Behavioral Ecology Vol. 11 No. 1: 40-43
© 2000 International Society for Behavioral Ecology
Risk taking during parental care: a test of the harm-to-offspring hypothesis
Department of Biology, University of Oslo, PO Box 1050 Blindern, N-0316 Oslo, Norway
Address correspondence to T. Slagsvold. E-mail: tore.slagsvold{at} bio.uio.no .
Received 9 February 1999; accepted 9 May 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Amount of risk taking during parental care is often explained in relation to the reproductive value of the offspring. The "harm-to-offspring hypothesis" focuses on the relative harm a period of no parental care can do to the offspring. According to this hypothesis, parents should take greater risks for offspring in poor condition than for offspring in good condition. We manipulated offspring condition in the pied flycatcher (Ficedula hypoleuca) and tested the harm-to-offspring hypothesis by exposing parents to a predator model (a sparrowhawk, Accipiter nisus). Time elapsed until a parent first entered the nest-box was used as a risk-taking measure. Parents spent significantly shorter time until first nest visit for offspring in poor condition than for offspring in good condition. Hence, the harm-to-offspring hypothesis was supported.
Key words: Ficedula hypoleuca, offspring condition, parental care, predation risk, reproductive value.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
In many birds, the single most important variable affecting fitness is nest predation (Ricklefs, 1969
). In altricial birds the chicks depend entirely on their
parents for food and protection. The adult birds face a dilemma when exposed
to a predator: should they defend the young and continue to feed them, or
should they hide and ensure their own survival? So far most studies on nest
defense have dealt with relationships between risk taking and reproductive
value of the brood (review by Montgomerie
and Weatherhead, 1988). Parents are assumed to increase their
defense and risk taking with increasing reproductive value of the brood.
Several studies have supported this hypothesis, though not all (e.g.,
Dale et al., 1996). Following
the review by Montgomerie and Weatherhead
(1988), Dale et al.
(1996) suggested a new
hypothesis to account for parental behavior when exposed to a predator.
According to the "harm-to-offspring hypothesis," level of risk
taking is adjusted to the harm the offspring will suffer from absence of
parental care (incubation, brooding or feeding in birds).
Here we present an experimental test of the harm-to-offspring hypothesis by
manipulating condition of nestling pied flycatchers (Ficedula
hypoleuca) and recording parental risk-taking behavior. We predict that
parents should take greater risks for offspring in poor condition than for
offspring in good condition (Dale et al.,
1996). This is because, regarding postfledging survival, the
marginal benefit gained by feeding offspring in poor condition would be
greater than that gained by feeding offspring in good condition (see
Magrath, 1991;
Slagsvold et al., 1995a).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
General
The study was carried out in Sørkedalen near Oslo in southern Norway during the breeding season of 1996. The study area consisted of a mixed coniferous and deciduous forest. All nestings took place in wooden nest-boxes fixed to trees about 1.5 m above ground with an entrance hole of 32 mm diam.
The pied flycatcher is a small (11-14 g), migratory, singlebrooded
passerine bird. It is short-lived, hole-nesting, and insectivorous. The female
builds the nest and incubates and broods alone, but both parents feed the
chicks, which fledge within 14-16 days of hatching
(Lundberg and Alatalo,
1992).
We captured and ringed pied flycatcher males on their arrival at nest-boxes with a metal ring and combinations colored rings. By making daily observations in the field, we collected data on the mating status of all birds, date of laying of first egg, clutch size, hatching date, and brood size.
Two stuffed specimens of female sparrowhawk (Accipiter nisus),
both in resting position, were used. There was no significant difference
between the response to the two predator models
(Listøen, 1997). The
sparrowhawk is a serious threat to adult pied flycatchers and is the only
common bird of prey in our study area
(Slagsvold et al., 1995b).
