• Female ornaments;
  • genetic quality;
  • male mate choice;
  • QTL;
  • sexual selection


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion

Understanding the evolution of sexual ornaments, and particularly that of female sexual ornaments, is an enduring challenge in evolutionary biology. Key to this challenge are establishing the relationship between ornament expression and female reproductive investment, and determining the genetic basis underpinning such relationship. Advances in genomics provide unprecedented opportunities to study the genetic architecture of sexual ornaments in model species. Here, we present a quantitative trait locus (QTL) analysis of a female sexual ornament, the comb of the fowl, Gallus gallus, using a large-scale intercross between red junglefowl and a domestic line, selected for egg production. First, we demonstrate that female somatic investment in comb reflects female reproductive investment. Despite a trade-off between reproductive and skeletal investment mediated by the mobilization of skeletal minerals for egg production, females with proportionally large combs also had relatively high skeletal investment. Second, we identify a major QTL for bisexual expression of comb mass and several QTL specific to female comb mass. Importantly, QTL for comb mass were nonrandomly clustered with QTL for female reproductive and skeletal investment on chromosomes one and three. Together, these results shed light onto the physiological and genetic architecture of a female ornament.

Sexual selection theory proposes that male ornament expression conveys information about the fitness benefits that a male may transmit to his reproductive partner(s) directly and/or to the offspring (Andersson 1994; Cotton et al. 2004; Mead and Arnold 2004; Andersson and Simmons 2006). Although the evolutionary significance of male sexual ornaments remains debated (Cameron et al. 2003; Cordero and Eberhard 2003; Kokko et al. 2003; Pizzari and Snook 2003), the expression of many sexual ornaments is not entirely sex-limited (e.g., Amundsen 2000; Bonduriansky 2001). With the exception of some sex role-reversed species (e.g., Amundsen and Forsgren 2001), the functional significance of the female expression of bisexual ornaments has received surprisingly little attention and has often been dismissed as a functional equivalent or a nonadaptive genetic byproduct of male ornaments (Amundsen 2000). For example, under some conditions (see Rice 1984), a female ornament that is not beneficial to females may in principle evolve if the same genes also code for a male ornament that is under strong sexual selection. However, there is mounting evidence that female ornaments may influence male reproductive decisions (Jones and Hunter 1993; Bonduriansky 2001; Jones et al. 2001; Griggio et al. 2003; Roulin 2004), suggesting that females too may be selected to express some ornamentation. Recent evolutionary models have demonstrated that female sexual ornaments may evolve in the absence of sex role reversal when ornament investment reflects costly reproductive investment (Chenoweth et al. 2006; Servedio and Lande 2006). Although theories for the evolution of sexual ornaments depend on assumptions on the genetic architecture of ornament expression, very little is known about the genetics of sexual ornaments, especially for the relatively neglected female sexual ornaments.

A recent breakthrough in the study of sexual ornaments is the use of genomic tools afforded by model study species (Tomkins et al. 2004; Andersson and Simmons 2006). In particular, quantitative trait locus (QTL) analysis enables the identification of progressively refined chromosomal regions associated with phenotypic variation in traits controlled by multiple genes with promising applications to the study of sexual ornaments and sexually selected traits. This approach has been recently used to study the genetics of sexually selected traits in some model species (Mundy 2005; Hughes and Leips 2006; Rosenthal and García de Léon 2006; Yeh et al. 2006; Johns and Wilkinson 2007). QTL analysis can be particularly useful to the study of female sexual ornaments because it enables one to explore two critical issues. First, we can test for QTL preferentially associated with variation in the female expression of an ornament. The presence of genes involved in the female, but not the male, expression of a bisexual ornament would undermine the idea that female ornamentation is entirely due to the female expression of the same genes that code for the ornament in males. Second, we can explore the genetic relationship between ornament investment and reproductive investment by measuring linkage and pleiotropic effects between QTL associated with ornament expression and those associated with reproductive investment.

Here, we study the female expression of a well-known avian bisexual ornament, the comb of the fowl, Gallus gallus. Using a combination of phenotypic analyses and genomics, we establish the relationship between ornament expression and reproductive investment and the genetic architecture of this relationship in a large population intercross.

The comb is a bisexual fleshy ornament (Parker et al. 1942; Bullock and Hall 1968), consisting of layers of epidermis, dermis and central connective tissue, with steroid-dependent expression (Parker et al. 1942; Bullock and Hall 1968; Blas et al. 2006). Although the function of the male comb has received considerable attention (e.g., Parker and Ligon 2003), less is known about the female equivalent. Males preferentially copulate with, and allocate sperm to females with relatively large combs (Pizzari et al. 2003; Cornwallis and Birkhead 2007). The functional significance of this male preference may be explained by the fact that female comb mass is associated with female social status, condition (Zuk et al. 1998; Cloutier and Newberry 2000), and—crucially for the evolution of male preference—female reproductive investment (Zuk et al. 1998; Tufvesson et al. 1999; Cloutier and Newberry 2000; Pizzari et al. 2003; Cornwallis and Birkhead 2006). However, the mechanisms underlying covariance between female comb and reproductive investment remain unexplored.

Female birds incur a range of costs associated with reproductive investment. However, identifying the physiological pathways mediating these costs has proven difficult and therefore the functional significance of individual variation in reproductive investment remains unclear (Williams 1994). A critical component of female reproductive investment in birds is calcium metabolism. Calcium metabolism in egg-laying female birds is unique (Sugiyama and Kusuhara 2001). The production of calcium carbonate (CaCO3) eggshell requires calcium that originates from the diet and from skeletal stores, in the form of medullary bone (Wilson and Duff 1990). The proportion of eggshell calcium of skeletal origin varies between species of birds (Reynolds et al. 2004). In some species the calcium present in a clutch of eggs extracts about 30% of the skeletal mass (Reynolds et al. 2004), whereas in the domestic fowl, artificial selection for egg production translates into over 40% of skeletal calcium reserves being invested in egg formation over prolonged periods (Taylor 1970). This extreme investment is likely to constrain female skeletal investment and bone quality. Calcium and other minerals required for egg production are stored beneath the hard outer cortical bone and within the endosteal cavity, where they occur mainly in the form of medullary and trabecular bone, with the medullary bone arranged on a framework of trabeculae (Dacke et al. 1993). During egg production, estrogen induces differentiation of endosteal cells and reduces the number of osteoclasts on the endosteal surface (Kusuhara and Schraer 1982; Ohashi et al. 1987), causing an increase of the endosteal cavity, an accumulation of trabecular and medullary bone, and the thinning of the cortical bone, and ultimately increasing calcium availability for shell production (Fig. 1) (Cransberg et al. 2001). Therefore, egg production in birds is expected to generate a steroid-mediated trade-off between female reproductive and skeletal investment, as calcium is transferred from cortical regions, which provide structural support, to egg production via temporary medullary storage. To the extent to which comb investment reflects female reproductive quality, we expect increased female comb investment to reflect female ability to sustain skeletal costs of reproductive investment. Further, for comb mass to evolve as a reliable predictor of female reproductive investment, we expect additive genetic variation underlying female investment in comb, reproduction, and skeletal quality, and additive genetic covariance between these traits.


