Abstract

For many animals, nest construction is a prerequisite for successful breeding. The choice of nesting materials is an important component of nest construction, because material properties can influence nest design and, potentially, reproductive success. Common waxbills are small African finches that select carnivore scat as a material to include in, on, and around their nests. I investigated the hypothesis that scat functions to reduce predation risk by documenting its use in a wild population of common waxbills and by conducting an artificial nest experiment. Among natural nests, scat was present in every nest that hatched young, and parents continued to add scat to nests throughout the nestling period. Among artificial nests, those that received experimental additions of carnivore scat survived at a significantly higher rate than did untreated nests, suggesting that scat functions to reduce predation risk. The mechanism by which nests are protected remains unclear, although it is likely that scat acts as an olfactory deterrent and/or camouflage. Researchers have long focused on the implications of nest site characteristics for avian life-history evolution. Results of the present study suggest that nest materials, similar to nest sites, may influence life histories of nest-building animals by affecting predation risk.

Nests are central to reproduction in a wide range of vertebrate and invertebrate taxa (Hansell, 1984). In birds, nearly every species constructs a nest in which the eggs, and often the young, of breeding adults are contained and protected. In the process of nest-building, individuals must make several decisions: they must choose a nest site, choose nesting materials from their environment, and then arrange those materials in a pattern to form a nest. Previous research suggests that nest site choices influence predation risk and that predation risk, in turn, influences life-history evolution in birds (Martin, 1995; Martin and Li, 1992; von Haartman, 1957). Choices of nest materials and nest designs may also have important consequences for avian life histories. But the evolutionary causes and consequences of those decisions remain relatively unexplored.

Birds regularly use nest materials to build a structural nest layer that accomodates the weight of eggs, incubating parents, and young. However, in many nests, parents add materials to the inner or outer surfaces of the structural layer that appear to provide little or no support to the nest (Hansell, 2000; Hansell and Deeming, 2002). A variety of functions have been proposed for these nonstructural components based on the physical properties of materials and their placement around the nest. Materials may affect breeding success by providing insulation (Lombardo et al., 1995; Reid et al., 2002; Winkler, 1993) or protecting young against parasites and pathogens (Clark and Mason, 1985, 1988; Petit et al., 2002; Wimberger, 1984). Alternatively, materials placed around nests and nestlike structures (e.g., bowers) may provide information about individual quality and influence mating or parental investment decisions (Borgia, 1985; Borgia, 1995; Fauth et al., 1991; Moreno et al., 1994; Soler et al., 1998a).

Nonstructural nest materials may also function to reduce the risk of nest predation. Lichen, moss, and silk materials that many birds use to cover nest exteriors may help to camouflage nests from visually oriented predators (Collias and Collias, 1984; Hansell, 1996). In addition, Bolles (1890) hypothesized that the snake skins gathered by many taxa (e.g., Myiarchus spp.: Baicich and Harrison, 1997; Kattan et al., 2000; Lanyon, 1997; Ptilorus victoriae: Frith and Frith, 1995; Ochetorhynchus certhioides: Zyskowski and Prum, 1999; Lamprotornis nitens: Schuetz JG, personal observation; and see Strecker, 1927) reduce the risk of nest predation by scaring off potential predators. Despite the diversity of functions ascribed to nonstructural nest materials, few studies have experimentally demonstrated the fitness consequences of their use and debate continues over the utility of specific materials, such as green plants (Brouwer and Komdeur, 2004; Clark and Mason, 1988; Fauth et al., 1991; Gwinner, 1997; Petit et al., 2002; Wimberger, 1984;).

Common waxbills (Estrilda astrild) are small African finches (Family: Estrildidae) that gather carnivore scat (i.e., fecal material that often contains indigestible hair and bones of vertebrate prey) and place it in, on, and around their nests. Based on observations of captive waxbills, Goodwin (1982) suggested that such materials serve to camouflage nests or to divert predators away from nests. In this study, I assessed the use of carnivore scat by wild common waxbills throughout the nesting cycle and conducted an artificial nest experiment in order to evaluate the specific hypothesis that scat gathered by common waxbills reduces the risk of nest predation.

