Nest ornamentation in blue tits: is feather carrying ability a male status signal?

Behavioral Ecology doi:10.1093/beheco/arq199 Advance Access publication 20 December 2010 Original Article Nest ornamentation in blue tits: is feather carrying ability a male status signal? Juan José Sanz a,b and Vicente García-Navas b a Departamento de Ecología Evolutiva, Museo Nacional de Ciencias Naturales (CSIC), Calle José Gutiérrez Abascal 2, 28006 Madrid, Spain and b Área de Zoología, Departamento de Ciencias Ambientales, Facultad de Medio Ambiente, Universidad de Castilla-La Mancha, Avenida Carlos III s/n, 45071 Toledo, Spain In a wide range of animal taxa, males build elaborate mating structures to attract females. For instance, some bird species use their nests or display sites as nonbodily ornaments (extended phenotypes) in intersexual communication. Recently, it has been proposed that feather carrying constitutes a postmating sexual display. Here, evidence supporting this hypothesis is presented for a species in which such behavior had not been previously reported, the blue tit Cyanistes caeruleus. In our Mediterranean study populations, males carry feathers to the nest and place them outside the nest cup. Here, we show experimentally and by means of observational data that feather carrying by male blue tits induced their mates to increase their clutch size. In addition, males from ornamented nests fed their offspring at a higher rate than those from nests in which this behavior was not detected. Feathercarrying effort was positively associated with male tarsus length. On the other hand, females paired with males engaging in this activity reduced their feeding rates but obtained direct fitness benefits: Ornamented nests resulted in more fledged young with better body condition than those without feathers carried by males. In summary, this study suggests that male feather-carrying ability constitutes an honest signal of their parental quality and a good predictor of breeding success in these blue tit populations. Key words: Cyanistes caeruleus, extended phenotype, feather carrying, Mediterranean, nest building, ornament, parental investment, Paridae, sexual selection, signaling. [Behav Ecol 22:240 247 (2011)] The use of extended phenotypes (i.e., nonbody ornaments) as signals is a widespread phenomenon across many animal taxa (fish: e.g., Barber et al. 2001, insects: e.g., LeBas et al. 2004, mammals: e.g., Sliwa and Richardson 1998, see Schaedelin and Taborsky 2009 for a review). In birds, elaborate structures that are used as nests or court areas are the most common nonbodily signals. In some species, the nest-building behavior plays a major role in mate choice because birds (mainly males) can use the time spent building nests (magpie Pica pica, De Neve and Soler 2002; penduline tit Remiz pendulinus, Szentirmai et al. 2005), the number of nests constructed (wren Troglodytes troglodytes, Evans and Burn 1996; Australian reed warbler Acrocephalus australis, Berg et al. 2006) or nest size (magpie, Soler et al. 2001; penduline tit, Hoi et al. 1994; Szentirmai et al. 2005; rufous bush robin Cercotrichas galactotes, Palomino et al. 1998; barn swallow Hirundo rustica, Soler, Cuervo, et al. 1998, 2007), to indicate their parental quality, experience, or genetic quality and thereby attract partners. Fascinating examples of intricate nest-building abilities used by males as intersexual signals to fertile females include the richly decorated structures (bowers) built by many members of the family Ptilonorhynchidae (e.g., Borgia 1985; Madden 2003). Male bowerbirds use colored objects such as berries, flowers, feathers, and glass to decorate their bowers as part of their courtship display (e.g., Borgia et al. 1987). This behavior, that is, the addition of odd materials to the nest to attract females, has also been reported in several other avian groups. Address correspondence to V. García-Navas. E-mail: vicente. garcianavas@uclm.es. Received 11 August 2010; revised 25 October 2010; accepted 4 November 2010. Ó The Author 2010. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com Nest ornamentation can consist of flower petals (red-backed fairy-wren Malurus melanocephalus, Karubian and Alvarado 2003), green material (European and spotted starling Sturnus unicolor and S. vulgaris, Gwinner 1997; Brouwer and Komdeur 2004; Veiga et al. 2005), or stones (black wheatear Oneanthe leucura, Moreno et al. 1994). The number, size, or rarity of these items could provide females with information on male condition or willingness to invest in reproduction (Soler, Møller, et al. 1998). On the other hand, maternally built nests are less common in comparison with paternally and biparentally built nests (Hansell 2000). The blue tit, Cyanistes caeruleus, belongs to this minority group of species in which females alone build the nests, which excludes the possibility that nest building per se can be used as a signal by females. The blue tit nest consists of a pad of moss often mixed with straw or dead grass at the base and a nest cup lined with fine dry grass and other materials, such as hair, wool, or bark shavings (Cramp 1998). In some cases, females actively incorporate fresh fragments of green plants at the end of nest construction until fledging (e.g., Mennerat et al. 2009). In our study area, we have observed that blue tits usually adorn their nests with flight feathers, a trait much less common in other Iberian populations (Sanz JJ and García-Navas V, personal observation). Feathers are not used as nest-building material (nest lining) because they are placed outside of the nest cup (Figure 1a,b). Sometimes, feathers are stuck into the nest with the plume upward or placed in a corner of the nest-box, which enhances the conspicuousness of this ornament (Sanz JJ and García-Navas V, personal observation). Hence, this puzzling behavior is unlikely to have a thermoregulatory purpose as occurs in other species (the insulating function hypothesis: e.g., Møller 1984, 1987; Lombardo et al. 1995; McGowan et al. 2004).

