Behavioral Ecology Vol. 11 No. 3: 309-314
© 2000 International Society for Behavioral Ecology
An experimental test of the prolonged brood care model in the tufted titmouse (Baeolophus bicolor)
Behavioral Ecology Group, Department of Evolution, Ecology, and Organismal Biology, The Ohio State University; 1735 Neil Avenue, Columbus, OH 43210-1293, USA
Address correspondence to E. Pravosudova, 604 Isla Pl., Davis, CA 95616. E-mail: epravosudova{at}hotmail.com .
Received 15 June 1999; accepted 24 September 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The prolonged brood care model rests on the assumption that retaining an offspring through the winter months in the face of a limited food supply should have a cost for parents. We tested this idea with a New World permanent-resident bird, the tufted titmouse (Baeolophus bicolor). Using DNA fingerprinting, we assessed the degree of relatedness between adult and juvenile birds in 17 winter groups, finding that in 8 of the groups no young bird was the offspring of the territorial pair. We compared the nutritional condition of territorial adult birds in small forest fragments from which their own offspring and other young had been removed with the nutritional condition of control birds from unmanipulated fragments. Contrary to the model's assumption, the nutritional condition of adults in treatment groups (young removed) appeared to be worse, not better, than in groups where a related juvenile was present. These results suggest that the prolonged brood care model may not be universal in its application and that under some ecological conditions, retaining offspring through the winter can result in a net benefit for territorial adults despite the necessity of sharing resources.
Key words: Baeolophus bicolor, nutritional condition, parental care, prolonged brood care model, winter social groups.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although factors contributing to the formation of social groups in animals have been frequently addressed (e.g., Axelrod and Hamilton, 1981
;
Brown, 1987; Emlen,
1982,
1997), most studies have
focused on social interactions during the breeding season. The adaptiveness of
nonbreeding aggregations has not received as much attention. In temperate
climates, some permanent-resident birds tend to form conspecific and/or
heterospecific social groups during the non-breeding season. The adaptive
significance of such flocking is usually explained in terms of predator
detection and/or improved foraging success (Dolby and Grubb,
1998,
2000;
Pulliam, 1973;
Thorpe, 1963).
Among theories explaining formation of winter social aggregations of
related conspecifics, the prolonged brood care model
(Ekman and Rosander, 1992;
Ekman et al., 1994) suggests
that parental control of natal dispersal can be an important factor
determining the size and composition of such groups. According to this model,
in the face of low resource abundance, socially dominant parents wintering on
their former breeding site do best by being competitive and retaining all
resources. In a situation where food competition is more relaxed, the model
predicts that sharing resources with independent offspring would be favored by
selection. Underlying such reasoning are the assumptions that delayed
dispersal is linked to relaxed winter competition and that offspring have a
higher probability of surviving their first winter if they stay in their
parents' territory than if they disperse. Even with the existence of this kin
bias in territorial pairs' behavior, the model assumes that there will be a
cost to adults from sharing the limited resources in their territory with
their young. That is, in order to retain their offspring through the winter,
adult birds should have to sacrifice some resources, resulting in lower levels
of nutrition for themselves. In contrast, territorial adults are expected to
increase their level of aggressiveness toward nonkin flock members and force
them to leave. The model suggests that nonkin young will be allowed to stay
only if resources are sufficient to be shared with more than one additional
flock member. Descriptive studies of Siberian (Perisoreus infaustus;
Ekman et al., 1994,
1996) and gray (Perisoreus
canadensis; Waite and Strickland,
1997) jays were consistent with the model in demonstrating that
these corvid species favored their own retained offspring over immigrants in
winter groups. Here we report the first controlled manipulative test of the
model.
The tufted titmouse (Baeolophus bicolor, formerly Parus
bicolor) is a common permanent resident of deciduous wood-land in eastern
North America, where it is a habitual member of heterospecific flocks
(Grubb and Pravosudov, 1994).
