Paternity and parental care in the black-throated blue warbler, Dendroica caerulescens

ANIMAL BEHAVIOUR, 2001, 62, 83–92 doi:10.1006/anbe.2001.1733, available online at http://www.idealibrary.com on

Paternity and parental care in the black-throated blue warbler, Dendroica caerulescens HELEN C. CHUANG-DOBBS*, MICHAEL S. WEBSTER* & RICHARD T. HOLMES†

*Department of Biological Sciences, University at Buffalo, State University of New York †Department of Biological Sciences, Dartmouth College (Received 16 August 1999; initial acceptance 30 October 1999; final acceptance 1 January 2001; MS. number: A8564)

Much debate surrounds the relationship between male parental care and paternity. We quantified parental care of yearling (second year: SY) and older (after second year: ASY) male black-throated blue warblers and determined parentage with microsatellites over four breeding seasons. ASY males that had sired all young in their broods fed 7-day-old nestlings at higher rates than ASY males that sired only some of the young in their broods. This relationship was not present in yearling males. A negative relationship between level of paternity and paternal care may arise if males facultatively adjust their care in response to cues of paternity, if poor-quality males are both more likely to be cuckolded and less able to provide parental care, or if males reduce both mate guarding and parental care to pursue extrapair matings. Parental care given by ASY males was not associated with additional mating opportunities during the nestling period. Thus, pursuit of extra matings is unlikely to account for the association between paternity and parental care. ASY males in better condition had higher levels of paternity, but did not guard their mates more closely nor feed their young at higher rates. These results suggest that ASY males may be able to assess their paternity, possibly by using cues related to levels of local synchrony during their females’ fertile periods, and adjust their care accordingly. Paternity obtained by SY males was not associated with local synchrony, however, suggesting that the apparent lack of a facultative response by males of this age class may be due to a lack of reliable cues to paternity. 

cues to assess the extent of his paternity in his brood and adjusts his care accordingly (Westneat & Sherman 1993). For instance, a male may be able to distinguish his own young from those of other males directly, although there is little evidence to support this in birds (Westneat & Sargent 1996; Sheldon & Ellegren 1998). Alternatively, a male may use indirect cues to assess his likelihood of paternity, such as the amount of access he had to his social mate during her fertile period, the amount of time the male spent guarding his mate during her fertile period, or perhaps the behaviour of the female (Møller 1988; Burke et al. 1989; Davies et al. 1992; Sheldon & Ellegren 1998; Whittingham & Dunn 1998). To date, correlational tests for relationships between male shares of paternity and parental care have yielded mixed results (reviewed in Kempenaers & Sheldon 1997). In part, this may be due to confounding factors that affect both paternity and parental care (Kempenaers & Sheldon 1997, 1998). For example, an association between parental care and paternity may arise if both factors independently covary with male condition. Male quality or body condition may affect either a male’s ability to guard

Parental investment theory postulates that individuals should balance costs and benefits of parental behaviour so as to maximize lifetime reproductive success (Winkler 1987). In birds, extrapair fertilizations (EPFs) are common (Westneat et al. 1990) and biparental care occurs in the majority of species (Lack 1968). As a consequence, males may frequently be faced with caring for unrelated offspring. Paternal investment therefore may not necessarily translate into fitness benefits to male care givers. Recent theoretical models predict that, under some circumstances, a decrease in a male’s paternity (i.e. relatedness of a male to his putative offspring) should result in a decrease in male parental care (Westneat & Sherman 1990; Whittingham et al. 1992; Xia 1992). Under the ‘facultative response hypothesis’, a male uses Correspondence and present addresses: M. S. Webster is now at the School of Biological Sciences, Washington State University, Pullman, WA 99164-4236, U.S.A. (email: [email protected]). H. C. ChuangDobbs is now at the Department of Biology, Southern Utah University, Cedar City, UT 84720, U.S.A. R. T. Holmes is at the Department of Biological Sciences, Dartmouth College, Hanover, NH 03755, U.S.A. 0003–3472/01/070083+10 $35.00/0

2001 The Association for the Study of Animal Behaviour

83



2001 The Association for the Study of Animal Behaviour

84

ANIMAL BEHAVIOUR, 62, 1

his mate during the female fertile period or a female’s propensity to seek EPFs from other males. Either of these routes may result in male quality affecting a male’s share of paternity (Ritchison et al. 1994; Westneat 1994; Riley et al. 1995; Burley et al. 1996; MacDougall-Shackleton et al. 1996). Male condition or quality may also translate separately into a male’s ability to provision young. Thus, as an alternative to the facultative response hypothesis given above, the ‘male quality hypothesis’ suggests that low-quality males are likely to lose paternity and also are likely to be poor provisioners, whereas high-quality males are likely to have high paternity and to be good provisioners. This hypothesis predicts that correlations between male condition and both paternity and paternal care will underlie a positive relationship between parental care and paternity. Alternatively, a negative relationship between paternity and paternal care may result from interactions between male quality and female control of investment. Following Burley’s ‘differential allocation model’, females mated to high-quality males may contribute more to parental investment than females mated to low-quality males. As a consequence, highquality males are likely to contribute less to care than low-quality males (Burley 1988). If females mated to high-quality males are also unlikely to seek EPFs, those males that contribute less care will also be those that have high paternity in their broods. Another possible covariate of paternity and parental care is the opportunity for males to engage in additional matings. If mate guarding is important in ensuring relatedness of a male to the young in his broods, males may risk their paternity if they invest less time in mate guarding and more time in other activities, such as the pursuit of additional matings (Westneat et al. 1990). Similarly, males may contribute less to parental care when there are fertile females available for copulations (Westneat et al. 1990; Wright & Cotton 1994; Hartley et al. 1995; Magrath & Elgar 1997). If, for a given male, many extrapair mating opportunities are available during both his mate’s fertile period and later, during the nestling stage, then the male may reduce both mate guarding and nestling feeding in order to pursue other matings (the converse would hold for a male with few extrapair mating opportunities). Thus, the ‘effort shift hypothesis’ predicts that (1) extrapair mating opportunities will lead to a reduction in mate guarding during the female fertile period, (2) extrapair mating opportunities will lead to a reduction in feeding rates during the nestling period, and (3) the frequency of extrapair mating opportunities will be correlated between these two periods of the reproductive cycle. In black-throated blue warblers EPFs occur at relatively high frequencies (Chuang et al. 1999), mate-guarding efforts of males appear to be important in assuring paternity (Chuang-Dobbs et al., in press), and both males and females contribute to provisioning young (Goodbred & Holmes 1996). In this study, we tested for a relationship between male parental care and paternity in this species. In addition, to elucidate the reasons for any such relationship and test among the hypotheses given above, we examined whether paternity and paternal care

