Length of feeding day and body weight of great tits in a singleand a two-predator environment

Transcription

1 Behav Ecol Sociobiol (2000) 48: Springer-Verlag 2000 ORIGINAL ARTICLE Indriķis Krams Length of feeding day and body weight of great tits in a singleand a two-predator environment Received: 8 December 1999 / Received in revised form: 15 March 2000 / Accepted: 31 March 2000 Abstract The risk of predation may influence the acquisition of energy and the feeding activity of animals. Feeding activity and body reserves of wintering great tits Parus major in response to the priority to food access were studied in two areas differing in incidence of predators. The one-predator area contained sparrowhawks Accipiter nisus only, whereas the two-predator area contained both sparrowhawks and pygmy owls Glaucidium passerinum, whose hunting periods overlap at dawn and dusk. In the two-predator area dominant great tits arrived at feeders significantly later in the morning, and left earlier in the evening, than their subordinate flock-mates. Hence, feeding day length of dominants was found to be significantly shorter. The reverse was true for the one-predator area. In addition, dominants carried significantly greater reserves than subordinates in the area inhabited by two predators. Factors constraining subcutaneous energy reserves were also studied in removal experiments. After the removal of dominant individuals, subordinate great tits did not reduce their body reserves in the two predator area. In contrast, subordinate great tits significantly reduced evening body reserves in the single-predator area. I concluded that the presence of the two predators increases unpredictability in feeding conditions for great tits. Dominant individuals responded to this by shortening their feeding day and increasing body reserves at dusk. Key words Predator-prey interactions Birds Foraging activity Body reserves Introduction Among resident birds in a seasonal environment, subcutaneous energy reserves increase in winter and decline Communicated by N.B. Metcalfe I. Krams ( ) Department of Sciences, Daugavpils University, Vienī bas iela 13, LV-5400 Daugavpils, Latvia krams@apollo.lv Fax: again in summer (King and Farner 1966; Lehikoinen 1987; Haftorn 1994). The fact that many species clearly fatten in response to increasing stress suggests that birds rarely carry the maximum amount of reserves (King 1972). This indicates that fattening strategies are under adaptive control. Energy reserves show considerable circadian variation, being depleted overnight and replenished during the day, and an upper limit to the amount of individual body reserves gathered during daily activities is generally considered to be set by a range of internal and external constraints, such as food resource availability or factors imposing fatness costs (Witter and Cuthill 1993). According to the optimal body mass hypothesis (Lima 1986) there is a trade-off between the risks of starvation and predation, the optimal solution being the mass that minimizes the joint risk. The hypothesis predicts that with increasing resource predictability or increasing predation risk, subcutaneous energy reserves should decrease (McNamara and Houston 1990; Houston and McNamara 1993; Bednekoff and Houston 1994; Clark and Ekman 1995), which is also supported by field data (Gosler et al. 1995) and experiments (Lilliendahl 1997, 1998). In dominance-structured flocks, the competitive superiority of dominants may add extra uncertainty to the feeding conditions of subordinate flock mates influencing the survival value of resources (Piper and Wiley 1990; Verhulst and Hogstad 1996). Consequently the optimal reserve level can dependent on rank (Krams 1998a, 1998b). Subordinates could be forced to carry extra reserves as a buffer against periods of high energy demand or food shortage when dominants are most likely to take advantage of their priority in food access (Ekman and Lilliendahl 1993). Forest birds live in environments with several types of predators, yet studies of the behaviour of wintering tits under the risk of predation focus almost exclusively on a single-predator system. This may be a reason why different studies have found different effects of dominance on fat reserves of passerine birds (Lima 1988a). In Northern Europe the sparrowhawk Accipiter nisus and the pygmy owl Glaucidium passerinum are considered to