In winter, titmice form coherent conspecific groups of two to eight
individuals. Such groups usually include two adult birds (a territorial pair)
and, often, one or more of their offspring and/or first-year individuals from
unknown natal sites (Brackbill,
1970; Brawn and Samson,
1983; Nice, 1930;
Pielou, 1957). Family groups
extend into the breeding season, and helping at the nest has been reported
(Pielou, 1957). Such helping
is relatively rare in temperate species and is known to exist in only two
other parids, the South African black tit (Parus niger;
Tarboton, 1981) and the New
World bridled titmouse (Baeolophus wollweberil)
(Nocedal and Ficken,
1998).
In our study area within the agricultural landscape of central Ohio, tufted
titmice wintering in forest fragments form small conspecific groups of two to
five birds. Based on Ekman and Rosander's
(1992) model, we predicted that
if all young birds were removed from a winter group residing in a small
woodland fragment, the remaining adults would not have to share the
non-renewing food supply through the winter and therefore would be in better
nutritional condition compared to territorial adults in an unmanipulated
control group containing at least one retained offspring of the pair.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A total of 65 tufted titmice (34 adults and 31 juveniles) from 17 different groups were captured during the winters of 1995-1996 and 1996-1997 in 10 forest fragments in Union County, Ohio. All woodlots were approximately the same size and were similar in topography and vegetation, consisting primarily of oaks (Quercus spp.), ashes (Fraxinus spp.), shagbark hickory (Carya ovata), sugar maple (Acer saccharum), and American beech (Fagus grandifolia). Each woodlot was completely isolated from other woodlands by cultivated fields. During 1995-1996, two woodlots contained two groups of titmice each; all other woodlots contained only one social group. At the beginning of the experiment in late November, each titmouse group consisted of an adult pair and one to three first-year birds.
At the beginning of each winter, each woodlot was randomly assigned to the
treatment or control group. Five of the woodlots were used during both years.
Treatment (5.18 ± 0.69 SD ha) and control (5.13 ± 0.71 SD ha)
woodlots did not differ significantly in area (t test, df = 13;
t = 0.13; p =.89). During early winter, we mist-netted
titmice at sunflower-seed feeders. To minimize any effect of food
supplementation on the nutritional condition of the adult titmice, we
maintained feeders in woodlots for only the time required to attract and catch
the birds (less than a week in every case). Each captured titmouse was fitted
with a U.S. Fish and Wildlife Service aluminum band and with colored plastic
leg streamers for individual identification. Age (first-year or adult) was
determined by skull pneumatization and plumage, and sex was determined by wing
length and behavior. A 50-µl blood sample was taken from a brachial vein of
each bird, shaken with 500 µl of lysis buffer (100 mM Tris, pH = 8.0, 100
mM EDTA, 10 mM NaCl, 5% sodium dodecyl sulfate;
Longmire et al., 1988), and
stored at ambient temperature.
While capturing adults, we banded and then removed all the young birds from the treatment woodlots and released them approximately 50 km away in suburban Columbus, Ohio. None of these young birds was seen again in the study area.
We used ptilochronology (Grubb,
1989,
1995) to compare the
nutritional condition of adult titmice in treatment and control fragments.
This method is based on the fact that as a feather grows, alternating light
and dark bands appear across its vane. A combination of one light and one dark
band is termed a growth bar and represents 24 h of feather growth
(Brodin, 1993). The width of
growth bars reflects the nutritional condition of a bird during the period
when a feather was grown: birds in better nutritional condition have wider
growth bars than birds in poorer nutritional condition. It has been previously
shown that in winter, free-ranging tufted titmice grow feathers with wider
growth bars in response to a continuously present artifically enhanced food
supply (Grubb and Cimprich,
1990). To assess nutritional condition, we removed the outermost
left and right tail feathers from all the adult titmice in both treatment and
control woodlots at the time of capture and allowed the birds to regenerate
the feathers over the course of the next 6 weeks. We then recaptured the
birds, removed the induced feathers, and stored the feathers individually in
paper envelopes.
To determine if the removal of young had an effect on the nutritional
condition of adults, we measured growth bar width, feather mass, and total
length of both original and induced feathers from all recaptured adult birds.
