Research note: Comparative estimation of genetic parameters for eggshell pigmentation traits

Abstract

Eggshell color exhibits considerable variation, with the diversity particularly rich in shells where protoporphyrin serves as the primary pigment. The measurement of eggshell color typically employs the Commission International del’Eclairage (CIE) L*a*b* color space system, which precisely describes color based on three dimensions: lightness (L*), red-green (a*), and yellow-blue (b*). However, in comparative studies of different egg colors, the lightness value (L*) is most frequently utilized, while the use of a*and b*values remains relatively insufficient. To comprehensively evaluate eggshell color, several comprehensive color indices, including quantities of true protoporphyrin IX (Q_p), Shell Color Index (SCI), Chroma (C*), and Hue Angle (h°), have been proposed. Nevertheless, it remains uncertain which index more suitable evaluates the eggshell color. This study involved 1,481 Rhode Island Red chickens and 1,938 Dwarf chicken. Eggshell color data were collected at 43, 60, and 80 weeks of age, and genetic parameter estimates were made for each comprehensive egg color index. The results indicated that, in terms of phenotypic correlation with L*, Q_p was superior to SCI, followed by h°, and finally C*. The Q_p value decreased with increasing age, validating the observed trend of eggshell color becoming lighter as hens age. Estimates of genetic parameters revealed that the heritability of Q_p ranged between 0.37 and 0.42, indicating a moderate to high level of heritability. The heritability of SCI ranged from 0.32 to 0.44, that of C* ranged from 0.23 to 0.41, and that of h° was 0.46. All comprehensive color indices showed medium to high heritability. In conclusion, Qp and SCI more accurately reflects variations in eggshell color. Therefore, it is recommended that Qp and SCI be adopted as the phenotypic parameter for eggshell color in breeding programs for brown- and pink-shelled laying hens to facilitate accelerated genetic progress.

Introduction

Eggshell color, as an externally visible morphological trait of avian eggs, exhibits distinct species- and breed-specific characteristics. The coloration is primarily determined by the types and total content of pigments deposited in the eggshell. Combinations of protoporphyrin IX and biliverdin at varying concentrations produce a broad spectrum of color patterns. Two primary hypotheses explain the evolutionary mechanisms underlying eggshell coloration. The first proposes that eggshell color provides camouflage for embryos, aiding in concealment or recognition. The second suggests that these pigments offer antioxidant protection to developing embryos, mitigating oxidative stress (Martínez et al., 2023).

Although eggshell color is not directly linked to internal egg quality, it serves as the most conspicuous external characteristic. Consumer preferences for specific shell colors vary across regions: brown-shell eggs dominate markets in China, the UK, France, Italy, and Portugal. Furthermore, uniformity in shell color within a batch is often demanded; for instance, some Japanese markets prefer standardized dark brown shells, whereas other regions favor light brown eggs (Zhou et al., 2025). In China, consumers have a wide range of preferences, and the color levels they like are usually between 3 and 8 on the standard color fans (Zhou et al., 2025).

Traditional methods for assessing eggshell color are diverse but lack standardization due to inherent limitations. Widely used techniques include eggshell color fans and the CIE L*a*b* color system. Color fans divide shell colors into 17 graduated segments for visual matching, but this method relies heavily on subjective judgment, leading to significant operational variability. The CIE L*a*b* system employs a handheld spectrophotometer to measure three points on the eggshell surface, with the average value representing the color. This objective approach reduces human error and improves efficiency compared to color fans. However, in practice, the lightness value (L*) is often used in isolation as a rough indicator of shell color, as it is easier to interpret within the same color lineage. The underutilization of the a*(red-green) and b* (yellow-blue) dimensions limit comparisons across different shell colors. Although L* does not fully represent eggshell color, its widespread use in practical production led us to adopt it as one of the reference benchmarks in this study. This allowed us to evaluate whether the new indices can provide a more comprehensive description of color while maintaining practical applicability.

To better utilize data from the L*a*b* system, several formulas have been developed to derive composite color indices. These include the Shell Color Index (SCI), Chroma (C*), and Hue Angle (h°). SCI is calculated through simple arithmetic operations on L, a, and b values, emphasizing the dominant role of L* in the color space. This index implies that lighter shells correlate with higher L* and lower a* and b* values, resulting in a higher SCI (Petracci and Baéza, 2007) (see Formula section).

Chroma and Hue Angle, in contrast, are derived solely from a* and b* values (see Formulas section). Chroma measures color saturation, where higher values indicate more vivid colors and may reflect pigment deposition levels (Dallaire-Lamontagne et al., 2024). Initially applied to assess yolk color, Hue Angle expresses color orientation based on the arctangent of the b/a ratio.

