Behavioral Ecology Vol. 11 No. 2: 189-195
© 2000 International Society for Behavioral Ecology
Body mass regulation in response to changes in feeding predictability and overnight energy expenditure
Centre for Behavioural Biology, School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, UK
Address correspondence to I. Cuthill. E-mail: I.Cuthill{at}bristol.ac.uk .
Received 16 March 1999; revised 22 July 1999; accepted 7 August 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Feeding and fat storage entail both costs and benefits. Benefits include minimizing the risk of starvation; costs include mass-dependent costs of locomotion and predation risk. An understanding of these costs and benefits is relevant not only to explanations of foraging patterns and fat storage, but to hoarding decisions, migration strategies, and population dynamics. Despite predictions from theoretical models, empirical tests of the assumptions and predictions of models have been tested only recently. However, published experiments on the effects of unpredictability have often confounded manipulations of mean, variability, and predictability of the food supply, all of which are predicted to affect foraging intensity and fat storage. In experiments on European starlings, Sturnus vulgaris, we manipulated the predictability of the food supply while holding the mean and average variability constant. We did this in conjunction with manipulation of overnight energy expenditure via simulated nocturnal wind exposure. Both greater unpredictability of food availability and higher overnight energy expenditure increased daily mass gain and dusk (lean and fat) mass, but in a purely additive fashion. Dawn mass only changed in response to predictability, not overnight energy expenditure. By introducing a probe day, with identical feeding experience for all treatments, we ascertained that the response to predictability was based on experience integrated over more than a single day.
Key words: fat storage, mass regulation, starvation, trade-offs, Sturnus vulgaris.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Trade-offs between the costs and benefits of fat storage have been used to explain variation between and within populations, seasonal and diurnal variation within individuals, and alternative strategies of energy budgeting in birds (reviews by Cuthill and Houston, 1997
; Lee, 1999;
Pravosudov and Grubb, 1997;
Witter and Cuthill, 1993). A
cost-benefit view of mass regulation is central to models of daily routines of
behavior (Bednekoff and Houston,
1994; Houston et al.,
1993; Hutchinson et al.,
1993; McNamara et al.,
1994). That theory and experiment have focused largely on birds is
not surprising, given the remarkable feat of 15-g endotherms surviving subzero
temperatures for the duration of a Scandinavian or Canadian winter (e.g.,
Lehikoinen, 1987;
Blem, 1990). Small birds in
winter face the double jeopardy of higher energetic demands (colder weather,
longer nights when the bird cannot feed, reduced shelter by foliage, higher
wind exposure) and reduced food supply (shorter days in which to feed, reduced
and more unpredictable food supply). All such deteriorations in the energetic
environment are likely to favor storage of greater energy reserves
(Bednekoff and Houston, 1994;
Houston and McNamara, 1993;
Houston et al., 1993;
Lima, 1986;
McNamara, 1990;
McNamara and Houston, 1990;
McNamara et al., 1994) for a
given level of cost of carrying these reserves (see, e.g.,
Witter and Cuthill, 1993;
Witter et al., 1994).
It is not clear which proximate factors produce adaptive changes in body
mass (Lee, 1999). Birds could
assess local conditions directly or use short-term predictive cues (such as
changes in the weather) or long-term cues (such as photoperiod). The optimal
strategy will depend on the predictive power of the cues, the cost of tracking
them, the benefit of energy storage if the conditions change, and the cost of
storing energy should conditions fail to change (see discussion and references
in Cuthill and Houston, 1997).
Certainly, there appears to be variation between species, some adjusting mass
in a preprogrammed manner with season, others responding to local changes in
the environment (Blem, 1990;
Dawson and Marsh, 1986;
Rogers et al., 1994;
Rogers, 1995;
Waite, 1992; see review by
Pravosudov and Grubb, 1997).
The goal of the present study was to investigate the effects of predictability
of the food supply, acting in conjunction with changes in overnight energetic
expenditure. This focus results from a lack of experiments separating effects
of foraging unpredictability from correlated changes and from studies
investigating potential nonadditive effects of simultaneous deterioration of
more than one factor in the environment.
