Abstract

Many animals have successfully adapted to human proximity, with dramatic increases in abundance as a consequence. Although such transitions imply a fitness advantage, the fitness benefits of associations between animals and humans have not been thoroughly investigated. In a comparative study of nest predation, I compared predation rates in 6874 nests of 11 species of birds with sympatric populations breeding indoors and outdoors. Mean nest predation rates were 23.5% outdoors, but only 1.0% indoors, because corvid nest predators never entered buildings. There was a negative correlation between nest predation rate and the proportion of individuals breeding indoors, implying that as species became more adapted to humans, and hence breeding indoors became more frequent, there was a significant decrease in nesting failure that translated into a difference in reproductive success due to reductions in nest predation. Finally, the difference in predation rate between outdoor and indoor nests was related to time since urbanization and number of generations since urbanization, implying that initially there was a large selection differential followed by reduced fitness differences between birds breeding outdoors and indoors due to gradual adaptation to human proximity by reproducing birds. With a high intensity of natural selection, these findings suggest that such adaptation to human proximity may only take a few hundred generations, as shown by several species that have only recently become associated with humans.

Large areas of natural habitats and farmland have been converted into urban areas during recent decades (Klausnitzer 1989; Shochat et al. 2006). Such urbanization has caused dramatic increases in areas occupied by cities and hence humans in North America, Europe, and Asia. These changes have forced animals and plants to either adapt to the novel conditions that include the proximity of humans or disappear. The characteristics of species that have adapted to the proximity of humans and hence urbanization, and the resulting changes in behavior, physiology, and life history, are not well understood (Diamond 1986; Klausnitzer 1989). Urban areas typically have longer growing seasons, reduced predator abundance, elevated resource abundance for the species that breed there, and higher population densities than rural areas (Shochat et al. 2006). In general, such differences between urban and rural habitats provide individuals belonging to species that manage the transition from natural to man-made habitats with large fitness benefits.

Key innovations that allow species to adapt to novel environments include novel ways of procuring food (e.g., Lefebvre et al. 1997) and novel sites used for reproduction (e.g., Nowakowski 1994; Nicolakakis et al. 2003; Yeh et al. 2007). Indeed, a high rate of feeding innovations is a characteristic of bird species that have become urbanized (Møller 2009). Some species of animals have adapted to the proximity of humans to the extent that only marginal populations are found in natural habitats (Klausnitzer 1989). That is the case for barn swallow Hirundo rustica, house sparrow Passer domesticus, and many other species of birds, whereas numerous others only rarely or never breed close to humans. The ability to tolerate close proximity of humans is elevated in urban compared with rural populations of the same species of birds (Møller 2008), and variance in tolerance of different species of birds to humans not only predicts successful urbanization but also changes in fearfulness after urbanization (Møller 2009). Several studies have suggested that humans can provide animals protection against predators (Klausnitzer 1989; Gliwicz et al. 1994), although there are no studies that have tested the generality of this phenomenon nor are there to the best of my knowledge information on the fitness advantages associated with this reproductive innovation. If a substantial fitness advantage existed, we should expect that species that have been adapting to man-made environments for a longer time to have a greater fraction of the population located in the modified habitats. Finally, we should expect that the reproductive advantage of being associated with humans should increase gradually with time since urbanization if this constituted adaptation to a novel environment.

The objective of this study of associations of breeding birds with humans was 1) to test for a general reduction in nest predation rate caused by the proximity of humans. To this end, I quantified nest predation rates for all 11 species of birds that have been found during the course of the study to naturally breed in trees, caves, or on the outside of buildings but are known to breed inside buildings such as barns (hereafter indoors), thus constituting sympatric populations that either breed inside buildings or outside in natural habitats. Other species like rock dove Columba livia and robin Erithacus rubecula that are known to breed indoors elsewhere were not recorded during the course of this study. Such an intraspecific analysis provides a strong test of the prediction that humans provide protection because populations that breed indoors and outdoors coexist and because the ancestral state is clearly breeding outdoors. Indoor nest sites are not limiting, so an absence of indoor breeding cannot be attributed to a lack of available habitat. 2) If breeding birds benefit from human association in terms of reduced predation risk, then overall nest predation rate should be negatively related to the proportion of breeding pairs of a given species that breed indoors. Such a negative relationship would imply that species that once bred outdoors and had high nest predation rates now have become associated with humans, enjoying a significant reduction in risk of predation. Because nest predation is a very important source of nest mortality, we should expect that any difference in nest predation would translate into a difference in reproductive success. We should also expect that gradual adaptation to the proximity of humans would result in the difference in nest predation rate between indoor and outdoor populations would depend on time since the first cases of breeding indoors and the number of generations since initial cases of indoor breeding. I tested these predictions using an extensive data set on nest predation rates of 6174 nests from paired sympatric populations of 11 species obtained during a period of 37 years. This design is particularly powerful because indoor and outdoor populations live under similar conditions in all other respects than nest sites. The most common nest predators in the study area are carrion crow Corvus corone, magpie Pica pica, jay Garrulus glandarius, rat Rattus norvegicus, marten Martes foina, and domestic cat Felis domesticus, as revealed by direct observations and experiments (Møller 1988, 1989, 1990, 1991). Although the mammalian predators occur both indoors and outdoors, breeding birds can seek shelter from avian nest predators by breeding indoors, thus providing a significant increase in fitness. It is this association with humans that shelters breeding birds from a significant segment of the nest predator community that is the subject of the present study.

