Behavioral Ecology Vol. 10 No. 2: 191-197
© 1999 International Society for Behavioral Ecology
Brood desertion in Kentish plover
the value of parental care
a Centre for Behavioural Biology, School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, UK b Behavioural Ecology Research Group, Department of Zoology, Kossuth University, Debrecen, H-4010, Hungary
Address correspondence to T. Székely, Centre for Behavioural Biology, School of Biological Sciences, University of Bristol, Woodland Road, Bristol, BS8 1UG, UK. E-mail: t.szekely{at} bristol.ac.uk
Received 18 January 1998; accepted 29 September 1998.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
To understand the evolution of parental care, one needs to estimate the payoffs from providing care for the offspring and from terminating care and deserting them. In this study we estimated the payoff from care provision, and in a companion paper we analyze the payoff from offspring desertion. In the current study we experimentally investigated the influence of the number and sex of attending parents on growth and survival of offspring in the Kentish plover Charadrius alexandrinus, in two sites (A and B). Either the male or the female parent was removed from some broods at hatching of the chicks (female-only and male-only broods, respectively), whereas in control broods both parents were allowed to attend their young. At site A survival of the chicks was lower in uniparental (male-only and female-only) broods than in control broods, whereas we found no difference in brood survival at site B. Brood survival decreased over the season. Removal of either parent did not influence the growth of the young, although growth varied over the breeding season, and it was significantly different between the sites. These results suggest that the payoff from parental care decreases over the breeding season and that the value of parental care (i.e., the contribution of parents to the survival of their young) may depend on the environment.
Key words: biparental care, Charadrius alexandrinus, Kentish plover, mate removal, offspring desertion, parental care, uniparental care.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Parental care is a highly variable trait in birds. In some species both parents care for the offspring, whereas in others either the male or the female provides the care on its own (reviewed by Clutton-Brock, 1991
). To
understand why some birds care for their young, whereas others do not, one
needs to know the payoffs from caring and from deserting (reviewed by
Clutton-Brock, 1991;
Székely et
al., 1996). These payoffs, however, are rarely known
(Clutton-Brock, 1991).
The payoff from caring may be different between male-only and female-only
broods and may vary between biparental and uniparental broods. In avian
studies, researchers have typically estimated the payoff from biparental care
and compared it against female-only care (reviewed by
Bart and Tornes, 1989;
Clutton-Brock, 1991;
Liker, 1995;
Wolf et al., 1988). One reason
for this bias may be that single males are often not willing (or are unable)
to incubate the clutch (e.g., in many waterfowl and passerines), so the payoff
from male-only incubation would be zero. The value of uniparental care by
either sex has rarely been studied experimentally (but see
Duckworth, 1992;
Erckmann, 1983;
Martin and Cooke, 1987;
Martin et al., 1985).
Experimental estimation of these payoffs is necessary because individual
qualities such as the ability to provide care or to attract a new mate may
influence the observed pattern of parental care. An experimental approach to
estimate these payoffs would be particularly valuable in a species in which
uniparental and biparental care both occur and neither sex is specialized to
provide care on its own (Clutton-Brock,
1991).
The Kentish plover Charadrius alexandrinus is one of the handful
of bird species in which either parent is capable of providing care on its
own, and biparental care and uniparental care all occur within the same
population (Fraga and Amat,
1996; Freudenthal and
Lessells, in press; Lessells,
1984;
Székely
and Lessells, 1993). To estimate the value of parental care
provided solely by the male, the female, or both parents, we removed either
the male or the female parent at hatching and left both parents with their
chicks in control broods. The value of parental care may vary over a breeding
season, so we were also interested in the variation of these values over this
period. In a complementary study, we analyze the remating opportunities of
deserting plovers (Székely et al., this
issue). Taken together our two studies are unique because they experimentally
estimate the payoffs from both caring and deserting for both sexes in the
natural environment. The only comparable studies to ours have been conducted
in fish and insects. In particular, in a series of laboratory and open-field
enclosure experiments in St. Peter's fish Sarotherodon galilaeus,
Balshine-Earn estimated both the benefit from caring
(Balshine-Earn, 1997) and the
benefit from deserting and spawning with a new mate
(Balshine-Earn, 1995). Also,
Thomas (1994) investigated the
costs and benefits of the care for either the male or the female parent in two
species of assassin bugs (Reduviidae), in which either only the male or the
female guards the eggs.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Experimental manipulation
The experiment was carried out in Tuzla Lake of the Çukurova Delta, Southern Turkey (36°42' N, 35°03' E), where Kentish plovers breed in the saltmarsh around the lake (Székely et al., this issue). The breeding population was about 1000 pairs. We used two sites on the north side of the lake (referred to as site A and site B). The distance between these sites was about 2.5 km (closest points) and 7 km (most distant points).
