Advertisement

Prey size and nestling gape size affect allocation within broods of the Mountain Bluebird

  • Jordyn A. StalwickEmail author
  • Karen L. Wiebe
Original Article

Abstract

The allocation of prey within broods of altricial nestlings may be the result of parents targeting certain offspring preferentially and/or the result of intrabrood competition related to features of nestlings such as body size. We investigated two mechanisms that could influence allocation of food within broods of Mountain Bluebirds (Sialia currucoides) by filming the size and type of prey that parents fed nestlings. According to the gape size constraint hypothesis, the inability of small nestlings to physically swallow large prey items might place junior brood members at a higher risk of starvation than older siblings if parents do not bring enough small items. Consistent with this idea, the frequency of ‘testing’, placing an item in the gape of a nestling but then withdrawing it, was positively correlated with prey size. The junior nestling of a brood was fed less often when parents brought larger prey items, and nestlings were tested more in the early nestling stage (0–4 days old) when their gapes were smaller than at older stages. Parents of many cavity-nesting species may feed the brood from the nest entrance and the resulting scramble competition among brood members to access the entrance hole also potentially increases the mortality of the least competitive (smallest) nestlings. Although the frequency of feeding from the entrance hole was prevalent during the late nestling stage (≥ 12 days old), feeding from the entrance was not associated with mortality rate in bluebird broods and the mortality of junior nestlings was not linked to the testing of prey. The results confirm that a mechanism of gape size constraint does direct large prey away from junior nestlings, but apparently small-sized prey items were abundant enough during our study so that this did not result in increased mortality of nestlings.

Keywords

Prey allocation Gape size constraint hypothesis Hatching asynchrony 

Zusammenfassung

Die Größe der Nahrung und der Schnabelöffnung beim Sperren beeinflussen beim Berghüttensänger die Verteilung der Nahrung auf die Brut.

Bei Nesthockern könnte die Verteilung des Futters auf die Nestlinge davon abhängen, ob Elterntiere einzelne Jungen bevorzugen. Sie könnte aber auch an der Konkurrenzsituation zwischen den Jungen liegen, die mit Eigenschaften wie z.B. der Körpergröße eines Nestlings zusammenhängt. Wir untersuchten zwei mögliche Mechanismen, die beim Berghüttensänger (Sialia currucoides) bei der Verteilung des Futters an einzelne Nestlinge eine Rolle spielen könnten. Hierfür filmten wir die Größe und Art der Beute, die die Elterntiere an die Nestlinge verfütterten. Entsprechend der „gape-size constraint hypothesis“(Hypothese der Schnabelöffnungs-Größe als einschränkendem Faktor) könnte die rein physikalische Unfähigkeit kleiner Nestlinge, größere Futterbrocken hinunterzuschlucken, für die jüngeren Nestlinge gegenüber ihren älteren Geschwistern ein größeres Risiko zu verhungern bedeuten, wenn die Eltern nicht auch ausreichend viele kleinere Nahrungsstücke beibringen. Im Einklang mit dieser Theorie korrelierte die „Test-Frequenz“, mit der ein Stückchen Futter in den aufgesperrten Schnabel eines Nestlings gelegt und wieder herausgezogen wurde, positiv mit der Größe der Stückchen. Innerhalb einer Brut erhielten die jüngeren Nestlinge seltener Nahrung, wenn die Eltern größere Nahrungsbrocken herantrugen, und die Nestlinge wurden öfter im frühen Nestlingsstadium (0–4 Tage alt) getestet, als ihre Schnäbel noch kleiner als in den späteren Stadien waren. Bei vielen Höhlenbrütern füttern die Eltern ihre Brut vermutlich vom Höhleneingang her, und das dabei entstehende Gedränge und die damit verbundene Konkurrenzsituation unter den Geschwistern, möglichst nahe an das Einflugloch zu kommen, würde potentiell die Sterblichkeitsrate unter den wettbewerbschwächsten (kleinsten) Nestlingen erhöhen. Obwohl die Häufigkeit des Fütterns vom Einflugloch her im späteren Stadium (> 12 Tage alt) vorherrschte, gab es aber keinen Zusammenhang mit der Mortalitätsrate der Berghüttensänger-Jungen, und die Sterblichkeitsrate der jungen Nestlinge hing nicht mit dem Nahrungsstückchen-Test zusammen. Diese Ergebnisse bestätigen, dass aufgrund der Einschränkung durch die Größe der Schnabelöffnung größere Futterbrocken in der Tat nicht bei den jüngeren Nestlinge ankommen, dass es aber während unserer Untersuchung offenbar ausreichend viele kleinere Nahrungsbrocken gab, so dass die größeren Brocken zu keiner höheren Sterblichkeitsrate bei den Nestlingen führte.

Introduction

When incubation begins before clutch completion, nestlings hatch over a span of 1 or more days. Having one or two eggs hatch later than the others in the clutch is common in passerines and this can result in the youngest nestling having reduced growth compared to its older siblings (Soley et al. 2011; Podlas and Richner 2013) and may lead to lower fledging mass or death (Maddox and Weatherhead 2008; Kim et al. 2010). Some of the many hypotheses for hatching asynchrony argue that it is adaptive, whereas others suggest that it is merely the result of energy constraints during incubation (reviews in Magrath 1990; Stenning 1996; Stoleson and Beissinger 1997). Irrespective of the adaptive significance of hatching asynchrony, the gape size of a nestling limits the size of prey that it can swallow (Clark 1995; Wiebe and Slagsvold 2012a). Junior nestlings within broods can then be disadvantaged when their parents bring relatively large prey that are sufficient to meet the energy demands of the larger, senior nestlings (Slagsvold and Wiebe 2007). Sometimes, passerine parents place a prey item into a nestling’s gape but withdraw it when the nestling does not swallow (Glassey and Forbes 2003); this ‘testing’ behaviour increases for insect prey with long wings or legs, perhaps due to the prey item being too large or cumbersome to swallow (Slagsvold and Wiebe 2007). Prey items that are tested but not fed to junior nestlings are often subsequently given to an older sibling that can swallow the prey (Wiebe and Slagsvold 2012a), which may intensify the disadvantage of the junior nestling. Nestlings may even prioritize the growth of their gapes to be more competitive (Gil et al. 2008), but as they age, gape size is predicted to be less of a constraint (García-Navas et al. 2012).

