Northeastern Naturalist Vol. 22, No. 3 B. Heinrich 2015 521 2015 NORTHEASTERN NATURALIST 22(3):521–529 Tree Swallows’ Feather-Lining Their Nest: An Anti-Parasitizing Strategy? Bernd Heinrich* Abstract. - I observed 12 nesting cycles of Tachycineta bicolor (Tree Swallow) with emphasis on the choice of nest materials used, timing of egg-laying, presence and behavior of conspecifics, and nesting success. All Tree Swallows built nests of dried grasses predominantly lined with large, plumed white feathers, but nest-lining behavior did not occur until after egg-laying began. When I experimentally removed feather-linings from one nest, the pair replaced them. In all nests, the birds tucked the feather quills under the eggs with the plumes arching over them. The Tree Swallows showed a strong preference both for shape of feathers and their color contrast from background. The Tree Swallows exhibited vigorous territorial behavior and nest-guarding from conspecifics before nest-building started, continued until early incubation, and then ceased when the young hatched. Birds other than the nest owners were frequent visitors in the nest area and they sometimes entered nests throughout the nesting cycle. Indirect evidence suggests that extra-pair eggs appearing in the nest and “extra” young accounted for nestling mortality . I here infer that although Tree Swallows experience strong nest parasitism, feather-lining behavior hides information on nest contents and thus reduces parasitism. Introduction Many swallows are gregarious and nest in colonies that may be dense (Brown and Brown 1996). Petrochelidon pyrrhonota Vielliot (Cliff Swallow) make clay retorts, a behavior which allows not only a choice of where to nest, but also placement of nests literally on top of one another. In contrast, Tachycincta bicolor Vielliot (Tree Swallow) is a secondary cavity nester, and their ability to nest depends on finding and then possessing often rare pre-existing nest sites. In eastern North America, they now nest primarily in human-provided bird boxes. Nest-site limitations would predictably favor evolution of territorial behavior as well as alternative reproductive strategies, such as extra-pair copulations (EPC) by males and egg-dumping by females that are unable to procure a suitable nest site. In turn, application of these strategies would promote selective pressure for evolution of co unter-strategies. Tree Swallow mating strategies have been studied extensively (Barber et al. 1996, Dunn et al. 2009, Vernier et al. 1993), and the rate of extra-pair mating is among the highest known for any species. In addition, brood parasitism by eggdumping females has also been reported (Lombardo 1988, Whittingham and Dunn 2001b). This study concerns another perhaps unusual phenomenon in these birds: unlike all other birds that use a variety of nesting materials to attach the nest to the substrate, give it form as a fortress, insulate it, attract mates, and/or provide medicinal properties (Heinrich 2010), Tree Swallows nesting in Maine and Vermont make *PO Box 153, Weld, ME 04285; bheinrich153@gmail.com. Manuscript Editor: Jean-Pierre Savard Northeastern Naturalist 522 B. Heinrich 2015 Vol. 22, No. 3 prominent use of white feathers to line their dry grass nests. The purpose of this study was to examine Tree Swallow nest-lining behavior to determine anticipated costs and benefits relevant to possible context of an anti-egg-d umping strategy. Methods I observed 12 Tree Swallow nesting cycles. Most of the observations reported here refer to a 4-ha forest clearing in western Maine. The clearing is surrounded by at least 2 km of unbroken forest in all directions. The 7 Maine nests that I examined in detail occurred sequentially at this site where I was present nearly full-time throughout the nesting periods. I monitored each nest for ~40 days by observing the birds during concentrated sessions of several hours each throughout the period of spring arrival, nest-building, and fledging. Tree Swallows were vocally conspicuous until incubation, and I recorded their seasonal first-arrival dates at the nest site based on visual and aural observations. I also observed in less detail 2 nests that were located in the 1 bird box available in a garden at the edge of a 10-ha beaver marsh in Hinesburg, VT. The Tree Swallow nest boxes opened to the side, and I assessed nest contents by direct observation. In any one year, I restricted all observations to 1 nest site only. My presence was part of the swallows’ environment from the day they arrived in the spring, and they generally ignored me, did not flush from the nest, and usually allowed me to touch them in it. Unless stated otherwise, feathers assessed in nests were those found and brought by the Tree Swallows themselves. Below, I descibe my methods