Behavioral Ecology Advance Access originally published online on September 29, 2006
Behavioral Ecology 2007 18(1):62-70; doi:10.1093/beheco/arl051
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Interval between clutches, fitness, and climate change
Laboratoire de Parasitologie Evolutive, CNRS UMR 7103, Université Pierre et Marie Curie, Bât. A, 7ème étage, 7 quai St Bernard, Case 237, F-75252 Paris Cedex 05, France
Address correspondence to A.P. Møller. E-mail: amoller{at}snv.jussieu.fr.
Received 2 February 2006; revised 4 July 2006; accepted 14 August 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Timing of optimal reproduction can be affected by the presence of multiple broods, with multi-brooded species breeding earlier (and later) than the optimal timing of breeding as compared with single-brooded species that only need to optimize the timing of a single brood. Approximately two-thirds of barn swallows Hirundo rustica produce 2 broods per year, and I tested whether the constraints on timing of reproduction were affected by climate change because climatic amelioration would allow both an earlier start and a later termination of reproduction. The duration of the interval between first and second clutches and the variance in the duration increased during 1971–2005 when temperature during spring, but not summer, increased rapidly. Interclutch interval was shorter when mean date of breeding was late and also among late-breeding individuals during individual years. When clutch size and brood size of the first clutch were large, interval until the second brood increased. Pairs with a long interval produced more fledglings than pairs with a short interval. Pairs with first broods with strong mean T-cell–mediated immune responses took shorter time to start their second clutch, whereas mean body mass or tarsus length of first broods were not significantly related to interclutch interval. Interclutch interval increased with the size of a secondary sexual character, the length of the outermost tail feathers of adult male barn swallows, but not with tail length of females, or with size of several other phenotypic characters in either sex. These findings are consistent with the hypothesis that the duration of the interclutch interval is determined by a combination of environmental conditions, reproductive effort, and sexual selection.
Key words: climate change, cost of reproduction, Hirundo rustica, life history, optimal timing of reproduction.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Organisms have to optimize the timing of their reproduction in seasonal environments, where resource abundance often has a seasonal peak. Lack (1954)
suggested that animals would be selected to reproduce at a time that coincided with maximum availability of resources for the offspring. Therefore, clutch size should covary with proximity to this optimal time of reproduction. In species with multiple broods during a single breeding season, animals may be unable to produce several broods so they coincide with the single seasonal peak of resources. Such constraints on optimization of timing of reproduction due to multiple broods should result in first broods often being produced relatively early and subsequent broods relatively late compared with peak availability of resources for rearing offspring. Consistent with this hypothesis, Crick et al. (1993) found that clutch size in single-brooded bird species tended to decline with date, whereas double- or multi-brooded species had a seasonal peak of clutch size during the middle of the breeding season.
Individuals of species with parental care usually only rear a single clutch at a given time, although overlap between clutches can be achieved when males provide parental care for the first clutch, whereas females initiate the second clutch (e.g., Verhulst et al. 1997; Wheelwright et al. 2003). There is usually selection for early breeding even in multi-brooded species, with early clutches producing more recruits than late clutches (e.g., Møller 1994; Sheldon et al. 2003; Dunn 2004). The production of two or more clutches in a season, even for late breeders, can be achieved by reducing the duration of the interval between the first and the second clutch. However, given that elevated parental effort induces costs of reproduction in general (Roff 1992; Stearns 1992) and in birds in particular (e.g., Lindén and Møller 1989; Erikstad et al. 1998; Dhondt 2001), due to the physiological effects of effort on a range of health parameters such as disease and parasitism, oxidative stress, and reductions in storage of essential nutrients for reproduction (Møller 1993a, 1997; von Schantz et al. 1999; Blount et al. 2000, 2003), only individuals in prime condition will be able to produce successive broods at short intervals. Previous experiments on birds have shown that an increase in the abundance of parasites in first clutch nests causes a delay in the production of the second clutch (Møller 1990a). An increase in size of the first clutch also delays the start of the second clutch (Tinbergen and Sanz 2004). Furthermore, simultaneous manipulation of parasite load and reproductive effort (the latter through brood size manipulation) had a multiplicative effect on the duration of the interclutch interval (Møller 1993a). This implies that the interval between clutches can respond to changes in environmental conditions, as reflected not only by the manipulation of parasite loads but also by an experimental increase in parental effort. Although experiments can demonstrate such trade-offs between life-history traits, phenotypic patterns of life-history variation often show effects of condition, with individuals in prime condition being able to lay early, lay many eggs, and still survive better than individuals in poor condition (van Noordwijk and de Jong 1986). When environmental conditions change, individuals of different phenotypic quality may be affected differently, with individuals in poor condition benefiting the most when the environment ameliorates.
