PHYSIOLOGY, ENDOCRINOLOGY, AND REPRODUCTION. Changes in Brown Eggshell Color As the Hen Ages

Transcription

1 PHYSIOLOGY, ENDOCRINOLOGY, AND REPRODUCTION Changes in Brown Eggshell Color As the Hen Ages A. Z. Odabaşi,* R. D. Miles, 1 M. O. Balaban,* 2 and K. M. Portier 3 *Food Science and Human Nutrition Department, University of Florida, Gainesville ; Department of Animal Sciences, University of Florida, Gainesville ; and Department of Statistics, University of Florida, Gainesville ABSTRACT The color of eggshells from eggs laid by creased, indicating that size of the egg was a major factor commercial-type Hy-Line brown hens 25 wk of age was studied over a period of 10 mo. Color measurements were made by a color machine vision system and were analyzed using a mixed model to calculate between and within hen variances and to investigate the effect of time on shell color. Hens laid eggs with lighter colored shells as the flock aged, as evidenced by the lightness (L*) values increasing in time. A decrease in pigmentation was associated with a decrease in the amount of redness (a*) in the eggshell. When L* and a* values were corrected for egg weight, the rate of change in the L* and a* values deaffecting the color of the eggshell. These findings quantified the observations that older hens lay lighter colored eggs due to an increase in egg size associated with no proportionate change in the quantity of pigment deposited over the shell surface. Using a 2-stage sampling analysis and the variances between and within hens, sample sizes required to estimate the color of eggshells within 5% of the true mean were calculated. Accordingly, 11 eggs would need to be collected from each of the 51 hens housed for a study of brown eggshell color using the L*, a*, and b* (yellowness) coordinates. Key words: hen age, eggshell color, brown egg layer, sample size 2007 Poultry Science 86: INTRODUCTION Although shell color is not an indication of internal quality of eggs, consumers in some markets throughout the world prefer brown eggs over white eggs (United Kingdom, Italy, Portugal, Ireland, Southeast Asia, Australia, and New Zealand). Shell color intensity within each country is dictated by consumer preference. For example, in the Japanese egg market there are standards for a uniform dark shell color, whereas other markets prize a uniform light brown tint to the shell that requires crossing of white and brown egg laying lines (Arthur and O Sullivan, 2005). When selecting egg-laying lines with specific dark and light brown shell tints for such markets, capability of objective color measurement becomes critical, especially as the flock ages and eggshell pigmentation declines. For this purpose, companies that produce these genetic lines of birds commonly use a colorimeter that measures the eggs lightness (L*) and hue [as a function of a red-green (a*) and a yellow-blue 2007 Poultry Science Association Inc. Received July 17, Accepted September 28, Corresponding author: rdmiles@ufl.edu 2 The Color Expert software was developed by author M. O. Balaban, who can be contacted about the color machine vision system at mob@ufl.edu 3 Current address: Statistics & Evaluation Center, American Cancer Society, Atlanta, GA (b*) scale; Arthur and O Sullivan, 2005]. We recommend the use of a Color Machine Vision (CMV) system to measure the color units more accurately. A colorimeter measures the color of only a small portion of the shell s surface area, whereas with CMV the color of the entire surface facing the camera is determined. The advanced CMV systems in use today have numerous advantages and uses. For example, it has recently been reported that the CMV system can successfully be used to detect different dirt stains on brown eggs (Mertens et al., 2005). As the commercial-type white and brown egg layers age, the weight of the eggs laid by these hens increases (Roland et al., 1975; Fletcher et al., 1983; Sell et al., 1987). Commercial Hy-Line brown hens, for example, commence egg production with eggs weighing approximately 45 g. Thereafter, egg weight and size increase sharply within the first 8 wk after the onset of egg production. Although the rate of increase in egg weight/ size slows down, eggs keep getting heavier and larger for another 20 wk. From then on, the egg weight of the Hy-Line Brown layers levels off at an average of about 66 to 67 g (Hy-Line International, 2005). A major contribution to the area of research involving factors influencing eggshell quality was that of Roland et al. (1975). Prior to their report, the exact relationship between eggshell calcium content and decline in eggshell quality, as related to the age of the hen, was not understood. These researchers reported that as the hen aged, the increase in egg weight with no proportionate 356