Experiment
We selected 20 monogamous pairs for the study. A pairwise design was used,
pairing broods with regard to date of first egg. The parents were exposed to
the models for the first time when the offspring were 8 days old (0 = day of
hatching). The trials were carried out 3 days in a row (days 8, 9, and 10) for
every pair, each day with different offspring conditions. Brood size (5-7
chicks) was never manipulated.
On the first day of the trial (day 8), offspring condition was unmanipulated (treatment U). On day 9, the two broods in a pair received different treatments S or F. In treatment S, all chicks in one nest were removed and starved by being held in a warming bag without food for about 3 h. In treatment F (fed), the chicks from the other nest were equally divided between the two nests. We assumed that these chicks would receive food from the adults and gain weight during the treatment time (3 h). On the final day (day 10), the treatments were switched. The brood which was starved on day 9 was now divided between the nests. The brood left was starved as described above. On day 9 and 10 we weighed the chicks before and after the treatment. After treatment we returned the chicks to their respective nests and began presentation the predator model.
When recording risk response, an observer settled in a hide 
10 m from
the nest-box. Each parent was allowed to visit the nest twice before the model
was presented to ensure that they had noticed the condition of the offspring.
These nest visits were brief, suggesting that the parents were quickly able to
judge the condition of the chicks. The model was then placed on top of the
nest-box and the observer went back to the hide. The model was removed after
both parents had spotted it. On average, the model was presented for 3 min.
Observation started when the observer had returned to the hide after removing
the model. The time elapsing until first entry in the nest-box (usually to
feed the nestlings) was used as a risktaking measure, assuming risk of
predation to be higher the sooner parents resume feeding after exposure to the
predator (Dale et al., 1996).
In the study area, sparrowhawks are serious threats to male and female pied
flycatchers, particularly during the nestling period
(Slagsvold and Dale, 1996;
Slagsvold et al., 1995b).
Parents may be particularly vulnerable when they enter the nest hole.
Nest-boxes were situated only 1.5 m above ground, and sparrowhawks attack from
above. We have witnessed hawks killing females perched at the entrance hole
(Slagsvold and Dale, 1996).
The trial ended if no parent had entered within 60 min. The predator model was
always presented simultaneously to the two nests of a pair (i.e., with two
observers and two models presented at the same time of day).
We decided which brood of a pair to be starved first by flipping a coin, except when the broods were of different ages (n = 2); then the oldest brood was starved first. The predator model to be used was also decided by flipping a coin. At the same nest the same model was used all days of the trial. When dividing broods uneven in size, the majority of the chicks were left in the original nest.
All trials were spread throughout the day (0700-1900 h) within the normal
feeding time of the pied flycatcher
(Lundberg and Alatalo, 1992).
Time until first nest visit was not related to time of day or time of breeding
season (26 June-9 July; Listøen,
1997). The model was never presented while it was raining, but
during a few trials it started to rain. A few trials (3 of 54) were postponed
a day because of bad weather, and for the same reason a few trials could not
be carried out and so sample size was reduced.
Time until first nest visit in untreated (U), S, and F groups met a normal distribution. All statistical tests are two tailed.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Response to the model
The parents reacted strongly to the presence of the predator model, making loud alarm calls. They generally stayed in the nest-box tree or in a tree nearby after the removal of the model and continued to call, and they usually visited the nest-box opening several times before entering. Some parents entered the nest-box after a few minutes; others spent more than half an hour before entering. First visits into the nest-box were of short duration, always < 1 min. There was no bias in which sex entered first (Listøen, 1997
). Second
visits were on average 4.4 min (SD = 5.3, n = 54) after the first
one. All broods lost weight during the 3 h of starvation, with a mean loss of 0.57 g per chick (Table 1), amounting to 4.5% of body mass. Fed chicks did not always gain weight during the treatment period but had a mean body mass increase of 0.18 g (1.4%). Hence, starving resulted in a significant change (paired t test; t17 = 12.63, p <.001) but feeding did not (t16 = 1.51, p =.15). Change in chick body mass during the treatment period was significantly different between S and F (same test, t15 = 6.30, p <.001).