Original feathers had been grown during the normal molting period the previous
autumn. To avoid bias, each feather was placed in a separate coded envelope
before being measured so that its identity was not known to the person
performing the measurements. To calculate the total length and average growth
bar width of a feather, we fixed it to an index card covered with a piece of
dark cloth. The dark background provided by the cloth increased the visibility
of growth bars. A size 0 insect-mounting pin was then pushed perpendicularly
down through the cloth and card at the proximal and distal ends of the feather
and at the margins of growth bars. Using the pin pricks on the card, we
measured total length of each feather to the nearest 0.1 mm and calculated the
mean value of the 10 daily growth bars centered on a point two-thirds of the
distance from the proximal end (Grubb,
1989). The mass of each feather was determined on an electronic
scale to the nearest 0.1 mg. To avoid any confounding effects of humidity or
temperature, we measured mass of all feathers on the same day. As both left
and right outermost tail feathers were measured for each bird, we used the
average of the left and right values for all three dependent variables. For
each of the three measurements, the correlation between a bird's left and
right feathers was high (all p <.005).
In preparation for statistical analysis, we took several measures to avoid any bias due to pseudoreplication. In the two cases where a woodlot contained two social groups of titmice, we randomly selected only one group for analysis. In cases where we collected induced feathers from both members of an adult pair, we randomly selected the feathers from either the male or the female for analysis. Three of the five woodlots that were used twice (in 1995-1996 and in 1996-1997) had new territorial adults during the second winter. For the two woodlots that had the same adult birds both winters, we randomly selected only one year's data for analysis. Thus, the primary sampling unit was one member of one territorial pair per woodlot-year with all birds used only once.
We used ANCOVA to determine the effect of removal of young on feather growth in adults. Treatment and sex were entered as factors, and to control for effects of bird size, we used dimensions of original feathers as covariates.
To determine relatedness among members of each social group, we used
multilocus minisatellite DNA fingerprinting. Before extraction, 250 µg of
proteinase K were added to each prepared blood sample; samples were then
incubated at 65°C overnight. Subsequently, two extractions with phenol,
two extractions with 25:24:1 phenol:chloroform:isoamyl alcohol, and one
extraction with 24:1 chloroform:isoamyl alcohol were performed. Following the
last extraction, the aqueous phase was dialyzed extensively against
TNE2 (10 mM Tris, pH = 7.4; 10 mM NaCl; 2 mM EDTA) for 4-6 h. Two
µg of DNA from each individual were digested with 7.5x excess
restriction enzyme HaeIII at 37°C for 4 h. Resulting fragments
were separated through a 0.8% agarose gel at 20 V for 65 h (until all
fragments smaller than 1,600 base pairs were run off the gel), and were then
transferred to nylon by Southern blot
(Southern, 1975) in 10x
SSC buffer and fixed to the membrane by UV crosslinking. Jeffreys's multilocus
minisatellite probe 33.15 (Jeffreys,
1985a,b)
was radiolabeled by primer extension. Hybridizations were run overnight, after
which hybridized filters were washed at 62°C in 1.5x SSC, 0.1%
sodium dodecyl sulfate and exposed to x-ray film at -20°C for several
days.
Samples from birds from the same winter social group were positioned next
to each other on a gel. Pairs of lanes on the resulting autoradiograph were
compared to examine the degree of band sharing between individuals. The band
sharing coefficient (x) reflects the degree of genetic similarity
between the two individuals under comparison
(Wetton et al., 1987). It can
be calculated as a proportion of the total number of bands in a dyad of lanes,
or x = 2S/(2S + A + B), where
S = the number of fragments of indistinguishable mobility and
intensity in the two lanes under comparison, A = the number of bands
unique to the first member of the dyad, and B = the number of bands
unique to the second member of the dyad.
To determine if first-year birds in a group were offspring of the
territorial adults, we used an independent set of band sharing values derived
from nestlings and their parents. In this second data set, seven families were
sampled during the 1996, 1997, and 1998 breeding seasons, and frequency
distributions of band sharing were created based on the known band sharing
coefficient values between confirmed first-order relatives and between
presumably unrelated individuals (e.g., mated pairs attending nests)
(Figure 1). The two
distributions overlapped at about x = 0.5. The lower value for the
95% confidence interval of the distribution for first-order relatives and the
upper value for the 95% confidence interval of the distribution for presumably
unrelated individuals coincided at 0.46. This value was then assigned as a
threshold so that if a band sharing coefficient between two birds of unknown
relatedness was >0.46, those two individuals were considered first-order
relatives. Birds with band sharing values <0.46 were considered unrelated.
