Copyright © NISC Pty Ltd
Ostrich 2004, 75(4): 211–216
Printed in South Africa — All rights reserved
OSTRICH
ISSN 0030–6525
Breeding biology of the Cape Bulbul Pycnonotus capensis: a 40-year
comparison
Oliver Krüger
Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, United Kingdom
e-mail: [email protected]
I examined the breeding biology of the Cape Bulbul Pycnonotus capensis at a study site in Port Elizabeth for a comparison
with the results obtained by Richard Liversidge from a study made 40 years ago. There were no differences in methods, so
results were directly comparable. No differences were found between the two study periods in the timing of breeding, clutch
size, number of fledglings per nesting attempt or successful nesting attempt, indicating no major shift in the life history strategy or breeding success of the population. During my study period, there was a strong calendar effect on breeding success:
early nests were more likely to be successful than late ones. I found strong differences between bush species with regard to
predation risk and brood parasitism risk by the Jacobin Cuckoo Clamator jacobinus. These differences emphasise the complex interaction between the timing of breeding, risk of predation and risk of brood parasitism, which influences decisions
about when to start nesting and nest site selection in the Cape Bulbul.
Introduction
The breeding biology of animals is influenced by a great
number of factors, such as food supply, predation risk and
nest site selection, amongst others (Newton 1998). Changes
in the breeding biology or even anatomy can be related to
changes in the abiotic (Hughes 2000, Yom-Tov 2001,
Gjerdrum et al. 2003) or biotic environment (Boveng et al.
1998). In order for a change to be detectable, long-term
studies or long-time comparisons are necessary (Sutherland
1996, Newton 1998, Thiollay 2000). However, long-term
studies are comparatively rarely done and long-time comparisons are plagued by differences in the methodology or
changes of the study site that render a comparison often
meaningless. In recent years, notably the timing of breeding
and the breeding success of many bird species have been
related especially to climatic factors, such as rising global
temperatures (McCleery and Perrins 1998, Coppack and
Both 2002, Gjerdrum et al. 2003). Most of these studies,
however, come from the Northern hemisphere and examples from southern Africa are few and far between (Gargett
1990, Ligon and Ligon 1990, are two examples), or deal with
population dynamics (Thiollay 1998 and 2000, KalejtaSummers et al. 2001).
This study took advantage of the fact that methods and
study site were almost identical despite the 40-year gap
between the two study periods. This made it possible to test
whether the breeding biology and breeding success of the
Cape Bulbul has changed during the last 40 years. The
Cape Bulbul, together with its close sister species, Darkcapped Bulbul Pycnonotus tricolor and African Red-eyed
bulbul P. nigricans, are amongst the most common bird
species in southern Africa but have been comparatively little
studied, except for hybridisation between the three species
(Lloyd et al. 1996 and 1997). The Cape Bulbul offers the
additional complexity of being the host of a brood parasite,
the Jacobin Cuckoo Clamator jacobinus (Liversidge 1970a
and 1970b). Hence its breeding biology might also be influenced by host-parasite co-evolutionary processes (Davies
1999). In a second part, additional information on factors
influencing reproductive success in the Cape Bulbul, such
as nest site selection and the interplay between predation
risk and brood parasitism risk, is provided.
Methods
Study site
The present study was conducted in the University of Port
Elizabeth nature reserve (34°01’S, 25°38’E), which is part of
the campus area of the university. The reserve consists of
coastal sand dunes interspersed with rocky outcrops and
belongs to the Cape St Francis coastal fynbos and thicket
mosaic habitat (Cowling and Heijnis 2001). The area
receives around 700mm of aseasonal rain annually.
Dominant shrub species in the reserve include Sea Guarri
Euclea racemosa, Brown Gonna Passerina filiformis, Dune
Olive Olea exasperata, White Bristle bush Metalasia muricata and Dune Crow-berry Rhus crenata.
The present study site is located approximately 3km
from the original study site Richard Liversidge used, the
Humewood golf course area. The one significant difference
between the two sites is that the alien invasive rooikrans
Acacia cyclops, the most common bush and nest bush
species in Liversidge’s study (Liversidge 1966, 1970a,
1970b), has been largely exterminated from the university
nature reserve, as part of the Working for Water programme.