|
On average, chicks did not gain weight from day 9 to day 10 (Figure 1). This shows that treatment and predator exposure not only influenced the immediate level of chick hunger but also had a more long-term effect on offspring condition. Measures in Figure 1 were from just after the 3-h period of brood treatment, just before model trials started.
|
The time elapsing until the parents visited the nest was significantly shorter when the offspring had been starved than during the initial trial (U; paired t test; t17 = 5.68, p <.001; Figure 2). Parents took significantly less time to enter the nest-box when offspring had been starved than when they had been fed (t15 = 3.65, p =.002). There was no significant difference in the time elapsing until first nest visit between the initial trial (U) and the F trial (F; same test; t16 = 1.02, p =.32).
|
On average, parents seemed to take less time before entering the nest-box on the last day of trial (day 10) than on the previous day (day 9). This decline was significant comparing S-treatment broods between days 9 and 10, but not when comparing F-treatment broods between the 2 days (Figure 2).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
After exposure to a predator model, parent pied flycatchers took significantly less time to make the first nest visit when the brood had been starved than before manipulation of offspring condition. Parents also took greater risks for chicks in poor condition than for fed chicks. Hence, these results provide experimental support to the harm-to-offspring hypothesis.
An alternative explanation is that parents of starved chicks may have
entered the nest sooner to stop nestlings from begging because begging may
attract nest predators (e.g., Haskell,
1994). This explanation is plausible but perhaps not likely in
case of pied flycatchers because they are hole nesters and not very vulnerable
to nest predation, at least not in our study area (Slagsvold T, unpublished
data). In an experimental study, a cost of begging was found for nests on the
ground but not for nests in trees (Haskell,
1994). Unfortunately, we did not record nestling begging when
parents were about to enter the nest for the first time after having been
exposed to the predator model. However, our impression was that the nestlings
did not beg at that time because parents were still giving alarm calls or had
recently been given such calls.
A third explanation is as follows. In absence of predators, frequency of
provisioning would increase when chicks have been starved. Hence, in case of
the S treatment, parents may simply have entered the nest sooner than in case
of the F treatment because they had shorter feeding trips before returning to
the nest. This explanation is not likely because pied flycatchers normally
enter the nest with food at intervals of only 1-2 min
(Lundberg and Alatalo, 1992).
In the present case the mean difference in entry time between the two
treatments was much longer (7 min; Table
1).
The harm-to-offspring hypothesis and the reproductive value hypothesis are
not mutually exclusive but focus on different costs and benefits. The present
study was not primarily designed to test the latter hypothesis because we
tried to keep offspring reproductive value as constant as possible. This was
done by the pairwise design where broods were paired with regard to time of
breeding and where brood size was not manipulated. In addition, all broods
were starved (on alternate days), and when there was a difference in age, the
oldest brood (i.e., with highest reproductive value) was starved first.
However, by manipulating offspring condition, we manipulated the reproductive
value of the brood; starvation of chicks not only had an immediate effect on
nestling hunger, but a more long-term influence on body mass, which in turn
might have consequences for offspring survival and breeding success (e.g.,
Magrath, 1991;
Slagsvold et al., 1995a).
Hence, our study allowed a test of the reproductive value hypothesis.
According to the hypothesis, parents should take greater risks the more
valuable the brood (Andersson et al.,
1980; Clutton-Brock,
1991; Lazarus and Inglis,
1986; Montgomerie and
Weatherhead, 1988; Onnebrink
and Curio, 1991). We found the opposite result because parents
took greater risks for chicks in poorer condition. Parents seemed to enter the
nest sooner on the last day of trial than on the previous day (but not
significantly so), and this may be interpreted as being willing to take
greater risk for older offspring, as predicted from the reproductive value
hypothesis. However, we suggest that parental behavior was rather related to
the low body masses of the chicks and their state of hunger (the
harm-to-offspring hypothesis).
That parents entered the nest sooner on the final day of trial might
indicate habituation, and if so this would affect our conclusion
(Knight and Temple, 1985).