of studying feather choice at 1 nest in relation to the results of the tests. Unless stated otherwise, all of these feathers came from an apparent hybrid of a domestic and Anas platyrhychos L. (Mallard) duck with both white and black feathers. Results Nest features Feathers, when present, were prominently visible in the nest (Fig.1). They were not brought into the nest until after egg-laying had started. They were carried into the nest only during the latter part of the 5–6-egg clutch deposition, but if I removed them, the birds sometimes continued to incorporate feathers during early incubation. During the latter part of the young’s nest-stay, the feathers were no longer visible after they had become pushed down and covered by a layer of feces. At that time, the adults had ceased to enter the nest and fed the young only at the nest entrance. I disassembled 5 nests after the young fledged and counted 111, 110, 46, 40, and 35 feathers in their linings. Of these, 206 were primarily white, and 125 were light-colored or had white markings on them. A large percentage of feathers appeared to be those of Aix sponsa L. (Wood Duck), Branta canadensis L. (Canada Goose), and possibly Ardea herodias L. (Great Blue Heron). At 1 nest, the Tree Swallows also used Meleagris gallopavo L. (Wild Turkey) and Dendragapus obscurus (Say) (Blue Grouse) feathers that I had experimentally spread on the ground within 10 m of the nest. Northeastern Naturalist Vol. 22, No. 3 B. Heinrich 2015 523 Feather-choice experiments The presence of feathers in the nest could be a function of time during the nesting cycle or feather availability. I tested timing at 1 nest by offering swallows Figure 1. Photograph of a nest and nest-lining after the beginning of egg-laying as revealed by a lateral view of a Tree Swallow nest box with the side panel removed to show the plumose feathers arching over the eggs. Northeastern Naturalist 524 B. Heinrich 2015 Vol. 22, No. 3 white plumose feathers (those being most prominent in their nests). The Tree Swallows showed no interest in them until the nest cup of fine grasses appeared to be finished (Fig. 2). Both the male and the female at that nest then began to approach white feathers tossed onto the ground, but they did not pick them up. After the nest contained 2 eggs, they still did not pick them up and there were still no feathers in the nest. I conducted the same test 2 days later; the birds rejected white down feathers, but immediately picked up 10–15-cm white wing-cover feathers and/or snatched them out of the air when I dropped them at test distance of 15 m from the Tree Swallows perched near the nest. The male then took 23 plumose feathers and carried them into the nest to the female who was spending most of the time in the nest box. When the female came out, she picked up 3 feathers and also carried them into the nest box. When I examined the nest, I found that all feathers were arranged with the long, curled plume tops arching over the eggs and the quills anchored under the eggs. I then removed all feathers from the nest to further test feather color choice relative to background color. I spread a white blanket on the ground with 10 white and 10 black feathers spread on it. Within a half-hour, the male took all 10 of the black feathers and 3 of the white ones, again carrying them to the female in the nest. After the male took all the feathers and left the area, I replaced the white blanket with a black tarpaulin, and spread on it the same number and kinds of feathers as before. In a half-hour, the male removed all 10 white feathers and 2 of the black ones. After incorporating the feathers into the nest, both members of the pair left. I examined the nest, found that the fourth egg had just been laid, and rem oved all feathers. After the female laid the fifth egg, the pair still showed occasional interest in feathers that were continuously available to them on the ground near their box, but when I continued to experimentally remove feathers from the nest, the Tree Swallows apparently made less effort to replenish them (Fig. 2). By the 7th day of incubation, the nest contained only 5 feathers and the pair then ignored all feathers offered; no more feathers were taken, even as daytime high temperatures dropped from 30 ºC to 9 ºC, and after a night of constant drizzle as th e cold continued. Figure 2. Feather-collecting during the nesting cycle, as determined by experimental presentation of feathers to the pair near their nest. Northeastern Naturalist Vol. 22, No. 3 B. Heinrich 2015 525 Territoriality In this study in Maine and Vermont, each year there was intense fighting for occupancy of a 4-ha clearing in the forest, and only 1 pair of Tree Swallows nested there despite the presence of sometimes up to 6 available nest boxes. The Tree