Parents of many bird species may offer food to senior siblings first (Smiseth et al. 2003; Mainwaring et al. 2011; Wiebe and Slagsvold 2012b), or allow these larger, more competitive, offspring to access the favoured positions closest to the feeding parent in scramble competition (Slagsvold and Rohwer 2000). Sometimes patterns of food allocation differ between male and female parents, with males feeding senior offspring and females evenly distributing food or prioritizing the junior nestlings (Budden and Beissinger 2009). For example, Blue Tit (Cyanistes caeruleus) nestlings move towards the male who usually feeds the closest nestling, whereas the female distributes food to nestlings farther away from herself, allowing smaller or less competitive nestlings to eat as well (Dickens and Hartley 2007). Female Blue Tits also deliver more spiders, i.e., high-quality prey, to junior than to senior nestlings, which could help younger offspring grow more quickly (García-Navas et al. 2014). Therefore, if females are attuned to their broods’ needs and have access to sufficient small, nutritious prey they may target these to junior offspring to reduce their mortality. Accordingly, the chance of a junior nestling surviving might depend not only on the delivery rate of prey, but also on the size and nutritional quality of the items (Wright et al. 1998).

Here, we examined two mechanisms that could affect the allocation of prey within broods of Mountain Bluebirds (Sialia currucoides). The gape size constraint hypothesis predicts that prey items will be ‘tested’ more frequently when offered to junior nestlings that have the smallest gapes, and given subsequently to senior siblings. Furthermore, the likelihood of testing should increase with size of the prey and with bulky prey types like grasshoppers and beetles compared to larvae and spiders. To date, the potential constraint of gape size on feeding patterns has only been examined in two very small passerines, Blue Tits (Slagsvold and Wiebe 2007; García-Navas et al. 2012) and Pied Flycatchers (Ficedula hypoleuca) (Wiebe and Slagsvold 2009). Since nestlings of larger-bodied species will hatch with larger gapes, they might not experience similar swallowing constraints. Hence, we were interested in whether gape size also limited swallowing ability in a larger thrush species, the Mountain Bluebird. Elsewhere we have documented that the size and type of insect prey differed between two habitats occupied by bluebirds, grasslands versus clearcuts (Stalwick 2018), so we were also interested in how habitat type might relate to food allocation within broods and perhaps, ultimately, the mortality of the junior nestling. Generally, we expected that younger nestlings would die sooner in broods with a high frequency of testing.

A second mechanism that could influence food allocation within broods of cavity nesters is whether the parents feed young from the entrance or actually enter the nest, where they are more capable of circumventing large and aggressive nestlings begging at the box entrance (Ryser et al. 2016). However, entering the box takes longer to distribute food, so if prey are scarce in a habitat, parents may try to maximize their foraging time and delivery rate by quickly transferring prey to young from the entrance hole instead. Western Bluebirds (Sialia mexicana) fed the hungriest, most intensely begging nestlings, not necessarily the largest ones (Smith et al. 2017), but to our knowledge, no one has examined the relationship between feeding at the nest hole entrance and delivery rates. If females are the sex that direct prey to the junior nestlings in asynchronous broods, we also predicted that they would provision from the nest entrance less than males. In addition, we expected mortality of the junior nestlings to be higher in broods where parents fed from the entrance more frequently.

Methods

Study site and study species

We studied Mountain Bluebirds in central British Columbia during the breeding seasons of 2016 and 2017. Clearcut and grassland sites were located near Bridge Lake (51°28′N, 120°43′W, 1140 m a.s.l.) and grassland sites near 100 Mile House (51°38′N, 121°17′W, 970 m a.s.l.) and Riske Creek (51°58′N, 122°31′W, 986 m a.s.l.). Over 300 plywood nest boxes were placed on retained trees in logging cuts (between 10–45 ha and 1–6 years old) and on fenceposts in grazed grasslands that contained native and introduced grasses and forbs. All boxes were at least 400 m apart. The clearcuts were originally forests that included Douglas fir (Pseudotsuga menziesii), trembling aspen (Populus tremuloides), lodgepole pine (Pinus contorta), and white and hybrid spruce (Picea glauca and Picea glauca × engelmannii).

The Mountain Bluebird is an insectivorous cavity-nesting thrush weighing about 30 g that readily settles in clearcuts despite its natural habitat being grasslands and burns (Johnson and Dawson 2018). Mountain Bluebirds arrive on our study areas after migration around mid-April and lay clutches of from five to six eggs on average. About 50% of broods are synchronous (hatch within 24 h) with the rest hatching within 2 days (Johnson and Dawson 2018). Both sexes provision the offspring with a variety of invertebrate prey, including Lepidoptera (adults and larvae), Arachnida, Coleoptera, Orthoptera, and Hymenoptera (Power 1980; Herlugson 1982), and the nestlings fledge after 15–22 days (Johnson et al. 2013a).

Fieldwork

Once nestlings hatched, adults were captured (n = 123 in 2016, 94 in 2017) using swing-door traps over the entrance hole to the box and were banded with a unique colour combination. Birds were aged as adults versus yearlings based on molt (Pyle 1997), weighed, and the length of six structures measured: head-bill, ninth primary, central rectrix, tarsus, culmen, and flattened wing chord. For each sex, these measures were entered in a principal components analysis (PCA) and the scores on the first axis (PCA1) were used as a measure of body size (Rising and Somers 1989). Body condition was the residual of mass regressed on PCA1 (Labocha and Hayes 2012; Wiebe and Vitousek 2015). Nest boxes were visited every 4–5 days to record nestling survival, and all nestlings were weighed before each filming period or at least three times during the nestling period.

