Assortative mating by multiple ornaments in northern cardinals (Cardinalis cardinalis)

Behavioral Ecology Vol. 14 No. 4: 515 520 Assortative mating by multiple ornaments in northern cardinals (Cardinalis cardinalis) Jodie M. Jawor, Susan U. Linville, Sara M. Beall, and Randall Breitwisch Department of Biology, University of Dayton, Dayton, OH 45469-2320, USA In positive assortative mating, individuals of similar phenotype mate together more frequently than expected by chance. Assortative mating by a variety of qualities, including ornamentation, is well documented in birds. Studies of assortative mating by ornaments have focused on single, highly conspicuous ornaments, but many species of birds possess multiple ornaments in both sexes. We compared ornament expressions between mates of northern cardinals (Cardinalis cardinalis) to determine if assortative mating occurred by one or more of the four ornaments displayed by both sexes. All cardinals possess tall head crests and redorange bills. In addition, males have black face masks and entirely red body plumage, whereas females have blackish face masks and red underwing coverts. We predicted that cardinals mate assortatively by plumage color because red plumage expression has been shown to indicate quality in both sexes. We found that cardinals mate assortatively by plumage and bill color, the two ornaments colored by carotenoid pigments, but not by mask expression or crest length. Whether this mating pattern arises by mutual mate choice or intrasexual selection is not known. Key words: assortative mating, Cardinalis cardinalis, carotenoids, multiple ornaments, nonrandom mating, northern cardinal. [Behav Ecol 14:515 520 (2003)] In positive assortative mating, individuals of similar phenotype or quality mate together more often than is expected by chance (Burley, 1983). This definition applies to the pattern of assortative mating, not the process by which it arises (see below). Positive assortative mating has been documented in many species of birds. This mating pattern may occur by size (Delestrade, 2001, and references therein), age (Warkentin et al., 1992, and references therein), social experience (Freeberg, 1996), color morph (Houtman and Falls, 1994), body condition (Bortolotti and Iko, 1992), parental behavior (Breitwisch, 1988; Filliater and Breitwisch, 1997; Nealen and Breitwisch, 1997), or hybrid status (Wiebe, 2000). Positive assortative mating also occurs based on ornamental characters, including feather length ornaments (Møller, 1993; Regosin and Pruett-Jones, 2001), melanin-based ornaments (Roulin, 1999), structural coloration (Amundsen et al., 1997; Andersson et al., 1998; Potti and Merino, 1996), and carotenoid-based ornaments (Hill, 1993; Mountjoy and Robertson, 1988). In all of these species, ornament expression relates to measures of individual quality, suggesting that by mating assortatively by ornamentation, individuals gain a mate of quality similar to their own. Positive assortative mating theoretically can arise by one of at least three processes. First, mating may occur via the preferential pairing of individuals of similar phenotype along the continuum of available phenotypes (Burley, 1983). Alternatively, such a mating pattern may arise via mutual male and female preferences for selection criteria (Johnstone et al., 1996). In this situation, although all individuals attempt to gain a mate of the highest quality, those of low quality are competitively constrained in mate choice. Last, intrasexual competition for territories by both sexes, in the absence of mate choice, may yield assortative mating (Creighton, 2001). The northern cardinal (Cardinalis cardinalis) is a sexually dichromatic, multiply ornamented, socially monogamous passerine. Both sexes possess head crests and red-orange bill Address correspondence to J.M. Jawor. E-mail: jmjawor@hotmail. com. Received 22 January 2002; revised 3 October 2002; accepted 25 October 2002. Ó 2003 International Society for Behavioral Ecology coloration. Males have black face masks and entirely red body plumage. Females have blackish face masks and red underwing coverts (their exposed body plumage is brown). Previous studies on cardinal ornamentation have focused on the red plumage of males and females. Wolfenbarger (1999c) found that redder males obtained higher quality mates and territories. Linville et al. (1998) found that redder females were higher quality parents. On the basis of these studies, we predicted that male and female cardinals would display positive assortative mating based on red plumage expression. The multiple ornaments of male cardinals are indicative of several aspects of individual quality (Jawor, 2002; Jawor et al., 2003). In addition, the multiple ornaments of female cardinals are indicative of individual quality and level of intrasexual aggression displayed during simulated territory intrusions (Jawor et al., 2003). On the basis of these findings, and Møller and Pomiankowski s (1993) hypotheses for the existence of multiple ornaments, we also asked whether cardinals mate assortatively