In all cases where a young bird in the experiment was highly related to both
territorial adults (x > 0.46), we counted the number of novel
bands in its profile to confirm the presumed parentage. The number of novel
bands in all such cases ranged from zero to two, a range of values
attributable to random mutation (e.g.,
Haydock et al., 1996;
Rabenold et al., 1990).
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In 8 of 17 groups, no young bird was related to either territorial adult. Four of these groups had one first-year bird, and four had two first-year birds. The other nine groups had the following composition: in three of the groups the only young bird present appeared to be an offspring of both adults; two groups had one offspring and one unrelated young; one group had one offspring and two unrelated young; and one group contained two offspring and one unrelated young. In the remaining two winter groups some young were related to only one member of the territorial pair. In one of these groups two of the three young birds were first-order relatives of the territorial female, and in the other, one of the three young was highly related to the male (Table 1). In these two cases, data from only the related adult were used in statistical analyses.
|
In statistical analyses, no measurement of original feathers differed significantly either between treatment and control birds or between males and females (two-way ANOVA, p >.1 in all cases).
Because the molecular analysis was performed only after the field manipulation, we could not be aware of the degree of relatedness between adult and first-year members of the various groups at the time we removed juveniles. Therefore, to test the model's main assumption, that parents incur a cost by permitting their young to remain with them through the winter, we initially used only adults that at the beginning of the experiment had had at least one retained offspring on their winter territory (kin-group adults). Thus, we first compared nutritional condition of control adults that shared their territory with at least one retained offspring with nutritional condition of treatment adults from whose territories such offspring (along with unrelated young, if present) had been removed. A total of eight recaptured birds (two from treatment and six from control groups) were available for this analysis. The differences between treatment and control adults were nonsignificant for all three parameters of induced feathers (ANCOVA, F1,1 = 1.13 p =.48 for growth bar width; F1,1 = 45.63, p =.094 for mass; F1,1 = 0.55, p =.543 for total length) although all three measures of feather growth were somewhat greater in control birds.
To see if this trend persisted with larger sample sizes, we added records from treatment adults that initially had lacked retained offspring in their social group (nonkin adults). For a number of reasons, combining the two kinds of treatment adults seemed justified. First, because of the removal, neither kin-group nor nonkin treatment adults were required to share resources with any first-year birds. Second, the original tail feather measurements of kin-group and nonkin adults did not differ significantly (ANOVA; F1,6 = 0.04, p =.839 for growth bar width; F1,6 = 0.21, p =.658 for mass; F1,6 = 0.03, p =.850 for total length), indicating that the two groups of adults were of similar quality. Third, we could not detect any difference in induced feather measurements between the two kin-group and five nonkin treatment adults (F1,4 = 3.15, p =.450 for growth bar width; F1,4 = 0.001, p =.972 for mass; F1,4 = 0.93, p =.380 for total length). The large p values reinforce the conclusion that nonkin and kin-group adults reacted similarly to the removal of young birds.
The comparison of treatment and control groups performed after lumping records for the two kinds of treatment adults verified the trend existing in the earlier comparison. Contrary to our prediction, average growth bar width of induced feathers in treatment-group adults was significantly narrower, not wider, than average growth bar width of controls (ANCOVA, F1,6 = 7.44, p =.034 for seven treatment and six control adults; Figure 2A). Induced feathers from treatment birds were also significantly lighter than those from controls (ANCOVA; F1,6 = 6.42, p =.044; Figure 2B), but the difference between treatment and control feathers was not significant for total length (ANCOVA, F1,6 = 1.61, p =.251; Figure 2C).
|
In kin-group adults, no difference could be detected between the induced feathers of males and females standardized for average values of original feathers. Four females and four males were compared (ANCOVA, F1,1 = 0.05, p =.860 for the growth bar width; F1,1 = 17.24, p =.150 for mass; F1,1 = 0.72, p =.553 for the total length).
