- Top of page
- Materials and methods
Birds time their breeding attempts so that the period of maximal nestling growth rate coincides with peak food availability. Birds (as other animals) use the vernal increase in photoperiod as a reliable cue to anticipate the onset of the breeding season so that the cascade of physiological and behavioural changes that must precede hatching is timed appropriately (Dawson et al. 2001; Ball & Balthazart 2003). Although increasing photoperiods during spring lead to gonadal maturation, later, long photoperiods cause gonadal regression through a mechanism known as photorefractoriness, and this essentially dictates the latest time that eggs can be laid. The beginning of the postnuptial moult is closely related to the time of gonadal regression.
Climate change may lead to increased spring temperatures and this will advance the time of the peak abundance of invertebrates (Buse et al. 1999), on which many species of birds rely to feed their young. If birds rely entirely on photoperiod to time breeding, they will be unable to compensate by adjusting the time of breeding and a mismatch will develop between the time of invertebrate abundance and peak nestling growth (Visser et al. 1998, Visser, Both & Lambrechts 2004; Coppack & Pulido 2004).
However, there is abundant evidence in many species that the time of egg-laying varies with temperature and that it has advanced over recent years, e.g. for Pied Flycatchers Ficedula hypoleauca (Both et al. 2004). Laying dates for Great Tits Parus major in Wytham Wood, UK, correlate with spring temperature (Perrins & McCleery 1989) and have advanced over the last four decades (Cresswell & McCleery 2003). However, this is not true for all European Parus study sites (Visser et al. 2003). In a population of non-migratory Common Starlings Sturnus vulgaris in the UK, first laying dates did not correlate with spring temperature and have not advanced over recent years (Feare & Forrester 2003). However, in a migratory population in southern Germany, there was a correlation with temperature immediately before egg-laying (Meijer et al. 1999). Analysis of breeding records for 65 species of birds in the UK shows that median egg-laying dates tended to get earlier between 1971 and 1995 and that this was significant for 20 species, including starlings (Crick et al. 1997). Median laying dates correlate with spring temperature (Crick & Sparks 1999). However, an advance in median laying date does not necessarily mean that first eggs are laid earlier – it could also reflect a decrease in the proportion of birds laying second or later clutches.
It is unclear how temperature may influence egg-laying dates. Does temperature directly modulate gonadal responses to photoperiod? Or is there an indirect effect, e.g. through changes in food supply? Perhaps surprisingly, there have been no detailed studies addressing whether temperature directly affects physiological responses to naturally increasing photoperiod during spring. In an early experiment on Great Tits, Soumalainen (1938) showed no apparent effect of increased temperatures on testicular size during early spring. A few studies have assessed the effects of temperature on gonadal maturation following an acute move from a short to a long photoperiod (Silverin & Viebke 1994; Wingfield et al. 1996; Wingfield et al. 1997; Wingfield et al. 2003– results are summarized in the Discussion) but such protocols may miss subtle effects of temperature on the rate of maturation during early spring.
The initial aim of this study was simply to determine whether temperature modulated photoperiodically induced gonadal maturation. The finding that temperature had no effect on the timing or rate of maturation was not unexpected. However, the additional finding that the duration of gonadal maturity was curtailed at higher temperatures was unexpected, but may explain results of recent field studies.
- Top of page
- Materials and methods
Extreme differences in temperature (15 °C and 10 °C) had no effect on the rate or timing of testicular maturation in starlings. In a population of non-migratory starlings in the UK, first laying dates did not correlate with spring temperature (Feare & Forrester 2003), but in a migratory population in southern Germany, there was a correlation with temperature immediately before egg-laying (Meijer et al. 1999). In the latter study, egg-laying could be advanced by heating nestboxes. So the exact timing of egg-laying may be influenced by temperature immediately before laying. However, if there is no effect of temperature on the timing and rate of gonadal maturation prior to then, the degree to which egg-laying can be advanced by temperature will be constrained. This leads to a potential problem for this species. Starlings feed their nestlings almost exclusively on leatherjackets (Tipula sp. larvae) and the timing of the increase in leatherjacket biomass during spring is advanced with increasing soil temperature (Dunnet 1955). If starlings rely entirely on photoperiod, they will breed too late. However, it is possible that temperature may advance breeding indirectly, for example through an advance in food supply (Meijer & Langer 1995). Increased food may act as a supplementary proximate cue rather than temperature.
