Hatching Asynchrony and Growth Trade-Offs Within Barn Swallow Broods

Abstract.

Hatching asynchrony results in age and size hierarchies within broods, and the subsequent asymmetric competition among siblings has important consequences for nestlings' fitness. In this study, we compare the growth of Barn Swallow (Hirundo rustica) nestlings in relation to their order of hatching. The aim was to test the prediction that early-hatched nestlings develop differently from late-hatched nestlings, which should be under greater pressure to trade investment in growth in favor of traits important to simultaneous fledging. Early-hatched nestlings were always larger than late-hatched nestlings, but when the age difference was taken into account, the two classes of nestling gained mass and head-bill length in similar ways, including having similar asymptotes, as predicted by nonlinear curve models. For wing length, however, late-hatched nestlings reached the inflection point of growth sooner than early-hatched nestlings, and although scaled rates of wing growth were similar, early-hatched nestlings had significantly longer wings, both before fledging, when the oldest nestling was 14 days old, and as suggested by the asymptotic, age-independent values derived from the nonlinear curve models. This finding suggests that nestlings hatched later preferentially develop body mass and the skeleton at the expense of wing feathers. As swallows rely on their wings for foraging and avoiding predators, this pattern of resource allocation is likely to have negative consequences for the late-hatched nestlings.

Resumen.

La asincronía en la eclosión de los huevos conduce a la existencia de jerarquías relacionadas con la edad y el tamaño dentro de las parvadas. La competencia asimétrica resultante entre los pichones tiene consecuencias importantes para su adecuación. En este estudio, comparamos el crecimiento de pichones de Hirundo rustica en relation con su position en el orden de eclosión. El objetivo fue poner a prueba la prediction de que los pichones que eclosionan más temprano se desarrollan de modo diferente que aquellos que eclosionan más tarde, los cuales deben estar más presionados para reducir la inversión en crecimiento, favoreciendo los rasgos que son importantes para el emplumamiento simultáneo. Los pichones que eclosionaron temprano siempre fueron más grandes que los que eclosionaron tarde. Sin embargo, tras tener en cuenta las diferencias en edad, las dos clases de pichones adquirieron masa y aumentaron su longitud de cabeza—pico de modos similares, incluyendo asíntotas similares predichas por modelos de curvas no lineales. En cambio, los pichones que eclosionaron tarde alcanzaron el punto de inflexión en el aumento de la longitud del ala antes que los pichones que eclosionaron temprano. Además, aunque las tasas de crecimiento de las alas a escala fueron similares, los pichones que eclosionaron tarde presentaron alas significativamente más largas tanto antes del emplumamiento (cuando el pichón más viejo tenia 14 días de edad) como con base en lo sugerido por los valores asintóticos e independientes de la edad que fueron derivados de los modelos no lineales. Este hallazgo indica que los pichones que eclosionan más tarde preferencialmente desarrollan su masa corporal y el esqueleto, a expensas de las plumas de las alas. Como las golondrinas dependen de sus alas para forrajear y evitar a los depredadores, es probable que este patrón de asignación de recursos tenga consecuencias negativas para los pichones que eclosionan tarde.

Introduction

Hatching asynchrony creates age and size hierarchies within broods and puts the later-hatched nestlings at a disadvantage in the competition for food provided by parents (Magrath 1990, Stoleson and Beissinger 1995, Stenning 1996). There is debate regarding whether hatching asynchrony evolved as a selected trait (e.g., brood-reduction hypothesis; Lack 1947) or as a side effect of selection for other traits (e.g., hurry-up hypothesis; Clark and Wilson 1981). Irrespective of the selective forces, the pattern in which nestling birds hatch is clearly controlled by their parents through the timing of the onset of incubation (Clark and Wilson 1981, Magrath 1990). Nestlings hatched earlier represent the minimum subset that parents can or will rear and vary little in growth and survival, while nestlings hatched later (runts) are much more susceptible to extrinsic and intrinsic variation in developmental conditions (review in Glassey and Forbes 2002). Most empirical studies have found that early hatching leads to superior survival (Mock and Forbes 1995, Glassey and Forbes 2002). Even when maternal manipulations influence nestlings' early growth rate, the effects are overridden by the size asymmetry caused by hatching asynchrony (Glassey and Forbes 2002).

