018; Larson et al., 2015; Knott et al., 2017). Platycercus elegans is a medium-sized (c. 35 cm long) parrot which occupies temperate, mesic habitats in south-eastern Australia, where it is an obligate cavity-nester (Krebs, 1998; Krebs & Magrath, 2000). The breeding season spans September to January, with most pairs laying 3–8 eggs (Krebs, 1998; Larson et al., 2015); incubation is 16–28 days (Krebs, 1998), with fledging occurring at 28–40 days post-hatching (Krebs, 1998; Larson et al., 2015). Adult P. e. adelaidae weigh 127.0 g ± 13.5 SD (females) to 130.4 g ± 8.0 SD (males), and tarsi lengths are 19.9 mm ± 1.0 SD (females) to 20.6 mm ± 0.7 SD (males) (our unpubl. data; no bill size data available). The time taken for nestlings to attain full size in this species varies depending on the trait being considered, and bills continue to grow linearly throughout the nestling period (Fig. 1). Figure 1. View largeDownload slide A, Nestling bill surface area (mm2) increases with nestling age (days) in Platycercus elegans adelaidae (N = 482 measuring events of 133 nestlings in 27 broods); growth in bill surface area was linear throughout the nestling stage. B, Nestling body weight (g) increases with nestling age (days) in P. e. adelaidae (N = 484 measuring events of 133 nestlings in 27 broods); an age of 22 days (vertical line) was considered the end of the linear growth phase for nestling weight. C, Nestling tarsus length (mm) increases with nestling age (days) in P. e. adelaidae (N = 484 measuring events of 133 nestlings in 27 broods); an age of 14 days (vertical line) was considered the end of growth for nestling tarsus length. Males are indicated by crosses and solid lines, females by circles and dashed lines; fit lines were derived from locally weighted scatterplot smoothing (LOESS). Figure 1. View largeDownload slide A, Nestling bill surface area (mm2) increases with nestling age (days) in Platycercus elegans adelaidae (N = 482 measuring events of 133 nestlings in 27 broods); growth in bill surface area was linear throughout the nestling stage. B, Nestling body weight (g) increases with nestling age (days) in P. e. adelaidae (N = 484 measuring events of 133 nestlings in 27 broods); an age of 22 days (vertical line) was considered the end of the linear growth phase for nestling weight. C, Nestling tarsus length (mm) increases with nestling age (days) in P. e. adelaidae (N = 484 measuring events of 133 nestlings in 27 broods); an age of 14 days (vertical line) was considered the end of growth for nestling tarsus length. Males are indicated by crosses and solid lines, females by circles and dashed lines; fit lines were derived from locally weighted scatterplot smoothing (LOESS). In this study, P. e. adelaidae nests were studied in dry sclerophyll eucalypt forests at Carey Gully, South Australia (34.97°S, 138.77°E). Sixty-one nest boxes were observed from September to December 2013. Of these nest boxes, 34 had been erected in 2004 as part of previous research projects (Joseph et al., 2008; Ribot et al., 2009, 2011; Berg & Bennett, 2010; Knott et al., 2013; Eastwood et al., 2014, 2015, 2017), while 27 were new boxes erected in July 2013 for the purpose of this study. Nest boxes were spaced at least 50 m apart and standardized for height (4-5 m above ground level), aspect (all faced south-east) and tree type (Eucalyptus spp., almost exclusively E. obliqua), as described by Larson et al. (2015). Nest boxes were constructed of 19-mm treated pine and were 24 cm wide × 28 cm deep × 42 cm high, with an entrance hole 7.5 cm in diameter and a sliding side door to allow observers access to nests. A c. 10-cm layer of Eucalyptus spp. wood chips had been placed in each nest box when erected as per Larson et al. (2015). Nestling measurements Nest boxes were checked every 2–3 days from 4 September 2013 until clutches were complete, and then weekly until 20 December 2013. Platycercus elegans lay eggs typically 2 days apart and have a high level of hatching asynchrony because incubation begins prior to clutch completion (Krebs, 1998, 1999). Nest box checks were more frequent (every 1–2 days) around the predicted hatching date to accurately determine hatching date. During this study, 283 eggs were laid, of which 149 hatched; reasons for failure to hatch included abandonment and predation/damage by possums. Thirty-three nestlings died (including those predated and 16 which died before any measurements were taken), leaving 116 to fledge successfully. Nestlings (N = 133 in 27 broods) were weighed using a spring scale (to the nearest 0.25 g; Pesola AG, Schindellegi, Switzerland) approximately every 7 days (mean 3.9 ± 1.1 SD visits per brood). At each weighing event (with the exception of two occasions), tarsus length, and bill length (BL), width (BW) and depth (BD) were also measured, by the same person (E.R.L.) each time, with callipers, to the nearest 0.1 mm. This allowed us to calculate bill surface area using eqn. 1, following previous studies on P. elegans and other species (Greenberg et al., 2012a, b; Greenberg & Danner, 2012; Campbell-Tennant et al., 2015). A limitation of this method is that it does not incorporate curvature of the bill; however, linear bill measurements show a high correlation (r = 0.98) with bill area in Australian parrots (Symonds & Tattersall, 2010). Nestlings were marked for individual identification, initially by trimming nails and down in specific combinations until approximately 10 days of age, when they were ringed with an individually numbered metal band (Australian Bird and Bat Banding Scheme). A small blood sample (< 100 µL) was taken from nestlings and used to determine sex, as described by Eastwood et al. (2015).  