Experimental manipulations were carried out between 4 May and 28 June 1996.
All broods were biparental at the time of manipulation. A randomized block
design was used to control for seasonal variation in survival. On any one date
(block), groups of three nests were randomly allocated to the treatments:
male-only (n = 14), female-only (n = 14), and control broods
(n = 14). In all broods both parents and their chicks were caught,
ringed, and measured. Each parent was ringed with a unique combination of
color rings. Their subcutaneous fat was scored using the method described by
Helms and Drury (1960). After
the measurements were taken we released the chicks and one parent (or both
parents in control broods). The released parent always returned to its brood.
The removed parent was taken to an aviary and released after its brood fledged
or died.
The experiment was designed to detect possible effects on chick survival in
a species with a naturally high level of chick mortality. The choice of sample
size was a compromise between minimizing additional chick mortality and
maximizing detection of the experimental treatment. The experiment was
licensed by Turkish Ministry for Natural Parks in a location where the Kentish
plover is locally abundant (Magnin and
Yarar, 1997).
We checked nests at least once a day near the expected date of hatching. Thirty-five broods were manipulated 0-2 days after the chicks hatched. The chicks in these broods were either in the nest-scrape or they were ringed on the day of hatching and recaptured for the manipulation 1 or 2 days later. The hatching date of seven broods was not known. We estimated the age of these chicks at manipulation from the growth of 12 control broods of our experiment, which were captured 34 times in total between hatching and the age of 25 days. We used tarsus length of these broods to estimate the age of unringed chicks because the tarsus grew linearly over the range of ages measured [age (in days) = 2.520 x tarsus (in mm) -48.341, r2 = 0.914, p <.0001].
Forty-one broods hatched three chicks each, and one chick was added on the day of hatching to a nest which hatched two chicks from three eggs. The age of the added chick was estimated to be 3.6 days at the time of manipulation. The added chick was readily accepted by its new parent.
Broods were checked at least every other day, and simultaneously the number
of chicks (and that of the attending parents) was recorded until the age of 25
days. Kentish plover chicks fledge at about 28 days
(Székely,
1996). Checking interval did not differ between male-only [2.6
± 0.2 days (SE)], female-only (2.6 ± 0.5 days), and control
broods (2.4 ± 0.2 days, ANOVA, F2,41 = 0.027,
p =.974). At each encounter the distance of the brood from the nest
or the first capture of the chicks was estimated. We recorded behavior of
parents and of chicks for 1 h every other day. During a behavioral sample, the
behavior of parent(s) and chicks was scanned every 30 s. In addition, the
distances between parent(s) and each chick were estimated every 5 min. We
attempted to capture chicks every fourth day for morphometric measurements. To
facilitate the relocation of broods in the large nonexperimental population,
we dyed the flanks of experimental parents with picric acid. This was
necessary because the parents often moved large distances with their brood
(see below).
Statistical procedures
We assessed body condition of parents at manipulation by regressing body
mass [in ln (g)] on tarsus length [in ln (mm); Pearson correlation
coefficient, males: r =.137, n = 42, p =.386;
females: r =.334, n = 42, p =.031] and taking the
residuals from these equations. These residuals correlated with the amount of
subcutaneous fat in both sexes (Spearman rank correlation, males:
rs =.309, n = 41, p =.049; females:
rs =.373, n = 42, p =.015).