2 CHANGES IN BROWN EGGSHELL COLOR 357 increase in shell deposition was the major factor that explained the age-related decline in shell quality of commercial egg-type laying hens. Later, Roland (1979) expanded the knowledge in this area when he reported that eggs that had the greater increase in size throughout the entire laying cycle also had the greater decline in shell quality. Roland (1979) also reported that shell quality at the end of the laying cycle was directly related to shell quality at the beginning of the cycle. Similarly, the pigments of the brown eggshell are deposited on a larger surface area as the hen ages. No evidence suggests a variation in the amount of pigment produced according to egg size (Solomon, 1997). As a result, the color of shells on eggs from a given flock turns paler with age of the flock (Lang and Wells, 1987; Solomon, 1997). In spite of the economic losses caused by variations in eggshell color, the changes in shell color as the hens age have not been studied quantitatively and reported (Lang and Wells, 1987). The objective of this study was to measure and quantify the changes in the shell color of eggs from commercial brown egg-type layers for 10 mo of production. MATERIALS AND METHODS Hens and Egg Collection This study was designed to monitor the shell color of eggs collected periodically from individual hens, during a 10-mo laying cycle. For this purpose, 240 Hy-Line brown layers were housed in individual cages. The hens were 25 wk old at the beginning of the study. A corn/ soybean meal based diet that included the nutrients and energy recommended by the National Research Council (1994) was fed during the experimental period. Because vanadium has been shown to have a negative bleaching effect on brown eggshells (Sutly et al., 2001; Odabasi et al., 2006), the diet was analyzed for vanadium and found to have a concentration of 3.1 ppm (DM basis). Vanadium was determined at a commercial laboratory by inductively coupled plasma spectroscopy following digestion with nitric and sulfuric acids. The dietary concentration of vanadium was in the normal range (<5 ppm) for a diet based on corn and soybean meal and a high quality inorganic phosphate source (Sell et al., 1982; Eyal and Moran, 1984; R. D. Miles, personal communication). The hens had ad libitum access to feed and water. The source of grain, soybean meal, and all other ingredients was the same throughout the entire experimental period. Hens were maintained on a lighting program of 15 h-continuous light each day. When hens were 25 wk of age, 3 eggs were collected from each hen on 3 consecutive days. The eggs from hens that did not lay consecutively on these 3 days were obtained on d4or5.attime of collection, each egg was marked according to hen, and then all eggs were taken to the laboratory and weighed. Eggs were refrigerated in the dark until color analysis. None of the eggs collected during this entire 10-mo study were stored for more than 1 wk. The 3-d egg collection from each hen was repeated during the first5dofeach month from November to September, for 10 mo. Data are not presented for the second month (December) because no eggs were collected due to a miscommunication among investigators. Similarly, mean shell weight data for the first month (November) are not presented. Once color analysis data were collected, eggs were broken and their shells cleaned of adhering albumen and allowed to air dry for 1 wk and shell weight (+membranes) determined. Color Analysis Measurement of Color. Color images of the set of 3 eggs obtained from each hen in a given month were captured using a color machine vision system, which consisted of a light box, an analog color video camera, a frame grabber, and a personal computer (Luzuriaga et al., 1997). The rectangular light box was aimed to provide uniform and constant lighting conditions for every sample and was 42.4 cm wide, 61 cm deep, and 68.6 cm high. The light box was made of 100% clear acrylic safety glazing sheets that had been painted white (flat white No. 1502, Krylon, Sherwin Williams