We placed microcameras (either GoPro Hero5 Session or custom-built motion-triggered digital mini cameras) in the ceiling of a nest box to film parental deliveries during 3-h filming blocks. The resolution of the videos was set at 1080 pixels and the roof of the box was replaced with one that had a small 2 × 3-cm ‘window’ covered with translucent plastic to let in enough natural light during filming to see the prey (Fig. 1). From the films we recorded the length and the width of the prey items using the known bill size of the adult as a reference. Prey volume was calculated using the equation of a cylinder: \(\pi \,(0.5{\text{ width}})^{2} \times {\text{length}}\) (Slagsvold and Wiebe 2007). Prey types were categorized by order except for larva and spiders, which were distinct categories. We also recorded whether parents tested a nestling when they placed the item into a gape and which nestling swallowed the prey. Parents always entered the box to feed nestlings when they were younger than 7 days, but when they were older, we also recorded if the parents fed while standing in the entrance hole or entered the nest box. Information on testing and entrance hole feeding was available for each filming event (n = 166 filming periods at 41 boxes in grasslands and 51 boxes in clearcuts). The number of provisioning visits recorded during filming bouts ranged from one to 90 (mean 26), but the few bouts with fewer than five deliveries were discarded.
Fig. 1

The two camera systems used for filming prey deliveries. Top panel GoPro Hero5 Session mounted in the nest box roof, bottom panel custom-built microcamera inserted through a hole in the side of the box and plugged into a small motion-sensitive digital video recorder fastened below the box

To determine if the junior nestling in asynchronous broods was disadvantaged by its relatively small gape size during the early stage of nestling development (Day 0–4), we marked its head with a small dot of white liquid paper for identification in the video and recorded whether it was offered, and ate, prey items. We did not quantify begging intensity because nestlings only raise their heads with an open gape at this age. Nestlings were not marked in synchronous broods in which no nestling weighed measurably less than the others. Sample size at the early nestling stage with a marked nestling was five clearcut and seven grassland boxes in 2016 and 19 clearcut and nine grassland boxes in 2017. In addition to the young nestling stage, we tried to film nest boxes two more times when nestlings were between 5 and 15 days old, on days with no rain, but poor weather or failures of cameras meant some broods were filmed only once with older nestlings. The sample size of filming bouts for all older nestlings (5–15 days old) was 47 films at 19 clearcut boxes and 19 grassland boxes in 2016, and 56 films at 20 clearcut and 15 grassland boxes in 2017.

Statistical analyses

We used R version 3.4.3 (R Core Team 2017) and report data as mean ± SE. Initial models included all two-way interactions but these were eliminated if non-significant (α > 0.05). We ran a general linear model (GLM) to determine if the degree of hatching asynchrony (smallest nestling mass/largest nestling mass) within a brood when offspring were 0–4 days old was associated with habitat type, brood size, laying date, female body condition, female age, nestling age, and year. We used the proportion of mass rather than actual mass differences because not all broods were measured at the same age.

Generalized linear mixed models [GLMMs; library lmerTest (Kuznetsova et al. 2017)] with a binomial distribution were used to analyse whether or not a prey item was tested and whether or not the junior nestling was fed in clearcut and grassland habitats; p-values were obtained using the R library car and the function Anova (Fox and Weisberg 2011). The random effect was nest box in each model. To examine factors associated with testing, the original model had brood size, nestling stage [early (0–4 days old), middle (5–11 days old), and late (≥ 12 days old)], prey volume (log transformed), prey type (larvae, spider, fly, grasshopper, beetle etc.), parent sex, and year as fixed effects. Only visits in which parents fed from inside the box were included in the testing model and the sex of the parent was that of the one that placed prey in the nestlings’ gapes, since males frequently passed food to the female during the early nestling stage. To examine patterns of testing at a finer scale, we used only data from the young nestling stage and constructed a model with habitat type, brood size, nestling age, prey volume (log transformed), prey type, parent sex, identity of the nestling offered the food (junior or not), and year as fixed effects. The GLMM for whether or not the junior nestling was fed had brood size, nestling age, prey volume (log-transformed), prey type, parent sex, and year as fixed effects.

We ran other GLMMs to determine if parents differed in the frequency with which they initially offered prey to the junior nestling using habitat type, brood size, prey volume (log transformed), prey type, parent sex, and year as fixed effects. We analysed which nestling (junior vs. senior) eventually ate the prey after an item was tested using two GLMMs, one where prey items were tested on the junior nestling and one where a senior nestling was tested with habitat type, brood size, nestling age, prey volume (log transformed), prey type, parent sex, and year as fixed effects. To see whether food items were allocated proportionally to junior nestlings, for each brood we calculated the proportion of: (1) tested items, and (2) all prey items that the junior nestling ate relative to that expected by equal allocation among siblings using paired t-tests or Wilcoxon Rank sum tests. The effect of testing frequency in the early nestling stage (0–4 days old) on fledgling success (proportion of the brood that fledged excluding nests that failed completely) was examined with a GLM with habitat type, hatching asynchrony (smallest nestling mass/largest nestling mass), and year as explanatory variables.

The GLMM for whether or not parents delivered items from the entrance hole or entered the box had habitat type, brood size, nestling stage, parent sex, and year as the fixed effects and nest box as the random effect. We tested whether this frequency of feeding from the hole affected fledging success using a GLM with the other explanatory variables of habitat, the proportion of feeding events that took place in the entrance hole, and year. Finally, we did two LMMs [one for the middle nestling stage (5–11 days old) and one for the late nestling stage (≥ 12 days old)] on delivery rates with parent sex, proportion of feedings from the entrance hole, and year as the fixed effects and nest box as the random effect.