Survival of each brood was estimated by a maximum likelihood method (see
details in
Noszály et
al., 1995). This estimate varied between 1.0 (all chicks survived
until the age of 25 days) and 0.0 (all chicks perished within the first 1-3
days after manipulation). Five chicks in four broods were lost from their
original brood and were adopted by one experimental brood and three foreign
broods. The survival of these chicks was not investigated and they were
excluded from data processing. Brood survival was not normally distributed;
thus we analyzed these data by nonparametric tests or applied parametric tests
on ranked transformed (RT) data (Zar,
1996). The use of the interaction term in RT analysis of variance
is legitimate only in 2 x 2 designs
(Seaman et al., 1994). To
conform to this assumption, we considered treatment to be a two-level factor
(uniparental versus control) in some of our RT analyses. Site also has two
levels, A and B.
The growth of young was investigated by analyzing the slope of linear growth curves. A linear curve was estimated for each brood in which at least one chick was recaptured at least once until the age of 25 days. Growth of tarsus is linear over the range of ages investigated. Change in mass is probably S-shaped, but as we have insufficient data points to estimate the curve accurately for each brood separately, we used the slope of linear fit as an approximation of rate of mass change.
Each brood was considered as the unit of analysis. If several observations
were available for a brood, we took the average of these observations. Size
and condition of parents were investigated by full mixed-factorial ANOVAs
(SPSS, 1988), in which the sex
of the parent (male, female) was the within-subject factor and site (A, B) was
the between-subject factor. In 9 out of 14 control broods both parents stayed
with the chicks until the chicks reached 25 days of age or died, whereas in
five broods the female parent deserted 10.2 ± 2.8 (SE) days after the
chicks hatched. In control broods the behavior of parents and chicks was
investigated while both parents stayed with the brood. Brood attendance was
calculated as the proportion of time that at least one parent was <10 m
from at least one chick. Feeding time was calculated as proportion of time
that the chicks pecked for food. Behavioral variables were arcsine square-root
transformed before analysis. The age of the brood influenced the behavior of
the parents (e.g., proportion of time spent on brooding: ANCOVA, factor: sex
of parent; covariate: brood age, F1,47 = 5.86, p
=.019) and that of their chicks (feeding: b = 0.019, n = 38,
p =.0002; being brooded: b = -0.023, n = 38, p
=.018). Thus we regressed all behavioral variables on brood age and used the
residuals of these linear regressions in MANOVAs, ANOVAs, and t
tests. In all two-way analyses of variance (MANOVA, ANOVA), we investigated
the interaction term (between type of care and site) and report if the
interaction term is significant (p <.05). Hatching dates are given
as number of days since 1 March. Distance of broods from their nest (or first
capture site) was log (x + 1) transformed. Correlations were
investigated either by Pearson correlation coefficient (r) or by
Spearman rank correlation coefficient (rs). Values are
given as means ± SE unless otherwise indicated. All tests were
two-tailed. We used SPSS for the Macintosh 4.0 and MINITAB
(1995) 10.1 for Windows for
data processing.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
At manipulation
We found no statistical difference among male-only, female-only, and control broods at the time of manipulation. Neither tarsus length of males and females, nor their body condition (measured either by residual body mass or subcutaneous fat), were different among male-only, female-only, and control broods (one-way ANOVAs, males: p =.533-.970, females: p =.317-.882). Also, tarsus length and body mass of chicks at manipulation and their hatching dates were not different among the three treatments (one-way ANOVAs, p =.426-.986), and we found no difference in the age of chicks at manipulation (Kruskal-Wallis test, p =.695). Furthermore, hatching date of male-only, female-only, and control broods were not different between site A and B (two-way ANOVA, site: F1,36 = 1.339, p =.255, type of care: F2,36 = 0.000, p = 1.000). Finally, neither tarsus length of parents (mixed-factorial ANOVA, site: F1,40 = 0.21, p =.648; sex of parent: F1,40 = 6.54, p =.014), nor their body condition were different between site A and B (site: F1,40 = 0.27, p =.609; sex: F1,40 = 0.08, p =.782).