MATERIALS AND METHODS

Study area

The study was conducted in Northern Jutland, Denmark, and Chernobyl, Ukraine (the latter area contains the only known records of indoor nests of Delichon urbica and Turdus philomelos ever reported, and both derived from a single farm in the village Pisky with more than 200 inhabitants). The results remain qualitatively similar if these 2 species from Chernobyl were excluded. I made extensive systematic searches for nests during the spring and summer months April–August 1970–2006. I relied on extensive help from amateurs with a good knowledge of birds in locating nests, and I requested help in locating nests of all commonly breeding species. In total, I obtained information on all 11 species with populations breeding both indoors and outdoors, in total 6174 nests (Table 1).

Table 1

Summary of sample sizes for nests, eggs, nest predation rates, mean brood size at fledging, body mass (g), and timing of urbanization of birds breeding indoors and outdoors

 Indoors
 
Outdoors
 
   
Species Number of nests Number of eggs Nest predation rate Mean brood size Number of nests Number of eggs Nest predation rate Mean brood size Body mass (g) Time urbanized Number of generations since urbanization 
Columba palumbus 14 28 0.030 1.94 134 266 0.494 0.97 494.50 1950 21 
Delichon urbica 26 0.000 4.62 260 1070 0.040 3.89 19.55 1850 90 
Fringilla coelebs 13 0.000 4.46 56 247 0.297 3.05 24.20 1950 21 
Hirundo rustica 4505 22 255 0.010 4.77 26 104 0.025 4.42 19.10 1850 100 
Muscicapa striata 26 0.000 4.38 38 0.030 3.21 15.50 1950 16 
Parus major 24 0.000 10.12 261 2144 0.120 7.44 18.50 1950 30 
Passer domesticus 333 1665 0.000 4.33 61 298 0.050 4.21 30.35 1880 58 
Streptopelia decaocto 0.000 1.97 12 24 0.349 1.16 201.50 1950 23 
Troglodytes troglodytes 12 96 0.000 5.87 46 0.122 5.87 8.90 1930 49 
Turdus merula 66 302 0.051 4.58 369 1690 0.384 3.51 95.85 1900 48 
Turdus philomelos 13 0.000 4.71 26 109 0.422 0.48 70.50 1960 26 
 Indoors
 
Outdoors
 
   
Species Number of nests Number of eggs Nest predation rate Mean brood size Number of nests Number of eggs Nest predation rate Mean brood size Body mass (g) Time urbanized Number of generations since urbanization 
Columba palumbus 14 28 0.030 1.94 134 266 0.494 0.97 494.50 1950 21 
Delichon urbica 26 0.000 4.62 260 1070 0.040 3.89 19.55 1850 90 
Fringilla coelebs 13 0.000 4.46 56 247 0.297 3.05 24.20 1950 21 
Hirundo rustica 4505 22 255 0.010 4.77 26 104 0.025 4.42 19.10 1850 100 
Muscicapa striata 26 0.000 4.38 38 0.030 3.21 15.50 1950 16 
Parus major 24 0.000 10.12 261 2144 0.120 7.44 18.50 1950 30 
Passer domesticus 333 1665 0.000 4.33 61 298 0.050 4.21 30.35 1880 58 
Streptopelia decaocto 0.000 1.97 12 24 0.349 1.16 201.50 1950 23 
Troglodytes troglodytes 12 96 0.000 5.87 46 0.122 5.87 8.90 1930 49 
Turdus merula 66 302 0.051 4.58 369 1690 0.384 3.51 95.85 1900 48 
Turdus philomelos 13 0.000 4.71 26 109 0.422 0.48 70.50 1960 26 

Life-history information

Adult birds carrying food in their beak were particularly used as a means of locating nests with nestlings. I deliberately attempted not to touch the nest or disturb the surrounding vegetation to avoid increasing the risk of nest predation. If nests contained eggs, when found, they were revisited after a duration of half the incubation period, and then again, if still not lost to predators, when nestlings were approximately two-thirds through the nestling period. The number of nest checks was in this way minimized to reduce any unnecessary predation due to investigators. A nest was considered to be depredated if nest contents were missing on 1 of the subsequent visits after the nest initially was found but before it could have produced fledglings.