Size differences within broods may strongly influence how parental care is distributed among siblings (e.g., Bengtsson and Rydén 1983, Parker et al. 1989, McRae et al. 1993, Leonard and Horn 1996, review in Glassey and Forbes 2002), and late-hatched offspring rarely grow fast enough to catch up with their older siblings. When parents respond to the competitive interactions of offspring (Royle et al. 2002), first-hatched and larger nestlings dominate and, during begging contests, often are able to secure a greater share of the food delivered by parents, primarily by reaching higher than their younger and smaller siblings (Smith and Montgomerie 1991, Teather 1992, Leonard and Horn 1996, Dearborn 1998). Nestlings hatched earlier reach key developmental landmarks earlier, including sight, sclerotised (hardened) legs, thermoregulation, and strength in the neck and gastrocnemius muscles (Glassey and Forbes 2002). All of these developmental advantages compound the competitive advantage of first-hatched nestlings over their younger and smaller siblings, enabling the older chicks to initiate begging sooner and maintain it for longer.

Alternatively, parents may choose to feed later-hatched nestlings preferentially (Godfray 1995, Kilner and Johnstone 1997), affecting the developmental priorities of late-hatched nestlings (Glassey and Forbes 2002). Subordinate nestlings, however, have been shown to trade off development of body mass in favor of wing growth (Zach 1982a, Nilsson and Svensson 1996, Lago et al. 2000, Nilsson and Gårdmark 2001). This prioritization of wing growth is possibly due to the need to fledge at the same time as the majority of the brood because once fledging has begun, parents' provisioning of young remaining in the nest decreases dramatically (Nilsson 1990). Late-hatched nestlings that fledge at relatively low body mass are, however, at high risk of early post-fledging mortality, suggesting a trade-off between growth allocation and survival (Nilsson 1990).

The aim of our study was to compare the growth patterns of early- and late-hatched nestlings in broods of the Barn Swallow (Hirundo rustica). Because late-hatched nestlings are generally smaller than their early-hatched siblings throughout the nestling period, we compared the growth patterns of morphological traits while controlling statistically for differences due purely to age. Compensating for age enabled us to search for evidence of trade-offs among traits and to shed light on whether nestlings hatched later are able to redress their initially bad position by using growth strategies that enable them to catch up to their more competitive siblings. In the Barn Swallow, the female starts to incubate after laying the penultimate or ultimate egg (Møller 1994). After 15 days of incubation, the nestlings hatch over a period of 1–3 days, often resulting in a clear size hierarchy within a brood. Nestlings hatched from the eggs laid last in a clutch (Saino et al. 2001) are therefore phenotypically disadvantaged, despite their eggs being relatively large (Saino et al. 2000). Although Barn Swallow nestlings show no significant sexual dimorphism in mass, size, or a number of physiological traits (Saino et al. 2002), there is evidence of sexual differences in begging behaviors and mouth coloration (Saino et al. 2003, 2008). Fledging usually takes place at 18–20 days of age, after which the nestlings become progressively more independent. Given that a previous study demonstrated a plastic growth response to ectoparasite abundance (Saino et al. 1998), individual nestlings should be expected to vary in growth patterns according to their hierarchical position within the brood. Therefore, our aim was to test the hypothesis that nestlings that hatch first and together develop differently from those that hatch later because the later-hatched nestlings are under greater pressure to trade off growth in mass in favor of traits important to simultaneous fledging. Early-hatched nestling Barn Swallows beg less intensely yet receive more food than their laterhatched siblings (Saino et al. 2001), implying that their parents respond to competition among offspring. Consequently, we predict that later-hatched nestlings prioritize wing growth at the expense of body mass (Zach 1982a, Nilsson and Svensson 1996, Lago et al. 2000, Nilsson and Gårdmark 2001) in order to fledge simultaneously with their older siblings.

Methods

From April to August 2005 we observed Barn Swallows breeding on farms surrounding Esthwaite Water in Cumbria, UK (54° 21′ N, 02° 59′ W). The study area consisted of upland hay meadows managed for silage production and livestock grazing. Barn Swallows nested in barns, sheds, and outbuildings associated with the farms.

