Behavioral Ecology Vol. 10 No. 6: 648-658
© 1999 International Society for Behavioral Ecology
Predation risk affects trade-off between nest guarding and foraging in Seychelles warblers
Zoological Laboratory, University of Groningen, PO Box 14, 9750 AA Haren, The Netherlands, and Department of Zoology, University of Melbourne, Parkville, Victoria 3052, Australia
Address correspondence to J. Komdeur, Zoological Laboratory, University of Groningen, PO Box 14, 9750 AA Haren, The Netherlands. E-mail: j.komdeur{at}biol.rug.nl .
Received 1 December 1998; revised 29 March 1999; accepted 19 April 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The fitness costs of egg loss for Seychelles warblers (Acrocephalus sechellensis) on Cousin Island are considerable because warblers have a single-egg clutch and no time to lay a successful replacement clutch. On the islands of Cousin and Cousine, with equal densities of Seychelles fodies (Foudia sechellarum), nearly 75% of artificial eggs placed in artificial nests were predated by fodies after 3 days. On Aride Island with no fodies present, loss of artificial eggs was not observed. Female warblers incubate the clutch, and male warblers guard the clutch when females are absent. Deterrence of fodies by male warblers is efficient: loss rate of eggs from unattended warbler nests was seven times as high as from attended nests, and the more nest guarding, the lower the egg loss and the higher the hatching success. Egg loss is independent of the amount of incubation by females. There is no trade-off between incubating and foraging by females. Nest guarding competes with foraging by males, and this trade-off has a more pronounced effect on egg loss when food availability is low. The transfer of breeding pairs from Cousin to either Cousine with egg-predating fodies or to Aride without fodies allowed us to experimentally investigate the presumed trade-off between nest guarding and foraging. On Cousine, individual males spent the same amount of time nest guarding and foraging as on Cousin, and egg loss was similar and inversely related to time spent nest guarding as on Cousin. Males that guarded their clutch on Cousin did not guard the clutch on Aride but allocated significantly more time to foraging and gained better body condition. Loss of warbler eggs on Aride was not observed. Time allocation to incubating and foraging by individual females before and after both translocations remained the same.
Key words: egg defense, egg predation, foraging, parental care, reproductive success, Seychelles warbler, trade-offs.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Parent birds can reduce the probability of predation by perching near the nest to guard and defend it (Ricklefs, 1969
; Greig-Smith,
1980; Hayes and Robertson,
1989; Li and Martin,
1991; Martin,
1992; Martin and Li,
1992; Nilsson,
1984, 1986).
However, the effect of guarding on breeding success depends on the extent to
which predation influences reproductive success and on the existence and
effectiveness of predator deterrence
(Martin, 1992;
Slagsvold, 1982). Parent birds
may either passively defend their young through use of vocalizations that
quiet the young and distract the predator
(East, 1981) or actively defend
the nest with clutch or young by attacking potential predators
(Andersson et al., 1980;
Barash, 1975;
McLean et al., 1986;
Montgomery and Weatherhead,
1988; Wiklund,
1990). Both types of defense may result in increased reproductive
success by reducing nest losses to predation. The effectiveness of guarding is
expected to be higher when both breeding partners guard by taking turns.
However, defense behavior can also be costly (e.g., higher risk of parent
being predated, less time for foraging). Allocation of time to nest guarding
and foraging are mutually exclusive. The time spent guarding may compete with
rate of food provisioning to young through time needed for finding food
(Martin, 1992;
Yom-Tov, 1974). Through this
trade-off, a life-history trade-off between current and future reproduction is
expected (Rose, 1983;
Stearns, 1989,
1992). Therefore, a pronounced
effect on reproductive success is expected when food availability is low
(Arcese and Smith, 1988;
Martin, 1987). Experimental
studies are needed to test the effectiveness of nest guarding and the effects
of predation on reproductive success and on the trade-off between guarding and
foraging of parents (Martin,
1992). So far, the consequences of predation risk for time
allocation during reproduction have not been examined
(Martin, 1993). In this study
we tested experimentally the effects of nest-guarding behavior on egg losses
and the effects of food availability and egg predators on time allocations to
guarding and foraging of parents in the Seychelles warbler (Acrocephalus
sechellensis).
The Seychelles warbler is a rare island endemic, known only from the
Seychelles Islands. Until, 1988, it occurred only on Cousin Island (29 ha),
but additional breeding populations were successfully established on the
islands of Aride (68 ha) and Cousine (26 ha) in September, 1988 and June,
1990, respectively (Komdeur,
1994a). Once paired, warblers reside permanently in the same
territory. Some individuals remain on their natal territories as helpers,
assisting their parents with raising younger siblings
(Komdeur, 1994b). Adult
warblers lack natural predators. The main egg predator is the Seychelles fody
(Foudia sechellarum), a weaver bird also endemic to the Seychelles
(Collar and Stuart, 1985). The
fodies are evenly distributed and equally abundant on the islands of Cousin
and Cousine, but absent on Aride Island
(Collar and Stuart, 1985;
Owen, 1986). The fitness costs
of egg loss are considerable because warblers usually have clutches of a
single egg (91.0%, n = 223), are single-brooded with restricted
breeding seasons, and have no time for a successful replacement clutch
(Komdeur, 1996b). Thus, the
loss of an egg results in the loss of an entire breeding season.