Nests were classified as indoors if they were placed inside barns, sheds, or other buildings and otherwise as being outdoor nests.

I estimated the proportion of nests being built indoors from the nest records, simply as the proportion of nests indoors over the total number of nests. Although nest search effort was concentrated indoors because of long-term studies of barn swallows and house sparrows, this index of the proportion of nests indoors does still provide a relative measure of association with humans.

Nest searching was exhaustive for D. urbica, H. rustica, Muscicapa striata, Parus major, and P. domesticus because these species were subject of specific studies. For the remaining 6 species, I made regular systematic searches of bushes and trees during the breeding season, and this was possible because total coverage of these habitats was less than 1% of the total 15 km2 study area. Thus, I have no reason to believe that there are inherent biases in sampling or sampling effort. I made an explicit attempt to estimate the reliability of estimates of the proportion of nests found indoors by calculating estimates for the years separately. A repeatability analysis (Becker 1984) based on a one-way analysis of variance using square root arcsine-transformed data provided a highly significant model (F = 447.57, degrees of freedom = 10, 209, P < 0.0001, repeatability [R] standard error [SE] = 0.96 [0.03]). Thus, there was evidence suggesting that estimates were highly consistent among years, although this resulted in a dramatic reduction in sample size for many years.

I used body mass of all species as an additional predictor variable based on my own field measurements or in the absence of data as reported by Cramp and Perrins (1977–1994).

Daily nest predation rate was estimated using the Mayfield (1975) method as modified by Johnson (1979). Total nest predation rate was then estimated from the daily nest predation rate estimates and the duration of the nestling periods, as reported by Cramp and Perrins (1977–1994).

Reproductive success was estimated as the number of nestlings produced per nesting attempt.

Timing of urbanization reflecting the time when a given species first started breeding commonly in cities may be an important parameter relevant for nest site choice because both phenomena involve wild animals adjusting to close proximity of humans. I adopted an approach suggested by Møller (2008) to estimate timing of urbanization. These estimates were obtained by requesting information on timing of urbanization from older ornithologists combined with information from the literature. Four independent estimates of timing of urbanization based on estimates by 2 experienced ornithologists from Denmark, myself, and a literature review covering observations of breeding birds in Copenhagen were strongly positively correlated, implying that these independent estimates reliably reflected the same phenomenon (see Møller, 2008, for full documentation). This information on timing of urbanization is provided in Møller (2008).

I extracted information on adult survival rate (S) and age at first reproduction (A), using Cramp and Perrins (1977–1994) as a source, and if information was missing, I used Glutz von Blotzheim and Bauer (1966–1997). I estimated generation time G as A + S/(1 − S). Thus, the number of generations since urbanization was estimated as the number of years since urbanization divided by generation time.

Statistics

Nest predation and the proportion of nests found indoors were square root arcsine transformed, whereas body mass and number of generations were log10 transformed. I tested for a difference in nest predation rate using a paired t-test, weighted by sample size. The association between nest predation rate and proportion of nests found indoors was tested with linear regression. I tested whether the difference in nest predation rate between indoor and outdoor nests translated into a difference in reproductive success by relating the difference between log10-transformed mean brood size at fledging for outdoor nests and log10-transformed mean brood size at fledging for indoor nests was related to the difference in square root arcsine-transformed nest predation rate of outdoor and indoor nests.

Many studies in behavioral ecology and ecology and evolutionary biology do not weight analyses by sample size (Garamszegi and Møller 2010). It is a common underlying assumption of most statistical approaches that each data point provides equally precise information about the deterministic part of total process variation, that is, the standard deviation (SD) of the error term is constant over all values of the predictor variables (Sokal and Rohlf 1995). The standard solution to violations of this assumption is to weight each observation by sampling effort in order to use all data by giving each datum a weight that reflects its degree of precision due to sampling effort (Draper and Smith 1981; Neter et al. 1996). Comparative analyses may be confounded by sample size if sampling effort is important and if sample size varies considerably among taxa. Therefore, I weighted regression by sample size (Møller and Nielsen 2006). In order to weight regressions by sample size in the analysis of contrasts, we calculated weights for each contrast by calculating the mean sample size for the taxa immediately subtended by that node (Møller and Nielsen 2006). Weighting regressions with N or square root of (N − 3) (Rosenthal 1991, p. 27–28) gave statistically similar conclusions.