Bill surface area=(BD+BW4)×BL  Nest box temperature Seventy HOBO pendant temperature data loggers (Onset Computer Corporation, Bourne, MA, USA) were deployed in nest boxes on 24 June 2013. Loggers were placed in the top left-hand (i.e. upper north-west, hence out of direct sun exposure) corner of the nest boxes, and set to record temperature, date and time every hour following Larson et al. (2015). Logger accuracy drift over time was previously found to be negligible (Larson et al., 2015). Temperature variables To test whether nest temperature predicted nestling growth, we chose two temperature variables to represent the distribution of temperatures within nest boxes: mean temperature (Tmean), and variability in temperature (Tvariability; standard deviation). We used standard deviation as our measure of variability, as we were interested in the effects of absolute variability in temperature (i.e. not relative to mean temperature). We calculated Tmean and Tvariability separately for day and night periods; separating these periods for analysis allowed us to distinguish effects arising from the higher, mainly increasing temperatures encountered during the day from those of the lower, mainly decreasing temperatures during the night when a need for retaining heat is more likely. Day and night periods were defined by the mean sunrise and sunset times for the approximately 1-week periods between the weighing/measurement event at each nest box, and were therefore different for each brood as the timing of broods was not synchronous across the season. For every nest box, these temperature variables (daytime Tmean and Tvariability, and night-time Tmean and Tvariability) were calculated over each of the approximately 1-week periods between the weighing/measurement events, which allowed us to analyse the association between growth and temperature for each nestling since it was last weighed/measured. Thus, we had four different temperature variables to use as predictors for each time a brood was weighed and measured. To provide additional context for our study conditions we also provide, for daytime and night-time periods, the Tmean, mean minimum and mean maximum temperatures, averaged across all days in the nestling period for each brood. Statistical analysis We used linear mixed models (GLM; proc. MIXED with REML estimation) to test which temperature variables were associated with nestling growth in terms of RBSA and body weight. Bill growth was linear throughout the nestling period (Fig. 1A), but growth in weight slows after 22 days (Fig. 1B). Therefore, analyses presented below were restricted to nestlings from 1 to 22 days old (following Krebs, 1998; Larson et al., 2015), although the relationships between bill size and temperature were qualitatively unchanged when analysed over the entire nestling period. This resulted in 354 weighing events (mean = 2.02 ± 0.94 SD per nestling) from 133 nestlings in 27 broods for analysis. To analyse growth in bill size relative to overall size, we tested bill size controlled for body weight (Symonds & Tattersall, 2010). We used bill surface area as the dependent variable, with nestling weight and age (in days) as fixed covariates. We also repeated these models with weight excluded to test effects on growth of absolute bill surface area, and using tarsus length (modelled as a quadratic effect) as an alternative measure of nestling size instead of body weight. Tarsus was not our preferred measure of body size to calculate RBSA, as tarsus growth in this species plateaus very early in the nestling period (Fig. 1C), and because the uninsulated legs of birds are appendages which may themselves play a role in thermoregulation (Symonds & Tattersall, 2010; Burness et al., 2013). To analyse nestling growth in terms of body weight, we used nestling weight as the dependent variable, with age as a fixed covariate. To test the effects of temperature, Tmean and Tvariability were included as fixed covariates, and daytime and night-time temperatures were tested in separate models. Because brood size may affect growth due to sibling competition for food, brood size was included in all models We used current brood size (i.e. at the time of each weighing/measuring event) as we expected this would provide the most robust indication of competition, but also report the outcome of using initial brood size (i.e. brood size at the time of hatching) instead. For the analysis of nestling weight, brood size was included as a fixed factor because parameter estimates suggested that the relationship between brood size and nestling weight was non-linear. Parameter estimates suggested an approximately linear relationship between bill size and brood size, so brood size was included as a covariate for RBSA growth models; however, using brood size as a fixed factor had no qualitative effect on our results (not shown). Hatching date was included as a fixed covariate in all models to control for seasonal effects on growth. Because adult male P. elegans have larger bills than females (Krebs, 1999), we hypothesized that there may be sex differences in growth, particularly of the bill, and therefore included sex as a fixed factor in all models also. No interactions between sex and temperature variables were significant predictors for growth in either RBSA (P > 0.067) or weight (P > 0.084), or qualitatively changed the effects of other parameters, so these interactions were not included in the final models. To account for the repeated measures from nestlings and for nestlings being clustered within nest boxes, both nestling identity and nest box identity were included as random intercepts. Platycercus elegans generally have a single breeding attempt each season (Krebs, 1998; Larson et al., 2015), and in our study most breeding attempts overlapped temporally, so we assumed that each breeding pair was included in the analysis only once. All analyses were performed using SPSS Statistics for Windows version 24 (IBM Corp., 2016, Armonk, NY, USA). Conformity to assumptions including normality and homoscedasticity was confirmed following Quinn & Keough (2002). Variance inflation factors were < 1.4, indicating no problematic collinearity among predictors (Tmean, Tvariability, brood size, hatching date, age). Dates were converted into Julian date for all analyses. Degrees of freedom from mixed models were rounded to whole numbers. Means and estimates are presented with standard error unless otherwise noted, and the significance level was set a priori at α = 0.05. RESULTS The Tmean experienced inside nest boxes during our study, averaged across broods (N = 27), was 17.5 °C ± 1.7 SD (range 14.4–22.7 °C) during the day and 13.9 °C ± 1.5 SD (11.3–18.0 °C) during the night. Mean maximum daytime temperature was 21.3 °C ± 1.9 SD (18.6–26.8 °C) and mean minimum night-time temperature was 11.8 °C ± 1.6 SD (9.4–15.8 °C). Both daytime and night-time mean temperature had positive associations with nestling RBSA growth (Table 1, Supporting Information Fig. S1), with each 1 °C increase in mean temperature estimated to increase RBSA by 0.37 and 0.43 mm2 for daytime and night-time temperatures, respectively. Temperature variability was not associated with RBSA growth (Table 1, Fig. S1). When not controlling for nestling weight, daytime and night-time mean temperatures still had significant positive effects on bill surface area growth (daytime mean: estimate 0.487 °C ± 0.148 SE, P = 0.001; night-time mean: 0.605 °C ± 0.178 SE, P = 0.001), but temperature variability did not (daytime: −0.526 °C ± 0.498 SE, P = 0.292; night-time: −0.215 °C ± 0.482 SE, P = 0.656). The relationships between bill surface area and temperature were qualitatively the same when using tarsus as an alternative measure of nestling size (effect of Tmean: P < 0.041; results not shown). Table 1. Effects of internal nest box temperature, hatching date, brood size and sex on relative bill surface area growth (bill surface area controlling for age and weight) of Platycercus elegans adelaidae nestlings Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 246  0.371  0.126  8.705  0.003  0.123  0.618    Tvariability  1, 224  −0.619  0.415  2.223  0.137  −1.437  0.199    Hatching date  1, 33  −0.050  0.047  1.126  0.296  −0.146  0.046  Brood size  1, 61  0.979  0.296  10.912  0.002  0.386  1.571  Sex  1, 90  −3.469  0.610  32.378  < 0.001  −4.680  −2.258    Age  1, 323  1.123  0.127  77.802  < 0.001  0.872  1.373    Weight  1, 316  0.383  0.022  293.565  < 0.001  0.339  0.427  Night  Tmean  1, 250  0.431  0.150  8.260  0.004  0.136  0.727    Tvariability  1, 254  0.177  0.403  0.194  0.660  −0.616  0.971    Hatching date  1, 33  −0.056  0.047  1.385  0.248  −0.152  0.041  Brood size  1, 65  0.859  0.302  8.085  0.006  0.256  1.463  Sex  1, 90  −3.421  0.613  31.156  < 0.001  −4.638  −2.203    Age  1, 326  1.129  0.125  81.175  < 0.001  0.882  1.375    Weight  1, 320  0.383  0.022  298.924  < 0.001  0.339  0.426  Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 246  0.371  0.126  8.705  0.003  0.123  0.618    Tvariability  1, 224  −0.619  0.415  2.223  0.137  −1.437  0.199    Hatching date  1, 33  −0.050  0.047  1.126  0.296  −0.146  0.046  Brood size  1, 61  0.979  0.296  10.912  0.002  0.386  1.571  Sex  1, 90  −3.469  0.610  32.378  < 0.001  −4.680  −2.258    Age  1, 323  1.123  0.127  77.802  < 0.001  0.872  1.373    Weight  1, 316  0.383  0.022  293.565  < 0.001  0.339  0.427  Night  Tmean  1, 250  0.431  0.150  8.260  0.004  0.136  0.727    Tvariability  1, 254  0.177  0.403  0.194  0.660  −0.616  0.971    Hatching date  1, 33  −0.056  0.047  1.385  0.248  −0.152  0.041  Brood size  1, 65  0.859  0.302  8.085  0.006  0.256  1.463  Sex  1, 90  −3.421  0.613  31.156  < 0.001  −4.638  −2.203    Age  1, 326  1.129  0.125  81.175  < 0.001  0.882  1.375    Weight  1, 320  0.383  0.022  298.924  < 0.001  0.339  0.426  Significant terms are shown in bold. View Large Table 1. Effects of internal nest box temperature, hatching date, brood size and sex on relative bill surface area growth (bill surface area controlling for age and weight) of Platycercus elegans adelaidae nestlings Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 246  0.371  0.126  8.705  0.003  0.123  0.618    Tvariability  1, 224  −0.619  0.415  2.223  0.137  −1.437  0.199    Hatching date  1, 33  −0.050  0.047  1.126  0.296  −0.146  0.046  Brood size  1, 61  0.979  0.296  10.912  0.002  0.386  1.571  Sex  1, 90  −3.469  0.610  32.378  < 0.001  −4.680  −2.258    Age  1, 323  1.123  0.127  77.802  < 0.001  0.872  1.373    Weight  1, 316  0.383  0.022  293.565  < 0.001  0.339  0.427  Night  Tmean  1, 250  0.431  0.150  8.260  0.004  0.136  0.727    Tvariability  1, 254  0.177  0.403  0.194  0.660  −0.616  0.971    Hatching date  1, 33  −0.056  0.047  1.385  0.248  −0.152  0.041  Brood size  1, 65  0.859  0.302  8.085  0.006  0.256  1.463  Sex  1, 90  −3.421  0.613  31.156  < 0.001  −4.638  −2.203    Age  1, 326  1.129  0.125  81.175  < 0.001  0.882  1.375    Weight  1, 320  0.383  0.022  298.924  < 0.001  0.339  0.426  Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 246  0.371  0.126  8.705  0.003  0.123  0.618    Tvariability  1, 224  −0.619  0.415  2.223  0.137  −1.437  0.199    Hatching date  1, 33  −0.050  0.047  1.126  0.296  −0.146  0.046  Brood size  1, 61  0.979  0.296  10.912  0.002  0.386  1.571  Sex  1, 90  −3.469  0.610  32.378  < 0.001  −4.680  −2.258    Age  1, 323  1.123  0.127  77.802  < 0.001  0.872  1.373    Weight  1, 316  0.383  0.022  293.565  < 0.001  0.339  0.427  Night  Tmean  1, 250  0.431  0.150  8.260  0.004  0.136  0.727    Tvariability  1, 254  0.177  0.403  0.194  0.660  −0.616  0.971    Hatching date  1, 33  −0.056  0.047  1.385  0.248  −0.152  0.041  Brood size  1, 65  0.859  0.302  8.085  0.006  0.256  1.463  Sex  1, 90  −3.421  0.613  31.156  < 0.001  −4.638  −2.203    Age  1, 326  1.129  0.125  81.175  < 0.001  0.882  1.375    Weight  1, 320  0.383  0.022  298.924  < 0.001  0.339  0.426  Significant terms are shown in bold. View Large Higher mean daytime and night-time temperatures were both positively associated with nestling growth in terms of body weight (Table 2), with each 1 °C increase in daytime mean temperature estimated to increase nestling weight by 0.61 g and each 1 °C increase in night-time mean temperature estimated to increase mean nestling weight by 0.88 g (Table 2). Temperature variability at night was negatively associated with growth (Table 2), with each 1 °C increase in temperature standard deviation estimated to reduce mean nestling weight by 2.2 g. Table 2. Effects of internal nest box temperature, hatching date, brood size and sex on nestling growth (weight controlling for age) of Platycercus elegans adelaidae nestlings Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 236  0.608  0.299  4.134  0.043  0.019  1.120    Tvariability  1, 215  −1.068  1.000  1.139  0.287  −3.039  0.904    Hatching date  1, 28  −0.182  0.125  2.129  0.156  −0.437  0.074  Brood size  7, 92      3.008  0.007      Sex  1, 99  −2.756  1.730  2.539  0.114  −6.189  0.676    Age  1, 241  5.380  0.087  3853.167  < 0.001  5.209  5.550  Night  Tmean  1, 241  0.880  0.360  5.992  0.015  0.172  1.589    Tvariability  1, 236  −2.218  0.963  5.301  0.022  −4.116  −0.320    Hatching date  1, 27  −0.168  0.126  1.772  0.194  −0.426  0.091  Brood size  7, 90      3.101  0.006      Sex  1, 98  −3.052  1.725  3.131  0.080  −6.475  0.371    Age  1, 236  5.343  0.087  3807.018  < 0.001  5.173  5.514  Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 236  0.608  0.299  4.134  0.043  0.019  1.120    Tvariability  1, 215  −1.068  1.000  1.139  0.287  −3.039  0.904    Hatching date  1, 28  −0.182  0.125  2.129  0.156  −0.437  0.074  Brood size  7, 92      3.008  0.007      Sex  1, 99  −2.756  1.730  2.539  0.114  −6.189  0.676    Age  1, 241  5.380  0.087  3853.167  < 0.001  5.209  5.550  Night  Tmean  1, 241  0.880  0.360  5.992  0.015  0.172  1.589    Tvariability  1, 236  −2.218  0.963  5.301  0.022  −4.116  −0.320    Hatching date  1, 27  −0.168  0.126  1.772  0.194  −0.426  0.091  Brood size  7, 90      3.101  0.006      Sex  1, 98  −3.052  