Why fight? Socially dominant jackdaws, Corvus monedula, have low fitness

Abstract

Social dominance is intuitively assumed to be associated with higher fitness, because social dominance implies better access to resources. We found that, in a colony of jackdaws, the dominant males consistently produced fewer fledglings, which had lower chances of survival to 1 year of age. Laying date and clutch size were independent of dominance, but females that mated with dominant males were in poorer condition and laid smaller eggs. Parental survival was independent of social dominance, and the frequency of extrapair fertilizations in jackdaws is negligible. Dominance was a stable trait of individuals, and not a state that all individuals eventually attained. We conclude that, in this colony, dominant jackdaws had lower fitness. To our knowledge, this is the first example of such a pattern in a free-living species. We hypothesize that the high density of our colony resulted in high testosterone titres, which suppressed paternal care of mate and offspring to the extent that it outweighed the benefits of higher resource access.

Full text

Why fight? Socially dominant jackdaws,
Corvus monedula, have low fitness
SIMONVERHULST*&H.MARTIJNSALOMONS*
Zoological Laboratory of the University of Groningen
(Received 3 October 2003; initial acceptance 26 November 2003;
final acceptance 31 December 2003; published online 25 August 2004; MS. number: 7872)
Social dominance is intuitively assumed to be associated with higher fitness, because social dominance
implies better access to resources. We found that, in a colony of jackdaws, the dominant males consistently
produced fewer fledglings, which had lower chances of survival to 1 year of age. Laying date and clutch size
were independent of dominance, but females that mated with dominant males were in poorer condition
and laid smaller eggs. Parental survival was independent of social dominance, and the frequency of
extrapair fertilizations in jackdaws is negligible. Dominance was a stable trait of individuals, and not a state
that all individuals eventually attained. We conclude that, in this colony, dominant jackdaws had lower
fitness. To our knowledge, this is the first example of such a pattern in a free-living species. We hypothesize
that the high density of our colony resulted in high testosterone titres, which suppressed paternal care of
mate and offspring to the extent that it outweighed the benefits of higher resource access.
Ó2004 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
Social groups are usually structured in the sense that some
individuals are consistently more successful at obtaining
resources when there is a conflict (Allee 1952; Drews
1993). Because socially dominant individuals, by defini-
tion, have priority of access to resources, it is generally
assumed that these individuals also attain the highest
reproductive success. The fitness advantage to socially
dominant individuals is crucial in understanding the
existence of dominance hierarchies; if dominants did
not benefit, investment in acquiring and maintaining
social dominance would be wasted (Pusey & Packer 1997).
Such investments are, for example, costly signals used in
agonistic interactions (Zahavi & Zahavi 1997) and harm-
ful side-effects of high androgen levels (Folstad & Karter
1992; Frank et al. 1995; Packer et al. 1995; Buchanan et al.
2001), and that agonistic interactions consume time and
energy and increase the risk of an injury. Such costs could
potentially outweigh the benefits of having priority of
access to resources, resulting in neutral or even negative
effects of social dominance on fitness (Rohwer & Ewald
1981; Ellis 1995). Many studies of primates have reported
positive or neutral effects of dominance on reproductive
success (reviewed in Ellis 1995), but this relation has been
little studied in other taxa (reviews in Ellis 1995; Piper
1997; Koivula 1999). Negative effects of dominance on
indicators of reproductive success have also been found,
but only in captive animals (Ellis 1995). However, the
greater access to resources ensured by being dominant is
unlikely to yield a fitness advantage when resources are
abundant, as is usually the case in captivity. For this
reason, and because free-living individuals face a multitude
of trade-offs and behavioural options that are absent in
captivity, the fitness consequences of dominance can be
assessed only in free-living animals.
We studied the life history consequences of social
dominance in a colony of free-living jackdaws. Jackdaws
spend much of their time in groups that have a strong
linear dominance hierarchy, in the sense that jackdaws
rarely lose a conflict with an individual with lower social
status (Tamm 1977; Ro
¨ell 1978; Wechsler 1988). They are
facultatively colonial during breeding, and colonies can be
found in buildings as well as under more natural con-
ditions, for example, in rabbit warrens (Ro
¨ell 1978; Vogel
2002). We determined social dominance by observing
interactions over an artificial food source. The outcome of
a conflict can be state dependent; hungrier birds may, for
example, win more conflicts over food (Andersson &
Ahlund 1991; Cristol 1992). However, jackdaws that
were successful in interactions over food also had
primary access to available nestboxes and succeeded in
defending more nestboxes during winter (Ro
¨ell 1978).