by any ornament(s) in addition to red plumage color. METHODS We conducted this study at Aullwood Audubon Center and Farm (398529 N, 848169 W), 15 km northwest of Dayton, Ohio, USA, during the breeding seasons of 7 years (1994 1997 and 1999 2001). Twenty-six pairs of cardinals were sampled for expression of all ornaments (2000 2001). Eighty-one pairs of cardinals were sampled for Munsell color chip score for plumage (1994 1997 and 1999 2001). Cardinals frequently remained paired over several breeding seasons, but we used only the initial pairings in the consideration of assortative mating. In the larger data set, 16 individuals remained on the study site for several years while mating with different individuals and are represented in the data set two (n ¼ 14), three (n ¼ 1), or four (n ¼ 1) times. Repeated measurements on these individuals in successive years are considered independent from one another. We captured individuals with potter traps and mist nets during the early breeding season. Each was banded with a U.S. Fish and Wildlife aluminum band and three colored leg bands in unique combinations. We did not use red color bands because bands similar in color to

516 Behavioral Ecology Vol. 14 No. 4 a species ornament have influenced choice in other species (Burley et al., 1982). Ornament measures We recorded two plumage color ornaments, a feather length ornament, and bill color for all birds. In 2000 and 2001, we measured the color of carotenoid-based plumage ornaments with a reflectance spectrophotometer (Digital SwatchbookÒ; X-Rite; Grandville, Michigan, USA), which yielded reflected spectra of areas of plumage. Three measurements were taken of the upper breast area of males and the underwing coverts of females. Spectral data were divided into the three components of hue, saturation, and brightness. Variables were derived following the method of Endler (1990) and of Pryke et al. (2001). Briefly, hue was estimated as the wavelength at which reflectance is halfway between its minimum and maximum (k[r50]). Saturation was estimated as the difference between two spectral segments, with the segment divider defined as kr(50). This difference was then divided by the total reflectance. Brightness was estimated as the total reflectance between 400 and 700 nm. The three values for each variable were then used to calculate a mean hue, saturation, and brightness for each individual. Ultraviolet (UV) reflectance was not considered because carotenoid-based plumage displays a minimal reflectance peak in the UV part of the spectrum (Wolfenbarger, 1999b; Hill GE, personal communication). Color variables were than entered into a principal component analysis (PCA) to yield a single color score (PC I) for plumage that was used in all analyses. The first principal component explained 48% of the variation in males and 54% in females. Hue and saturation loaded positively, whereas brightness loaded negatively for both sexes. We also scored plumage color by using the Munsell color chip system. Before 2000 we did not have access to equipment that produced spectral data, and the Munsell color chip system has been considered a reliable color measurement tool (e.g., see Burley and Coopersmith, 1987; Linville et al., 1998). Because color is a three-dimensional categorization, we converted Munsell scores to a one-dimensional continuum similar to that in Burley and Coopersmith (1987). Color brightness was weighted primary, with hue secondary and saturation tertiary, to prevent the possibility of a highly saturated pink outranking a darker, less-saturated red-orange (as in Linville et al., 1998). Female cardinal face masks frequently have indistinct borders. Face mask was scored in females based on color (grayish to black) and relative size of the mask; a larger darker mask received a higher score. Male cardinal face masks have obvious borders, and area of male masks was determined from three standardized digital photographs (left and right sides and front of mask). Photographs were loaded into Adobe Photoshop, and a 1-cm grid was laid over images on the monitor via the Adobe Photoshop software. The front versus side of a mask was defined by a line tangent to both the medial margin of the eye and the posterior extent of the lower mandible. Grid squares that were more than half covered by mask were added together for area measurements. We measured crests (in millimeters) from the point of feather attachment to the end of the longest feathers. Crest length was regressed on tarsus length, yielding a predicted length for this ornament based on body size. Crest length residuals were used in all analyses of this ornament. Bill color was determined by matching the bill to the most similar Munsell color chip (see above for weighting technique). We were unable to use the reflectance spectrophotometer to measure bill color owing to the convexity and size of the bill relative to the size of the measurement window of the instrument. We determined if ornament expression changed predictably with age in cardinals by comparing ornament measures from individuals