2 148 be the most important airborne predators of tits in the majority of studies on anti-predator behaviour (e.g. Ekman 1986; Suhonen 1993; Suhonen et al. 1993; Kullberg 1995, 1998a; Krams 1996). The sparrowhawk is widespread in the Baltic States, occurring in all types of forest, where it hunts for small passerines throughout the year (Transehe 1965; Priednieks et al. 1989). This diurnal predator usually requires good light conditions to attack its prey (Newton 1986). The pygmy owl in turn attacks parids near dawn and dusk (Mikkola 1983). The owl is not uncommon in the extensive coniferous forest found mainly in northern and eastern parts of Latvia (Transehe 1965; Priednieks et al. 1989). Thus, the forestwintering tits may occur in habitats containing either the sparrowhawk only or both the airborne predators. Some studies have demonstrated that there may be a relationship between dominance and the activity times (Lima 1988a; Summers and Feare 1995; Lahti et al. 1997). Under predation risks caused by the sparrowhawk, dominant great tits Parus major start feeding earlier than their subordinate flock-mates (De Laet 1985). There may be more predators about in the two-predator environment, especially in the dim light of dawn and dusk, because both nocturnal and diurnal predators may be active simultaneously (Mikkola 1983). Further, feeding at dusk and dawn is a risky affair since predatory attacks may be more difficult to detect (Lima 1988b). Consequently, this may increase uncertainty in foraging prospects of the dominance-group members, affecting their fattening strategies (Bednekoff and Krebs 1995; Gosler 1996; Cuthill and Houston 1997). I hypothesise that (1) dominant great tits may postpone the initiation of feeding activities in the morning in habitats containing both sparrowhawks and pygmy owls. Furthermore (2) the daily activity of dominant great tits may start earlier than the feeding of subordinate great tits in the forest areas with the sparrowhawk as the only predator. In line with this I predict that (3) dominant great tits should carry greater body reserves towards the end of daily activity in the two-predator habitat than the dominant tits wintering in the forest areas containing only sparrowhawks. To find out what kind of rank-dependent variation, if any, there is in the amount of reserves carried by members of winter groups of great tits in habitats differing in predation risk, I measured evening and morning body weights of free-living birds of known social status. The sequences of the first arrival to feeders in the morning and termination of feeding in the evening were also recorded. To test whether the correlation between rank in a basic flock and body reserves represents a causal relationship, I performed removal experiments. Methods Study site and species I studied individually colour-marked members of great tit basic flocks (Saitou 1979) wintering in the forest near the town of Krā slava (55 47 N, E) and at the Nature Reserve of Teiči (56 33 N, E), eastern Latvia. The first study site is referred to as one predator site containing the sparrowhawk only, whereas the second site contained both the sparrowhawk and the pygmy owl (Strazds et al. 1998). The bulk of the data was collected during winters of 1995/1996, 1996/1997 and 1998/1999. Although basic flocks are unstable, my observations show that forest-wintering great tits can form stable units for about 2 months in mid-winter. These great tits are usually associated with smaller parids. Therefore I studied body reserves and foraging behaviour from the beginning of December to the end of January only. The climate during the study was cold with daytime temperatures around 10 C (range 34 C to 5 C) and the snow cover m deep. Daylength at the winter solstice is 6.7 h. All members of the 18 basic flocks of the great tit (mean number of individuals=3.6, SE=0.28) had been sexed and aged (as first-winter or adult birds) either in the previous breeding season or during the study period. The birds were trapped by mist nets or baited traps. Wings (maximum chord) (Svensson 1984) were measured to 0.5 mm with a stopped steel rule to give a measure of body size (Garnett 1976). Dominance ranks Dominance order was measured within each flock using pairwise interactions between birds at temporary feeders prior to the study. The feeders were provided with sunflower seeds and fat. During observations at feeders I recorded between 58 and 88 