Changes of fat reserves with experimental manipulation of local
environmental conditions have been investigated with regard to mean
temperature (Ekman and Hake,
1990; Kontogiannis,
1967; Kendeigh et al.,
1969), variability in temperature
(Bednekoff et al., 1994),
sudden changes in overnight temperature
(Lilliendahl et al., 1996),
the length of the entire feeding day
(Bednekoff and Krebs, 1995),
interruptions of feeding during the day
(Dall and Witter, 1998;
Ekman and Hake, 1990;
Witter and Swaddle, 1997;
Witter et al., 1995), and
unpredictability of food supply (Hurly,
1992; Witter et al.,
1995). However, in no case has the effect of predictability been
separated from variability, and in several cases the unpredictable/variable
treatment was also one with poorer average conditions than the control (e.g.,
Dall and Witter, 1998;
Ekman and Hake, 1990), or with
greater extremes (Bednekoff and Krebs,
1995; Bednekoff et al.,
1994). Witter et al.
(1995) sought to discriminate
effects of unpredictability from the actual time of interruption: fixed-length
interruptions could occur in the morning, in the afternoon, in the morning or
afternoon with equal probability, or not at all (control). This experimental
design was effective in disentangling effects of variable timing of
deprivation from the time of day at which deprivation occurred (see discussion
in Cuthill and Houston, 1997),
but it could not separate effects of variability from predictability. That is,
food deprivation that occurs in the morning or afternoon could influence fat
storage because the deprivation can occur at different times of day
(variability) regardless of whether the timing of shortfall is predictable or
not. In this context, it is notable that Witter et al.
(1995) did not systematically
record the actual timing of deprivations on the days immediately before each
mass measurement; they simply recorded mass in between deprivation days. If
response to, say, afternoon and morning deprivation is asymmetric, then
recording the mean mass in the variable deprivation treatment cannot reveal
whether the birds are responding to variability over and above the additive
responses to morning and afternoon deprivations in isolation. That is, the
response to variability may in fact be explicable in terms of the sum of
responses to local conditions within any one day. To separate these
possibilities we used an experimental design in which (1) treatment groups
experienced equivalent histories of variability in deprivations, but differed
in the predictability of shortfalls, and (2) local (within-day) conditions on
the test day were identical, all that differed being the prior history of
predictability. Furthermore, we investigated the influence of these effects
acting in conjunction with different levels of overnight energetic
expenditure.
Temperate birds in winter experience simultaneous deterioration of several
factors expected to influence energy budgets (see above). The joint action of
two factors negatively influencing the energetic environment is predicted to
have a synergistic effect on starvation risk
(McNamara and Houston, 1990).
Accordingly, our experiment was designed to investigate possible nonadditive
effects of manipulating both feeding predictability and energetic expenditure
on fat storage and body mass. The aspect of energetic expenditure manipulated
in the current experiment was nocturnal wind exposure. There is correlational
evidence that this affects body mass in the chosen study animal, the European
starling Sturnus vulgaris, in the wild. Peach et al.
(1992), measuring wild
starlings captured at roost sites, found that mean body masses of adult
starlings were higher on relatively windy nights. Furthermore, controlling for
this relationship, mean body mass was negatively correlated with both
long-term average temperature and, in the case of adult males at least,
prevailing temperature. A direct proximate influence of anticipated nocturnal
wind exposure on body mass was demonstrated by Witter et al.
(1994), so we had some
confidence that the treatment would affect mass regulation in this
species.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We used six wild-caught, adult female starlings (Sturnus vulgaris), housed at 16 ± 2°C on a 10 h light: 14 h dark photoperiod. Under these conditions starlings are photosensitive (Nicholls et al., 1988
), and
in this physiological state they have previously been shown to respond to
manipulations of the energetic environment by changing body mass
(Witter et al., 1995). Birds
were housed singly in 1.0 x 0.3 x 0.4 m cages, in visual but not
acoustic isolation, and had ad libitum access to drinking water. After the
experiment, they were acclimatized in large, communal flight aviaries open to
ambient environmental conditions and, after veterinary inspection,
released.
Each starling obtained food from an operant feeder (Campden Instruments
model 442 pellet dispenser), which delivered an aliquot of approximately 0.1 g
turkey crumbs each time the bird pushed open a 0.03 x 0.03 m hinged
transparent plastic window in front of a hopper. However, the feeder was only
active, and the birds could only obtain food, when a green cue light above the
window was illuminated. Birds were trained to this feeding response by 2 days
of exposure to automatic delivery of crumbs at random intervals between 1 and
120 s. Each delivery was paired with illumination of the cue light, the latter
being extinguished when the birds pushed open the window to eat the food item.