that were caught in two successive years. For Munsell color scores, we compared scores from 32 males and 30 females that were each captured twice between 1994 and 1997 or between 1999 and 2001. For crest residual length (15 males, 13 females) and mask expression (11 females), we considered individuals captured twice between 1999 and 2001. Male mask area, bill color for both sexes, and spectral color measurements for both sexes were not analyzed owing to sample sizes of fewer than 10 individuals with repeated measurements. We also present data from several 1-year-old birds banded as nestlings and recaptured the following breeding season. Very few nestlings in this open population remain to breed. Individual quality measures We collected several measures of individual quality. These included average feather growth during the previous fall molt, body size measures (wing chord and tarsus length), and body condition. Average feather growth during molt was determined by measuring 10 growth bars of the right outermost rectrix (Grubb, 1989). Individual health and nutritional status are reflected in the rate at which individuals complete molt (Grubb, 1995). Body condition was determined by linear regression of body weight on a skeletal measurement (tarsus length), yielding a predicted body weight based on structural size (Brown, 1996). Residual of body weight was used as a measure of condition. We recorded reproductive success for pairs as number of broods that successfully fledged (zero to two nests) because the primary cause of nestling mortality is predation of the entire brood (Filliater et al., 1994). We asked if more ornamented pairs have higher reproductive success by comparing ornament expression to the number of successfully fledged broods. Ornaments within each sex were ranked, and ranks for each ornament were averaged between pair members. Mean ornament ranks for pairs were compared with reproductive success. Statistical analysis We used SigmaStatÒ, version 1.0 ( Jandel Corp.), for Spearman rank correlations to compare ornament expressions, body size, and individual quality measures between mates. We used Kruskal-Wallis one-way ANOVA to analyze the association between reproductive success and pair ornamentation. Plumage reflectance spectra were analyzed by using SAS 8.0 (SAS Institute) for PCAs. We analyzed changes between years in Munsell color scores of both sexes, and in female mask rank with sign tests. Paired t tests were used for analysis of changes in crest residual lengths of both sexes. RESULTS Cardinal pairs mated assortatively by plumage color measured as reflectance spectra (r s ¼ 0:58; p ¼ :005; n ¼ 22 [incomplete data for remaining pairs]) (Figure 1). As a validation of this result, we also tested for assortative mating based on the much larger sample of Munsell color scores. Pairs also mated assortatively based on Munsell color scores, which are weighted to emphasize plumage brightness, as described above (r s ¼ 0:39; p, :0001; n ¼ 81) (Figure 2). Cardinal pairs also mated assortatively by bill color (r s ¼ 0:52; p ¼ :02; n ¼ 20) (Figure 3). Our overall hypothesis was that cardinals mated assortatively by one or more ornaments. Because any one ornament of the four would satisfy the prediction, we

Jawor et al. Assortative mating in cardinals 517 Figure 1 Breast color and underwing covert color for mated pairs of cardinals, based on scores from principal component analysis (r s ¼ 0:58; p ¼ :005; n ¼ 22). corrected the significance of the individual tests of the hypothesis by the Bonferroni method (corrected critical value, p ¼ :0125). This correction means that the mating pattern based on bill color is of marginal significance. Pairs did not mate assortatively by mask expression or crest residual length (for both, p. :05). Cardinal pairs did not mate assortatively by body size (wing chord or tarsus length), body condition, or average feather growth during fall molt (for all, p. :05). When testing whether more ornamented pairs had higher reproductive success, we considered only the two ornaments by which pairs mated assortatively, plumage color (reflectance spectra and Munsell score) and bill color. There were no significant differences in reproductive success between moreand less-ornamented pairs (for both, p. :05). There was no predictable change between successive years in Munsell plumage color scores for either sex (sign tests; n ¼ 32 males and 30 females; for both, p. :05). Birds in those samples were of unknown age. In addition, the seven 1-year-old birds (three males and four females) each displayed a Munsell plumage color score above the median score for their sex (i.e., Figure 2 Munsell color chip score of mated pairs of cardinals, scoring system described in Methods (r s ¼ 0:39; p, :0001; n ¼ 82). Figure 3 Bill colors of mated pairs of cardinals, scoring system described in Methods (r s ¼ 0:52; p ¼ :02; n ¼ 26). they showed higher expression of the ornament). Although anecdotal, these few data on known age birds suggest that 1-year-old cardinals do not predictably display duller plumage than