After increasing the sample size by adding values from nonkin treatment birds, we had available records from eight females and five males. Average induced growth bar width standardized for average growth bar width of original feathers was significantly greater in males (ANCOVA, F1,6 = 6.28, p =.046; Figure 3A), as was standardized average total feather length (ANCOVA, F1,6 = 8.49, p =.027; Figure 3C). Standardized average induced feather mass was not significantly different between males and females (ANCOVA, F1,6 = 1.50, p =.267; Figure 3B). There was no significant interaction (all p >.1) between treatment and sex for any dependent variable in any two-factor ANCOVA analysis.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
According to the prolonged brood care model, during the winter adults should tolerate their own young more than nonkin young. In this experiment, all of the wintering groups of tufted titmice initially contained a pair of territorial adults and at least one first-year bird. However, not all the groups included an offspring of the adult pair. The fact that the only first-year birds in 8 of 17 groups were unrelated to both adults suggests that in this species and habitat one or more mechanisms in addition to adult aggression toward young may be important in determining group composition. First, reproductive success may vary even in fragments of the same size (e.g., Lynch and Whigham, 1984
). Our
study began in the winter, so we do not know whether those adults lacking
related young in their groups had reproduced successfully, yet it seems that
the proportion of nonkin winter groups in our study is too high to be a result
of reproductive failure alone. However, it has been shown that fragmentation
can reduce dispersal rates in avian species in some habitats (e.g.,
Lens and Dhondt, 1994). In
agricultural woodland fragments, the tendency for young to disperse
voluntarily rather than stay with their parents may be stronger (e.g.,
Berg, 1997). Also, during the
onset of dispersal, parents might be less willing to share resources with
their young in fragmented habitat (since the resource base might be limited in
a fragment), and force their offspring to leave the territories that are poor
in resources. In such cases, it is quite possible that adults may be joined by
unrelated dispersing juveniles later on. It is also possible that such
immigrant first-year birds persist in nonkin winter groups even in the face of
substantial aggression from the resident adults (Ekman J, personal
communication). Finally, adults could have tolerated the presence of any
young, kin or nonkin in order to gain advantages of living in a larger flock
(Pulliam, 1973;
Thorpe, 1963).
Unfortunately, we do not have quantitative data on the availability of
titmouse food in our study woodlots. Thus, it remains possible that the food
supply in some woodlots was so low that territorial adults forced out
offspring to potentially spend the winter in a better territory. It is also
possible that young birds decided on their own to disperse in search of a
better territory. Immigrant young that join adults on worse territories may
have been low-ranking juveniles from other woodlots that failed to establish
on better grounds. In some species dominant juveniles actively force
subordinate siblings from the natal territory
(Strickland, 1991).
Furthermore, for those adults whose young had left or been expelled, it may
still have paid to accept additional flock members with whom they could share
minimal amounts of food without potentially suffering inclusive fitness
costs.
The model assumes that sharing limited winter resources with kin young has
a cost for adult birds. Based on this assumption, we predicted that adults
from manipulated groups (young removed) would do better nutritionally than
adults sharing food with their offspring. The results of our experiment
suggest that this was not the case; instead, such adults seemed to have done
worse. There might be several reasons for this disparity between prediction
and result. First, even under a limited food supply, a larger group size may
confer a fitness advantage. For a territorial adult, the benefits of improved
foraging efficiency and better predator detection may outweigh the costs of
sharing resources. Second, a deciduous forest habitat such as our study area
may provide a relatively high winter supply of resources. In deciduous
habitats territorial adults might not have been selected to be as despotic
toward subordinate flock members as would be the case in species living in
more harsh northern coniferous woodlands
(Ekman et al, 1994;
Waite and Strickland, 1997).
The adaptiveness of maintaining a larger flock size may outweigh that of
monopolizing scarce resources. Thus, resident adult tufted titmice may
tolerate one or more juveniles in their winter flock without significantly
sacrificing their food supply.
Although, because of the model's assumptions, control groups in which all young were unrelated to adults had to be excluded from the test of the prediction, we did use adults from treatment groups that had contained only nonkin young before manipulation to increase our sample size. We have shown that kin-group and nonkin treatment adults did not differ significantly in nutritional condition during the experiment. This result is not likely due to small sample sizes because, for the same group of treatment adults, the differences between males and females for all three feather parameters were highly significant, with all the values being higher in males (ANOVA, F1,7 = 478.66, p =.0002 for growth bar width; F1,7 = 14.93, p =.006 for mass; F1,7 = 7.15, p =.032 for total length).