METHODS

Study species and study site

Common waxbills occur throughout much of sub-Saharan Africa, inhabiting a wide range of grassland and woodland habitats. In southern Africa, socially monogamous pairs breed throughout much of the year (September–June; Maclean, 1993), presumably in response to the timing of regional rainfall and the subsequent availability of seeds. Common waxbills build a nearly spherical nest chamber of grass stems on or near the ground. Entry to the nest chamber is achieved through a stiff grass tunnel that often opens onto a small bare spot of earth or rock. The top of the nest frequently accommodates another structure, the “cock's nest,” the function of which has been debated (Goodwin, 1982).

I studied scat use by common waxbills at two sites (Hilton and Midmar) in the midlands of KwaZulu-Natal, South Africa, during 1999–2000 and 2001–2002. The Hilton site was located on a forest plantation owned by the Mondi Forest Company (29°33′ S, 30°17′ E), and the Midmar site straddled three privately owned farms (29°32′ S, 30°03′ E). The sites were of comparable elevation (approximately 1000 m) and were dominated by expanses of grasses (e.g., Paspalum, Setaria, Eragrostis) bordered with stands of Pinus, Eucalyptus, or Acacia trees. The study species was common at both sites.

Description of scat use by common waxbills

I searched for nests during the 1999–2000 and 2001–2002 breeding seasons by walking through suitable nesting habitat and observing the behavior of common waxbills. Nests were also found by inspecting appropriate vegetation without specific cues from the birds. I revisited nests soon after I expected their clutches to be complete. To avoid causing abandonment of nests by parents, I did not visit nests regularly during the incubation period. Nests were revisited on the expected hatch day and every 3 days after hatching to assess nest success and nestling survival. Throughout the 2001–2002 season, I scored the amount of scat present in, on, or around the nest chamber during each nest visit (scat score: 0, none; 1, a few pieces; 2, some; and 3, many pieces).

Artificial nest experiment

I manufactured artificial nests using domed bamboo nest baskets frequently used by aviculturalists for breeding finches. Nest baskets were comparable in size (12 × 9 cm; entrance, 3 × 4.5 cm) to common waxbill nests and were similar in having a closed nest chamber rather than an open cup. All nests were lined with grass typical of that used in common waxbill nests. Over several weeks, I collected carnivore scat similar to that I had seen gathered by waxbills and stored it in a freezer. The shape and size of the scat, its location, and the prevalence of rodent fur and bones contained within suggested that the majority of scat was produced by serval (Felis serval), although feral cats (F. domesticus) and African wild cats (F. sylvestris lybica) also occurred in the area. Just before the experiment, collected scats were thawed, homogenized, and divided into packets of roughly equal volume, though masses varied (mean ± SD, 6.71 ± 1.46 g, N = 41).

I placed a total of 82 artificial nests in six transects at the Hilton study area. Transects 1 and 2 were separated by a small stream, were more than 50 m from each other, and were more than 300 m from any other transects. Transects 3 and 4 were separated by a stream, were more than 50 m from each other, and were 200 m from any other transect. Transects 5 and 6 were situated 75 m from each other and 200 m from transects 1 and 2. All transects were situated in habitat that was suitable for common waxbill nesting, and I chose appropriate nest sites (i.e., I placed nests on the ground, nestled between or against vegetation) within a 50-cm radius of each 25 m point along transects. Thus, within transects, nests were positioned from 24 m to 26 m apart.

Transects 1–4 were established and visited between 22 February and 11 March 2000, and transects 5 and 6 were established and visited between 14 March and 1 April 2000. Timing of the experiment coincided with the latter part of the waxbill breeding season in KwaZulu-Natal. The absence of experimental nests early in the breeding season is unlikely to bias results, because predation risk at natural nests is constant throughout the breeding season (Schuetz JG, unpublished data).

Two days after establishing transects, I placed scat inside, on top of, and at the entrance of every other nest in a pattern similar to that found in wild nests. The amount of scat used was comparable to that on natural nests receiving a scat score of three. I then baited all nests with two society finch (Lonchura striata) eggs that I acquired from a South African aviculturalist. The aviculturalist had removed the eggs from society finch parents within a day of their being laid and kept them refrigerated until I collected them for the experiment. Because the eggs were never incubated, embryos did not undergo development and were incapable of experiencing pain and suffering. The eggs would have been discarded or boiled and used as “dummy” eggs had I not asked to use them in the present study. Society finch eggs (15.7 × 11.5 mm) are only slightly larger than common waxbill eggs (12.3 × 9.6 mm) and are similar in their white coloration. As a result, predator responses to society finch and common waxbill eggs are unlikely to differ.