The most striking current change in environmental conditions is that due to changing climate. Climate change is occurring at an unprecedented pace, with mean temperature having increased by 0.6 °C during the last century, and even larger changes are predicted for the current century (Houghton et al. 2001). This increase in temperature has caused the duration of the growing season in several parts of the temperate and arctic zones to increase, providing a benefit to reproducing animals in terms of the possibility of reproduction during a longer time period. There is considerable empirical information suggesting that animals have adjusted phenotypically or microevolutionarily to such climate change through adjustments in timing of reproduction, clutch size, or egg size (review in Dunn 2004). These adjustments also work at a much finer scale such as through changes in the timing of start of incubation, as demonstrated for great tits Parus major in England (Cresswell and McCleery 2003), or changes in the interval between spring arrival of migratory birds and start of reproduction (Both and Visser 2001; Møller 2004). However, several studies also suggest that climate change has been the cause of mistiming of reproduction by birds because the cues that are used for initiating reproduction no longer provide reliable information about the stage of phenology and change in phenology of the main food (Visser et al. 1998, 2004; Nielsen and Møller 2006).
Here I analyze how changes in environmental conditions, as reflected by climate change, have affected the duration of the interclutch interval, using the multi-brooded barn swallow Hirundo rustica as a model system. First, I tested whether interclutch interval increased during 1971–2005, when temperatures increased during early spring, causing an increase in the duration of the growing season and hence in the duration of the period during which breeding might be possible. A longer breeding season would also increase the variance in duration of the interclutch interval because poor-quality individuals with long intervals would be able to produce a second clutch at a later date than they would under more severe environmental conditions.
Second, interclutch interval could be predicted by life-history traits such as timing of reproduction, clutch size, and brood size. If the duration of the interval is optimized depending on costs and benefits of rapid production of a second clutch, we should expect the interval to be longer in years when reproduction is early due to an increase in the duration of the growing season and among early reproducing individuals. In addition, we should expect the interval between clutches to be longer when the first clutch and in particular the first brood was large because a large clutch or brood would imply large parental effort. In contrast, if differences in individual quality accounted for the data (if there are inherent quality differences among individuals), we should expect high-quality individuals to be able to produce a second clutch after a relatively short interval since the laying of the first clutch. Finally, if there is a net fitness benefit associated with a long interclutch interval, we should expect total production of fledglings in first and second broods to increase with the duration of the interclutch interval.
Third, parents may differ inherently in their ability to rear offspring. For example, adults that are resistant to virulent parasites (as shown by Møller 1990b) may produce resistant offspring and still initiate a second clutch shortly after the first clutch. Therefore, I expected measures of parasite resistance to be negatively related to duration of interclutch interval. One such component of parasite resistance is T-cell–mediated immune response, which Saino et al. (1997a) have shown to have a quantitative genetic component. Thus, I predicted based on individual differences in quality that interclutch interval would be negatively related to nestling quality measured in terms of T-cell response but not in terms of body size or body mass because T-cell response better reflects condition than body size or body mass (Saino et al. 1997a). In contrast, if trade-offs accounted for these patterns, a longer interval should follow the production of a brood's high strong T-cell responses.
Finally, adults may differ inherently in their ability to reproduce, and such differences may affect the ability to rapidly produce a second clutch. A particularly likely candidate as a phenotypic marker of this ability is the size of secondary sexual characters. Male barn swallows with long tails provide less parental care than short-tailed males (Møller 1992a, 1994; de Lope and Møller 1993; Kose and Møller 1999; Kose et al. 1999). This implies that females mated to males with long tails provide a disproportionately large share of parental care (Møller 1992a, 1994; de Lope and Møller 1993; Kose and Møller 1999; Kose et al. 1999), and the costs of such care would have to be paid as a delay in the start of the second clutch. Therefore, we can predict a positive relationship between interclutch interval and tail length of males, but not for tail length of females, or for other morphological characters of either sex. These predictions were tested by investigating (1) patterns of life-history decisions of individuals by analyzing phenotypic traits of individuals, while controlling for differences among years and (2) temporal trends in life-history traits across generations as a means of determining whether phenotypes have changed in a consistent manner either due to phenotypic plasticity or due to microevolutionary change.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study species
The barn swallow is a long-distance migratory passerine that breeds throughout the Western Palaearctic, but winters mainly south of the Sahara (Møller 1994
). It is an aerial insectivore, with the main food being Diptera, with smaller amounts of Hymenoptera, Coleoptera, and other flying insects, making it particularly susceptible to the influence of current weather conditions (Møller 1994). Tail length is a secondary sexual character, and males with long tails enjoy superior mating success, extrapair paternity, and differential parental investment by females (Møller 1988, 1989, 1994; de Lope and Møller 1993; Møller and Tegelström 1997; Saino et al. 1997b; Møller et al. 1998). Tail length in the barn swallow is condition-dependent as reflected by effects of body mass, hematophagous mites, radiation, and senescence on tail length (Møller 1990b, 1991, 1993b; Møller and de Lope 1999). Males usually attract a single mate with sexual displays, and both partners build a nest, whereas the female incubates the clutch alone (Møller 1994). Both sexes contribute to food provisioning of the offspring, and a second clutch can be reared during a single season. The interval between first and second clutches is a minimum of the duration of the laying period, the incubation period, and the nestling period. Because modal clutch size is 5 eggs in the first clutch, modal incubation period is 14 days, and modal nestling period is 20 days (Møller 1994), we should expect the interclutch interval to be at least 39 days among pairs with a first clutch that produced at least one fledgling. That calculation is based on the assumption that females only can start a second clutch when the first brood has fledged. However, females may start laying before fledging of the first brood either in the same nest or a different, causing the interval to be as short as 30 days when the first clutch produced at least one fledgling.