Correspondence: S. Verhulst, P.O. Box 14, 9750 AA Haren, The
Netherlands (email: s.verhulst@biol.rug.nl).
*The authors contributed equally to this paper.
777
0003–3472/03/$30.00/0 Ó2004 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
AN IMAL BE HAV IOUR , 2004, 68, 777–783
doi:10.1016/j.anbehav.2003.12.020

This information, together with the observation that the
hierarchy is highly stable over successive years, indicates
that dominance in competition over food reflects re-
source-holding potential (Parker 1974), rather than some
transient effect of state.
There are two previous studies of the relation between
social dominance and reproductive success in jackdaw
colonies. Henderson & Hart (1995) showed that dominant
pairs produced more fledglings. In contrast, Ro
¨ell (1978)
reported evidence suggesting that dominant pairs had
fewer fledglings; neither study compared parental sur-
vival. However, Ro
¨ell assessed dominance in June, during
or after chick production, and he suggested that high
dominance was the consequence of being without chicks
(Ro
¨ell observed temporary rank shifts in the breeding
season that were subsequently reversed in autumn).
Furthermore, Ro
¨ell presented only the cumulative fledg-
ling production of eight males over 5 years. These males
were selected from a much larger sample, because they
were present in all 5 years of the study, and these males
may therefore constitute a biased sample of the popula-
tion. Data on fledgling quality were not available, and low
fledgling production might have been compensated by
producing high-quality fledglings. However, finding a spe-
cies where the fitness consequences of social dominance
depend on the ecological or social setting to the extent
that dominance even results in low fitness would offer
a valuable tool to study the costs and benefits of domi-
nance and increase our understanding of the evolution of
dominance and social structures. We therefore studied the
association between dominance and reproductive success
in the same colony as did Ro
¨ell, but we determined social
dominance before the breeding season, and we also
measured parental survival and offspring quality in
relation to social dominance.
METHODS
Study Population
We studied free-living jackdaws in the colony at the
Zoological Laboratory in Haren, The Netherlands, a semi-
urban environment. The colony was established in 1965
and enlarged to 36 nestboxes in 1996 when the study was
resumed. Data on dominance and reproduction reported
in this study were collected in 1998 and 2000, and
survival was measured up to 2003. The study was carried
out under licence from the Ethical Committee for animal
experiments of the University of Groningen.
Nestboxes were checked daily, starting in the first week
of April, until the clutch was complete. To determine from
which egg a chick had hatched (for another study), we
moved clutches to an incubator 1–2 days before the
estimated hatching date (temperature 37.7 C, RH 75%).
Clutches were exchanged for hardboiled quail eggs, which
birds readily accepted. Length and width of the eggs were
measured (G0.1 mm), and egg volume (V,incm
3
) was
estimated using the formula: VZ(p!A
2
!L!K)/6,
where Ais width, Lis length and, for jackdaws, the
constant, K, is 0.00096 (Soler 1988). Eggs in the incubator
were checked at least every 2 h during the daylight period.
Eggs that did not hatch were dissected to check for
presence of an embryo. Hatching success of eggs with an
embryo was 95.7% (NZ211, years combined). Hatch-
lings were weighed, individually marked by clipping the
tip of one or two nails and returned to their nests. An
equal number of quail eggs was subsequently removed
from the nest.
We checked the survival of the chicks in the nest every 5
days (day 1 Zhatch date of the first egg). At days 10, 20
and 30, we also weighed the chicks and measured tarsus
and wing length (days 20 and 30). At day 30, shortly
before fledging, the chicks were ringed.
Breeding birds were individually marked with colour
rings and a metal numbered ring. Birds were caught in
a large baited cage or in their nestbox using trap doors.
Early in the nestling period (day 5), a sample of adults was
captured in 1998, and most breeding birds were captured
in 2000. Biometric characteristics (tarsus and wing length,
mass) were measured, a small blood sample (!60 ml) was
taken for DNA from the brachial vein, and most birds were
released within 20 min.