This study was designed to answer four questions. First, what is the influence of Seychelles fodies on egg loss? This was studied by comparing survival of artificial eggs placed in experimental nests on the islands of Cousin and Cousine, with similar density of fodies, and on Aride, without fodies. Second, do Seychelles warblers deter egg-predating Seychelles fodies during incubation? If so, nest guarding against fodies is expected to be effective in lowering egg predation. This was measured by relating nest-guarding intensity with egg loss rates and tested by comparing egg loss rates from warbler's nests with egg loss rates from unguarded artificial nests. Third, is there a trade-off between nest guarding and foraging? In other words, do warblers increase foraging time and reduce nest-guarding time as food abundance is lower or risk of egg predation from fodies is absent? Fourth, what are the consequences of this trade-off for the parents and their (future) reproductive success? The transfers of warbler pairs from Cousin to Cousine and from Cousin to Aride allowed us to test whether a change in predation risk affects allocation of time to foraging and guarding behaviors by comparing changes in time allocation of the same pairs before and after transfer under the same food and climatic conditions. In addition, the transfers allowed us to determine the effect of egg predators on parents' reproductive success.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Study area and data collection
All Seychelles warbler groups were studied from May to August, the main breeding season, each year. On Cousin, 15-18 warbler groups were studied from 1986 to 1996. Over the period of 23-29 September 1988, 29 color-ringed adult Seychelles warblers (16 males and 13 females) were transferred from Cousin to Aride by motorboat, a journey of about 45 min. During the 2 years after the transfer, five pairs that were breeding pairs on Cousin remained pairs on Aride. During a 3-day period from 29 June to 1 July 1990, 29 color-ringed adult warblers (15 males and 14 females) were transferred from Cousin to Cousine, a journey of about 15 min. During the 2 years after the transfer, four pairs that were breeding pairs on Cousin remained pairs on Cousine. On Aride, 13-43 warbler groups were studied from 1988 to 1996, and on Cousine, 15-17 groups from 1990 to 1996. Data presented here were based on unassisted breeding pairs and on individually color-ringed birds. In June or July each year, the number of warbler territories in high- and very high-quality habitat (see below) were counted accurately on the three islands.
Breeding activity
We recorded the breeding activity by monitoring all territories on Cousin,
Aride, and Cousine every 2 weeks during the research period for nest-building
activity by observing females for 30 min. This observation period was long
enough to determine whether birds are nest building
(Komdeur, 1991). During the
nest-building and incubation period, nests were checked every day for presence
of an egg or a nestling, sometimes with the help of a mirror attached to a
stick. Egg tossing by warblers does not occur
(Komdeur et al., 1999;
Kraaijeveld and Smit, 1997),
so missing eggs were considered to be predated. Observations on incubation,
nest guarding (when the bird was perched <2.5 m from the nest while the egg
was not incubated), and agonistic interactions between Seychelles fodies and
warblers were conducted simultaneously 3-5 days after initiation of egg
laying. Each observation period comprised three 1-h segments in periods
equally spaced over the day: 0630-1030, 1030-1430, and 1430-1830 h. For each
half minute we recorded whether each bird was taking part in each of the
above-mentioned categories. Agonistic interactions between fodies versus male
and/or female warblers were always measured when fodies were present within
2.5 m radius from the nest. We classified the interactions into three
categories: warbler chases fody, fody chases warbler, and no chases. To test
whether time spent incubating and nest guarding affect foraging time, foraging
studies were conducted at the same day when incubation and nest guarding were
monitored. For males and females foraging observations were taken during a
30-min period using the one-zero sampling method
(Martin and Bateson, 1986). At
half minute intervals it was noted whether foraging had taken place. The
warblers are remarkably tame and easily observed. Incubating, nest guarding,
and foraging were expressed as the percentage of total time the warbler had
been observed.
Territory quality
We measured territory quality in terms of the insect prey abundance because
this correlated positively with adult survival and reproductive success
(Komdeur, 1991). The
Seychelles warblers are insectivorous, and so territory quality was expressed
as mean number of prey invertebrates available within a territory (for method,
see Komdeur, 1991,
1994a). The quality of all
territories on Cousin was assessed monthly from May 1987 to June 1991, and in
June from 1993 to 1997. The quality of territories on Cousin and Aride was
assessed simultaneously from September 1988 to September 1990 and in June
1993-1997; on Cousine, it was assessed from June 1990 to June 1991, and in
June 1993-1997. Territories on Cousin and Cousine were divided into low
(0-1500 insects present), medium (1501-3000 insects present), and high quality
(3001-4500 insects present). All territories studied on Aride were of very
high-quality (>12,000 insects present).