Study area
Barn swallows were studied during 1971–2005 at Kraghede (57°12'N, 10°00'E), Denmark (Møller 1994). This study area of ca., 55 km2 consists of open farmland habitat with scattered mixed plantations, groves, and bogs. Barn swallows breed at farms where the number of breeding pairs ranges from single pairs to more than 50.
General field procedures
Barn swallows were captured shortly after arrival to the breeding grounds using mist nets with the same capture and measurement procedures being followed 1971–2005.
A number of morphological measurements were made, including the length of the outermost tail feathers, the central tail feathers, and the flattened wing, with a ruler with an accuracy of 1 mm. Tarsus length was measured to the nearest 0.01 mm with a digital caliper and body mass recorded to the nearest 0.1 g with a Pesola spring balance. All measurements were made by A.P.M., thus avoiding interobserver variability in measurements. More detailed information on methods of data collection and measurements than given above can be found in Møller (1992b, 1994). Measurements of adults were recorded for all individuals during the years 1984–2005. They were provided with a numbered aluminum ring, a color ring, and a combination of color codes on the white ventral plumage. Individuals were assigned to nests using individual identification.
I excluded all individuals with broken or damaged feathers from the analyses (less than 3% in any given year). The tip of the outermost tail feathers of barn swallows is rounded, and because feathers are composed of small barbs, any broken barb immediately leaves an irregular shape of the feather that is readily visible to anyone familiar with birds. This is unlikely to have seriously biased the samples of birds in any particular way.
All individuals involved in experiments with the exception of untreated controls were excluded from the analyses. Because treatments were assigned randomly, this cannot have caused any bias in the data. Individuals receiving a noncontrol treatment in any given year only involved a small fraction of individuals.
Recording reproductive performance
All individuals were considered to be yearlings when first captured. This assumption is justified by 174 of 175 recruits being yearlings when returning to a breeding site within the study area. Breeding philopatry was high because only 4 out of 3365 adult barn swallows ever moved from one breeding site to another within or between years and then only to the nearest neighboring site at a maximum distance of 400 m (within a study area of 55 km2).
All potential breeding sites were visited at least weekly to record reproductive events. Laying date was the date when the first egg of a clutch was laid, assuming that 1 egg was laid daily, which was the case with the exception of rare laying breaks (Møller 1994).
Clutch size was the maximum number of eggs found in a nest after laying had ceased, whereas brood size at hatching and fledging, respectively, was the number of nestlings present at hatching and at fledging. I recorded laying date, clutch size, and brood size for second clutches as described above. Annual production of offspring was the number of fledglings per pair in both the first and the second brood.
Survival rates for yearlings and older individuals of the 2 sexes were calculated using recaptures and resightings of already ringed individuals. Previous capture–mark–recapture analyses have shown that more than 98% of birds present are recaptured (Møller and Szép 2002; Møller AP and Szép T, unpublished data). This implies that recaptures and resightings provide reliable estimates of apparent survival rate. Because breeding dispersal is extremely limited (see above), estimates of apparent survival are not confounded by dispersal.
Recording nestling phenotype
When nestlings were 12 days old, thus having achieved adult body size, tarsus length, and body mass were recorded for a subsample as described above for adults.