Closely related species are more likely to have similar phenotypes than species that are more distantly related. Therefore, species cannot be treated as statistically independent observations in comparative analyses because apparent phenotypic correlations among species may result from species sharing a common ancestor rather than convergent evolution.

I controlled for similarity in phenotype among species due to common phylogenetic descent by calculating standardized independent linear contrasts (Felsenstein 1985) using the software CAIC (Purvis and Rambaut 1995). All branches of the phylogenies were assigned uneven branch lengths, assuming a gradual evolution model as implemented in the software, although a second set of analyses based on similar branch length produced qualitatively similar results to those reported here. I tested the statistical and evolutionary assumptions of the comparative analyses (Garland et al. 1992) by regressing absolute standardized contrasts against their SDs. In order to test for effects of problems of heterogeneity in variance, I repeated analyses with the independent variable expressed in ranks. These analyses are conservative tests of the null hypothesis, explicitly investigating the robustness of the conclusions. These new analyses did not change any of the conclusions, and they are therefore not reported here.

The composite phylogeny used in the comparative analyses was based on Sibley and Ahlquist (1990) (Supplementary Figure S1).

I calculated weights for each contrast by calculating the mean sample size for the taxa immediately subtended by that node (Møller and Nielsen 2006). Thus, if 2 sister species had sample sizes of 2 and 10, sample size at the node subtended from these 2 species would be (2 + 10)/2 = 6.

I used multiple regression to find the best-fit model using the software JMP (2000). There was no evidence of collinearity between variables (Tabachnick and Fidell 1996).

Regressions of standardized linear contrasts were forced through the origin because comparative analyses assume no evolutionary change in a character when the predictor variable has not changed (Purvis and Rambaut 1995).

RESULTS

Nests found indoors, nest predation, and breeding success

Nest predation rate ranged from 0.008 to 0.450 among species (Table 1). Mean nest predation rate indoors was 0.008 (SE = 0.005) for the 11 species, whereas it was 0.212 (SE = 0.054) for outdoor populations of the same species. The means, weighted by sample size, were on average 0.010 (SE = 0.002), whereas it was 0.235 (SE = 0.054) for outdoor populations of the same 11 species. A paired t-test weighted by sample size showed a significant difference (paired t = 2.89, df = 10, P = 0.016), with all species having higher predation rates outdoors than indoors (Figure 1, binomial test, 11 of 11 species, when 5.5 is expected at random: P < 0.001).

Figure 1

Nest predation rate of 11 bird species when breeding indoors and outdoors. The line is y = x.

Figure 1

Nest predation rate of 11 bird species when breeding indoors and outdoors. The line is y = x.

The overall nest predation rate experienced by different species was negatively related to the proportion of nests being found indoors (Figure 2, linear regression weighted by sample size: F = 16.66, df = 1,9, r2 = 0.65, P = 0.0027, slope [SE] = −2.26 [0.55]). That was also the case after including body mass as an additional variable (multiple regression: F = 34.09, df = 2,8, r2 = 0.89, P < 0.0001; proportion of nests indoors: slope [SE] = −0.16 [0.05], t = 11.38, P = 0.0097; body mass: slope [SE] = 0.37 [0.08], t = 19.00, P = 0.0024). An analysis of standardized independent contrasts showed that only the proportion of nests indoors significantly predicted variation in nest predation rate (linear regression weighted by sample size: F = 5.34, df = 1,9, r2 = 0.37, P = 0.046, slope [SE] = −3.32 [1.44]). Therefore, species with currently small nest predation rates breed indoors to a larger extent.

Figure 2

Nest predation rate for 11 different bird species in relation to the proportion of nests found indoors.

Figure 2

Nest predation rate for 11 different bird species in relation to the proportion of nests found indoors.

The difference in reproductive success estimated as brood size at fledging between outdoor and indoor nests was predicted by the difference in predation rate between outdoor and indoor nests (F = 7.38, df = 1,9, r2 = 0.45, P = 0.024, slope [SE] = −0.867 [0.319]) and that was also the case in a model weighted by sample size (F = 9.70, df = 1,9, r2 = 0.52, P = 0.012, slope [SE] = −0.391 [0.125]). Therefore, not only predation rate but also reproductive success was affected by nest location.