Predation experiment
The purpose of the predation experiment was to test the extent to which the
presence of Seychelles fodies contributes to egg loss from Seychelles
warblers' nests. We used artificial nests and artificial eggs to test the
effect of fodies on egg predation rate. The artificial nests were sisal
baskets mimicking the shape, size, and color of warbler's nests. The
artificial eggs were made of candle wax and had the size (length: 21 mm,
width: 14 mm), color, and marking of warblers' eggs. On the three islands we
monitored egg losses from experimental nests simultaneously during two
periods, 10-15 January 1997 and 23-28 February 1997, which both coincided with
the breeding season of Seychelles warblers. Each nest contained one artificial
egg, which is the most common clutch size of warblers. The nests were placed
at 3.5 m height in Pisonia grandis in high-quality territories on
Cousin and Cousine and in very high-quality territories on Aride each 30 m
apart along a fixed transect. Pisonia grandis was chosen because it
is the most common nest tree; 3.5 m height was chosen because this is the
average nesting height of warblers, and 30 m intervals were chosen because
this is the mean distance between warbler nests in high-quality habitat
(Komdeur, 1991). For 3
consecutive days we revisited the nests daily between 0600-0700 and 1800-1900
h and checked for egg loss.
To test for possible shortcomings of the experimental design, we conducted
three tests in high-quality habitat on Cousin Island in January 1997. First,
we tested for the effect of nest monitoring on predation rate. If there are no
effects, this allows us to make direct comparisons of egg loss of nests
checked daily with egg loss of nests checked every 3 days. For this purpose
warbler nests containing one egg incubated by warbler pairs were selected. Of
these, 10 were monitored daily between 0600-0700 and 1800-1900 h over a 3-day
period, and 13 were checked between 0600-0700 h of the first day and between
1800-1900 h of the third day. Second, we tested whether artificial nests were
perceived in the same way by predators as natural nests. We placed.an
artificial egg in 10 artificial nests and in 10 empty warbler's nests. On the
third day egg loss was monitored for both nest types. The warbler's nests used
were nests from which young had successfully fledged in the week before the
start of the experiment. Third, we tested whether artificial eggs were
predated to the same degree as warbler's eggs. After natural egg loss warblers
continue incubation when an artificial or a conspecific egg is placed in the
nest within 2-3 h (Kraaijeveld and Smit,
1997). In nests of warbler pairs we placed an artificial egg (9
nests) or a conspecific egg (10 nests) immediately after egg loss. After 3
days nests were monitored for the loss of incubated artificial eggs and
natural eggs.
Trade-off experiment: Seychelles fodies and significance of
nest-guarding behavior
The translocation of Seychelles warblers and the fact that some breeding
pairs from Cousin remained together on the new islands enabled us to measure
the influence of Seychelles fodies on allocation of time to guarding and
foraging behaviors, by comparing pre- and post-transfer data for the same
pairs. The experimental units consisted of four Cousin pairs removed to
Cousine and five Cousin pairs removed to Aride. The control units consisted of
six and seven breeding pairs, respectively, remaining on Cousin. Time
allocation between guarding, incubation, and foraging of the male and the
female of each pair was exactly known for the 2 years before and 17 months
after translocation to Cousine, and in the 2 years before and 2 years after
the translocation to Aride. All the pairs were experienced breeders because
they had raised fledglings in 2 consecutive years before transfer. During the
whole experimental period, no helpers were present with the breeding pairs,
and the breeding pairs remained the same. Therefore the internal control used
in this study (comparing nest guarding and foraging pre- and post-transfer of
the same breeding pairs) is an explicit matched-pairs control for group size
and experience of the breeding pairs.
Seychelles fody census
We estimated the density of Seychelles fodies on Cousin and Cousine in June
each year from 1986 to 1991. A fixed transect (1.5 km) was walked slowly in
low-, medium- and high-quality habitat, and the number of birds present within
10 m on both sides of the transect was counted. During both periods of the
predation experiments, 10-15 January 1997 and 23-28 February 1997, the density
of Seychelles fodies along a fixed transect (1.5 km) on high-quality habitat
on Cousin and Cousine was assessed three times on different days.
Data analyses
Important factors affecting reproductive success are territory quality
(Komdeur, 1992), the bird's
age (Komdeur, 1996a), and the
presence of helpers (Komdeur,
1994b). On Cousine the investigated birds were occupying
high-quality territories and on Aride very high-quality territories. We
restricted our analysis to breeding males and females whose ages remained
within the 3- to 7-year age interval, which is the period during which
foraging and reproductive success remain constant
(Komdeur, 1996a). All
investigated birds were paired with the same partners during the whole period
and did not receive assistance from helpers. Tests on incubation, nest
guarding, foraging, egg loss, and hatching success were based on the first
breeding attempt observed for each breeding bird between 3 and 7 years of age
to avoid pseudoreplication. Percentages were arcsine transformed for
statistical analyses. The factors, which may affect egg loss and hatching
success, were analyzed by logistic regression. The factor enters into the
equation only if the probability (p) associated with the G
test on the decrease in scaled deviance (D) is <.05. The
coefficient of variation is calculated as the quotient of standard deviation
(SD) and the mean multiplied by 100%. A multivariate analysis of variance
designed for repeated measures (MANOVA) was applied to assess the effect of
presence of Seychelles fodies on data on foraging, nest guarding, and
incubating. Means are expressed with standard deviations; unless stated
otherwise probability values are two-tailed, and the null hypothesis was
rejected at p <.05. Of the statistical tests with small sample
sizes, we assessed the post-hoc statistical power (1-ß), which is the
probability of rejecting the null hypothesis (no differences between groups)
when it is in fact false and should be accepted (Type II error;
Zar, 1984). For the power
analyses we set the effect size, which is the minimum true difference to be
detected, at medium (
2 test 
= 0.30; paired-sample
t tests: d = 0.50; repeated-measures MANOVA:
f2 = 0.15; Cohen,
1988). If (1-ß) > 0.50, then the probability of rejecting
the null hypothesis, given that the alternative hypothesis is true, is high
(Lipsey, 1990;
Thomas and Krebs, 1997). Power
(1-ß) for (paired-sample) t tests, 2, and
repeated-measures MANOVA were determined by G*POWER analysis
(Buchner et al., 1997).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Effect of Seychelles fodies on egg loss
Including all incubation observations by all breeding groups on Cousin Island (n = 712), 12 egg predations were actually observed, all by Seychelles fodies. All these predations occurred when warblers were absent from the nest. In all cases fodies perched themselves on the rim of the nest and destroyed the warbler's egg by tossing out the egg and subsequently eating the contents on the forest floor. The average time between arrival and tossing an egg was 10.9 ± 1.3 s (range: 4-17 s, n = 9). Because only fodies predate warbler eggs, the prediction is that egg loss should be significantly lower in the absence of fodies. Given the short time necessary to predate an egg, nest defense by warblers against fodies should be intense.