older birds. Therefore, age apparently is not the variable responsible for the bimodality of the frequency distributions for plumage color scores for both males and females (Figure 4). Regarding other ornaments, there was no predictable changes in either female face mask expression or crest residual length (for both, p. :05), but male crest residual lengths increased with age (paired t test: t ¼ 2:17; df ¼ 14; p ¼ :05). There was significant yearly variation in plumage coloration in cardinals. The colors recorded for males across years were significantly different in an overall analysis (Kruskal-Wallis ANOVA: H ¼ 11; df ¼ 5; p ¼ :052), although median color in no single year was found to be different from any other year. The colors recorded for females across the years were also significantly different (H ¼ 14:5; df ¼ 5; p ¼ :013). Interobserver variability in color scoring was reduced by requiring agreement between two individuals (either S.U.L. and R.B., or J.M.J. and R.B.). Differences in color between years may be owing to interyear variation in carotenoid availability during molt (Linville and Breitwisch, 1997). DISCUSSION We have documented that cardinal pairs mate assortatively by both plumage and bill color. Members of cardinal pairs did not mate assortatively by face mask expression or head crest length. This demonstration of assortative mating by both plumage and bill color is the first documentation of positive assortative mating in a bird species by more than one ornament. The mating pattern observed in cardinals may arise in one of at least three ways. Positive assortative mating may arise if individuals of both sexes choose mates based on ornaments that display aspects of individual quality. Mutual mate choice is not well documented in birds, although it may be widespread (see Burley, 1981; Hunt et al., 1999; Jones and Hunter, 1993; Marzluff and Balda, 1988). Individual quality probably varies on a continuum, and all individuals attempt to obtain mates of the highest quality (Johnstone et al., 1996). However, only those of the highest quality are free to choose mates of similar high quality. Those of lower quality will have no option but to mate with others of similarly low quality. Mutual mate choice by high-quality individuals produces a pattern of positive assortative mating.

518 Behavioral Ecology Vol. 14 No. 4 Figure 4 Frequency distributions of plumage color (Munsell scores) and bill color for male and female cardinals. Both plumage and bill color are indicative of multiple aspects of individual quality in cardinals (Jawor, 2002; Jawor et al., 2003). Higher-quality individuals, as demonstrated by these ornaments, may be free to choose individuals of equal quality, producing the assortative mating pattern observed in cardinals. Alternatively, birds all along the ornamental continuum may choose mates of similar ornamentation (Burley, 1983). This process would also produce a pattern of positive assortative mating. Studies that demonstrate positive assortative mating in birds (see above) do not distinguish between these alternatives. Testing these hypotheses would involve an assessment of the mating pattern at different levels of ornamentation. The hypotheses predict assortative mating either along the ornamental continuum or only among birds of high ornamental expression ( Johnstone et al., 1996). Unfortunately, our data are both too few and too variable to test statistically these alternatives (Figures 1 3). Finally, assortative mating can arise from intrasexual competition for territories or mates, as observed in European blackbirds, Turdus merula (Creighton, 2001). If intrasexual competitive ability is signaled by ornamentation, positive assortative mating by ornamentation may independently occur if individuals of similar competitive ability settle on the same territory. Cardinals follow the typical pattern of males fighting for territories and females settling on those territories and repelling other females. Wolfenbarger (1999b) has shown that dominance among male cardinals in aviaries was associated with expression of red plumage color, and male plumage color was predictive of body size (measured as wing chord length), which, in turn, may influence competitive ability ( Jawor, 2002). Wolfenbarger (1999c) also demonstrated that redder males possessed territories with the densest vegetation at nesting heights and that these males were mated to females that bred earlier in the season. Mate-choice experiments in the laboratory, however, provided no evidence that female cardinals prefer redder males (Wolfenbarger, 1999a). Thus, it seems likely that females may be choosing territories rather than males and defending their territory choice from other prospecting females. In support of this suggestion, we have observed highly aggressive responses by female cardinals both to intruding neighboring females and to stuffed museum skins of females placed as intruders ( Jawor et al., 2003). At present, we do not know if assortative mating in cardinals is owing to intersexual or intrasexual selection or both.