To estimate survival and identify nestbox owners, we
regularly observed colour-ringed birds from mid-February.
Whenever possible, other colour-ringed jackdaws visiting
the colony and the surrounding areas were also identified.
Resighting probability is the product of the probability
that a bird is still alive and the probability that a living
individual is seen, and survival analysis usually requires
estimating both these parameters using capture–recapture
analysis. However, the probability of observing individu-
als known to be alive was high (91–100% over the study
years 1998–2003 for both sexes, except for 1999, when
fieldwork intensity was low). We therefore used the
Kaplan–Meijer test to compare survival between dominant
and subdominant birds. Resighting probability of an
individual was independent of the fate of its partner in
all years (0.34 !P!0.99).
Dominance
Agonistic interactions were recorded during March and
the first half of April of 1998 and 2000, until the first egg in
the colony was laid. Conflicts are resolved in different ways,
either through displacement, threat or fights, and these
were all scored to obtain a rank order (Ro
¨ell 1978). To stage
conflicts, we offered food in small pits (diameter 10 cm, one
in 1998, two in 2000) approximately 10 m from the nearest
nestbox. The pits were 30 m apart. At these pits, only one
jackdaw or a jackdaw pair could eat at a time. The feeding
pits were filled only preceding an observation period. We
used only interactions at the feeding pit in which both
participants were ringed to calculate social rank
(XGSD ¼80:0G58:2 interactions=male, NZ42 males).
The success in agonistic interactions (R) of bird J was
calculated following Henderson & Hart (1995):RZ(N
interactions won by J/Ninteractions lost by J) !(N
individuals supplanted by J/Nindividuals that supplanted
J). This equation takes both the proportion of interactions
won and the proportion of individual birds supplanted
ANIMAL BEHAVIOUR, 68, 4
778

into account; other methods to assign rank yielded almost
identical results. Based on the success score (R), a rank
number was assigned to each bird. We then scaled rank
between 0 and 1 (most and least dominant male, re-
spectively), because the number of birds in the hierarchy
differed slightly between years.
Male jackdaws are dominant over females, and the
females’ success in conflicts is highly dependent on the
rank and proximity of her partner (Lorenz 1931; Ro
¨ell
1978; Wechsler 1988). Consequently, we cannot deter-
mine female rank independently and instead used the
rank of the male to characterize the rank of the pair.
Statistical Analysis
A number of individuals bred in the colony in both
1998 and 2000, and different breeding attempts from the
same individuals cannot be considered independent sam-
ples. Data were therefore analysed with a repeated meas-
ures hierarchical linear model to avoid pseudoreplication,
using the program MLwiN (version 1.10, Rabasch et al.
2000). Statistical significance of variables was assessed
from the increase in deviance (DDev) when the variable
was removed from the model. The change in deviance is
asymptotically distributed as c
2
with corresponding
change in degrees of freedom (Snijders & Bosker 1999).
Unless stated otherwise, data from both years (1998 and
2000) were combined in the analysis. Separate regressions
(calculated using SPSS 9.0) are also shown in Table 1 for
comparison between years.
RESULTS
The Hierarchy
Birds present in both 1998 and 2000 had very similar
ranks in the 2 years (F
1,7
Z12.5, PZ0.01; Fig. 1). Despite
becoming 2 years older, males did not acquire higher
dominance between 1998 and 2000 (mean difference Z
0.03, paired ttest: t
8
Z0.41, PZ0.7). The rank of
birds determined at both feeding pits separately was
highly correlated (Pearson: r
12
Z0.90, P!0.001; data
from 2000), indicating that rank was not site dependent
on the scale of the colony. Observations at the two feeding
pits were therefore combined. The constancy of domi-
nance rank in space and time shows that our method
yields highly repeatable estimates of rank.