At this age, nestlings were injected intradermally in the wing web (patagium) with 0.2 mg of phytohemagglutinin (PHA)-P (Sigma, L-8754, Sigma Inc., Saint Quentin Fallavier, France) in 0.04 ml isotonic saline (the antigen injection) to obtain an estimate of T-cell–mediated immune response. This was done during 1996–2005. The left wing web was injected with the same amount of saline only (a control injection). Injection sites were marked with permanent ink before injection. The thickness of wing webs was measured immediately before and 24 h after injection in inoculated sites using a pressure-sensitive caliper (Alpa SpA, Milano, cod. SM112, Teclock, Japan) with an accuracy of 0.01 mm. The reaction to PHA was controlled for the effect of injection per se and thickening due to saline injection. This was done by calculating the difference between the change in thickness of the right PHA-injected wing web (thickness 24 h after injection minus thickness before injection) and the corresponding change in thickness of the left wing web, only injection with saline (Saino et al. 1997a). The thickness of the wing web was measured 3 times before and after PHA injection, and the average of these 3 measurements was used in calculations. The repeatability (Falconer and Mackay 1996) of the wing web index was high and highly significant (R [SE] = 0.95 [0.02], F = 37.47, degrees of freedom [df] = 401,1206, P < 0.0001), as in previous studies (Saino et al. 1997a).
Recording local climate conditions
I obtained mean temperature for April–August from Tylstrup located 7 km from the study area based on measurements provided by the Danish Meteorological Institute. This period was chosen because it encompasses the months with reproduction.
Mean temperature in April increased significantly during 1971–2004 (F = 17.40, df = 1,32, r2 = 0.35, P = 0.0002, slope [SE] = 0.066 [0.016]), whereas that was not the case for mean temperature in May, June, or July (F < 2.23, df = 1,32, r2 < 0.07, P > 0.15). Mean temperature in August also increased significantly during the study (F = 4.67, df = 1,32, r2 = 0.13, P = 0.038, slope [SE] = 0.056 [0.026]).
The North Atlantic Oscillation (NAO) is a major source of atmospheric mass balance between pressure centers over Ponta Delgada, Azores, and Stykkisholmur, Reykjavik, Iceland (Hurrell 1995). An index of NAO is estimated as the difference in normalized sea-level pressures by division of each monthly pressure by the long-term standard deviation (1865–1984). High index values are associated with high winter temperatures and high levels of precipitation in Denmark (Hurrell 1995). NAO data for the period April–August were obtained from http://www.cru.uea.uk/ftpdata/nao/htm. The results presented here did not change if NAO data for December–March or for the entire year were used instead of the April–August index.
Statistical methods
Analyses of annual mean estimates and variances of interclutch intervals were based on data for 2705 pairs recorded during the period 1971–2005. Sample sizes per year varied between 9 and 91 pairs with 2 clutches. Analyses of interclutch intervals of individual pairs were based on 1623 pairs recorded during the years 1984–2005 because detailed information on morphology was only available for all adults during that period. Because all information was unavailable for all pairs, sample sizes differ among tests. All individuals were only included in their first year of life, thereby eliminating any pseudoreplication.
I investigated patterns of temporal autocorrelation in annual mean interclutch interval and coefficient of variation in interval but found only a weak, significant autocorrelation with a time lag of 1 year for mean interval (JMP 2000). That was also the case for mean temperature in April and mean laying date.
I developed best-fit, minimal models with laying date of the second clutch as the dependent variable and a number of predictor variables and their 2-way interactions as independent variables. Best-fit models were developed using a stepwise approach, and forward and backward elimination procedures produced similar models. Values reported are means (standard error [SE]).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Temporal trends in interclutch interval
Interclutch interval was on average 49.3 days (SE = 0.23), N = 1631, with a range of 9–86 days in the total sample of both unsuccessful and successful first clutches. Intervals for pairs with unsuccessful first clutches were shorter (mean [SE] = 37.3 days [1.00], N = 146) than for pairs with successful first clutches that produced at least one fledgling (Figure 1; mean [SE] = 50.4 days [0.21], N = 1485; t = 17.67, df = 1629, P < 0.0001).
|
Interval between clutches increased significantly during 1984–2005, when the analysis was based on mean interval duration for each year (Figure 2A). This implies an increase by 8.0 days of 19% since 1971. There was also an increase in duration of interval when temporal change in mean laying date of the first clutch was controlled statistically in a multiple regression (F = 39.24, df = 1,32, r2 = 0.55, P < 0.0001, slope [SE] = 0.22 [0.03]). The frequency of pairs producing 2 clutches remained constant during the period 1971–2005 (analysis based on square-root arcsine-transformed data: F = 1.15, df = 1,33, r2 = 0.03, P = 0.29), implying that change in interval could not be attributed to differences in the fraction of birds producing 2 broods.
|
The variance in interval duration increased significantly during 1971–2005 (Figure 2B). Mean laying date for first clutches advanced slightly during 1971–2005 (F = 5.34, df = 1,33, r2 = 0.14, P = 0.027, slope [SE] = -0.15 [0.07]), whereas mean laying date for second clutches was delayed during the same period (F = 10.40, df = 1,33, r2 = 0.24, P = 0.0028, slope [SE] = 0.19 [0.06]). This implies that first clutches were 5 days earlier in 2005 than in 1971, whereas second clutches in 2005 were almost 7 days later than in 1971.