The starlings rapidly learned to push open the window to access the food and
to associate the availability of food with the green cue light. In the
terminology of operant conditioning, training consisted of a random interval
schedule with food delivery paired with illumination of a cue light, and the
experimental schedule was a fixed ratio of 1 (i.e., continuous reinforcement)
but available only when the cue light was illuminated. During the experiment
the starlings therefore had control of how much they ate, but the periods when
food was available were under experimental control, with availability obvious
from illumination of the cue lights. During both training and the experiment,
food was never available during the last 15 min of the day, to allow birds to
settle to roost, and for the first 45 min of the day, to allow the cage lining
paper and drinking water to be changed and all equipment to be checked. (In
practice this disturbance lasted less than 15 min.) Operant control and
recording of feeding attempts were via a computer running the Spider system
(Paul Fray Ltd.; see, e.g., Cuthill et al.,
1990; Dall et al.,
1997). Although the operant feeders detected bird-induced food
deliveries, rather than consumption per se, we never found unconsumed items in
the hoppers at the end of the day, so these data accurately reflect
consumption.
There were four experimental treatments, corresponding to the 2 x 2 factorial combination of manipulations of both predictability of food availability and overnight energetic expenditure. Each bird experienced each of the four treatments, using one-and-a-half Latin squares to balance order as far as possible; that is, four birds experienced treatment orders according to a Latin square, and the remaining two birds received treatments according to two rows of another Latin square. With four treatments and six birds, order and treatment are potentially confounded, but we tested for this explicitly after the experiment. Each treatment lasted for 8 days, with the last day constituting a probe day during which conditions were identical for all birds, and when the key experimental data were collected (see below).
Food availability could be either predictable or unpredictable. In the
predictable treatment, a bird's feeder was either active or inactive in
alternating half-hour blocks throughout the day
(Figure 1). In the
unpredictable treatment, the feeding day was similarly divided into half-hour
blocks, but whether the feeder was active during any one block was
probabilistic, with p =.5. The cue light was illuminated when the
feeder became active so that a bird did not have to sample to check the
feeder's state. In addition, although feeder availability was random, we
screened the computer-generated sequences in advance to check that no day had
a particularly high proportion of time when food was unavailable. There were
two reasons for this: for the birds' welfare (a day with several sequences of
nonavailability of the feeder could lead to a real danger of starvation), and
also to ensure that predictability of food availability (the desired
manipulation) was not confounded with total duration of food availability (see
Hurlbert, 1984, on random
interspersion of treatments). Thus both treatments offered the same number of
hours of food availability (4.5 h), but in one treatment the bird experienced
a perfectly predictable sequence of availability and nonavailability, and in
the other the bird could not predict when the feeder would be active
(Figure 1).
|
As the predictable and unpredictable treatments necessarily provided different short-term experience of food gain during each day, any effects on body mass we detected could have been a response to local (within-day) conditions rather than predictability. For example, a bird on the unpredictable treatment may, by chance, not be able to feed during the middle portion of the day and thus feed intensively when food, again by chance, becomes available toward the end of the day. Thus a higher dusk mass in the unpredictable treatment may be a response not to the average predictability of feeding conditions, but to the immediate run of bad luck it experienced in the middle of the day. To deconfound within-day experience from average predictability, we gave birds a probe day on the eighth (final) day of each treatment. On this day, all birds experienced identical within-day conditions of food availability—namely, the perfect alternating food availability of the predictable treatment. For birds on the predictable treatment, day 8 would be just another predictable day. For birds on the unpredictable treatment, the perfect alternation would be just another peculiar sequence of half-hour blocks of availability and nonavailability. Thus any differences in mass accumulation between treatments on the probe days would have to be due to differences in experienced feeding predictability over the preceding week, rather than a response to immediate (within-day) conditions.
Overnight energetic expenditure was manipulated by using electric fans to
simulate wind exposure, a technique successfully used by Witter et al.