Nest predation, time since urbanization, and number of generations

The proportion of nests found indoors increased with the number of years since urbanization (F = 15.55, df = 1,9, r2 = 0.63, P = 0.0034, slope [SE] = 0.013 [0.003]). Likewise, the proportion of nests found indoors increased with the number of generations since urbanization (F = 19.89, df = 1,9, r2 = 0.69, P = 0.0016, slope [SE] = 0.016 [0.004]).

The difference in nest predation rate between outdoor and indoor populations of the same species increased with time since urbanization (Figure 3A; F = 9.85, df = 1,9, r2 = 0.52, P = 0.012, slope [SE] = 0.0037 [0.0012]) and that was also the case in a model weighted by sample size (F = 41.22, df = 1,9, r2 = 0.82, P = 0.0001, slope [SE] = 0.0045 [0.0007]). An analysis of linear contrasts provided a similar conclusion (F = 6.19, df = 1,9, r2 = 0.41, P = 0.035, slope [SE] = 0.0033 [0.0013]), as did a model weighted by sample size (F = 9.69, df = 1,9, r2 = 0.52, P = 0.013, slope [SE] = 0.0031 [0.0010]). Therefore, there was evidence of gradual adaptation to human cohabitation with an initial large fitness advantage, followed by decreasing fitness benefits.

Figure 3

Difference in nest predation rate between outdoor and indoor nests for 11 different bird species in relation to (A) time since urbanization and (B) number of generations since urbanization.

Figure 3

Difference in nest predation rate between outdoor and indoor nests for 11 different bird species in relation to (A) time since urbanization and (B) number of generations since urbanization.

The difference in nest predation rate between outdoor and indoor populations decreased with the number of generations since urbanization (Figure 3B; F = 5.88, df = 1,9, r2 = 0.40, P = 0.038, slope [SE] = −0.50 [0.21]) and that was also the case in a model weighted by sample size (F = 60.42, df = 1,9, r2 = 0.87, P < 0.0001, slope [SE] = −0.77 [0.10]). An analysis of contrasts did not reach statistical significance (F = 2.08, df = 1,9, r2 = 0.19, P = 0.18), but a model weighted by sample size provided a significant relationship (F = 9.28, df = 1,9, r2 = 0.51, P = 0.014, slope [SE] = −0.55 [0.18]). Finally, there was also a significant relationship between difference in brood size between outdoor and indoor nests and number of generations since urbanization (F = 5.33, df = 1,9, r2 = 0.37, P = 0.046, slope [SE] = 0.0020 [0.0009]).

DISCUSSION

Populations of birds that breed indoors in barns and other buildings enjoyed reduced nest predation rates, compared with outdoors, because human proximity eliminated corvids as nest predators from buildings with consequences for reproductive success. Nest predation effects might be partially compensated by reduced starvation rates. However, this alternative could be dismissed because the difference in reproductive success between nests indoors and outdoors was predicted by the difference in nest predation rate. The dramatic difference in success among birds that breed next to each other constitutes an enormous selection differential with respect to being associated with humans. This change in reproductive success represented a gradual change in nest predation rate and reproductive success with time since urbanization. The second major finding of this study was that overall nest predation rates of different bird species were reduced when individuals of a given species placed an increasing proportion of their nests indoors. Given that this correlation does not allow inference about causality, the alternative interpretation that the proportion of nests being found indoors is directly affected by nest predation rate is equally likely. Finally, the difference in predation rate between nests found outdoors and indoors was directly related to time since urbanization, implying that there has been gradual adaptation to human proximity by reproducing birds and that many species that became urbanized early (and hence associated with humans long time ago) enjoy smaller fitness differentials from breeding indoors than species that only recently started to breed indoors.

This study was based on information on nest site choice, nest predation, reproductive success, and timing of urbanization of only 11 species. Although this is admittedly a small sample, and sample sizes for some species were small because they represented very high or very low frequencies of nests found indoors, this study represents so far the largest sample, and the sample is exhaustive because no species breed indoors that were not included here. Indeed, previous studies of similar phenomena have often only been based on a single species. The findings reported here concern a breeding innovation in terms of close proximity to humans when birds make their choice of nest site. Although it is easy to understand that it takes very little for animals to change their breeding sites from overhanging banks, cracks, crevices, or holes to human dwellings and buildings, it is much more difficult to understand how wild animals have become adapted to the close proximity of humans. In the present study, species that have only recently become urbanized, as defined by Møller (2009), enjoyed the largest benefits in term of protection of nests against nest predators, with this effect decreasing with time since urbanization. These results imply that there are very large fitness benefits to be acquired by the first individuals settling inside buildings and that any individual that makes this first step toward human cohabitation experiences large fitness advantages. Although it is possible that the ability to cope with human proximity has spread from urban to rural habitats, I consider this possibility unlike simply because urban birds generally have much reduced dispersal distances compared with their rural counterparts (Klausnitzer 1989). Thus, it is much more likely that this ability to cope with human proximity has evolved independently in urban birds and in birds breeding indoors in rural areas.