Swallows usually arrived in the clearing in mid- to late April when there was still snow on the ground. The first birds to return were usually pairs that immediately examined nest boxes. However, nest building did not begin until 2–4 weeks later when warm weather arrived and the snow melted. Nevertheless, wi thin minutes of arrival and weeks before nest building began, the pair took possession of a bird box and attacked other Tree Swallows arriving at, in, or over the clearing’s air space. In the often several routine daily chases, both the pursued and pursuers vanished from my sight, to at least a distance of 300 m. Chase flights sometimes escalated to physical contact during which birds tangled in the air and then fell to the ground. Territory-guarding began at dawn when the pair arrived (apparently from a distant roost) and perched side-by-side where they were highly visible in the top of a tree at the edge of the clearing from which they chased Tree Swallows that came near the clearing. One or both left their perch to chase and attack the intruder. The female usually chased the intruder alone, and the male flew down to the nest box where he remained vocal and perched in the nest-box entrance. However, the Tree Swallows did not forage in the clearing, and so there were also absences from the nest site by one or both members of the pair during feeding flights. As soon as eggs were being laid and before the clutch was complete, the pair took turns relieving each other at the nest, except when a Tree Swallow approached the clearing, in which case the male immediately flew down to perch in the nest e ntrance. In 2014, only a single Tree Swallow arrived in the clearing on 11 April, and left after a brief look into the nest box used the previous year. In the next 2 days, it returned to perch on a branch, but did not examine bird boxes. A pair finally came on 6 May, and immediately examined several of the available bird boxes. A pair was then present for the rest of the nesting season. Starting the next day and during the the following 5 days, there were numerous fights between 2 females that involved aerial grappling in which they fell to the ground. Nest-building did not take place until 20 days later. Until then, several daily incursions of both individuals and pairs of other Tree Swallows continued throughout the nesting cycle. Other females regularly approached the nest entrances of the 2 sequentially active (but not the 4 inactive) nest boxes, and looked into the nest-box entrances. They were tolerated without interference after the breeding pair’s young had hatched. They did not, however, at any time take part in food provisioning of the young. Nest parasitism At 1 nest where 4 eggs were laid from 16 to 20 May, 2 more were added on 2 June to make 6 eggs total, but on 3 June, 1 was then missing (leaving 5 total). Two more were laid in this nest by 7 June for a total of 7 eggs. To test if the Tree Swallows would reject foreign eggs, I added 2 nearly same-sized spotted Hirundo Northeastern Naturalist 526 B. Heinrich 2015 Vol. 22, No. 3 rustica (L.) (Barn Swallow) and Cistothorus palustris (Wilson) (Marsh Wren) eggs. When I examined the nest on 11 June, the 2 foreign eggs had been pushed to the side and the nest contained 1 less Tree Swallow egg. On 17 June, 2 new (translucent) Tree Swallow eggs had been laid in the nest, and the next day a baby hatched from one of the previously laid eggs. On 19 June, the nest contained the baby and 5 swallow eggs, in addition to the foreign eggs still on the periphery of the nest, which appeared to have been re-built. Due to my subsequent absence from the site, I was unable to determine the fate of the nest. After the young fledged at another nest, it still contained 2 eggs that had not hatched, 1 of which contained a partiallydeveloped embryo. Nesting success The time from hatching to fledging in the 2 nests was 17 days (Table 1) during which the young encountered no major weather disturbance. All of these young appeared to grow at the same rate and to the same size. In contrast, at 3 nests that experienced several days of continuous cold rain, time until fledging was 3 days longer. At these nests, some young were either left dead in the nest or were too weak to fly off with the rest of the fledglings (Fig. 3). Also in the last several days before fledging at these nests, only the young that perched in the entrance hole received food; the parents no longer entered the nest box. In one nest where no young fledged, there was a notable size difference among the dead nestlings; they measured 4.9, 6.1, 6.5 and 6.8 cm from tip of bill to tip of ta il. Discussion Birds build a diversity of nests (Hansell 2000) for holding, hiding, protecting and incubating their eggs; and housing their young. Much of nest diversity