Brood survival
Brood survival was significantly influenced by the interaction of parental
care and study site (two-way RT ANOVA, F1,38 = 5.981,
p =.019), so we analyzed the influence of parent removal separately
for the two sites (Figure 1).
Brood survival was different among male-only, female-only, and control at site
A (RT ANOVA, F2,24 = 3.421, p =.049); we found no
such difference at site B (F2,12 = 0.104, p
=.902). We further analyzed the differences among treatment groups within site
A by using orthogonal contrasts. Survival of uniparental broods (male-only and
female-only) was significantly lower than that of the control ones (RT ANOVA,
t = 2.611, p =.015), whereas there was no difference in
survival between male-only and female-only broods (t = 0.083,
p =.934). Male-only broods survived less well at site A than at site
B (RT t tests, t = 4.05, p =.002), whereas the
survival of control and female-only broods was not different between the sites
(control: t = 0.46, p =.657, female-only: t = 1.85,
p =.089).
|
We know the fate of seven chicks. Five chicks were found dead, and these chicks were probably killed by adult Kentish plovers. Two out of five chicks had been severely pecked by unringed adults when these chicks were seen alive for the last time. Some of the attacking adults attended their own brood in adjacent territories. One chick was taken by a red-backed shrike Lanius collurio, and another one starved to death after a clam attached itself to its toe and thus restricted the chick's movement.
Brood survival tended to decline within season at both sites (site A: rs = -.304, n = 27, p =.123; site B: rs = -.752, n = 15, p =.001). Overall, brood survival declined during the season (rs = -.528, n = 42, p <.001; Figure 2).
|
These conclusions remained unaltered when we excluded the broods in which all chicks died within 1-3 days of manipulation (2 out of 14 broods in each of the three treatments; Figure 2). Brood survival remained significantly different among male-only, female-only, and control broods at site A (F2,19 = 5.219, p =.016), whereas brood survival still did not differ at site B (F2,11 = 0.003, p =.997). Brood survival declined over the season at both sites (site A: rs = -.625, n = 22, p =.002; site B: rs = -.692, n = 14, p =.006).
Growth of young
We fitted quadratic equations to chick growth rates of broods hatching at
different times in the breeding season
(Figure 3). The quadratic
equations provided better fits for changes in both body mass (site A:
r2 =.637, site B: r2 =.537) and tarsus
length (site A: r2 =.549, site B: r2
=.418) than the linear equations (body mass: site A: r2
=.587, site B: r2 =.210, tarsus length: site A:
r2 =.525, site B: r2 =.004). Note that
chicks in late broods, particularly at site A, grew slowly and actually lost
mass (Figure 3).
|
To investigate the influence of parent removal on the growth of chicks, first we set up full ANCOVA models. In the full models the change in either body mass or tarsus length was the dependent variable, the type of care (male only, female only, and control) and site (A, B) were the factors, and a linear and a quadratic term of hatching date [(hatching date — mean hatching date)2] were the covariates. In the full models the effects of linear and quadratic covariates were combined, and all second- and third-order interactions between factors and covariates were investigated. Second, the nonsignificant interactions (p >.05) were eliminated stepwise from the full models, and we report these restricted ANOVAs below.
Removal of either parent did not influence the growth rate of chicks, measured by change in either body mass or tarsus length (Table 1). However, the effects of both site and hatching date were significant on the changes in body mass and tarsus length, and there was a significant interaction between site and hatching date (Table 1, Figure 3). In particular, growth of chicks decreased over the season at site A, whereas at site B the growth was initially slow and peaked in chicks that hatched between 15 May and 10 June (Figure 3).
|
Behavior of parents and chicks
Single males spent more time attending their brood than males sharing care
with their females (Table 2,
Figure 4). The response of
females to removal of their mates was different between sites. At site A
single females spent more time with their brood than control females
(Figure 4), whereas at site B
females did not change significantly the time they spent attending their brood
(Figure 4). Thus the time when
at least one parent cared for the brood was not different among male-only,
female-only, and control broods at site A (ANOVA on residual attending time,
F2,21 = 1.179, p =.327), but it was different
among broods at site B (F2,11 = 5.013, p =.028).