Environmental conditions and interval
There was no significant relationship between mean duration of interclutch interval and NAO (F = 0.35, df = 1,32, r2 = 0.01, P = 0.56). However, the mean interval became longer as temperature in April increased in recent years (Figure 3). There was no significant relationship for interclutch interval and mean temperature for any of the other months or for mean temperature during the entire summer.
|
Life-history traits and interval
Mean interclutch interval was longer in years when mean laying date was early (F = 4.97, df = 1,33, r2 = 0.13, P = 0.033, slope [SE] = -0.26 [0.12]). No other variable like laying date for the second clutch, clutch size for the first or the second clutch, or the proportion of clutches that produced at least one fledgling entered as significant predictors in this model. Likewise, laying date of the second clutch was later, and hence interclutch interval was longer among individuals that started laying their first clutch early (Figure 4A, Table 1A). The fit of the latter model was not improved by considering a nonlinear relationship between laying date of the second clutch and laying date. The earlier laying date of the second clutch for pairs that bred late did not compensate for delay instart of laying because the slope was less than 1 (t = 17.50, df = 1603, P < 0.0001). This effect of laying date of the second clutch was independent of the size of the first clutch, which did not enter as a significant predictor variable (Table 1A).
|
|
The interval until laying of the second clutch was longer when first clutches were large (Figure 4B, Table 1A). This effect was even stronger for the size of the first brood (F = 303.28, df = 1,1629, r2 = 0.16, P < 0.0001, slope [SE] = 2.27 [0.13]). In contrast, there was a negative relationship between size of the second clutch and interclutch interval (Table 1A).
The interval between clutches increased with the total number of fledglings in the 2 clutches (Figure 4C). This means that production of fledglings increased with increasing duration of the interval, providing a fecundity advantage to pairs that were able to delay the start of a second clutch.
Survival of adults as reflected by recaptures at the breeding sites were marginally related to interval duration, but in opposite directions in the 2 sexes as indicated by the significant sex by interval interaction (Table 2). Males tended to survive less well when intervals were long, whereas females tended to survive better when intervals were long, as shown by the significant sex by interval interaction.
|
Nestling phenotype and interval
Interclutch interval was negatively related to mean T-cell response of nestlings in the first brood (Figure 5, Table 1B). No other nestling variable entered as a significant predictor. The year effect was not a significant predictor (F = 0.78, df = 8,290, P = 0.62).
|
Adult phenotype and interval
Interval was significantly positively related to tail length in males (Figure 6, Table 1C) but not to tail length in females or any other phenotypic variable. An analysis based on all pairs for which there were data on tail length for both partners revealed a model that explained 33% of the variance (F = 22.92, df = 29,1378, P < 0.0001). There was a nonsignificant effect of tail length (F = 3.67, df = 1,1378, P = 0.06), but a significant difference in the effect of tail length on duration of interclutch interval between the sexes (F = 5.53, df = 1,1378, P = 0.019).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The main findings of this study were that interclutch intervals in the barn swallow became longer and more variable as spring temperature increased during the period 1971–2005. Intervals were longer when barn swallows started breeding early and when their first clutches and in particular first broods were large. Barn swallows with long intervals produced more fledglings during the breeding season. Intervals were longer when females were mated to sexually attractive males as reflected by long tails. These findings are consistent with the hypothesis that the duration of the interclutch interval is determined by costs and benefits of a short interval due to a combination of effects of environmental conditions, reproductive effort, and individual differences in the ability to pay such costs.
The interclutch interval not only increased in duration but also became more variable, as mean temperature in April increased by more than 2.2 °C since 1971. This observation is consistent with the hypothesis that it is the duration of the reproductive season that is a contributing factor determining the duration of the interval between subsequent clutches. Whereas changes in local climate as reflected by temperature in early spring were dramatic during a period of only 35 years, there was no similar change in temperature during May–July, which then can be considered to represent control periods that did not affect the duration of the interclutch interval. Second clutches are laid in July and to a small extent in August, but environmental conditions at that time as reflected by temperature did not affect the duration of the interclutch interval. Thus, it is not the environmental conditions, when the interclutch interval is determined (i.e., when the second clutch is laid), that are important but the conditions at the start of the breeding season. Because there was a slight advance in laying date of the first clutch during 1971–2005, and an even greater delay in laying date of the second clutch during the same period, this implies that it is the change in laying date of both the first and the second clutch that determines the duration of the interclutch interval.