The Shell Color Index (SCI) involves only simple arithmetic operations, which can lead to substantial errors in practical applications. Chroma (C*) is primarily useful for comparisons within the same hue domain, assessing saturation levels but lacking the capacity to differentiate between distinct colors; consequently, it is unsuitable for evaluating eggshell colors with significant hue variations, such as brown eggs (Kojima et al., 2022). The Hue Angle (h°), initially developed for assessing yolk color, indicates color bias through angular values. However, its sensitivity is limited within similar hue ranges, making it inadequate for detecting subtle variations in saturation within the same hue (Kojima et al., 2022). In current production settings, the lightness value (L*) is often used as a rough proxy for overall eggshell color due to its simplicity.

The previous research developed the Q_p evaluation system by correlating protoporphyrin and biliverdin content in the eggshell with L*a*b* measurements, establishing a direct link between color coordinates and pigment concentration (Zeng et al., 2022).

Given the multitude of existing eggshell color evaluation methods, each with distinct emphases, there is currently a lack of comparison between different eggshell color methods. This study aims to identify more comprehensive eggshell color indicators, distinguish eggshell colors, compare various eggshell color indicators from the perspective of genetic parameter estimation, and provide support for eggshell color selection in breeding.

Materials and methods

Experimental animals and housing conditions

All hens were from Beijing Zhongnong Bangyang Layer Breeding Co., Ltd. The birds were housed in a clean and well-ventilated poultry house under a single-cage rearing system. The housing environment utilized natural ventilation supplemented with an artificial lighting system. The photoperiod was maintained at 16 hours of light and 8 hours of darkness daily, with a light intensity ranging from 18 to 22 lux. The temperature within the house was controlled between 18°C and 21°C, with a relative humidity of 55% to 70%.

Experimental birds and Sample Collection

A total of 1,481 pure-line Rhode Island Red hens and 1,938 pure-line Dwarf chicken were selected as experimental birds, Eggs were collected at three ages: 43, 60 weeks. A total of 371 Rhode Island Red laying hens and 433 Dwarf laying hens were used for egg collection at 80 weeks of age. Each egg was clearly labeled with the individual hen's identification number and its family lineage number for subsequent traceability and data analysis.

Determination of eggshell color

Eggshell color was measured using a handheld spectrophotometer (CM-700d, Konica Minolta, Inc., Japan). Measurements were taken at three points on each egg: the blunt end, the equatorial region (middle), and the sharp end. The L* (lightness), a* (red-green component), and b* (yellow-blue component) values were recorded, and the average of the three measurement points per egg was calculated. For each hen, the color values from three eggs were averaged and used for subsequent analysis.

Formulas for comprehensive eggshell color indices

Four formulas are utilized for assessing Q_p, SCI, C*, and h°, respectively.

$$SCI=L*-a*-b*$$

$$C*=\surd \left[{\left(a*\right)}^2+{\left(b*\right)}^2\right]$$

$$h^{\circ }=arctan\left(b*/a*\right)$$

$$Q_p=1.042+0.0137L*a*+0.0763a*b*-0.0014L*a*b*$$

Statistical analysis

Data analysis was performed using R language (RStudio version 4.3.2). The eggshell color data were cleaned using the “tidyr” package, and results were visualized using the ggplot2 package. Differences in the data across different weeks of age were compared using one-way analysis of variance (ANOVA). A significance threshold of P < 0.05 was applied to determine statistical significance.

Estimation of genetic parameters

Genetic parameters for Q_p, SCI, C*, and h°were estimated using the data collected at 43 and 60 weeks of age. The analysis was performed with DMU version 6.0, applying the DMUAI module which implements the Residual Maximum Likelihood (REML) method for variance component estimation. A single-trait repeatability model was fitted for the analysis. A single-trait repeatability model was employed in this study, with week age and breed included as fixed effects, and individual additive genetic effects and permanent environmental effects fitted as random effects. This model is appropriate for repeated measures data and facilitates the comparison of genetic parameters across different color indices.

Results and discussion

The changing patterns of different egg color indicators during the laying period

Eggshell color was measured for eggs collected from Rhode Island Red and Dwarf chicken at 43, 60, and 80 weeks of age. The results are presented in Fig. 1. A general lightening of eggshell color was observed as the hens aged. This phenomenon is primarily attributed to an age-related decline in the capacity for pigment synthesis and deposition in the shell gland. The results indicated that the L* and SCI increased with advancing hen age. In contrast, the C*, h°and Q_p exhibited a decreasing trend over the same period (P < 0.01). Concurrently, the SCI increased significantly with week age (P < 0.01). Both C and h° decreased significantly with increasing week age (P < 0.01). The observed trends for all four indices (SCI, C*, h°, Q_p) are consistent with findings reported in previous studies (Li et al., 2019). The decline in Q_p values with advancing age reliably reflects the lightening of eggshell color.

Fig. 1.