We further analyzed which groups differed from each other at site B and found
that the only significant difference was between female-only and control
broods (Tukey HSD test, p <.05).
|
|
Behavior of parents was not different among male-only, female-only, and
control broods (MANOVAs on residuals of three parental behaviors, male: Wilk's
= 0.850, p =.344; female: Wilk's = 0.933,
p =.719, univariate analyses are in
Table 2). In particular, there
were no differences in the time spent on brooding, being alert, or defending
the young from conspecifics among male-only, female-only, and control broods
(Table 2). Males tended to
fight less in male-only broods than in control broods at site B
(t test, t = 3.27, p =.011) but not at site A
(t = 0.45, p =.663), although we should view these
univariate results with caution because the overall MANOVA was not
significant. Chicks received as much brooding in male-only broods (21.1
± 6.3% of chicks' time) as in female-only (26.8 ± 6.7%) and
control broods (26.3 ± 5.1%; ANOVA on residual time being brooded,
F2,32 = 0.417, p =.663). Also, there was no
difference in feeding time of chicks between male-only (11.9 ± 2.4% of
chicks' time), female-only (8.2 ± 2.1%), and control broods (12.5
± 1.9%; ANOVA on residual feeding time, F2,32 =
1.390, p =.264).
Finally, type of parental care did not influence how far the brood moved from their nest or first capture site (two-way ANOVA, type of care: F2,30 = 0.27, p =.765), although broods moved a longer distance at site A (229 ± 31 m, n = 22 broods) than at site B (78 ± 25 m, n = 14 broods, two-way ANOVA, site: F1,30 = 33.66, p <.0001).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The benefits from brood care
Two parents may increase the survival chances of their off-spring more than a single parent in a number of ways. First, two parents may feed their chicks more often than a single parent, thus chicks in biparental broods may grow faster and survive better than in uniparental broods (reviewed by Bart and Tornes, 1989
;
Clutton-Brock, 1991). The help
by the male parent may be particularly important when food is scarce
(Bart and Tornes, 1989;
Duckworth, 1992). Although a
single parent often attempts to compensate for the lost help of its mate, this
compensation may not be complete (Markam
et al., 1996; Wright and
Cuthill, 1989). Second, parents sharing brood care with their mate
may be able to spend more time on brooding their chicks than single parents.
Brooding of the young is particularly important before the young reach thermal
independence, or where the environment is hostile such as very cold or hot
(Liker, 1995). Third, two
parents may defend their brood from predators and conspecifics better than a
single parent (e.g., by driving the enemies away or distracting their
attention) (Schneider and Lamprecht,
1990).
The results at site A support a previous nonexperimental study that showed
that after desertion by the female parent, brood survival decreased in the
Kentish plover
(Székely
and Williams, 1995). We argue that these results are mostly
consistent with the brood defense hypothesis. First, Kentish plovers are
precocial, thus their chicks are not fed. It is true that two parents may be
able to take their chicks to a site where food is abundant, although this is
unlikely at our study sites because the growth of the young was not influenced
by the type of parental care. Second, brooding time of the parents was not
different among male-only, female-only, and biparental broods, although the
power of our tests was low (males: ß = 0.167, females: ß = 0.324).
Brooding behavior may be more important during incubation than during brood
care in precocial birds. For example, experimental removal of one parent in
biparentally incubating species such as western sand-piper Calidris
mauri, killdeer Charadrius vociferus, and Kentish plover showed
that a single parent was unable to complete incubation because its body
condition deteriorated (western sandpiper;
Erckmann, 1983), although the
response of the remaining parent was not uniform in killdeer and Kentish
plover (Brunton, 1988;
Lessells, 1983).