Reproduction
Rank was not correlated with laying date of the first egg
or clutch size (Table 1). However, more dominant pairs
produced smaller eggs (Fig. 2) and smaller hatchlings
(Table 1). Nestling mortality was higher in nests of
dominant pairs, especially in 1998. This resulted in fewer
fledglings produced by dominant birds (Fig. 3a). Further-
more, chicks of dominant pairs fledged with lower mass
(Fig. 3b) and smaller size (Table 1). Too few fledglings were
observed in later years to test directly whether dominant
birds produced fewer surviving offspring, and we therefore
Table 1. Dominance rank and reproduction
Dependent variables
1998 2000 Years combined
rNP rNPDDev NP
Laying date 0.26 16 0.33 0.17 22 0.45 1.81 38 0.18
Clutch size 0.21 16 0.44 0.22 22 0.33 1.70 38 0.19
Egg volume 0.49 16 0.056 0.46 22 0.03 9.24 38 0.002
Hatchling mass 0.55 14 0.04 0.36 22 0.10 7.12 36 0.008
Fledgling mass 0.72 9 0.03 0.63 18 0.005 14.02 27 !0.001
Fledgling tarsus length 0.42 9 0.27 0.39 18 0.11 3.71 27 0.054
Fledgling wing length 0.75 9 0.02 0.39 18 0.11 7.60 27 0.006
Number of fledglings 0.54 16 0.03 0.15 22 0.50 4.56 38 0.03
Results are calculated for each year separately, and years were combined using hierarchical linear models. All tests for
years combined are with 1 df. Sample sizes for fledgling mass and size are lower because not all nests resulted in
fledged young. Significant correlations are in bold.
0 0.2 0.4 0.6
Y = X
0.8 1
Dominant Subdominant
Dominance rank 1998
0
0.2
0.4
1
0.6
0.8
Dominance rank 2000
Figure 1. Dominance rank of jackdaws in 1998 and 2 years later.
Low rank indicates high dominance. Line indicates equal values
(YZX).
VERHULST & SALOMONS: LOW FITNESS IN DOMINANT BIRDS 779

verified whether fledging mass predicted survival pros-
pects. For fledglings in 1996–2002, we compared fledging
mass between fledglings found dead within a few weeks of
fledging (XGSD mass at fledging ¼155:0G36:3g,NZ35),
fledglings whose fate was unknown (194.8 G31.3 g,
NZ107) and fledglings seen alive the next year or later
(212.2 G20.3 g, NZ24). Using hierarchical linear mod-
els (with year and parents as hierarchical levels), we
found that fledglings that survived their first year
weighed more than fledglings whose fate was unknown
(DDev
1
Z8.08, P!0.005), which in turn weighed more
than fledglings found dead within a few weeks of
fledging (DDev
1
Z27.95, P!0.001). When this analysis
was repeated using mean fledgling mass per brood as the
independent variable, and either the proportion of
fledglings surviving or the proportion of fledglings
found dead as the dependent variable, then the same
result emerged (both DDev
1
O7.4, P!0.01).
Parental Condition
Social dominance was independent of body size (length
of tarsus and wing, males: F
1,18
!1.1, P> 0.3; females:
F
1,22–21
!0.2, P> 0.6). Enlarging the sample size by in-
cluding birds captured outside the 1998 or 2000 breeding
seasons did not change this result. Condition (residuals
from regression of mass on tarsus) was lower in females
that mated with dominant males (F
1,20
Z6.28, PZ0.02;
Fig. 4a), but male condition was independent of rank
(F
1,15
Z0.85, PZ0.4; Fig. 4b). The relation between
dominance rank and condition differed significantly
between the sexes (rank by sex interaction: F
1,36
Z4.20,
PZ0.048).
To check whether effects of female condition could
explain lower reproductive success of dominant jackdaws,
we included female condition in the model for egg
volume and fledgling quality. Fledgling production was
not correlated with female condition (data not shown),
perhaps because most condition data were from 2000, and
the effect of dominance on fledgling production was weak
in that year (Table 1). The effect of social dominance on
egg volume was mediated through condition of the female
(DDev
1
Z7.52, NZ23, P!0.01; Fig. 5), because social
dominance did not explain additional variation when
female condition was taken into account (DDev
1
Z1.8,
PZ0.2). Fledgling mass was correlated with egg volume,
but there was also a significant effect of social dominance
(Table 2). Thus, female condition and egg volume ex-
plained only part of the correlation between social
dominance and fledgling quality.
Parental Survival
Dominant jackdaws could compensate for lower annual
reproductive output with a longer reproductive life span.