1.725  3.131  0.080  −6.475  0.371    Age  1, 236  5.343  0.087  3807.018  < 0.001  5.173  5.514  Significant predictors are shown in bold. Brood size was treated as a categorical predictor. View Large Table 2. Effects of internal nest box temperature, hatching date, brood size and sex on nestling growth (weight controlling for age) of Platycercus elegans adelaidae nestlings Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 236  0.608  0.299  4.134  0.043  0.019  1.120    Tvariability  1, 215  −1.068  1.000  1.139  0.287  −3.039  0.904    Hatching date  1, 28  −0.182  0.125  2.129  0.156  −0.437  0.074  Brood size  7, 92      3.008  0.007      Sex  1, 99  −2.756  1.730  2.539  0.114  −6.189  0.676    Age  1, 241  5.380  0.087  3853.167  < 0.001  5.209  5.550  Night  Tmean  1, 241  0.880  0.360  5.992  0.015  0.172  1.589    Tvariability  1, 236  −2.218  0.963  5.301  0.022  −4.116  −0.320    Hatching date  1, 27  −0.168  0.126  1.772  0.194  −0.426  0.091  Brood size  7, 90      3.101  0.006      Sex  1, 98  −3.052  1.725  3.131  0.080  −6.475  0.371    Age  1, 236  5.343  0.087  3807.018  < 0.001  5.173  5.514  Time period  Predictor  d.f.  Estimates  SE  F  P  Lower 95% CI  Upper 95% CI  Day  Tmean  1, 236  0.608  0.299  4.134  0.043  0.019  1.120    Tvariability  1, 215  −1.068  1.000  1.139  0.287  −3.039  0.904    Hatching date  1, 28  −0.182  0.125  2.129  0.156  −0.437  0.074  Brood size  7, 92      3.008  0.007      Sex  1, 99  −2.756  1.730  2.539  0.114  −6.189  0.676    Age  1, 241  5.380  0.087  3853.167  < 0.001  5.209  5.550  Night  Tmean  1, 241  0.880  0.360  5.992  0.015  0.172  1.589    Tvariability  1, 236  −2.218  0.963  5.301  0.022  −4.116  −0.320    Hatching date  1, 27  −0.168  0.126  1.772  0.194  −0.426  0.091  Brood size  7, 90      3.101  0.006      Sex  1, 98  −3.052  1.725  3.131  0.080  −6.475  0.371    Age  1, 236  5.343  0.087  3807.018  < 0.001  5.173  5.514  Significant predictors are shown in bold. Brood size was treated as a categorical predictor. View Large Nestling sex was a significant predictor of RBSA growth (Table 1) but not growth in weight (Table 2), with males having higher RBSA growth. No effects of hatching date on growth in nestling RBSA (Table 1) or weight (Table 2) were found. Brood size (at the time of weighing/measuring) significantly predicted nestling growth in terms of both RBSA (Table 1) and weight (Table 2); broods with two to six nestlings had lower growth than smaller and larger brood sizes. In contrast, initial brood size (at hatching) did not significantly affect growth in RBSA (P > 0.157) or weight (P > 0.073), nor did replacing current brood size with initial brood size qualitatively change the conclusions about any other variables (results not shown). As expected, because nestling bills and weights were measured during a period of rapid growth, age always had a significant effect on both RBSA (Table 1) and weight (Table 2). After controlling for age, RBSA was still positively predicted by weight (Table 1) or tarsus length (P < 0.045). DISCUSSION Bills are important thermoregulatory structures in both adult and developing birds, with bill size and vascularity affecting the radiative capacity of the bill (Burness et al., 2013; Tattersall et al., 2009, 2017). Recent studies of RBSA in adult birds have shown interspecific variation in relation to geographical distribution, and within-species variation in relation to habitat (Symonds & Tattersall, 2010; Greenberg et al., 2012a), seasonal temperatures (Greenberg et al., 2012b; Danner & Greenberg, 2015; Friedman et al., 2017) and long-term climate change (Campbell-Tennant et al., 2015). However, the contribution of temperature-dependent development to such patterns has been little reported (Burness et al., 2013). Our study revealed that, in nestling Adelaide rosellas, growth in bill size and RBSA was positively related to mean temperature inside the nest, and this was true for both daytime and night-time temperatures. Temperature variability was not related to RBSA, but nestling growth in terms of body weight was negatively related to night-time temperature variability. Our study was conducted in temperate conditions in south-eastern Australia, and the temperatures that nestlings were exposed to inside nest boxes during our study, particularly during the day, were within the range where radiant heat dissipation through the bill is possibly effective and related to bill size (Tattersall et al., 2009, 2017; Ryeland, Weston & Symonds, 2017; but see van de Ven et al., 2016), and also where effects of temperature on bill growth have been previously demonstrated in the laboratory (Burness et al., 2013). Temperatures were also generally below critical thresholds above which nestling growth or survival was impaired in other species (e.g. Cunningham et al., 2013). To our knowledge, our study is the first to investigate the effects of temperature during rearing on bill ontogeny in situ in free-living birds, and to compare the effects of day and night temperatures on bill size for adult or juvenile birds. This may be important, as historical data and climate change projections indicate night-time temperatures are increasing faster than daytime temperatures globally (Davy et al., 2017). Growth in relative bill size Larger RBSA is expected to permit better heat dissipation by increasing the radiative capacity of the bill, and may result from a developmental response to temperature (Burness et al., 2013; Tattersall et al., 2017; Tattersall, Chaves & Danner, 2018). Therefore, we predicted that greater exposure to high temperatures inside the nest, especially during the day, would be associated with greater RBSA growth. Our results were consistent with this prediction, as we found that higher daytime and night-time mean temperature was positively associated with growth in RBSA (controlled for nestling weight), although we found no effects of temperature variability. This pattern corresponds to Allen’s rule, which states that in warmer environments animals possess larger appendages in order to eliminate excess heat (Allen, 1877). Our findings suggest that relative bill size may at least partially reflect an ontogenetic response to variation in rearing temperature resulting from nest microclimates. If developmental differences in bill size or thermoregulatory function persist into adulthood, as shown by Burness et al. (2013), then such developmental plasticity may be an underappreciated contributor to widely observed patterns of relative bill size, which have often been attributed to local adaptation or an evolutionary response to climate