(1994). Fans (Carlton model
Air 12) were placed 1 m in front of each cage and swiveled back and forth
automatically so that the starlings could not easily avoid the air flow. The
wind speed was between 0.1 and 0.6 m/s, depending on the exact position within
the cage. As all cages were within the same room, the disturbance from noise
was similar for all birds, so the level of air flow was the only difference
between treatments. These were simply "fan on" or "fan
off." Wind exposure increases not only energetic expenditure but also
evaporative water loss, so changes in body mass reflect changes in both fat
and water content. We estimated body fat independently of body mass (see
below) to separate these effects.
Body mass was recorded automatically throughout the day using perches
attached to electronic balances (Ohaus model E400) interfaced to a computer.
The measurements were accumulated and stable readings estimated to the nearest
0.1 g using Nestbug software
(Szép et
al., 1995). As starlings sat on the perches intermittently, and
not all readings stabilized, we used median values for each 1-h block in the
final analyses. In addition to automatic recording, at dusk on each probe day
we caught the birds and weighed them manually to the nearest 0.1 g. At the
same time, we measured total body electrical conductivity (TOBEC) by placing
each bird in an EM-Scan Inc. model SA-1 small animal body composition
analyzer. Each bird was measured five times, removing and replacing the animal
between each measurement to randomize error due to subject placement relative
to the electromagnetic coil (Asch and Roby,
1995). As fat has low conductance, the TOBEC reading is
proportional to lean mass, which can be estimated using a calibration equation
derived for the starling by Scott et al.
(1991). We calculated fat mass
by subtracting TOBEC-derived lean mass from directly measured total mass. The
reliability of TOBEC as an estimator of lean mass is quite variable for a bird
the size of a starling (Asch and Roby,
1995; Meijer et al.,
1994), but the error should have been low in our study because the
birds were fully hydrated, dry, warm, and close to the maximum size for the
measuring chamber (Asch and Roby,
1995). Use of the calibration equation of Meijer et al.
(1994) produced equivalent
results, both in pattern and significance.
In the analyses of probe day data we used the manually measured dusk
masses, these being closer to true dusk mass than the electronic balance
measures of average mass in the last hour of the day. For dawn mass, we used
the mass measured electronically in the first hour of the following day (i.e.,
the first after experiencing a probe day). For these reasons, the dawn and
dusk masses for the probe day analyses
(Figure 2) and the equivalent
values for the within-day mass trajectories (based on probe day electronic
data; see Figure 4) are not
identical. All statistical analysis was by repeated-measures ANOVA using
Minitab (Ryan and Joiner,
1994), after verifying the appropriateness of parametric
tests.
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mass at dusk, as measured manually after catching the birds at the end of the probe days, was higher for birds in the unpredictable than in the predictable treatment (Figure 2a; main effect of predictability, F1,5 = 86.56, p <.001). Dusk mass was also higher for birds in the fan-on than in the fan-off treatments (main effect of nocturnal wind exposure, F1,5 = 52.39, p <.001), and there was no significant interaction between feeding predictability and wind exposure (F1,5 = 2.93, p =.147). The changes in dusk mass were due to effects of these treatments on both fat and lean mass (Figure 3; fat mass, main effect of predictability, F1,5 = 19.22, p =.007; of fan, F1,5 = 29.00, p =.003; interaction, F1,5 = 0.19, p =.683; lean mass, main effect of predictability, F1,5 = 13.60, p =.014; of fan, F1,5 = 8.14, p =.036; interaction, F1,5 =.55, p =.493). From electronic measurements of dawn mass on the first day after the probe day, there were effects of predictability (Figure 2b; birds on the unpredictable treatment had higher dawn masses; F1,5 = 11.52, p =.019), but not wind exposure (F1,5 = 0.29, p =.611) or an interaction between these factors (F1,5 = 4.54, p =.086). The differences in dusk mass were not simply due to different masses at the start of the probe day, as birds gained significantly more mass on probe days (Figure 2c) when experiencing nocturnal wind exposure (F1,5 = 9.08, p =.030) and feeding unpredictability (F1,5 = 6.80, p =.048), with no significant interaction (F1,5 = 2.35, p =.186). However, the differences in mass gain did not seem to be due to differences in food consumption. Although the trends were in the same direction as for mass gain, there were no statistically significant effects of treatment on the number of food items consumed over the day, as measured by the operant feeders (Figure 2d; predictability, F1,5 = 0.93, p =.378; wind exposure, F1,5 = 0.13, p =.729; interaction, F1,5 = 0.39, p =.562).
|
The data also allow us to calculate overnight mass loss on the probe day. Wind exposure induced greater mass loss overnight (F1,5 = 51.97, p <.001), in the face of whatever compensatory responses the birds employed. Interstingly, birds also lost more overnight mass when in the unpredictable treatment (F1,5 = 16.27, p <.010), again with no predictability x fan interaction, F1,5 = 2.10, p =.207).