aggressive encounters per flock (total 1188), and within each dyad the individual was considered to be dominant if it won significantly more interactions than the other (2-tailed sign-test). A clear dominance hierarchy was found in all the basic flocks. Old males were the highest, in the hierarchies, with no exceptions. Males were superior to females in both age groups of the tit species. This pattern was observed to be the same in free-roaming tits away from feeders, confirming previous reports (Saitou 1979; Gosler 1987; Krams 1998c). I divided the flock members into two dominance categories: dominants and subordinates. For this purpose, alpha and beta birds were treated as dominants and the rest as subordinates. Body weight When examining the feeding efficiency of great tit individuals, I used repeated weighings of individuals on different days. This method has been used in several other studies of body reserves (e.g. Ekman and Lilliendahl 1993). The weight of the bird was recorded on an electronic balance (BLTK-500), with a precision of 0.1 g. I collected weight data during the last hour of the birds daily activity period and during the first visit of an individual to the feeder in the morning. I recorded only one weight per individual tit per morning and evening. The members of a flock were weighed on the same day and I usually continued data collection for a week. When the birds were weighed, sunflower seeds were embedded in solidified fat and this food mixture was placed on the balance. Firstly, this made the body weight records more reliable and, secondly, the amount of food retrieved by tits was so small that its effect on daily pattern of body mass gain could presumably be neglected. To avoid the influence of temperature and habitat, every 2nd week I switched the collection of the field records from one patch to another for weight collection at sites of different predation risk. A set of 281 weighings of individual great tits (mean per individual=7.1, SE=0.06, n=42, number of basic flocks=10) was obtained in habitats with sparrowhawk only. Another set of 183 weighings of individual great tits (mean per individual=6.3, SE=0.04, n=30, number of basic flocks=8) was obtained in habitats including the both predators. The sampling unit was the flock and I compared body weight of dominants and subordinates pairwise within each flock. Tests that indicated significant differences were further analysed by comparisons using a Bonferroni procedure (Sokal and Rohlf 1995). A positive relationship between size and social rank has been re-

3 149 ported in the great tits, as well as in some other tit species (Garnett 1976; Järvi and Bakken 1984; Hegner 1985; Hogstad 1987). Therefore, I controlled for size by calculating a body weight index (BWI) for which the body weight (Table 1) was divided by the third power of the wing-length (Ekman and Lilliendahl 1993). Since extra fat may affect escape behaviour (Witter et al. 1994; but see Kullberg 1998b; Veasey et al. 1998), wing-length should be of biological significance for fat reserves (McNamara and Houston 1990). Observations of foraging activity I recorded the arrival time of each individual at the feeders in the morning and the departure time at the end of the day. Day length decreased during the study from about 11.5 h at the beginning of October to 8 h in the middle of November. The birds were trained to come to the permanent feeders in the territories when they heard a sound signal, and food at feeders was supplied only during observational hours. One could, of course, argue that the birds behaviour was different from normal. However, any possible effects would be similar in each individual, since each was taught to arrive when called for. Usually I played the signal continuously while walking around in the territory well before sunrise to minimize the possibility that some individuals would not hear the call. In any case, the call was loud enough to make sure it was heard in all parts of the territories. Continuous calling was employed to make sure the birds arrived at the feeders as soon as they departed from their roosts. Moreover, the amount of food acquired by the tits from the feeders was small. As a result, the birds used only natural food sources during the day and this minimized the predictability of feeders. Therefore I assumed that the supplemental food did not affect the duration of roosting. In the evening, I never