ORIGINAL ARTICLE OA
Functional
Ecology 1998
12, 607–612
000
EN
Air pollution fades the plumage of the Great Tit
T. EEVA, E. LEHIKOINEN and M. RÖNKÄ
Section of Ecology, Department of Biology, University of Turku, FIN-20014 Turku, Finland
Summary
1. Great Tits (Parus major) derive the carotenoid pigments for their yellow plumage via
the prey items in their diet. Air pollution is known to affect the abundance of many forest
insects, e.g. green caterpillars, which are an important source of food and pigments for
tits. This study investigates whether air pollutants indirectly affected the intensity of the
yellow colour in P. major plumage via the reduced access to carotenoid sources.
2. The intensity of the yellow colour in the plumage of P. major nestlings was scored
and the relative abundance of green herbivorous larvae in territories around a polluting
copper smelter in SW Finland was simultaneously measured.
3. Both the intensity of yellow colour in nestling plumage and caterpillar abundance
increased with increasing distance from the pollution source. The colour intensity
correlated significantly with the density of green herbivorous larvae in a territory.
4. Parus major nestlings were significantly heavier at distant sites than close to the
pollution source which suggests that the future survival probability of pale nestlings
may be lowered.
5. Young birds, after their first moult, were studied for the relationships between condition, size and plumage colour by the means of ptilochronology. The plumage colour
intensity did not correlate with the size corrected width of the growth bars in fifth rectrix
(condition at moult) but was correlated positively with the length of the rectrix (size).
6. The implications of colour change for survival and mate choice are discussed.
Key-words: Carotenoids, caterpillar food, heavy metal pollution, Parus major, plumage colour
Functional Ecology (1998) 12, 607–612
Introduction
© 1998 British
Ecological Society
In addition to direct toxic effects, air pollution can
indirectly affect birds through, for example, habitat
changes (Morrison 1986), increased amount of parasites (Eeva, Lehikoinen & Nurmi 1994) or reduced
amount of suitable food (Graveland 1990; Hörnfeld
& Nyholm 1996). Although xenobiotics most
severely affect the early stages of life cycle (e.g.
Scheuhammer 1987), indirect effects may be present
in adulthood as well. In this paper we describe a
mechanism of how environmental contaminants may
indirectly affect birds by changing their plumage
colour, and discuss the consequences of colour
change for individual’s fitness.
Several studies have shown that female birds show
preference for colourful males, which consequently
are of high adaptive value (e.g. Hill 1990; Palokangas
et al. 1994). Bright plumage colour has been shown to
signal a good nutritional condition (Hill &
Montgomerie 1994), high hunting activity
(Palokangas et al. 1994) or low parasite load (Dufva
& Allander 1995; Sundberg 1995). In general, birds
cannot synthesize the necessary carotenoid pigments
themselves but need to obtain them from the diet
(Brush 1978). The access to the carotenoid sources
may be affected by air pollutants if they reduce the
amount of carotenoid-rich food items. Anthropogenically limited access to the carotenoid sources
might thus affect individual’s fitness later in life, e.g.
via mate choice or social dominance.
Great Tits (Parus major L.) feed their nestlings a
great number of green caterpillars and sawfly larvae
(van Balen 1973; Perrins 1991) and these are shown to
be an important source of carotenoids for growing and
moulting birds. Slagsvold & Lifjeld (1985) showed
that the more the P. major nestlings ate green lepidopteran larvae, the yellower their plumage became.
A few studies have demonstrated decreased numbers
of caterpillars around the point source discharges of
air pollutants (Heliövaara & Väisänen 1988;
Ruohomäki et al. 1996; Eeva, Lehikoinen &
Pohjalainen 1997). As a result of some preliminary
observations in the field we considered it possible that
paler P. major nestlings were produced in a polluted
area than in natural environments. In summer 1996 we
simultaneously measured the intensity of yellow
colour in the plumage of P. major nestlings, the
nestling condition (body mass) and the relative
amount of caterpillars in the territories at different
607
608
T. Eeva et al.
distances from an air pollution source, a copper
smelter. Our focus is on the insect larvae in Scots Pine
(Pinus sylvestris) foliage which in our study area is
the predominant foraging site of P. major during the
nestling period (Eeva et al. 1997). Birds were also
trapped in the autumn to see whether the colour intensity still reflected the condition of an individual during the moult and to determine how the colour
intensity possibly changes during the first moult.