Predictability and wind exposure both affected the daily pattern of mass gain on the probe day (Figure 4; predictability x hour, F8,40 = 4.67, p <.001; fan x hour, F8,40 = 2.40, p =.032), with no significant interaction (predictability x fan x hour, F8,40 = 0.27, p =.973). The differences in mass between treatments show a similar pattern from dawn to dusk, but the differential is greater in the last few hours of the day (Figure 4). In the predictable/fan-off treatment, birds ceased to gain mass in the last couple of hours of the day, whereas the effect of either feeding unpredictability or forthcoming nocturnal wind exposure was to prolong mass gain. In the most energetically stressful treatment (unpredictable/fan on), birds gained mass at a fairly constant rate throughout the day.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The starlings in our experiment responded to both an unpredictable food supply and overnight wind exposure by gaining more mass during the day and reaching a higher body mass at dusk. As mass is higher at both dawn and dusk, and daily mass gain is greater, this result is in accordance with the "true winter fattening" model of Lehikoinen (1987
). The response to
overnight wind exposure is not novel, having been documented in wild starlings
(Peach et al., 1992) and
experimentally demonstrated by Witter et al.
(1994). However, this is the
first study to demonstrate clearly an effect of unpredictability as opposed to
variability in food supply and the first to investigate the effects of two
factors simultaneously affecting energy budgets. Previous experiments have not
separated the effects of unpredictability and variability in food
availability, although some have separated the effects of variable deprivation
and deprivation per se. However, it is noteworthy that in the latter studies,
treatment differences could be interpreted as a response to the worst
conditions recently experienced rather than variability itself
(Bednekoff and Krebs, 1995;
Bednekoff et al., 1994;
Witter et al., 1995).
Although our results cannot be attributed to different feeding experiences
within the probe day, we still cannot conclude that the birds are encoding
predictability. By "encode" we mean that they have an internal
representation, either cognitive or physiological, for this particular quality
of the external environment (cf. discussions on risk-sensitivity in foraging
decisions; Kacelnik and Bateson,
1996). In the unpredictable treatment before the probe day, birds
experienced the same average amount of feeding time per day as in the
predictable treatment, but experienced some individually longer periods of
time without food ("runs of bad luck"; see
Ydenberg, 1984). The
predictable treatment never exposed birds to more than 30 min without food,
whereas in the unpredictable treatment birds experienced longer runs of both
good and bad luck. This is a plausible mechanism by which such patterns of
unpredictable food availability could evoke behavioral or physiological
adjustments of energy use. With a regular pattern of food availability, the
birds can rapidly adapt to a regular routine of feeding (starlings readily
learn patterns of food availability at different temporal scales; e.g.,
Bateson and Kacelnik, 1995;
Dall et al., 1997;
Wenger et al., 1991). The
irregular availability in the unpredictable treatment will result in a
mismatch between motivation, or attempts, to feed and the ability to do so. In
birds, feeding disruption has been shown to stimulate production of
glucocorticosteroids and other hormones in the hypothalamo-pituitary-adrenal
cascade, which in turn affects body mass
(Wingfield and Ramenofsky,
1997; Wingfield et al.,
1998). A direct prediction of our tentative mechanistic
explanation of the effects of unpredictability would be that such
interruptions would promote higher, or more prolonged, corticosterone release
than predictable interruptions of equivalent duration.
Although half-hour periods without food might seem a trivial interruption
to foraging, the possibility of several sequential episodes increases
starvation risk, and is predicted to affect the optimal level of energy
reserves (Hutchinson et al.,
1993); McNamara,
1990; McNamara and Houston,
1990; McNamara et al.,
1994). Thus, the effect of unpredictable food availability on body
mass observed in this experiment is, at face value, what state-dependent
models of mass regulation predict (see also
Lehikoinen, 1987;
Lima, 1986). However, these
models have assumed, not unreasonably, that mass gain is via increased food
intake, whereas the different mass gains in our experiment were not
attributable to any statistically discernible differences in foraging rates.