Results

Hatching asynchrony and brood size

The average mass of the junior nestlings when measured during the early nestling stage (0–4 days old) was 5.54 g ± 0.26 and the mass of the largest nestling was 7.43 g ± 0.34, so junior siblings were on average 0.75 ± 0.01 the mass of the largest siblings within the first 4 days of hatching. This relative mass difference within broods did not differ between habitats (F1,65 = 3.69, P = 0.06), ages (F1,63 = 0.11, P = 0.74) or years (F1,62 = 1.62, P = 0.21) and was not related to female age or body condition (condition effect: F1,64 = 0.07, P = 0.80). Brood size averaged 5.0 ± 0.1 (n = 83, range 2–7) across all boxes, but it was not significant in any model of hatching asynchrony, testing or food allocation, so for simplicity, it was removed as a factor in all subsequent models.

Testing

Habitat type was not significant in a GLMM with just habitat type as a fixed factor (\(\chi_{1}^{2}\) = 0.04, P = 0.84), nor in any other model of provisioning with other explanatory variables so it was deleted in subsequent models for simplicity. Of 3205 items delivered, 28% were tested and the likelihood of testing was positively correlated with prey volume (Table 1) but did not depend on prey type (taxon). Parents only ate 0.83% of the prey items brought to the nestlings, so self-feeding would not have affected the testing results. Male and female parents tested items at the same rate in the data pooled across nestling stages and likelihood of testing decreased during the nestling period (Fig. 2). The junior nestling was the first target of 19% of provisioning visits, but this frequency did not differ between parents (\(\chi_{1}^{2}\) = 0.30; P = 0.58), prey sizes (\(\chi_{1}^{2}\) = 0.02; P = 0.90), or years (\(\chi_{1}^{2}\) = 0.21; P = 0.65). Since males sometimes passed food items to females, we checked if the sex that delivered the prey item (as opposed to the one that fed the nestling) affected testing, but we found that there was still no significant difference in testing between sexes (GLMM: \(\chi_{1}^{2}\) < 0.01; P = 0.98). Since we were more interested in which nestling the parent was directing the prey to, in subsequent models we used the parent that tested and fed nestlings.
Table 1

Generalized linear mixed models (GLMMs) for prey ‘testing’ (placing item into a gape and then withdrawing it) in Mountain Bluebirds

Dependent variable

Model

Coefficienta ± SE

df

χ 2

P

Overall testingb

Nestling stage

− 1.73 ± 0.13 (Middle)

− 2.65 ± 0.26 (Late)

2

202.89

< 0.001*

Prey volume

0.46 ± 0.05

1

94.03

< 0.001*

Parent sex

− 0.10 ± 0.09 (Male)

1

1.16

0.28

Year

− 0.15 ± 0.14 (2017)

1

1.12

0.29

Early stage testingc

Nestling age

− 0.83 ± 0.22

1

14.12

< 0.001*

Prey volume

0.46 ± 0.09

1

29.37

< 0.001*

Parent sex

0.36 ± 0.18 (Male)

1

4.14

0.042*

Nestling rankd

1.03 ± 0.21 (junior)

1

24.01

< 0.001*

Year

0.10 ± 0.35

1

0.08

0.78

Fed junior

Prey volume

− 0.17 ± 0.08

1

4.08

0.043*

Parent sex

0.06 ± 0.20 (Male)

1

0.09

0.76

Year

0.02 ± 0.20 (2017)

1

0.01

0.91

Rank that ate item after junior testing

Nestling age

0.26 ± 0.25

1

1.10

0.29

Prey volume

− 0.43 ± 0.19

1

5.04

0.025*

Parent sex

− 0.13 ± 0.40 (Male)

1

0.10

0.75

Year

0.07 ± 0.40 (2017)

1

0.03

0.87

Rank that ate item after senior testing

Nestling age

0.84 ± 0.34

1

6.13

0.013*

Prey volume

− 0.49 ± 0.25

1

3.74

0.053

Parent sex

− 0.05 ± 0.51 (Male)

1

0.01

0.91

Year

0.07 ± 0.51 (2017)

1

0.02

0.89

*P < 0.05

aCoefficient estimates (unstandardized) derived from the summary function in R (χ2- and p-values from the Anova function). For categorical variables, coefficients for the category in parentheses refer to the intercept (the category of the variable not in parentheses)

bOverall testing refers to testing that happened across all three nestling stages

cEarly stage testing refers to testing that occurred when nestlings were 0–4 days old

dNestling rank in the early stage refers to whether a junior or senior nestling was initially offered the item first

Fig. 2

The proportion of prey items that were ‘tested’ by male and female Mountain Bluebird parents at each nestling stage (Early 0–4 days old, Middle 5–11 days old, Late ≥ 12 days old). Error bars are SE. The number of boxes filmed was 63 in the early nestling stage, 75 in the middle, and 28 in the late nestling stage

Focusing on the young stage, when most testing occurred, the likelihood of testing decreased from age 0 to 4 days (n = 63 boxes) and was positively associated with prey size (Table 1; Fig. 3). At this early stage, males tested nestlings more than females, but the magnitude of this difference was small since the average proportion of items males tested was 0.41 ± 0.02 and for females was 0.39 ± 0.02. There was more testing if the item was offered first to a junior nestling (0.56 ± 0.04 items tested) than to a senior nestling (0.36 ± 0.02). With data pooled across broods, of the 124 times a junior nestling was tested, it eventually ate the item in 36% percent of cases, whereas the food was transferred to an older nestling otherwise. In contrast, of the 186 times a larger nestling was tested, it ate the item itself in 48% of cases. Thus, there was a trend that junior nestlings lost prey to siblings more often than did senior nestlings (χ2-test: \(\chi_{1}^{2}\) = 3.59; P = 0.06). Within broods at the young stage, the junior nestling was not fed less than what would be expected through random allocation of prey by the parents, either when including only prey that was tested (W = 118.5; P = 0.38) or all prey items regardless of testing (t36 = − 1.83; P = 0.07). However, the junior nestling was fed less often after a nestling was tested (either the junior or a senior nestling) when the prey item was larger, although this relationship was not quite significant when a senior nestling was tested first (Table 1). Junior nestlings were fed more often after a senior nestling was tested when the nestlings were older, but this pattern was not seen when the junior nestling was tested first.
Fig. 3