Co., Solon, OH) to reflect light in all directions so that shadows were minimized. The linear light sources consisted of two 45.7 cm long, 15-W Chroma 50 fluorescent lights (F15T8-C50, General Electric, Cleveland, OH) aligned parallel to each other. The bulbs were located at the top of the light box, and they were separated from the chamber by a mm-thick white translucent Polycast acrylic sheet (No. 2247, Polycast Technology Corp., Stanford, CT). The 51% light transmission property of this sheet provided diffused light inside the chamber. For more details on the design of the light box, the readers are referred to Luzuriaga (1995). A charged couple device video camera (Sony SSC- S20, Sony Corp., Japan) positioned at an equal distance between the 2 light bulbs captured 24-bit color images of the eggshell surface. The camera settings (brightness, contrast, hue, and saturation) were adjusted to have a match as close as possible between the color of the image and the color of the eggshell. Through the S-video output of the camera, analog signals were sent to a color frame grabber (Matrox Meteor, Matrox, Canada) in the computer. With its accompanying software, the frame grabber digitized the analog signals into pixel images. The 3 eggs collected from a given hen were positioned on the bottom surface of the light box, with their blunt end facing the door of the light box. Each egg was placed on a circular pad with a hole in the middle and a sticky back so the eggs would not roll away as the image was being captured. The pads also positioned the location of each egg so that all eggs were placed in the exact location in all images. A color tile was positioned on an L-shaped wooden stand to minimize specular reflection off of the tile surface (Figure 1). The location of the tile

3 358 ODABAŞI ETAL. Figure 1. Color Machine Vision System with eggs and color tile. Camera and lens can be seen at top of the light box. was also kept constant for all images by a mark that identified the exact position of the wooden stand in the light box. The described procedure was used each month to evaluate the color of the shells from the 3 eggs collected from each of the 240 hens. Using Color Expert software, the color of each individual egg was calculated from the computer images and reported as the L*, a*, and b* values of the eggshell surface facing the camera. The L* value represents lightness and ranges from 0 to 100, with 0 corresponding to black and 100 to white. The chromatic attributes, redness-greenness and yellowness-blueness, are measured by a* and b*, respectively. Positive values of a* represent the amount of redness of the shell color, whereas negative values of a* indicate the amount of greenness in the shell color. Similarly, the yellow and blue components in any color are represented by positive and negative values of b*, respectively. Calibration of Color Data. A 1-point (hitching post) color calibration was used to correct for possible differences among the images due to changes in the light source or changes in the camera s sensitivity (Luzuriaga, 1999). This involved the comparison of the average L*, a*, and b* values of the color tile saved in each image file with known L*, a*, and b* values of the tile. The difference was applied to all of the objects in the image. The L*, a*, and b* values of the orange tile against which the color of all eggs was standardized were 61.94, 43.48, and 50.66, respectively. The calibrated images were saved for further analysis. Statistical Analyses The eggshell color data for each hen were the average of the color values of 3 eggs obtained from that hen in a given month. Average L*, a*, and b* values obtained Table 1. Change in color of eggshells [Lightness (L*), redness (a*), yellowness (b*)] 1, egg weight, and shell weight during the laying cycle 2 Mean egg Mean shell Month Mean (L*) ± SE Mean (a*) ± SE Mean (b*) ± SE weight (g) ± SE weight (g) ± SE 1 (Nov.) 62.0 e ± b ± e ± d ± (Jan.) 71.5 c ± ab ± ab ± c ± cd ± (Feb.) 70.2 d ± a ± ab ± b ± b ± (Mar.) 70.7 cd ± a ± bc ± a ± a ± (Apr.) 73.1 b ± cd ± c ± ab ± bc ± (May) 72.9 b ± e ± d ± ab ± c ± (Jun.) 73.6 b ± de ± bc ± ab ± d ± (Jul.) 73.2 b ± c ± a ± ab ± c ± (Aug.) 75.4 a ± f ± cd ± b ± e ± 0.03 a f Means within a column lacking a common superscript differ (P 0.01). 1 A higher L* value means lighter color; a higher a* value means a redder color; a higher b* value means a more yellow color.