change (Symonds & Tattersall, 2010; Campbell-Tennant et al., 2015; Friedman et al., 2017; Tattersall et al., 2017, 2018). Within- and between-species comparisons have observed strong positive relationships between higher temperatures and larger bill size in several taxa, including parrots, galliforms, penguins and gulls (Symonds & Tattersall, 2010; Campbell-Tennant et al., 2015), Meliphagoidea (Gardner et al., 2016; Friedman et al., 2017), salt-marsh sparrows (Ammodramus caudacutus) (Greenberg et al., 2012b), song sparrows (Melospiza melodia) (Danner & Greenberg, 2015), hornbills (Tockus leucomelas) (van de Ven et al., 2016) and starlings (Sturnus vulgaris) (Cardilini et al., 2016). However, all of these studies were based on observations of adult birds, whereas we studied nestlings during development. Our findings complement a previous laboratory experiment on Japanese quail, a precocial species not confined to a nest site (Burness et al., 2013). That study reported that juvenile quail which had been raised in captivity at 15 °C developed shorter bills than those raised at 30 °C. One strength of that study is that it also followed birds into maturity and measured bill surface temperatures, revealing that although previously cold-reared birds exhibited catch-up growth when subsequently housed at a common temperature, the difference in thermoregulatory function of the bills in cold- and warm-reared birds persisted, possibly due to differences in vascularity of the bill. Thus, the results suggested that differences in bill development related to rearing temperature may result in persistent, possibly lifelong, physiological changes (Burness et al., 2013). As we could not follow birds after fledging and it appears the bills were still growing (linearly) at fledging (Fig. 1A), it remains unknown whether bill differences (in size or thermoregulatory function) which develop during the nestling stage are maintained after fledging under natural conditions, when birds are free to choose microclimates and diets in maturity. To resolve this, future studies should follow juveniles under natural conditions. If differences are maintained, then bill size plasticity in nestlings could have implications for the resilience of birds to climate change (Campbell-Tennant et al., 2015), and a mismatch between temperatures experienced during development and adulthood may result in heat loss or heat deficits through the bill in adult birds. Moreover, diet, foraging strategies and song characteristics can all be affected by bill size (Christensen, Kleindorfer & Robertson, 2006). Hence, it is possible that ontogenetic variation in bills, if maintained in adulthood, could have considerable fitness effects throughout life. Our study suggests that natural differences in microclimate between nest sites, which are much smaller than the experimental temperature differences used by Burness et al. (2013), may also impact on bill size development. Unlike Burness et al. (2013), our study was correlational, so we are unable to fully exclude other factors which may be correlated with nest microclimates, such as brood size (although this was controlled for statistically in our models) or behaviour. Humidity was not measured in the present study, but may also be of considerable importance to avian thermoregulation, and thus bill development, because it mediates evaporative water loss (e.g. Gerson et al., 2014). Bill size, which contributes to non-evaporative heat dissipation, may become more important at high humidity (van de Ven et al., 2016). Experimental manipulations of nest temperature (e.g. using insulation or heating pads), possibly combined with brood swaps to control for genetic influences, would be a useful next step to overcome such limitations. A further limitation of our study, and most studies on bill size, is that the calculation of bill surface area is based on linear measurements and does not account for variation in the curvature of the bill. More accurate methods for measuring surface area (e.g. from photos) may yield more robust results. We found that larger brood size at the time of measuring events was associated with greater RBSA (and weight) growth, which we speculate may occur due to the increased heat generated by larger broods and limitations on evaporative cooling imposed by more crowded nest cavities. Interestingly, no effects of brood size at hatching were found, reinforcing the view that variation in brood size was directly affecting short-term growth rather than reflecting indirect effects of nest or parental quality or investment. Nestling sex was also a strong predictor of RBSA, which concurs with the larger head–bill size observed in adult males (41.3 mm ± 0.9 SD, N = 35) compared to females (38.5 mm ± 0.9 SD, N = 47) which we have previously observed in this study population (t80 = 14.114, P < 0.001; our unpubl. data). A recent comparative study of shorebirds has shown that behavioural thermoregulation is mediated by bill size at the species level, with placement of the bill within the plumage for insulation while resting being more common in species with larger bills relative to their body weight (Ryeland et al., 2017). Therefore, it would also be interesting for future studies to investigate the possible behavioural implications of temperature-related ontogenetic variation in bills. Similarly, future research could also assess whether behavioural responses to temperature by parents, such as time spent in foraging or parental care activities (Cunningham et al., 2013; Du Plessis et al., 2012), are associated with the bill development of nestlings. Such effects on development may even begin in the incubation stage: in a recent study of zebra finches, Mariette & Buchanan (2016) found that higher nest temperatures were associated with higher weight throughout the nestling stage, but this pattern was reversed when embryos were experimentally exposed to conspecific incubation calls prior to hatching. Growth in body weight We found that growth in body weight (weight controlled for age) increased with higher mean temperatures in the nest during growth, which is consistent with previous research in other species (e.g. Siikamaki, 1996; Eeva et al., 2002; Dawson et al., 2005; Pipoly et al., 2013; Chausson et al., 2014; Mariette & Buchanan, 2016; Martin et al., 2017). However, the opposite pattern has been found in some studies. For example, in their experimental study of Japanese quail, Burness et al. (2013) found that birds raised at 15 °C gained more body weight than those raised at 30 °C, and retained a small (approximately 1%) but significant