Life-history traits were predicted to be associated with interclutch interval because laying date, clutch size, and brood size are all determined not only by trade-offs between different fitness components but also by individual differences in quality, as suggested by life-history theory (Roff 1992; Stearns 1992). The interval until start of the second clutch became shorter with later start of the first clutch. Because late-breeding individuals are of poor phenotypic quality (Møller 1994), we would expect late breeders to have long intervals if only phenotypic quality accounted for the observations. That was clearly not the case. Although intervals became shorter with later laying of the first clutch, this effect could not fully compensate for the delay of laying of the second clutch by late-breeding pairs. I found an independent effect of size of the first clutch on interval duration. This effect was even more pronounced for size of the first brood. A similar effect for size of the first brood has been reported for the great tit (Verboven and Verhulst 1996). This effect was expected if parental effort increased the duration of the interval. There was evidence that pairs with long intervals, independent of timing of their first clutch, adjusted clutch size and brood size for late reproduction because the size of the second clutch was negatively related to interval duration. This implies that pairs with long intervals laid smaller second clutches and raised smaller second broods than pairs with short intervals.
T-cell response of nestlings from first clutches, but not body mass or tarsus length of the same offspring, predicted duration of the interclutch interval. Shorter intervals for pairs producing offspring with stronger T-cell responses does not imply that such offspring are cheap to produce because experimental manipulation of brood size in the barn swallow has shown that production of offspring with strong T-cell responses is associated with a reduction in parental survival prospects (Saino et al. 1999). Therefore, the negative relationship between T-cell response of nestlings and interval duration is likely to reflect that certain parents are able to both lay a second clutch after a short interval and simultaneously produce offspring of superior quality. If life-history trade-offs accounted for this relationship, we should expect the interval to be longer after the production of offspring with strong T-cell responses. That was clearly not the case. Saino et al. (1997a) have shown a genetic component to T-cell response in nestling barn swallows, and such partly genetically determined immunity may result in production of offspring with strong immune responses because offspring will not only inherit the immune response of their parents but also provide parents with superior health status thus allowing them to lay a second clutch after a short interval. This mechanism may allow such parents to avoid the costs of parasitism and therefore rapidly initiate a second clutch. Alternatively, effects of parental effort on offspring quality may be due to a trade-off between duration of parental care of the first clutch and early start of a second clutch. Previous studies of the great tit have shown that the duration of parental care for first brood nestlings was longer when second clutches were removed (Verhulst et al. 1997). Furthermore, the interclutch interval increased when great tit fledglings from the first brood were heavy (Verboven and Verhulst 1996).
Females prefer to mate with male barn swallows with long tails (Møller 1988, 1994). Such males enjoy a number of mating advantages, including differential parental investment by females into provisioning of offspring (Møller 1992a, 1994; de Lope and Møller 1993; Kose and Møller 1999; Kose et al. 1999). Females pay costs for higher investment into reproduction with attractive males. A measure of such costs is the time until the next reproductive event. I found evidence of the interclutch interval being longer when the male partner had a long rather than a short tail. Across the range of tail lengths in males (85–138 mm), this effect gave rise to a difference in interclutch interval of (138 – 85 mm) x 0.13 days/mm = 6.9 days on average. Interestingly, female tail length did not have an independent effect on duration of the interval, showing that this effect is not a mere effect of tail length per se. Likewise, a number of other adult phenotypic characters such as length of the short central tail feathers, wing, and tarsus or body mass in either sex were not significantly related to the duration of the interclutch interval. This provides evidence of interclutch interval being related to a specific phenotypic character in one sex, as predicted a priori from sexual selection theory.
The duration of the interclutch interval may change across generations as a consequence of phenotypic plasticity and microevolution due to the fitness costs and benefits of the trait. Here I have shown that long interclutch intervals are associated with higher reproductive success than short intervals. This implies that individual barn swallows acquire a fitness benefit by extending the duration of the interval between the first and the second clutch. This benefit is most likely acquired through elimination of a time constraint caused by climate change. This interpretation is supported by survival prospects of adult barn swallows changing in response to variation in interclutch interval. Whereas females survived better when the interclutch interval was long, their male partners survived less well. Adult female barn swallows invest disproportionately in reproduction by providing half of all nest building, laying eggs, performing incubation, and brooding and providing half of all feeding of nestlings (Møller 1994). Adult males have longer tail feathers than females, and these take longer to molt than female tails (Møller 1994; Møller et al. 1995). These differences between the sexes may explain the sex difference in relationship between interclutch interval and survival prospects. This difference may also suggest the presence of a sexual conflict (Trivers 1972) over duration of the interclutch interval, with females benefiting from a long interval and males benefiting from a short one. The fact that females ultimately determine when laying occurs would make it likely that females determine the duration of the interclutch interval. However, males may indirectly influence the duration of the interval through their contribution to rearing of offspring and thus to male effects on timing of independence of the offspring of the first clutch.