To get information on abiotic factors, I obtained monthly
minimum and maximum temperatures as well as monthly
rainfall data for the seven years of study, from the South
African Weather Service. The monthly data were subse-
212
Field work
During the start of the 2001 field season, the author was
taught how to locate bulbul nests by Richard Liversidge, so
differences in field methods between the two study periods
are negligible. Field work was conducted mainly from sunrise to about 10h00 and from 15h00 to sunset. Territorial
males were not colour-ringed but mapped while singing from
prominent perches, while nests were found by either watching females with nest material or by following the Jacobin
Cuckoo which parasitizes the Cape Bulbul. The latter technique had been used extensively by Liversidge (1970a and
1970b) in his study. Once a nest was found it was checked
on a daily or two-day interval basis to determine clutch size,
incubation period, hatching success, nestling period and
fledging success. For those few nests where no egg was
laid, I used the day the nest was finished plus two days to
assign the time of the nest attempt. The additional two days
were based on both Liversidge (1970a) and my own observations that commonly two days pass between a finished
nest and the first egg being laid. In 2002 and 2003 all successfully fledged chicks were ringed with a unique combination of SAFRING metal rings and darvic colour plastic rings.
Sometimes I was able to identify the predator species that
preyed on either eggs or chicks, because the common eggeater snake (Dasypeltis scabra) regurgitates the egg shell in
the vicinity of the nest. Striped mice (Rhabdomys pumilio)
can be often identified as the predator because they just
make a hole into the egg and often leave droppings in the
nest. Predation by either small grey mongoose (Galerella
pulverulenta) or yellow mongoose (Cynictis penicillata) normally leaves the nest partially or totally dismantled.
Statistical analysis
Differences between the two study periods were assessed
using non-parametric tests only (χ2 and Mann-Whitney
tests), because of small sample size (four vs three years).
Time trends were assessed using Spearman’s non-parametric rank correlation co-efficient. Whenever parametric
tests were used the data were checked for deviance from
normality. With regard to the correlation between time of the
year and reproduction, all breeding attempts were treated as
independent, but the result did not change qualitatively
when replacement clutches were excluded from the analysis. I calculated the length of the breeding season using the
MacArthur index (Ricklefs 1966). Daily nest mortality rate
and nesting success were calculated using the Mayfield
index (Mayfield 1975). All statistical analyses were performed in SPSS for Windows and P = 0.05 (two-tailed) was
used as a significance threshold throughout.
Results
Comparison between the two study periods
With regard to incubation and nestling periods, I found similar values compared to those observed 40 years ago
(Liversidge 1970a). Incubation time ranged between 11 and
14 days with a median incubation period of 12 days (n = 13,
sd = 1.2). Nestling period was recorded to range between 11
and 15 days with a median nestling period of 13 days (n =
15, sd = 1.7).
There was no significant shift in the distribution of nesting attempts across the season between the two study periods (Figure 1). The peak month for nesting attempts was
November 40 years ago, while it was October during my
study period. However, there was no overall difference
between the two study periods with regard to nesting time
(χ2 = 5.649, df = 4, P = 0.168). Combining both study periods, the length of the Cape Bulbul breeding season can be
estimated at 104 days, using the MacArthur index. Daily
nest mortality rates and nesting success were 0.075 and
0.113 for 2001, 0.022 and 0.536 for 2002 and 0.041 and
0.314 for 2003. The three-year mean daily nest mortality rate
and nesting success were 0.045 and 0.275, respectively.
Hence, around 27.5% of nests where at least one egg was
laid can be expected to be successful.
Comparing the two study periods with regard to their
breeding biology parameters reveals mostly similarities
(Table 1). Overall, from 233 eggs laid in Richard Liversidge’s
study, 63 chicks fledged (27.0%) whereas in my study period, out of 213 eggs recorded, 70 chicks fledged (32.9%).
Hatching success in my study period was 83.5%. Predation
rates were also similar between the two study periods: 60%
of nests were preyed on 40 years ago while 59% were
preyed on in my study period. Predation on eggs accounted
for 85.5% of predation events in my study period while predation on chicks was much rarer (14.5%). These figures
were more biased towards predation on eggs compared to
those reported by Liversidge (predation on eggs: 71.9% of
PERCENTAGE OF NESTS (%)
quently condensed into an annual figure and a figure for the
months when breeding occurred (August to December).
There were no significant differences in weather variables
between the two study periods, with the exception that maximum mean annual temperature was higher in my study
period compared to Liversidge’s study (22.2°C vs 22.5°C, Z
= 2.223, P = 0.026). Hence, there has been no overall
change in the abiotic environment at the study site between
the two periods.