The underlying process accounting for temporal change in nest site choice could be learning or an evolutionary response to selection. Although learning would imply a change in phenotype that could happen over very short time given the large selective benefits, adaptation would be based on gradual changes in gene frequencies over time. Indeed, the difference among species in frequency of nests found indoors was related linearly to time since urbanization (another phenomenon that involves the ability to cope with human proximity). Such a linear change during more than 150 years is unlikely to be caused by learning, which would result in a rapid change over time (as seen in milk bottle opening by tits). In contrast, a gradual change that has not come to completion after more than 150 years is consistent with microevolutionary change. An evolutionary response would require significant heritability for which there is no information. However, a requirement for animals to cope with human proximity is a reduction in stress response, a reduction in aggression, and several other changes well known to happen during domestication (Belyaev 1969, 1979; Kohane and Parsons 1986, 1987, 1988; Campler et al. 2009; Trut et al. 2009; Wirén et al. 2009).

What is the duration of such close associations between birds and humans? We can use the barn swallow as an example. Generation time G can be estimated as A + S/(1 − S), with adult survival rate (S = 35%) and age at first reproduction (A = 1 year), from Cramp and Perrins (1977–1994), providing an estimate of 1.59 years per generation. Barn swallows have been breeding in association with humans in Denmark at least since the beginning of the Neolithic or at least 4000 years, as shown by subfossil remains (Møller 1992). This gives approximately 2500 generations, which should provide plenty of time for adaptation to humans, with current populations having more than 99% of nests situated indoors. Another example is the blackbird Turdus merula, which has only been breeding in human proximity for less than 100 years. With an adult survival rate of 56% and age at first reproduction of 1 year, we can estimate generation time as 2.27 years per generation, giving an estimate of 44 generations for evolving a level of 15% of nests being found indoors.

Given the enormous fitness advantage that individual birds breeding indoors enjoy, why do not larger fractions of populations and more species exclusively breeding indoors? One possibility is that there are both costs and benefits of such associations. For example, Partecke et al. (2006) have shown that urban blackbirds have lower levels of corticosterone when subject to a stressful event than rural blackbirds, although other studies of corticosterone responses have not shown such a difference between urban and rural birds (Bonier et al. 2007; Schoech et al. 2007; Fokidis et al. 2009; Heiss and Clark 2009). Successful association with humans may rely on modification of stress responses to humans, allowing for close proximity without disruption of daily routines of birds. Selection experiments on Japanese quail Coturnix japonica have shown that fear responses and hormone levels can change readily (e.g., Gil and Faure 2007), and similar effects are expected to underlie the evolution of domestication (Belyaev 1969, 1979; Kohane and Parsons 1986, 1987, 1988; Campler et al. 2009; Trut et al. 2009; Wirén et al. 2009). Møller (2008) showed that birds from urban populations are less weary of proximity of humans compared with individuals from nearby rural populations. Furthermore, Møller (2009) showed that fearfulness to humans was already lower in rural populations of species that subsequently became urbanized, implying that certain species were preadapted to colonization of urban areas. Finally, species that eventually became urbanized experienced a reduction in the variance in fearfulness, followed by a subsequent expansion of variation in fearfulness among individuals (Møller 2009).

If close association between humans and animals involve physiological adjustments that reduce the costs of frequent disturbance, then a number of predictions can be made. There should be an interspecific association between the level of corticosterone released during exposure to humans and the proportion of individuals breeding indoors. We should also expect that species closely associated with humans show reduced fear reactions and that standardized measures of fearfulness such as the frequency of alarm calls when approached by a human and the distance at which an individual flies away from an approaching human should be reduced in species that have a higher frequency of nests placed indoors.

Most studies of animals that have adapted phenotypically or evolutionarily to human proximity are based on birds, although studies of mammals and invertebrates exist (review in Klausnitzer, 1989). Such changes in phenotype are often described for single populations that have adjusted to human proximity (e.g., Gliwicz et al., 1994), making it difficult to draw general conclusions. However, if the underlying mechanisms for such adaptation are as proposed above, these should be equally applicable to other taxa.