Regardless of the reasons that some of the groups lacked kin young in early winter (e.g., adults' reproductive attempt had failed or all the resident young had dispersed), analysis of original feathers indicated that the nutritional condition of adult titmice in all the woodlots was about the same at the start of the manipulation. The results of our experiment thus suggest that the significantly lower average growth bar width and mass of induced feathers in the treatment-group birds can be attributed to the experimental manipulation. Whether adults spending the winter with retained offspring differ in nutritional condition from adults spending the winter in nonkin groups is beyond the scope of our original question, but all induced feather growth parameters were not significantly different between the seven kin-group and four nonkin control adults (p =.365 for the growth bar width, p =.136 for mass, and p =.826 for total length of induced tail feathers).
Male titmice grew their induced feathers faster and to a greater total
length than did females, a result holding even after standardization to values
of original feathers. Grubb and Cimprich
(1990) found a similar sex
difference in feather regeneration in this species. As adult female tufted
titmice are socially subordinate to adult males in winter flocks
(Grubb and Pravosudov, 1994),
it is not surprising that females in our sample were in comparatively poorer
nutritional condition. However, since no significant interaction was shown
between treatment and sex, there is no evidence that the manipulation affected
sexes differently.
In conclusion, our results do not support the key assumption of the prolonged brood care model. Our experiment demonstrated that the nutritional condition of territorial adults improved, rather than suffered, in the presence of related conspecific first-year members of a winter flock. Therefore, in some systems, the nutritional benefits for adults of associating with young in winter may outweigh the costs of sharing food. However, many other factors contributing to the formation of social groups in this species remain unclear. For example, it will be important to determine how adults share resources with relatives as compared to unrelated flock members (Ekman J, personal communication). Nutritional condition of retained offspring and levels of adult aggressive behavior toward them could be compared with the same parameters in immigrant young residing in similar-sized forest fragments. However, although such an analysis could show whether adults differentiate between kin and nonkin in sharing extra resources, it would not diminish the fact that adults benefited nutritionally from having conspecific young in their territories rather than suffering a cost from sharing resources, as the prolonged brood care model suggests.
|
ACKNOWLEDGEMENTS |
|---|
We thank Patty Parker for allowing us to use her lab and for great help with earlier drafts of the manuscript. We also thank the Conklin, Evans, Fitzpatrick, Geyer, Lowe, Michaels, Schmitter, and Scott families for allowing us to use their property. N. Arguedas, J. Diaz, A. Dolby, T. Jones, K. Lundy, D. Sillick, R. Tuttle, and B. Worden helped with the field and lab work. Helpful comments from Anders Brodin, Jan Ekman, A.S. Gaunt, the members of our lab group and the Parker lab group, and an anonymous reviewer improved the manuscript. This study was supported by The American Ornithologists' Union, National Science Foundation grant IBN-9522064, the North American Bluebird Society, The Ohio State University, the Society of Sigma Xi, and The Wilson Ornithological Society.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Axelrod R, Hamilton WD, 1981. The evolution of cooperation. Science 211: 1390-1396.
Berg A, 1997. Diversity and abundance of birds in relation to forest fragmentation, habitat quality and heterogeneity. Bird Study 44: 355-366.
Brackbill H, 1970. Tufted titmouse breeding behavior. Auk 87: 522-536.
Brawn JD, Samson FB, 1983. Winter behavior of tufted titmice. Wilson Bull 95: 222-232.
Brodin A, 1993. Radio-ptilochronology: tracing radioactively labeled food in feathers. Ornis Scand 24: 167-173.
Brown JL, 1987. Helping and communal breeding in birds: ecology and evolution. Princeton, New Jersey: Princeton University Press.
Dolby AS, Grubb TC Jr, 1998. Benefits to satellite members in mixed-species foraging groups: an experimental analysis. Anim Behav 56: 501-509.[Web of Science][Medline]
Dolby AS, Grubb TC Jr, 2000. Social context affects
risk-taking by a satellite species in a mixed-species foraging group.
Behav Ecol 11:
110-114.