Krüger
45
40
35
1960-1962
2001-2003
30
25
20
15
10
5
Aug.
Sep.
Oct.
Nov.
Dec.
Figure 1: Comparison of the percentage of nests initiated in each
month from August to December between the two study periods
Ostrich 2004, 75: 211–216
213
Table 1: Comparison of the breeding success of the Cape Bulbul between 1959 to 1962 and 2001 to 2003. Data for the years 1959 to 1962
have been taken from Liversidge (1966) and Liversidge (1970b) and modified for comparison. The columns ‘attempts per pair’ and ‘fledged
per pair’ refer only to those pairs where there was high certainty that all nesting attempts were found during a given year. To calculate the parasitism rate, only nests with at least one bulbul egg laid were included
Pairs
1959
1960
1961
1962
2001
2002
2003
13
19
13
15
31
22
31
Nests
found
22
39
21
33
41
22
38
Attempts
/pair
1.69
2.05
1.62
2.20
1.86
1.00
1.81
Clutch size
(se)
2.40
2.71
2.50
2.62 (0.11)
2.25 (0.10)
2.44 (0.09)
predation events and 28.1% of predation events on chicks).
Of those cases where the predator species could be identified, the great majority of predation could be assigned to
striped mice.
There were no significant differences in number of nesting attempts (Z = 0.707, P = 0.480), clutch size (Z = 0.655,
P = 0.513), number of fledglings per nesting attempt (Z =
1.414, P = 0.157) and number of fledglings per pair (Z =
0.707, P = 0.480). There was a weak trend towards lower
parasitism rates in my study period (Z = 1.768, P = 0.077).
There was no large difference in the number of parasitic
eggs per parasitised nest, being 1.22 in Liversidge’s study
compared to 1.11 in my study period. There were also no
significant time trends, combining both study periods, with
the exception that parasitism rate decreased significantly
over time (rs = –0.857, n = 7, P = 0.014). Looking at correlations between breeding parameters and climate data
revealed only three significant associations out of a total of
30 tested (nests per pair, clutch size, fledged young per
nesting attempt, fledged young per pair and parasitism rate
were correlated with the six weather variables); nests per
pair increased with annual mean minimum temperature (rs =
0.847, n = 7, P = 0.016), fledged young per nest increased
with annual rainfall levels (rs = 0.857, n = 7, P = 0.014) and
parasitism rate decreased with annual mean maximum temperature (rs = –0.898, n = 7, P = 0.006). However, adjusting
the significance levels for multiple testing using a sequential
Bonferroni correction (Rice 1989) would make even these
three correlations non-significant.
Further aspects of Cape Bulbul breeding biology
Although there was no significant forward shift of the breeding season between the two study periods, an early start still
seems to be advantageous because there was a significant
negative correlation between reproduction and time (Figure
2, r = –0.286, df = 99, P = 0.004). This decline of reproduction over the season was not due to a decrease in clutch
size (r = –0.139, df = 84, P = 0.203). As can be seen from
the regression line in Figure 2, the key difference is that late
nests have an increasingly higher probability of failing, which
is due to predation, rather than fledging fewer chicks.
Indeed, comparing the mean date of the first egg between
successful and failed nests reveals a highly significant difference (F1,95 = 14.471, P < 0.001): successful nests had the
Fledged
/attempt
0.68
0.59
0.09
0.27
0.51
1.00
1.05
REPRODUCTION (juv/nest)
Year
Fledged/succ
attempt (se)
1.91 (0.25)
2.00 (0.23)
2.00 (0.15)
Fledged
/pair
1.15
1.21
0.15
0.60
0.95
1.00
1.90
% Nests
parasitised
72.7
33.3
38.1
12.1
7.7
30.0
5.9
3.0
2.0
y = -0.013x + 1.455
R2= 0.082
1.0
-10 0
10
20 30 40 50 60 70 80
TIME (01 Sep. = 1)
90
Figure 2: Relationship between nest initiation date and nest success (number of young fledged), combining all data for the three
years
first egg laid on average on 9 October whereas failed nests
had the first egg laid on average on 27 October. One additional contributing factor to the decrease in reproduction with
time was that replacement clutches in my study period were
significantly smaller than the original clutch (Figure 3, paired
t-test: t12 = 2.551, P = 0.025), which contrasts to the finding
40 years ago that replacement clutches can both be smaller
and larger (Liversidge 1970a, Keith et al. 1992). I also found
that, not surprisingly, the re-nesting interval after failure
(mean = 19.8 days and sd = 2.6) was significantly shorter
than the re-nesting interval after a successful breeding
attempt (mean = 33 days and sd = 1.5, F1,15 = 5.184, P =
0.038).