covaried with either male quality or additional mating opportunities. METHODS

Study Area and Field Methods We studied a population of black-throated blue warblers from 1995 to 1998 at the Hubbard Brook Experimental Forest in West Thornton, New Hampshire, U.S.A. Males arrive at breeding grounds in early May and females arrive approximately one week later and settle with males to form monogamous pairs (Holmes 1994). Males guard their mates during the nest-building and egg-laying stages, when females are most fertile (ChuangDobbs et al., in press). Despite social monogamy and male paternity guards, rates of EPFs are relatively high (Chuang et al. 1999). Microsatellite analyses presented elsewhere (Webster et al., in press) show that, during the period covered by this study, 23% of 390 nestlings were sired by extrapair males (range 15–32% per year), and 37% of 122 broods contained extrapair young (range 28–57% per year) (average clutch size: 3.6, range 2–5 nestlings). We mapped territories by following individuals and recording presence and vocal advertisements by males. Females build their nests primarily in shrub layers, within 1–1.5 m off the ground (Holmes 1994). We located nests by following females as they carried nest-building materials or by searching regions with considerable undergrowth. We captured adults with mist nets and marked each with a numbered aluminium U.S. Fish and Wildlife Service (USFWS) band and a unique combination of coloured leg bands. We also classified adults by plumage as yearling (second year: SY) or older (after second year: ASY) (Pyle et al. 1987; Holmes 1994). All adults were weighed to the nearest 0.1 g with a Pesola scale and their tarsus lengths were measured to the nearest 0.01 mm with a digital calliper. For molecular analyses, we collected ca. 20
l of blood from the brachial vein of each individual. We collected blood samples from 6-day-old nestlings and marked them with a USFWS band. Blood samples were either kept frozen in liquid nitrogen in the field and then at
80 C, or stored in lysis buffer (100 mM Tris, 10 mM NaCl, 100 mM EDTA, 0.5% sodium dodecyl sulfate, SDS) at 4 C until analysis. To quantify mate guarding, we used focal-animal sampling methods (Altmann 1974) to determine rates of males following females during the female fertile period. Because females are most fertile during the nest-building and egg-laying period (Birkead & Møller 1992), we restricted our observations to this period for each focal male. In 1998, we followed 11 males for up to 1 h, during which we recorded the number of times that a male moved to be within 1 m of his mate within 5 s of the female’s initial movement away from the male. We also recorded proximity of focal males to their social mates every 2 min as an instantaneous sample (Altmann 1974). For this measure, we estimated distances between individuals as being less than or greater than 20 m. The latter category included those instances in which focal males were visible, but their social mates were not.

CHUANG-DOBBS ET AL.: PATERNITY AND PARENTAL CARE

We videotaped nests to assess parental provisioning patterns. This was done on the fifth and seventh days after nestlings hatched in 1995, fourth and fifth days in 1996, and seventh day in 1997 and 1998. Only seven of the videotapings from 1995 were obtained from the same nests (i.e. the majority of nests were not videotaped more than once on different days). We mounted SONY Handycam camcorders with 24 zoom lenses on tripods and covered them with camouflaged burlap. If necessary, we also used SONY Sportspaks to protect cameras from excessive moisture. We set cameras 10–15 m away from the nest, and pinned back vegetation that obscured the view of the nest. Recording sessions lasted for an average time (SD) of 2.340.94 h. We did not include in our analyses any nests with evidence that parents were disturbed (N=4; e.g. chipping notes by female, repeated and rapid visits to the nest by parents without food delivery or brooding). From the videotapes, we quantified numbers of feeding trips and biomass of prey brought to the nest per visit. Length of prey (l) was converted into biomass (b) using the formula b=0.004(l2.64) as described in Rodenhouse (1986) and Goodbred & Holmes (1996).

Microsatellite Methods We extracted nuclear DNA from blood that had been stored at
80 C by incubating 5
l of blood at 65 C for 1 h in 650
l TNE buffer (10 mM Tris, 10 mM NaCl, 2 mM EDTA), 5
l of 20 mg/ml proteinase K, and 16
l of 20% SDS, followed by standard phenol-chloroform extractions (see Westneat 1990). We isolated DNA from blood that had been stored in lysis buffer by incubating 40
l of blood-buffer mix at 65 C for 3 h in 360
l TNE buffer and 10
l of 20 mg/ml proteinase K, followed by the same extraction protocols. We amplified 1
l of DNA from each individual in a 10
l PCR reaction containing 100
M dNTP (each), 0.25
M primers (each), 1.5
Ci [33P]dATP (NEN Life Science Products), and a PCR reaction mix (1 unit Taq DNA polymerase, 3.0 mM MgCl2, 50 mM KCl, 10 mM Tris-HCl). We amplified DNA from each individual using PCR primers developed from the genome of the black-throated blue warbler, D. caerulescens (Dca 24, Dca 28, Dca 32; Chuang 1999; Webster et al., in press) and from the genome of the yellow warbler, D. petechia (Dp
01 and Dp
16; Dawson et al. 1997). Cycling parameters were as follows: 3 min at 94 C followed by 30 cycles of 94 C for 60 s, 60 s annealing (the annealing temperature for Dca 32, Dp
01, and Dp
16 was 60 C, for Dca 24 was 55 C, and for Dca 28 was 64 C), 45 s extension at 72 C, and finally an extra 5 min at 72 C for complete extension of products. We ran products on 6% denaturing polyacrylamide gels containing 7M urea at 50 C. An M13 DNA sequence labelled with [33P]dATP served as a size reference. Gels were dried for 1 h and exposed to autoradiography film for 2–3 days. We quantified the size of bands (i.e. alleles) at each locus by comparing each band to the M13 DNA size standard. Multilocus DNA fingerprinting analyses indicate that intraspecific brood parasitism is rare or absent in this species (Chuang et al. 1999). Therefore, we assumed