SUPPLEMENTARY MATERIAL

Supplementary material can be found at http://www.beheco.oxfordjournals.org/.

E. Flensted-Jensen, K. Klarborg, and W. C. Årestrup kindly helped find nests. The editor and 2 referees provided constructive comments.

References

Becker
WA
Manual of quantitative genetics
 , 
1984
Pullman (WA)
Academic Enterprises
Belyaev
DK
Domestication of animals
Sci J
 , 
1969
, vol. 
5
 pg. 
47
 
Belyaev
DK
Destabilizing selection as a factor in domestication
J Hered
 , 
1979
, vol. 
70
 (pg. 
301
-
308
)
Bonier
F
Martin
PR
Sheldon
KS
Jensen
JP
Foltz
SL
Wingfield
JC
Sex-specific consequences of life in the city
Behav Ecol
 , 
2007
, vol. 
18
 (pg. 
121
-
129
)
Campler
M
Jongren
M
Jensen
P
Fearfulness in red junglefowl and domesticated White Leghorn chickens
Behav Processes
 , 
2009
, vol. 
81
 (pg. 
39
-
43
)
Cramp
S
Perrins
CM
The birds of the Western Palearctic
 , 
1977
Oxford
Oxford University Press
Diamond
JM
Rapid evolution of urban birds
Nature
 , 
1986
, vol. 
324
 (pg. 
107
-
108
)
Draper
NR
Smith
H
Applied regression analysis
 , 
1981
2nd ed
New York
John Wiley
Felsenstein
J
Phylogenies and the comparative method
Am Nat
 , 
1985
, vol. 
125
 (pg. 
1
-
15
)
Fokidis
HB
Orchinik
M
Deviche
P
Corticosterone and corticosteroid binding globulin in birds: relation to urbanization in a desert city
Gen Comp Endocrinol
 , 
2009
, vol. 
160
 (pg. 
259
-
270
)
Garamszegi
LZ
Møller
AP
Forthcoming
Effects of sample size and intraspecific variation in phylogenetic comparative studies: a meta-analytic review
Biol Rev Camb Philos Soc
 , 
2010
Garland
T
Jr
Harvey
PH
Ives
AR
Procedures for the analysis of comparative data using phylogenetically independent contrasts
Syst Biol
 , 
1992
, vol. 
41
 (pg. 
18
-
32
)
Gil
D
Faure
JM
Correlated response in yolk testosterone levels following divergent selection for social behaviour in Japanese quail
J Exp Zool
 , 
2007
, vol. 
307A
 (pg. 
91
-
94
)
Gliwicz
J
Goszczynski
J
Luniak
M
Characteristic features of animal populations under synurbanization: the case of the blackbird and the striped field mouse
Mem Zool
 , 
1994
, vol. 
49
 (pg. 
237
-
244
)
Glutz von Blotzheim
UN
Bauer
KM
Handbuch der Vögel Mitteleuropas
 , 
1966
Wiesbaden, Germany
Aula-Verlag
Heiss
RS
Clark
AB
Growth and nutritional state of American crow nestlings vary between urban and rural habitats
Ecol Appl
 , 
2009
, vol. 
19
 (pg. 
829
-
839
)
JMP
JMP
 , 
2000
Cary (NC)
SAS Institute Inc
Johnson
DH
Estimating nest success: the Mayfield method and an alternative
Auk
 , 
1979
, vol. 
96
 (pg. 
651
-
661
)
Klausnitzer
B
Verstädterung von Tieren
 , 
1989
Wittenberg-Lutherstadt
Neue Brehm-Bücherei
Kohane
MJ
Parsons
PA
Environment-dependent fitness differences in Drosophila melanogaster: temperature, domestication and the alcohol-dehydrogenase locus
Heredity
 , 
1986
, vol. 
57
 (pg. 
289
-
304
)
Kohane
MJ
Parsons
PA
Mating ability in laboratory-adapted and field-derived Drosophila melanogaster: the stress of domestication
Behav Genet
 , 
1987
, vol. 
17
 (pg. 
541
-
558
)
Kohane
MJ
Parsons
PA
Domestication: evolutionary change under stress
Evol Biol
 , 
1988
, vol. 
23
 (pg. 
30
-
48
)
Lefebvre
L
Whittle
P
Lascaris
E
Finklestein
A
Feeding innovations and forebrain size in birds
Anim Behav
 , 
1997
, vol. 
53
 (pg. 
549
-
560
)
Mayfield
HF
Suggestions for calculating nest success
Wilson Bull
 , 