Materials and methods
© 1998 British
Ecological Society,
Functional Ecology,
12, 607–612
The data were collected in the surroundings of the
town Harjavalta (61°20'N, 22°10'E), SW Finland. The
main source of local air pollutants is a factory complex producing copper, nickel and fertilizers in the
centre of Harjavalta. Sulphuric oxides and heavy metals (especially Cu, Zn, Ni and Pb) are common pollutants in the area (Kubin 1990; Jussila & Jormalainen
1991). During the 1940s, sulphur dioxide produced in
the process was released into the atmosphere, which
led to a decline of forest in the vicinity of the factory.
Later on, most SO2 was utilized to produce sulphuric
acid resulting in a reduction in emissions. During
1995 about 3300 t SO2 was emitted into the air. The
pollution gradient around the factory has been confirmed by analyses of the SO2 content of pine needles
and observed reduction in the species number of bark
lichen (Laaksovirta & Silvola 1975), by analyses of
heavy metals with the moss bag method (Hynninen
1986), and by analyses of moss and rain water samples from the forest floor (Jussila & Jormalainen
1991). In addition, heavy metals content in P. major
nestling faeces decreases exponentially with increasing distance from the factory complex (Eeva &
Lehikoinen 1996).
Ten study sites, each with 45–55 nestboxes, were
originally established during 1991–95 along an air
pollution gradient in three main directions (SW, SE
and NW) away from the smelter complex up to the
distance of 6 km (Fig. 1). In order to take account of
the possible habitat effects on plumage colour, relative
proportions of the three main tree species (Scots Pine
Pinus sylvestris, birch Betula sp. and Spruce Picea
abies) in territories were estimated: four abundance
classes were used: 0 = not present, 1 = sparse, 2 =
moderate, 3 = dominant. For a more detailed description of the study area and the pollution gradient see
Eeva & Lehikoinen (1995).
During the breeding period, nestboxes were visited
at least once a week to determine the laying and hatching dates and to measure the nestlings. At the age of
15 (± 1) days, nestlings were weighed, and scored for
the intensity of yellow colour in their plumage
(flanks). In autumn, after their first moult, young birds
were trapped with nets from feeders to determine
whether the intensity of plumage coloration still
reflected the condition of an individual during the
moult. The condition of birds during moult was
Fig. 1. Location of the study area and 10 study sites around
the copper smelter (in the middle).
measured by the means of ptilochronology (Grubb
1989). The width of the daily growth bars in feathers
is considered to be a good indicator of an individual’s
nutritional condition (Grubb 1995), and the width of
the growth bars in the right fifth rectrix was used,
which was grown after the postjuvenile moult. The
width of the growth bars was measured using the
method of Grubb (1989). The mean of nine bars was
calculated and the means were standardized to bird
size by the regression between the length of the rectrix
and mean bar width. It was not possible to recapture
the individuals scored as nestlings in sufficient numbers to determine how the plumage colour possibly
changed during the moult.
The colour intensity was measured using a six-class
colour scale (1 = pale, 6 = bright), which was printed
on a paper with grey background. The scale was preserved in a case protected from light. Direct sunlight
was avoided when scoring the nestlings. During the
summer, all colour determinations were made by the
same observer. In the autumn, colour determinations
were made by four persons and the repeatability of the
colour measurements within and among observers
were tested with museum skins. Each of four persons
scored 30 individuals twice. The samples were mixed
between the determinations. The repeatability (Harper
1994) within persons was high (from 76% to 85%) for
three persons and low (36%) for one person, who
scored only five birds in the field. The repeatability
among the four persons was 61%. The nestlings could
not be sexed but the colour differences between sexes
were probably small since the colour variation within
broods was small: in 35% of the broods all the
nestlings were scored to the same colour class (see
also Slagsvold & Lifjeld 1985). In the postjuvenile
plumage males were brighter than females.