lead when housed subsequently at a common temperature. In that experiment, the birds had ad libitum access to food, which may allow the cold-reared birds to compensate with a larger food intake. Another explanation for relationships between growth and temperature is that temperature-mediated changes in parental foraging activity or efficiency could lead to reduced provisioning to offspring and resultant decrease in nestling growth rates (Ricklefs & Hainsworth, 1968; Murphy, 1985; Du Plessis et al., 2012; Cunningham et al., 2013). These patterns may therefore be expected to differ between altricial and precocial species, and to depend on the correlation between nest microclimates and the ambient temperatures experienced by parents while foraging outside the nest. Relationships between temperature and growth are also dependent on the magnitude of temperatures experienced, and a positive relationship between temperature and growth arising from lower energy demands for thermoregulation may reverse once ambient temperature surpasses a critical threshold (Cunningham et al., 2013). In our study, estimated effects of mean temperature were slightly larger for the night, and only night-time variability in temperature affected growth, suggesting that temperatures at night are particularly influential on nestling weight. This may reflect increased energetic costs of thermoregulation for nestlings which experience low temperatures at night (Dawson et al., 2005; Cunningham et al., 2013). Our current results concur with the findings of our previous study on P. e. elegans, which found that minimum temperature experienced at night during the nestling period, followed by mean temperature, were the most important predictors of nestling growth (Larson et al., 2015). A relationship between growth and temperature variability was also identified by Larson (2015), although that study is not comparable in this context as it did not separate the daytime and night-time periods, and thus greater temperature standard deviation could arise either from more extreme daytime temperatures or lower night-time temperatures. Many studies that have investigated the relationship between nest-site temperature and nestling growth have been undertaken in the northern hemisphere at higher latitudes than our study, but our results indicate that even under mild temperate conditions (mean day and night temperatures in nest boxes in our study were 17.5 and 13.9 °C, respectively) nestling growth may be positively related to higher mean temperatures. However, such benefits may be tempered by an increasing frequency or magnitude of temperature extremes under a warming climate (Cunningham et al., 2013). CONCLUSIONS We found that nest microclimate predicts nestling growth in terms of both body weight, and bill surface area relative to body weight, in an Australian parrot. Higher mean temperature was positively related to growth in weight and RBSA, and this was true for both daytime and night-time temperatures. In contrast, greater night-time temperature variability negatively affected weight of nestlings. Recent studies have shown that bills play an important role in thermoregulation for birds (Tattersall et al., 2009; Burness et al., 2013; van de Ven et al., 2016; Ryeland et al., 2017). Differences in RBSA of adult birds have been reported within and between species in relation to temporal and geographical variation in temperature, in line with Allen’s rule (Tattersall et al., 2017). Our results support the hypothesis that plasticity in bill development has the potential to contribute to such patterns due to variation in nest microclimate. Future work could build on these results through experimental manipulations of nest microclimate, and by following into adulthood the thermoregulatory capacity of bills and related fitness consequences. Further study of the developmental plasticity of bills, the potential for temperature during rearing and nest microclimate to influence bill morphology, and the implications for thermoregulation in adult birds, will facilitate a better understanding of avian responses to increasing temperatures and climate variability due to climate change. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s web-site. Figure S1. Relationships between temperature variables and predicted bill surface area (mm2) of nestling P. elegans adelaidae derived from the models shown in Table 1. Symbols indicate measurement events at each brood (solid circles: first measurement; open circles: second measurement; crosses: third measurement; diamonds: fourth measurement; triangle: fifth measurement). Overall fit lines are for illustrative purposes, and do not account for repeated measures of nestlings. ACKNOWLEDGEMENTS We are grateful to the landholders who allowed us to work on their properties or assisted with the research, particularly Brian and Maggie Colton, Robert and Jennie Henzell, Andrew Stanley and Chic Stirling Phillips, and George and Alice Avramidis. We also thank Martin and Virginia Gare, Raoul Ribot, Kate Buchanan, and Bill Buttemer for assistance, and the many field volunteers who helped collect data used in this study. Juan Amat, Ross Goldingay, Peter Korsten, anonymous reviewers and the editor provided valuable comments on earlier drafts of the manuscript. This work was funded by a Deakin University postgraduate scholarship to E.R.L., an Australian Research Council Linkage grant (LP140100691) to A.T.D.B. and M.L.B., Birdlife Australia, Holsworth Research Endowment and Deakin University. A.T.D.B. and M.L.B. were also supported by Australian Research Council Discovery grant DP180103494. We are grateful to the Australian Bird and Bat Banding Scheme for the supply of bands and support. All applicable institutional and/or national guidelines for the care and use of animals were followed. The research reported in this study conformed to the laws of Australia and the state of South Australia (Deakin University Animal Ethics Committee approval A33-2008 and State research permit 10004759-3). The authors declare no conflicts of interest. REFERENCES Allen JA. 1877. The influence of physical conditions in the genesis of species. Radical Review  1: 108– 140. Ardia DR, Perez JH, Clotfelter ED. 2006. Nest box orientation affects internal temperature and nest site selection by Tree Swallows. Journal of Field Ornithology  77: 339– 344. Google Scholar CrossRef Search ADS   Ardia DR, Perez JH, Clotfelter ED. 2010. Experimental cooling during incubation leads to reduced innate immunity and body condition in nestling tree swallows. Proceedings of the Royal Society of