4 CHANGES IN BROWN EGGSHELL COLOR 359 Figure 2. Eggs with color values [lightness (L*), redness (a*), and yellowness (b*)] representative of the mean color values recorded in each month. from the Color Expert software were used to study the changes in eggshell color with the age of the hens. A mixed linear model where time was the fixed effect and the hen was the random effect was used. The model was implemented with the mixed procedure (PROC MIXED) of the SAS System (v.8, SAS Inst. Inc., Cary, NC). Mixed methodology has been suggested to be the most appropriate method for analyzing repeated measures data with valid standard errors and efficient statistical tests (Littell et al., 1998). The repeated measures data in this study are color values of the eggs laid by the same hen over time. Normally, color values obtained from any 2 eggs laid by the same hen are correlated, and this correlation is higher if the color readings are taken close to each other in time. Thus, repeated measures data analyses are appropriate. The first step in the mixed model methodology is to model the covariance structure. Different structures (i.e., compound symmetry, unstructured, or a combination of autoregressive structure within hens and random effects between hens) can be fit to the data, and the one that provides the best fit as indicated by the goodness of fit criteria in the PROC MIXED output is the one to be used in the rest of the analyses. Next, time trends are analyzed by estimating and comparing means (Littell et al., 1998). At each step, once a time trend model for the means has been specified, fitting the model involves estimating the trend model parameters and reestimating the covariance structure (K. M. Portier, personal communication). Where required, least square means were computed and multiple comparisons of the least square means were carried out with the Bonferroni adjustment to prevent the inflation of the experiment-wise Type I error. RESULTS AND DISCUSSION Change in Eggshell Color, Egg Weight, and Shell Weight Over Time As the hens aged, the shells of the eggs they laid became lighter in color. Table 1 shows the average L*, a*, and b* of the eggshells. There was a marked increase in the lightness of the shells from the value recorded at the onset of lay for the young hens (mo 1). The lightness of the eggs, as indicated by a greater L* value, increased through the first 5 mo. Between mo 6 and 9, the mean lightness values were not significantly different from one another. However, this does not mean the lightness values had reached a plateau as evidenced by the significant increase in the mean lightness value observed in mo 10. The change in the redness values was in the opposite direction of the change in lightness. The redness

5 360 ODABAŞI ETAL. Table 2. Pearson correlation coefficients of lightness (L*), redness (a*), and yellowness (b*) values across months Month Corr Corr Corr Corr Corr Corr (n) (L n,a n ) (L n,b n ) (a n,b n ) (L n,l 10 ) (a n,a 10 ) (b n,b 10 ) * 0.14* 0.40* * * 0.64* 0.56* 0.55* * 0.26* * 0.57* 0.57* * * 0.67* 0.60* 0.64* * * 0.79* 0.58* 0.60* * * 0.84* 0.79* 0.79* * 0.19* 0.41* 0.81* 0.74* 0.69* * * 0.83* 0.76* 0.70* * 0.26* 0.48* *Indicates a significant correlation (α = 0.05) between the color values shown at the heading of the corresponding column. values, as indicated by smaller a* values, decreased with months, and the lightness values showed an increasing trend over time, indicating an overall decrease in the pigmentation of the eggs as the hens aged. The redness value did not reach a plateau and was still decreasing in mo 10. The changes in the yellowness of the eggs were less than the changes in lightness and redness of the shell after mo 1. To provide a visual image of the change in eggshell color, Figure 2 was generated. Each egg shown has color (L*, a*, and b*) values approximately equal to the mean color (L*, a*, and b*) values obtained in the indicated month (Table 1). The steep rise in lightness and yellowness from mo 1 (November) to mo 3, the decrease in redness from mo 5 (March) to mo 6 (April) and from mo 9 (July) to mo 10 are visible in this figure. Table 1 also shows the change in egg weight and shell weight over time. Egg weight increased through the first 5 mo, with the highest egg weight value occurring in mo 5, then plateaued. The highest value for mean shell weight occurred in mo 5, and the lowest value occurred in mo 10. In this study, the increased egg weight and shell weight in mo 5 and the decreased