Nestling mass was used as an indicator of nestling
condition. Since the age of the nestlings varied from
609
Air pollution
fades plumage of
Great Tit
14 to 16 days, age-corrected mass residuals were
calculated to make values comparable. A logistic
growth curve was fitted to the whole data from 1996
following Ricklefs (1983): M(x) = A/{1 + [(A – I)/I]*
exp (– K * x)}, where M(x) = mass at age x, A =
asymptotic mass, I = initial mass, K = rate constant. A
fixed initial mass of 1·8 g was used. Logistic curves
were fitted by non-linear least squares estimation in
the data analysis program AXUM (TriMetrix Inc.
1994). For the best fit curve A = 16·0, K = 0·40 and
r2 = 0·45. Mass residuals (observed mass – estimated
mass) are given as percentage deviations from the
expected mass.
The amount of herbivorous larvae in tree foliage
was measured in each territory by the frass-fall
method (Southwood 1978). The amount of falling
frass was used as a relative measure of caterpillar
density in the tree canopy. Round plastic funnels
(34-cm diameter) were mounted with wire at a height
of 1 m to trunks. Similar sized trees (diameter of
15–25 cm at 1-m height) were selected when possible.
Tree length, trunk diameter and the height of green
foliage were measured. Sampling took place between
11 and 26 June and coincided with the late nestling
period of P. major: the mean fledging date was 22
June. Funnels were placed in triangles within a radius
of 15 m from nestboxes. Under the funnel there was a
container where the frass accumulated during the collection period. Contents were dried and stored in
paper bags until the frass was separated from the litter
and weighed. As the ‘caterpillar index’ the frass mass
per collecting time (h), per height of green tree foliage
(m) was used. Proportioning the frass mass to the
height of tree foliage was used to make this measure
reflect the density rather than the biomass of larvae in
a tree. Hoop-netting of Scots Pine foliage at different
sites was used to determine the species composition
among herbivorous larvae.
Statistical tests were made using SAS statistical
system (SAS Institute 1989). For regressions and
analysis of covariance (ANCOVA) the GLM procedure
(with SS3) was used. The normality of residuals was
checked for each model using UNIVARIATE procedure.
Spearman correlation coefficients (rs) were used for
the analyses of colour scores of postjuvenile plumage
because of non-normal distributions. All means are
given with their standard deviations (SD).
Results
Fig. 2. The mean intensity of nestling plumage colour in P. major broods along the
pollution gradient (y = 0·26x + 2·67, df = 46, r2 = 0·43, F = 34·3, P < 0·0001). Each dot
is a brood average.
Fig. 3. The logarithmic caterpillar index (frass mass (mg)/time (h)/height (m) of tree
foliage) at P. major territories along the pollution gradient during the late nestling
period (y = 0·20x – 6·62, df = 31, r2 = 0·19, F = 7·5, P = 0·010).
The plumage colour of P. major nestlings ranged from
score 2 to score 5 (mean = 3·55, SD = 0·76, n = 328).
The intensity of yellow colour in plumage increased
with increasing distance from the pollution source
(Fig. 2). The density of caterpillars during the late
nestling period was significantly lower close to the
pollution source than further away (Fig. 3). Two insect
species dominated in hoop-net samples (84 larvae)
collected during the same period: a sawfly Neodiprion
sertifer (61%) and a moth Panolis flammea (37%).
The mean colour intensity in P. major broods correlated significantly with the density of caterpillars in a
canopy of Scots Pine (Fig. 4). Parus major nestlings
were significantly lighter in mass near the pollution
source than further away from it (Fig. 5). Since both
body mass and colour intensity were correlated with
distance from the pollution source, there was also a
correlation between body mass and colour intensity
(r = 0·28, n = 48, P = 0·049). To study the relationship
between body mass and colour intensity a regression
model was run where nestling mass and distance were
used to explain the variation in colour intensity. Body
mass did not explain the variation in colour intensity
(df = 1, F = 0·11, P = 0·75) whereas the distance did
(df = 1, F = 29·44, P < 0·0001). This means that
although nestlings were brighter and heavier further
610
T. Eeva et al.
from the pollution source, body mass alone cannot
explain the differences in plumage colour.