To search for nests we visited the farms weekly from the beginning of April. Once located, nests were checked every 2–3 days to establish the timing of reproduction; we backcalculated the date of the first egg by assuming one egg was laid per day (Møller 1994). If we knew the date the first egg had been laid, we checked the nest for hatching in the morning around a date predicted from a 13-day incubation; if we did not know the date of the first egg, we checked the nest daily. On the day of hatching, we gave each nestling in a brood a unique marking on its back with indelible ink, and 6 days after hatching we fitted each with an individually numbered band. At 2-day intervals from hatching to day 14, we measured nestlings' body mass (±0.1 g with an electronic balance), head-bill length, gape length, gape width, tarsus (from the depression in the angle of the joint between the tibia and tarsus to the end of the folded foot) (±0.01 mm with dial calipers), and length of the right wing, right fourth primary, and right central rectrix (±0.5 mm with a fixed rule). We followed the same route from nest to nest every day to ensure that broods were weighed and measured at approximately the same time of day. All measurements were taken by one observer (MCM). To avoid the risk of premature fledging, we did not visit nests after the earlyhatched nestlings reached day 14 but checked the nests around day 20 (± 1) to establish fledging success.

Statistical Analysis

We found a total of 44 nests but excluded 12 from the analyses because either the whole brood died before reaching an age of 14 days or only one nestling hatched. The 95 early-hatched nestlings were defined as those that hatched on the first day of hatching, and the 25 late-hatched nestlings were defined as those that hatched on a later day (Mock and Forbes 1995). The number of late-hatched nestlings is lower than the number of broods included in the analyses because several broods hatched on a single day. Nestlings' growth parameters were derived iteratively by fitting logistic growth curves in nonlinear regression in the statistical package S-PLUS, version 6. We fitted logistic models of the form size ∼ a/(1 + exp[-k × (age - b)]), where a is the asymptote that the nestling's growth approaches, b is the inflection point on the time axis in which growth changes from accelerating to decelerating, and k is the constant scale parameter for rate of growth (Pinheiro and Bates 2000, Remeš and Martin 2002). This method has been criticized previously on the grounds that the curve should be replaced by a more complex curve that accounts for the shape as well as the growth rate (Brisbin et al. 1987). However, these logistic growth curves have been used traditionally as they produce only three parameters (a, b, and k) that are readily interpretable biologically (Ricklefs 1968, Remeš and Martin 2002, Roff et al. 2005). Logistic growth curves provide estimates of the asymptote of the size values (a in the model) for individual nestlings, but because of the age differences created by hatching asynchrony, the measurements of the later-hatched nestlings may not encompass the later leveling off, or even recession, of the curve and so overestimate the asymptotic values of later-hatched nestlings. This problem is particularly prevalent among aerial foragers (Remeš and Martin 2002), so we analyzed the pre-fledging sizes (day 14 for the early-hatched nestlings) as well as the asymptotic values estimated from the growth curves. We estimated the age of individual nestlings by visiting nests on a daily basis around the time of hatching and by assuming that nestlings hatched at 06:00; Barn Swallow nestlings hatch between 04:00 and 08:00 (Turner 2006). Consequently, we calculated the growth parameters for early- and late-hatched nestlings by nonlinear regression and compared them with linear mixed-effects models. We used mixed models as they allow both fixed and random terms to be defined (Crawley 1993). Random terms allow the analysis to account for repeated measures or related data; brood identity was fitted as a random term in this case. All statistical tests are two-tailed, means are presented ±1 standard error, and a critical P-value of 0.05 is applied throughout.

Results

A principal-components analysis of all of the data revealed that of the seven characters recorded, mass, head-bill length, and wing length explained 98.4% of the variation in nestlings' growth. Consequently, further analyses considered only the growth of these characters, which represent body size, skeletal growth, and feather growth, respectively.

The inflection points, scale parameters, and asymptotes of the growth trajectories of both mass gain and head-bill growth of early- and late-hatched Barn Swallow nestlings did not differ significantly (Table 1, Figs. 1 and 2). Given that early- and late-hatched nestlings grew in a similar way, it is not surprising that when the early-hatched nestlings were 14 days old, they were significantly heavier (4%) and had larger (2%) head—bill lengths than the younger, later-hatched nestlings (Table 1, Figs. 1 and 2).