London, Series B: Biological Sciences  277: 1881– 1888. Google Scholar CrossRef Search ADS   Berg ML, Bennett ATD. 2010. The evolution of plumage colouration in parrots: a review. Emu  110: 10– 20. Google Scholar CrossRef Search ADS   Brewster CL, Sikes RS, Gifford ME. 2013. Quantifying the cost of thermoregulation: thermal and energetic constraints on growth rates in hatchling lizards. Functional Ecology  27: 490– 497. Google Scholar CrossRef Search ADS   Burness G, Huard JR, Malcolm E, Tattersall GJ. 2013. Post-hatch heat warms adult beaks: irreversible physiological plasticity in Japanese quail. Proceedings of the Royal Society of London, Series B: Biological Sciences  280: 20131436. Google Scholar CrossRef Search ADS   Burton NHK. 2007. Intraspecific latitudinal variation in nest orientation among ground-nesting passerines: a study using published data. Condor  109: 441– 446. Google Scholar CrossRef Search ADS   Campbell-Tennant DJE, Gardner JL, Kearney MR, Symonds MRE, Ladle R. 2015. Climate-related spatial and temporal variation in bill morphology over the past century in Australian parrots. Journal of Biogeography  42: 1163– 1175. Google Scholar CrossRef Search ADS   Cardilini AP, Buchanan KL, Sherman CD, Cassey P, Symonds MRE. 2016. Tests of ecogeographical relationships in a non-native species: what rules avian morphology? Oecologia  181: 783– 793. Google Scholar CrossRef Search ADS   Chausson A, Henry I, Almasi B, Roulin A. 2014. Barn Owl (Tyto alba) breeding biology in relation to breeding season climate. Journal of Ornithology  155: 273– 281. Google Scholar CrossRef Search ADS   Christensen R, Kleindorfer S, Robertson J. 2006. Song is a reliable signal of bill morphology in Darwin’s small tree finch Camarhynchus parvulus, and vocal performance predicts male pairing success. Journal of Avian Biology  37: 617– 624. Google Scholar CrossRef Search ADS   Cunningham SJ, Martin RO, Hojem CL, Hockey PA. 2013. Temperatures in excess of critical thresholds threaten nestling growth and survival in a rapidly-warming arid savanna: a study of common fiscals. PLoS ONE  8: e74613. Google Scholar CrossRef Search ADS   Danner RM, Greenberg R. 2015. A critical season approach to Allen’s rule: bill size declines with winter temperature in a cold temperate environment. Journal of Biogeography  42: 114– 120. Google Scholar CrossRef Search ADS   Davy R, Esau I, Chernokulsky A, Outten S, Zilitinkevich S. 2017. Diurnal asymmetry to the observed global warming. International Journal of Climatology  37: 79– 93. Google Scholar CrossRef Search ADS   Dawson RD, Lawrie CC, O’Brien EL. 2005. The importance of microclimate variation in determining size, growth and survival of avian offspring: experimental evidence from a cavity nesting passerine. Oecologia  144: 499– 507. Google Scholar CrossRef Search ADS   du Plessis KL, Martin RO, Hockey PAR, Cunningham SJ, Ridley AR. 2012. The costs of keeping cool in a warming world: implications of high temperatures for foraging, thermoregulation and body condition of an arid-zone bird. Global Change Biology  18: 3063– 3070. Google Scholar CrossRef Search ADS   Eastwood JR, Berg ML, Ribot RFH, Raidal SR, Buchanan KL, Walder KR, Bennett ATD. 2014. Phylogenetic analysis of beak and feather disease virus across a host ring-species complex. Proceedings of the National Academy of Sciences USA  111: 14153– 14158. Google Scholar CrossRef Search ADS   Eastwood JR, Berg ML, Spolding B, Buchanan KL, Bennett ATD, Walder K. 2015. Prevalence of beak and feather disease virus in wild Platycercus elegans: comparison of three tissue types using a probe-based real-time qPCR test. Australian Journal of Zoology  63: 1– 8. Google Scholar CrossRef Search ADS   Eastwood JR, Ribot RFH, Rollins LA, Buchanan KL, Walder K, Bennett ATD, Berg ML. 2017. Host heterozygosity and genotype rarity affect viral dynamics in an avian subspecies complex. Scientific Reports  7: 13310. Google Scholar CrossRef Search ADS   Eeva T, Lehikoinen E, Ronka M, Lummaa V, Currie D. 2002. Different responses to cold weather in two pied flycatcher populations. Ecography  25: 705– 713. Google Scholar CrossRef Search ADS   Friedman NR, Harmáčková L, Economo EP, Remeš V. 2017. Smaller beaks for colder winters: thermoregulation drives beak size evolution in Australasian songbirds. Evolution  71: 2120– 2129. Google Scholar CrossRef Search ADS   Gardner JL, Symonds MRE, Joseph L, Ikin K, Stein J, Kruuk LEB. 2016. Spatial variation in avian bill size is associated with humidity in summer among Australian passerines. Climate Change Responses  3: 11. Google Scholar CrossRef Search ADS   Gerson AR, Smith EK, Smit B, McKechnie AE, Wolf BO. 2014. The impact of humidity on evaporative cooling in small desert birds exposed to high air temperatures. Physiological and Biochemical Zoology  87: 782– 795. Google Scholar CrossRef Search ADS   Goldingay RL. 2015. Temperature variation in nest boxes in eastern Australia. Australian Mammalogy  37: 225– 233. Google Scholar CrossRef Search ADS   Greenberg R, Cadena V, Danner RM, Tattersall GJ, Tattersall G. 2012a. Heat loss may explain bill size differences between birds occupying different habitats. PLoS ONE  7: e40933. Google Scholar CrossRef Search ADS   Greenberg R, Danner R, Olsen B, Luther D. 2012b. High summer temperature explains bill size variation in salt marsh sparrows. Ecography  35: 146– 152. Google Scholar CrossRef Search ADS   Greenberg R, Danner RM. 2012. The influence of the California marine layer on bill size in a generalist songbird. Evolution  66: 3825– 3835. Google Scholar CrossRef Search ADS   Hagan AA, Heath JE. 1980. Regulation of heat loss in the duck by vasomotion in the bill. Journal of Thermal Biology  5: 95– 101. Google Scholar CrossRef Search ADS   Joseph L, Dolman G, Donnellan S, Saint KM, Berg ML, Bennett ATD. 2008. Where and when does a ring start and end? Testing the ring-species hypothesis in a species complex of Australian parrots. Proceedings of the Royal Society of London B: Biological Sciences  275: 2431– 2440. Google Scholar CrossRef Search ADS   Knott B, Berg ML, Ribot RFH, Endler JA, Bennett ATD. 2017. Intraspecific geographic variation in rod and cone visual pigment sensitivity of a parrot, Platycercus elegans. Scientific Reports  7: 41445. Google Scholar CrossRef Search ADS   Knott B, Davies WI, Carvalho LS, Berg ML, Buchanan KL, Bowmaker JK, Bennett ATD, Hunt DM. 2013. How parrots see their colours: novelty in the visual pigments of Platycercus elegans. Journal of Experimental Biology  216: 4454– 4461. Google Scholar CrossRef Search ADS   Krebs EA. 1998. Breeding biology of crimson rosellas (Platycercus