Possible habitat effects on the colour intensity were
studied using an ANCOVA model in which colour intensity was used as the dependent variable, proportions
of three main tree species as the independent variables
and distance to the pollution source as a covariate.
Distance to the pollution source was not correlated
with the abundance of any tree species. The relative
abundances of the three tree species did not explain
the intensity of yellow colour in the plumage of
nestlings (Scots Pine, F3,37 = 0·77, P = 0·52; birch,
F3,37 = 1·21, P = 0·32; Spruce, F3,37 = 0·69, P = 0·56),
but the effect of the covariate (distance) was significant (F1,37 = 18·32, P < 0·0001).
The mean colour score in the postjuvenile plumage
of P. major was 4·17 (SD = 0·68, n = 202) for females
Fig. 4. The dependence of P. major nestling plumage colour intensity on the logarithmic caterpillar index (frass mass (mg)/time (h)/height (m) of tree foliage) at territories
(y = 0·46x + 6·16, df = 30, r2 = 0·32, F = 14·1, P = 0·0007).
Fig. 5. The residual masses of P. major nestlings (see methods) along the pollution
gradient (y = 3·33x – 6·42, df = 46, r2 = 0·20, F = 11·7, P = 0·0013).
and 4·43 (SD = 0·83, n = 200) for males. The size-corrected widths of the growth bars in the fifth rectrix did
not correlate with the plumage colour intensity in
either sex (males, rs = – 0·27, P = 0·13, n = 33;
females, rs = 0·27, P = 0·19, n = 25), and plumage
colour did not reflect the nutritional condition during
the moult. However, the length of the rectrix correlated positively with the plumage colour intensity in
both sexes (males, rs = 0·35, P = 0·045; females,
rs = 0·44, P = 0·027; Fig. 6). Larger birds were therefore yellower than smaller ones.
Discussion
It was found that P. major nestlings were paler near the
pollution source, and that the plumage colour intensity
correlated positively with the abundance of green herbivorous larvae. The amount of carotenoids ingested
and the intensity of plumage colour have been shown
to correlate more or less linearly – under some proximal control which causes, for example, the differences
between sexes (Slagsvold & Lifjeld 1985; Hill 1992;
Hill et al. 1994). In an earlier study it was also found
that the proportion of green larvae in nestling diet was
smaller in the vicinity of the pollution source than further away (Eeva et al. 1997). This supports the idea
that the observed correlation between the abundance of
larvae and plumage colour is causal. The correlation
we found is the first, although indirect, evidence of the
relationship between anthropogenically limited availability of carotenoid-rich prey items and plumage
colour in birds.
However, pollutants may also directly affect
carotenoid metabolism and/or transportation to feathers. For example, the pale plumage might have
occurred if the movement of carotenoids is blocked by
contaminants between ingestion and eventual deposition in feathers (see Hill et al. 1994). Several studies
have shown that polyhalogenated hydrocarbons can
affect the concentrations of carotenoid derivatives in
organs (Peakall 1992). Unfortunately, there is little
information available on the effects of heavy metals
on transportation of carotenoids, and this alternative
cannot be excluded in our study.
Slagsvold & Lifjeld (1985) noted that P. major
nestlings reared in a deciduous forest were yellower
than those reared in coniferous woodlands. This could
have been a confounding factor in our study, since
there are also habitat gradients in our study area parallel with the pollution gradient, e.g. an increase of the
proportion of Spruce towards unpolluted sites (see
Eeva et al. 1997). However, this was not the case in P.
major territories. In no case was the abundance of any
of the three main tree species correlated with the distance from the pollution source, nor did the proportions of any tree species correlate with the plumage
colour intensity.