social mothers were biological mothers in all cases and compared genotypes of nestlings to their mothers. A nestling’s allele that was the same size as one of the social mother’s alleles was considered to be a ‘match’ between individuals. The remaining (i.e. nonmaternal) allele of each nestling was then compared to the alleles of the nestling’s social father. We calculated the average probabilities of matches between nonsires and nestlings across all alleles at each locus. For each locus, the average probability (across all alleles) that a nonsire would possess the nestling’s nonmaternal allele (calculated following Jamieson 1996) ranged from 0.112 to 0.417 (Webster et al., in press). Based upon these averages, the combined probability that a nonsire would match at all five microsatellite loci was 0.0002 (i.e. the probability of detecting a nonsire was 0.9998). Thus, we were able to identify extrapair offspring with high certainty.

Statistical Methods Male mate-guarding efforts were quantified as incidences of males following females per hour. We also measured extent of male mate guarding as a function of the proportion of scans in which we observed a male within 20 m of his mate. Parental provisioning was quantified separately for males and females in terms of per capita nestling feeding rates (feeding trips per nestling per hour) and per capita biomass delivery rates (grams per nestling per hour). These measures were obtained by standardizing brood delivery rates to brood size (i.e. by dividing the total number of deliveries to the brood by the number of nestlings). To estimate male body condition, we used a measure of mass per unit body size (Jakob et al. 1996). To obtain this, we regressed weights on tarsus lengths and used the resulting residuals either as a continuous measure of body condition or as a categorical variable. In the latter case, males whose values fell above the regression line were treated as males in ‘good condition’, and males with values falling below the regression line as ‘poor’. We defined a female’s fertile period as the time beginning 3 days before the first egg of a clutch was laid and ending the day the penultimate egg was laid. We also defined neighbours as individuals in territories having an estimated boundary within 50 m of the boundary of a focal territory (territory size range 1–4 ha; Holmes 1994). The number of opportunities for additional matings during the fertile period of a male’s mate (i.e. local synchrony) equalled the number of neighbouring females whose fertile periods overlapped the focal female’s by at least 1 day (Chuang et al. 1999). We also examined opportunities for additional matings during the nestling period. These were measured as the number of neighbouring females whose fertile periods overlapped the focal pair’s nestling period by at least 1 day, with nestling periods beginning the day the nestlings hatched and ending the day of fledging, abandonment by parents, or loss of nests due to predation. For statistical analyses concerning local synchrony, only females with known neighbours were used; territories at the edge of the study plot were excluded. We used opportunities for additional

85

86

ANIMAL BEHAVIOUR, 62, 1

Table 1. Results of analyses of variance tests for differences in male per capita parental care with paternity, male age and interactions on different days of the nestling period Male nestling feeding rate (per capita)

Male biomass delivery rate (per capita)

df

MS*

F

P

df

MS

F

P

Day 4 EPY Male age EPY*Male age Residual

1 1 1 9

0.001 0.47 0.24 0.26

0.003 1.79 0.92

0.96 0.21 0.36

1 1 1 8

3.91 28.14 12.19 18.73

0.21 1.50 0.65

0.66 0.26 0.44

Day 5 EPY Male age EPY*Male age Residual

1 1 1 21

1.38 0.46 0.00 0.50

2.77 0.92 0.00

0.11 0.35 0.99

1 1 1 18

98.99 100.77 0.02 58.86

1.68 1.71 0.00

0.21 0.21 0.99

Day 7 EPY Male age EPY*Male age Residual

1 1 1 50

0.31 0.18 5.14 0.93

0.33 0.19 5.52

0.57 0.67 0.02

1 1 1 44

22.65 42.25 269.68 55.96

0.41 0.76 4.82

0.53 0.39 0.03

See Results for post hoc analyses. *Mean square values. Degrees of freedom and mean square values for residuals are provided.

matings either as a continuous variable or as a categorical measure; for the latter, opportunities were determined to be low (if the number of fertile female neighbours was less than or equal to the population median) or high (if the number of fertile neighbours was greater than the population median). Finally, we tested for possible effects of seasonality on body condition, availability of fertile females, frequency of EPF and nestling feeding rates. We used Statview 5.0 for all analyses (SAS Institute 1998). RESULTS

Paternity and Parental Care Analyses of male feeding rates revealed little effect of paternity or age class on male nestling feeding rates on days 4 and 5 of the nestling period (Table 1). However, feeding rates for days 4 and 5 had a much higher variance than data from day 7, most likely due to smaller sample sizes for the earlier days. For 7-day-old nestlings, a significant interaction existed between male age class and paternity on nestling feeding rates (Table 1): ASY males with EPFs in their broods fed nestlings at lower rates than ASY males without EPFs (Scheffe’s test: nestling feeding rates: t28 =2.27, P=0.03; biomass delivery rates: t27 =2.14, P=0.04; Fig. 1), but the feeding rates of SY males were not associated with paternity (nestling feeding rates: t22 =1.14, P=0.27; biomass delivery rates: t17 =1.10, P=0.29; Fig. 1). Female feeding rate was significantly higher at nests with EPFs than at nests without EPFs, but again on only day 7 of the nestling period (Table 2, Fig. 1). This pattern of female parental care held regardless of female age class. Because the significant relationship between paternity and parental care occurred only on day 7 and differed

with male age, we restricted the remainder of analyses to 7-day-old nestlings and conducted these analyses separately for SY and ASY males.