I censused the contents of each nest 2, 4, 8, 12, and 16 days after baiting them. The 16-day trial period was comparable to the total time that common waxbill eggs are exposed to predators in natural nests. I also recorded vegetation height around each nest and accessibility to the eggs in an effort to account for environmental variation that might affect predation risk. Vegetation height was evaluated as the average height of all vegetation in a 50-cm radius around the nest and was scored as one (less than 50 cm), two (50–100 cm), or three (more than 100 cm). Vegetation directly in front of the nest entrance was characterized as presenting one (little or no obstruction), two (moderate obstruction), or three (substantial obstruction) to potential predators. I interpreted the disappearance or destruction of one or both eggs as a predation event (results did not differ if I inferred predation only when both eggs disappeared). With the exception of the initial scat treatment, I handled and checked all nests in the same manner on each nest visit.

Statistical analyses

I analyzed data on success of artificial nests by using product-limit survival analysis with scat-treated and untreated nests forming two groups. If nests were depredated, I recorded the day of the trial period on which I discovered the egg(s) were missing. If nests remained intact over the 16-day trial period, I recorded them as surviving until day 16 and censored those data (censoring allows for the inclusion of cases in which predation did not occur by the end of the observation period; for discussion of survival analysis, see Allison, 1995; SAS, 2002a). I also investigated the effects of vegetation height, nest obstruction, and time of breeding season on nest success using proportional hazards. To assess whether predation risk changed over the 16-day trial period, I calculated the proportion of intact nests depredated during each interval and plotted values at the midpoint of each interval. All analyses were performed in JMP 5.0 (SAS, 2002b).

RESULTS

Description of scat use by common waxbills in natural nests

Scat was present (scat score ≥ 1) on 23.9% (11/46) of nests that were found or visited before egg-laying and on 91.9% (57/62) of nests found or visited during the egg stage. In all 67 nests that reached the nestling stage, scat was present inside the nest chamber, on the outer surface of the nest, or on the ground near the nest entrance (Figure 1). The increased prevalence of carnivore scat in waxbill nests throughout the nesting period was not owing to biased subsampling of nests containing scat. When controlling for repeated observations on individual nests, there was a significant positive association between scat scores and the number of days passed since clutch initiation (mixed model regression: F1,338 = 275.04, p < .0001, random effect = nest, N = 87 nests), indicating that parents add scat to nests throughout the egg and nestling periods (Figure 2). Although nests that successfully fledged young eventually contained more scat than did those that were abandoned or depredated (Wilcoxon sign-rank test: χ2 = 4.397, p = .036, df = 1, N = 87), there was no detectable difference between successful and unsuccessful nests in the rate at which they accumulated scat (mixed model regression: F1,337 = 0.269, p = .60, N = 87 nests). Furthermore, there was no association between the number of common waxbill young fledging from successful nests (i.e., those that produced at least one chick) and the rate of scat accumulation on those nests (mixed model regression: F1,162 = 0.960, p = .33, N = 30 nests).

Figure 1

Proportions of nests found or visited during nest-building (N = 46), egg-laying (N = 62), and nestling stages (N = 67) that contained at least some scat (scat score ≥ 1).

Figure 1

Proportions of nests found or visited during nest-building (N = 46), egg-laying (N = 62), and nestling stages (N = 67) that contained at least some scat (scat score ≥ 1).

Figure 2

Accumulation of carnivore scat on natural nests. I scored the amount of scat present in, on, or around the nest chamber during each nest visit (scat score: 0, none; 1, a few pieces; 2, some; 3, many pieces) and standardized observations with respect to clutch initiation (day = 0) of each nest. Sample sizes at overlapping points are indicated by circle diameters (range = 1–16; N = 426 scat scores, 87 nests).