Figure 1. Stratec pQCT XCT photographs illustrating the differences between female femurs with high cortical bone and low trabecular bone (A), and low cortical bone and high trabecular bone (B) (pictures were taken with a 280/400 density setting, resolution 70 μm to 1 pixel, Bone A is 6.96 mm in diameter and bone B is 6.57 mm in diameter, tograph credit Andreas Kindmark).

Download figure to PowerPoint

Red junglefowl, Gallus gallus ssp., the wild ancestor of the domestic chicken, Gallus g. domesticus (Fumihito et al. 1994; Nishibori et al. 2005), evolved under intense sexual selection (Pizzari et al. 2002). Following domestication, some lines have undergone intense artificial selection for female reproductive investment (i.e., egg production, e.g., Etches 1996). We studied the F2 population generated from a large-scale intercross of two populations divergent for female reproductive investment: a red junglefowl population and a domestic line (White Leghorn), artificially selected for egg production. The high phenotypic and genetic variation resulting from the intercross of these divergent lines increases the probability of identifying informative gene regions involved in the traits of interest (Andersson 2001). Females from the domestic line produced significantly more egg mass than female junglefowl (Schütz et al. 2002). This cross therefore enables one to detect genetic differences that arise between a “wild-derived” population (which has been subject to far more recent natural selection) and a population that has undergone intense artificial selection.

In this study, we first test the phenotypic relationships between female somatic investment in comb mass, reproductive investment, and, given the hypothesized link between female reproductive investment and skeletal investment, skeletal quality. Second, we examine the genetic architecture of comb mass, using QTL analysis, by identifying QTL associated with comb mass, and by exploring the relationship between QTL associated with comb mass, and those associated with female reproductive and skeletal investment.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion


We studied the F2 progeny (n= 811) generated from the crossing of a White Leghorn line artificially selected for egg mass (SLU13, with egg mass defined as an increase in both the number of eggs produced as well as the weight of eggs produced) and red junglefowl. Detailed information on the origin and management of the two parental populations and their intercross is presented in Schütz et al. (2002). Basic characteristics of the parental populations are described in Schütz et al. (2002) and Kerje et al. (2003). The F2 birds were hatched in six batches between May and December 1999, and raised indoors under standardized conditions at the research station of the Swedish University of Agricultural Sciences, Skara. The F2 population has been analyzed for various phenotypic traits (e.g., Keeling et al. 2004), including fecundity and skeletal characteristics (e.g., Kerje et al. 2003; 2004; Rubin et al. 2006; Wright et al. 2006). Due to time and logistic constraints, only a random subset of the whole F2 population was used for this study.


Comb somatic investment

Comb mass is difficult to measure reliably in live birds. Therefore, we measured comb mass postmortem, by surgically removing and weighing the comb to the nearest 0.01 g on a total of 474 F2 birds (122 females, 352 males), and five parental females for each population. All birds were sacrificed at 230 days of age and stored in freezers at −20° C.

Female reproductive investment

We measured female reproductive investment in three ways: (1) the number, (2) mean mass, and (3) total mass of the eggs produced by a female over a one week period (Schütz et al. 2002; Kerje et al. 2003). Egg mass is a critical determinant of embryo viability and chick survival across different species of birds (e.g., Bolton 1991; Williams 1994; Wagner and Williams 2007). In the fowl, egg mass is a particularly important predictor of embryo mass in the second half of the incubation period, and chick body mass at hatching (Wilson 1991). Therefore, the product of clutch size and the mass of individual eggs, or the total (cumulative) mass of eggs produced by a female can be an informative measure of female reproductive investment. Reproductive investment data were available for a total of 377 F2 females. Of the 377 F2 females for which reproductive investment was measured when they were alive, 122 were also phenotyped for comb mass post-mortem.

Skeletal investment

A total of 37 different measures of bone density, strength, and composition were taken postmortem in 337 F2 birds (159 females, 178 males; Rubin et al. 2006). Four of these measures: (1) bone mineral density of the entire carcass (i.e., total bone mineral density) measured by dual energy x-ray absorptiometry (DXA bmd), (2) bone mineral content of the entire carcass (i.e., total bone mineral content) measured by DXA (DXA bmc), (3) endosteal circumference, and (4) periosteal circumference of the femoral midshaft (the latter two measured by peripheral Quantitative Computerized Tomography) were selected a priori to test a trade-off between female reproductive and skeletal investment. In addition, we performed: (1) one principal component analysis (PCA) to summarize the cortical content, density, area and thickness of the femoral bone, with the first axis explaining 78% of the variation present in females and with an eigenvalue of 3.12 (cortical pca); (2) one PCA to summarize metaphyseal area and trabecular bone area of the femur, with the first axis explaining 83% of the variation in females and with an eigenvalue of 1.67 (trabecular area pca); and (3) a PCA to summarize metaphyseal bone mineral density, metaphyseal bone mineral content and trabecular bone mineral density of the femur, with a first axis explaining 58% of the variation present in females and with an eigenvalue of 1.75 (trabecular bmd pca). The first axes of each of these PCAs (1–3) (i.e., cortical pca, trabecular area pca, trabecular bmd pca) were also used as additional measures of skeletal investment. Skeletal investment was measured in 102 of the 122 F2 females for which both reproductive and comb investments were measured.

Correlations between phenotypic traits were performed using a General Linear Model (GLM) in SPSS version 10.1. Batch (i.e., cohort) was included as a factor and body mass was included as a covariate. Due to nonnormality of the error distribution, a permutation procedure with 1000 replicates was used to establish significance thresholds, using R (ver. 2.2). Similarly, GLMs and permutation tests were also used to check for correlations when determining the degree of multiple testing corrections that were required, with Bonferroni corrections applied for all noncorrelated traits that were used in the analysis. Phenotypic analysis was performed using comb mass, the three measures of female reproductive traits, and the seven skeletal traits, as detailed above. Reproductive investment and comb analysis was performed on females only. Correlations between skeletal traits were calculated for males and females separately.