What consequences might the pale plumage have
for future fitness of a nestling? Sexual selection is
611
Air pollution
fades plumage of
Great Tit
regarded as one of the most important factors in the
evolution of bright colours in birds (Savalli 1995). The
width of the black central breast stripe is an important
cue for social status, sex recognition and sexual selection in P. major (e.g. Järvi & Bakken 1984; Norris
1990; Slagsvold 1993) but there are no studies concerning the effects of yellowness on mate choice. Recent
studies suggest that bright colour may act as a cue for
low haematozoan parasite load in birds (Hamilton &
Zuk 1982; Dufva & Allander 1995; Sundberg 1995).
We found that the mass of nestlings was correlated with
the plumage colour and it is probable that both mass
and colour changes have a common reason: the scarcity
of caterpillars in the polluted area. Nestling mass has
been shown to correlate with the future survival of an
individual (e.g. Perrins 1965; Tinbergen & Boerlijst
1990; Verhulst, Van Balen & Tinbergen 1995).
Yellowness, as a condition-dependent character, may
therefore also be an important cue in sexual selection.
Whether the plumage colour differences of nestlings
persist after their first moult was not established.
In addition to mate choice the plumage colour might
affect male–male competition, social dominance in
winter flocks, or the ability to avoid predation via
crypticism. No correlation was found between the
width of the growth bars in tail feather (i.e. condition
during the moult) and plumage colour intensity after
postjuvenile moult. Instead, the colour intensity correlated with size so that, for both sexes, birds with long
tails were brighter than those with short tails. Garnett
(1981) showed that in P. major there is already a relationship between body size and dominance of an individual in flocks two weeks after fledging (but see
Lambrechts & Dhondt 1986). A pale plumage may
therefore predict reduced winter survival.
In addition to the effect on plumage colour, the deficiency of carotenoids can have direct effects on
Fig. 6. The median lengths (mm) of the fifth rectrices of P. major males (n = 33) and
females (n = 25) in three classes of plumage colour intensity. Boxes include 50% of
the data and bars show the highest and lowest value within the step multiplied by 1·5.
The width of the box denotes the proportional number of observations. The deviant
data points for two females are indicated with symbol ×.
health. The deficiency of vitamin A (retinoids) is associated with a diversity of anomalies, including reproductive impairment, embryonic mortality, growth
retardation and bone deformities (Peakall 1992). Since
animals are unable to synthesize retinoids they need to
get them via carotenoids in food. A connection
between the lowered carotenoid levels and increased
embryonal mortality have been demonstrated in wild
salmon (e.g. Pettersson & Lignell 1995). These
changes have been suggested to involve environmental
toxins and/or nutritional disturbances, but the mechanisms are not yet completely understood (e.g.
Börjeson, Förlin & Norrgren 1995). It is therefore possible that lower growth rates of P. major nestlings in
polluted area may also reflect a vitamin-deficient diet.
Acknowledgements
We thank Simo Veistola, Jürgen Wiehn, David Currie
and two anonymous referees for reading and
commenting on the earlier versions of the manuscript.
We are also grateful to Toni Laaksonen, Jorma Nurmi
and Julia Bojarinova, who helped us with field work.
This study was financially supported by the Academy
of Finland and by the Laboratory of Animal Ecology
in Turku University.
References
Börjeson, H., Förlin, L. & Norrgren, L. (1995) Investigation
of antioxidants and pro-oxidants in salmon affected by the
M74 syndrome. Report from the Second Workshop on
Reproduction Disturbances in Fish, 20–23 November
1995, Stockholm, Sweden (eds B.-E. Bengtsson, C. Hill
& S. Nellbring), pp. 95–96. Swedish Environmental
Protection Agency, Stockholm.
Brush, A.H. (1978) Avian pigmentation. Chemical Zoology
(ed. A.H. Brush), pp. 141–164. Academic Press, New York.