Sanz and García-Navas Nest ornamentation by male blue tits 241 Figure 1 (a) Unmanipulated nest of blue tit with a covert feather of wood pigeon added in a natural way prior to laying. (b) Top view of a natural blue tit nest decorated with feathers belonging to different species (such as red-legged partridge or chaffinch). Feathers are placed surrounding the nest cup (like a border) that makes them very perceptible from the nest-box entrance. In both cases, it should be noted that feathers are arranged outside of the nest cup that prevents them from coming into direct contact with the eggs or the incubating female. As far as we know, only a few studies have addressed the use of foreign feathers with a signaling function (Borgia and Gore 1986; Frith CB and Frith DW 1990; Veiga and Polo 2005; Polo and Veiga 2006) despite the fact that feathers are an important nest-building material for many bird species (Hansell 2000). Polo and Veiga (2006) reported in a study on starlings that females use flight feathers to decorate the nest in response to a male display (carrying of green plant material) and that the amount of feathers carried was related to female reproductive experience and laying date, 2 variables correlated with female body condition. However, the use of unusual materials during nest building is more common in males, the sex that typically exhibits ornaments and other secondary sexual characters (Andersson 1994). The blue tit is not an exception; feather-carrying display is carried out by the male (see RESULTS). According to the sexual selection theory, if the nest itself or the nest-building behavior conveys information on male reproductive condition, this trait could be used by females (which usually invest more than males in reproduction) to assess the relative quality of their partner and modulate their parental effort (Burley 1986). Thus, if the carrying of feathers to the nest is a good indicator of male quality, we expect this signal to prompt females to invest more in reproduction, trading off between current and future reproductive attempts and maximizing overall fitness. This was tested by employing data from both a descriptive and an experimental approach. First, we conducted an observational study on the use of feathers across a breeding season; second, we documented possible bias in feather selection with the aim to discern whether blue tits show preference for a particular color or not; third, we report correlational evidence of the effect of feather ornamentation on certain reproductive parameters; and fourth, we experimentally manipulated the presence of feathers in blue tit nests. MATERIALS AND METHODS Study areas The study was carried out at the Cabañeros National Park (Ciudad Real and Toledo Provinces, central Spain) and Quintos de Mora (Toledo Province, central Spain), a government-owned game reserve. In Cabañeros, a total of 350 wooden nest-boxes distributed in 2 nest-box plots were monitored during 3 consecutive years (2006 2008). The first plot (Anchurones, 150 nest-boxes) was located in a floodplain dominated by holm and cork oaks (Quercus rotundifolia and Q. suber) scattered in a savannah-like configuration. The second nest-box plot (El Brezoso, 100 nest-boxes) was settled in a riverbank where the Pyrenean oak (Q. pyrenaica) is the prevalent species. In Quintos de Mora, we managed 200 and 280 nest-boxes during the breeding seasons of 2008 and 2009, respectively. Nest-boxes were distributed in 2 Pyrenean oak forests with opposite exposure (Gil García and Valdeyerno, 100 nest-boxes in each one) to which we added a new plot (Las Navas, 80 nest-boxes) located in a zeen oak (Q. faginea) woodland in the last study year. General field methods From late March onward, nest-boxes were monitored to determine the onset of egg laying, clutch size, hatching date, the number of hatchlings, and the number of fledged young. During the nest-building stage and the subsequent laying period, attention was paid to the nest appearance. Nests were visited daily to determine the presence of feathers in the nests and were grouped accordingly into 2 categories: nests with or without feathers. In addition, we recorded the number of feathers present in each nest in one of the study areas (Cabañeros). We only considered those feathers placed outside the nest cup, that is, those that were not in contact with the eggs. This was the case in the majority (.85%) of nests sampled. After clutch completion (except in the 2006 breeding season), we measured the height of the nest (average value of 4 measurements corresponding to each corner of the nest), and each was weighed with an electronic balance to the nearest 0.1 g. Adult birds were captured with spring traps while feeding nestlings 8 days old. All birds were identified with metal rings, sexed by presence or absence of a brood patch, and aged according to plumage characteristics as yearlings (first year) or older breeders. At ringing, we took the following morphometric measurements: body mass with an electronic balance (60.1 g) and tarsus length with a digital caliper (60.01 mm). Furthermore, in one of the study areas (Cabañeros), birds were fitted individually with passive integrated transponder (PIT) tags (Trovan ID 103; length: 11.6 mm, mass: 0.1 g) glued to 2 plastic color bands to quantify parental feeding effort (see below). When the nestlings were 13 days old, they were ringed, weighed, and the length of their tarsi measured in a similar way as parents. Around day