Changes in eggshell color parameters across different weeks of age.​​ (A) Rhode Island Red hens; (B) Dwarf chicken. Data are presented as mean ± SEM. * indicates significant difference (P< 0.05), and ​**​ indicates highly significant difference (P < 0.01).

Estimation of genetic parameters for different eggshell color indices

Given the pronounced influence of age-related decline on eggshell color at 80 weeks and the common practice of early selection in practical breeding programs, data from this age were excluded from the genetic parameter estimation to minimize interference from environmental variation on heritability estimates. Genetic parameters were estimated using data collected from Rhode Island Red and Dwarf chicken at 43 and 60 weeks of age. The results are presented in Fig. 2.

Fig. 2.

Estimates of genetic parameters. (A) Data from Rhode Island Red hens at 43 weeks of age; (B) Data from Dwarf chicken at 43 weeks of age; (C) Data from Rhode Island Red hens at 60 weeks of age; (D) Data from Dwarf chicken at 60 weeks of age. The lower triangle of the matrix represents genetic correlations, the upper triangle represents phenotypic correlations, and the diagonal elements indicate heritability estimates.

Phenotypic correlation analyses revealed specific relationships. The upper triangle of the correlation matrices indicated that in Rhode Island Red hens, the phenotypic correlations between L* and SCI, C*, h°, and Q_p were 0.89, −0.49, −0.75, and −0.96, respectively. In Dwarf chicken, the corresponding correlations were 0.82, −0.25, −0.66, and −0.93. This pattern demonstrates that, concerning the phenotypic correlation with L*, the Q_p index was superior to SCI, followed by h°, and finally C*. This suggests that Q_p more effectively captures the variation in eggshell brightness and aligns well with practical observations in production. In both the Q_p and SCI formulas, all three parameters (L*, a*, and b*) from the CIELAB color space are utilized, which contributes to their stronger correlation with the L value. In contrast, the correlations of C* and h° with Lare indirectly manifested through the correlations of a* and b* with L*, consequently resulting in lower correlation values (Guo et al., 2020).

Heritability estimates, located on the diagonal of the matrices, showed that in Rhode Island Red hens, all eggshell color indices exhibited low to moderate heritability (h²_L* = 0.26, h²_SCI = 0.32, h²_C* = 0.23, h²_h° = 0.46, h²_Qp = 0.37). In Dwarf chicken, the indices displayed moderate heritability (h²_L* = 0.27, h²_SCI = 0.44, h²_C* = 0.41, h²_h° = 0.46, h²_Qp = 0.42). Our results are consistent with those of Guo et al., who found the additive heritability of h²_L* = 0.23 (Guo et al., 2020). The generally higher heritability of the composite indices compared to L* alone may be attributed to the fact that L* solely reflects shell brightness and may not fully represent the pigment-based coloration. The refined formulas, potentially integrating more color information, consequently showed higher heritability (Brahimi et al., 2022). On the other hand, from a biological perspective and based on the definitions of the formulas, Qp reflects the pigment content in the eggshell. An increase in pigment content generally leads to a decrease in lightness (L*), thereby explaining the observed correlation between Qp and L*. In the case of the SCI formula, higher pigment content is typically associated with a reduction in L* and an increase in a* and b* values. These underlying relationships may also contribute to the higher heritability estimates observed for these indices. This study did not include direct biochemical assays for protoporphyrin or biliverdin in the eggshell. The Qp values were derived from a previously established regression model that predicts pigment content based on L*a*b* color coordinates. Consequently, any discussion of pigment deposition mechanisms should be regarded as indirect inference. The results at 60 weeks of age (Fig. 2C, D) showed a consistent trend with those at 43 weeks. However, as hens age, the influence of environmental factors on these traits typically increases. Therefore, 43 weeks of age appears to be a more suitable time point for selection.

Conclusion

This study assessed and, for the first time, estimated the heritability of four comprehensive eggshell color indices. It was found that the Q_p and SCI, which incorporate all three parameters (L*, a*, and b*) of the CIELAB color space, exhibited superior comprehensive performance compared to the C* and h° indices. Meanwhile, all indices showed moderate to high heritability. It should be noted that the conclusions of this study are based on brown-shelled and pink-shelled laying hen breeds. Further validation is recommended before extending these findings to other eggshell colors. In conclusion, it is recommended that Q_p and SCI be used as phenotypic parameter for eggshell color in breeding programs for brown and pink shelled laying hens to accelerate genetic improvement.

CRediT authorship contribution statement

Runzhe Wang: Writing – original draft. Jiahui Lai: Investigation. Honglei Jin: Investigation. Bingxin Luo: Investigation. Chuanwei Zheng: Investigation. Zhiqiong Mao: Investigation. Guiyun Xu: Investigation. Jiangxia Zheng: Writing – review & editing.

Disclosures

The authors declare that they have no known competing financial interests or personal relationships that could appear to influence the work reported in this paper.