The effect of nestling age on the average proportion of prey items delivered by Mountain Bluebird parents that were tested during the early nestling stage. Nestling age 0 Day of hatching. Error bars are SE. The number of boxes filmed is given above the bars. Different lowercase letters indicate significant difference (P < 0.05)

Nestling mortality and patterns of feeding

Excluding broods that failed completely because of abandonment or depredation, and considering only first nesting attempts, 28.9% of 38 broods in clearcuts and 24.1% of 58 broods in the grasslands experienced the mortality of at least one nestling. Within these broods, if mortality occurred, typically only one nestling died (12.5% of cases) although a few broods had two (7.3%) or three (6.3%) nestlings die. Most of the nestlings died in the late nestling stage (≥ 12 days old) in both habitats (30 nestlings), whereas only two nestlings died in the early nestling stage (0–4 days old). Although we were only able to sex seven nestlings when they died, three of these were male and four were female, so there was no strong sex bias. The proportion of the brood that fledged was not associated with the proportion of prey that was tested during the early nestling stage (GLM, \(\chi_{1,44}^{2}\) = 1.40, P = 0.24), habitat (\(\chi_{1,45}^{2}\) = 1.12, P = 0.29), hatching asynchrony (\(\chi_{1,43}^{2}\) = 1.93, P = 0.16), or year (\(\chi_{1,42}^{2}\) = 1.19, P = 0.28). Similarly, whether or not mortality occurred within broods did not differ based on habitat (\(\chi_{1,45}^{2}\) = 0.76, P = 0.38), frequency of testing (\(\chi_{1,44}^{2}\) = 0.70, P = 0.40), degree of hatching asynchrony (\(\chi_{1,43}^{2}\) = 0.36, P = 0.55), or year (\(\chi_{1,42}^{2}\) = 1.53, P = 0.22).

Excluding an outlier at 1 day old, the earliest nestling age at which parents started feeding from the hole was 6 days in both grasslands and clearcuts. The films showed that, during the late nestling stage, nestlings were usually sleeping when a parent landed in the entrance hole, and in response to the noise, they stretched their necks towards the parent immediately, jostling with their siblings. The likelihood of parents feeding from the entrance hole did not vary with brood size or with habitat type, but occurred more often in later nestling stages than earlier nestling stages (Table 2; Fig. 4). The habitat × parent sex interaction resulted from females feeding from the entrance hole less often than males in grasslands but not in clearcuts (Fig. 4). The frequency of feeding from the entrance hole was not associated with fledging success in the brood (Table 2). Whether or not mortality occurred in a brood was not associated with habitat type (middle stage: \(\chi_{1,69}^{2}\) = 0.04, P = 0.84; late: \(\chi_{1,25}^{2}\) = 2.48, P = 0.12) or frequency of feeding from the entrance hole (middle: \(\chi_{1,68}^{2}\) = 0.11, P = 0.74; late: \(\chi_{1,24}^{2}\) = 0.30, P = 0.58). Feeding from the entrance hole was not associated with delivery rates during the middle nestling stage but increased entrance feeding was associated with increased delivery rates in the late nestling stage (Table 2).
Table 2

Models for the frequency that parents fed from the entrance hole (entrance feeding; EF) versus inside the nest box, and provisioning rates for Mountain Bluebirds in grassland and clearcut habitats

Dependent variable

Model

Coefficienta ± SE

df

F or χ2

P

Entrance feeding

Habitat type

− 0.83 ± 0.63 (Grass)

1

0.60

0.44

Brood size

0.08 ± 0.18

1

0.19

0.66

Nestling stage

6.87 ± 0.84 (Middle)

9.96 ± 0.88 (Late)

2

241.88

< 0.001*

Parent sex

0.06 ± 0.17 (Male)

1

18.22

< 0.001*

Habitat type × parent sex

0.80 ± 0.23 (Grass and male)

1

11.77

< 0.001*

Fledging proportion

Habitat type

0.35 ± 0.38 (Grass)

1.66

0.85

0.36

Entrance feeding

0.28 ± 0.62

1.65

0.20

0.65

Year

1.31 ± 0.43 (2017)

1.64

10.83

< 0.001*

Delivery rates

Middle stage EF

− 0.06 ± 0.99

1.125

0.003

0.95

Parent sex

− 0.97 ± 0.51 (Male)

1.90

3.70

0.057

Year

− 0.59 ± 0.59 (2017)

1.110

1.01

0.32

Delivery rates

Late stage EF

3.35 ± 1.08

1.39

9.67

0.003*

Parent sex

− 3.05 ± 0.67 (Male)

1.24

20.54

< 0.001*

Year

0.01 ± 0.86 (2017)

1.34

< 0.001

0.99

The EF model is a GLMM and the model for the proportion of the brood that fledged was a GLM and excluded nests that failed completely. Delivery rates were analysed with LMM models. The GLMM and GLM have χ2-statistics and the LMMs have F-values

*P < 0.05

aCoefficient estimates (unstandardized) from the summary function in R. The df, and test statistics, were obtained using the Anova function (or the anova function for LMM models). For categorical variables, the coefficient is calculated for the category in parentheses to compare it to the intercept (the category of the variable that is not indicated in parentheses)

Fig. 4

The average proportion ± SE of feeding events at the entrance hole by Mountain Bluebirds at each nestling stage (Early 0–4 days old, Middle 5–11 days old, Late ≥ 12 days old) separated by habitat. The number of boxes with filming data for clearcuts was 37 for early, 40 for middle, and 12 for the late nestling stage. The number of boxes filmed in grasslands was 26 for early, 35 for middle, and 16 for the late nestling stage