Figure 2

Accumulation of carnivore scat on natural nests. I scored the amount of scat present in, on, or around the nest chamber during each nest visit (scat score: 0, none; 1, a few pieces; 2, some; 3, many pieces) and standardized observations with respect to clutch initiation (day = 0) of each nest. Sample sizes at overlapping points are indicated by circle diameters (range = 1–16; N = 426 scat scores, 87 nests).

Artificial nest experiment

Four nests were removed from the analysis of nest survival because they were stolen before the first census. For the remaining 78 nests, those treated with scat survived significantly better than did those that were left untreated (product-limit survival, log-rank test between groups: χ2 = 5.637, p = .018, df = 1, N = 78) (Figure 3). On average, scat-treated nests lasted 14.8 days before being depredated, whereas untreated nests lasted only 12.4 days (Wilcoxon sign-rank test: χ2 = 6.713, p = .010, df = 1, N = 78). Survival of nests was not affected by vegetation height around nests (likelihood ratio test: χ2 = 1.562, p = .21, df = 1, N = 78) or perceived obstruction to the nest entrance (likelihood ratio test: χ2 = 0.057, p = .81, df = 1, N = 78). In addition, there was no seasonal change in predation risk for artificial nests. Survival of nests baited in February (transects 1–4) was no different than survival of nests baited in March (transects 5,6; likelihood ratio test: χ2 = 0.014, p = .91, df = 1, N = 78).

Figure 3

Survival of artificial nests. The proportions of scat-treated nests (dashed line, N = 39) and untreated nests (solid line, N = 39) that remained intact (i.e., no eggs were removed by predators) during the artificial nest experiment. The experiments lasted 16 days from when nests were baited with society finch eggs (day = 0) to the final nest check (day = 16).

Figure 3

Survival of artificial nests. The proportions of scat-treated nests (dashed line, N = 39) and untreated nests (solid line, N = 39) that remained intact (i.e., no eggs were removed by predators) during the artificial nest experiment. The experiments lasted 16 days from when nests were baited with society finch eggs (day = 0) to the final nest check (day = 16).

The distribution of predation risk over the course of the 16-day experiment appeared to differ between treatment groups (Figure 4). Nests that were treated with scat seemed to suffer increased predation risk as they aged (linear regression: R2 = .777, F1,3 = 10.471, p = .048). The same was not true of untreated nests, which showed consistent predation risk throughout the 16-day trial period (linear regression: R2 = .345, F1,3 = 1.645, p = .29). When the data are grouped, however, the apparent interaction between treatment and age is not significant, possibly owing to small sample size (N = 5 intervals).

Figure 4

Hazard rates of artificial nests. Data points represent the proportion of intact artificial nests (i.e., those containing two eggs) that were depredated between nest visits. Filled circles indicate artificial nests that did not receive scat. Open circles and dashed regression line indicate artificial nests that received the scat treatment. The risk of scat-treated nests being depredated increased significantly (p = .048) over the course of the experiment. The number of nests used to generate each data point decreased throughout the trial as nests were depredated (scat-treated nests, N = 39, 38, 38, 37, 31 nests intact at the beginning of each interval; untreated nests, N = 39, 35, 34, 29, 21 nests intact at the beginning of each interval).

Figure 4

Hazard rates of artificial nests. Data points represent the proportion of intact artificial nests (i.e., those containing two eggs) that were depredated between nest visits. Filled circles indicate artificial nests that did not receive scat. Open circles and dashed regression line indicate artificial nests that received the scat treatment. The risk of scat-treated nests being depredated increased significantly (p = .048) over the course of the experiment. The number of nests used to generate each data point decreased throughout the trial as nests were depredated (scat-treated nests, N = 39, 38, 38, 37, 31 nests intact at the beginning of each interval; untreated nests, N = 39, 35, 34, 29, 21 nests intact at the beginning of each interval).

The proportion of scat-treated nests surviving the 16-day trial was similar to the proportion of natural nests surviving an egg-laying and incubation period of equal duration (χ2 = 0.360, p = .55, df = 1, N = 106). There was a trend for artificial nests without scat to have worse survival than natural nests over the same period (χ2 = 2.961, p = .087, df = 1, N = 106).