DNA samples were available for all but four of the females for which comb, reproductive, and skeletal investment was measured. DNA preparation and QTL analysis followed Kerje et al. (2003). The number of markers was increased from 104 in Kerje et al. (2003) to 160 in the present study, which covered 29 autosomes and the Z chromosome, with the total map distance covered increasing to 3356 cM with an average interval distance of 21.0 cM. A full list of markers used is available in Wright et al. (2006). The skeletal investment QTLs including the four skeletal traits used to measure skeletal investment (DXA bmd, DXA bmc, endosteal circumference, periosteal circumference) and the bone measurements used to derive three PCAs (see phenotypic analysis above) had been calculated previously (see Rubin et al. 2006). The fecundity QTLs had been calculated previously using the reduced 104 marker map (Kerje et al. 2003).

QTL analysis consisted of both univariate and multivariate analyses. Univariate analysis was performed using standard interval mapping (through qtl express – as well as epistatic analysis (as detailed in Carlborg et al. 2000). Two-QTL analyses were also conducted on traits with more than one QTL present on a chromosome. This approach used qtl express to test the presence of two QTL versus no QTL and two QTLs versus one QTL. As in the phenotypic analysis, batch and sex were entered as fixed effects in all QTL models, although body weight at 200 days of age was entered as a covariate. Differences between sexes were tested using a “QTL × sex” interaction model, fitted using the qxpak version 2.16 software package (Perez-Enciso and Misztal 2004), with each QTL fitted within each sex. Where this resulted in a greater significance than the standard analysis with sex as a fixed effect only, the two models (i.e., “QTL × sex interaction” vs. “sex as fixed effect only”) were tested using a likelihood ratio (LR) test with one degree of freedom to detect a significant sex difference.

We controlled for the possibility of allometric confounds in the comb QTL analysis in two ways. First, we performed a bivariate analysis of comb mass, with body mass included as an additional trait rather than a covariate, using qxpak version 2.16. This revealed no change in the LR, confirming that the comb mass QTLs detected were independent of body mass. Second, we controlled for the possibility of spurious effects due to nonlinear allometric effects by replicating the comb QTL analyses using log-transformed comb mass. Rather than reduce the effect size of the comb QTL, this analysis did not change the results qualitatively but resulted in a larger effect size for the female comb QTL at 193 cM on chromosome 1 (from LOD 2.6 to 3.3), a small increase for the female comb QTL at 87 cM on chromosome 1 (from LOD 3.4 to 3.5), and no changes for the other QTL. In the case of trabecular bone density, total egg mass and body mass were initially included as covariates, but removed because they had no significant effect on any of the QTL.

Multivariate analysis was also used to test for possible effects of pleiotropy between the traits being analyzed. These tests were incorporated using qxpak version 2.16. When contrasting pleiotropic versus linkage models, traits on each chromosome were taken in a pairwise fashion, and the likelihood of two QTLs (one for each trait) was compared to the likelihood of only one QTL affecting both traits (as indicated in Perez-Enciso and Misztal 2004), using a LR test with one degree of freedom.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion


In this section, we analyze the relationship between female somatic investment in comb, reproductive, and skeletal investment.

Consistent with the idea that increased female somatic investment in the comb is associated with increased reproductive investment, somatic investment in comb mass was higher in females of the domestic parental population with higher reproductive output than in female junglefowl (Table 1, Mann–Whitney U-test between populations for female comb mass, Z=−2.61, P= 0.009, n1= 5, n2= 5).

Table 1.  Trait means ± SD of the red junglefowl (RJF), White Leghorn (WL) and F2 populations for the traits used in the QTL analysis.
Comb mass (g) 1.1±0.3 7.3±0.6 1.8±1.0 
Egg number (no./wk) 2.6±2.3 6.0±1.7 4.8±1.6 
Mean egg mass (g)  23±19.857.5±15.243.2±11.1 
Total egg mass (g/wk)97.3±96.6367.1±109.6221.9±77.8 
Endosteal circumference (mm)14.7±1.113.3±4.316.4±1.617.8±1.4  19.7±1.5 17.8±1.6
Periosteal circumference (mm)20.0±1.322.3±1.620.3±1.423.4±1.3  25.3 ± 1.2 23.2±1.5
DXA bmc (mg/mm)  35±8 72 ± 19 39±8  56±6    71±3  56±9
DXA bmd (mg/cm3) 331±34467±78329±35 376±20   389 ± 40 383±33
Body mass (g)799.5±130.11629.3±110.4981±1351119.1±136.32107.2±148.31262±168

In the F2 generation of the intercross, somatic investment in female comb was strongly associated with some measures of female reproductive investment, explaining a significant proportion of both, the number (GLM with female body mass as covariate, F1, 114= 17.0, P< 0.001, R2= 0.20, B= 1.79, Fig. 2A) and total mass (F1, 114= 9.9, P= 0.002, R2= 0.28, B= 73.2, Fig. 2B) of eggs produced by a female, but not mean egg mass (F1, 114= 2.21, P= 0.14, B=−0.69). The fact that the relationships were better described by a logarithmic function (increasing the R2 from 0.17 to 0.20 for egg number, and from 0.26 to 0.28 for total egg mass) indicates that further somatic investment in comb beyond a certain threshold investment translates into diminishing fecundity increments.


Figure 2. Female comb mass predicts female fecundity. There was a significant curvilinear positive relationship between female somatic investment in comb mass (here visually expressed as comb mass over body mass) and both, the number of eggs produced (A), and total egg mass produced (B). We analyzed these relationships with GLMs in which comb mass and body mass were entered as covariates and batch as a factor. Zero values of egg number/mass were excluded from the analysis, their inclusion however, strengthened the results of the analysis.

Download figure to PowerPoint

We further explored the significance of female reproductive investment by analyzing its effect on skeletal investment. Endosteal circumference was positively correlated with the trabecular area pca and negatively correlated with the cortical pca (Table 2A). The hypothesis of a trade-off was further confirmed by the fact that the trabecular area pca covaried positively but DXA bmd covaried negatively with mean egg mass, although endosteal circumference covaried positively with both total egg mass and egg number (Table 2B). Together, these results indicate that, as female reproductive investment increases, calcium is stripped off the cortical bone and accumulated as medullary/trabecular bone as temporary store for egg production, resulting in thinner bones with larger endosteal cavities.

Table 2.  (A) Correlations between skeletal measurements associated with female reproductive investment for females (above) and males (below diagonal line). Note that these are provided to give relative strengths of correlation and are corrected for body mass, the actual probability was calculated using a GLM and bootstrapping procedure with body mass also included as a covariate (see methods). (B) Association (F-value and regression coefficient (when significant) between skeletal traits and female reproductive investment are presented.
 DXA bmdDXA bmcendost. circ.periost. circ.trab. area pcatrab. bmd pcacortical pcabody mass
  1. NS=P>0.05, *P<0.05, **P< 0.01, ***P<0.001.

  2. 1Trait that was significant prior to the addition of other factors to the model.