Not only does the selection of a time to start a nesting
attempt have fitness consequences, but also the selection of
the nest site, i.e. the nest bush. Table 2 lists all the nest bush
species used in my study period and the success rate with
the number of fledglings per nest and successful nest. By far
the two most commonly used bush species were Sea Guarri
and Brown Gonna. There were large differences between
bush species in the success rate and one important contributing factor was the leaf architecture of the nest bush.
The two nest bush species with needle-type leaves had significantly lower success rates than the other nest bush
species with broad leaves (Figure 4, χ2 = 6.279, df = 1, P =
214
Krüger
Table 2: Plant species used for nesting in 2001 to 2003 with their frequency and success rate. Scientific and common names were taken from
Van Wyk and Van Wyk (1997)
Scientific name
Enuclea racemosa
Passerina filiformis
Olea esasperata
Rhus crenata
Rhus glauca
Metalasia muricata
Chrysanthemoides monilifera
Acacia cyclops
Cynanchum ellipticium
Dovyalis caffra
Rapanea gilliana
Osyris compressa
Didelta spinosa
Rhus laevigata
Cassine tetragona
Gymnosporia buxifolia
Mystroxylon aethiopicum
Rhus pyroides
Cussonia thyrsiflora
Common name
Sea Guarri
Brown Gonna
Dune Olive
Dune Crow-berry
Blue Kuni-bush
White Bristle Bush
Bush-tick Berry
Red Eye
Monkey Rope
Kei-apple
Dwarf Cape beech
Coastal Sumach
Thorny Salad bush
Dune currant
Climbing Saffron
Common Spike-thorn
Kooboo-berry
Common wild currant
Cape Coast cabbage
No. of nests (%)
33 (35.5)
14 (15.1)
6 (6.5)
6 (6.5)
6 (6.5)
4 (4.3)
4 (4.3)
3 (3.2)
3 (3.2)
2 (2.2)
2 (2.2)
2 (2.2)
2 (2.2)
1 (1.1)
1 (1.1)
1 (1.1)
1 (1.1)
1 (1.1)
1 (1.1)
% successful
45.5
14.3
66.7
50.0
16.7
0.0
25.0
66.7
66.7
0.0
0.0
50.0
50.0
0.0
0.0
0.0
100.0
100.0
0.0
45
40
PERCENTAGE (%)
CLUTCH SIZE
2.5
2.0
1.5
1.0
Fledged/nest
0.97
0.29
1.50
0.83
0.33
0.00
0.75
1.00
1.33
0.00
0.00
0.50
0.50
0.00
0.00
0.00
2.00
1.00
0.00
Fledged/succ. nest
2.13
2.00
2.25
1.67
2.00
0.00
3.00
1.50
2.00
0.00
0.00
1.00
1.00
0.00
0.00
0.00
2.00
1.00
0.00
Successful nests
Parasitised nests
35
30
25
20
15
10
5
0.5
Needle leaves
First clutch
Broad leaves
NEST BUSH TYPE
Replacement clutch
Figure 3: Comparison of clutch size between first and replacement
clutches (n = 13)
Figure 4: Comparison of success rate (% of nests fledging young)
and parasitism rate between nests placed in either shrubs with needle-type leaves (Passerina filiformis and Metalasia muricata) or
shrubs with broad-type leaves (all other bush species)
0.014). While this result should lead to a strong selection
pressure to nest in broad-leaved bushes, there was also
some evidence for a counteracting selection pressure. As
can be seen from Figure 4, I did not record a single incident
of brood parasitism by the Jacobin Cuckoo in a needleleaved bush. Although this distribution of parasitism is not
significantly different from expected (χ2 = 2.596, df = 1, P =
0.135), this is simply due to the small sample size of parasitised nests recorded so far (n = 11).
Discussion
Comparing the breeding biology of the Cape Bulbul over a
40-year period, the amount of similarities found between
both studies seems to indicate that there has been no significant shift in a major abiotic or biotic factor influencing bulbul reproduction. Many such responses to change have
been documented for various life history traits in other bird
species, notably in the Northern hemisphere (Gjerdrum et al.