DISCUSSION

The function of carnivore scat in common waxbill nests

Use of carnivore scat as a nonstructural nest material was widespread in the common waxbill populations that I studied. Every nest that successfully hatched young contained at least some scat, and waxbills continued to gather the material throughout the nesting period. In support of Goodwin's (1982) antipredator hypothesis, artificial nests that were treated with scat survived significantly better than did nests without scat. Patterns of predation at artificial nests could not be accounted for by differences in vegetation among nest sites or by changes in the seasonal risk of predation. Predation rates at natural nests and artificial nests treated with scat were not significantly different, suggesting that my experiment accurately characterized the behavior of waxbill nest predators. Together, these results support the hypothesis that selection to reduce the risk of nest predation has shaped nest material choices of common waxbills.

Other adaptive explanations for the use of carnivore scat seem less plausible in this system, though my findings do not specifically preclude them. The scat collected by common waxbills contains a significant amount of mammal fur, a material that some birds appear to use as an insulative nest lining (Hansell, 2000; Skowron and Kern, 1980). However, the capability for scat to influence the microclimate of waxbill nests seems limited by its placement around the nest. Only material included in the nest chamber is likely to affect nest temperatures, and this amount is typically smaller than the amount placed outside the nest chamber. Alternatively, the ability to collect scat, which may be a relatively rare material in the environment, could be interpreted as a sexually selected signal between actual or potential mates (Soler et al., 1998b). However, both males and females collect scat throughout egg and nestling development, well after pairing has occurred and the parentage of offspring has been determined. In addition, scat scores on natural nests were not associated with fledgling numbers as might be expected in a sexual selection scenario, and waxbills never used scat in sexual or nesting-intention displays even though other materials are frequently used in this context (Goodwin, 1982; Schuetz JG, personal observation).

Mechanisms of nest protection

Identifying nest predators will be important for determining the mechanism(s) by which scat conveys protection against predation. Both natural and artificial nests showed little if any disturbance when depredated, suggesting that most nest failures were owing to small rodents and snakes capable of passing through the narrow entrances to nest chambers. Depending on the specific predators involved, scat may cause avoidance of nests or reduce the detectability of nests. Numerous studies show that rodents actively avoid the chemical cues associated with cat feces and urine (see Drickamer et al., 1992; for review, see Kats and Dill, 1998), presumably to reduce their risk of being eaten. Odors associated with scat may also mask the smell of nests and make it difficult for predators to locate eggs or young. If scat acts either as an olfactory deterrent or as a form of olfactory camouflage, we could account for the observations that untreated nests suffer constant risk and scat-treated nests suffer increasing risk (Figure 4) with one simple assumption: scat loses its olfactory efficacy over time. Given that artificial nests were treated with scat only at the beginning of a 16-day trial, this hypothesis seems reasonable and may help to explain why common waxbills add scat to natural nests throughout the entire nesting cycle. Future work describing the chemical constituents of scat and their physical properties will be of considerable value in determining the mechanisms of nest protection.

Nests materials and life-history evolution

The evolutionary consequences of scat use by common waxbills are interesting to consider because variation in the risk of nest predation appears to be linked with variation in avian life histories (see Ghalambor and Martin, 2002). Comparative analyses suggest that relatively high predation rates select for shorter incubation (Martin, 2002) and nesting periods (see Ricklefs, 1984), faster nestling growth (Remes and Martin, 2002), and earlier nestling development (Bosque and Bosque, 1995; Remeš and Martin, 2002). All of these traits interact with survival and reproduction strategies of parents (Lack, 1968; O'Connor, 1984; Stearns, 1992).

Compared with other southern African birds, common waxbills exhibit a peculiar ecology and life history that may have originated, or may persist, as a result of scat effectively reducing predation risk. Unlike the vast majority of altricial taxa that breed in the region (85.3% of 587 species), common waxbills nest on the ground (Schuetz JG, unpublished data extracted from Maclean, 1993). And despite their small size, common waxbill nestlings grow and develop relatively slowly compared with nestlings of other ground-nesting species (Schuetz JG, unpublished data extracted from Maclean, 1993). If scat use has contributed to the evolution or persistence of this unusual combination of traits, we would expect ground-nesting species in the region that do not use scat to experience higher rates of nest predation than those experienced by common waxbills.