Wing growth of late- and early-hatched nestlings differed. Their scale parameter did not differ significantly, but in contrast to early-hatched nestlings, late-hatched nestlings reached the inflection point of growth at a significantly earlier age and had a lower (6%) mean asymptote (Table 1, Fig. 3). When the early-hatched nestlings were 14 days old, they had significantly longer (10%) wings than their younger siblings (Table 1). This result suggests that early- and late-hatched Barn Swallow nestlings have different modes of wing development and that for late- hatched nestlings wing development is a lower priority than skeletal growth or mass gain.

Discussion

Ultimately, the nonlinear-curve models predicted that when fledging, Barn Swallow nestlings that hatched early would be larger and heavier than their later-hatched siblings, so throughout the nestling period later-hatched nestlings were at a competitive disadvantage when begging from provisioning parents. Furthermore, our data show that late- and earlyhatched nestlings did not put on mass or grow their skeleton in a way significantly different from each other, yet the feathers of late-hatched nestlings slowed in growth sooner and were shorter at fledging than those of early-hatched nestlings. So the evidence suggests that late-hatched nestlings not only failed to catch up with the development of their older siblings but also suffered a lack of feather development as a consequence of their position in the hierarchy. This further suggests that in a trade-off among the development of the skeleton, body mass, and feathers, feathers are likely to be the least important for a growing Barn Swallow nestling.

This finding contradicts the original prediction that laterhatched nestlings prioritize wing growth at the expense of body mass in order to fledge simultaneously with their older siblings. Our findings, however, are consistent with those hypotheses that view hatching asynchrony as a selected trait (e.g., brood-reduction hypothesis; Lack 1947) but not with those that view size hierarchy as a side effect of selection for other traits (e.g., hurry-up hypothesis; Clark and Wilson 1981), as late-hatched nestlings failed to catch up with the development of their early-hatched siblings. This disparity contravenes the fact that within a Barn Swallow clutch both egg mass (Saino et al. 2004) and albumen content (Ferrari et al. 2006) increase with laying and hatching order. Consequently, Barn Swallows are among those species pursuing a “brood-survival” strategy, which means that they may at least partly compensate for the initial disadvantage of late-hatched nestlings by, for example, increasing egg size with laying order (Stoleson and Beissinger 1995, Stenning 1996). In contrast, a decline in egg size with laying sequence allows for parents to adopt a “broodreduction” strategy (Lack 1947, Clark and Wilson 1981). Like ours, a study of the Tree Swallow (Tachycineta bicolor) also found that later-hatched nestlings were unable to match earlier-hatched nestlings in either body mass or feather growth, despite this species also employing a “brood-survival” strategy (Zach 1982a). A sex-biased hatching order coupled with sex-specific growth patterns may explain our results, although laying order does not predict the sex of Barn Swallows, presumably because there is no sexual dimorphism in mass, size, or a number of physiological traits (Saino et al. 2002). Although there is evidence of sex-specific begging behaviors and mouth coloration in Barn Swallow nestlings (Saino et al. 2003, 2008), there should be little selection for females to bias the sex of last laid, marginal nestlings.

At fledging, Barn Swallows have wings approximately 80% the length of the adults', and fledglings can attain adult size by about day 33 (Turner 2006). Moreover, juvenile Barn Swallows undergo a complete molt on the wintering grounds before returning to the breeding range the following spring (Møller 1994, Turner 2006). Such opportunities for compensatory growth might allow young birds to redress any deficiencies in their plumage state (Metcalfe and Monaghan 2001), whereas deficiencies in skeletal growth cannot be remedied because bones ossify in the first few weeks after hatching (Schew and Ricklefs 1998). Alternatively, our results might indicate that in this species wing growth is plastic, as two previous studies have found that in nests where dipteran ectoparasites were added, nestlings developed wings significantly more quickly than control nestlings in order to fledge sooner and avoid the ectoparasites (Møller 1994, Saino et al. 1998). Moreover, the availability of a nutrient required for the development of a specific morphological character may facilitate the development of one character rather than an alternative. For example, a diet ample in protein may enhance feather development, whereas one ample in calcium may enhance bone development. The majority of studies, however, have demonstrated that late-hatched nestlings are not able to compensate fully for hatching last and usually compromise body mass in favor of wing and/or feather development, as in the House Wren (Troglodytes aedon; Zach 1982a, 1982b, Lago et al. 2000), American Goldfinch (Carduelis tristis; Skagen 1987), Tree Swallow (Johnson et al. 2003), and Collared Flycatcher (Ficedula albicollis; Rosivall et al. 2005). These findings from a diverse array of species and our results support the idea that late-hatched nestlings are not able to compensate fully for hatching last but differ in terms of the morphological character compromised.