2003, Jenni and Kery 2003). One factor that might influence
the amount of phenotypic plasticity in the Cape Bulbul is its
Ostrich 2004, 75: 211–216
long life-span (maximum of 13 years recorded in the wild,
Yom-Tov et al. 1994). It is well established that Southern
hemisphere passerines live longer than their equally-sized
Northern hemisphere counterparts (Yom-Tov et al. 1994),
and a long life-span coupled with iteroparity selects for
rather invariable investment into reproduction (Stearns
1992) in a rather stable environment. Rainfall, one crucial
factor influencing bird reproduction in southern Africa
(Liversidge 1966 and 1970a), is much less erratic at the
study site than it is further inland (Liversidge 1966). This
could well explain why there were no marked differences in
clutch size and reproductive success. However, an alternative feasible explanation is that differences were just not
large enough to be detected by comparing four with three
years of data.
The one-time trend documented indicated a decrease of
brood parasitism by the Jacobin Cuckoo. Whether this is a
real trend is doubtful for two reasons. First, the Jacobin
Cuckoo is very erratic in its annual density and presence in
any given area (Liversidge 1970b) and so a real trend could
only be established firmly with more years of data. Second,
it might be that I was not able to find parasitised nests with
the same frequency as Richard Liversidge did 40 years ago,
so differences in parasitism might also reflect differences in
nest-finding abilities. This could also explain the significantly smaller number of nests per pair I found in my study period.
One aspect of the breeding biology not covered by
Richard Liversidge 40 years ago is the large difference
between nest bush species with regard to breeding success
and brood parasitism. From my results it seems that the
selection of a bush to nest in can have profound fitness consequences but that brood parasitism by the Jacobin Cuckoo
counteracts the selection pressure exerted by predation risk
to some extent. Although more data are needed to establish
the relative importance of brood parasitism and predation as
selective forces influencing the choice of nest bush species
in the bulbul, there seems to be a trade-off between parasitism and predation risk. At current levels of parasitism and
predation, the selective force of brood parasitism, however,
seems to be rather low. Liversidge (1970b) estimated that
despite the high parasitism rate in his study, Jacobin
Cuckoos depressed the reproduction of the bulbul population by just 7%. It is known from other studies that cuckoos
sometimes do not parasitize host nests at random (Øien et
al. 1996, Honza et al. 1998, Soler et al. 1999, Moskat and
Honza 2000), but the strong avoidance of nests in certain
bush species I found in the Jacobin Cuckoo is striking. Why
broad-leaved bushes provide a safer nest site than needleleaved bushes is unclear at present, but one potential explanation might be that the more rigid branches in the needleleaved bushes are easier to climb for striped mice, which are
most likely the main predators of bulbul eggs at the study
site.
Acknowledgements — I am greatly indebted to Richard Liversidge for
all his initial help and support during the study. I am also deeply
indebted to Adri Barkhuysen for the many hours of fieldwork he
shared with me and all the logistic and moral support. Graham Kerley
kindly granted permission to conduct the study in the nature reserve.
Colleen de Villiers kindly supplied the weather data. Penn Lloyd and
Terry Oatley greatly improved earlier versions of this manuscript. This
215
study was supported through a Marie Curie Postdoctoral Fellowship
of the European Union, a Churchill College Junior Research
Fellowship and a Royal Society Research Fellowship.
References
Boveng PL, Hiruki LM, Schwartz MK and Bengtson JL 1998.