Male Quality, Mate Guarding and Parental Care Male body condition was not associated with male age class (Scheffe’s test: t42 =0.94, P=0.35). Rates of males following females during the fertile period were significantly related to relative male body condition, with males in relatively better condition following females at lower rates than males in poor condition (Mann–Whitney U test: U=1.00, N1 =N2 =4, P=0.04). Male proximity to females was unrelated to relative male body condition (Mann–Whitney U test: U=5.00, N1 =N2 =4, P=0.38). Sample sizes for mate-guarding analyses were insufficient to split results based on male age class. For ASY males, male body condition was negatively associated with the likelihood of EPF occurring in broods (log-likelihood ratio test: 21 =4.12, P=0.04). This relationship was not apparent in SY males (21 =2.29, P=0.13). Male body condition was not associated with nestling feeding rates by ASY males (linear regression: R2 =0.03, F1,22 =0.56, P=0.41) or by SY males (R2 =0.005, F1,18 =0.08, P=0.77). There was also no relationship between male body condition and male biomass delivery rates (linear regression: ASY males: R2 =0.14, F1,21 =3.11, P=0.08; SY males: R2 =0.09, F1,13 =1.39, P=0.27). Although the relationship between male body condition and biomass delivery rates in older males approached significance, the regression coefficient was not in the direction predicted by the male quality hypothesis (regression coefficient=
5.98). In addition to these single-factor analyses, we conducted an analysis of covariance (ANCOVA) in which

Biomass delivery rate (g per nestling per h)

Nestling feeding rate (trips per nestling per h)

CHUANG-DOBBS ET AL.: PATERNITY AND PARENTAL CARE

20

EPY no EPY

3

19

we tested the effect of paternity (EPF present versus absent) on male feeding rates with male body condition as a covariate. For SY males, neither paternity nor male condition were significant factors on male nestling feeding rates or biomass delivery rates (Table 3). For ASY males, paternity was associated significantly with both per capita nestling feeding rates and biomass delivery rates; male condition did not covary with nestling feeding rates, but was a significant covariate for male biomass delivery rates (Table 3). However, as in the univariate analyses, the relationship between body condition of ASY males and biomass delivery was negative (regression coefficient=
7.65). Thus, our measures of paternal care were not associated with male body condition in the manner predicted by the male condition hypothesis.

9

35

2 15 11 1

0

Opportunities for Matings and Parental Care 20 18 18

7 15

11 12

10 31 5

0

Male (ASY)

Male (SY)

Female

Figure 1. Mean±SE per capita nestling feeding rates (top) and per capita biomass delivery rates (bottom) for male and female blackthroated blue warblers with and without extrapair young (EPY) in their broods. Sample sizes are indicated above standard error bars and all measures of parental care correspond to provisioning of 7-day-old nestlings. Age divisions of males correspond to older (ASY: after second year) and yearling (SY: second year) males. Parental care by ASY males with EPY was significantly less than care by ASY males without EPY (Scheffe’s test: nestling feeding rates: P=0.03; biomass delivery rates: P=0.04).

Local breeding synchrony is a measure of females available to a male for additional matings during his mate’s fertile period. Opportunities for a male to engage in additional matings with nearby females, represented by the number of neighbouring fertile females, were significantly greater during his female’s fertile period than during the parental care period (Fig. 2). This was true of both ASY (paired t test: t52 =5.27, P<0.0001) and SY males (t25 =4.45, P=0.0001). There was no association between a male’s opportunities to engage in extrapair copulations with neighbours during the fertile period and his opportunities during the period of parental care (linear regression: ASY males: R253 =0.02, P=0.26; SY males: R226 =0.006, P= 0.69). In this species, an increase in local synchrony is associated with a decrease in a male’s subsequent share of paternity (Chuang et al. 1999). Logistic regressions of the level of local synchrony on presence or absence of EPFs using data from 1995 to 1998 revealed a positive relationship between local synchrony and presence of EPFs in nests of older males (log-likelihood ratio test: 21 =4.42, P=0.03) but not in nests of SY males (21 =0.11, P=0.74).

Table 2. Comparisons of per capita provisioning of females with and without EPFs in their broods on different days of the nestling period Female nestling feeding rate (per capita) Day 4 5 7

Female biomass delivery (per capita)

EPF?*

Mean

df

t†

P

Mean

df

t

P

Present Absent Present Absent Present Absent

0.67 0.90 1.20 1.40 2.70 2.00

10

0.53

0.61

9

1.02

0.34

23

0.48

0.63

20

1.88

0.07

53

1.98

0.05

1.93 3.75 4.04 10.00 13.65 6.08

47

3.95

0.003

*Presence or absence of extrapair fertilizations in broods. †t statistic values from unpaired t tests.

87

ANIMAL BEHAVIOUR, 62, 1

Table 3. Results of analyses of covariance tests for effect of presence of extrapair young on ASY and SY male provisioning rates with male body condition as a covariate Male nestling feeding rate (per capita)

Male biomass delivery rate (per capita)

df

MS*

F

P

df

MS

F

P

ASY males EPY Male condition Residual

1 1 18

4.49 1.29 0.92

4.90 1.41

0.04 0.25

1 1 18

407.22 246.78 53.11

7.67 4.65

0.01 0.04

SY males EPY Male condition Residual

1 1 18

0.94 0.35 1.10

0.86 0.32

0.37 0.58

1 1 13

6.31 40.60 49.25

0.21 0.82

0.73 0.38

Interaction effects were not significant and were dropped from initial ANCOVAs. All analyses correspond to day 7 feeding rates. *Mean square values. Degrees of freedom and mean square values for residuals are provided.