Dufva, R. & Allander, K. (1995) Intraspecific variation in
plumage coloration reflects immune response in Great Tit
(Parus major) males. Functional Ecology 9, 785–789.
Eeva, T. & Lehikoinen, E. (1995) Egg shell quality, clutch
size and hatching success of the great tit (Parus major)
and the pied flycatcher (Ficedula hypoleuca) in an air pollution gradient. Oecologia 102, 312–323.
Eeva, T. & Lehikoinen, E. (1996) Growth and mortality of
nestling Great Tits (Parus major) and Pied Flycatchers
(Ficedula hypoleuca) in a heavy metal pollution gradient.
Oecologia 108, 631–639.
Eeva, T., Lehikoinen, E. & Nurmi, J. (1994) Effects of
ectoparasites on breeding success of great tits (Parus
major) and pied flycatchers (Ficedula hypoleuca) in an
air pollution gradient. Canadian Journal of Zoology 72,
624–635.
Eeva, T., Lehikoinen, E. & Pohjalainen, T. (1997) Pollutionrelated variation in food supply and breeding success in
two hole-nesting passerines. Ecology 78, 1120–1131.
Garnett, M.C. (1981) Body size, its heritability and influence on juvenile survival among great tits, Parus major.
Ibis 123, 31–41.
Graveland, J. (1990) Effects of acid precipitation on reproduction in birds. Experientia (Basel) 46, 962–970.
Grubb, T.C.J. (1989) Ptilochronology: feather growth bars
as indicators of nutritional status. The Auk 106,
314–320.
612
T. Eeva et al.
© 1998 British
Ecological Society,
Functional Ecology,
12, 607–612
Grubb, T.C.J. (1995) Ptilochronology. A review and
prospects. Current Ornithology (ed. D. M. Power), pp.
89–114. Plenum Press, New York.
Hamilton, W.D. & Zuk, M. (1982) Heritable true fitness and
bright birds: a role for parasites? Science 218, 384–387.
Harper, D.G.C. (1994) Some comments on the repeatability
of measurements. Ringing and Migration 15, 84–90.
Heliövaara, K. & Väisänen, R. (1988) Interactions among
herbivores in three polluted pine stands. Silva Fennica 22,
283–292.
Hill, G.E. (1990) Female house finches prefer colourful
males: sexual selection for a condition-dependent trait.
Animal Behaviour 40, 563–572.
Hill, G.E. (1992) Proximate basis of variation in carotenoid
pigmentation in male house finches. The Auk 109, 1–12.
Hill, G.E. & Montgomerie, R. (1994) Plumage colour
signals nutritional condition in the house finch.
Proceedings of the Royal Society of London Series B –
Biological Sciences 258, 47–52.
Hill, G.E., Montgomerie, R., Inouye, C.Y. & Dale, J. (1994)
Influence of dietary carotenoids on plasma and plumage
colour in the house finch: intra- and intersexual variation.
Functional Ecology 8, 343–350.
Hörnfeldt, B. & Nyholm, N.E.I. (1996) Breeding performance of Tengmalm’s owl in a heavy metal pollution
gradient. Journal of Applied Ecology 33, 377–386.
Hynninen, V. (1986) Monitoring of airborne metal pollution
with moss bags near an industrial source at Harjavalta,
southwest Finland. Annales Botanici Fennici 23, 83–90.
Järvi, T. & Bakken, M. (1984) The function of the variation
in the breast stripe of the great tit (Parus major). Animal
Behaviour 32, 590–596.
Jussila, I. & Jormalainen, V. (1991) Spreading of heavy
metals and some other air pollutants at Pori-Harjavalta
district in SW-Finland. SYKEsarja B 4, 1–58.
Kubin, E. (1990) A survey of element concentrations in the
epiphytic lichen Hypogymnia physodes in Finland in
1985–86. Acidification in Finland (eds P. Kauppi, P.
Anttila & K. Kenttämies), pp. 421–446. Springer-Verlag,
Berlin.
Laaksovirta, K. & Silvola, J. (1975) Effect of air pollution
by copper, sulphuric acid and fertilizer factories on plants
at Harjavalta, W. Finland. Annales Botanici Fennici 12,
81–88.