Therefore we infer that the predicted alterations in daily mass gain were
achieved either by altered digestive efficiency or (more likely given the
speed of response) reduced activity and metabolic expenditure. The latter has
been found in great tits (Parus major) responding to variable
overnight temperatures and night lengths
(Bednekoff and Krebs, 1995;
Bednekoff et al., 1994). The
higher dusk masses in our experiment were due to increases in both fat and
lean mass. The higher lean mass, being a measure of all water-soluble material
in the bird, could be due both to greater undigested gut contents and water
retention. As nocturnal wind exposure must increase evaporative water loss, it
would be interesting to investigate whether birds can respond to this by
anticipatory water consumption, just as they respond to higher overnight
energy expenditure by anticipatory weight gain. The present data cannot
distinguish between elevated lean mass due to water and that due to undigested
food.
Higher mass at dawn in response to feeding unpredictability was associated
with higher mass gains over the probe day, but birds also maintained a higher
dawn mass in this treatment. This, the 'true winter fattening' response of
Lehikoinen (1987), is what
might be expected when a run of bad luck (no feeding) is possible at dawn
(Hutchinson et al., 1993;
McNamara et al., 1994). It is
noteworthy, therefore, that nocturnal wind exposure elevated daily mass gain
and mass at dusk, but not dawn mass. With anticipated higher energy
expenditure at night, the starlings reached a dusk mass, and/or adjusted
energy loss overnight, so as to attain the same dawn mass as in the 'no fan'
treatments. As nocturnal wind exposure increases the possibility of overnight
starvation but not, by itself, an energy shortfall during the day, this is
also consistent with the logic of state-dependent models.
Nocturnal wind exposure and feeding unpredictability have additive effects
on mass regulation. Unpredictable foraging increases daily weight gain and
dusk mass by the same amount whether the forthcoming night is likely to incur
high or low energetic expenditure through wind chill. Simultaneous
deterioration of two aspects of the energetic environment are predicted to
have a synergistic effect on starvation risk
(McNamara and Houston, 1990).
However, starvation risk need not be linearly related to energy reserves, far
less body mass, so it is hard to make clear predictions about the anticipated
effects in our experiment without a model tailored to the experimental
environment. As such, we simply present this as an empirical finding which may
or may not be consistent with current theory. Clearly, small birds face
multiple deteriorations in their energetic environment in winter, and this
issue deserves further theoretical and empirical investigation.
The daily mass trajectory of starlings in the mildest (predictable/fan-off)
treatment showed minimal mass gain in the later hours of the day. It is here
that the differences from the other treatments were most marked, with birds
gaining mass at a fairly constant rate throughout the day in the harshest
(unpredictable/fan-on) treatment. These differences are what one might expect
in a low-predation-risk environment (or at least an environment with low
mass-dependent predation risk), as here the premium is on minimizing the
probability of starvation (McNamara et
al., 1994). However, there is no reason to expect that captive
birds are acting as if predation risk is low
(Houston and McNamara, 1989),
and the predicted optimal daily mass trajectories can be quite parameter
sensitive (Hutchinson et al.,
1993; McNamara et al.,
1994). Thus, it may be unwise to treat our observed mass
trajectories as confirming theoretical predictions. Nevertheless, the elevated
reserves at dusk and dawn are qualitatively what one would predict for a small
bird facing both high nocturnal energy expenditure and the risk of diurnal
feeding interruptions of unpredictable duration
(Bednekoff and Houston, 1994;
Houston and McNamara, 1993;
Lehikoinen, 1987;
McNamara and Houston, 1990;
McNamara et al., 1994).
|
ACKNOWLEDGEMENTS |
|---|
We are particularly grateful to Zoltan Tøth (Department of Genetics, Eötvös University, Müzeum krt. 4/A, 1088 Budapest, Hungary; e-mail tothz@falco.elte.hu) for providing us with the Nestbug software for body mass recording, and for customizing the software for our electronic balances. Many thanks also to Mark Witter, with whom this research program was initiated, for years of enjoyable brainstorming, and likewise to John McNamara, Alasdair Houston, and other members of the Centre for Behavioural Biology at Bristol. This research was funded by Natural Environment Research Council grant GR3/ 8924 to I.C.C. and was completed during a sabbatical funded by the Nuffield Foundation.
|
FOOTNOTES |
|---|
E. Jones is now at Silsoe Research Institute, Silsoe, Bedford, MK45 4HS, UK.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Asch A, Roby DD, 1995. Some factors affecting precision of the total body electrical conductivity technique for measuring body composition in live birds. Wilson Bull 107: 306-316.