We were unable to find potential shortcomings in the experimental design,
although test power was low (Table
1). The observer's visitation rate of warbler's nests was of no
influence on egg loss rate. Loss rate of warbler eggs tended by warbler pairs
was similar for nests visited in the morning and the evening daily during a
3-day period and for nests visited in the morning of day 1 and the evening of
day 3 only. Artificial nests were perceived similarly by fodies as natural
nests. Loss rate of artificial eggs placed in artificial nests was similar to
loss rate of artificial eggs placed in old warbler's nests
(Table 1). During this
experiment, three egg predations, all by fodies, were observed. In one case,
the artificial egg was tossed from the artificial nest, and in two cases the
artificial egg was tossed from the old warbler nest. The average time between
arrival and tossing the artificial egg was 11.0 ± 2.7 s (n =
3). Fodies were observed predating both artificial and natural eggs. Loss rate
of artificial eggs was similar to loss rate of warbler eggs, both being
incubated and guarded by warbler pairs. Thus we assume that the predation
experiments discussed below reflect the natural situation
(Major and Kendal, 1996).
|
The densities of Seychelles fodies in high-quality habitat on Cousin and Cousine were similar in the 2 months in which the predation experiments were conducted, although the test power was low (Table 2). In addition, on the islands of Cousin, Cousine, and Aride, loss rates of artificial eggs from artificial nests were independent of both periods, and test power to detect significant differences on Cousine and Aride was high (Table 2). Therefore, for each island data on egg losses were pooled over both periods. On Cousin and Cousine, no difference in loss rates of artificial eggs was detected (Table 2, Figure 1). The average egg loss was high, with 73.3% of eggs on Cousin and 73.5% of eggs on Cousine being lost 3 days after the start of the experiment. On Aride Island, with no fodies present, no egg losses were recorded during both 3-day experimental periods (Table 2). The difference between the change in egg loss measured on day 3 on Cousin and Aride was significant (Figure 1).
|
|
Effectiveness of predator deterrence on Cousin Island
The density of Seychelles fodies on Cousin each year from 1986 to 1991 was
25, 30, 21, 26, 28, and 27 birds/ha, respectively. Because fody density was
similar for each year, all breeding records from all male and female
Seychelles warblers were pooled over the entire study period. The breeding
performance by Seychelles warblers during the incubation period and the
agonistic interactions between Seychelles warblers and Seychelles fodies are
presented in Table 3. Warblers
have defined differences in the roles of the sexes in the nesting effort. The
female does all of the incubation, and the male does nearly all of the nest
guarding (Table 3). The amount
of nest guarding by the male is low when the female incubates the egg.
However, males spent significantly more time nest guarding when the female was
absent from the nest than when the female was incubating the egg (mean percent
time nest guarding: 72.5 ± 1.6 and 1.3 ± 1.2, respectively;
paired-sample t test: t = 32.14, df = 123, p
<.001). The variance in total amount of time spent incubating by the female
was less than the variance in total amount of time spent nest guarding by the
male (coefficient of variation: 11.1% and 40.5%, respectively). On average,
87.5% ± 13.1% of time was spent near or on the nest by either one of
the birds. During the incubation period males spent more time foraging than
females (Table 3).
|
The results obtained from the predation experiment suggest that when
Seychelles fodies are present and warblers are incubating a clutch, deterrence
of fodies close to the nest by warblers should be intense. The presence of
fodies close to the warbler's nest caused disturbance to the breeding pairs
(Table 3). If a fody was
present within a radius of 2.5 m from the warbler's nest, and if at least one
warbler was present, agonistic interactions between both species have been
observed in 62.0% of cases. In all these cases the warbler successfully chased
the fody. This success is not due to the warbler being heavier than the fody
(Seychelles warber: 16.1 g; Sechelles fody: 16.4 g;
Komdeur, 1991). When a fody
was present near the nest while the female was incubating and the male absent
from the nest, only in 26.7% of cases did the female leave the nest to chase
the fody. However, when the female was absent and the male was guarding the
nest, in nearly all cases the male chased the fody. When both male and female
warblers were present, the proportion of fodies chased by males was
significantly higher than that chased by females (2 = 234.89,
df = 1, p <.001; Table
3). A fody was never observed chasing an incubating female off the
nest when the male was absent, chasing a guarding male from the nest when the
female was absent, or landing on the nest while one or both warblers were
present close to the nest (Table
3). However, when the breeding pair was absent from the nest, none
of the fodies present near the nest was chased.