  3. DXA bmd, whole body DXA bone mineral density; DXA bmc, whole body DXA bone mineral content; periost. circ., periosteal circumference; endos. circ., endosteal circumference; trab. area pca, trabecular area first principal component; trab. bmd pca, trabecular bone mineral density first principal component; cortical pca, cortical bone first principal component.

  DXA bmd0.73***− 0.4***0.5***0.32***
  DXA bmc0.70*** 0.080.37***0.31*** 0.26***0.43***0.50***
  Endost. circ.***0.59***−0.38***−0.30***0.47***
  Periost. circ.0.200.24**0.91*** 0.64***−0.070.35***0.57***
  Trab. area pca0.120.23*0.46***0.48***−0.31***0.070.35***
  Trab. bmd pcaNANANANANA 0.46***0.2**
  Cortical pca0.32***0.34***−0.20**0.28***0.07NA0.2**
  Body mass0.53***0.79***0.64***0.72***0.62***NA0.42*** 
  Mean eggF=6.5** F=11.4*** F=34.2***
 10.73  0.48  0.003
  Total egg F=16.0*** 
  Egg numberNSNSF=13.1***NS1NSNSNSNS1

Importantly, after controlling for body mass and egg number, a small but significant portion of residual variance in female cortical pca was explained by female comb mass (F1, 97= 5.1, P= 0.019, R2= 0.12), suggesting that females with relatively large combs were able to maintain relatively thick cortical bones despite high egg production, consistent with the idea that somatic investment in comb reflects female ability to sustain a trade-off between skeletal and reproductive investment.


In this section we: (1) analyze the genetic architecture of female somatic investment in male comb mass, by identifying QTL associated with sex-specific and bisexual somatic investment in comb; and (2) identify QTL for skeletal investment and female reproductive investment, and test whether they are linked to QTL controlling female comb mass.

Comb QTL

We found two female-specific comb QTLs on chromosome 1 (at 87 and 193 cM, respectively, Table 3). We tested for the possibility that these two QTL may in fact reflect only one single locus, by performing two-QTL tests to ascertain the significance of these loci simultaneously. In the case of the female comb QTLs on chromosome 1, both the two- versus no-QTL test (F= 16.3) and the two- versus one-QTL test (F= 13.7) were significant at the genome-wide threshold, with an overall LOD score of 6.2, rejecting the hypothesis that the observed results are better explained by a single QTL. In addition, the position of both female comb mass QTLs was unaltered by the two-QTL analysis.

Table 3.  QTL for comb mass, female reproductive, and skeletal investment and adult body mass in the red junglefowl × domestic F2 cross. Positive additive effects indicate greater effects in red junglefowl alleles, negative indicate greater effects in domestic (White Leghorn) alleles. The female comb mass QTL on chromosome 1 is displayed, (A) without, and (B) with total egg mass entered as covariate in the analysis.
TraitChromosomeposition (cM)Additive model LODAdditive and dominance model LODR2Additive ± SEDominance ± SEConfidence interval (cM)N
  1. Confidence interval is calculated by the region indicated by a one-LOD drop of the highest LOD score.

  2. *Significant at the 5% genome-wide level of significance, **Significant at the 1% genome-wide level of significance.

  3. R2=Residual variance explained by QTL.

  4. 1Suggestive at the 20% genome-wide level of significance.

  5. 2Significant at the 1% single position level of significance.

Female comb mass11932.812.60.10  0.42±0.120.01±0.22174 – 216118
Trabecular bmd11383.8*4.6*0.14  0.03±0.01−0.02±0.01122 – 152147
Total egg mass11282.9*3.110.04  21.5±5.911.4±9.7112 – 162373
Body mass 200 days11095.3**5.5**0.03−44.1±8.9−17.1±12.0106 – 118792
Female comb mass-A1 873.4*3.6*0.14  0.55±0.13−0.19±0.19 75 – 118118
Female comb mass-B1 824.1**4.4**0.17 118
Endosteal circumference1 744.9**5.0*0.07−0.73±0.160.10±0.23 62 – 88332
 (female effect) 
Body mass 200 days1 7258.9**61.7**0.31−143.1±7.944.6±12.2 68 – 76792
Mean egg mass1 669.8**11.9**0.14−2.77±0.381.69±0.53 58 – 72373
Mean DXA bmd31023.114.110.06  0.01±0.004−0.02±0.01 79 – 112308
Female comb mass3 672.022.020.08−0.41±0.13−0.1±0.25 53 – 91118
Male comb mass3 6611.2**11.4**0.15−5.44±0.74−1.06±1.30 53 – 71338
Mean egg mass3 523.5*3.510.04−1.43±0.360.16±0.75 29 – 86373
Male comb mass8 772.912.80.04−3.63±0.990.23±2.21 48 – 103338

The lack of any significant male QTL in the same region would suggest that the two female comb mass QTL on chromosome 1 are female specific. To test female specificity, a QTL × sex interaction analysis was performed on both male and female data combined, with the fitted QTL varying within sex. This model was then compared to a model with sex only included as a fixed effect, with a LR test used to check significance. In the case of the female comb mass QTL at 87 cM, a strongly significant effect was found (LR difference = 37.8, P < 0.001), although in the case of the suggestive female comb mass QTL at 194 cM the effect was also significant (LR difference = 7.7, P= 0.006). Both of these female comb mass QTLs had alleles derived from the red junglefowl line giving a greater increase in comb mass (Table 3; Fig. 3).


Figure 3. QTL for female investment in comb mass, reproductive, and skeletal quality on Chromosome 1. The y-axis covers the total length of the chromosome (498 cM), the x-axis represents percentage of the F2 population mean. The region containing a QTL cluster has been exploded for greater clarity. QTL effects are expressed as percentage of the F2 population mean, with twice the additive variance of each QTL used in the calculation (i.e., the full difference existing between lines). Effects greater in the red junglefowl direction (red) are positive, whereas those for the domestic line (blue) are negative. Error bars indicate the QTL confidence interval as defined by a drop in LOD score by 1.

Download figure to PowerPoint

A major QTL for male comb mass was located on chromosome 3 (66 cM, Table 3), explaining more than 14% of the total variation present in male comb mass. Due to the large effect of this QTL, the same position in females was analyzed with a single position threshold, which revealed that a significant QTL for female comb mass also occurred at this position, suggesting that this QTL may be associated with variation in the bisexual expression of comb mass in this intercross.