Third, two parents may be better able to defend their young against
predators and conspecifics than a single parent. Fraga and Amat
(1996) argued that two Kentish
plovers may protect their young better than a single one by chasing
gull-billed terns Sterna nilotica, a major chick predator, away from
the brood and by distracting the predator's attention. Early detection of
predators may also be important, particularly at night when the chicks are
constantly brooded and predators can approach the brood within a short
distance. Mammals kill adult plovers at night
(Fraga and Amat, 1996;
Székely,
1996), which suggests that both the parents and their young are
vulnerable at this time. Also, two parents may defend their young better
against conspecifics than one parent. Infanticide by adults has been reported
in several populations (Fraga and Amat,
1996;
Székely,
1996; this study). Freudenthal
and Lessells (in press) argued that at high breeding density, both
parents are selected to stay with the young to provide efficient protection
from harassment by neighboring pairs. A major advantage of biparental care of
young is that while one parent fights against neighbors, the other can brood
the chicks. Such sharing of tasks may be particularly important at high
breeding density where fights are common, and shortly after hatching when the
chicks require almost constant brooding.
Finally, two parents are able to swap on-duty and off-duty roles so that
the chicks are attended by at least one parent most of the time. When the
female deserted the brood, the male increased its own feeding time while
attending their brood
(Székely
and Williams, 1995). Closely attending the chicks for a prolonged
period of time seems to be difficult, as the best foraging areas are often
separated from the brood-rearing ones. We observed that in uniparental broods
the parent often left the chicks on their own for a variable period of time
and fed 200-300 m away. While the parent is away the chicks are exposed to
predators and harassment by other adult Kentish plovers.
Parents may gain a further benefit from biparental brood care by retaining
the mate for future breeding (Hannon,
1984; Martin and Cooke,
1987). In our study in three control pairs in which both parents
stayed with the young until the chicks died (one pair), or fledged (two
pairs), both members of the pair bred again in 1996. In two out of three
pairs, the pair renested together. Because finding a mate may take a long
period of time (Székely et al., this issue), a
major advantage of renesting with the previous mate is that neither sex spends
time on searching for a new mate and that both birds gain quick access to a
mate that they chose in the first place. In the third pair, however, the
parents stayed together until their young fledged, although both of them bred
with a different mate afterward. Divorce from a mate is not unusual among
birds, although it often occurs between, and not within, breeding seasons
(Choudhury, 1996;
Ens et al., 1996).
In this study we also found that brood survival was not different between
male-only and female-only care in both sites. However, single males were more
successful in raising their young than single females in a Portuguese
population of Kentish plover
(Székely,
1996). A possible explanation for the difference between these
studies may relate to breeding density. Fraga and Amat
(1996) suggested that single
males may be better at protecting their young from conspecifics than are
single females. This difference between the sexes may be important only at
high breeding density such as in Portugal (5-16 nests/ha;
Székely
and Williams, 1995) and not at a lower breeding density such as in
Turkey (2-6 nests/ha; Székely T and Cuthill I,
unpublished data).
Seasonal variation in habitat quality
Unlike site A, at site B we found no difference among male-only,
female-only, and control broods. We offer three explanations for the
differences between sites. First, brood mortality (either due to predation or
infanticide) was lower at site B; thus the presence of both parents did not
increase the success of the brood. Second, the abilities of adults to provide
care may be different between the sites. If high-quality parents (e.g., older
plovers) bred at site B, whereas the young or inexperienced ones bred at site
A, then a single high-quality parent may be as successful in raising young as
two parents. However, the lack of difference in body size and body condition
between parents at site A and B does not support the latter argument. Finally,
we investigated fewer broods at site B (15) than at site A (27), thus the
statistical power of the tests such as the ANOVA was lower for broods at site
B (ß = 0.063) than at site A (ß = 0.586). Nevertheless, the
inspection of ranges in brood survival between uniparental and control broods
suggests that no difference would be detected even with larger sample sizes
(Figure 1).
Another difference between the sites was in chick growth. At site A chick growth declined over the season, whereas at site B it peaked in mid-season. Thus broods tended to grow faster early in the season at site A than at B, whereas later in the season the reverse was true. These results suggest that the abundance of food gradually declined at site A, whereas at B it peaked in mid-season. In line with this argument, broods fed on several temporary pools which dried out by the end of breeding season at site A. Broods fed mostly on the edge of the lake at site B, which probably provides less variable food over the season. The argument also agrees with our finding that broods moved a longer distance from their nest in site A than in B.