In conclusion, I have shown in analyses of long-term time series of interclutch intervals that the time constraints on duration of such intervals can be changed when temperatures increase due to climate change. Such climate change seems to have affected the mean and the variance in interclutch interval and the costs and the benefits of particular interval durations.
|
ACKNOWLEDGEMENTS |
|---|
I gratefully acknowledge help with fieldwork by N. Cadée, E. Flensted-Jensen, and C. Spottiswoode. My research was supported by grants from the French Biodiversity Program.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Blount JD, Houston DC, Møller AP. (2000) Why egg yolk is yellow. Trends Ecol Evol 15:47–49.[CrossRef][Medline]
Blount JD, Houston DC, Surai PF, Møller AP. (2003) Egg-laying capacity is limited by carotenoid pigment availability in wild gulls Larus fuscus. Proc R Soc Lond B Biol Lett 271:S79–S81.
Both C and Visser ME. (2001) Adjustment of climate change is constrained by arrival date in a long-distance migrant bird. Nature 411:296–298.[CrossRef][Medline]
Cresswell W and McCleery RH. (2003) How great tits maintain synchronization of their hatch date with food supply in response to long-term variability in temperature. J Anim Ecol 72:356–366.[CrossRef]
Crick HQP, Gibbons DW, Magrath RD. (1993) Seasonal changes in clutch size in British birds. J Anim Ecol 62:263–273.[CrossRef]
de Lope F and Møller AP. (1993) Female reproductive effort depends on the degree of ornamentation of their mates. Evolution 47:1152–1160.[CrossRef][ISI]
Dhondt AA. (2001) Trade-offs between reproduction and survival in tits. Ardea 89:155–166.
Dunn P. (2004) Breeding dates and reproductive performance. In Møller AP, Fiedler W, Berthold P (Eds.). Effects of climatic change on birds(Elsevier, Amsterdam) pp. 69–87.
Erikstad KE, Fauchald P, Tveraa T, Steen H. (1998) On the cost of reproduction in long-lived birds: the influence of environmental variability. Ecology 79:1781–1788.[CrossRef]
Falconer DS and Mackay TFC. (1996) Introduction to quantitative genetics 4th ed. (Longman, New York).
In Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Dai X, Maskell K, Johnson CA (Eds.). Climate Change 2001: The Scientific Basis (2001) (Cambridge University Press, Cambridge (UK)).
Hurrell JW. (1995) Decadal trends in the North Atlantic Oscillation: regional temperatures and precipitation. Science 269:676–679.
JMP. (2000) JMP(SAS Institute Inc., Cary (NC)).
Kose M, Mänd R, Møller AP. (1999) Sexual selection for white tail spots in the barn swallow in relation to habitat choice by feather lice. Anim Behav 58:1201–1205.[CrossRef][ISI][Medline]
Kose M and Møller AP. (1999) Sexual selection, feather breakage and parasites: the importance of white spots in the tail of the barn swallow. Behav Ecol Sociobiol 45:430–436.[CrossRef]
Lack D. (1954) The natural regulation of animal numbers(Clarendon, Oxford).
Lindén M and Møller AP. (1989) Clutch size, cost of reproduction, and covariation of life history traits in birds. Trends Ecol Evol 4:367–371.[CrossRef]
Møller AP. (1988) Female choice selects for male sexual tail ornaments in the monogamous barn swallow. Nature 332:640–642.[CrossRef]
Møller AP. (1989) Viability costs of male tail ornaments in a swallow. Nature 339:132–135.[CrossRef]
Møller AP. (1990a) Effects of parasitism by the haematophagous mite Ornithonyssus bursa on reproduction in the barn swallow Hirundo rustica. Ecology 71:2345–2357.[CrossRef]
Møller AP. (1990b) Effects of an haematophagous mite on the barn swallow (Hirundo rustica): a test of the Hamilton and Zuk hypothesis. Evolution 44:771–784.[CrossRef]
Møller AP. (1991) Sexual selection in the monogamous barn swallow (Hirundo rustica). I. Determinants of tail ornament size. Evolution 45:1823–1836.[CrossRef]
Møller AP. (1992a) Sexual selection in the monogamous swallow (Hirundo rustica). II. Mechanisms of intersexual selection. J Evol Biol 5:603–624.[CrossRef]
Møller AP. (1992b) Parasites differentially increase the degree of fluctuating asymmetry in secondary sexual characters. J Evol Biol 5:691–699.[CrossRef]
Møller AP. (1993a) Parasites increase the cost of reproduction in their hosts. J Anim Ecol 62:309–322.[CrossRef]
Møller AP. (1993b) Morphology and sexual selection in the barn swallow Hirundo rustica in Chernobyl, Ukraine. Proc R Soc Lond B Biol Sci 252:51–57.