Relationship between QTL for comb mass, reproductive, and skeletal investment

All QTLs associated with variation in comb mass were clustered in two chromosomal regions, one on chromosome 1 and one on chromosome 3. These regions were also associated with female reproductive and skeletal investment QTL (Table 3;Figs. 3, 4; see also Rubin et al. 2006 for skeletal QTL). Both female comb mass QTL on chromosome 1 (87 and 193 cM) were roughly equidistant from the total egg mass and trabecular bone mineral density QTL. Like the female comb mass QTL, the total egg mass and trabecular bone mineral density also had increased effects in the red junglefowl. It is therefore possible that the association between female comb mass and fecundity is entirely phenotypically plastic, comb mass changing over time to match the oviposition status of a female. To test for this possibility we entered total egg mass as a covariate in the QTL analysis for female comb mass. If comb mass was entirely dependent on egg production, the QTL on chromosome 1 should be significantly reduced. However, with this additional covariate the significance of the female comb mass QTL actually increased, rather than decreased, exceeding the 1% threshold (Table 3), suggesting that variation in female comb mass and fecundity has independent genetic components.


Figure 4. QTL for investment in female and male comb mass, female reproductive and skeletal quality on chromosome 3. The y-axis covers the total length of the chromosome (335 cM), the x-axis represents percentage of the F2 population mean. The region containing a QTL cluster has been exploded for greater clarity. QTL effects are expressed as percentage of the population mean, with twice the additive variance of each QTL used. Effects greater in the red junglefowl direction (red) are negative, whereas those greater for the domestic line (blue) are positive.

Download figure to PowerPoint

The major QTL for bisexual expression of comb mass on chromosome 3 is in the vicinity (< 14 cM) of a QTL associated with mean egg mass. A QTL for mean egg mass on chromosome 1 was located in close proximity to the female comb mass QTL. Furthermore, several QTL, also identified in this cross, associated with skeletal investment were located on each of chromosomes 1 and 3 (Rubin et al. 2006). Most notably, a QTL for trabecular bone mineral density was located on chromosome 1 at 138 cM, in close proximity to the locus affecting total egg mass. Both the bisexual comb mass QTL and the mean egg mass QTL cause an increase, whereas the bone density QTL causes a decrease in the domestic line artificially selected for egg production (Fig. 4). The fact that clusters of QTL for female comb mass, reproductive, and skeletal investment show greater allelic effects in the same direction (i.e., greater effects of red junglefowl alleles on chromosome 1, greater effects of domestic line alleles on chromosome 3) may reflect the signal of differential selection (natural and artificial, respectively) on cosegregating genes associated with female reproductive investment in the parental populations. No two-way epistatic interactions were detected between any of the identified QTL.

We tested whether the two QTL clusters on chromosome 1 and 3 are pleiotropic in origin, with a single locus affecting several different traits, or rather represent multiple linked QTLs affecting different traits. In each of these two QTL clusters, pairs of traits were fitted into a multivariate model (see Table 4). If any QTL are indicated to be pleiotropic it is then possible to build further traits into the model and determine whether such pleiotropic effects extend over several traits. Despite the fact that in all pairwise cases the linkage model had a higher likelihood, only 11 of the 18 of these pairs were significantly different using a LR test (Table 4). Of these, the majority of significant linkage models were present in the chromosome 1 cluster. Therefore, multivariate analysis does indicate a greater probability of linkage for some of the loci involved, but is not conclusive for all traits, with the cluster on chromosome 1 most likely to represent a collection of linked QTL, whereas the cluster on chromosome 3 is largely indeterminate as to the potential effects of pleiotropy or linkage. Due to the relatively small sample size of individuals used to measure female comb mass however, such results must be interpreted with caution: pleiotropic effects may be present, but larger sample sizes are required to provide more recombination events that allow us to better distinguish possible pleiotropy from linkage.

Table 4.  Multivariate analysis of the QTLs on chromosomes 1 and 3. In each region, traits were taken in a pairwise fashion to check for possible pleiotropy. Likelihood ratio and significance value are given, as well as the model (linkage or pleiotropy) with the highest likelihood.
Pairwise traitsPleiotropy versus linkageLRP
chromosome 1
Female comb mass versus egg mean massLinkage 5.70.02
Female comb mass versus endosteal circumferenceLinkage 3.90.05
Female comb mass versus trabecular bmdLinkage 8.20.00
Female comb mass versus total egg massLinkage 9.40.00
Trabecular bmd versus total egg massLinkage 1.20.28
Trabecular bmd versus endosteal circumferenceLinkage16.70.00
Trabecular bmd versus egg mean massLinkage17.60.00
Endosteal circumference versus mean egg massLinkage 0.70.41
chromosome 3
Comb mass versus egg mean massLinkage 0.90.34
DXA bmd versus comb massLinkage11.70.00
DXA bmd versus egg mean massLinkage 4.20.04

Given that the multivariate tests performed did not detect any significant effects of pleiotropy, and indeed indicated a significantly greater probability of linkage on chromosome 1, a valid question remains as to what is the probability of such QTL all occurring in a broadly similar region by chance. To test the likelihood of such clustering, a Poisson distribution based on the total number of marker intervals and the size of the two intervals on chromosome 1 and 3 was used to calculate the probability of this clustering occurring by chance, given the total number of QTLs detected in the cross. All skeletal, fecundity, and comb mass QTLs were included in the analysis (not just those detected on chromosomes 1 and 3). These intervals (covering eight marker intervals on chromosome 1 and five marker intervals on chromosome 3) were chosen because they covered all the QTL in each given interval, as well as included the minimum and maximum confidence interval of the highest and lowest QTL (thereby increasing the size of the region that contained the QTL and leading to a more conservative estimate). Marker intervals rather than chromosomal length were used to account for varying marker density between chromosomes. In the case of both the QTL cluster on chromosome 1 and that on chromosome 3, the probability of chance clustering was 0.003 (Po 8, 2.45 chromosome 1, Po 5, 0.82, chromosome 3).


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion

Studies of QTL for sexual ornaments are scarce and almost exclusive to male ornaments, for example, courtship songs (Gleason et al. 2002; Moehring and Mackay 2004; Huttunen et al. 2004), abdominal pigmentation (Foley et al. 2007), and cuticular hydrocarbons (Kopp et al. 2003) in Drosophila. Information on number and location of loci controlling ornament variation is crucial to feed back to quantitative genetic models of sexual selection. Although the sample sizes of the present study may make these results vulnerable to overestimations of the effect size of detected QTL (the “Beavis” effect, Beavis 1998), the present study strongly suggests that at least one major effect locus and several other loci of lesser effect are involved in the variation in comb mass between two populations highly divergent in this trait.

The present study also reveals that a sexual ornament that is expressed in both males and females is partly controlled by sex-specific physiological and genetic mechanisms. The female expression of bisexual ornaments, has often been dismissed as nonadaptive byproduct of male ornaments (Amundsen 2000), whereas a small—but increasing—number of studies have considered the possibility that the female expression of bisexual ornaments may have a function similar to the male expression, creating potential for mutual mate choice (e.g., Jones and Hunter 1993; Jones et al. 2001). The present results would indicate that variation in female comb mass is controlled by a region of chromosome 3 that also controls male comb mass, and by regions on chromosome 1 that were not associated with male comb mass, suggesting that the genetic architecture of the comb is at least partially discrete between the sexes. These results are not consistent with the idea that the female expression of bisexual ornaments is necessarily exclusively a genetic byproduct of the male expression. Importantly, we show that female comb mass was not functionally equivalent to male comb mass but reflected female-specific measures of reproductive investment at the expense of skeletal investment.