Results of the present study complement previous work on the relationship between nests and life-history evolution (Martin, 1995; Martin and Li, 1992; von Haartman, 1957) by demonstrating that the choice of nest materials, not just nest sites, can affect the success of nesting birds. We will only be able to assess the relative importance of nest material choices for life-history evolution after documenting the use and function of nest materials in a wide range of taxa. However, future studies that account for the fitness consequences of nest material choices are likely to provide a richer view of the relationship between nest structures and life-history evolution in birds.

This work was supported by an National Science Foundation Graduate Fellowship and a travel grant from the Einaudi Center at Cornell University. I would like to thank the KwaZulu-Natal Nature Conservation Society for assistance throughout the project. The Mondi Forest Company at Mountain Home as well as the Bate, Nash, and Hanbury families graciously allowed me to work on their properties. T. Bruce, S. Burton, L. Merrill, M. Swett, B. Taft, and L. Yang provided excellent help in the field. Willie Evans was kind enough to provide the society finch eggs. L. Chan, D. Hawley, G. Langham, B. Safran, D. Winkler, and anonymous reviewers offered helpful comments on the manuscript.

References

Allison PD,
1995
. Survival analysis using the SAS system. Cary, North Carolina: SAS Institute.
Baicich PJ, and Harrison CJO,
1997
. A guide to the nests, eggs, and nestlings of North American birds. San Diego, California: Academic Press.
Bolles F,
1890
. Snake skins in the nests of Myiarchus crinitus.
Auk
 
7
:
288
.
Borgia G,
1985
. Bower quality number of decorations and mating success of male satin bowerbirds Ptilonorhynchus violaceus: an experimental analysis.
Anim Behav
 
33
:
266
–271.
Borgia G,
1995
. Complex male display and female choice in the spotted bowerbird: specialized functions for different bower decorations.
Anim Behav
 
49
:
1291
–1301.
Bosque C, and Bosque MT,
1995
. Nest predation as a selective factor in the evolution of developmental rates in altricial birds.
Am Nat
 
145
:
234
–260.
Brouwer L, and Komdeur J,
2004
. Green nesting material has a function in mate attraction in the European starling.
Anim Behav
 
67
:
539
–548.
Clark L, and Mason JR,
1985
. Use of nest material as insecticidal and anti-pathogenic agents by the European starling.
Oecologia
 
67
:
169
–76.
Clark L, and Mason JR,
1988
. Effect of biologically active plants used as nest material and the derived benefit to starling nestlings.
Oecologia
 
77
:
174
–180.
Collias NE, and Collias EC,
1984
. Nest building and bird behavior. Princeton, New Jersey: Princeton University Press.
Drickamer LC, Mikesic DG, and Shaffer KS,
1992
. Use of odor baits in traps to test reactions to intra- and interspecific chemical cues in house mice living in outdoor enclosures.
J Chem Ecol
 
18
:
2223
–2250.
Fauth PT, Krementz DG, and Hines JE,
1991
. Ectoparasitism and the role of green nesting material in the European starling.
Oecologia
 
88
:
22
–29.
Frith CB, and Frith DW,
1995
. Notes on the nesting biology and diet of Victoria's riflebird Ptilorus victoriae.
Emu
 
95
:
162
–174.
Ghalambor CK, and Martin TE,
2002
. Comparative manipulation of predation risk in incubating birds reveals variability in the plasticity of responses.
Behav Ecol
 
13
:
101
–108.
Goodwin D,
1982
. Estrildid finches of the world. Ithaca, New York: Cornell University Press.
Gwinner H,
1997
. The function of green plants in nests of European starlings Sturnus vulgaris.
Behaviour
 
134
:
337
–351.
Hansell MH,
1984
. Animal architecture and building behaviour. London: Longman.
Hansell MH,
1996
. The function of lichen flakes and white spider cocoons on the outer surface of birds' nests.
J Nat Hist
 