shell weight in mo 10 are possibly related to overall average feed intake (data not shown) being highest in mo 5 and lowest in mo 10. Table 3. Terms and their corresponding values used for the calculation of sample size Correlation Between Color Components To determine if a linear relationship existed between the color components (L*, a*, and b*), Pearson correlation coefficients were calculated. Table 2 shows the correlations between lightness/redness, lightness/yellowness, and redness/yellowness of eggshells from eggs collected during a given month in columns 2-4. For example, column 2 shows that the lighter the shell color, the less redness it contains. This negative correlation becomes stronger as the hens age, from a value of 0.75 in mo 1 to 0.93 in mo 10. As column 3 shows, because there is no asterisk, the correlation between the lightness/ yellowness of the eggshells was not significant during mo 3, 5, 6, 7, and 9. The correlation between redness/ yellowness was significant but not as strong as the correlation between redness/lightness (column 4). Column 5 shows the correlation of the lightness of the shells with the lightness of the shells in a future month (mo 10). Accordingly, hens laying lighter colored eggs will be laying lighter colored eggs in the future. Columns 6 and 7 show correlations between present and future values for redness and yellowness. When the color data from each month are correlated with those from last month, it can be observed that beginning with the third month significant correlations were obtained. There was no cor- Egg Shell Item L* 1 a* 1 b* 1 weight weight Equation 1 S 2 1, variance among hens S 2 2, variance within hens c 1, ratio of cost of sampling a hen over the cost of sampling an egg c M, number of eggs that can be obtained from a hen during the study Equation 2 r, relative precision x, sample mean S 2 1, variance among hens α, type 1 error L* (lightness), a* (redness), and b* (yellowness). 2 Estimates of these variances were obtained from PROC MIXED (SAS Inst. Inc., Cary, NC). 3 Sample mean was obtained as the intercept term in the solution for fixed time effect in PROC MIXED.

6 CHANGES IN BROWN EGGSHELL COLOR 361 Figure 3. Change in shell color [lightness (L*), redness (a*), and yellowness (b*)] after the correction for the change in egg weight. The power term (two-thirds or 0.67) is used to convert from a function of volume (egg weight) to a function of surface area. relation between the first month s color values and those of the last month. One possible explanation for this lack of correlation is because after the first month egg size increases far beyond that observed in the first month and the color is diluted over a larger surface area. Mean egg weight increased by 6.81 g from mo 1 to 3 (58.53 vs g, respectively); from mo 3 to 10, the change in mean egg weight only ranged from 1 to 1.96 g. Overall, the significant correlations in Table 2 indicate that hens that lay eggs with a lighter color in the early part of the laying cycle will also lay lighter colored eggs later in the cycle. These lighter colored eggshells (higher L*) will have less redness (a*). Eggshell Color Corrected for Egg Weight Just as Roland (1979) concluded that the decline in eggshell quality with age of the hen is a result of an increase in egg size without a proportionate increase in calcium carbonate deposition in the eggshell, the data collected in the present study provide evidence that the change in the eggshell color as the laying hen ages is also attributed to an increase in egg size without an accompanying increase in pigmentation. As a result, more shell surface is covered with a given amount of the pigment as the hen ages and lays larger eggs. Figure 3 shows the change in color values corrected for egg size. The power term (two-thirds or 0.67) in the correction shown in the legend of Figure 3 was used to convert from a function of volume (egg weight) to a function of surface area. When regression lines are fitted to the data points in Figure 3, it was observed that the R 2 values obtained for corrected lightness and corrected redness were The regression line for corrected lightness had a slope of 0.04, indicating that when corrected by egg weight, lightness increased by only 1% of the initial value recorded at mo 1. Similarly, the redness of the eggshells decreased by only 3% of the initial value recorded at mo 1 when the change in egg weight was taken into consideration. Looking at this figure, the slopes of the lines for