Møller AP. (1994) Sexual selection and the barn swallow(Oxford University Press, Oxford).
Møller AP. (1997) Parasites and the evolution of host life history. In Clayton D and Moore J (Eds.). Host-parasite evolution: general principles and avian models(Oxford University Press, Oxford) pp. 105–127.
Møller AP. (2004) Protandry, sexual selection and climate change. Glob Change Biol 10:2028–2035.[CrossRef]
Møller AP and de Lope F. (1999) Senescence in a short-lived migratory bird: age-dependent morphology, migration, reproduction and parasitism. J Anim Ecol 68:163–171.[CrossRef]
Møller AP, Magnhagen K, Ulfstrand A, Ulfstrand S. (1995) Phenotypic quality and moult in the barn swallow Hirundo rustica. Behav Ecol 6:242–249.
Møller AP, Saino N, Taramino G, Galeotti P, Ferrario S. (1998) Paternity and multiple signaling: effects of a secondary sexual character and song on paternity in the barn swallow. Am Nat 151:236–242.[CrossRef]
Møller AP and Szép T. (2002) Survival rate of adult barn swallows Hirundo rustica in relation to sexual selection and reproduction. Ecology 83:2220–2228.
Møller AP and Tegelström H. (1997) Extra-pair paternity and tail ornamentation in the barn swallow Hirundo rustica. Behav Ecol Sociobiol 41:353–360.[CrossRef][ISI]
Nielsen JT and Møller AP. (2006) Effects of food abundance, density and climate change on reproduction in the sparrowhawk Accipiter nisus. Oecologia 149:505–518.[CrossRef][ISI][Medline]
Roff DA. (1992) The evolution of life histories(Chapman and Hall, New York).
Saino N, Calza S, Møller AP. (1997a) Immunocompetence of nestling barn swallows in relation to brood size and parental effort. J Anim Ecol 66:827–836.[CrossRef]
Saino N, Calza S, Ninni P, Møller AP. (1999) Barn swallows trade survival against offspring condition and immunocompetence. J Anim Ecol 68:999–1009.[CrossRef]
Saino N, Primmer C, Ellegren H, Møller AP. (1997b) An experimental study of paternity and tail ornamentation in the barn swallow (Hirundo rustica). Evolution 51:562–570.[CrossRef][ISI]
Sheldon BC, Kruuk LE, Merilä J. (2003) Natural selection and inheritance of breeding time and clutch size in the collared flycatcher. Evolution 57:406–420.[CrossRef][ISI][Medline]
Stearns SC. (1992) The evolution of life histories(Oxford University Press, Oxford).
Tinbergen JM and Sanz JJ. (2004) Strong evidence for selection for larger brood size in a great tit population. Behav Ecol 15:525–533.
Trivers RL. (1972) Parental investment and sexual selection. In Campbell B (Ed.). Sexual selection and the descent of man: 1871–1971(Heinemann, London) pp. 136–179.
van Noordwijk AJ and de Jong G. (1986) Acquisition and allocation of resources: their influence on variation in life history tactics. Am Nat 128:137–142.[CrossRef][ISI]
Verboven N and Verhulst S. (1996) Seasonal variation in the incidence of double broods: the date hypothesis fits better than the quality hypothesis. J Anim Ecol 65:264–273.[CrossRef]
Verhulst S, Tinbergen JM, Daan S. (1997) Multiple breeding in the Great Tit. A trade-off between successive reproductive attempts? Funct Ecol 11:714–722.[CrossRef]
Visser ME, Both C, Lambrechts MM. (2004) Avian timing of reproduction, food availability and climate change. In Møller AP, Fiedler W, Berthold P (Eds.). Effects of climatic change on birds(Elsevier, Amsterdam) pp. 89–110.
Visser ME, van Noordwijk AJ, Tinbergen JM, Lessells CM. (1998) Warmer springs lead to mistimed reproduction in great tits (Parus major). Proc R Soc Lond B Biol Sci 265:1867–1870.[CrossRef]
von Schantz T, Bensch S, Grahn M, Hasselquist D, Wittzell H. (1999) Good genes, oxidative stress and condition-dependent sexual signals. Proc R Soc Lond B Biol Sci 266:1–12.[Medline]
Wheelwright NT, Tice KA, Freeman-Gallant CR. (2003) Postfledging parental care in Savannah sparrows: sex, size and survival. Anim Behav 65:435–443.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. R. Ridley and N. J. Raihani Task partitioning increases reproductive output in a cooperative bird Behav. Ecol., August 8, 2008; (2008) arn097v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||