Effect of food availability on time allocation to nest guarding,
incubating, and foraging on Cousin Island
Data obtained from the artificial nest experiments revealed that nest
attendance by Seychelles warblers is important to prevent egg loss
(Table 1). Loss rate of eggs
not tended by warbler pairs was significantly higher than loss rate of eggs
incubated and guarded by breeding pairs (average loss rate after 3 days: 75%,
n = 20 and 10.5%, n = 19, respectively; 2 =
16.47, df = 1, p <.001). Because males do not participate in
incubating and females do not participate in nest guarding, nest attendance is
a combination of nest guarding by the male and incubating by the female. Nest
guarding by the male is negatively correlated with the amount of food
available in his breeding territory (Figure
2A). There is evidence for the existence of a trade-off between
nest guarding and foraging (Figure
2A). The better the quality of the breeding territory, the less
time was spent foraging and the more time nest guarding by the male. Food
availability had no effect on the amount of incubating and foraging by females
paired with these males (Figure
2B). In addition, there was a negative correlation between food
availability and egg loss (Figure
2C). This was not due to differences in fody densities in
different-quality territories (average fodies/ha, 1986-1991: low-quality
territories: 24.3 ± 2.7; medium-quality territories: 27.7 ± 2.7;
high-quality territories: 26.3 ± 5.9). There is evidence that egg loss
and hatching success on Cousin Island are independent of the amount of
incubation by the female but dependent on the amount of nest guarding by the
male (Figures 2 and
3). The more time spent nest
guarding by the male, the lower the risk of the egg being lost and the higher
the probability of hatching (Figure
3).
|
|
Effect of Seychelles fodies on egg loss and trade-off between nest
guarding and foraging
Seychelles fodies and time spent nest guarding and incubating The
density of Seychelles fodies on Cousine Island each year from 1986 to 1991 was
28, 32, 21, 29, 22, and 28 birds/ha, respectively. The average density of
Seychelles fodies on Cousine was the same as that on Cousin (mean density/ha:
26.7 ± 4.3 versus 26.2 ± 3.1, respectively; two-tailed
t test: t = 0.36, df = 5, p =.733). In addition,
fody densities (Table 2) and
egg loss from artificial nests (Figure
1) measured in 1997 were similar on Cousine and Cousin. Given that
fody densities on Cousin and Cousine are equal, fraction egg loss on Cousine
should be negatively related to the amount of nest guarding, like on Cousin
(Figure 3). Because we used
high-quality territories only for the analyses, the amount of nest guarding
should be the same as on Cousin. In contrast, on Aride Island with no fodies
present and no losses of artificial eggs from artificial nests, warbler eggs
should not be lost and nest guarding should not occur. In principle, because
food availability per territory on Aride is significantly higher than on
Cousin, warblers could allocate more time to nest guarding. In addition, egg
loss on Cousine and Aride should be independent of the amount of incubation by
the female.
The average clutch size of the investigated breeding pairs on high-quality territories on Cousin and Cousine was nearly twice as small as that of pairs breeding in very high-quality territories on Aride (Table 4). On Cousine Island females spent the same percentage of time incubating, and males spent the same percentage of time nest guarding as on Cousin, and the fraction of eggs lost and eggs hatched was the same as on Cousin (Table 4). The fraction of eggs lost and eggs hatched on Cousine was independent of the amount of time spent incubating by the female, but inversely related to the amount of nest guarding by the male (Figure 3). An increase in time spent nest guarding results in a decrease in egg loss and consequently in an increase in hatching success. On Aride Island, females spent the same percentage of time incubating as on Cousin, but males spent significantly less time nest guarding as on Cousin (Table 4). The warblers on Aride had no egg loss and significantly higher hatching success as compared to warblers on Cousin (Table 4). Hatching success on Aride was independent of time spent incubating by the female and nest guarding by the male (Figure 3). The sum of the percentage of eggs lost and the percentage of eggs hatched on both Cousin and Aride did not equal 100% because of the presence of a few infertile eggs. When considering the eggs being laid by all warbler pairs on Aride during the 2 years after translocation, none was lost during the incubation period, and 89.3 % of eggs hatched (n = 112). Young produced by the investigated breeding pairs on Cousin and Cousine with egg predators but without nestling predators fledged 5 days later than young on Aride (Table 4). During the nestling phase the weights of adult male and female parents were similar on Cousine and Cousin, but on Aride adult male and female parents were significantly heavier than male and female parents on Cousin (Table 4). However, sex-specific differences in weight on Aride versus Cousin were more pronounced for males (7.5% weight difference) than for females (2.0% weight difference; see also Table 4).
|
During the 12-year study period on Cousin, territory density on high-quality habitat and the average amount of time spent incubating and nest guarding by warblers remained the same each year (Figure 4). Time spent nest guarding and incubating was independent of territory density (Figure 4). On the islands of Cousine and Aride, the number of territories on high- and very high-quality habitat, respectively, increased considerably, but the average amount of time spent incubating and nest guarding by warblers occupying territories in these habitats remained the same over the 8- and 10-year period, respectively, after the translocation (Figure 4).
|
Seychelles fodies and trade-off between nest guarding and
foraging
In the presence of Seychelles fodies, egg loss is affected by food
availability and by the amount of time spent nest guarding by male warblers.