We analysed survival of jackdaws up to the breeding
season of 2003. Birds were assigned to two groups of
equal size, which comprised the top and bottom 50% of
the hierarchy. Data for both years were combined (first
breeding season, either 1998 or 2000, was assigned 0, and
each bird was used only once in the analysis). Survival was
independent of rank in both sexes (females: log
0 0.2 0.4 0.6 0.8 1
Dominant Subdominant
Dominance rank
–2
–1
2
0
1
Egg volume (cm3)
Figure 2. Egg volume in relation to parental dominance rank. Thin
and dashed lines are regression lines for data from 1998 (C) and 2000
(B), respectively; bold line shows linear regression for data from both
years combined. Data shown are differences between observed values
and annual means. Low rank indicates high dominance.
Dominant Subdominant
Dominance rank
0.2 0.4 0.6 0.8 1
0
1
2
5(b)
3
4
Number of fledglings
0 0.2 0.4 0.6 0.8 1
–100
–50
100 (a)
0
50
Fledglings mass
Figure 3. Dominance and reproductive success. (a) Fledgling mass
(relative to annual mean). (b) Number of fledglings. Thin and
dashed lines are regression lines for data from 1998 (C) and 2000
(B), respectively; bold line shows linear regression for data from
both years combined. Low rank indicates high dominance.
ANIMAL BEHAVIOUR, 68, 4
780

rank Z1.39, NZ30, PZ0.24; males: 0.38, NZ30,
PZ0.54; Fig. 6). Average survival rate was 72.7% for
females (NZ30) and 70.8% for males (NZ30), in
agreement with estimates on the basis of ring recoveries
(69%; Dobson 1990).
DISCUSSION
Social dominance was associated with low fledgling pro-
duction and low fledgling quality (Fig. 3). For example,
the five least dominant males annually fledged 1.2 young
of viable quality ( fledging mass O170 g), but the five most
0 0.2 0.4 0.6 0.8 1
Dominant Subdominant
Dominance rank
–20
–15
20 (b)
0
15
10
5
–5
–10
Male condition (g)
0.2 0.4 0.6 0.8 1
–30
–20
30 (a)
0
20
10
–10
Female condition (g)
0
Figure 4. Social dominance and condition (residuals from regression
of mass on tarsus) of (a) females and (b) males. Low rank indicates
high dominance.
–20 –10 0 10 20 30
Female condition (g)
–2
–1
2
0
1
Egg volume (cm3)
Figure 5. Egg volume in relation to female condition. Egg volume is
deviation from annual mean egg volume, and female condition is
the residual from a regression of mass on tarsus.
Table 2. Effects of dominance, egg volume and female condition on
fledgling mass
DDev Coefficient (SE) Ddf P
Null model 179.51
Final model 158.46 15
Constant 39.55 (10.93) C1 0.001
Dominance rank C10.42 70.88 (18.87) C1 0.001
Egg volume C6.32 16.57 (6.02) C1 0.012
Rejected term
Female condition 0.59 1 0.44
Data (mean g per brood) were analysed using hierarchical linear
models. Null model includes the constant only. Final model includes
all significant parameters. Changes in deviance (DDev) and in
degrees of freedom (Ddf ) indicate the changes when parameters are
dropped from the final model one at a time (or added to the final
model for rejected terms). Female condition is the residual from
a regression of mass on tarsus. NZ18 broods.
012345
Years
0.05
0.1
0.2
0.4
0.6
0.8
1
Male survival
012345
0.05
0.1
0.2
0.4
0.6
0.8
1
Female survival
(b)
(a)
Figure 6. Parental survival in relation to social dominance in (a)
females and (b) males. All birds were assigned to two groups of
equal size on the basis of their rank (dominants: C; subdominants:
B). Yaxis (log scale) shows proportion of birds still alive. Year = 0
denotes first year that rank was known (1998 or 2000). Broken line
between years 3 and 4 is because the sample size is lower for years 4
and 5, with only birds that bred in 1998.