242 Behavioral Ecology 20, nests were visited again to determine the productivity (number of fledged young) of each pair. Parental feeding effort Parental investment (75 females and 75 males) during the broodrearing period (11 12 days posthatching) was estimated as the number of feeding events within a 24-h period. These were recorded by means of a transponder reader system consisting of an antenna fitted to the nest-box entrance and connected to a datalogging module and a 12 V battery. When a tagged bird alights at its nest entrance, it interrupts the optical barrier triggering the reading of the bird s corresponding PIT, and a visit is stored in the memory with the exact time at which it took place and the bird s identity (unique code). The data logger and the unit power were mounted in a plastic box beneath the nest-box and camouflaged with leafy branches or fallen leaves. In order to minimize disturbance to feeding birds, the sensor module was integrated into a preexisting nest-box door. The system was mounted on the day before recording feeding activity (day 10) to allow birds to become accustomed to this equipment. The installation of this device took about 10 min, and the blue tits feeding activity resumed 10 20 min later. In the analysis, each visit was not considered as a feeding because when a bird is perched at the nest entrance, it could interrupt the optical barrier several times causing multiple records. To extract reliable feeding rates from our data set, we excluded all recordings that were less than 6 s apart following a similar protocol to that employed by Johnsen et al. (2005). We have found a strong correlation between provisioning rates obtained by means of this procedure and those observed from video recordings (García-Navas et al. 2009). Thus, this method seems to be a good approach to quantify parental invesment in offspring feeding. Behavioral observations: colored feathers experiments Testing color preference of blue tits In2008, bluetit pairsbreedinginquintosdemorawere supplied with commercial colored feathers (Almacenes Cobián SA, Madrid, Spain) of similar size (80 90 mm in length) in order to test whether blue tits gather feathers irrespective of their color characteristics (insulating function hypothesis) or they select those that may be more detectable in dark environments (signaling hypothesis, see also Avilés et al. 2006). Just after nest construction was completed and before egg laying, we deposited a total of 10 feathers under the nest-box (2 red, 2 orange, 2 green, 2 violet, and 2 black). These colors vary largely in their reflectance spectra (Figure 2). If blue tits select certain colors (i.e., they gather feathers in a nonrandom way), their use frequency should account for more than 20% (equal representation in relation to the available feathers of each color) of the total number of gathered feathers. Feathers were placed directly on the ground, and they were surrounded with stones to impede wind disturbance. During the egg-laying period, nests (Valdeyerno n ¼ 35 and Gil García n ¼ 26) were visited daily in search for feathers. We annotated the number of feathers carried and the color of each. We then looked at whether the carrying of colored feathers into the nests had an influence on birds reproductive parameters. Figure 2 Reflectance spectra for artificial color feathers. Each curve represents the average for 10 spectra. Videotaping blue tits response to provisioning with colored feathers During the 2009 breeding season, blue tits nesting in Quintos de Mora were again supplied with colored feathers in order to videotape their response to the pile of colored feathers for at least 1 h. A video camera (Sony handy-cam) was placed 10 15 m away from the pile of feathers. We provided colored feathers prior to egg laying in a total of 18 nests. Meanwhile, in a subsample of nests (n ¼ 7), colored feathers were offered after clutch completion, once the incubation period began. When watching the videotapes, we noted the time elapsed between the moment at which colored feathers were deposited and a bird was observed approaching them. Thus, we examined whether blue tit parents react immediately to the presence of feathers in the vicinity of their nest and which of the sexes made use of this resource. Each pair member was identified on the basis of plumage characteristics (male blue tits usually appear brighter blue than females; e.g., Perrins 1979) and by ring details (location: left or right leg and presence or absence of color bands). Thereby, we were able to determine whether the addition of feathers into the nest is usually carried out by the male or by the female. Experimental manipulation In the spring of 2008, we performed an experiment with the blue tit populations breeding at Cabañeros. We selected nest dyads matched by laying date, and these were subjected to either a feather addition or a feather removal treatment, once the nest-building period was complete. In those nests assigned to the feather addition treatment (n ¼ 16), we added 5 upperwing covert feathers of rock dove Columba livia of similar size (60 75 mm) before the onset of egg laying. Feathers were arranged outside the nest cup with the concave side upward (Addition group). In the second experimental group (n ¼ 16), we removed all natural feathers that were found when checking nest-boxes during the laying period (Removal group). We also considered 2 control treatments: 1) unmanipulated nests (n ¼ 14) in which we found feathers carried by the male (Control 1) and 2) unmanipulated nests (n ¼ 16) without feathers carried by the male (Control 2). Nests were checked on a daily basis to record the number of feathers present in each. We replaced feathers in those nests (Addition group) in which they were removed by birds to ensure that all nests subjected to this treatment contained a similar amount of feathers. Statistical analyses A total of 257 nests (Cabañeros; 2006 2008: 147 and Quintos de Mora; 2009: 110) were classified into 2 categories: nests with or without feathers (n ¼ 134 and 122, respectively). By means of generalized linear models, we explored the existence of differences in general reproductive parameters (laying date, clutch size, length of the incubation period, and