Therefore we expect that time allocation between nest guarding and foraging by
males on Cousine (with fodies present) will be the same as on Cousin. However,
on Aride, with higher food abundance but with no fodies present and less nest
guarding, males may spend more time foraging. Mean percentage time spent
foraging, nest guarding, and incubating for unassisted male and female
warblers on high-quality territories on Cousin Island before translocation,
and of the same pairs in high-quality territories on Cousine Island and in
very high-quality territories on Aride Island after translocation are
presented in Table 5. For all
comparisons the test power to detect significant differences was high
(Table 5). Mean percentage time
spent foraging, nest guarding, and incubating by males and females was the
same before both translocations. Over the period after the transfer to
Cousine, time allocation to foraging, nest guarding, and incubating of both
sexes on Cousine was the same as that of the same birds on Cousin before the
transfer (Table 5). After the
transfer to Aride, time allocation to foraging and incubating by females on
Aride was the same as that of the same females on Cousin before the transfer
(Table 5). However, after
translocation to Aride, the same males spent less time nest guarding and more
time foraging on Aride as on Cousin before the transfer
(Table 5). The differences
between the changes in time spent foraging and nest guarding of males on Aride
and of males remaining on Cousin were significant
(Table 5).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Nest guarding behavior as an antipredator adaptation
The presence of Seychelles fodies has a significant effect on egg loss in the Seychelles warbler. On Cousin and Cousine, both with similar fody predation pressure, nearly three out of the four unguarded artificial nests were predated after 3 days. On Aride Island, without fody predation pressure, however, none of the unguarded artificial nests was predated. Because all egg predations took a short time when warblers were absent from the nest, the warblers' nest attendance is important to egg survival. On Cousin Island, unguarded warblers' nests experienced a 7.1 times higher egg loss rate than guarded nests. In the Seychelles warbler only females incubate the clutch. Fodies were never observed chasing an incubating female from the nest, which suggests that active nest defense by females during incubation should not occur. However, during incubation when males were not guarding and when a fody came close to the nest, in 26.7% of cases females ceased incubation to chase the fody. When the females were absent from the clutch, males guarded the nest intensively, whereas females spent little time guarding. Nearly all fodies present close to the nest were chased by the nest-guarding male warbler. The amount of time spent incubating by different females varied relatively little (31-56%) and had no effect on egg loss. Male warblers spent significantly more time near the nest with the clutch when the female was absent from the nest than when she was incubating the clutch. An increase in time spent nest guarding by the male results in a decrease in egg loss and hence in an increase in hatching success. However, the amount of time spent nest guarding varied considerably between males (5-78%) and was inversely related with food abundance in the territory. These results were controlled for group size, age, breeder experience, and the reproductive value of the egg (egg age). The availability of effective egg defense behavior by the male enables the warbler to make optimal use of a limited breeding opportunity.
Food and predation trade-offs: short-term and evolutionary
responses
It is argued that, if nest defense is energetically costly or entails costs
on future reproduction, nest-guarding parents should reduce nest-guarding time
and increase foraging time as food gets scarce or as risk of predation
decreases from predators that can be deterred by parents
(Martin, 1993). On Cousin
Island with Seychelles fodies being equally abundant in Seychelles warbler
territories of different food quality, a food-related trade-off was observed
between time allocated to foraging and to time allocated to nest guarding by
male warblers. However, no food-related trade-off was observed between
foraging and incubating by female warblers.
The transfer of Seychelles warbler breeding pairs from Cousin to Cousine (no change in predation pressure or food availability) and from Cousin to Aride (reduction in predation pressure and increase in food availability) confirmed the plasticity in nest-guarding and foraging behavior. On Cousine, sex-specific time allocation of pairs during reproduction was the same as that of the same pairs on Cousin before the translocation. On Aride, males that spent a significant amount of time nest guarding on Cousin did not guard the clutch and spent significantly more time foraging. Females on Aride, on the other hand, spent the same amount of time incubating and foraging as they did on Cousin before translocation.
Parents should optimize the parental investment mechanisms by which they
can maximize their lifetime reproductive success. Defense of eggs and
offspring against predators is an important form of parental investment in
many species (Martin, 1993;
Trivers, 1972). In the
Seychelles warbler where hatching success is determined by the nest-guarding
properties of males, nest-guarding and foraging by males are adjusted to the
availability of food and to the presence of predators. Time allocation to
incubation and foraging by females was independent of food availability and
predator presence. These results were controlled for group size, age, breeder
experience, genetic differences between birds, and egg age. The amount of time
nest guarding and incubating on the three islands was independent of territory
density. The lack of nest-guarding behavior by males on Aride cannot be
explained by increased extrapair sexual behavior for three reasons: (1)
immediately after the transfer, extrapair sexual behaviors did not occur
(Kraaijeveld and Smit, 1997);
(2) radio-telemetry studies showed that territorial males never leave their
territories during incubation (van
Katwijk, 1995); and (3) parentage analyses showed that nestlings
produced by pairs were always sired by the territorial male
(Kappe, 1998).