Nonrandom clustering of comb QTL with the QTL associated with both female reproductive and skeletal investment suggests a genetic basis for the relationship between female investment in comb and reproduction, mediated by linkage disequilibria (or possibly pleiotropy in the case of the region on chromosome 3) between QTL in the F2 hybrids. It is difficult to extrapolate intrapopulation patterns from interpopulation crosses, and whether the same genetic relationship exists within parental (and other fowl) populations remains to be established. However, these results indicate the potential for a female ornament to be genetically linked to female ability to invest in reproduction, and would suggest that the preference of male fowl for females with relatively large combs (Pizzari et al. 2003; Cornwallis and Birkhead 2007) may be selected directly (more and better eggs), and indirectly through two mechanisms: (1) more successful offspring (due to higher maternal investment), and (2) daughters that inherit superior reproductive efficiency. In addition, the comb of a male will convey information on the genetic reproductive quality of his female relatives. Consistent with this idea, male comb mass was genetically positively correlated with female comb mass and with both the number and the mass of eggs produced by females in a line of domestic fowl artificially selected for male comb mass (von Schantz et al. 1995), and thus female preference for large male combs (Parker and Ligon 2003) may be partly explained by these female effects.

Most models of sexual selection propose that ornament expression is a reliable indicator of an individual's breeding value for attractiveness and/or viability (Andersson 1994; Cotton et al. 2004; Mead and Arnold 2004; Andersson and Simmons 2006). Despite mounting evidence for male partner choice, both overt through selection of copulation partners (Jones and Hunter 1993; Amundsen and Forsgren 2001; Bonduriansky 2001; Jones et al. 2001; Griggio et al. 2003) and cryptic, through differential sperm allocation (Wedell et al. 2002; Pizzari et al. 2003; Cornwallis and Birkhead 2007), the functional significance of female sexual ornaments has received relatively little attention. Recent population genetic models demonstrate that male preference for a female ornament can only be maintained if female ornament reflects higher fecundity or viability (Chenoweth et al. 2006; Servedio and Lande 2006).

Importantly, the results of the present study strongly suggest that selection for female reproductive investment appears to favor high calcium mobilization and low relative skeletal investment. Calcium mobilization and medullary bone formation are induced by both estrogen and androgen. Comb expression is also steroid dependent (Parker et al. 1942; Bullock and Hall 1968), suggesting that positive selection for female fecundity may result in correlated selection for female comb mass and calcium mobilization.

The large effect of the bisexual comb mass QTL on chromosome 3 will facilitate the identification of the underlying nucleotide variation in this region, giving molecular insights not only into the mechanism of development, but also in the relative effects of pleiotropy and linkage. The confidence interval of this QTL is around 19 cM using a standard one-lod drop, decreasing to 12 cM (54–66 cM) with bootstrap resampling. This interval contains five known genes. Of these, one of particular note is a gene encoding a bone morphogenetic protein (BMP2). BMP2 is part of the transforming growth factor (TGF) superfamily and induces bone formation when transgenically inserted into rats (Wang et al. 1990), as well as being able to cause osteogenic transformation in a mouse pluripotent cell line (Cheng et al. 2003).

Although QTL emerging from an intercross of divergent populations reflect the genetic differences between the parental populations, three factors suggest that the results of the present study may identify general mechanisms. First, the use of a wild-type (red junglefowl) and domestic line artificially selected for female reproductive investment is likely to reveal basic genetic differences associated with domestication and with changes in selection on female reproductive investment. Second, consistent with the present study, a previous study (Schreiweis et al. 2005) based on a different fowl intercross, also detected a QTL for trabecular bone density in the same region on chromosome 3 in which we detected a QTL for DXA bone mineral density, indicating that the QTL identified by our study may not be specific to genetic differences of this intercross, but of general biological relevance. Third, considering that in several vertebrates sexual ornaments are steroid dependent and female reproductive investment is associated with estrogen and calcium mobilization, these results are likely to be of broad relevance, and open up novel approaches to study the evolution of both male and female ornaments and reproductive investment. A striking example of the generality of calcium mobilization in female ornamentation revealed by the present study is provided by female barn owls, Tyto alba, in which eumelanic plumage ornaments reflect female offspring viability (Niecke et al. 2003). This eumelanin pigmentation requires high calcium concentration and more ornamented females have higher calcium concentration in their humerus bone (Roulin et al. 2006). It is worth noting that the domestic line used in our intercross is homozygous for an allele at the Dominant white locus (PMEL17) that knocks out the synthesis of eumelanin (Keeling et al. 2004; Kerje et al. 2004). It is tempting to speculate that this may have evolved as a response to higher mobilization of calcium demanded by artificial selection for super-optimal egg production.

In the future, it will be important to identify the genes involved in steroid-mediated calcium mobilization associated with female reproductive investment and analyze their molecular evolution across avian species characterized by different patterns of female reproductive investment and ornamentation.