Based on these results we propose the following hypothesis. In early season food is abundant (particularly in site A), and there is little competition for food between families; thus brood survival is high at both sites. Biparental care makes little difference because chicks do not need to wander in search of food, and single parents do not need to leave their chicks for long when feeding. On the other hand, late in the season, temporary pools dry up (particularly at site A), and thus broods have to wander more to find food, and they get attacked. Single parents also have to move farther in search of food, so they leave chicks unprotected for longer. Therefore, chick mortality is expected to increase in uniparental broods late in the season, particularly at site A due to both predation and infanticide.
In conclusion, we showed that the payoffs from male-only, female-only, and
biparental care may depend on the environment. In particular, biparental care
may be more successful than uniparental care when mortality of young is high,
either due to predation or infanticide. We propose that the frequency of
infanticide is influenced by the spatial and temporal distribution of food. We
also found that brood survival decreased over the season. The latter result
may help to understand the seasonal variation in deserting time of female
Kentish plovers
(Székely
and Williams, 1994). In particular, broods hatched either early in
the season or at the end of the season were deserted after a short period of
time, whereas broods hatched in mid-season were attended for a longer period
of time. Our current study and the complementary one
(Székely et al., this issue) suggest that
females gain different rewards from deserting early and late broods. In
particular, females deserting early in the season quickly remate and renest
because their chance of getting a new mate is high, and their new brood will
likely fledge. On the other hand, females deserting late in the season desert
their brood because the survival chance of these broods is low and perhaps
they gain more by deserting, entering molt, and preparing for migration.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The project was funded by a Leverhulme Trust grant to A. I. Houston, I.C.C., and J. M. McNamara (F/182/AP) and by an Országos Tudományos Kutatási Alap grant to T.S. (T020036). T.S. was also supported by a Natural Environment Research Council grant to A. I. Houston, I.C.C., and J. M. McNamara (GR3/10957). Rings were provided by P.R. Evans (Durham University) and Radolfzell Vögelwarte, Germany. We thank to J. Kis for his assistance in the field. Ö. Karabaçak (Milli Parklar, Adana), M. Yarar (DHKD, Istanbul), G. Sarigül (DHKD, Tasucu) and V. van den Berk (National Reference Centre for Nature, Wageningen) helped us by providing practical information on feasibility of the field study. J. M. C. Hutchinson (Bristol University) wrote a program to calculate maximum likelihood estimates of brood survival. We thank A. I. Houston and two referees for their comments on the manuscript.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Balshine-Earn S, 1995. The evolution of parental care in cichlid fishes (PhD dissertation). Cambridge: University of Cambridge.
Balshine-Earn S, 1997. The benefits of uniparental versus biparental mouth brooding in Galilee St. Peter's fish. J Fish Biol 50:371-381.
Bart J, Tornes A, 1989. Importance of monogamous male birds in determining reproductive success. Behav Ecol Sociobiol 24:109-116.
Brunton DH, 1988. Energy expenditure in reproductive effort of male and female killdeer (Charadrius vociferus).Auk 105:553-564.
Choudhury S, 1996. Divorce in birds: a review of the hypotheses. Anim Behav 50:413-429.
Clutton-Brock TH, 1991. The evolution of parental care. Princeton, New Jersey: Princeton University Press.
Duckworth JW, 1992. Effects of mate removal on the behavior and reproductive success of reed warblers Acrocephalus scirpaceus. Ibis 134:164-170.
Ens BJ, Choudhury S, Black JM, 1996. Mate fidelity in monogamous birds. In: Partnerships in birds (Black JM, ed). Oxford: Oxford University Press; 344-401.
Erckmann WJ, 1983. The evolution of polyandry in shorebirds: An evaluation of hypotheses. In: Social behavior of female vertebrates (Wasser SK, ed). New York: Academic Press;113 -168.
Fraga RM, Amat JA, 1996. Breeding biology of a Kentish plover (Charadrius alexandrinus) population in an inland saline lake.Ardeola 43:69-85.