Adult warblers on Aride were significantly heavier than adult warblers on
Cousin, which was probably due to the higher food abundance on Aride. On
Aride, by investing less in guarding the clutch and more in foraging, male
parents may gain better condition by which they may maximize their
reproductive success. Sex-specific differences in weight on Aride versus
Cousin were more pronounced for males than for females. There is evidence that
body condition of warblers is correlated with parental investment level. Food
provision rates to young by male parents during the entire period of
dependence (about 3 months) were significantly higher on Aride than on Cousin
and Cousine (Komdeur, 1991).
By investing more in provisioning fledglings, males on Aride may reduce the
female's workload to provisioning. Food provision rates by females were
similar on Cousin and Cousine, but significantly lower on Aride
(Komdeur, 1991). This may well
be the reason that females on Aride were able to invest in another clutch
immediately after their first clutch had fledged
(Komdeur, 1996b), thereby
improving the reproductive output for themselves and their partners. However,
whether male condition determines the level of paternal investment, and
whether female warblers are able to adjust their clutch size to the
forthcoming paternal investment, needs to be investigated. In addition, the
long-term effects of egg predators on the parents' survival and lifetime
reproductive success are yet unknown.
Clutch size: a predation perspective?
Life-history theory and empirical studies (see
Lima, 1987;
Martin, 1988) suggest that the
rate of nest predation affects clutch size. Higher nest predation rates should
favor smaller clutch sizes for two reasons: to obtain increased growth rates
of nestlings to minimize fledging time and hence exposure to predation
(Case, 1978;
Lack, 1948,
1968), and to minimize costs
and save more energy to be invested in several nesting attempts rather than a
single attempt (Martin and Li,
1992; Slagsvold,
1982). Although the occurrence of smaller clutch sizes of
Seychelles warblers on islands with egg predators is in agreement with the
theoretical expectation, this cannot be explained by the predation-hypothesis.
First, on the islands with egg-predating fodies it is not important to
decrease fledging time of nestlings by laying smaller clutches because
nestlings have no predators (Komdeur,
1991). On the contrary, young produced by the investigated
breeding pairs on Cousin and Cousine with egg predators but without nestling
predators fledged significantly later than young on Aride. Second, on the
islands of Cousin and Cousine the warblers usually raise one clutch per year,
whereas on Aride with high food availability year-round, warblers prolonged
their reproductive season (most bred year-round), and increased the annual
number of broods (on average warblers raise six clutches per year)
(Komdeur, 1996b). The
differences in clutch sizes of warblers on the three islands cannot be
explained by compensation for increased predation. The larger clutch size
observed on Aride is possibly a cause of the higher food abundance in warbler
territories (Komdeur,
1996b).
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We thank Joost Tinbergen, Serge Daan, Simon Verhulst (Groningen University, The Netherlands) and three anonymous referees for their constructive criticism of an early draft of the manuscript. We are also grateful to Michael Rands (BirdLife International, UK), who arranged permission to work on Cousin Island and to transfer warblers, to James Cadbury and the late Christopher Cadbury, and to Rober Vogel, who generously provided support and agreed to have warblers on the islands of Aride and Cousine, respectively. We thank Robby Bresson of Cousin, Michael and Susan Betts of Aride, and Peter Hitchins and Stella le Maitre of Cousine for helping with logistics of the project, and we are grateful to Marc de Goede and Bart Hekman for help with predation experiments. The study was supported by grants from BirdLife International, the Danish Natural Science Research Council, and the Netherlands Foundation for the Advancement of Tropical Research (WOTRO).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Arcese M, Smith JNM, 1988. Effects of population density and supplemental food on reproduction in song sparrows. J Anim Ecology 57:119-136.
Andersson M, Wiklund G, Rundgren H, 1980. Parental defence of offspring: a model and an example. Anim Behav 28:536-542.
Barash DP, 1975. Evolutionary aspects of parental behavior: distraction behavior of alpine accentor. Wilson Bull 87:367-373.
Buchner A, Erdfelder E, Faul F, 1997. How to use G*Power [WWW document]. URL http://www.psychologie.uni-trier.de:8000/projects/gpower/how_to_use_gpower.html.
Case TJ, 1978. On the evolution and adaptive significance of postnatal growth rates in the terrestiral vertebrates.Q Rev Biol 53:243-282.[Medline]
Cohen J, 1988. Statistical power analysis for the behavioral sciences, 2nd ed. Hillsdale, New Jersey: Lawrence Erlbaum Associates.
Collar NJ, Stuart SN, 1985. Threatened birds of Africa and related islands. ICBP/IUCN Red Data Book, Part 1. Cambridge: International Council for Bird Preservation.
East M, 1981. Alarm calling and parental investment in the robin Erithacus rubecula. Ibis 123:223-230.
Greig-Smith PW, 1980. Parental investment in nest defence by stonechats (Saxicola torquata). Anim Behav 28:604-619.