Jawor et al. Assortative mating in cardinals 519 Positive assortative mating by ornamentation can reflect age differences. If ornament expression predictably increases with age and if individuals mate assortatively by age (e.g., firstyear adults versus older adults) (Marzluff and Balda, 1988; Perrins and McCleery, 1985; Warkentin et al., 1992), then a pattern of matched ornament expressions arises as a consequence. We have little data on birds of known age in this population, but anecdotal observations suggest that plumage color does not change in any predictable manner with age, thus providing no support for the mating-by-age hypothesis. Further, ornament expressions of cardinals of unknown age did not change in any predictable manner in successive years. Both bill color and plumage color (Munsell scores) display bimodal frequency distributions in which there is a group of very brightly colored birds and a separate group of less brightly colored birds for each sex (Figure 4). However, this distribution is not obviously related to age in cardinals (i.e., duller birds are not all first-year adults), further supporting the conclusion that individuals in this population are not mating assortatively by age. Multiple ornaments can convey important information to a receiver, either providing multiple messages of quality components or redundantly signaling overall quality of an individual (Jawor, 2002; Møller and Pomiankowski, 1993). Receivers may be members of the same or opposite sex. In cardinals, bill and plumage color, both of which are associated with individual quality in both males and females, can function as temporal indicators of quality. Molt in cardinals occurs once yearly in the fall, and plumage color can potentially indicate an individual s condition at that time (i.e., condition 6 months before the beginning of the breeding season). Bill color likely represents a more changeable ornament than plumage color, as in zebra finches (Taeniopygia guttata; Burley et al., 1992), and can potentially provide information on current quality. If so, then these two ornaments can together provide more complete information on the quality of an individual than either alone, as proposed by Møller and Pomiankowski (1993). What are the fitness consequences of this mating pattern? In contrast to the findings for several other species (Møller, 1993; Regosin and Pruett-Jones, 2001; Roulin, 1999), highly ornamented pairs of cardinals did not achieve higher reproductive success than that of less ornamented pairs. This population of cardinals experiences high levels of nest predation (Filliater et al., 1994), which may confound any association between pair reproductive success and ornamentation (as in Thusius et al., 2001). Cardinals did not mate assortatively by either the expression of melanin-based face masks or head crest length. Melaninbased ornaments frequently function in intrasexual interactions but apparently less so in intersexual interactions (Badyaev and Hill, 2000; but see Roulin, 1999). Avian head crests are, in general, poorly studied ornaments (however, see Brown, 1963; Jones and Hunter, 1993, 1999). In male cardinals, expressions of both of these ornaments are related to components of individual quality ( Jawor, 2002). Lande s (1980) genetic correlation hypothesis for female ornaments may apply to cardinal head crests and face masks. That is, neither may serve a function in female cardinals, and the absence of assortative mating by these two ornaments is the result. We suggest that in multiply ornamented species, ornaments that are sexually selected in both sexes can coexist with ornaments that are sexually selected in one sex and functionless in the other sex. In any case, demonstration of assortative mating in cardinals by two carotenoid-based ornaments enriches our view of the evolution of multiple ornaments in socially monogamous, biparental species of birds. This research was supported with grants