heard great tit vocalizations after the last individual had left. Still, the birds left when the feeder was accessible as well as when it was absent, suggesting that they were not detained by the food. I compared the activity of dominants and subordinates pairwise within each flock. Removal experiment In order to see how dominance rank was related to body reserves and foraging activity under different predation risk, at the end of November 1998 I temporarily removed the local dominants in five experimental groups in a habitat containing only sparrowhawks and five groups in a habitat with both avian predators. The birds were caught by baited traps. Some subordinates in control groups were also removed so that the numbers of birds removed and remaining was the same. The remaining individuals were then observed for 5 days. Males removed from experimental flocks were caught early in the morning, immediately brought to indoor cages, weighed, and provided with ad libitum sunflower seeds and dried arthropods. After the experiments the birds were released at the place of capture. I performed repeated-measures ANOVAs, with the before and after BWI values as the repeated measures (the within-subject effect) and the type of bird (dominant from control flocks or subordinate from removal flocks) as the between-subject effect. Predators I was able to identify the presence of the pygmy owl and the sparrowhawk by regular field observations at the study site since Outside the breeding season sparrowhawks were seen or heard almost every day both in Scots pine Pinus sylvestris and Norway spruce Picea abies forests, hunting mainly for tits and great spotted woodpeckers Dendrocopos major (Krams 1996; I. Krams, personal observations). In general, it was more difficult to confirm the presence of the pygmy owl because of its smaller size and behaviour. However, in the study area all of the owls were found at least at the beginning of the previous breeding season. The flocks at the nature reserve (n=8) all overlapped with pygmy owl territories. To test the presence of pygmy owls by prey responses, I exposed all of the basic flocks to playbacks of the predator each month during the study period, using a Uher 4200 Report Stereo IC tape-recorder and S-30 loudspeaker. Before an experimental playback, a control playback consisting of a recorded series of the bullfinch Pyrrhula pyrrhula vocalizations was played. The tits (n=25) in basic flocks within territories of pygmy owls gave more alarm calls after hearing the predator s vocalizations than the control playback (binomial test, one-tailed P=0.001). Their high rate of calling was effective in attracting free-ranging conspecifics and heterospecifics to the immediate vicinity of the loudspeaker. In habitats without the pygmy owl, only 1 alpha male out of 35 members of ten basic flocks showed a weak response to the pygmy owl playback (binomial test, one-tailed P>0.05). Pygmy owls were not known to live in the 100 km between the two study areas. In general, pygmy owls avoid extensive pine forests such as forests near Krā slava, preferring mixed spruce and pine stands. Mixed forest around the bog of Teiči has been well-known as a habitat of the pygmy owl since the first Latvian Breeding Birds Atlas project (Priednieks et al. 1989). Mammalian predators such as small mustelids (Mustela erminea and M. nivalis) may kill a considerable number of great tits roosting in nest-boxes (Orell 1989). However, I did not find any evidence of mustelid predation on great tits during my study. Results Initiation and termination of foraging activity In the one-predator area dominants started to forage 28.62±1.23 min (mean, SE) earlier than their subordinate flock-mates (2-tailed paired t-test: t=3.42, df=9, P<0.01). Dominants also continued their foraging activities for 24.45±2.18 min (mean, SE) longer than subordinates (2- tailed paired t-test: t=3.29, df=9, P<0.01). By contrast, dominants initiated their daily activity 32.87±3.06 min (mean, SE) later than subordinate flock members in the two-predator area (2-tailed paired t-test: t=3.83, df=7, P<0.01) and terminated the activity day usually ±2.08 min (mean, SE) earlier than subordinates in the two-predator area (2-tailed paired t-test: t=3.58, df=7, P<0.01). Body weight At dawn dominant great tits were significantly heavier than their subordinate flock-mates in the both areas, and dominant individuals of the