1975
, vol. 
87
 (pg. 
456
-
466
)
Møller
AP
Nest predation and nest site choice in passerine birds in habitat patches of different size: a study of magpies and blackbirds
Oikos
 , 
1988
, vol. 
53
 (pg. 
215
-
221
)
Møller
AP
Nest site selection across field-woodland ecotones: the effect of nest predation
Oikos
 , 
1989
, vol. 
56
 (pg. 
240
-
246
)
Møller
AP
Nest predation selects for small nest size in the blackbird
Oikos
 , 
1990
, vol. 
57
 (pg. 
237
-
240
)
Møller
AP
Clutch size, nest predation and distribution of avian unequal competitors in a patchy environment
Ecology
 , 
1991
, vol. 
72
 (pg. 
1336
-
1349
)
Møller
AP
The hirundines
Nat Mus
 , 
1992
, vol. 
31
 (pg. 
1
-
32
)
Møller
AP
Flight distance of urban birds, predation and selection for urban life
Behav Ecol Sociobiol
 , 
2008
, vol. 
63
 (pg. 
63
-
75
)
Møller
AP
Successful city dwellers: a comparative study of the ecological characteristics of urban birds in the Western Palearctic
Oecologia
 , 
2009
, vol. 
159
 (pg. 
849
-
858
)
Møller
AP
Interspecific variation in fear responses predicts urbanization in birds
Behav Ecol
 , 
2010
, vol. 
21
 (pg. 
365
-
371
)
Møller
AP
Nielsen
JT
Prey vulnerability in relation to sexual coloration of prey
Behav Ecol Sociobiol
 , 
2006
, vol. 
60
 (pg. 
227
-
233
)
Neter
J
Kutner
MH
Nachtsheim
CJ
Wasserman
W
Applied linear statistical models
 , 
1996
Chicago (IL)
Irwin
Nicolakakis
N
Sol
D
Lefebvre
L
Behavioural flexibility predicts species richness in birds, but not extinction risk
Anim Behav
 , 
2003
, vol. 
65
 (pg. 
445
-
452
)
Nowakowski
JJ
The impact of human presence on the nest distribution of Turdus merula and song trush T. philomelos
Acta Ornithol
 , 
1994
, vol. 
29
 (pg. 
59
-
65
)
Partecke
J
Schwabl
I
Gwinner
E
Stress and the city: urbanization and its effects on the stress physiology in European blackbirds
Ecology
 , 
2006
, vol. 
87
 (pg. 
1945
-
1952
)
Purvis
A
Rambaut
A
Comparative analysis by independent contrasts (CAIC)
Comp Appl Biosci
 , 
1995
, vol. 
11
 (pg. 
247
-
251
)
Rosenthal
R
Meta-analytic procedures for social research
 , 
1991
New York
Sage
Schoech
SJ
Bowman
R
Bridge
ES
Boughton
RK
Baseline and acute levels of corticosterone in Florida scrub-jays (Aphelocoma coerulescens): effects of food supplementation, suburban habitat, and year
Gen Comp Endocrinol
 , 
2007
, vol. 
154
 (pg. 
150
-
160
)
Shochat
E
Warren
PS
Faeth
SH
McIntyre
NE
From patterns to emerging processes in mechanistic urban ecology
Trends Ecol Evol
 , 
2006
, vol. 
21
 (pg. 
186
-
191
)
Sibley
CG
Ahlquist
JE
Phylogeny and classification of birds: a study in molecular evolution
 , 
1990
New Haven (CT)
Yale University Press
Sokal
RR
Rohlf
FJ
Biometry
 , 
1995
3rd ed
New York
Freeman
Tabachnick
BG
Fidell
LS
Using multivariate statistics
 , 
1996
New York
HarperCollins
Trut
L
Oskina
I
Kharlamova
A
Animal evolution during domestication: the domesticated fox as a model
Bioessays
 , 
2009
, vol. 
31
 (pg. 
349
-
360
)
Wirén
A
Gunnarsson
U
Andersson
L
Jensen
P
Domestication-related genetic effects on social behavior in chickens: effects of genotype at a major growth quantitative trait locus
Poultry Sci
 , 
2009
, vol. 
88
 (pg. 
1162
-
1166
)
Yeh
P
Hauber
ME
Price
TD
Alternative nesting behaviours following colonization of a novel environment by a passerine bird
Oikos
 , 
2007
, vol. 
116
 (pg. 
1473
-
1480
)