from the Animal Behaviour Society, the Dayton Audubon Society, and University of Dayton Graduate and Faculty Summer Fellowships. Geoff Hill and an anonymous reviewer provided helpful comments on a early draft of this paper. We thank C. Krueger and J. Wilson of Aullwood Audubon Center and Farm for assistance and permission to use facilities and the field site. We thank K. Beal for assistance with statistical analyses and J. Endler for recommendations concerning spectral analyses. J. Banks, B. DeTardo, P. Donahoo, L. Fullencamp, B. Getsfred, N. Gray, K. Johnstone, J. LeClair, N. Mohlman, A. Schilling, B. Wallace, and K. Zaiden provided much-needed assistance in the field. This work was conducted under University of Dayton Animal Care and Use Committee approval number A3092-01, USFWS permits MB798701-0 and 22351, and Ohio DNR permits 319 and 55723. REFERENCES Amundsen T, Forsgren E, Hansen LTT, 1997. On the function of female ornaments: male bluethroats prefer colourful females. Proc R Soc Lond B 264:1579 1586. Andersson S, Örnborg J, Andersson M, 1998. Ultraviolet sexual dimorphism and assortative mating in blue tits. Proc R Soc Lond B 265:445 450. Badyaev AV, Hill GE, 2000. Evolution of sexual dichromatism: contribution of carotenoid- versus melanin-based coloration. Biol J Linn Soc 69:153 172. Bortolotti GR, Iko WM, 1992. Non-random pairing in American kestrels: mate choice versus intra-sexual competition. Anim Behav 44:811 821. Breitwisch R, 1988. Sex differences in defence of eggs and nestlings by northern mockingbirds (Mimus polyglottos). Anim Behav 36:62 72. Brown JL, 1963. Ecolgeographic variation and introgression in an avian visual signal: the crest of the Steller s jay, Cyanocitta stelleri. Evolution 17:23 39 Brown ME, 1996. Assessing body condition in birds. Curr Ornith 13, 67 135. Burley N, 1981. Mate selection by multiple criteria in a monogamous species. Am Nat 117:515 528. Burley N, 1983. The meaning of assortative mating. Ethol Sociobiol 4:191 203. Burley N, Coopersmith CB, 1987. Bill color preferences of zebra finches. Ethology 76:133 151. Burley N, Krantzberg G, Radman P, 1982. Influence of colour-banding on the conspecific preferences of zebra finches. Anim Behav 30:444 455. Burley NT, Price DK, Zann RA, 1992. Bill color, reproduction and condition effects in wild and domesticated zebra finches. Auk 109:13 23. Creighton E, 2001. Mate acquisition in the European blackbird and its implications for sexual strategies. Ethol Ecol Evol 13:247 260. Delestrade A, 2001. Sexual size dimorphism and positive assortative mating in alpine choughs (Pyrrhocorax graculus). Auk 118:553 556. Endler JA, 1990. On the measurement and classification of colour in studies of animal colour patterns. Biol J Linn Soc 41:315 352. Filliater TS, Breitwisch R, 1997. Nestling provisioning by the extremely dichromatic northern cardinal. Wilson Bull 109:145 153. Filliater TS, Breitwisch R, Nealen PM, 1994. Predation on northern cardinal nests: does choice of nest site matter? Condor 96:761 768. Freeberg TM, 1996. Assortative mating in captive cowbirds is predicted by social experience. Anim Behav 52:1129 1142. Grubb TC, 1989. Ptilochronology: feather growth bars as indicators of nutritional status. Auk 106:314 320. Grubb TC, 1995. Ptilochronology: a review and prospectus. Curr Ornith 12:89 114. Hill GE, 1993. Male mate choice and the evolution of female plumage coloration in the house finch. Evolution 47:1515 1525. Houtman AM, Falls JB, 1994. Negative assortative mating in the whitethroated sparrow, Zonotrichia albicollis: the role of mate choice and intra-sexual competition. Anim Behav 48:377 383. Hunt S, Cuthill IC, Bennett ATD, Griffiths R, 1999. Preferences for ultraviolet partners in the blue tit. Anim Behav 58:809 815. Jawor JM, 2002. Multiple ornaments and sexual selection in a socially monogamous passerine, the northern cardinal (Cardinalis cardinalis) (PhD dissertation). Dayton, Ohio: University of Dayton.