30
:
303
–311.
Hansell MH,
2000
. Bird nests and construction behaviour. Cambridge: Cambridge University Press.
Hansell MH, and Deeming DC,
2002
. Location, structure and function of incubation sites. In: Avian incubation: behaviour, environment, and evolution (Deeming DC, ed.) Oxford: Oxford University Press; 8–27.
Kats LB, and Dill LM,
1998
. The scent of death : chemosensory assessment of predation risk by prey animals.
Ecoscience
 
5
:
361
–394.
Kattan GH, Alvarez-Lopez H, Gomez N, and Cruz L,
2000
. Notes on the nesting biology of the apical flycatcher, a Colombian endemic.
J Field Ornithol
 
71
:
612
–618.
Lack DL,
1968
. Ecological adaptations for breeding in birds. London: Methuen.
Lanyon WE,
1997
. Great Crested Flycatcher (no. 300) In: The birds of North America (Poole A, Gill F, eds.) Philadelphia: The Birds of North America; 1–20.
Lombardo MP, Bosman RM, Faro CA, Houtteman SG, and Kluisza TS,
1995
. Effect of feathers as nest insulation on incubation behavior and reproductive performance of tree swallows (Tachycineta bicolor).
Auk
 
112
:
973
–981.
Maclean GL,
1993
. Roberts' birds of southern Africa. Cape Town: The Trustees of the John Voelcker Bird Book Fund.
Martin TE,
1995
. Avian life-history evolution in relation to nest sites, nest predation, and food.
Ecol Monogr
 
65
:
101
–127.
Martin TE,
2002
. A new view of avian life-history evolution tested on an incubation paradox.
Proc R Soc Lond B
 
269
:
309
–316.
Martin TE, and Li PJ,
1992
. Life-history traits of open-nesting vs. cavity-nesting birds.
Ecology
 
73
:
579
–592.
Moreno J, Soler M, Møller AP, and Linden M,
1994
. The function of stone carrying in the black wheatear, Oenanthe leucura.
Anim Behav
 
47
:
1297
–1309.
O'Connor RJ,
1984
. The growth and development of birds. New York: Wiley.
Petit C, Hossaert-McKey M, Perret P, Blondel J, and Lambrechts MM,
2002
. Blue tits use selected plants and olfaction to maintain an aromatic environment for nestlings.
Ecol Lett
 
5
:
585
–589.
Reid JM, Cresswell W, Holt S, Mellanby RJ, Whitfield DP, and Ruxton GD,
2002
. Nest scrape design and clutch heat loss in pectoral sandpipers (Calidris melanotos).
Func Ecol
 
16
:
305
–312.
Remeš V, and Martin TE,
2002
. Environmental influences on the evolution of growth and development rates in passerines.
Evolution
 
56
:
2505
–2518.
Ricklefs RE,
1984
. The optimization of growth rate in altricial birds.
Ecology
 
65
:
1602
–1616.
SAS,
2002
a. JMP statistics and graphics guide. Cary, North Carolina: SAS Institute Inc.
SAS,
2002
b. JMP, version 5.0. Cary, North Carolina: SAS Institute Inc.
Skowron C, and Kern M,
1980
. The insulation in nests of selected North American songbirds.
Auk
 
97
:
816
–824.
Soler JJ, Cuervo JJ, Møller AP, and De Lope F,
1998
a. Nest building is a sexually selected behaviour in the barn swallow.
Anim Behav
 
56
:
1435
–1442.
Soler JJ, Møller AP, and Soler M,
1998
b. Nest building, sexual selection and parental investment.
Evol Ecol
 
12
:
427
–441.
Stearns SC,
1992
. The evolution of life histories. Oxford: Oxford University Press.
Strecker JK,
1927
. Birds and snake-skins. In: Contributions from Baylor University Museum. Waco, Texas: Baylor University 11:1–12.
von Haartman L,
1957
. Adaptation in hole-nesting birds.
Evolution
 
11
:
339
–347.
Wimberger PH,
1984
. The use of green plant material in bird nests to avoid ectoparasites.
Auk
 
101
:
615
–618.
Winkler DW,
1993
. Use and importance of feathers as nest lining in tree swallows (Tachycineta bicolor).
Auk
 
110
:
29
–36.
Zyskowski K, and Prum RO,
1999
. Phylogenetic analysis of the nest architecture of neotropical ovenbirds (Furnariidae).
Auk
 
116
:
891
–911.