VERHULST & SALOMONS: LOW FITNESS IN DOMINANT BIRDS 781

dominant males fledged only 0.5 viable young. This
decrease was not compensated with greater longevity,
because survival was independent of social dominance
(Fig. 6). In theory, males could also compensate for low
success with their own partner by having extrapair
fertilizations, but jackdaws have an exceptionally strong
pair bond (Lorenz 1931), and the frequency of extrapair
fertilizations is practically zero (Liebers & Peter 1999;
Henderson et al. 2000). Dominance was a permanent trait
of individuals, and not a state that all surviving indivi-
duals eventually attained. These findings led us to conclude
that dominant jackdaws in our colony had lower fitness
than did subdominants. To our knowledge, this is the first
example of such a pattern in any free-living species.
Hypotheses that explain low fitness of dominant birds
are based on one of two assumptions. Either it is assumed
that there is a causal relation between dominance and
reproductive success, or it is assumed that a third factor
causes both high dominance and low reproductive suc-
cess. One possible third variable is senescence, when aging
is associated with a decline in reproductive success and an
increase in dominance. Although Henderson & Hart
(1995) reported a correlation between age and dominance,
we consider this explanation unlikely because dominance
was independent of age in our colony, both in recent years
(Fig. 1; S. Verhulst & H. M. Salomons, unpublished data)
and in the 1970s (Ro
¨ell 1978), which may result from the
near absence of breeding yearlings in our colony. We next
consider hypotheses that assume a causal relation be-
tween dominance and reproductive success, although we
acknowledge that we cannot rule out an unknown third
factor.
In his extensive review, Ellis (1995) emphasized the
variation in the fitness consequences of dominance, and
proposed that dominance enhances reproductive success
in particular when food availability is low. In agreement
with this proposition, Henderson & Hart (1995) found in
jackdaws that dominance had a stronger effect on fledg-
ling production in years when overall success was low.
Using simple optimality reasoning, this proposition can
logically be extended to explain a negative association
between dominance and reproductive success: when costs
are associated with acquiring and maintaining domi-
nance, and these costs are not compensated with in-
creased resource access (because resources are abundant
regardless of status), the net effect of dominance on
reproductive success will be negative. However, there is
no evidence that resources were abundant in our colony.
One or more nestlings starved in almost every nest, and
fledgling production was approximately equal in our
colony and the colony studied by Henderson & Hart,
where dominance enhanced reproductive success. Thus,
high resource abundance (Ellis 1995) is unlikely to explain
the negative association that we found between domi-
nance and reproductive success.
Females paired with dominant males had poorer condi-
tion and produced smaller eggs, and this explained at least
part of the effect of dominance on reproductive success
(Table 2). Dominant males could have partners in poor
condition because high-quality females prefer subdomi-
nant males as partners, and low-quality females are
‘making the best of a bad job’ by pairing with a dominant
male (Qvarnstro
¨m & Forsgren 1998). Another possibility is
that females suffer from having a dominant male as
partner. These hypotheses are complementary, because if
females suffer from having a dominant partner, this would
explain why such males are less attractive. Dominant
jackdaws do show more aggression towards their partner
than do subdominant males (1978), and female prefer-
ence of subdominant males to avoid intrapair aggression
has also been demonstrated experimentally in Japanese
quail, Coturnix coturnix japonica (Ophir & Galef 2003). This
evidence lends credibility to these hypotheses, but for
a rigid test, mate choice experiments with jackdaws are
required.
Henderson & Hart (1995) reported a positive effect of
dominance on reproductive success, and a comparison of
our colonies may provide further indications why domi-
nance had a negative effect on reproductive success in our
colony. Both colonies had approximately the same num-
ber of breeding pairs and comparable reproductive suc-
cess, but nestboxes were approximately 8 m apart in their
colony in the U.K. and only 1.5–3 m apart in our colony.
This detail may have far-reaching consequences if closer
proximity induces more interactions. It is well established
that having close neighbours and more agonistic inter-
actions results in higher testosterone titres in birds (Ball &
Wingfield 1987; Wingfield et al. 1990; Beletsky et al.
1992). Testosterone suppresses paternal care (Hegner &
Wingfield 1987; Ketterson & Nolan 1992; Ketterson et al.
1992) and could also explain higher intrapair aggression.
Data to compare aggression levels or testosterone between
colonies are not available, but dominant jackdaws do
participate more in agonistic interactions (Tamm 1977),
including in our colony (effect of rank on number of
interactions: F
1,36
Z17.65, P!0.001). We consider a tes-
tosterone-based mechanism the most parsimonious ex-
planation of our results, but further experiments, such as
manipulation of colony density and measuring testoster-
one, are required to test this hypothesis. Such experiments
may increase our understanding of the evolution of
dominance and social structures.