The number of fertile female neighbours available during the fertile period of the male’s mate (i.e. local synchrony) was negatively associated with male nestling feeding rates during the subsequent nestling period. This relationship existed for ASY males only (linear regression: 30

nestling feeding rates: R2 =0.26, F1,17 =5.88, P=0.02; biomass delivery rates: R2 =0.36, F1,17 =9.56, P=0.007); provisioning by SY males was unrelated to local synchrony during their females’ fertile periods (nestling feeding rates: R2 =0.09, F1,8 =0.67, P=0.44; biomass delivery

(a)

2.5 (a) 2

10

0

40

0

1

2

3

4

5

(b)

30

20

Nestling feeding rate (trips per nestling per h)

20

Frequency

88

1.5 1 0.5 0 2.5 (b) 2 1.5 1 0.5

10

0 0

0

1

2

3

4

5

Number of fertile females Figure 2. Histogram of number of fertile female black-throated blue warblers on neighbouring territories (a) during a female’s fertile period and (b) during the parental care period. The median number of fertile female neighbours during any given female’s fertile period was two, and the median number of fertile neighbours during the nestling period was one.

Older

Yearling Male age class

Figure 3. Effects of high ( ) and low () levels of additional mating opportunities on male parental care (a) during the fertile period of a male’s mate and (b) during the nestling period. Levels of opportunity refer to values above (high) or below (low) the median number of fertile female neighbours for each period. Patterns for male per capita biomass delivery rates were similar and are not shown. Nestling feeding rates correspond to provisioning of 7-day-old nestlings.

CHUANG-DOBBS ET AL.: PATERNITY AND PARENTAL CARE

rates: R2 =0.04, F1,6 =0.20, P=0.68). Thus, levels of local synchrony were associated strongly with subsequent male parental care in ASY males (Fig. 3a); a male that had more than the median number of fertile female neighbours (median=2) when his own female was fertile was likely to provision his brood less than a male with the median number or fewer fertile neighbours during his female’s fertile period (unpaired t test: nestling feeding rates: t17 =3.26, P=0.005; biomass provisioning rates: t17 =3.06, P=0.007). As before, this relationship did not hold for yearling males (Fig. 3a; unpaired t test: nestling feeding rates: t7 =0.20, P=0.68; biomass provisioning rates: t5 =0.26, P=0.80). In contrast, there was no relationship between numbers of fertile female neighbours during the parental care period and day 7 provisioning rates by ASY males (linear regression: nestling feeding rates: R2 =0.07, F1,19 =1.37, P=0.26; biomass delivery rates: R2 =0.09, F1,19 =2.36, P=0.14) or SY males (nestling feeding rates: R2 =0.19, F1,10 =2.41, P=0.15; biomass delivery rates: R2 =0.33, F1,6 =2.90, P=0.14). We tested for differences in parental care between males surrounded by the median number of fertile females or fewer and those surrounded by more than the median number of fertile females during the parental care period (Fig. 3b): opportunities for additional matings during the parental care period were unrelated to provisioning rates in both ASY males (unpaired t test: nestling feeding rates: t19 =0.55, P=0.59; biomass delivery rates: t19 =1.18, P=0.25) and SY males (nestling feeding rates: t10 =0.73, P=0.48; biomass delivery rates: t6 =0.85, P=0.43).

Effects of Seasonality Male body condition was not associated with capture date for either ASY males (linear regression: R2 =0.13, F1,22 =0.35, P=0.56) or for SY males (R2 =0.03, F1,18 =0.013, P=0.91). Although our measures of male body condition did not vary with season, changes in male body condition from the time of capture until the nestling period were possible. There were no effects of time of season on nestling feeding rates by females (linear regression: nestling feeding: R2 =0.03, F1,59 =1.79, P=0.19; biomass delivery: R2 =0.001, F1,52 =0.09, P=0.87), by ASY males (nestling feeding: R2 =0.02, F1,33 =0.71, P=0.40; biomass delivery: R2 =0.02, F1,30 =0.61, P=0.44), or by SY males (nestling feeding: R2 =0.02, F1,22 =0.52, P=0.48; biomass delivery: R2 =0.001, F1,17 =0.017, P=0.90). Finally, there was no seasonal effect on EPF frequency (loglikelihood ratio test: 21 =0.001, P=0.94), consistent with results previously reported for this species (Chuang et al. 1999). DISCUSSION Extrapair fertilizations, which occur at relatively high frequency in black-throated blue warblers, were associated with a decrease in male parental care and an increase in female parental care. However, reductions in paternal care only occurred with older males and not SY males.

Recent controversy over the value of correlational approaches in examining relationships between a male’s level of parental care and share of paternity (e.g. Kempenaers & Sheldon 1998; Lifjeld et al. 1998; Wagner et al. 1998) have arisen in part because of possible covariation between parental care and paternity with other factors (Kempenaers & Sheldon 1997). Thus, because our results were based on correlational findings, we also examined three separate hypotheses that may account for the relationship between paternity and feeding rates observed for ASY males. First, both the male quality hypothesis and the differential allocation hypothesis predict that the condition of ASY males will correlate with the presence of EPF in a brood due to either the ability of a male to guard his mate or the likelihood of a female to seek EPFs. Because SY males did not show a relationship between paternity and parental care, these hypotheses further predict that these relationships will not hold for SY males. In this study, ASY males in better condition were more likely to have full shares of paternity in their broods than ASY males in poorer condition, whereas condition of SY males did not affect their share of paternity. Evidence from other species suggest a balance of male and female control over copulations (e.g. Currie et al. 1998, 1999; Osorio-Beristain & Drummond 1998) and female choice of extrapair copulatory partners may depend on the quality of the female’s social mate (Burley et al. 1996; Kempenaers et al. 1997). In the present study, despite having higher paternity, males in better condition followed their females less than males in poorer condition, and proximity of a male to his mate was not related to male condition. Therefore male mate guarding, although shown to affect paternity (Chuang-Dobbs et al., in press), is unlikely to be the only factor influencing extrapair siring success. Although we did not test female choice directly, females mated to high-quality males were both unlikely to have EPFs in their nest and unlikely to be guarded closely by their mates. These results suggest that female choice of withinand extrapair mates affects EPF, with male quality being associated positively with paternity. The male quality hypothesis also requires that male condition be positively related to levels of provisioning. Parent health, quality or condition has been shown to affect the provisioning of nestlings in some cases (Erikstad et al. 1997) but not in others (Freeman-Gallant 1997; Moreno et al. 1997; Wagner et al. 1997). In the present study, we also found that male body condition was significantly related to one measure of parental care (biomass delivery by ASY males). However, in contrast to the prediction of the male quality hypothesis, our results indicate that ASY males that delivered biomass at high rates were in poorer condition than those that delivered at low rates. Therefore, the observed relationship between parental care and paternity of ASY males cannot be explained by the male quality hypothesis. Burley’s differential allocation model differs from the male quality hypothesis by predicting a negative relationship between quality of males and provisioning. Our results fit these predictions: ASY males in better body condition had lower biomass delivery rates than ASY