Sanz and García-Navas Nest ornamentation by male blue tits 243 number of fledged young), parental effort (measured as feeding rates during a 24-h period), and body condition of young and adult birds between both groups. In all analysis years, nest-box plot and nest type (with or without feathers) were included as fixed factors. Female or male age was entered as a categorical variable. The following covariates were also included, as necessary, in the initial models: laying date, clutch size, brood size on day 8, and nestling tarsus length. To avoid pseudoreplication, bird identity was entered as a random factor. Degrees of freedom may differ between analyses because we did not have all measurements for all birds. We explored the effect of the addition of colored feathers provided to the birds on some of the abovementioned variables in a pilot study carried out in Quintos de Mora during the 2008 breeding season. We then tested experimentally whether the addition or removal of feathers had an effect on breeding traits of blue tits. Regrettably, the initial sample size of experimental treatments (n ¼ 16) was reduced for some of the variables considered (e.g., feeding frequency) due to predation by ladder snakes Rhinechis scalaris and nest usurpation by rock sparrows Petronia petronia. In all analyses, backward stepwise procedure was used to remove nonsignificant variables and interactions from the full models. Only final models are shown. We used Fisher s least significant difference (LSD) test for a posteriori pairwise comparisons. Pearson s correlations were employed to test the relationships between the number of feathers carried and both nest measurements (height and weight) and adult size (tarsus length). As variables fulfilled normality requirements, all tests were 2 tailed. Descriptive statistics are given as the mean 6 standard deviation (SD), unless stated otherwise. RESULTS Observational data Use of natural feathers More than half of blue tit nests (52%) were adorned. The average number of flight feathers found at the nest was approximately 6 (range: 0 30). We found that blue tits incorporated feathers from different species into their nests (see Supplementary Data). The most common were those from wood pigeon Columba palumbus (mainly upperwing coverts) and red-legged partridge Alectoris rufa (contour feathers, grayish with the point colored red). The presence of feathers was not significantly related to other nest features, such as its weight (Pearson s correlation t 98 ¼ 20.13, P ¼ 0.90) or height (t 98 ¼ 20.51, P ¼ 0.61). Use of color feathers We observed colored feathers in 64% of nests (n ¼ 61) in which we provided this resource after the nest-building period. Orange feathers were the most often selected; 24 of the 72 feathers found inside the nests were of this color. Thus, the majority of blue tit nests (81.8%) in which commercial feathers were detected contained at least one of orange color. The next most preferred colors were violet (54.5%) and red (50.0%). Green and black feathers were only found in 40.9% and 27.3%, respectively, of the nests in which we detected carrying of colored feathers. In sum, only orange feathers were found in a proportion (38.5%) greater than expected by chance (the violet, red, green, and black feathers supposed the 21.4, 19.7, 10.8, and 9.5% of the total, respectively). The presence of colored feathers seemed to have an effect on female productivity because larger clutch sizes were found in those nests in which colored feathers were added (mean 6 SD, with colored feathers: 8.31 6 0.34, n ¼ 22 and without colored feathers: 7.65 6 0.35, n ¼ 39; F 1,58 ¼ 2.31, P ¼ 0.02). There was no significant effect of the addition of colored feathers on either nestling body mass (F 1,56 ¼ 0.10, P ¼ 0.75) or number of chicks fledged (F 1,57 ¼ 1.06, P ¼ 0.30). In 10 of 18 nests filmed, we observed that feathers attracted blue tits, and at least one individual was recorded while moving or carrying feathers. In some cases, less than 5 min passed between the start of the video recording and the arrival of a pair member to the pile of feathers. This suggests that feathers constitute a very detectable and attractive resource for blue tits. The fact that in some nests, birds did not appear or were late in appearing is not surprising because at the end of the nest-building phase and during the egg-laying period, both pair members (mainly the male) are usually absent for long periods of time, and it rare to find them in the nest surroundings. In some nests, both pair members approached the pile to move feathers, but only the male (n ¼ 7) was observed carrying feathers to the nest. On the other hand, when feathers were provided to the birds once the laying was completed, they did not show interest in feathers (blue tits were not detected inspecting or carrying feathers in any of these nests). Effect of natural feathers on breeding parameters The presence of natural feathers in the nests was independent of the timing of breeding, although we detected a slight decrease in the prevalence of this trait as the season progressed (see Table 1). Clutch size varied significantly between nests in which such behavior was detected and those in which was not; females from nests adorned with feathers produced larger clutches when controlling for laying date (Table 1). The length of the incubation period did not seem to be affected by the presence of feathers in the nest. However, we found a nest-box plot 3 feathers interaction revealing contrasting patterns in these woodlands (Table 1). Post hoc comparisons showed that differences were only statistically significantly in El Brezoso where the incubation period was longer in those nests containing feathers (Fisher s LSD test, Anchurones: P ¼ 0.46, El Brezoso: P ¼ 0.02). There were differences between nest groups in nestling condition: Chicks from nests decorated with feathers were significantly heavier than those from nests in which such behavior was not detected (Table 1). Moreover, female body mass was higher in nests decorated with feathers than in those nests without feathers (Table 1). For males, there was no effect of the presence of feathers on their body mass. However, male, but not female, tarsus length significantly differed between these groups of nests (Table 1) with smaller values for males that did not carry feathers to the nest. In fact, we found a positive relationship between the number of feathers present in the nest and male tarsus length (Figure 3; Pearson s correlation t 97 ¼ 2.75, P, 0.01). On the other hand, there was no significant effect of the reproductive experience of the parents on the presence of feathers in the nest (female age: F 1,128 ¼ 0.30, P ¼ 0.83 and male age: F 1,128 ¼ 0.30, P ¼ 0.83). The presence of feathers carried by males during the nestbuilding period seemed to have a significant effect on parental feeding rates. Males from nests decorated with feathers fed their offspring more often than those that did not carry feathers to the nests (Table 1; Figure 4). On the other hand, females paired with males that carried feathers to the nest significantly reduced the number of feeding trips (Table 1; Figure 4). When considering the relative contribution of each parent to offspring feeding, we found that male contribution was significantly higher in nests decorated with feathers than those in which such behavior was not detected (male proportion [%] of feeding rates; with feathers: 62.44 6 2.57 and without feathers: 52.46 6 2.68; F 1,73 ¼ 6.94, P ¼ 0.01). Finally, the presence of