Hayes PA, Robertson RJ, 1989. The impact of male parental care on female eastern kingbird reproductive success. Wilson Bull 101:462-467.
Kappe A, 1998. Detecting genetic variation; application of molecular techniques in conservation biology (PhD dissertation). Wageningen: Ponsen and Looijen.
Komdeur J, 1991. Cooperative breeding in the Seychelles warbler (PhD dissertation). Cambridge: Cambridge University.
Komdeur J, 1992. Importance of habitat saturation and territory quality for evolution of cooperative breeding in the Seychelles warbler. Nature 358:493-495.
Komdeur J, 1994a. Conserving the Seychelles warbler Acrocephalus sechellensis by translocation from Cousin Island to the islands of Aride and Cousine. Biol Conserv 67:143-152.
Komdeur J, 1994b. Experimental evidence for helping and hindering by previous offspring in the cooperative breeding Seychelles warbler (Acrocephalus sechellensis). Behav Ecol Sociobiol 34:31-42.
Komdeur J, 1996a. Influence of age on reproductive
performance in the Seychelles warbler. Behav Ecol
7:417-425.
Komdeur J, 1996b. Seasonal timing of reproduction in a
tropical bird, the Seychelles warbler: a field experiment using translocation.J Biol Rhythms
11:333-346.
Komdeur J, Kraaijeveld-Smit F, Kraaijeveld K, Edelaar P,1999 . Explicit experimental evidence for the role of mate guarding to minimise loss of paternity in the Seychelles warbler. Proc R Soc Lond B (in press).
Kraaijeveld K, Smit F, 1997. Timing of mate-guarding and reproduction in the Seychelles warbler on Cousine Island in 1996 (MSc dissertation). Groningen: Groningen University.
Lack D, 1948. The significance of clutch size. Part 3.Ibis 90:25-45.
Lack D, 1968. Ecological adaptations for breeding in birds. London: Methuen Press.
Li P, Martin TE, 1991. Nest-site selection and nesting success of cavity-nesting birds in high elevation snow-melt forest.Auk 108:405-410.
Lima SL, 1987. Clutch size in birds: a predation perspective. Ecology 68:1062-1070.
Lipsey MW, 1990. Design sensitivity: statistical power for experimental research. Newbury Park, California: Sage.
Major RE, Kendal CE, 1996. The contribution of artificial nest experiments to understanding avian reproductive success: a review of methods and conclusions. Ibis 138:298-307.
Martin P, Bateson P, 1986. Measuring behaviour: an introductory guide. Cambridge: Cambridge University Press.
Martin TE, 1987. Food as a limit on breeding bird: a life-history perspective. Annu Rev Ecol Syst 18:453-487.[ISI]
Martin TE, 1988. Nest placement: implications for selected life history traits, with special reference to clutch size. Am Nat 132:900-910.
Martin TE, 1992. Interaction of nest predation and food limitation in reproductive strategies. Curr Ornithol 9:163-197.
Martin TE, 1993. Nest predation, nest sites and birds: new perspectives on old patterns. Bioscience 43:523-532.
Martin TE, Li P, 1992. Life-history traits of open- versus cavity-nesting birds. Ecology 73:579-592.
McLean IG, Smith JNM, Stewart KG, 1986. Mobbing behaviour, nest exposure and breeding success in the American robin.Behavour 96:171-186.
Montgomery R, Weatherhead PJ, 1988. Risks and rewards of nest defence by parent birds. Q Rev Biol 63:167-187.
Nilsson SG, 1984. The evolution of nest-site selection among hole-nesting birds: the importance of nest predation and competition.Ornis Scand 15:167-175.
Nilsson SG, 1986. Evolution of hole-nesting in birds: on balancing selection pressures. Auk 103:432-435.
Owen HV, 1986. Management plan of Cousin Island, Seychelles. Cambridge: International Council for Bird Preservation.
Rose MR, 1983. Theories of life history evolution.Am Zool 23:15-23.
Ricklefs RE, 1969. An analysis of nesting mortality in birds. Smithsonian Contrib Zool 9:1-48.
Slagsvold T, 1982. Clutch size variation in passerine birds: the nest predation hypothesis. Oecologia 54:159-169.
Stearns SC, 1989. Trade-offs in life history evolution. Funct Ecol 3:259-268.
Stearns SC, 1992. The evolution of life histories. Oxford: Oxford University Press.
Thomas L, Krebs CJ, 1997. Technological tools.Bull Ecol Soc Am 78:126-139.
Trivers R, 1972. Parental investment and sexual selection. In: Sexual selection and the descent of man 1871-1971 (Campbell B, ed). Chicago: Aldine Press;139 -179.
van Katwijk Q, 1995. The population demography of the Seychelles warbler on Cousine Island, a saturating environment (MSc dissertation). Groningen: Groningen University.
Wiklund CG, 1990. Offspring protection by merlin Falco columbarius females: the importance of brood size and expected offspring survival for defence of young. Behav Ecol Sociobiol 26:217-223.
Yom-Tov Y, 1974. The effects of food and predation on breeding density and success, clutch size and laying date of the crow (Corvus corone L.). J Anim Ecol 43:479-498.
Zar JH, 1984. Biostatistical analysis. Englewood Cliffs, New Jersey: Prentice-Hall.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||