two-predator area were significantly heavier than dominants of the one-predator area (2-tailed paired t-test: t=2.04, df=7, P<0.05, Table 1). In the two-predator area dominant great tits were heavier than dominant individuals in the one-predator area towards the end of their daily activity (2-tailed paired t-test: t=2.16, df=7, P<0.05, Table 1). Dominant great tits were heavier than subordinates both in the one- and the two-predator areas (Table 1). The subordinates gained on the average 1.92 g mass during their activities in the sparrowhawk area while dominants gained about 0.22 g more than subordinates (Table 1). Dominant individuals gained about 0.54 g more than subordinates in the twopredator area. Although subordinates gained less body weight during the activity day, they lost less relative

4 150 Table 1 Morning body weight, evening body weight and daily weight gain of individuals in two habitats differing in predator incidence One-predator habitat Two-predator habitat Dominants Subordinates t df P a Dominants Subordinates t df P a Mean SE Mean SE Mean SE Mean SE Morning weight (g) < <0.01 Evening weight (g) < <0.01 Weight gain (g) < <0.01 a 2-tailed paired t-test Table 2 Body reserves (BWI) of great tits at dawn and dusk and relative weight loss by roosting individuals in two habitats differing in predator incidence One predator habitat Two predator habitat Dominants Subordinates T df P a Dominants Subordinates T df P a Mean SE Mean SE Mean SE Mean SE BWI at dawn (kg m 3 ) < n.s. BWI at dusk (kg m 3 ) < n.s. Relative weight loss (%) n.s <0.05 a 2-tailed Wilcoxon s matched-pairs signed-ranks test weight than dominant great tits during the night in the both areas (Table 2). However, these differences are partially due to differences in body size between birds of different status (see next section). Body weight index Fig. 1 The effect on the body weight (mass) index (kg m 3 ) at dusk of subordinate ( ) great tits that rose in rank after removal of dominant individuals, and dominant ( ) great tits in control flocks after removal of subordinates in one- and two-predator areas. Bars show means±se Dominants were larger than subordinates among the birds weighed when wing length was used as a measure of size (one-predator area: dominants=78.74±0.39 mm, subordinates: 75.02±0.43 mm, 2-tailed paired t-test: t=5.57, df=9, P<0.01; two-predator area: dominants: 78.58±0.24 mm, subordinates=75.16±0.28 mm, 2-tailed paired t-test: t=4.18, df=7, P<0.01). Using the BWI, dominants were found to carry significantly lower body reserves that subordinates at dawn in the one predator area (Table 2). The BWIs of dominants and subordinates did not differ in the two-predator area. Hence, dominants and subordinates have roughly equal reserves at the beginning of daily activities under conditions of higher risk. The dawn BWI of dominants was significantly different at the two sites (2-tailed Wilcoxon s matched-pairs signed-ranks test: t=2, n=8, P<0.05), whereas morning reserves of subordinates were found to be similar (2- tailed Wilcoxon s matched-pairs signed-ranks test: t=6, n=8, P>0.05). The pattern of larger reserves in subordinates also became evident in comparisons of fatness at dusk where individual size was controlled through the BWI, which increased from dominants to subordinates. BWIs of subordinates and dominants did not differ at dusk in the two-predator area. By contrast dominants had lower reserves than subordinates in the one-predator site, and their body reserves were also significantly lower than those carried by dominant individuals in the two-predator area (Table 2). Removal experiments In the one-predator area body reserves of great tits at dusk were affected both by the type of bird (subordinate from removal flocks or dominant from control flocks) and time (Fig. 1). The interaction between type of bird and change in BWI was significant (before removal or after removal) (repeated ANOVA: type of bird:

5 F 1,23 =6.77, P<0.05; time effect: F 1,23 =28.24, P<0.0001; type of bird time: F 1,23 =41.34, P<0.0001) showing that the extent of changes in BWI was dependent on the type of bird. A separate analysis for the two-predator area showed a significant difference for differences between types of birds but no significant effect of time (before removal or after removal), nor was there a significant interaction (repeated ANOVA: type of bird: F 1,23 =4.32, P<0.05; time effect: F 1,23 =1.16, P>0.05; type of bird time: F 1,23 =0.85, P>0.05). Discussion The increased BWI of dominants in a two-predator patch only 100 km away is a result that conflicts with the theoretical predictions. At high latitudes food availability is considered to be a key factor constraining strategic decisions where food supplies are not sufficiently high, and the optimal level of reserves is therefore reduced (Koivula et al. 1995). However, there is evidence that food resources were similar in the two areas differing in predator incidence (I. Krams, T. Krams and J. Cernihovics, unpublished work). Hence, food may not cause a remarkable change in the fattening strategy of dominant great tits. However, it was found that the duration of the activity day of dominants was significantly shorter than the duration of activity of subordinate tits in the twopredator patch. Greater body reserves of dominants in the two-predator area could be a consequence of the increased feeding unpredictability (Bednekoff and Krebs 1995). What was the cause of this increased unpredictability and why were dominant great tits active longer in the single-predator patch? It is considered that the amount of reserves carried by wintering birds depends on food availability and exposure to predators (Lima 1986; McNamara and Houston 1990). However, it must depend also upon the individual s ability to assess the relative risk of starvation and mass-dependent predation, which may vary between years and locations (Bednekoff et al. 1994). Information on feeding conditions can be integrated almost constantly over time. Since the predators were seen or heard every day in the vicinity of feeders, I judge that each individual great tit had complete information about predation occurring within the population with which to assess the danger to itself. Hence I assume that both the starvation risk and the predation risk can be equally well predicted and therefore provide accurate values for the birds internal model, which integrates the relative risks with regulation of body reserves. The results of this study are in accordance with findings of some previous studies showing that subordinate individuals carry more body reserves towards the end of their activity day (Ekman and Lilliendahl 1993; Gosler 1996; Krams 1998a, 1998b), as found in the sparrowhawk area. It shows that the extra body reserves of subordinates corresponded to more unpredictable access to food resources, suggesting that body reserves are under 151 the control of a foraging strategy and are not simply related to food availability (McNamara and Houston 1990; Bednekoff and Krebs 1995). This fattening strategy no doubt lowers the probability of energetic shortfalls. Extra reserves in subordinate great tits could be a buffer against times of limited food resources, for example, when dominants claim their right of access to food (Clark and Ekman 1995). The reduced fat reserves of dominant individuals could be explained by foraging predictability rather than the result of a dominance cost (Hogstad 1987) or starvation (Jansson et al. 1981). Otherwise, dominants with their lower body reserves and greater energy demand while roosting (Røskaft et al. 1986) would be at a higher risk of an energetic shortfall than subordinates. The results of removal experiments in the sparrowhawk area also confirmed the suggestion that dominant individuals foraged optimally, maintaining a neutral energy balance. This is evident from the declining BWI of subordinates after temporal removal of dominant individuals. Dominant individuals are usually considered to arrive earlier in the morning than subordinates in habitats where the sparrowhawk is present (De Laet 1985). This was confirmed in the present study. Dominant tits indeed arrived in the darkness and started to feed significantly earlier than subordinates in the sparrowhawk habitat. Predator avoidance in twilight conditions allowed the birds to restore the energy reserves before sparrowhawks started hunting (Lima 1988a). It is worthwhile mentioning that dominants handled sunflower seeds at the feeders, which was not observed during the normal daylight when they usually handled seeds in the nearest bushes. If they postponed their feeding activity till dawn at the one-predator site, the dominant tits might have a greater probability of being attacked by the sparrowhawks. In the two-predator area dominants arrived to feeders significantly