Associate Editor: J. Wolf


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion

We are grateful to A. Watchman and A. Blomqvist for great help with animal husbandry and data collection, to Ö. Carlborg for advice and help with the QTL analysis, to R. Bustling, T. Parker, A. Roulin, J. Wolf, C. Rubin, and two anonymous referees for useful comments on the manuscript. This work was financed by a scholarship from the Swedish University of Agricultural Sciences and a FORMAS grant to TP, and FORMAS grants to PJ. TP conceived the study, collected the comb mass data, wrote the first draft of the manuscript and conducted the first phenotypic comb and egg analyses. DW conducted all the final phenotypic and QTL analyses, and wrote the final draft of the manuscript with TP, HB and AK conducted all the bone measurements, SK constructed the QTL linkage map and genotyped the F2 birds, PJ and KS collected phenotypic measures of reproductive investment and body mass on the F2 birds, PJ and LA designed the red junglefowl–White Leghorn intercross.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  • Amundsen, T. 2000. Why are female birds ornamented? Trends Ecol. Evol. 15:149155.
  • Amundsen, T., and E. Forsgren. 2001. Male mate choice selects for female coloration in a fish. Proc. Natl. Acad. Sci. USA 98:1315513160.
  • Andersson, M. 1994. Sexual Selection. Princeton Univ. Press, Princeton .
  • Andersson, L. 2001. Genetic dissection of phenotypic plasticity in domestic animals. Nature Rev. Genet. 2:1308.
  • Andersson, M., and L. W. Simmons. 2006. Sexual selection and mate choice. Trends Ecol. Evol. 21:296302.
  • Beavis, W. D. 1998. QTL analyses: power, precision and accuracy. 49th Annual Corn and Sorghum Industry Research Conference. ASTA, Washington DC . 250266.
  • Blas, J., L. Perez-Rodriguez, G. R. Bartlett, J. Vineland, and T. A. Merchant. 2006. Testosterone increases bioavailability of carotenoids: insights into the honesty of sexual signalling. Proc. Natl. Acad. Sci. USA. 49:1863318637.
  • Bolton, M. 1991. Determinants of chick survival in the lesser black-backed gull: relative contributions of egg size and parental quality. J. Anim. Ecol. 60:949960.
  • Bonduriansky, R. 2001. The evolution of male mate choice in insects: a synthesis of ideas and evidence. Biol. Rev. 76:305339.
  • Bullock, J., and J. C. Hall. 1968. Effect of steroid esters on comb growth. World Poult. Sci. J. 24:328.
  • Cameron, E., T. Day, and L. Rowe. 2003. Sexual conflict and indirect benefits. J. Evol. Biol. 16:10551060.
  • Carlborg, Ö., L. Andersson, and B. Kinghorn. 2000. The use of a genetic algorithm for simultaneous mapping of multiple interacting quantitative trait loci. Genetics 155:20032010.
  • Cheng, H., W. Jiang, F. M. Phillips, R. C. Haydon, Y. Peng, L. Zhou, H. H. Luu, N. L. An, B. Breyer, P. Vanichakarn, et al. 2003. Osteogenic activity of the fourteen types of human bone morphogenetic proteins (BMPs). J. Bone Joint. Surg. 85:15441552.
  • Chenoweth, S. F., P. Doughty, and H. Kokko. 2006. Can non-directional selection male mating preferences facilitate honest female ornamentation? Ecol. Lett. 9:179184.
  • Cotton, S., K. Fowler, and A. Pomiankowski. 2004. Do sexual ornaments demonstrate heightened condition-dependent expression as predicted by the handicap hypothesis? Proc. R. Soc. Lond. B 271:771783.
  • Cloutier, S., and R. C. Newberry. 2000. Recent social experience, body weight and initial patterns of attack predict the social status attained by unfamiliar hens in a new group. Behaviour 137:705726.
  • Cordero, C., and W. G. Eberhard. 2003. Female choice of sexually antagonistic male adaptations: a critical review of some current research. J. Evol. Biol. 16:16.
  • Cornwallis, C. K., and Birkhead, T. R. 2007. Experimental evidence that female ornamentation increases the acquisition of sperm and signals fecundity. Proc. R. Soc. Lond. B 274:583590.
  • Cransberg, P. H., G. B. Parkinson, S. Wilson, and B. H. Thorp. 2001. Sequential studies of skeletal calcium reserves and structural bone volumes in a commercial layer flock. Brit. Poult. Sci. 42:260265.
  • Dacke, C. G., S. Arkle, D. J. Cook, I. M. Wormstone, S. Jones, M. Zaidi, and Z. A. Bascal. 1993. Medullary bone and avian calcium regulation. J. Exp. Biol. 184:6388.
  • Etches, R. J. 1996. Reproduction in poultry. CAB International, Wallingford , UK .
  • Foley, B., S. F. Chenoweth, S. V. Nuzhdin, and M. W. Blows. 2007. Natural genetic variation in cuticular hydrocarbon expression in male and female Drosophila melanogaster. Genetics. 175:14651477.
  • Fumihito A, T. Miyake, S. I. Sumi, S. Ohno, and N. Kondo. 1994. One species of the red jungelfowl (Gallus gallus gallus) suffices as the matriarchic ancestor of all domestic breeds. Proc. Natl. Acad. Sci. USA 91:1250512509.
  • Gleason, J. M., S. V. Nuzhdin, and M. G. Ritchie. 2002. Quantitative trait loci affecting a courtship signal in Drosophila melanogaster. Heredity 89:16.
  • Griggio, M., G. Matessi, and A. Pilastro. 2003. Male rock sparrow (Petronia petronia) nest defence correlates with female ornament size. Ethology 109:659669.
  • Hughes, K. A., and J. Leips. 2006. Quantitative trait locus analysis of male mating success and sperm competition in Drosophila melanogaster. Evolution 60:14271434.
  • Huttunen, S., J. Aspi, A. Hoikkala, and C. Schloterrer. 2004. QTL analysis of variation in male courtship song in Drosophila virilis. Heredity 92:263269.
  • Johns, P. M., and G. S. Wilkinson. 2007. X chromosome influences sperm length in the stalk-eyed fly Cyrtodiopsis dalmanni. Heredity 99:5661.
  • Jones, I. L., and F. M. Hunter. 1993. Mutual sexual selection in a monogamous seabird. Nature 362:238239.
  • Jones, K. M., P. Monaghan, and R. G. Nager. 2001. Male mate choice and female fecundity in zebra finches. Anim. Behav. 62:10211026.
  • Keeling, L., L. Anderson, K. E. Schütz, S. Kerje, R. Fredriksson, Ö. Carlborg, C. K. Cornwallis, T. Pizzari, and P. Jensen. 2004. Chicken genomics: feather-pecking and victim pigmentation. Nature 431:645646.
  • Kerje, S., Ö. Carlborg, L. Jacobsson, K. Schütz, C. Hartmann, P. Jensen, and L. Andersson. 2003. The twofold difference in adult size between the red junglefowl and White Leghorn chickens is largely explained by a limited number of QTLs. Anim. Genet. 34:264274.
  • Kerje, S., P. Sharma, U. Gunnarsson, H. Kim, S. Bagchi, R. Fredriksson, K. Schütz, P. Jensen, G. Von Heijne, R. Okimoto, et al. 2004. Dominant white, Dun and Smoky color variants in chicken are associated with insertion/deletion polymorphisms in the PMEL17 gene. Genetics 168:15071518.