also involves adaptations to location and availability of materials that dictate or influence attachment to substrate, exterior design, and anti-predator or anti-cuckoldry features (Collias and Collias 1965). Nest linings of different species may serve in cushioning and insulating the eggs. Chickadees’ use of fur, feathers, and plantdown presumably insulates the nest, but their loose nest lining is also used to cover and hide the eggs when they are left temporarily without an adult to attend them Table 1. Progression and fate of nesting by pairs of Tree Swallows in a Maine clearing. In 2013, the original pair deserted its nest after a series of cold rainy days (2013a) and a second pair moved in and nested successfully (2013b). Otherwise only 1 pair nested once per year at this site. 2010 2011 2012 2013a 2013b 2014 Return date - 24 Apr 7 May 15 Apr 11 Apr Nest start - 7 May 12 May 4 May - 13 May First egg 22 May 15 May 18 May 19 May 16 June 28 May Nest feathers at first egg - 0 0 1 0 0 Nest feathers total 110 - 46 - 44 - Eggs (Total/hatched) 6/6 6/6 6/6 0/5 5/5 5/5 Young (total/fledged) 5/6 5/6 6/6 0 0 3/5 Flyers fledged 4 5 6 0 0 3 Northeastern Naturalist Vol. 22, No. 3 B. Heinrich 2015 527 (Smith 1991). In some cases, such as Sturnus vulgaris L. (European Starling), nest linings can also contain aromatic plants that function in controlling ectoparasites (Clark and Mason 1985, Gwinner 1997, Wimberger 1984) and may also be used in mate attraction (Brouwer and Kumdeur 2004). Buteo platypterus (Broad-winged Hawks), insert flat-shaped greens into their nest after the eggs hatch, which maintain a clean surface that may function to retard spoilage of dismembered prey in the nest (Heinrich 2013). Hole-nesters such as woodpeckers build no nests in their excavated tree holes. The role of feathers in the nest lining of Tree Swallows is not obvious from casual inspection. They do not appear to be related to pairing because both males and females bring feathers to the nest; structurally and by their placement in the nest, they are not suited for insulation; they do not have known medicinal or miticidal properties, but rather could pose a risk of infecting the nest. One possible function of these feathers could be to hide the eggs from potential predators. However, the nest is not in the open but sequestered inside at least a shallow hole with a narrow entrance where few predators could or would likely enter. Also, the feather-lining behavior does not precede egg-laying as is usual in other birds (B. Heinrich, pers. observ.). Several considerations converge to suggest that the nest-lining behavior is part of a strategy to reduce egg-dumping by females. Feathers of any color, if large and plumose (such as those preferred by the Tree Swallows) could camouflage the swallows’ white eggs, because they are inserted into the nest with quills down and plumes arching over the eggs. White feathers Figure 3. A just-fledged Tree Swallow at age = 20 days that is too weak to fly and had dropped into the weeds below the nest-box. Its 5 siblings were expert flyers and had left the nest. Northeastern Naturalist 528 B. Heinrich 2015 Vol. 22, No. 3 would camouflage white eggs better than black if the eggs were incompletely covered. As predicted, Tree Swallows preferred the large, curled white feathers, and placed them into the nest in a way that covered the eggs. Tree Swallows nest in generally rare but conspicuous nest sites, making it difficult to conceal nests from potential egg-dumpers. Guarding by constant vigilance near the nest may become necessary to counter nest-parasitism. Nest-guarding by both members of the pair, and especially the attacks by the female on other Tree Swallow females, is intense. Rarity of suitable cavities in appropriate habitat could therefore limit reproduction and would likely provide strong selective pressure for nest parasitism. To be successful, a parasite egg needs to be inserted during the egg-laying cycle of the host. Therefore, it is during egg-laying that the host needs to be attentive and guard against parasitism. For Tree Swallows, this vigilance is particularly difficult given feeding constraints related to egg production. During the period of at least 5–6 days when a female Tree Swallow lays her clutch, she also needs to forage to replenish nutrients, especially because males do not feed their mates. Therefore, the male’s otherwise unusual role in helping to procure rare nest-lining material, functions to reduce his mate’s time spent nest guarding. If the male did not bring feathers, the female would have to hunt for them herself. In addition, that task might cost the male parentage because the female would be exposed to the possibility of extra-pair mating during the search for feathers. The incidence/intensity of Tree Swallows