Bateson M, Kacelnik A, 1995. Preferences for fixed and variable food sources: variability in amount and delay. J Exp Anal Behav 63: 313-329.[ISI][Medline]
Bednekoff PA, Biebach H, Krebs J, 1994. Great tit fat reserves under unpredictable temperatures. J Avian Biol 25: 156-160.
Bednekoff PA, Houston AI, 1994. Avian daily foraging patterns: effects of digestive constraints and variability. Evol Ecol 8: 36-52.
Bednekoff PA, Krebs JR, 1995. Great tit fat reserves: effects of changing and unpredictable feeding day length. Funct Ecol 9: 457-462.
Blem CR, 1990. Avian energy storage. Curr Ornithol 7: 59-113.
Cuthill IC, Houston AI, 1997. Managing time and energy. In: Behavioural ecology, 4th ed (Krebs, JR, Davies, NB, eds), Oxford: Blackwell Scientific; 97-120.
Cuthill IC, Kacelnik A, Krebs JR, Haccou P, Iwasa Y, 1990. Starlings exploiting patches: the effect of recent experience on foraging decisions. Anim Behav 40: 625-640.
Dall SRX, Cuthill IC, Cook N, Morphet M, 1997. Learning about food: starlings, Skinner Boxes and earthworms. J Exp Anal Behav 67: 181-192.
Dall SRX, Witter MS, 1998. Feeding interruptions, diurnal mass changes and daily routines of behaviour in the zebra finch. Anim Behav 55: 715-725.[ISI][Medline]
Dawson WR, Marsh RL, 1986. Winter fattening in the American goldfinch and the possible role of temperature in its regulation. Physiol Zool 59: 357-368.
Ekman JB, Hake MK, 1990. Monitoring starvation risk:
adjustments of body reserves in greenfinches (Carduelis chloris)
during periods of unpredictable foraging success. Behav Ecol
1: 62-67.
Houston AI, McNamara JM, 1989. The value of food: effects of open and closed economies. Anim Behav 37: 546-562.
Houston AI, McNamara JM, 1993. A theoretical investigation of the fat reserves and mortality levels of small birds in winter. Ornis Scand 24: 205-219.
Houston AI, McNamara JM, Hutchinson JMC, 1993. General results concerning the trade-off between gaining energy and avoiding predation. Phil Trans R Soc Lond B 341: 375-397.
Hurlbert SH, 1984. Pseudoreplication and the design of ecological field experiments. Ecol Monogr 54: 187-211.
Hurly TA, 1992. Energetic reserves of marsh tits
(Parus palustris): food and fat storage in response to variable food
supply. Behav Ecol 3:
181-188.
Hutchinson JMC, McNamara JM, Cuthill IC, 1993. Song, sexual selection, starvation and strategic handicaps. Anim Behav 45: 1153-1177.
Kacelnik A, Bateson M, 1996. Risky theories: the effects of variance on foraging decisions. Am Zool 36: 402-434.
Kendeigh SC, Kontogiannis JE, Maza A, Roth R, 1969. Environmental regulation of food intake by birds. Comp Biochem Physiol 31: 941-957.
Kontogiannis JE, 1967. Day and night changes in body weight in the white-crowned sparrow (Zonotrichia albicollis). Auk 84: 390-395.
Lee SJ, 1999. Functional and mechanistic approaches to mass regulation in birds. Biol. Rev (in press).
Lehikoinen E, 1987. Seasonality of the daily weight cycle in wintering passerines and its consequences. Ornis Scand 18: 216-226.
Lilliendahl K, Carlson A, Welander J, Ekman JB, 1996. Behavioural control of daily fattening in great tits (Parus major). Can J Zool 74: 1612-1616.
Lima SL, 1986. Predation risk and unpredictable feeding conditions: determinants of body mass in birds. Ecology 67: 377-385.
McNamara JM, 1990. The starvation-predation trade-off and some behavioural and ecological consequences. In: Behavioural mechanisms of food selection, NATO ASI Series A, Life Sciences (Hughes RN, ed). New York: Springer-Verlag; 39-59.