However, nestlings' body masses on days 9 and 10 did not differ
(Figure 1), even though pied
flycatcher nestlings are expected to gain 1-1.5 g per day at this age
(Lundberg and Alatalo, 1992).
Hence, chicks were on average in poorer condition before treatment on day 10
than on day 9. From the harm-to-offspring hypothesis, the parents were
therefore predicted to take slightly greater risks on day 10 than on day 9, as
observed. Note also that the strong difference in risk-taking behavior between
the two treatments S and F on day 10
(Figure 2) cannot be explained
by habituation because all nests had been exposed the same number of times
(two) to a predator model before the final trial. The harm-to-offspring
hypothesis, on the other hand, predicts this increased difference because
starving chicks that already were in poor condition may have resulted in a
stronger effect of treatment on day 10.
In conclusion, manipulation of offspring condition made possible an
experimental test of the harm-to-offspring hypothesis. The hypothesis was
supported, confirming two recent descriptive studies of risk-taking behavior
in passerine birds (Bures and Pavel,
1997; Dale et al.,
1996), and an experimental study in which body temperature of 1-
to 3-day-old nestlings were experimentally reduced
(Bures and Horackova,
1998).
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We thank Wibecke Johansen, Helene Lampe, and Helge Rinden for field assistance and Svein Dale for comments on the manuscript.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Andersson M, Wiklund CG, Rundgren H, 1980. Parental defence of offspring: a model and an example. Anim Behav 28:536-542.[Web of Science]
Bures S, Horackova, K 1998. A test of the harm to offspring hypothesis in exothermic pied flycatcher (Ficedula hypoleuca) nestlings. Folia Zool 47:75-78.
Bures S, Pavel V, 1997. The effect of nestling condition on risk-taking in meadow pipits. Anim Behav 54:1531-1534.[Web of Science][Medline]
Clutton-Brock TH, 1991. The evolution of parental care. Princeton, New Jersey: Princeton University Press.
Dale S, Gustavsen R, Slagsvold T, 1996. Risk taking during parental care: a test of three hypotheses applied to the pied flycatcher. Behav Ecol Sociobiol 39:31-42.
Haskell D, 1994. Experimental evidence that nestling
begging behaviour incurs a cost due to nest predation. Proc R Soc Lond
B 257:161
-164.
Knight RL, Temple SA, 1986. Why does intensity of avian nest defense increase during the nesting cycle? Auk 103:318-327.[Web of Science]
Lazarus J, Inglis IR, 1986. Shared and unshared parental investment, parent-offspring conflict and brood size. Anim Behav 34:1791-1804.
Listøen C, 1997. Risk taking during parental care: a test of the harm to offspring hypothesis (Cand scient thesis). Oslo: University of Oslo.
Lundberg A, Alatalo RV, 1992. The pied flycatcher. London: Poyser.
Margrath RD, 1991. Nestling weight and juvenile survival in the blackbird Turdus merula. J Anim Ecol 60:335-351.
Montgomerie R, Weatherhead PJ, 1988. Risks and rewards of nest defence by parent birds. Q Rev Biol 63:167-187.[Web of Science]
Onnebrink H, Curio E, 1991. Brood defence and age of young: a test of the vulnerability hypothesis. Behav Ecol Sociobiol 29:61-68.[Web of Science]
Ricklefs RE, 1969. An analysis of nesting mortality in birds. Smithson Contrib Zool 9:1-48.
Slagsvold T, Amundsen T, Dale S, 1995a. Costs and benefits of hatching asynchrony in blue tits Parus caeruleus.J Anim Ecol 64:563-578.
Slagsvold T, Dale S, 1996. Disappearance of female pied flycatchers in relation to breeding stage and experimentally induced molt. Ecology 77:461-471.
Slagsvold T, Dale S, Kruszewicz A, 1995b. Predation favours cryptic coloration in breeding male pied flycatchers. Anim Behav 50:1109-1121.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||