bringing feathers into the nest peaked after initiation of egg-laying, and waned quickly thereafter, whereas the incidence/ intensity of nest defense peaked long before nest-building started, waned during incubation, and ceased by hatching. Such timing suggests that the selective pressure for the addition of feathers to the nest has evolved to camouflage the process of egg-laying. The behavior of hunting for specific kinds of feathers at just the time of egglaying raises questions: Does it account, for example, for the apparent weak mate guarding by males (Chek and Robertson 1994) if the female must hunt far from the nest for her nutrients as well as feathers in her fertile period? Do her wide-ranging travels before egg-laying (Dunn and Whittingham 2005, 2007) relate to food- and feather-searches, instead of for extra-pair matings? Finally, is the high incidence of embryo mortality (Whittingham and Dunn 2001a) related not to egg-laying asynchrony as such, but to egg-discrimination against unwanted “ext ra” eggs instead? Author's addendum note: As this manuscript was going to press, in the summer of 2015, I observed a 13th Tree Swallow nesting. I conducted the following experiment of feather choice ~10 m in front of the nest box. Two 3 m x 5 m tarpaulins, one black and the other white, were spread side by side. Each was at the same time provided with 2 white and 2 black feathers, which I replaced continuously as they were removed and taken by a swallow into the nest box. Only the female took feathers; the male perched within 30 m the entire time. She took 31 white feathers from the black tarp and 4 from the white tarp, and 21 black feathers from the white tarp and 5 from the black tarp. Four of 5 egss hatched, and the young Northeastern Naturalist Vol. 22, No. 3 B. Heinrich 2015 529 fledged successfully without any appearing starved at any time. None came up to the nest entrance except occasionally, starting on the day before fledging. Acknowledgments I thank John Alcock and Jean-Pierre L. Savard for reviewing the manuscript and making helpful suggestions for improving it. Lynn Jennings provided the Blue Grouse feather. Literature Cited Barber, C.A., R.J. Robertson, and P.T. Boag. 1996. The high frequency of extra-pair paternity in Tree Swallows is not an artifact of nest boxes. Behavioral Ecology and Sociobiology 38:425–430. Brouwer, L., and J. Komdeur. 2004. Green nesting material has a function in mate attraction in European Starlings. Animal Behaviour 67:539–48. Brown, C.B., and M.B. Brown. 1996. Coloniality in the Cliff Swallow: The Effect of Group Size and Social Behavior. The University of Chicago Press, Chicago, IL. 566 pp. Chek, A.A., and R.J. Robertson. 1994. Weak mate-guarding in Tree Swallows: Ecological constraint or female control? Ethology 98:1–13. Clark. L., and J.R. Mason. 1985. Use of nest material as insecticide and anti-pathenogenic agents by the European Starling. Oecologia 67:169–176. Collias, N.E., and E.C. Collias. 1964. Evolution of nest building in the Weaverbirds (Ploceidae). University of California Publications in Zoology73:1–2 39. Dunn, P.O., and L.A. Whittingham. 2005. Radio-tracking of female Tree Swallows prior to egg-laying. Journal of Field Ornithology 76:260–264. Dunn, P.O. and L.A. Whittingham. 2007. Search costs influence spatial distribution but not level of extra-pair matings in Tree Swallows. Behavioral Ecology and Sociobiology 61:449–454. Dunn, P.O., J.T. Lifjeld, and L.A. Whittingham. 2009. Multiple paternity and nestling quality in Tree Swallows. Behavioral Ecology and Sociobiology 63:911–922. Gwinner, H. 1997. The function of green plants in nests of European Starlings (Sturnus vulgaris). Behavior 134:337–351. Hansell, M. 2000. Bird Nests and Construction Behavior. Cambridge University Press, Cambridge, UK. 280 pp. Heinrich, B. 2010. The Nesting Season: Cuckoos, Cuckolds, and the Invention of Monogamy. Harvard University Press, Cambridge, MA. 352 pp. Heinrich, B. 2013. Why does a hawk build with green nesting material? Northeastern Naturalist 20:209–218. Lombardo, M.P. 1988. Evidence of intraspecific brood parasitism in the Tree Swallow. Wilson Bulletin 100:126–128. Smith, S.M. 1991. The Black-capped Chickadee: Behavioral Ecology and Natural History. Cornell University Press. Ithaca, NY. Vernier, L., P.O. Dunn, J.T. Liejeld, and R.J. Robertson. 1993. Behavioral patterns of extrapair copulations in Tree Swallows. Animal Behaviour 45:412–415. Whittingham, L.A., and P.O. Dunn. 2001a. Survival of extra-pair and within-pair young in Tree Swallows. Behavioral Ecology:496–500 . Whittingham, L.A., and P.O. Dunn. 2001b. Female responses to intraspecific brood parasitism in the Tree Swallow. Condor 103:166–170. Wimberger, P.H. 1984. The use of green plants in bird nests to avoid ectoparasites. Auk 101:615–618.