The question remains why males invest in acquiring
dominance when they do not benefit. Apparently, the
increase in resource access associated with being dominant
is outweighed by other factors associated with dominance.
This hypothesis suggests that the decision rule that
jackdaws use in conflicts is maladaptive when it is assumed
that birds could also ‘choose’ not to fight, at least in the
specific circumstances (e.g. high density) of our colony.
However, on a global scale, the decision rule may, on
average, still be optimal, as illustrated by the positive
association found between dominance and reproductive
success in the colony studied by Henderson & Hart (1995).
Alternatively, the decision rule that jackdaws use in
conflicts could be part of a genetically determined behav-
ioural syndrome (as in great tits, Parus major:Verbeek et al.
1996; Drent et al. 2003) in the sense that variation in
dominance reflects different behavioural syndromes.
Variation between colonies in the fitness consequences
of dominance could then contribute to the maintenance
of genetic variation in behavioural syndromes.
ANIMAL BEHAVIOUR, 68, 4
782

Acknowledgments
Peter Visser measured dominance in the 1998 season, and
many students participated in the fieldwork. Many people
reported dead and live jackdaws, and Jeroen Reneerkens
helped to catch adults. Comments from Cor Dijkstra and
Ton Groothuis improved the manuscript. S.V. was sup-
ported by the Technology Foundation Stichting Techni-
sche Wetenschappen, applied science division of
Nederlandse Organisatie voor Wetenschappelijk Onder-
zoek and the technology programme of the Ministry of
Economic Affairs.
References
Allee, W. C. 1952. Dominance and hierarchy in societies of
vertebrates. Colloques Internationaux du CNRS. XXXIV Structure et
Physiologie des Socie
´te
´s Animales, 157–182.
Andersson, S. & Ahlund, M. 1991. Hunger affects dominance
among strangers in house sparrows. Animal Behaviour,41, 895–
897.
Ball, G. F. & Wingfield, J. C. 1987. Changes in plasma levels of
luteinizing hormone and sex steroid hormones in relation to
multiple-broodedness and nest-site density in male starlings.
Physiological Zoology,60, 191–199.
Beletsky, L. D., Orians, G. H. & Wingfield, J. C. 1992. Year-to-year
patterns of circulating levels of testosterone and corticosterone in
relation to breeding density, experience, and reproductive success
of the polygynous red-winged blackbird. Hormones and Behavior,
26, 420–432.
Buchanan, K. L., Evans, M. R., Goldsmith, A. R., Bryant,
D. M. & Rowe, L. V. 2001. Testosterone influences basal
metabolic rate in male house sparrows: a new cost of dominance
signalling? Proceedings of the Royal Society of London, Series B,268,
1337–1344.
Cristol, D. A. 1992. Food deprivation influences dominance status
in dark-eyed juncos, Junco hyemalis.Animal Behaviour,43,
117–124.
Dobson, A. 1990. Survival rates and their relationship to life-history
traits in some common British birds. Current Ornithology,7,
115–146.
Drent, P. J., van Oers, K. & van Noordwijk, A. J. 2003. Realized
heritability of personalities in the great tit (Parus major). Proceed-
ings of the Royal Society of London, Series B,270, 45–51.
Drews, C. 1993. The concept and definition of dominance in animal
behaviour. Behaviour,125, 283–313.
Ellis, L. 1995. Dominance and reproductive success among non-
human animals: a cross-species comparison. Ethology and Socio-
biology,16, 257–333.
Folstad, I. & Karter, A. J. 1992. Parasites, bright males, and the
immunocompetence handicap. American Naturalist,139,
603–622.
Frank, L. G., Weldele, M. L. & Glickman, S. E. 1995. Masculini-
zation costs in hyenas. Nature,377, 584–585.
Hegner, R. E. & Wingfield, J. C. 1987. Effects of experimental
manipulation of testosterone levels on parental investment and
breeding success in male house sparrows. Auk,104, 462–469.
Henderson, I. G. & Hart, P. J. B. 1995. Dominance, food acquisition
and reproductive success in jackdaws Corvus monedula.Ornis
Scandinavica,24, 142–148.