244 Behavioral Ecology Table 1 Differences between nests with and without feathers in general breeding parameters, body condition of nestlings and adult birds, and parental feeding effort (nestling provisioning rates during a 24-h period) for blue tits breeding at Montes de Toledo (central Spain) Nest type Without feathers With feathers df F P Laying date (1 ¼ 1 April) 19.30 6 1.42 14.36 6 1.53 Year 3,252 3.60 0.01 Feathers 1,252 3.25 0.07 Clutch size 7.62 6 0.20 8.76 6 0.21 Nest-box plot 4,250 8.67 <0.001 Laying date 1,250 42.34 <0.001 Feathers 1,250 17.85 <0.001 Incubation period (days) 13.78 6 0.16 14.16 6 0.17 Nest-box plot 1,142 5.53 0.02 Laying date 1,142 76.74 <0.001 Feathers 1,142 0.99 0.32 Nest-box plot 3 feathers 1,142 5.88 0.02 Nestling body mass (g) 10.40 6 0.10 10.71 6 0.10 Feathers 1,143 5.10 0.02 Female body mass (g) 10.26 6 0.07 10.44 6 0.06 Year 2,126 18.99 <0.001 Tarsus length 1,126 5.81 0.02 Feathers 1,129 4.07 0.04 Female tarsus length (mm) 15.87 6 0.07 15.95 6 0.06 Feathers 1,129 0.80 0.37 Male body mass (g) 10.51 6 0.07 10.51 6 0.06 Feathers 1,118 0.14 0.70 Male tarsus length (mm) 16.18 6 0.06 16.39 6 0.06 Feathers 1,121 5.50 0.02 Female feeding rates 150.64 6 14.76 109.06 6 14.18 Laying date 1,71 11.58 <0.01 Brood size on day 8 1,71 5.51 0.02 Feathers 1,71 5.60 0.02 Male feeding rates 155.92 6 12.83 179.72 6 12.33 Nest-box plot 1,71 7.96 <0.01 Brood size on day 8 1,71 6.04 <0.01 Feathers 1,71 4.12 0.04 Number of fledged young 5.73 6 0.29 6.67 6 0.27 Feathers 1,143 0.93 0.02 Generalized linear models with backward stepwise model selection (see MATERIALS AND METHODS). Only the terms contained in the final models are shown here. df ¼ degrees of freedom. Significant effects are reported in bold. feathers in the nest had a significant effect on productivity: The absolute number of fledged young was higher in those nests in which feather carrying was detected (Table 1). Experimental manipulation of feather presence Experimentally adding feathers to nests significantly affected clutch size: Larger clutches were found in those nests to which we added feathers in comparison with those in which they Figure 3 Male tarsus length in relation to the number of feathers found in their nest for 2 nest-box populations of blue tits breeding at Cabañeros National Park, central Spain (n ¼ 97). To avoid pseudoreplication, when an individual bird was sampled more than once, we randomly selected values from only 1 year. Sample sizes are given above the bars. Means 6 standard error are given. Figure 4 The relative parental offspring provisioning (number of feeding events during a 24-h period) in nests with and without feathers. Empty dots: males and filled dots: females. Means 6 standard error are given.

Sanz and García-Navas Nest ornamentation by male blue tits 245 Figure 5 Differences in clutch size among nests assigned to 4 different treatments: nests with feathers added experimentally (Addition), nests in which these were removed (Removal), nests with feathers added naturally that were left undisturbed (Control 1), and nests without feathers that were not manipulated (Control 2). were removed (t-test t 15 ¼ 22.65, P ¼ 0.01). When considering the 4 experimental groups, we also found statistically significant differences in clutch size (Figure 5; laying date: F 1,57 ¼ 14.05, P, 0.001 and treatment: F 3,57 ¼ 6.31, P, 0.001). Females from unmanipulated nests lacking feathers laid smaller clutches in comparison with the remaining categories (Fisher s LSD test, Ps, 0.05). Post hoc comparisons also revealed that clutches from nests adorned naturally or artificially with feathers were larger than in those in which they were removed (Ps, 0.05). On the other hand, there was no effect of the experimental treatment on nestling body mass (F 3,49 ¼ 0.81, P ¼ 0.49) or the number of fledglings (F 3,52 ¼ 0.21, P ¼ 0.89). Neither parental investment during the brood-rearing period differed significantly among treatments ( female feeding rates ; brood size at day 8: F 1,26 ¼ 5.80, P ¼ 0.02, treatment: F 3,26 ¼ 0.51, P ¼ 0.68 and male feeding rates ; brood size at day 8: F 1,23 ¼ 9.17, P, 0.01, treatment: F 3,23 ¼ 0.35, P ¼ 0.79). However, this lack of differences could be a result of low statistical power due to the small sample size (power test; female feeding rates: 0.15 and male feeding rates: 0.12). Sample size was reduced as a consequence of predation and nest desertion. Interestingly, we found a high proportion of widowed females in those nests in which we added feathers; half of the males that were fitted with transponders were not detected feeding nestlings. Male desertion rate was significantly or slightly higher in those nests that were manipulated (by adding or removing feathers) in comparison with the control treatments (Control 1: 0%, Control 2: 6.2%, Removal: 18.75%, Addition: 31.2%; Removal vs. Control 1: v 2 : 1.14, P ¼ 0.28; Removal vs. Control 2: v 2 : 2.92, P ¼ 0.08; Addition vs. Control 1: v 2 : 3.28, P ¼ 0.07; Addition vs. Control 2: v 2 : 5.25, P ¼ 0.02). With regard to this latter, we observed that in half of the nests (50%, n ¼ 16) in which feathers were added experimentally, these were removed (presumably by the male) from one day to the next, a behavior not reported in nests belonging to other groups. DISCUSSION Relatively few studies have documented the use of extended purely ornamental phenotypes in birds with the gathering of decorative objects such as colored feathers, flower petals, pieces of glass, or snail shells being an unusual behavior (e.g., Borgia et al. 1987; Karubian and Alvarado 2003). Here, we have shown evidence in favor of the hypothesis that the presence of ornamental feathers in the nest may function as a male signal in blue tits, which has not been described