McNamara JM, Houston AI, 1990. The value of fat reserves and the tradeoff between starvation and predation. Acta Biotheor 38: 37-61.
McNamara JM, Houston AI, Lima SL, 1994. Foraging routines of small birds in winter: a theoretical investigation. J Avian Biol 25: 287-302.
Meijer T, Mohring FJ, Trillmich F, 1994. Annual and daily variation in body mass and fat of starlings Sturnus vulgaris. J Avian Biol 25: 98-104.
Nicholls TJ, Goldsmith AR, Dawson A, 1988. Photorefractoriness in birds and comparison with mammals. Physiol Rev 68: 133-176.
Peach WJ, Hodson DP, Fowler JA, 1992. Variation in the winter body mass of starlings Sturnus vulgaris. Bird Study 39: 89-95.
Pravosudov VV, Grubb TC Jr, 1997. Energy management in passerine birds during the nonbreeding season. A review. Curr Ornithol 14: 189-234.
Rogers CM, 1995. Experimental evidence for temperature-dependent winter lipid storage in the dark-eyed junco (Junco hyemalis oreganus) and song sparrow (Melospiza melodia morphna). Physiol Zool 68: 277-289.
Rogers CM, Nolan V, Ketterson ED, 1994. Winter fattening in the dark-eyed junco: plasticity and possible interaction with migration trade-offs. Oecologia 97: 526-532.
Ryan BF, Joiner BI, 1994. MINITAB handbook. Belmont, California: Duxbury Press.
Scott I, Grant M, Evans PR, 1991. Estimation of fat-free mass of live birds: use of total body electrical conductivity (TOBEC) measurements in studies of single species in the field. Funct Ecol 5: 314-320.
Szép T, Barta Z, Tøth Z, Søvari Z, 1995. Use of an electronic balance with bank swallow nests: a new field technique. J Field Ornithol 66: 1-11.
Waite TA, 1992. Winter fattening in gray jays: seasonal, diurnal and climatic correlates. Ornis Scand 23: 499-503.
Wenger D, Biebach H, Krebs JR, 1991. Free-running circadian rhythm of a learned feeding pattern in starlings. Naturwiss 78: 87-89.
Wingfield JC, Maney DL, Breuner CW, Jacobs JD, Lynn S, Ramenofsky M, Richardson RD, 1998. Ecological bases of hormone-behavior interactions: the "emergency life history stage." Am Zool 38: 191-206.
Wingfield JC, Ramenofsky M, 1997. Corticosterone and facultative dispersal in response to unpredictable events. Ardea 85: 155-166.
Witter MS, Cuthill IC, 1993. The ecological costs of avian fat storage. Phil Trans R Soc Lond B 340: 73-92.[ISI][Medline]
Witter MS, Cuthill IC, Bonser RHC, 1994. Experimental investigations of mass-dependent predation risk in the European starling, Sturnus vulgaris. Anim Behav 48: 201-222.
Witter MS, Swaddle JP, 1997. Mass regulation in juvenile starlings: response to change in food availability depends on initial body mass. Funct Ecol 11: 11-15.
Witter MS, Swaddle JP, Cuthill IC, 1995. Periodic food availability and strategic regulation of body mass in the European starling, Sturnus vulgaris. Funct Ecol 9: 568-574.
Ydenberg RC, 1984. Great tits and giving-up times: decision rules for leaving patches. Behaviour 90: 1-24.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Asher and M. Bateson Use and husbandry of captive European starlings (Sturnus vulgaris) in scientific research: a review of current practice Lab Anim, April 1, 2008; 42(2): 111 - 126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Speakman Obesity: The Integrated Roles of Environment and Genetics J. Nutr., August 1, 2004; 134(8): 2090S - 2105S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Lange and O. Leimar Social stability and daily body mass gain in great tits Behav. Ecol., July 1, 2004; 15(4): 549 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Kelly and W. W. Weathers Effects of feeding time constraints on body mass regulation and energy expenditure in wintering dunlin (Calidris alpina) Behav. Ecol., November 1, 2002; 13(6): 766 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Polo and L. M. Bautista Daily body mass regulation in dominance-structured coal tit (Parus ater) flocks in response to variable food access: a laboratory study Behav. Ecol., September 1, 2002; 13(5): 696 - 704. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||