elegans) on Black Mountain, Australian Capital Territory. Australian Journal of Zoology  46: 119– 136. Google Scholar CrossRef Search ADS   Krebs EA. 1999. Last but not least: nestling growth and survival in asynchronously hatching crimson rosellas. Journal of Animal Ecology  68: 266– 281. Google Scholar CrossRef Search ADS   Krebs EA. 2001. Begging and food distribution in crimson rosella (Platycercus elegans) broods: why don’t hungry chicks beg more? Behavioral Ecology and Sociobiology  50: 20– 30. Google Scholar CrossRef Search ADS   Krebs EA, Magrath RD. 2000. Food allocation in crimson rosella broods: parents differ in their responses to chick hunger. Animal Behaviour  59: 739– 751. Google Scholar CrossRef Search ADS   Larson ER, Eastwood JR, Buchanan KL, Bennett ATD, Berg ML. 2015. How does nest box temperature affect nestling growth rate and breeding success in a parrot? Emu Austral Ornithology  115: 247– 255. Google Scholar CrossRef Search ADS   Mariette MM, Buchanan KL. 2016. Prenatal acoustic communication programs offspring for high posthatching temperatures in a songbird. Science  353: 812– 814. Google Scholar CrossRef Search ADS   Martin TE, Boyce AJ, Fierro-Calderón K, Mitchell AE, Armstad CE, Mouton JC, Bin Soudi EE. 2017. Enclosed nests may provide greater thermal than nest predation benefits compared with open nests across latitudes. Functional Ecology  31: 1231– 1240. Google Scholar CrossRef Search ADS   Mihailova M, Berg ML, Buchanan KL, Bennett ATD. 2014. Odour-based discrimination of subspecies, species and sexes in an avian species complex, the crimson rosella. Animal Behaviour  95: 155– 164. Google Scholar CrossRef Search ADS   Mihailova M, Berg ML, Buchanan KL, Bennett ATD. 2018. Olfactory eavesdropping: The odor of feathers is detectable to mammalian predators and competitors. Ethology  124: 14– 24. Google Scholar CrossRef Search ADS   Murphy MT. 1985. Nestling eastern kingbird growth: effects of initial size and ambient temperature. Ecology  66: 162– 170. Google Scholar CrossRef Search ADS   Pike DA. 2014. Forecasting the viability of sea turtle eggs in a warming world. Global Change Biology  20: 7– 15. Google Scholar CrossRef Search ADS   Pipoly I, Bókony V, Seress G, Szabó K, Liker A. 2013. Effects of extreme weather on reproductive success in a temperate-breeding songbird. PLoS ONE  8: e80033. Google Scholar CrossRef Search ADS   Quinn GP, Keough MJ. 2002. Experimental design and data analysis for biologists . Cambridge: Cambridge University Press. Google Scholar CrossRef Search ADS   Ribot RFH, Berg ML, Buchanan KL, Bennett ATD. 2011. Fruitful use of bioacoustic alarm stimuli as a deterrent for Crimson Rosellas (Platycercus elegans). Emu  111: 360– 367. Google Scholar CrossRef Search ADS   Ribot RFH, Berg ML, Buchanan KL, Bennett ATD. 2013. Is there variation in the response to contact call playbacks across the hybrid zone of the parrot Platycercus elegans? Journal of Avian Biology  44: 399– 407. Google Scholar CrossRef Search ADS   Ribot RFH, Berg ML, Buchanan KL, Komdeur J, Joseph L, Bennett ATD. 2009. Does the ring species concept predict vocal variation in the crimson rosella, Platycercus elegans, complex? Animal Behaviour  77: 581– 593. Google Scholar CrossRef Search ADS   Ribot RFH, Buchanan KL, Endler JA, Joseph L, Bennett ATD, Berg ML. 2012. Learned vocal variation is associated with abrupt cryptic genetic change in a parrot species complex. PLoS ONE  7: e50484. Google Scholar CrossRef Search ADS   Ricklefs RE, Hainsworth FR. 1968. Temperature dependent behavior of the cactus wren. Ecology  49: 227– 233. Google Scholar CrossRef Search ADS   Ryeland J, Weston MA, Symonds MRE. 2017. Bill size mediates behavioural thermoregulation in birds. Functional Ecology  31: 885– 893. Google Scholar CrossRef Search ADS   Salaberria C, Celis P, López-Rull I, Gil D. 2014. Effects of temperature and nest heat exposure on nestling growth, dehydration and survival in a Mediterranean hole-nesting passerine. Ibis  156: 265– 275. Google Scholar CrossRef Search ADS   Sedgeley JA. 2001. Quality of cavity microclimate as a factor influencing selection of maternity roosts by a tree-dwelling bat, Chalinolobus tuberculatus, in New Zealand. Journal of Applied Ecology  38: 425– 438. Google Scholar CrossRef Search ADS   Şekercioğlu ÇH, Primack RB, Wormworth J. 2012. The effects of climate change on tropical birds. Biological Conservation  148: 1– 18. Google Scholar CrossRef Search ADS   Serrat MA, King D, Lovejoy CO. 2008. Temperature regulates limb length in homeotherms by directly modulating cartilage growth. Proceedings of the National Academy of Sciences, USA  105: 19348– 19353. Google Scholar CrossRef Search ADS   Sidhu LA, Dann P, Chambers L, Catchpole EA. 2012. Seasonal ocean temperature and the survival of first-year little penguins Eudyptula minor in south-eastern Australia. Marine Ecology Progress Series  454: 263– 272. Google Scholar CrossRef Search ADS   Siikamaki P. 1996. Nestling growth and mortality of Pied Flycatchers Ficedula hypoleuca in relation to weather and breeding effort. Ibis  138: 471– 478. Google Scholar CrossRef Search ADS   Symonds MRE, Tattersall GJ. 2010. Geographical variation in bill size across bird species provides evidence for Allen’s rule. The American Naturalist  176: 188– 197. Google Scholar CrossRef Search ADS   Tattersall GJ, Andrade DV, Abe AS. 2009. Heat exchange from the toucan bill reveals a controllable vascular thermal radiator. Science  325: 468– 470. Google Scholar CrossRef Search ADS   Tattersall GJ, Arnaout B, Symonds MRE. 2017. The evolution of the avian bill as a thermoregulatory organ. Biological Reviews  92: 1630– 1656. Google Scholar CrossRef Search ADS   Tattersall GJ, Chaves JA, Danner RM. 2018. Thermoregulatory windows in Darwin’s Finches. Functional Ecology  32: 358– 368. Google Scholar CrossRef Search ADS   van de Ven TMFN, Martin RO, Vink TJF, McKechnie AE, Cunningham SJ. 2016. Regulation of heat exchange across the Hornbill beak: functional similarities with toucans? PLoS ONE  11: e0154768. Google Scholar CrossRef Search ADS   Vasseur DA, DeLong JP, Gilbert B, Greig HS, Harley CDG, McCann KS, Savage V, Tunney TD, O’Connor MI. 2014. Increased temperature variation poses a greater risk to species than climate warming. Proceedings of the Royal Society B: Biological Sciences  281: 1– 8. © 2018 The Linnean Society of London, Biological Journal of the Linnean Society This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Biological Journal of the Linnean SocietyOxford University Press

Published: May 29, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off