89

90

ANIMAL BEHAVIOUR, 62, 1

males in worse body condition. However, the differential allocation model also predicts that high-quality males will have (1) high paternity in their own nests (due to female control of copulations) and (2) low provisioning rates at these nests (due to larger parental investment by females), and vice versa for low-quality males. Instead, our results show a positive correlation between paternity and provisioning by ASY males. In addition, females fed more at nests belonging to males of lower quality (i.e. males that had been cuckolded) than they did at nests belonging to males of higher quality. These combined results run contrary to predictions of the differential allocation model. The effort shift hypothesis suggests that opportunities for additional matings may covary with paternity and parental care such that a positive relationship appears between the two. In this species, the majority of extrapair copulations occur between neighbours (Chuang 1999; Webster et al., in press), and similar patterns have been reported in other passerines (Dunn et al. 1994; Hasselquist et al. 1996; Perreault et al. 1997; Stutchbury et al. 1997). Numbers of fertile female neighbours are therefore likely to be indicative of a male’s opportunities for additional matings. The effort shift hypothesis would have been supported if these additional mating opportunities were related separately to both paternity and parental care. Additional mating opportunities during the fertile period of a male’s mate (i.e. local synchrony) were indeed associated with lowered shares of paternity for ASY males, as predicted. However, we did not find further support for the effort shift hypothesis because mating opportunities were not associated with levels of paternal care. This may have been due to timing within the breeding season; numbers of fertile females in neighbouring territories were far greater during the fertile period of a male’s mate than during the nestling period, and overall, additional mating opportunities during the fertile period were unrelated to opportunities during the nestling period. Our results suggest that male parental behaviour may be adjusted in response to indirect cues to a male’s share of paternity (i.e. a facultative response). Parental care rates of ASY males, although unrelated to additional mating opportunities during the nestling period, were strongly associated with extrapair mating opportunities earlier, during their mates’ fertile periods. High local synchrony is a good predictor of lowered levels of paternity (Chuang et al. 1999), but results presented here show that this holds for ASY but not SY males. Males of other species have been shown to use access to their social mates as an indirect cue of paternity (Burke et al. 1989; Davies et al. 1992; Sheldon & Ellegren 1998; Whittingham & Dunn 1998). The number of chases or the amount of ‘interest’ that a female receives from other males (Møller 1988) or displays towards potential extrapair mates during her fertile period also may be used by her mate to assess his share of paternity. In general, high local synchrony may result in an increase in encounter rates between males, females and intruders. In addition, males may foray off of their territories more and guard their own mates less when fertile females are present in

neighbouring territories (Westneat 1988; Chuang-Dobbs et al., in press). Therefore, increased encounters with neighbouring males and decreased attendance of the female resulting from different degrees of local synchrony could potentially serve as reliable cues to an older male’s share of paternity and allow for adjustment of subsequent care. Theoretical models predict that a facultative response requires not only reliable cues to a male’s share of paternity, but also for benefits of nonparental activities to outweigh costs of reduced care for young (Whittingham et al. 1992; Westneat & Sherman 1993). Costs of reduced care will be lessened if other individuals (e.g. the male’s mate) compensate for the male’s reduced care. In the present study, females had higher provisioning rates at nests with EPFs than without, suggesting at least partial compensation by females for decreases in male care. In addition, the relationship between paternal care and paternity in ASY males and the increase in female parental care with presence of EPFs occurred late but not early in the nestling period. These latter results may have been due to the lower power of analyses of data from days 4 and 5 arising from smaller sample sizes. Alternatively, in this species, demands of nestlings and parental expenditure increase with nestling age (Goodbred & Holmes 1996) and costs of paternal care are likely to be low early in the nestling period of each social pair. Why was a relationship between paternity and parental care not found in yearling (SY) males? Whereas caring for unrelated young may be beneficial under certain conditions (e.g. Gori et al. 1996; Freeman-Gallant 1997), we suggest a simpler explanation for this species. Although a relationship between local synchrony and paternity existed for ASY males, this was not the case for SY males. Thus, unlike older males, SY males might not have had a dependable cue by which to assess their shares of paternity. Reasons underlying the lack of association between local synchrony and paternity for SY males are unclear. In general, paternity may be related to local synchrony due to trade-offs between pursuit of additional matings and mate guarding, optimization of female choice of extrapair mates, or both (Westneat et al. 1990; Stutchbury & Morton 1995; Slagsvold & Lifjeld 1998; Chuang et al. 1999). Skill at breeding may be honed with age (Nur 1984) and SY males, being inexperienced breeders, may show high variance in their abilities to guard mates, pursue additional matings, or care for young. If this is the case, then interactions and behaviours between individuals arising from high local synchrony may not affect paternity of SY males in a predictable manner and thus may not be a reliable cue to paternity. In summary, male facultative responses to paternity may be complex and responses may differ between groups of individuals. Although covariation of paternity and parental care with male quality and extrapair mating opportunities are possible, we did not find this to be the case in our study. Difficulties may arise from correlational studies (Kempenaers & Sheldon 1997, 1998), but problems may arise from experimental tests of the effects of a male’s perceived paternity and subsequent parental care as well (Jamieson & Quinn 1997). For future