previously for this species (Perrins 1979; Cramp 1998). The fact that feathers usually were placed outside the nest cup (Figure 1a,b; Sanz JJ and García-Navas V, personal observation) and that such behavior only took place for a limited period of time (from the end of nest building to clutch completion) but throughout the entire breeding season (from late March to mid-june) does not constitute support for the nest-insulating function hypothesis. Because ornamental feathers are not in contact with the eggshells, this also excludes the possibility that feathers are introduced to diminish the probability of embryo infection by pathogenic bacteria as suggested by a recent paper (Peralta- Sánchez et al. 2010). Meanwhile, we found that when supplied with colored feathers, blue tits showed a clear preference for certain colors (mainly orange), whereas black feathers were rarely selected presumably due to their low reflectance (Eaton and Lanyon 2003). The results shown here support the findings of recent studies carried out with starling colonies in central Spain for which it has been suggested that feather carrying is a sex-specific signaling behavior involved in intersexual communication (Veiga and Polo 2005; Polo and Veiga 2006). The main result of our study was the effect that the addition of feathers to the nest had on female productivity. We found observational and experimental evidence that feather carrying by male blue tits induced females to lay more eggs. Because in this species male contribution to nest building can be considered as merely anecdotal, the addition of feathers to the nest when it is near completion or during early laying could provide males an opportunity to induce females to increase their reproductive investment. In this sense, our results agree with a previous study carried out on magpies, a species with biparental nest building, where an experimental increase in nest size (which mimicked increased male effort during nest building) resulted in females laying larger clutches (Soler et al. 2001). On the other hand, a recent study on pied flycatchers Ficedula hypoleuca, a species in which nest building is thought to be done exclusively by females, has reported that some males also contribute to nest construction and that the females paired with them laid heavier eggs (Martínez-de la Puente et al. 2009). Thus, nest-building activity or abilities related to this period may be used by the males as a postmating sexual display. However, what benefits do females obtain from responding to the male display? If nest building or gathering of ornamental materials is costly, only individuals in good condition should be able to successfully carry out such a task. This possibility is supported in the present study; we found that the number of feathers carried to the nest increased significantly with male tarsus length. That is, blue tits of larger size were more predisposed (or showed greater abilities) to search for and carry feathers to the nests. Such a result may arise for 2 reasons. First, birds of larger size may be capable of spending more time in search for this scarce resource than smaller birds due to the latter being more susceptible to energy constraints. Alternatively, because feathers constitute a difficult-to-find resource when a large number of them are available naturally (in our study areas, feather patches from sparrow hawk Accipiter nisus attacks are the main source of feathers), birds of larger size could monopolize this resource. This is in accordance with previous studies in which it has been found that body size may constitute an advertisement of male resource-holding potential (i.e., priority to access for feathers, see references given in Kölliker et al. 1999). Thus, feather provisioning may also play an important role in competitive interactions among male blue tits (see more below). The framework developed for animal communication based on bodily signals in an intersexual context holds equally well for nonbodily ornaments. Hence, the fact that female blue tits increased their clutch size in response to the addition of feathers (either experimentally or naturally) suggests that males

246 Behavioral Ecology exhibiting a higher feather-carrying effort could be perceived as more vigorous or more attractive individuals (i.e., highquality males) by the females. In this sense, several studies have shown that females mated to attractive males allocated more resources in the phase of egg production, which is strictly under female control, increasing their productivity (e.g., Petrie and Williams 1993), egg size (Cunningham and Russell 2000; Osorno et al. 2006), or testosterone content (e.g., Gil et al. 1999; Kingma et al. 2009). However, to our knowledge, no other studies, apart from Johnsen et al. (2005) and Szentirmai et al. (2005), have demonstrated that females of species with biparental care enhance their parental effort (offspring feeding rates; parental decision: care/desert the clutch) in response to male attractiveness independently from the partner s contribution. Here, we did not find evidence that feather carrying leads females to invest more in feeding their progeny. Rather, the opposite is true; our results indicate that feeding frequencies by females were lower in those nests in which we observed feathers than in those lacking this ornament. In addition, we found that females from nests decorated with feathers were heavier, which suggests that, indeed, they may have relaxed their parental responsibilities. In this sense, it has been suggested that the rate at which one parent supplies food can modify the provisioning behavior of its partner (the partner-compensation hypothesis; Witte 1995). Our results support this hypothesis because we found that males that decorated their nests with feathers fed their offspring at a significantly higher rate than those that did not engage in such activity. Despite the fact that the opposite situation (i.e., females are compelled to increase their parental investment when their mate contributes less care) is perhaps more common, our results suggest that females would change their provisioning behavior at the expense of what their mate was doing. This does not mean that females work less, but they could opt to reduce the