  • Kokko, H., R. Brooks, M. D. Jennions, and J. Morley. 2003. The evolution of mate choice and mating biases. Proc. R. Soc. Lond. B 270:653664.
  • Kopp, A., R. M. Graze, S. Z. Xu, S. B. Carroll, and S. V. Nuzhdin. 2003. Quantitative trait loci responsible for variation in sexually dimorphic traits in Drosophila melanogaster. Genetics 163:771787.
  • Kusuhara, S., and H. Schraer. 1982. Cytology and autoradiography of estrogen-induced differentiation of avian endosteal cells. Calcified Tissue Int. 34:352358.
  • Mead, L. S., and S. J. Arnold. 2004. Quantitative genetic models of sexual selection. Trends Ecol. Evol. 19:264271.
  • Moehring, A. J., and T. F. C. Mackay. 2004. The quantitative genetic basis of male mating behavior in Drosophila melanogaster. Genetics 167:12491263.
  • Mundy, N. I. 2005. A window on the genetics of evolution: MC1R and plumage colouration in birds. Proc. R. Soc. Lond. B 272:16331640.
  • Niecke, M., S. Rothlaender, and A. Roulin. 2003. Why do melanin ornaments signal individual quality? Insights from metal element analysis of barn owl feathers. Oecologia 137:153158.
  • Nishibori, M., T. Shimogiri, T. Hayashi, and H. Yasue. 2005. Molecular evidence for hybridization of species in the genus Gallus except for Gallus varius. Anim. Genet. 36:367375.
  • Ohashi, T., S. Kusuhara, and K. Ishida. 1987. Effects of oestrogen and anti-oestrogen on the cells of the endosteal surface of male Japanese quail. Brit. Poult. Sci. 28:727732.
  • Parker, J. E., F. F. McKenzie, and H. L. Kempster. 1942. Development of the testis and combs of White Leghorn and New Hampshire cockerels. Poult. Sci. 21:3544.
  • Parker, T. H., and J. D. Ligon. 2003. Female mating preferences in red junglefowl: a meta-analysis. Ecol. Ethol. Evol. 15:6372.
  • Pérez-Enciso, M., and I. Misztal. 2004. Qxpak: a versatile mixed model application for genetical genomics and QTL analyses. Bioinformatics 20:27922798.
  • Pizzari, T., and R. R. Snook. 2003. Sexual conflict: chasing away paradigm shifts. Evolution 57:12231236.
  • Pizzari, T., D. P. Froman, and T. R. Birkhead. 2002. Pre- and post-insemination episodes of sexual selection in the fowl. Heredity 88:112116.
  • Pizzari, T., C. K. Cornwallis, H. Løvlie, S. Jakobsson, and T. R. Birkhead. 2003. Sophisticated sperm allocation in male fowl. Nature 426:7074.
  • Reynolds, S. J., R. Mänd, and V. Tilgar. 2004. Calcium supplementation of breeding birds: directions for future research. Ibis 146:601614.
  • Rice, W. R. 1984. Sex chromosomes and the evolution of sexual dimorphism. Evolution 38:735742.
  • Rosenthal, G. G., and F. J. García De León. 2006. Sexual behavior, genes, and evolution in Xiphophorus. Zebrafish 3:8590.
  • Roulin, A. 2004. Proximate basis of the covariation between a melanin-based female ornament and offspring quality. Oecologia 140:668675.
  • Roulin, A., T. Dauwe, R. Blust, M. Eens, and M. Beaud. 2006. A link between eumelanism and calcium physiology in the barn owl. Naturwissenschaften 93:426430.
  • Rubin, C.-J., H. Brändström, D. Wright, S. Kerje, U. Gunnarsson, K. Schütz, R. Fredriksson, P. Jensen, L. Andersson, C. Ohlsson, et al. 2006. Quantitative Trait Loci for trabecular bone mineral density and bone strength in an intercross between domestic and wild-type chicken. J. Bone Min. Res. 22:375384.
  • Schreiweis, M. A., P. Y. Hester, P. Settarand, and D. E. Moody. 2005. Identification of quantitative trait loci associated with egg quality, egg production, and body weight in an F2 resource population of chickens. Anim. Genet. 37:106112.
  • Schütz, K., S. Kerje, Ö. Carlborg, L. Jacobsson, L. Andersson, and P. Jensen. 2002. QTL analysis of a red junglefowl × White Leghorn intercross reveals trade-off in resource allocation between behavior and production traits. Behav. Genet. 32:423433.
  • Servedio, M. R., and R. Lande. 2006. Population genetic models of male and mutual mate choice. Evolution 60:674685.
  • Sugiyama, T. and S. Kusuhara. 2001. Avian calcium metabolism and bone function. Asian-Aust. J. Anim. Sci. 14:8290.
  • Taylor, T. G. 1970. How an eggshell is made. Sci. Am. 222:8994.
  • Tomkins, J. L., J. Radwan, J. S. Kotiaho, and T. Tregenza. 2004. Genetic capture and resolving the lek paradox. Trends Ecol. Evol. 19:323328.
  • Tufvesson, M., B. Tufvesson, T. Von Schantz, K. Johansson, and M. Wilhelmson. 1999. Selection for sexual male characters and their effects on other fitness related traits in White Leghorn chickens. J. Anim. Breed. Genet. 116:127138.
  • Von Schantz, T., M. Tufvesson, G. Göransson, M. Grahn, M. Wilehelmson, and H. Wittzell. 1995. Artificial selection for increased comb size and its effects on other sexual characters and viability in Gallus domesticus (the domestic chicken). Heredity 75:518529.
  • Wagner, E. C., and T. D. Williams. 2007. Experimental (antiestrogen-mediated) reduction in egg size negatively affects offspring growth and survival. Physiol. Biochem. Zool. 80:293305.
  • Wang, E. A., V. Rosen, J. S. D'Alessandro, M. Bauduy, P. Cordes, T. Harada, D. I. Israel, R. M. Hewick, K. M. Kerns, P. LaPan, et al. 1990. Recombinant human bone morphogenetic protein induces bone formation. Proc. Natl. Acad. Sci. 87:22202224.
  • Wedell, N., M. J. G. Gage, and G. A. Parker. 2002. Sperm competition, male prudence and sperm-limited females. Trends Ecol. Evol. 17:313320.
  • Williams, T. D. 1994. Intraspecific variation in egg size and egg composition in birds: effects on offspring fitness. Biol. Rev. 68:35– 59.
  • Wilson, H. R. 1991. Inter-relationships of egg size, chick size, posthatching growth and hatchability. Worlds Poult. Sci. J. 47:520.
  • Wilson, S., and S. R. I. Duff. 1990. Morphology of medullary bone during the egg formation cycle. Res. Vet. Sci. 48:216220.
  • Wright, D., S. Kerje, K. Lundström, J. Babol, K. Schütz, P. Jensen, and L. Andersson. 2006. Quantitative trait loci analysis of egg and meat production traits in a red junglefowl × White Leghorn cross. Anim. Genet. 37:529534.
  • Yeh, S. D., S. R. Liou, and J. R. True. 2006. Genetics of divergence in male wing pigmentation and courtship behavior between Drosophila elegans and D. gunungcola. Heredity 96:383395.
  • Zuk, M., T. Kim, S. I. Robinson, and T. S. Johnsen. 1998. Parasites influence social rank and morphology, but not mate choice, in female red junglefowl, Gallus gallus. Anim. Behav. 56:493499.