520 Behavioral Ecology Vol. 14 No. 4 Jawor JM, Gray N, Beall SM, Breitwisch R, 2003. Multiple ornaments correlate with aspects of condition and behaviour in female northern cardinals (Cardinalis cardinalis). Anim Behav (in press). Johnstone RA, Reynolds JD, Deutsch, JC, 1996. Mutual mate choice and sex differences in choosiness. Evolution 50:1382 1392. Jones IL, Hunter FM, 1993. Mutual sexual selection in a monogamous seabird. Nature 362:238 239. Jones IL, Hunter FM, 1999. Experimental evidence for mutual interand intrasexual selection favouring a crested auklet ornament. Anim Behav 57:521 528. Lande R, 1980. Sexual dimorphism, sexual selection and adaptation in polygenic characters. Evolution 34:292 305. Linville SU, Breitwisch R, 1997. Carotenoid availability and plumage coloration in a wild population of northern cardinals. Auk 114:796 800. Linville SU, Breitwisch R, Schilling AJ, 1998. Plumage brightness as an indicator of parental care in northern cardinals. Anim Behav 55:119 127. Marzluff JM, Balda RP, 1988. Pairing patterns and fitness in a freeranging population of pinyon jays: what do they reveal about mate choice? Condor 90:201 213. Møller AP, 1993. Sexual selection in the barn swallow Hirundo rustica, III: female tail ornaments. Evolution 47:417 431. Møller AP, Pomiankowski A, 1993. Why have birds got multiple sexual ornaments? Behav Ecol Sociobiol 32:167 176. Mountjoy DJ, Robertson RJ, 1988. Why are waxwings waxy? delayed plumage maturation in the cedar waxwing. Auk 105:61 69. Nealen PM, Breitwisch R, 1997. Northern cardinal sexes defend nests equally. Wilson Bull 109:269 278. Perrins CM, McCleery R, 1985. The effects of age and pair bond on the breeding success of great tits, Parus major. Ibis 127:306 315. Potti J, Merino S, 1996. Decreased levels of blood trypanosome infection correlate with female expression of a male secondary sexual trait: implications for sexual selection. Proc R Soc Lond B 263:1199 1204. Pryke SR, Andersson S, Lawes MJ, 2001. Sexual selection of multiple handicaps, I: the red-collared widowbird: female choice of tail length but not carotenoid display. Evolution 55:1452 1463. Regosin JV, Pruett-Jones S, 2001. Sexual selection and tail-length dimorphism in scissor-tailed flycatchers. Auk 118:167 175. Roulin A, 1999. Nonrandom pairing by male barn owls (Tyto alba) with respect to a female plumage trait. Behav Ecol 10:688 695. Thusius KJ, Peterson KA, Dunn PO, Whittingham LA, 2001. Male mask size is correlated with mating success in the common yellowthroat. Anim Behav 62:435 446. Warkentin IG, James PC, Oliphant LW, 1992. Assortative mating in urban-breeding merlins. Condor 94:418 426. Wiebe KL, 2000. Assortative mating by color in a population of hybrid northern flickers. Auk 117:525 529. Wolfenbarger LL, 1999a. Female mate choice in northern cardinals: is there a preference for redder males? Wilson Bull 111: 76 83 Wolfenbarger LL, 1999b. Is red coloration of male northern cardinals beneficial during the nonbreeding season?: A test of status signaling. Condor 101:655 663. Wolfenbarger LL, 1999c. Red coloration of male northern cardinals correlates with mate quality and territory quality. Behav Ecol 10: 80 90.