lightness (L*) and redness (a*) in the eggshell clearly show the very small change in these values, which are 1 to 2 orders of magnitude smaller than their initial value at mo 1. This means that once corrected for egg weight, very little change in eggshell pigmentation was observed. Calculation of Sample Size Required for Measurement of Eggshell Color The data collected in this experiment allow for the calculation of the sample size to be used in similar studies of eggshell color. In a 2-stage sampling set-up, where the first stage is the sampling of the hens and the second stage is the sampling of the eggs, the equations [1] and [2] can be used to determine the sample size of eggs and hens, respectively (K. M. Portier, personal communication). m opt = S 2 c 1 c 2 S2 1 S2 2 M where c 1 = the cost for sampling a first-stage unit (hen); c 2 = the cost of sampling a second stage unit (egg) once [1]

7 362 ODABAŞI ETAL. Table 4. Sample sizes required to estimate the color values [lightness (L*), redness (a*), yellowness (b*)] of eggshells and the weight of the eggs and the eggshells Egg Shell Number of L* a* b* weight weight Hens, n Eggs, m opt the first stage unit (hen) has been obtained; S 2 1 and S 2 2 = the first and second stage population variances; M = the number of second stage units in each first stage unit; and m opt = the optimum number of second stage units measured from each of the first stage units in the sample. n o = t2 n 0 1,α/2S 2 1 r 2 x 2 [2] where r = the relative precision; x = the sample mean; and S 2 1 = the population variation. Equations 2 and 1 were used to calculate the number of hens to be used in an experiment and the number of eggs to be sampled from each hen, respectively. The numerical values used for each of the terms in these equations are shown in Table 3. The first and the second stage variances in equation [1] are the between-hen and within-hen variances, respectively. These variances were obtained from the solution of the linear mixed model. The sample mean is the intercept term in the solution for the fixed effect of time on color values of the eggshell, egg weight, or shell weight. The ratio of the cost of first stage sampling (establishing the hen house, buying layers, flock management, etc.) to the cost of second stage sampling (obtaining eggs from already established layers) was assumed to be 100. For M, which is the number of eggs that can be obtained from each hen, 15 was used, based on the assumption that each hen would lay an egg a day in a 15-d study. However, it was observed that the same results are obtained for larger values of M, even for M. The results of these calculations are shown in Table 4. The sample size required to estimate the mean a* of eggshells within 5% of the true mean turned out to be the largest when compared with the requirements for L* and b*. This is because the variability between the hens relative to the sample mean is the largest for a*. However, it is not a large sample size compared with the number of hens housed for research studies. Also, the sample sizes calculated are only valid for studies that involve a single strain of brown egg layer because the presence of hens in a flock from different strains would increase the variability in the data. The changes in the shell color of brown eggs obtained from commercial-type brown egg layers in their first production year were studied. It was observed that time had a significant effect on all 3 components of shell color: L*, a*, and b*. Lightness of the shell color increased as the flock aged, whereas a* decreased. Changes in the b* component of shell color were not substantial. The shells of the eggs laid within the first month of the laying cycle were much darker in color compared with the color of eggshells obtained later in the cycle. Also, there was a significant increase in egg weight from mo 1 to 3 and a continual significant increase in egg weight through mo 5. As was expected based on the previous literature, egg weight plateaued for the remainder of the laying cycle. The L*, a*, and b* components of brown eggshells were found to be correlated. The correlation of the a* with the L* was found to be stronger than the correlation between the a* and b*. Future studies relating the concentration of the specific pigment(s) to the L*, a*, and b* values of the eggshells are recommended for a better understanding of the correlation among the color components of shells from hens laying brown-shelled eggs. As discovered by Roland (1979), hens