Discussion

Similar to the pattern found in smaller passerines (Slagsvold and Wiebe 2007; Wiebe and Slagsvold 2012a), testing of prey was most likely within the first few days of hatching when the gape size of nestlings was smallest, and the likelihood of testing increased with prey volume, supporting the gape size constraint hypothesis. The relationship between likelihood of testing and the size of the prey relative to the gape shows that testing is not simply the parent trying to find the hungriest brood member. Indeed, junior Mountain Bluebird nestlings consumed fewer large prey items as a result of testing than did older siblings during the early nestling stage (0–4 days old); however, they were not strongly disadvantaged because mortality rate was not associated with the frequency of testing during the early nestling stage. Perhaps because the degree of hatching asynchrony in broods was small (span < 2 days) and the number of small prey was sufficient, the skew of provisioning that resulted from testing was not strong enough to cause many young nestlings to starve in either habitat type. The degree of hatching asynchrony did not differ between clearcuts and grasslands, and since hatching pattern in bluebirds relates to incubation onset (Johnson et al. 2013b), females in the two habitats likely had similar incubation behaviour and perhaps similar energy constraints during laying.

Factors influencing testing

Consistent with predictions of gape size constraints on swallowing ability (Slagsvold and Wiebe 2007), larger prey items were tested more frequently than smaller prey at all stages of the nestling period and the junior nestling was tested more often than older siblings. Gapes widen as nestlings grow enabling them to swallow larger prey and so, similar to a case in Blue Tits (García-Navas et al. 2014), the frequency of testing also declined with nestling age in bluebirds. Hence, the most critical bottleneck for gape size constraints in Mountain Bluebirds occurred within the first 4 days (early nestling stage) and especially within the first 2 days after hatch. In Pied Flycatchers, rates of testing decreased substantially in the first 4 days after hatching as well (Slagsvold and Wiebe 2007).

Although female Mountain Bluebirds, on average, brought smaller prey to the nest than males (Stalwick 2018) and hence should have tested nestlings less, we found the sexes had similar rates of testing. This pattern was complicated because males sometimes passed prey items to the female when she was brooding (personal observation). However, testing rates did not differ according to which sex caught the prey either, so apparently the difference in prey size between male and female parents was not large enough to translate to different rates of testing between the sexes. Parents did not offer food to junior offspring first and, in the early nestling stage, females did not allocate more food to the junior nestling than males, in contrast to some other studies (Budden and Beissinger 2009; Ryser et al. 2016). Hence, parent bluebirds did not appear to be intentionally helping junior offspring. Likewise, Wiebe and Slagsvold (2009) found that female Pied Flycatchers did not preferentially target junior nestlings and that there is a theoretical basis for why parents might not prefer to feed junior offspring (Mock et al. 2011).

When a prey item was small, although the junior nestling might be tested, it was more likely to get the prey item eventually than when the prey was large. This is probably because small items that were tested were not in an optimum position in the parent’s bill for feeding of the nestling but, after repositioning, could then fit into the nestling’s gape and be swallowed. In contrast, if the item was simply too large to fit into the gape of a junior nestling, the only option for the parent was to move onto a larger nestling (Wiebe and Slagsvold 2009). Similar to the situation in Slagsvold and Wiebe (2007), there were few instances of prey transferred from senior to junior nestlings after testing, but this proportion increased when nestlings aged from 0 to 4 days, as the gapes of all nestlings grew and they became more capable of swallowing larger prey. So, the advantage of being a senior nestling was short-lived in our study.

The frequency of testing during the early nestling stage was not associated with the proportion of the brood that fledged, as few junior nestlings died of starvation. Despite higher rates of testing, junior nestlings appeared to consume prey with a frequency expected by equal allocation among siblings, so there was enough small prey that they could eventually swallow. Johnson and Dawson (2018) suggested that junior Mountain Bluebird nestlings might catch up with their older siblings in mass within 1 day of hatching if fed sufficiently. The mortality of junior nestlings in asynchronous broods of other passerines often occurs when nestlings are quite young (Granbom and Smith 2006; Slagsvold and Wiebe 2007; Mock et al. 2009). In contrast, most nestling mortality in our study occurred in the late nestling stage, probably because most broods were fairly synchronous, and because energy demands were higher then. High energy demands of older insectivorous nestlings may also trigger mortality at that stage if cold weather impedes provisioning, like in Tree Swallows (Tachycineta bicolor) (Winkler et al. 2013).

Position of parents during feeding

Parents fed from the entrance hole more often as nestlings grew, as has been found for a few other species like Tree Swallows (Slagsvold and Rohwer 2000) and Syrian Woodpeckers (Dendrocopos syriacus) (Mersten-Katz et al. 2012). Nestlings closest to the entrance hole are fed more often when parents do not completely enter the nest (Leonard and Horn 1996; Smith et al. 2017), which could mean that parents favour the larger or most competitive nestlings (Kacelnik et al. 1995). Equal food distribution among offspring is not always the goal of avian parents, which may prioritize the largest offspring (Smiseth et al. 2003; Mainwaring et al. 2011; Wiebe and Slagsvold 2012b) that may be the highest quality.

The prevalence of feeding from the entrance hole did not differ between habitat types, but, at least in grasslands, the greater propensity of males than females to feed rapidly and from the entrance is a pattern noted in a few other species, including Tree Swallows (Whittingham et al. 2003) and Eurasian Hoopoes (Upupa epops) (Ryser et al. 2016). Budden and Beissinger (2009) found that Green-rumped Parrotlet (Forpus passerinus) males favoured larger, more competitive nestlings, while females took time to provision the smaller, less competitive nestlings more often than males. Similarly, male Blue Tits also fed the closest, more competitive nestling, whereas females distributed food more evenly, although this feeding took place within the nest box (Dickens and Hartley 2007; Dickens et al. 2008). Despite any competition to access the nest box entrance, feeding from the entrance hole did not affect fledging success in Mountain Bluebirds, so junior nestlings apparently still received adequate amounts of food.