later than subordinates. Dominants usually started to feed in good light conditions when the maximum hunting activity of the pygmy owl supposedly was already over (Lahti et al. 1997). Overall, the results strongly suggest that the early morning initiation of daily feeding is perceived by tits as a potentially risky affair. Thus, the dominant tits avoided feeding in the dim light when both nocturnal and diurnal predators are abroad simultaneously (Mikkola 1983; I. Krams, personal observations). Such predators may use the cover of dimness to initiate attacks since they undoubtedly have a relative advantage over potential prey at this time. Studies indicate that social birds can benefit directly from their own vigilance (Lima 1994; Krams 1998c; Lima and Bednekoff 1999). Under good light conditions a relationship between rank and ability to choose the safest foraging behaviour may be crucial to detect attacking predators. Trading safety for feeding efficiency, dominants decreased the duration of their activity day and the changes in their fattening strategy followed as a result. Starvation risk in the morning because of the reduced length of the activity day should increase more than any mass-dependent costs leading to greater

6 152 body reserves carried by dominant great tits at dusk (Clark and Ekman 1995; Lima 1988a, 1988b; Lahti et al. 1997). The amount of reserves is also dependent on the ability of an individual to restore them while dominants were free to feed, displacing subordinates according to the definition of social dominance (Wilson 1975). The results presented show the complexity of ways in which different types of predators may combine to influence behavioural decision making under the risk of predation. Acknowledgements I thank Markku Orell, Peter A. Bednekoff, Jukka Suhonen, Ilva Everte and two anonymous referees for valuable comments which greatly improved the manuscript. Financial support was received from the Science Council of Latvia. All animal manipulations reported were carried out in accordance with the legal and ethical standards of the Republic of Latvia. References Bednekoff PA, Houston AI (1994) Optimizing fat reserves over the entire winter: a dynamic model. Oikos 71: Bednekoff PA, Krebs JR (1995) Great tit fat reserves: effects of changing and unpredictable feeding day. Funct Ecol 9: Bednekoff PA, Biebach H, Krebs J (1994) Great tit fat reserves under unpredictable temperatures. J Avian Biol 25: Clark CW, Ekman J (1995) Dominant and subordinate fattening strategies: a dynamic game. Oikos 72: Cuthill IC, Houston AI (1997) Managing time and energy. In: Krebs JR, Davies NB (eds) Behavioural ecology: an evolutionary approach. Blackwell, Oxford, pp De Laet JF (1985) Dominance and anti-predator behaviour of great tits Parus major: a field study. Ibis 127: Ekman J (1986) Tree use and predator vulnerability of wintering passerines. Ornis Scand 17: Ekman JB, Lilliendahl K (1993) Using priority to food access: fattening strategies in dominance-structured willow tit (Parus montanus) flocks. Behav Ecol 4: Garnett MC (1976) Some aspects of body size in the great tit. DPhil thesis, University of Oxford Gosler AG (1987) Some aspects of bill morphology in relation to ecology in the great tit, Parus major. DPhil thesis, University of Oxford Gosler AG (1996) Environmental and social determinants of winter fat storage in the great tit Parus major. J Anim Ecol 65:1 17 Gosler AG, Greenwood JD, Perrins C (1995) Predation risk and the cost of being fat. Nature 377: Haftorn S (1994) Diurnal rhythm of passerines during the polar night at Pasvik, North Norway, with comparative notes from South Norway. Fauna Norv Ser C Cinclus 17:1 8 Hegner RE (1985) Dominance and anti-predator behaviour in blue tits (Parus caeruleus). Anim Behav 33: Hogstad O (1987) Social rank in winter flocks of willow tits Parus montanus. Ibis 129:1 9 Houston AI, McNamara JM (1993) A theoretical investigation of the fat reserves and mortality levels of small birds in winter. Ornis Scand 24: Jansson C, Ekman J, Brömssen A von (1981) Winter mortality and food supply in tits Parus spp. Oikos 37: Järvi T, Bakken M (1984) The function of the variation in the breast stripe of the great tit (Parus major). Anim Behav 32: King JR (1972) Adaptive periodic fat storage by birds. Proc Int Ornithol Congr 15: King JR, Farner DS (1966) The adaptive role of winter fattening in the white crowned sparrow with