CHUANG-DOBBS ET AL.: PATERNITY AND PARENTAL CARE

examinations of the relationship between parental care and paternity, we therefore stress the careful examination of potential covarying factors. Acknowledgments The field component of this study would not have been possible without T. S. Sillett, J. J. Barg, and numerous field assistants from the University at Buffalo and Dartmouth College. Field work was conducted in the Hubbard Brook Experimental Forest, which is administered by the U.S. Forest Service, Northeast Research Station, Radnor, Pennyslvania, U.S.A. H. A. Murphy and K. S. Phipps provided invaluable laboratory assistance. We thank M. A. Coffroth, R. C. Dobbs, H. R. Lasker, P. G. Parker, D. J. Taylor and two anonymous referees for suggestions that have improved this manuscript. This study was supported by National Science Foundation grants to the State University of New York at Buffalo and to Dartmouth College. The research presented here was described in Animal Research Protocol No. A-3354-01, approved on 10 March 1997 by the Institutional Animal Care and Use Committee of the University at Buffalo. References Altmann, J. 1974. Observational study of behavior: sampling methods. Behaviour, 49, 227–267. Birkhead, T. R. & Møller, A. P. 1992. Sperm Competition in Birds. London: Academic Press. Burke, T., Davies, N. B., Bruford, M. W. & Hatchwell, B. J. 1989. Parental care and mating behaviour of polyandrous dunnocks Prunella modularis related to paternity by DNA fingerprinting. Nature, 338, 249–251. Burley, N. 1988. The differential-allocation hypothesis: an experimental test. American Naturalist, 132, 611–628. Burley, N., Parker, P. G. & Lundy, K. 1996. Sexual selection and extrapair fertilization in a socially monogamous passerine, the zebra finch (Taemiopygia guttata). Behavioral Ecology, 7, 218–226. Chuang, H. C. 1999. Extra-pair fertilizations and parental care in the black-throated blue warbler (Dendroica caerulescens): patterns and consequences of behavior. Ph.D. thesis, University at Buffalo. Chuang, H. C., Webster, M. S. & Holmes, R. T. 1999. Extrapair paternity and local synchrony in the black-throated blue warbler. Auk, 116, 726–736. Chuang-Dobbs, H. C., Webster, M. S. & Holmes, R. T. In press. The effectiveness of mate guarding by male black-throated blue warblers. Behavioral Ecology. Currie, D. R., Burke, T., Whitney, R. L. & Thompson, D. B. A. 1998. Male and female behaviour and extra-pair paternity in the wheatear. Animal Behaviour, 55, 689–703. Currie, D. R., Krupa, A. P., Burke, T. & Thompson, D. B. A. 1999. The effect of experimental male removals on extrapair paternity in the wheatear, Oenanthe oenanthe. Animal Behaviour, 57, 145– 152. Davies, N. B., Hatchwell, B. J., Robson, T. & Burke, T. 1992. Paternity and parental effort in dunnocks Prunella modularis: how good are male chick-feeding rules? Animal Behaviour, 43, 729–745. Dawson, R. J. G., Gibbs, H. L., Hobson, K. A. & Yezerinac, S. M. 1997. Isolation of microsatellite DNA markers from a passerine bird, Dendroica petechia (the yellow warbler), and their use in population studies. Heredity, 79, 506–514.

Dunn, P. O., Robertson, R. J., Michaud-Freeman, D. & Boag, P. T. 1994. Extra-pair paternity in tree swallows: why do females mate with more than one male? Behavioral Ecology and Sociobiology, 35, 273–281. Erikstad, K. E., Asheim, M., Fauchald, P., Dahlhaug, L. & Tveraa, T. 1997. Adjustment of parental effort in the puffin; the roles of adult body condition and chick size. Behavioral Ecology and Sociobiology, 40, 95–100. Freeman-Gallant, C. R. 1997. Parentage and paternal care: consequences of intersexual selection in savannah sparrows? Behavioral Ecology and Sociobiology, 40, 395–400. Goodbred, C. O. & Holmes, R. T. 1996. Factors affecting food provisioning of nestling black-throated blue warblers. Wilson Bulletin, 108, 467–479. Gori, D. F., Rohwer, S. & Caselle, J. 1996. Accepting unrelated broods helps replacement male yellow-headed blackbirds attract mates. Behavioral Ecology, 7, 49–54. Hartley, I. R., Davies, N. B., Hatchwell, B. J., Desrochers, A., Nebel, D. & Burke, T. 1995. The polygynandrous mating system of the alpine accentor, Prunella collaris. II. Multiple paternity and parental effort. Animal Behaviour, 49, 789–803. Hasselquist, D., Bensch, S. & vonSchantz, T. 1996. Correlation between male song repertoire, extra-pair paternity and offspring survival in the great reed warbler. Nature, 381, 229–232. Holmes, R. T. 1994. Black-throated blue warbler (Dendroica caerulescens). In: The Birds of North America. No. 87 (Ed. by A. Poole & F. Gill), pp. 1–24. Philadelphia: Academy of Natural Sciences; Washington, D.C.: American Ornithologists’ Union. Jakob, E. M., Marshall, S. D. & Uetz, G. W. 1996. Estimating fitness: a comparison of body condition indices. Oikos, 77, 61–67. Jamieson, A. 1994. The effectiveness of using co-dominant polymorphic allelic series for (1) checking pedigrees and (2) distinguishing full-sib pair members. Animal Genetics Supplement (1), 25, 37–44. Jamieson, I. G. & Quinn, J. S. 1997. Problems with removal experiments designed to test the relationship between paternity and parental effort in a socially polyandrous bird. Auk, 114, 291–295. Kempenaers, B. & Sheldon, B. C. 1997. Studying paternity and paternal care: pitfalls and problems. Animal Behaviour, 53, 423– 427. Kempenaers, B. & Sheldon, B. C. 1998. Confounded correlations: a reply to Lifjeld et al. and Wagner et al. Animal Behaviour, 55, 241–244. Kempenaers, B., Verheyen, G. R. & Dhondt, A. A. 1997. Extrapair paternity in the blue tit (Parus caeruleus): female choice, male characteristics, and offspring quality. Behavioral Ecology, 8, 481– 492. Lack, D. 1968. Ecological Adaptations for Breeding in Birds. London: Methuen. Lifjeld, J. T., Anthonisen, K., Blomqvist, D., Johnsen, A., Krokene, C. & Rigstad, K. 1998. Studying the influence of paternity on paternal effort: a comment on Kempenaers & Sheldon. Animal Behaviour, 55, 235–238. MacDougall-Shackleton, E. A., Robertson, R. J. & Boag, P. T. 1996. Temporary male removal increases extra-pair paternity in eastern bluebirds. Animal Behaviour, 52, 1177–1183. Magrath, M. J. L. & Elgar, M. A. 1997. Paternal care declines with increased opportunity for extra-pair matings in fairy martins. Proceedings of the Royal Society of London, Series B, 264, 1731– 1736. Møller, A. P. 1988. Paternity and paternal care in the swallow, Hirundo rustica. Animal Behaviour, 36, 996–1005. Moreno, J., Barbosa, A., Potti, J. & Merino, S. 1997. The effects of hatching date and parental quality on chick growth and creching age in the chinstrap penguin (Pygoscelis antarctica): a field experiment. Auk, 114, 47–54.