An experimental manipulation of the food supply is required to test the idea that feeding from the hole is a direct response to foraging demands on the parent. Brood size was not associated with the frequency of feeding from the entrance, but if parents adjust brood size to their own foraging capacity, such trade-offs may be obscured. However, delivery rates increased with increased frequency of feeding from the entrance during the late nestling stage, possibly indicating that feeding from the entrance hole is a time-saving strategy for Mountain Bluebirds when food demand is highest. Other explanations for feeding from the entrance hole could be that it is a strategy to decrease the risk of predation and parasite infections in the adult (Ryser et al. 2016). Parents may also feed from outside the box simply because of physical constraints of crowding in the box when the nestlings are near fledging. Female Common Kestrels (Falco tinnunculus) had difficulty entering the nest cavity due to overcrowding later in the nestling period (Steen et al. 2012), and the vigorous pushing and jostling of large Tree Swallow nestlings tended to block the entrance hole (Slagsvold and Rohwer 2000).

In sum, we found support for the idea that gape size constraints affect food allocation within broods of bluebirds at young nestling ages. Thus, such constraints can influence food allocation and mortality for insectivorous species as large as 30 g. However, the impact of such constraints on nestling mortality in Mountain Bluebirds was small, perhaps because broods were fairly synchronous and gape constraints operated mostly within the first 2 days of hatching. Despite different insect prey in clearcuts versus grasslands in British Columbia (Stalwick 2018), feeding patterns were similar between these habitat types, but future studies of gape size constraints could focus on patterns of feeding and nestling mortality in areas where prey types are relatively large and bulky and where overall prey availability is lower.

Notes

Acknowledgements

Thanks to the Scherrers, Arendals and Bridge Creek Ranch for access to their properties. S. Srayko helped with insect identification. The study was funded by a Natural Sciences and Engineering Research Council of Canada grant (203177) to K. L. W. This study complied with the current laws of Canada.

Compliance with ethical standards

Ethical approval

All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. All procedures performed in the studies involving animals were in accordance with the ethical standards of the institution and conducted under Animal Care Permit 20160018.