number of feeding trips in order to increase their prey selectivity (which may reduce the overall food delivery rate; e.g., Grieco 2002). Thus, females could have responded to low brood demand (as a consequence of the high provisioning effort of the males) by increasing the size and quality of food brought to the nest (Hurtrez-Boussès et al. 2000; García-Navas and Sanz 2010). This would explain why nestlings from nests decorated with feathers were heavier than those from nests lacking ornamentation. Hence, our results are in line with what the sexual selection hypothesis posits when explaining the maintenance of elaborated or costly displays. This does not exclude the possibility that indirect benefits are also involved in the maintenance of postmating signals. The above-mentioned relationship between feather carrying and paternal care is noteworthy because few studies have shown a clear association between male secondary traits and parental effort devoted to young in birds. In this sense, there is controversy over whether males should reliably signal the direct benefits they offer to females (see Møller and Jennions 2001 for a review). According to Kokko (1998), male signals are more likely to be honest if the opportunities for multiple mating are low. In the blue tit, a monogamous species, it has been shown that other secondary sexual characters reliably reflect the foraging ability and health status of the bearer (Senar et al. 2002). The addition of bizarre material to the nest may also function as an honest indicator of male quality and thus as a signal of good parents in this species. Our results are in agreement with previous studies on black wheatears in which it has been shown that stone-carrying display reflects male health status (Soler et al. 1999) and their disposition to providing parental care (Moreno et al. 1994). On the other hand, when manipulating nest ornamentation, we found a higher male desertion rate in those nests in which we added feathers in comparison with the other experimental groups. This suggests that males could infer the existence of a second male (i.e., extrapair mating) from the presence of foreign feathers and thus invest less in the care of the brood due to reduced paternity confidence (see Kempenaers and Sheldon 1997 and references therein). In fact, we observed that in some nests in which feathers were added experimentally, these were removed or buried with nest material within a 24-h period. The possibility that feather carrying could also act as a signal in male male competition (i.e., intrasexual communication) or could at least be used by researchers as a way of manipulating male certainty of paternity deserves further attention. In conclusion, the results presented here suggest that feather carrying to the nest with ornamental purpose is likely to convey information on parental quality, and this behavior can be used by female blue tits for making parental investment decisions. Further studies on this issue in populations in which this puzzling behavior has not been described previously (i.e., novel environments) may provide excellent opportunities for gaining an understanding of how certain sexual signals may have evolved and the selective agents or mechanisms responsible for their divergence among populations (e.g., Yeh 2004). SUPPLEMENTARY MATERIAL Supplementary material can be found at http://www.beheco. oxfordjournals.org/. FUNDING An FPI predoctoral fellowship (Ministerio de Ciencia e Innovación-European Social Fund to V.G.N.); projects 69/2003 (Organismo Autónomo de Parques Nacionales-Ministerio de Medio Ambiente) and GCL2007-61395 (Ministerio de Ciencia e Innovación). We are grateful to José Jiménez, Manuel Carrasco, Ángel Gómez- Manzaneque (Cabañeros National Park), José Manuel Sebastián, and Carlos R. Vigal (Centro Quintos de Mora) for the facilities offered to us to work in these protected areas. Gil Rosenthal and 2 anonymous reviewers provided constructive comments and improved the quality of the manuscript. REFERENCES Andersson M. 1994. Sexual selection. Princeton (NJ): Princeton University Press. Avilés JM, Soler JJ, Pérez-Contreras T. 2006. Dark nests and egg colour in birds: a possible functional role of ultraviolet reflectance in egg detectability. Proc R Soc Lond B Biol Sci. 273:2821 2829. Barber I, Nairn D, Huntingford FA. 2001. Nests as ornaments: revealing construction by male sticklebacks. Behav Ecol. 12:390 396. Berg ML, Beintema NH, Welbergen JA, Komdeur J. 2006. The functional significance of multiple-nest building in the Australian reed warbler Acrocephalus australis. Ibis. 148:395 406. 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, Gore MA. 1986. Feather stealing in the satin bowerbird (Ptilonorhynchus violaceus); mate competition and the quality of display. Anim Behav. 34:727 738. Borgia G, Kaatz IM, Condit R. 1987. Flower choice and bower decoration in the satin bowerbird Philonorhynchus violaceus a test of hypotheses for the evolution of male display. Anim Behav. 35:1129 1139. Brouwer L, Komdeur J. 2004. Green nesting material has a function in mate attraction in the European starling. Anim Behav. 67:539 548. Burley N. 1986. Sexual selection for aesthetic traits in species with biparental care. Am Nat. 127:415 445.

Sanz and García-Navas Nest ornamentation by male blue tits 247 Cramp S. 1998. The complete birds of the Western Palearctic. Oxford: Optimedia, Oxford University Press. Cunningham EJA, Russell AF. 2000. Egg investment is influenced by male attractiveness in the mallard. Nature. 404:74 77. De Neve L, Soler JJ. 2002. Nest-building activity and laying date influence female reproductive investment in magpies: an experimental study. Anim Behav. 63:975 980. Eaton MD, Lanyon SM. 2003. The ubiquity of avian ultraviolet plumage reflectance. Proc R Soc Lond B Biol Sci. 270:1721 1726. Evans MR, Burn JL. 1996. An experimental analysis of mate choice in the wren: a monomorphic, polygynous passerine. Behav Ecol. 7:101 108. Frith CB, Frith DW. 1990. Archbold s bowerbird Archboldia papuensis (Ptilonorhynchidae) uses plumes from king of saxony bird of paradise Pteridophora alberti (Paradisaeidae) as bower decoration. Emu. 90:136 137. García-Navas V, Ortego J, Sanz JJ. 2009. Heterozygosity-based assortative mating in blue tits (Cyanistes caeruleus): implications for the evolution of mate choice. Proc R Soc Lond B Biol Sci. 276: 2931 2940. García-Navas V, Sanz JJ. 2010. Flexibility in the foraging behavior of blue tits in response to short-term manipulations of brood-size. Ethology. 