Ekman J, Rosander B, 1992. Survival enhancement through food sharing: a means of parental control of natal dispersal. Theor Popul Biol 42: 117-129.[Web of Science][Medline]
Ekman J, Sklepkovych B, Tegelstrom H, 1994. Offspring
retention in the Siberian jay (Perisoreus infaustus): the prolonged
brood care hypothesis. Behav Ecol 5:
245-253.
Ekman J, Brodin A, Bylin A, Sklepkovych B, 1996.
Selfish long-term benefits of hoarding in the Siberian jay. Behav
Ecol 7:
140-144.
Emlen ST, 1982. The evolution of helping: I. An ecological constraints model. Am Nat 119: 29-39.[Web of Science]
Emlen ST, 1997. Predicting family dynamics in social vertebrates. In: Behavioural ecology: an evolutionary approach (Krebs JR, Davies NB, eds). Oxford: Blackwell Science; 228-253.
Grubb TC Jr, 1989. Ptilochronology: feather growth bars as indicators of nutritional status. Auk 106: 314-320.
Grubb TC Jr, 1995. Ptilochronology: a review and prospectus. Curr Ornithol 12: 89-114.
Grubb TC Jr, Cimprich DA, 1990. Supplementary food improves the nutritional condition of wintering woodland birds: evidence from ptilochronology. Ornis Scand 21: 277-281.
Grubb TC Jr, Pravosudov VV, 1994. Tufted titmouse (Parus bicolor). In: The birds of North America, no. 86 (Poole A, Gill F, eds). Philadelphia: American Ornithological Union.
Haydock JP, Parker PG, Rabenold KN, 1996. Extra-pair paternity uncommon in the cooperatively breeding bicolored wren. Behav Ecol Sociobiol 38: 1-16.
Jeffreys AJ, Wilson V, Thein SL, 1985a. Hypervariable 'minisatellite' regions in human DNA. Nature 314: 67-73.[Medline]
Jeffreys AJ, Wilson V, Thein SL, 1985b. Individual-specific 'fingerprints' of human DNA. Nature 316: 76-79.[Medline]
Lens L, Dhondt AA, 1994. Effects of habitat fragmentation on the timing of crested tit Parus cristatus natal dispersal. Ibis 136: 147-152.
Longmire JL, Lewis AK, Brown NC, Buckingham JM, Clark LM, Jones MD, Meinke LJ, Meyne JM, Ratiff RL, Ray FA, Wagner RP, Moyziz RK, 1988. Isolation and molecular characterization of a highly polymorphic centromeric tandem repeat in the family Falconidae. Genomics 2: 14-24.[Medline]
Lynch JF, Whigham DF, 1984. Effects of forest fragmentation on breeding bird communities in Maryland, USA. Biol Conserv 28: 287-324.
Nice MM, 1930. Winter range of tufted titmouse. Wilson Bull 45: 87.
Nocedal J, Ficken MS, 1998. Bridled titmouse (Baeolophus wollweberi). In: The birds of North America, no.375 (Poole A, Gill F, eds.). Philadelphia: American Ornithological Union.
Pielou WP, 1957. A life-history study of the tufted titmouse, Parus bicolor Linnaeus (PhD dissertation). East Lansing: Michigan State University.
Pulliam HR, 1973. On the advantages of flocking. J Theor Biol 38: 419-422.[Web of Science][Medline]
Rabenold PP, Rabenold KN, Piper WH, Haydock J, Zack SW, 1990. Shared paternity revealed by genetic analysis in cooperatively breeding tropical wrens. Nature 348: 538-540.
Southern EM, 1975. Detection of specific sequence among DNA fragments separated by gel electrophoresis. J Mol Biol 98: 503-517.
Strickland D, 1991. Juvenile dispersal in gray jays—dominant brood member expels siblings from natal territory. Can J Zool 69: 2935-2945.
Tarboton WR, 1981. Cooperative breeding and group territoriality in the black tit. Ostrich 52: 216-225.
Thorpe WH, 1963. Learning and instinct in animals. London: Methuen.
Waite TA, Strickland D, 1997. Cooperative breeding in gray jays: philopatric offspring provision juvenile siblings. Condor 99: 523-525.
Wetton JH, Carter RE, Parkin DT, Walters D, 1987. Demographic study of a wild house sparrow population by DNA fingerprinting. Nature 327: 147-149.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||