Although an increase in temperature did not advance testicular maturation in starlings, this is not true for all species. Also, the rate of ovarian development in females may be more temperature dependent. Wingfield et al. (1992) have argued that non-photoperiodic supplementary factors, such as temperature, are more likely to affect the rate of gonadal maturation in species with longer and more flexible breeding seasons. A series of studies in which different subspecies of White-Crowned Sparrow (Zonotrichia leucophrys sp.) were transferred from a short to a long photoperiod at different temperatures has leant support for this. In Z. l. gambelii, which has a short predictable breeding season at high latitudes, temperature had no effect on the rate of testicular or ovarian maturation (Wingfield et al. 1996). In Z. l. pugetensis, which breeds at mid-latitudes, temperature did not affect testicular growth but did affect ovarian development, and in Z. l. oriantha, which breeds at lower latitudes but high altitudes and has a flexible breeding season, temperature affected gonadal maturation in both sexes (Wingfield et al. 2003). In Willow Tits (Parus montanus) high temperature accelerated testicular maturation, but in Great Tits (P. major) there was little effect (Silverin & Viebke 1994). However, in all of these studies, the experimental paradigm was an acute transfer from a short to a long photoperiod, rather than a natural increase in photoperiod. When Song Sparrows (Melospiza melodia) were exposed to naturally increasing photoperiods, higher temperature slightly enhanced testicular growth in birds from a mountain population, but had no effect in birds from a coastal population (Perfito et al. 2005).
Starlings have a short and predictable breeding season. Even if ovarian development is more temperature dependent than testicular development, fertile eggs could not be laid before full testicular maturation. The time of full testicular maturation in this study corresponded to first egg dates in free-living birds, and was not advanced by higher temperature.
Although temperature had no effect on the timing of testicular maturation in starlings, it did have a dramatic and significant effect on the timing of regression and on the timing of the start of postnuptial moult. The consequence of this is that increasing temperatures will decrease the duration of the breeding season. The number of pairs of starlings laying second clutches in the colony studied by Feare & Forrester (2003) has indeed decreased, but so too has the number of first clutches, reflecting a decrease in the population of this species. Analysis of data for starlings in the Nest Record Scheme of the British Trust for Ornithology shows that median egg-laying date has significantly advanced over the last three decades and that this does correlate with spring temperatures (H.Q.P. Crick, personal communication). Temperature during January, February and March had little effect, which is in accord with this study. April temperature was much more influential. However, first eggs are laid in early April so the scope for April temperature to influence first egg date is limited. The scope for April temperatures to affect median laying date through subsequent effects on the timing of the end of the breeding season is possibly greater.
No advance in the start of the breeding season, but a decrease in its duration, is what has been reported in some recent studies. During the last four decades, the beginning of the Turtle Dove Streptopelia turtur breeding season has not changed, but the end has advanced by 12 days, with fewer breeding attempts per pair (Browne & Aebischer 2003). There is evidence that first egg date has advanced in a population of Great Tits in the UK over the last four decades (Cresswell & McCleery 2003), but the timing of first clutches has advanced in only 5 of 13 European Great Tit study sites and in only 3 of 11 Blue Tit P. caeruleus sites (Visser et al. 2003). However, there has been a decrease in the proportion of pairs that start a second clutch (Visser et al. 2003). The number of breeding attempts per year may be an important factor in the recent decline in some bird species in the UK (Siriwardena et al. 2000).
The mechanism which leads to earlier regression at a higher temperature is unknown, but it may involve prolactin. High temperatures enhance prolactin secretion (Maney et al. 1999; Gahali, El Halawani & Rozenboim 2001) and high prolactin concentrations cause gonadal regression (Dawson & Sharp 1998). Prolactin also has a role in the control of moult (Dawson & Sharp 1998). Birds at lower temperatures started to moult later than those at higher temperatures in experiments 1 and 2, but then moulted more rapidly. The increased rate of moult was not a direct effect of temperature (experiment 3). Rather, it is an effect of photoperiod, and is dictated by the timing of the start of moult (Dawson et al. 2000; Dawson 2004). Consequently, birds in the outdoor aviary (experiment 1), which started to moult at the same time as the 5 °C group, then moulted at the same rate as the 5 °C group.
This study demonstrated that increased temperature does not advance the timing of gonadal maturation in starlings but does advance the timing of gonadal regression. This implies that there may be two detrimental consequences of increasing spring temperatures – an insufficient advance in breeding seasons to match the advance in invertebrate food supplies and breeding seasons that end sooner. The recorded advances in median egg-laying dates for many species (Crick et al. 1997; Crick & Sparks 1999) may result in part from an advance in the end of breeding seasons rather than reflecting successful adaptation. This may therefore understate the consequences of climate change for bird populations. The implication is that successful adaptation to climate change will require evolutionary changes in the physiological responses to photoperiod.