Avian growth consists of the coordinated development of a number of structures, and several studies have demonstrated that developmental plasticity can accommodate trade-offs between these structures. We found that late-hatched nestlings develop body mass and skeletal characters similarly to earlierhatched nestlings. Late-hatched nestlings, however, did not match early-hatched nestlings in terms of wing development, which may indicate that during early development either body mass and skeletal development are of highest importance or that wing development is of less importance. Late-hatched nestlings being able to maintain body mass and skeletal development may be the result of Barn Swallows employing the brood-survival strategy. Late-hatched nestlings' different patterns of growth are likely to have both benefits and costs. A benefit of maintaining body mass, and therefore body size, is that late-hatched nestlings are able to compete effectively with their siblings within the nest and thereby gain access to food from provisioning parents by jostling or reaching (e.g., McRae et al. 1993, Lotem 1998, Dickens et al. 2008). Alternatively, late-hatched nestlings may maintain similar masses to avoid being the “tasty” nestling. The “tasty-nestling” hypothesis suggests that in the coevolutionary arms race between hosts and their parasites, parents may maximize their reproductive success by sacrificing one or more late-hatched nestlings to minimize parasite attacks on the early-hatched nestlings (Simon et al. 2003). Support for this hypothesis comes from a study of the Blue Tit (Cyanistes caeruleus), in which the nestlings' ectoparasite load was found to be negatively correlated with their mass (Simon et al. 2003). These findings also concur with the idea that although growth rates are plastic (Schew and Ricklefs 1998), there is a strong genetic component to asymptotic body size (van Noordwijk and Marks 1998).

Although maintaining body mass may facilitate effective competition in the nest, compromising wing development is likely to have costs, as demonstrated by Saino et al. (1998), who found that when dipteran ectoparasites were added to nests, nestlings developed their wings significantly more quickly than control nestlings in order to fledge sooner and avoid the ectoparasites. Moreover, stunting wing growth may reduce flight maneuverability and escape speed. Aerial insectivores must fly efficiently as soon as they fledge in order to feed and avoid aerial predators such as hawks (Turner 2006). Fledglings, however, may not need full-sized wings immediately as they are provisioned by their parents for at least 10 days after fledging (Medvin and Beecher 1986). In addition, they are often accompanied in flight by a parent (Medvin and Beecher 1986), which could reduce the risk from aerial predators. Thus late-hatched nestlings may use the post-fledging period to catch up with wing growth before they must begin foraging and avoiding aerial predation on their own. Nevertheless, such compensatory growth is likely to have a detrimental effect on an individual's ability to complete the trans-Saharan migration to the winter range (Møller 1994, Turner 2006). Given that conditions experienced during early development largely determine social dominance as an adult (Lindström 1999), the long-term consequences of alternative growth strategies need to be considered.

Acknowledgments

We are grateful to Pip Barr, Mike Brass, Gary Dixon, Alistair Irvine, Myles Sandys, Chris Taylforth, Gary Thomason, Richard Walker, and Mike Woodhouse for allowing us access to Barn Swallow nests on their land, to Ken Wilson, Ian Owens, Donald Dearborn, Scott Forbes, Michael Patten, Stephanie Strickler, and anonymous referees for useful comments on previous versions of the manuscript; and to the Natural Environment Research Council for funding a standard grant (NEC00034x/1) and a studentship to Mark C. Mainwaring(NER/S/A.2003/11263).

Literature Cited