91

92

ANIMAL BEHAVIOUR, 62, 1

Nur, N. 1984. Increased reproductive success with age in the California gull: due to increased effort or improvement of skill? Oikos, 43, 407–408. Osorio-Beristain, M. & Drummond, H. 1998. Non-aggressive mate guarding by the blue-footed booby: a balance of female and male control. Behavioral Ecology and Sociobiology, 43, 307– 315. Perreault, S., Lemon, R. E. & Kuhnlein, U. 1997. Patterns and correlates of extrapair paternity in American redstarts (Setophaga ruticilla). Behavioral Ecology, 8, 612–621. Pyle, P. S., Howell, N. G., Yunick, R. P. & DeSante, D. R. 1987. Identification Guide to North American Passerines. Bolines, California: Slate Creek Press. Riley, H. T., Bryant, D. M., Carter, R. E. & Parkin, D. T. 1995. Extra-pair fertilizations and paternity defence in house martins, Delichon urbica. Animal Behaviour, 49, 495–509. Ritchison, G., Klatt, P. H. & Westneat, D. F. 1994. Mate guarding and extra-pair paternity in northern cardinals. Condor, 96, 1055– 1063. Rodenhouse, N. L. 1986. Food limitation for forest passerines: effects of natural and experimental food reductions. Ph.D. thesis, Dartmouth College. SAS Institute 1998. Statview Guide. 2nd edn. Cary, North Carolina: Statview. Sheldon, B. C. & Ellegren, H. 1998. Paternal effort related to experimentally manipulated paternity of male collared flycatchers. Proceedings of the Royal Society of London, Series B, 265, 1737– 1742. Slagsvold, T. & Lifjeld, J. T. 1997. Incomplete female knowledge of male quality may explain variation in extra-pair paternity in birds. Behaviour, 134, 353–371. Stutchbury, B. J. M. & Morton, E. S. 1995. The effect of breeding synchrony on extra-pair mating systems in songbirds. Behaviour, 132, 675–690. Stutchbury, B. J. M., Piper, W. H., Neudorf, D. L., Tarof, S. A., Rhymer, J. M., Fuller, G. & Fleischer, R. C. 1997. Correlates of extra-pair fertilization success in hooded warblers. Behavioral Ecology and Sociobiology, 40, 119–126.

Wagner, R. H., Davidar, P., Schug, M. D. & Morton, E. S. 1997. Do blood parasites affect paternity, provisioning and mate-guarding in purple martins? Condor, 99, 520–523. Wagner, R. H., Schug, M. D. & Morton, E. S. 1998. Studying paternity and paternal care: the value of negative results. Animal Behaviour, 55, 239–240. Webster, M. S., Chuang-Dobbs, H. C. & Holmes, R. T. In press. Microsatellite identification of extra-pair sires in a socially monogamous songbird. Behavioral Ecology. Westneat, D. F. 1988. Male parental care and extrapair copulations in the indigo bunting. Auk, 105, 149–160. Westneat, D. F. 1990. Genetic parentage in the indigo bunting: a study using DNA fingerprinting. Behavioral Ecology and Sociobiology, 27, 67–76. Westneat, D. F. 1994. To guard mates or go forage: conflicting demands affect the paternity of male red-winged blackbirds. American Naturalist, 144, 343–354. Westneat, D. F. & Sargent, R. C. 1996. Sex and parenting: the effects of sexual conflict and parentage on parental strategies. Trends in Ecology and Evolution, 11, 87–91. Westneat, D. F. & Sherman, P W. 1993. Parentage and the evolution of parental behavior. Behavioral Ecology, 4, 66–77. Westneat, D. F., Sherman, P. W. & Morton, M. L. 1990. The ecology and evolution of extra-pair copulations in birds. Current Ornithology, 7, 331–369. Whittingham, L. A. & Dunn, P. O. 1998. Male parental effort and paternity in a variable mating system. Animal Behaviour, 55, 629–640. Whittingham, L. A., Taylor, P. D. & Robertson, R. J. 1992. Confidence of paternity and male parental care. American Naturalist, 139, 1115–1125. Winkler, D. 1987. A general model for parental care. American Naturalist, 130, 526–543. Wright, J. & Cotton, P. A. 1994. Experimentally induced sex differences in parental care: an effect of certainty of paternity? Animal Behaviour, 47, 1311–1322. Xia, X. 1992. Uncertainty of paternity can select against paternal care. American Naturalist, 139, 1126–1129.