References

  1. Budden AE, Beissinger SR (2009) Resource allocation varies with parental sex and brood size in the asynchronously hatching Green-rumped Parrotlet (Forpus passerinus). Behav Ecol Sociobiol 63:637–647CrossRefGoogle Scholar
  2. Clark AB (1995) Gapes of sexually dimorphic blackbird nestlings do not show sexually dimorphic growth. Auk 112:364–374CrossRefGoogle Scholar
  3. Dickens M, Hartley IR (2007) Differences in parental food allocation rules: evidence for sexual conflict in the Blue Tit? Behav Ecol 18:674–679CrossRefGoogle Scholar
  4. Dickens M, Berridge D, Hartley IR (2008) Biparental care and offspring begging strategies: hungry nestling Blue Tits move towards the father. Anim Behav 75:167–174CrossRefGoogle Scholar
  5. Fox J, Weisberg S (2011) An {R} companion to applied regression, 2nd edn. Sage, Thousand Oaks. http://socserv.socsci.mcmaster.ca/jfox/Books/Companion. Accessed 12 July 2017
  6. García-Navas V, Ferrer ES, Jose Sanz J (2012) Prey selectivity and parental feeding rates of Blue Tits Cyanistes caeruleus in relation to nestling age. Bird Study 59:236–242CrossRefGoogle Scholar
  7. García-Navas V, Ferrer ES, Serrano-Davies E (2014) Experimental evidence for parental, but not parentally biased, favouritism in relation to offspring size in Blue Tits Cyanistes caeruleus. Ibis 156:404–414CrossRefGoogle Scholar
  8. Gil D, Bulmer E, Celis P, López-Rull I (2008) Adaptive developmental plasticity in growing nestlings: sibling competition induces differential gape growth. Proc R Soc B-Biol Sci 275:549–554CrossRefGoogle Scholar
  9. Glassey B, Forbes S (2003) Why Brown-headed Cowbirds do not influence Red-winged Blackbird parent behaviour. Anim Behav 65:1235–1246CrossRefGoogle Scholar
  10. Granbom M, Smith HG (2006) Food limitation during breeding in a heterogeneous landscape. Auk 123:97–107CrossRefGoogle Scholar
  11. Herlugson CJ (1982) Food of adult and nestling Western and Mountain Bluebirds. Murrelet 63:59–65CrossRefGoogle Scholar
  12. Johnson LS, Dawson RD (2018) Mountain Bluebird (Sialia currucoides), version 2.0. In: Rodewald PG (ed) The birds of North America. Cornell Lab of Ornithology, IthacaGoogle Scholar
  13. Johnson LS, Hebert RM, Napolillo FM, Allen A (2013a) The process of fledging in the Mountain Bluebird. J Field Ornithol 84:367–376CrossRefGoogle Scholar
  14. Johnson LS, Napolillo FM, Kozlovsky DY, Hebert RM, Allen A (2013b) Variation in incubation effort during egg laying in the Mountain Bluebirds and its association with hatching asynchrony. J Field Ornithol 84:242–254CrossRefGoogle Scholar
  15. Kacelnik A, Cotton PA, Stirling L, Wright J (1995) Food allocation among nestling starlings: sibling competition and the scope of parental choice. Proc R Soc B-Biol Sci 259:259–263CrossRefGoogle Scholar
  16. Kim M, Furness RW, Nager RG (2010) Hatching asynchrony is constrained by parental nest attendance during laying. Behav Ecol Sociobiol 64:1087–1097CrossRefGoogle Scholar
  17. Kuznetsova A, Brockhoff PB, Christensen RHB (2017) lmerTest package: tests in linear mixed effects models. J Stat Softw 82:1–26CrossRefGoogle Scholar
  18. Labocha MK, Hayes JP (2012) Morphometric indices of body condition in birds: a review. J Ornithol 153:1–22CrossRefGoogle Scholar
  19. Leonard M, Horn A (1996) Provisioning rules in Tree Swallows. Behav Ecol Sociobiol 38:341–347CrossRefGoogle Scholar
  20. Maddox JD, Weatherhead PJ (2008) Egg size variation in birds with asynchronous hatching: is bigger really better? Am Nat 171:358–365CrossRefGoogle Scholar
  21. Magrath RD (1990) Hatching asynchrony in altricial birds. Biol Rev Camb Philos 65:587–622CrossRefGoogle Scholar
  22. Mainwaring MC, Lucy D, Hartley IR (2011) Parentally biased favouritism in relation to offspring sex in Zebra Finches. Behav Ecol Sociobiol 65:2261–2268CrossRefGoogle Scholar
  23. Mersten-Katz C, Barnea A, Yom-Tov Y, Ar A (2012) The woodpecker’s cavity microenvironment: advantageous or restricting? Avian Biol Res 5:227–237CrossRefGoogle Scholar
  24. Mock DW, Schwagmeyer PL, Dugas MB (2009) Parental provisioning and nestling mortality in House Sparrows. Anim Behav 78:677–684CrossRefGoogle Scholar
  25. Mock DW, Dugas MB, Strickler SA (2011) Honest begging: expanding from signal of need. Behav Ecol 22:909–917CrossRefGoogle Scholar
  26. Podlas KA, Richner H (2013) The adaptive function of hatching asynchrony: an experimental study in Great Tits. Anim Behav 86:567–576CrossRefGoogle Scholar
  27. Power HW (1980) The foraging behavior of Mountain Bluebirds with emphasis on sexual foraging differences. Ornithol Monogr 28:1–72Google Scholar
  28. Pyle P (1997) Identification guide to North American birds. Part 1. Slate Creek Press, BolinasGoogle Scholar
  29. R Core Team (2017) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. https://www.R-project.org/. Accessed 12 July 2017
  30. Rising JD, Somers KM (1989) The measurement of overall body size in birds. Auk 106:666–674CrossRefGoogle Scholar
  31. Ryser S, Guillod N, Bottini C, Arlettaz R, Jacot A (2016) Sex-specific food provisioning patterns by parents in the asynchronously hatching European Hoopoe. Anim Behav 117:15–20CrossRefGoogle Scholar
  32. Slagsvold T, Rohwer S (2000) Behavioral domination of food delivery by Tree Swallow nestlings. Wilson Bull 112:278–281CrossRefGoogle Scholar
  33. Slagsvold T, Wiebe KL (2007) Hatching asynchrony and early nestling mortality: the feeding constraint hypothesis. Anim Behav 73:691–700CrossRefGoogle Scholar
  34. Smiseth PT, Bu RJ, Eikenaes AK, Amundsen T (2003) Food limitation in asynchronous Bluethroat broods: effects on food distribution, nestling begging, and parental provisioning rules. Behav Ecol 14:793–801CrossRefGoogle Scholar
  35. Smith MG, Dickinson JL, Rush A, Wade AL, Yang DS (2017) Western Bluebird parents preferentially feed hungrier nestlings in a design that balances location in the nest. Behav Ecol Sociobiol 71:58–64CrossRefGoogle Scholar
  36. Soley N, Siefferman L, Navara KJ, Hill GE (2011) Influence of hatch order on begging and plumage coloration of nestling Eastern Bluebirds. Wilson J Ornithol 123:772–778CrossRefGoogle Scholar
  37. Stalwick JA (2018) Provisioning patterns, diet, and reproduction of Mountain Bluebirds (Sialia currucoides) in clearcut versus grassland habitats. Dissertation, University of SaskatchewanGoogle Scholar
  38. Steen R, Sonerud GA, Slagsvold T (2012) Parents adjust feeding effort in relation to nestling age in the Eurasian Kestrel (Falco tinnunculus). J Ornithol 153:1087–1099CrossRefGoogle Scholar
  39. Stenning MJ (1996) Hatching asynchrony, brood reduction and other rapidly reproducing hypotheses. Trends Ecol Evol 11:243–246CrossRefGoogle Scholar
  40. Stoleson SH, Beissinger SR (1997) Hatching asynchrony, brood reduction, and food limitation in a neotropical parrot. Ecol Monogr 67:131–154CrossRefGoogle Scholar
  41. Whittingham LA, Dunn PO, Clotfelter ED (2003) Parental allocation of food to nestling Tree Swallows: the influence of nestling behaviour, sex and paternity. Anim Behav 65:1203–1210CrossRefGoogle Scholar
  42. Wiebe KL, Slagsvold T (2009) Parental sex differences in food allocation to junior brood members as mediated by prey size. Ethology 115:49–58CrossRefGoogle Scholar
  43. Wiebe KL, Slagsvold T (2012a) Brood parasites may use gape size constraints to exploit provisioning rules of smaller hosts: an experimental test of mechanisms of food allocation. Behav Ecol 23:391–396CrossRefGoogle Scholar
  44. Wiebe KL, Slagsvold T (2012b) Parents take both size and conspicuousness into account when feeding nestlings in dark cavity nests. Anim Behav 84:1307–1312CrossRefGoogle Scholar
  45. Wiebe KL, Vitousek MN (2015) Melanin plumage ornaments in both sexes of Northern Flicker are associated with body condition and predict reproductive output independent of age. Auk 132:507–517CrossRefGoogle Scholar
  46. Winkler DW, Luo MK, Rakhimberdiev E (2013) Temperature effects on food supply and chick mortality in Tree Swallows (Tachycineta bicolor). Oecologia 173:129–138CrossRefGoogle Scholar
  47. Wright J, Both C, Cotton PA, Bryant D (1998) Quality vs. quantity: energetic and nutritional trade-offs in parental provisioning strategies. J Anim Ecol 67:620–634CrossRefGoogle Scholar

Copyright information

© Dt. Ornithologen-Gesellschaft e.V. 2018

Authors and Affiliations

  1. 1.Department of BiologyUniversity of SaskatchewanSaskatoonCanada

Personalised recommendations