Lambrechts, M. & Dhondt, A.A. (1986) Male quality, reproduction, and survival in the great tit (Parus major).
Behavioral Ecology and Sociobiology 19, 57–63.
Morrison, M.L. (1986) Bird populations as indicators of
environmental change. Current Ornithology 3, 429–451.
Norris, K.J. (1990) Female choice and the evolution of the
conspicuous plumage coloration of monogamous male
great tits. Behavioral Ecology and Sociobiology 26,
129–138.
Palokangas, P., Korpimäki, E., Hakkarainen, H., Huhta, E.,
Tolonen, P. & Alatalo, R.V. (1994) Female kestrels gain
reproductive success by choosing brightly ornamented
males. Animal Behaviour 47, 443–448.
Peakall, D. (1992) Animal Biomarkers as Pollution
Indicators. Chapman & Hall, London.
Perrins, C.M. (1965) Population fluctuations and clutch-size
in the Great Tit, Parus major L. Journal of Animal
Ecology 34, 601–647.
Perrins, C.M. (1991) Tits and their caterpillar food supply.
Ibis 133, 49–54.
Pettersson, A. & Lignell, Å. (1995) Decreased astaxanthin
levels in the Baltic salmon and the M74 syndrome. Report
from the Second Workshop on Reproduction
Disturbances in Fish, 20–23 November 1995, Stockholm,
Sweden (eds B.-E. Bengtsson, C. Hill & S. Nellbring), pp.
28–29. Swedish Environmental Protection Agency,
Stockholm.
Ricklefs, R.E. (1983) Avian postnatal development. Avian
Biology VII (eds D. S. Farner, J. R. King & K. C. Parkes),
pp. 1–83. Academic Press, London.
Ruohomäki, K., Kaitaniemi, P., Kozlov, M., Tammaru, T. &
Haukioja, E. (1996) Density and performance of Epirrita
autumnata (Lepidoptera: Geometridae) along three air
pollution gradients in northern Europe. Journal of
Applied Ecology 33, 773–785.
SAS Institute Inc. (1989) SAS/STAT User’s Guide, 4th edn,
Vols 1–2. SAS Institute Inc., Cary, NC.
Savalli, U.M. (1995) The evolution of bird coloration and
plumage elaboration. A review of hypotheses. Current
Ornithology, Vol. 12, pp. 141–190. Plenum Press, New
York.
Scheuhammer, A.M. (1987) The chronic toxicity of
aluminium, cadmium, mercury, and lead in birds: A
review. Environmental Pollution 46, 263–295.
Slagsvold, T. (1993) Sex recognition and breast stripe size in
great tits. Ardea 81, 35–41.
Slagsvold, T. & Lifjeld, J.T. (1985) Variation in plumage
colour of the Great Tit Parus major in relation to habitat,
season and food. Journal of Zoology (London) 206,
321–328.
Southwood, T.R.E. (1978) Ecological Methods. With
Particular Reference to the Study of Insect Populations.
Chapman and Hall, London.
Sundberg, J. (1995) Parasites, plumage coloration and reproductive success in the yellowhammer, Emberiza citrinella. Oikos 74, 331–339.
Tinbergen, J.M. & Boerlijst, M.C. (1990) Nestling weight
and survival in individual great tits (Parus major).
Journal of Animal Ecology 59, 1113–1127.
TriMetrix Inc. (1994) Axum. Technical Graphics And Data
Analysis. User’s Guide, 1st edn. TriMetrix, Inc., Seattle.
Van Balen, J.H. (1973) A comparative study of the breeding
ecology of the Great Tit Parus major in different habitats.
Ardea 61, 1–93.
Verhulst, S., Van Balen, J.H. & Tinbergen, J.M. (1995)
Seasonal decline in reproductive success in the Great Tit:
variation in time or quality. Ecology 76, 2392–2403.
Received 11 April 1997; revised 6 October 1997; accepted
12 November 1997