Henderson, I. G., Hart, P. J. B. & Burke, T. 2000. Strict monogamy
in a semi-colonial passerine: the jackdaw Corvus monedula.Journal
of Avian Biology,31, 177–182.
Ketterson, E. D. & Nolan, V. 1992. Hormones and life histories: an
integrative approach. American Naturalist,140, S33–S62.
Ketterson, E. D., Nolan, V., Wolf, L. & Ziegenfus, C. 1992.
Testosterone and avian life histories: effects of experimentally
elevated testosterone on behavior and correlates of fitness in
the dark-eyed junco (Junco hyemalis). American Naturalist,140,
980–999.
Koivula, K. 1999. Ecological consequences of social dominance in
birds. Proceedings of the 22nd International Ornithological Congress,
Durban, 1580–1591.
Liebers, D. & Peter, H.-U. 1999. Intraspecific interactions in
jackdaws Corvus monedula: a field study combined with parentage
analysis. Ardea,86, 221–235.
Lorenz, K. 1931. Beitra¨ge zur Ethologie sozialer Corvidae. Journal fu
¨r
Ornithologie,80, 50–98.
Ophir, A. G. & Galef, B. G., Jr. 2003. Female Japanese quail that
‘eavesdrop’ on fighting males prefer losers to winners. Animal
Behaviour,66, 399–407.
Packer, C., Collins, D. A., Sindimdo, A. & Goodall, J. 1995.
Reproductive constraints on aggressive competition in female
baboons. Nature,373, 60–63.
Parker, G. A. 1974. Assessment strategy and the evolution of
fighting behaviour. Journal of Theoretical Biology,47, 223–243.
Piper, W. H. 1997. Social dominance in birds: early findings and
new horizons. Current Ornithology,14, 125–187.
Pusey, A. E. & Packer, C. 1997. The ecology of relationships. In:
Behavioural Ecology: An Evolutionary Approach. 4th edn (Ed. by
J. R. Krebs & N. B. Davies), pp. 254–283. Oxford: Blackwell
Scientific.
Qvarnstro
¨m, A. & Forsgren, E. 1998. Should females prefer
dominant males?. Trends in Ecology and Evolution,13, 498–501.
Rabasch, J., Browne, W., Goldstein, H., Yang, M., Plewis, I.,
Healy, M., Woodhouse, G., Draper, D., Langford, I. & Lewis, T.
2000. A User’s Guide to MLwiN. 2nd edn. London: Institute of
Education.
Ro
¨ell, A. 1978. Social behaviour of the jackdaw, Corvus monedula,in
relation to its niche. Behaviour,54, 1–124.
Rohwer, S. & Ewald, P. W. 1981. The cost of dominance and
advantage of subordination in a badge signaling system. Evolution,
35, 441–454.
Snijders, T. A. B. & Bosker, R. J. 1999. Multilevel Analysis: An
Introduction to Basic and Advanced Multilevel Modeling. London:
Sage.
Soler, M. 1988. Egg size variation in the jackdaw Corvus monedula in
Granada, Spain. Bird Study,35, 69–76.
Tamm, S. 1977. Social dominance in captive jackdaws (Corvus
monedula). Behavioural Processes,2, 293–299.
Verbeek, M. E. M., Boon, A. & Drent, P. J. 1996. Exploration,
aggressive behaviour and dominance in pair-wise confrontations
of juvenile male great tits. Behaviour,133, 945–963.
Vogel, R. 2002. Kauwen Corvus monedula in konijnenholen in 2002.
Limosa,75, 53–56.
Wechsler, B. 1988. Dominance relationships in jackdaws (Corvus
monedula). Behaviour,106, 252–264.
Wingfield, J. C., Hegner, R. E., Dufty, A. M. & Ball, G. F. 1990. The
‘challenge hypothesis’: theoretical implications for patterns of
testosterone secretion, mating systems, and breeding strategies.
American Naturalist,136, 829–846.
Zahavi, A. & Zahavi, A. 1997. The Handicap Principle: A Missing Piece
of Darwin’s Puzzle. Oxford: Oxford University Press.
VERHULST & SALOMONS: LOW FITNESS IN DOMINANT BIRDS 783