Population growth of Antarctic fur seals: limitation by a top predator, the leopard seal? Ecology 79: 2863–2877
Coppack T and Both C 2002. Predicting life-cycle adaptation of
migratory birds to global climate change. Ardea 90: 369–378
Cowling RH and Heijnis CE 2001. The identification of broad habitat units as biodiversity entities for systematic conservation planning in the Cape Floristic Region. South African Journal of Botany
67: 15–38
Davies NB 1999. Cuckoos and cowbirds versus hosts: coevolutionary lag and equilibrium. Ostrich 70: 71–79
Gargett V 1990. The Black Eagle: a Study. Acorn Books, Randburg
Gjerdrum C, Vallee AMJ, Cassady St. Clair C, Bertram DF,
Ryder JL and Blackburn GS 2003. Tufted puffin reproduction
reveals ocean climate variability. Proceedings of the National
Academy of Sciences USA 100: 9377–9382
Honza M, Øien IJ, Moksnes A and Røskaft E 1998. Survival of
Reed Warbler Acrocephalus scirpaceus clutches in relation to
nest position. Bird Study 45: 104–108
Hughes L 2000. Biological consequences of global warming: is the
signal already apparent? Trends in Ecology and Evolution 15:
56–61
Jenni L and Kery M 2003. Timing of autumn bird migration under
climate change: advances in long-distance migrants, delays in
short-distance migrants. Proceedings of the Royal Society
London B 270: 1467–1471
Kalejta-Summers B, Allan DG and Longrigg TD 2001. Long-term
trends, seasonal abundance and energy consumption of waterbirds at Rietvlei, Western Cape, South Africa, 1950–1997. Ostrich
72: 63–79
Keith S, Urban EK and Fry CH 1992. The Birds of Africa. Vol 4.
Academic Press, London
Ligon JD and Ligon SH 1990. Green woodhoopoes: life history
traits and sociality. In: Stacey PB and Koenig WD (eds) Co-operative Breeding in Birds. pp 31–66. Cambridge University Press,
Cambridge
Liversidge R 1966. Fluctuations in a breeding population in the
Eastern Cape. Ostrich Supplement 6: 419–424
Liversidge R 1970a. The Ecological Life History of the Cape Bulbul.
PhD thesis, University of Cape Town, South Africa
Liversidge R 1970b. The biology of the Jacobin Cuckoo Clamator
jacobinus. Ostrich Supplement 8: 117–137
Lloyd P, Craig AJFK, Hulley PE, Essop MF, Bloomer P and
Crowe TM 1997. Ecology and genetics of hybrid zones in the
southern African Pycnonotus bulbul species complex. Ostrich 68:
90–96
Lloyd P, Hulley PE and Craig AJFK 1996. Comparisons of the
vocalizations and social behaviour of southern African
Pycnonotus bulbuls. Ostrich 67: 118–125
Mayfield HF 1975. Suggestions for calculating nest success.
Wilson Bulletin 87: 456–466
McCleery R and Perrins CM 1998. Temperature and egg-laying
trends. Nature 391: 30–31
Moskat C and Honza M 2000. Effect of nest and nest site characteristics on the risk of cuckoo Cuculus canorus parasitism in the
Great Reed Warbler Acrocephalus arundinaceus. Ecography 23:
335–341
Newton I 1998. Population Limitation in Birds. Academic Press, San
Diego
Øien IJ, Honza M, Moksnes A and Røskaft E 1996. The risk of
parasitism in relation to the distance from reed warbler nests to
216
cuckoo perches. Journal of Animal Ecology 65: 147–153
Rice WR 1989. Analyzing tables of statistical tests. Evolution 43:
223–225
Ricklefs RE 1966. The temporal component of diversity among
species of birds. Evolution 20: 235–242
Soler JJ, Martinez JG, Soler M and Møller AP 1999. Host sexual
selection and cuckoo parasitism: an analysis of nest size in sympatric and allopatric magpie Pica pica populations parasitized by
the Great Spotted cuckoo Clamator glandarius. Proceedings of
the Royal Society London B 266: 1765–1771
Stearns SC 1992. The Evolution of Life Histories. Oxford University
Press, Oxford
Sutherland WJ 1996. From Individual Behaviour to Population
Ecology. Oxford Series in Ecology and Evolution. Oxford
University Press, Oxford
Krüger
Thiollay JM 1998. Long-term dynamics of a tropical savanna bird
community. Biodiversity and Conservation 7: 1291–1312
Thiollay JM 2000. Stability and long-term changes in a west African
raptor community. In: Meyburg BU and Chancellor RD (eds)
Raptors at risk. pp 15–25. Proceedings of the 5th World Congress
on Birds of Prey. Pica Press, Berlin
Van Wyk B and Van Wyk P 1997. Field Guide to Trees of Southern
Africa. Struik Publishers, Cape Town
Yom-Tov Y 2001. Global warming and body mass decline in Israeli
passerine birds. Proceedings of the Royal Society London B 268:
947–952
Yom-Tov Y, McCleery R and Oatley T 1994. The survival rates of
some southern African passerines. Ostrich 65: 329–332
Received January 2004, accepted April 2004
Editor: MD Anderson