Freudenthal RF, Lessells CM, in press. Mate desertion of Kentish plovers, Charadrius alexandrinus, breeding in southern Portugal.Ardea .
Hannon SJ, 1984. Factors limiting polygyny in the willow ptarmigan. Anim Behav 32:153-161.
Helms CW, Drury WD Jr, 1960. Winter and migratory weight and fat: field studies on some North American buntings.Bird-Banding 31:1-40.
Lessells CM, 1983. The mating system of Kentish plovers Charadrius alexandrinus: some observations and experiments.Wader Study Group Bull 39:43.
Lessells CM, 1984. The mating system of Kentish plovers Charadrius alexandrinus. Ibis 126:474-483.
Liker A, 1995. Monogamy in precocial birds: a review.Ornis Hung 5:1 -14.
Magnin G, Yarar M, 1997. Important bird areas in Turkey. Istanbul: DHKD.
Markam S, Yom-Tov Y, Wright J, 1996. The effect of male removal on female parental care in the orange-tufted sunbird. Anim Behav 52:437-444.
Martin K, Cooch FG, Rockwell RF, Cooke F, 1985. Reproductive performance in lesser snow geese: are two parents essential?Behav Ecol Sociobiol 17:257-263.
Martin K, Cooke F, 1987. Bi-parental care in willow ptarmigan: a luxury? Anim Behav 35:369-379.
MINITAB, 1995. Reference manual. Release 10Xtra. State College: MINITAB.
Noszály G, Székely T, Hutchinson JMC, 1995. Brood survival of Kentish plovers (Charadrius alexandrinus) in alkaline grasslands and drained fish-ponds. Ornis Hung 5:15-21.
Schneider J, Lamprecht J, 1990. The importance of biparental care in a precocial monogamous bird the bar-headed goose Anser indicus. Behav Ecol Sociobiol 27:415-420.
Seaman JW, Walls SC, Wise SE, Jaeger RG, 1994. Caveat emptor: rank transform methods and interaction. Trends Ecol Evol 9:261-263.
SPSS, 1988. SPSS-X user's guide. Chicago: SPSS.
Székely T, 1996. Brood desertion in Kentish plover Charadrius alexandrinus: an experimental test of parental quality and remating opportunities. Ibis 138:749-755.
Székely T, Lessells CM,1993 . Mate change by Kentish plovers Charadrius alexandrinus. Ornis Scand 24:317-322.
Székely T, Webb JN, Houston AI, McNamara JM, 1996. An evolutionary approach to offspring desertion in birds. Current Ornithol 13:271-330.
Székely T, Williams TD,1994 . Factors affecting timing of brood desertion by female Kentish plovers Charadrius alexandrinus. Behaviour 130:17-28.
Székely T, Williams TD,1995 . Costs and benefits of brood desertion in female Kentish plovers, Charadrius alexandrinus. Behav Ecol Sociobiol 37:155-161.
Thomas LK, 1994. The evolution of parental care in assassin bugs (PhD dissertation). Cambridge: University of Cambridge.
Wolf L, Ketterson ED, Nolan V Jr, 1988. Paternal influence on growth and survival of dark-eyed junco young: do parental males benefit? Anim Behav 36:1601-1618.
Wright J, Cuthill I, 1989. Manipulation of sex differences in parental care. Behav Ecol Sociobiol 25:171-181.
Zar JH, 1996. Biostatistical analysis. London: Prentice-Hall.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. E. Garcia-Pena, G. H. Thomas, J. D. Reynolds, and T. Szekely Breeding systems, climate, and the evolution of migration in shorebirds Behav. Ecol., September 1, 2009; 20(5): 1026 - 1033. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Eldegard and G. A Sonerud Female offspring desertion and male-only care increase with natural and experimental increase in food abundance Proc R Soc B, May 7, 2009; 276(1662): 1713 - 1721. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Szekely, I. C. Cuthill, S. Yezerinac, R. Griffiths, and J. Kis Brood sex ratio in the Kentish plover Behav. Ecol., January 1, 2004; 15(1): 58 - 62. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||