116:744 754. Gil D, Graves J, Hazon N, Wells A. 1999. Male attractiveness and differential testosterone investment in zebra finch eggs. Science. 286:126 128. Grieco F. 2002. How different provisioning strategies result in equal rates of food delivery: an experimental study of blue tits. J Avian Biol. 33:331 341. Gwinner H. 1997. The function of green plants in nests of European starlings Sturnus vulgaris. Behaviour. 134:337 351. Hansell MH. 2000. Birds nests and construction behaviour. Cambridge: Cambridge University Press. Hoi H, Schleicher B, Valera F. 1994. Female mate choice and nest desertion in penduline tits, Remiz pendulinus. Anim Behav. 48: 743 746. Hurtrez-Boussès S, Renaud F, Blondel J, Perret P, Galan M- J. 2000. Effects of ectoparasites of young on parents behaviour in a Mediterranean population of blue tits. J Avian Biol. 31:266 269. Johnsen A, Delhey K, Schlicht E, Peters A, Kempenaers B. 2005. Male sexual attractiveness and parental effort in blue tits: a test of the differential allocation hypothesis. Anim Behav. 70:877 888. Karubian J, Alvarado A. 2003. Testing the function of petal-carrying in the red-backed fairy-wren (Malurus melanocephalus). Emu. 103: 87 92. Kempenaers B, Sheldon BC. 1997. Studying paternity and paternal care: pitfalls and problems. Anim Behav. 53:423 427. Kingma SA, Komdeur J, Vedder O, von Engelhardt N, Korsten P, Groothuis TGG. 2009. Manipulation of male attractiveness induces rapid changes in avian maternal yolk androgen deposition. Behav Ecol. 20:172 179. Kokko H. 1998. Should advertising parental care be honest? Proc R Soc Lond B Biol Sci. 265:1871 1878. Kölliker M, Heeb P, Werner I, Mateman AC, Lessells CM, Richner H. 1999. Offspring sex ratio is related to male body size in the great tit (Parus major). Behav Ecol. 10:68 72. LeBas NR, Hockham LR, Ritchie MG. 2004. Sexual selection in the gift-giving dance fly, Rhamphomyia sulcata, favours small males carrying small gifts. Evolution. 58:1763 1772. Lombardo MP, Bosman RM, Faro CA, Houtteman SG, Kluisza TSK. 1995. Effect of feathers as nest insulation on incubation behavior and reproductive performance of tree swallows (Tachycineta bicolor). Auk. 112:973 981. Madden JR. 2003. Bower decorations are good predictors of mating success in the spotted bowerbird. Behav Ecol Sociobiol. 53:269 277. Martínez-de la Puente J, Merino S, Lobato E, Moreno J, Tomás G, Morales J. 2009. Male nest-building activity influences clutch mass in pied flycatchers Ficedula hypoleuca. Bird Study. 56: 264 267. McGowan A, Sharp SP, Hatchwell BJ. 2004. The structure and function of nests of long-tailed tits Aegithalos caudatus. Funct Ecol. 18:578 583. Mennerat A, Perret P, Bourgault P, Blondel J, Gimenez O, Thomas DW, Heeb P, Lambrechts MM. 2009. Aromatic plants in nests of blue tits: positive effects on nestlings. Anim Behav. 77:569 574. Møller AP. 1984. On the use of feathers in bird nests: predictions and tests. Ornis Scand. 15:38 42. Møller AP. 1987. Nest lining in relation to the nesting cycle in the swallow Hirundo rustica. Ornis Scand. 18:148 149. Møller AP, Jennions MD. 2001. How important are direct fitness benefits of sexual selection? Naturwissen. 88:401 415. Moreno J, Soler M, Møller AP, Linden M. 1994. The function of stone carrying in the black wheatear, Oenanthe leucura. Anim Behav. 47:1297 1309. Osorno JL, Morales J, Moreno J, Merino S, Tomás G, Vásquez R. 2006. Evidence for differential maternal allocation to eggs in relation to manipulated male attractiveness in the pied flycatcher (Ficedula hypoleuca). J Ornithol. 147:605 611. Palomino JJ, Martín-Vivaldi M, Soler M, Soler JJ. 1998. Functionalsignificance of nest size variation in the rufous bush robin Cercotrichas galactotes. Ardea. 86:177 185. Peralta-Sánchez JM, Møller AP, Martín-Platero AM, Soler JJ. 2010. Number and color composition of nest lining feathers predict eggshell bacterial community in barn swallows nests. An experimental study. Funct Ecol. 24:426 433. Perrins CM. 1979. British tits. London: Collins. Petrie M, Williams A. 1993. Peahens lay more eggs for peacocks with larger trains. Proc R Soc Lond B Biol Sci. 251:127 131. Polo V, Veiga JP. 2006. Nest ornamentation by female spotless starlings in response to a male display: an experimental study. J Anim Ecol. 75:942 947. Schaedelin CF, Taborsky M. 2009. Extended phenotypes as signals. Biol Rev. 84:293 313. Senar JC, Figuerola J, Pascual J. 2002. Brighter yellow blue tits make better parents. Proc R Soc Lond B Biol Sci. 269:257 261. Sliwa A, Richardson PRK. 1998. Responses of aardwolves, Proteles cristatus, Sparrman 1783, to translocated scent marks. Anim Behav. 56:137 146. Soler JJ, Cuervo JJ, Møller AP, de Lope F. 1998. Nest building is a sexually selected behavior in the barn swallow. Anim Behav. 56: 1435 1442. Soler JJ, Martín-Vivaldi M, Haussy C, Møller AP. 2007. Intra- and interspecific relationships between nest size and immunity. Behav Ecol. 18:781 789. Soler JJ, Møller AP, Soler M. 1998. Nest building, sexual selection and parental invesment. Evol Ecol. 12:427 441. Soler JJ, De Neve L, Martínez JG, Soler M. 2001. Nest size affects clutch size and the start of incubation in magpies: an experimental study. Behav Ecol. 12:301 307. Soler M, Martín-Vivaldi M, Marín JM, Møller AP. 1999. Weight lifting and health status in the black wheatear. Behav Ecol. 10:281 286. Szentirmai I, Komdeur J, Székely T. 2005. What makes a nest-building male successful? Male behaviour and female care in penduline tits. Behav Ecol. 16:994 1000. Veiga JP, Polo V. 2005. Feathers at nests are potential female signals in the spotless starling. Biol Lett. 1:334 337. Veiga JP, Polo V, Viñuela J. 2005. Nest green plants as a male status signal and courtship display in the spotless starling. Ethology. 112:1 9. Witte K. 1995. The differential-allocation hypothesis: does the evidence support it? Evolution. 49:1289 1290. Yeh P. 2004. Rapid evolution of a sexually selected trait following population establishment in a novel environment. Evolution. 58: 166 174.