that lay eggs with good quality shells at the beginning of the laying cycle will lay eggs with good shell quality later in the laying cycle. Similarly, in this present study, the correlations of the lightness in the last month with those of previous months suggested that the hens that lay eggs with more pigment on the shell (darker) would continue to lay darker eggs in the future. This, of course, is assuming that the many factors known to negatively affect eggshell pigmentation throughout the laying cycle are considered (Lang and Wells, 1987; Butcher and Miles, 1995). However, the linear relationships between the color components of the last month with those of the preceding months were not strong enough to enable good prediction of the color of the eggshells as a flock of hens aged. When the color components were corrected by egg weight for the change in egg size as hens aged, the color components were found to be practically stable in time. The data collected in this study indicated that eggshell pigmentation decreased throughout the first 10 mo of the laying cycle. The larger eggshell surface area, due to the increase in egg weight, resulted in lighter colored eggs. This is similar to what Roland et al. (1975) concluded about the decline in eggshell quality with age of the hen because a constant amount of eggshell calcium deposition occurred as the hen aged. These data collected in this long-term study indicated that a decline in eggshell pigmentation should be expected as the laying cycle continues. However, it must be kept in mind that there are numerous factors that have a negative bleaching effect on eggshell pigmentation (Lang and Wells, 1987; Butcher and Miles, 1995). These factors must be controlled so that the impact on the normal decline in pigmentation will be minimal. REFERENCES Arthur, J. A., and N. O Sullivan Breeding chickens to meet egg quality needs. Int. Hatchery Pract. 19:7 9. Butcher, G. D., and R. D. Miles. Factors causing poor pigmentation of brown-shelled eggs Cooperative Extension Service Fact Sheet VM94. Inst. Food and Agric. Sci., Univ. Florida, Gainesville.

8 CHANGES IN BROWN EGGSHELL COLOR 363 Eyal, A., and J. R. Moran Egg changes associated with reduced interior quality because of dietary vanadium toxicity in the hen. Poult. Sci. 63: Fletcher, D. L., W. M. Britton, G. M. Pesti, and A. P. Rahn The relationship of layer flock age and egg weight on egg component yields and solids content. Poult. Sci. 62: Hy-Line International Hy-Line variety brown- Commercial management guide Hy-Line Int., West Des Moines, IA. Lang, M. R., and J. W. Wells A review of eggshell pigmentation. World s Poult. Sci. J. 43: Littell, R. C., P. R. Henry, and C. B. Ammerman Statistical analysis of repeated measures data using SAS procedures. J. Anim. Sci. 76: Luzuriaga, D. A Development and testing of an automated quality evaluation device for shrimp. MS Thesis. Univ. Florida, Gainesville. Luzuriaga, D. A Application of computer vision and electronic nose technologies for quality assessment of color and odor of shrimp and salmon. PhD Diss. Univ. Florida, Gainesville. Luzuriaga, D. A., M. O. Balaban, and S. Yeralan Analysis of visual quality attributes of white shrimp by machine vision. J. Food Sci. 62: , 130. Mertens, K., B. De Ketelaere, B. Kamers, F. R. Bamelis, B. J. Kemps, E. M. Verhoelst, J. G. De Baerdemaeker, and E. M. Decuypere Dirt detection on brown eggs by means of color computer vision. Poult. Sci. 84: National Research Council Nutrient Requirements of Poultry. 9th rev. ed. Natl. Acad. Press, Washington, DC. Odabasi, A. Z., R. D. Miles, M. O. Balaban, K. M. Portier, and V. Sampath Vitamin C overcomes the detrimental effect of vanadium on brown egg shell pigmentation. J. Appl. Poult. Res. 15: Roland, D. A., Sr Factors influencing shell quality of aging hens. Poult. Sci. 58: Roland, D. A., Sr., D. R. Sloan, and R. H. Harms The ability of hens to maintain calcium deposition in the egg shell and egg yolk as the hen ages. Poult. Sci. 54: Sell, J. L., C. R. Angel, and F. Escribano Influence of supplemental fat on weights of eggs and yolks during early egg production. Poult. Sci. 66: Sell, J. L., J. A. Arthur, and I. L. Williams Adverse effect of dietary vanadium, contributed by dicalcium phosphate, on albumen quality. Poult. Sci. 61: Solomon, S. E Egg and eggshell quality. Iowa State Univ. Press, Ames. Sutly, J. P., R. D. Miles, C. W. Comer, and M. O. Balaban The influence of vanadium on pigmentation of brownshelled eggs. Poult. Sci. 80:1039. (Abstr.)