comments on its regulation. Am Nat 100: Koivula K, Orell M, Rytkönen S, Lahti K (1995) Fatness, sex and dominance: seasonal and daily body mass changes in willow tits. J Avian Biol 26: Krams I (1996) Predation risk and shifts of foraging sites in mixed willow and crested tit flocks. J Avian Biol 27: Krams I (1998a) Rank-dependent fattening strategies of willow tit Parus montanus and crested tit P. cristatus mixed flock members. Ornis Fenn 75:19 26 Krams I (1998b) Individuals adjust their body reserves to dominance position within mixed flocks of the willow (Parus montanus) and the crested tit (P. cristatus): a field experiment. Pol J Ecol 46: Krams I (1998c) Dominance-specific vigilance in the great tit. J Avian Biol 29:55 60 Kullberg C (1995) Strategy of pygmy owl while hunting avian and mammalian prey. Ornis Fenn 72:72 78 Kullberg C (1998a) Spatial niche dynamics under predation risk in the willow tit Parus montanus. J Avian Biol 29: Kullberg C (1998b) Does diurnal variation in body mass affect take-off ability in wintering willow tits? Anim Behav 56: Lahti K, Koivula K, Orell M (1997) Dominance, daily activity and winter survival in willow tit: detrimental cost of long working hours? Behaviour 134: Lehikoinen E (1987) Seasonality of the daily weight cycle in wintering passerines and its consequences. Ornis Scand 18: Lilliendahl K (1997) The effect of predator presence on body mass in captive greenfinches. Anim Behav 53:75 81 Lilliendahl K (1998) Yellowhammers get fatter in the presence of a predator. Anim Behav 55: Lima SL (1986) Predation risk and unpredictable feeding conditions: determinants in body mass in birds. Ecology 67: Lima SL (1988a) Initiation and termination of daily feeding in dark-eyed juncos: influences of predation risk and energy reserves. Oikos 53:3 11 Lima SL (1988b) Vigilance during the initiation of daily feeding in dark-eyed juncos. Oikos 53:12 16 Lima SL (1994) On the personal benefits of anti-predatory vigilance. Anim Behav 48: Lima SL, Bednekoff PA (1999) Back to the basics of antipredatory vigilance: can nonvigilant animals detect attack? Anim Behav 58: McNamara JM, Houston AI (1990) The value of fat reserves and the tradeoff between starvation and predation. Acta Biotheor 38:37 61 Mikkola H (1983) Owls of Europe. Poyser, Calton Newton I (1986) The sparrowhawk. Poyser, Calton Orell M (1989) Population fluctuations and survival of great tits Parus major dependent on food supplied by man in winter. Ibis 131: Piper WH, Wiley RH (1990) The relationship between subcutaneous fat and annual survival in wintering whitethroated sparrows (Zonotrichia albicollis). Behav Ecol Sociobiol 26: Priednieks J, Strazds M, Strazds A, Petrinš A (1989) Latvian breeding bird atlas. Zinā tne, Rīga Røskaft E, Järvi T, Bakken M, Bech C, Reinertsen RE (1986) The relationship between social status and resting metabolic rate in great tits (Parus major) and pied flycatcher (Ficedula hypoleuca). Anim Behav 34: Saitou T (1979) Ecological study of social organization in the great tit, Parus major L. III. Home range of the basic flocks and dominance relationships of the members in a basic flock. Misc Rep Yamashina Inst Ornithol 11: Sokal RR, Rohlf FJ (1995) Biometry. Freeman, San Francisco

7 153 Strazds M, Avotiņš A, Ķuze J, Cī rulis V (1998) Breeding bird atlas of the Krustkalni state nature reserve. Putni Dabā 7:7 52 Suhonen J (1993) Predation risk influences the use of foraging sites by tits. Ecology 74: Suhonen J, Halonen M, Mappes T (1993) Predation risk and the organization of the Parus guild. Oikos 66: Summers RW, Feare CJ (1995) Roost departure by European starling Sturnus vulgaris: effects of competition and choice of feeding site. J Avian Biol 26: Svensson L (1984) Identification guide to European passerines. Naturhistoriska Riksmuseet, Stockholm Transehe N (1965) Vogelwelt Lettlands. Dohren, Hannover Veasey JS, Metcalfe NB, Houston, DC (1998) A reassessment of the effect of body mass upon flight speed and predation risk in birds. Anim Behav 56: Verhulst S, Hogstad O (1996) Social dominance and energy reserves in flocks of willow tits. J Avian Biol 27: Wilson EO (1975) Sociobiology. Belknap, Cambridge Witter MS, Cuthill IC (1993) The ecological costs of avian fat storage. Philos Trans R Soc Lond B 340:73 92 Witter MS, Cuthill IS, Bonser RHC (1994) Experimental investigations of mass-dependent predation risk in the European starling, Sturnus vulgaris. Anim Behav 48: