FACTORS AFFECTING SELECTION PROGRESS FOR SHELL STRENGTH. K. W. Washburn Department of PoultryScience The University of Georgia Athens, GA 30602

Transcription

1 FACTORS AFFECTING SELECTION PROGRESS FOR SHELL STRENGTH K. W. Washburn Department of PoultryScience The University of Georgia Athens, GA INTRODUCTION The per capita egg consumption has decreased considerably over the last 15 years. A major reason for this decrease is the concern over cholesterol and changing eating habits of today's family. However, I feel that some of this decline may be due to consumer dissatisfaction with the shell quality and subsequent rejection of eggs as a food item they wish to purchase. This negative effect on consumption added to the approximately 7% loss of eggs from the point of lay to the consumer, provides an enormous opportunity for dramatic improvement in economic return by improving the shell quality of commercial layers. In previous talks on the subject of egg shell strength, it has been stated that the average egg breakage (including cracks) from point of lay to consumers' use was 7%. Recent estimates indicate that this value has changed very little. Breeders of commercial egg layers are all placing selection pressure on improving shell strength. Thus, the question needs to be asked as to why so many of our eggs are lost because of weak shells. A number of questions, summarized in Table 1, come to mind in evaluating the genetic approach to improving shell strength. 136

2 The first question to be addressed is, "What is the status of the genetic Variation in shell strength?" A discussion of genetic variation can be divided into 5 categories: 1. Differences between brown and white shell color populations. 2. Differences between genetic groups. 3. Heritability estimates. 4. Long term selection experiments. 5. Specific long term selection experiments which used a commercial-type population, assessed the extent of diherences in different environments, and assessed changes in other important traits. Differences between brown and white shell color populations: A number of studies have shown differences in shell strength between brown and white shell color populations. These can be divided into pure breed comparisons and comparisons between commercial strains. Taylor and Martin (1928) found that the percent shell of brown eggs (from Barred Plymouth Rocks) was lower than that of white shelled eggs from White Leghorns. Tyler and Geake (1958) reported that the white shelled eggs from White Leghorns had thicker shells than the brown shelled eggs from Rhode Island Reds. Eggs from Brown Leghorns (white shell) was similar to those from the Rhode Island Red breed. Perek and Snapir (1970) reported that the mg per cm shell was greater in White Leghorns than in White Plymouth Rock (brown shell). Rodda (1972 found that a White Leghorn strain had better shell strength than a Rhode Island Red strain. Shell strength of commercial white egg strains were in general better than that of brown egg strains 137

3 (Potts and Washburn, 1974) and eggs from white egg commercial strains in random sample tests had better shell strength than those from brown egg strains (Hunton, 1980). Differences in shell strength between genetic groups (non-commercial) have been shown in a number of studies. Significant differences among dam families in percent shell ash were reported by Munro (1938). Genetic variation in shell thickness in a White Leghorn population was observed by Taylor and Lerner (1941). Significant differences in shell thickness between various breeds and lines were found by Farnsworth and Nordskog (1955). Significant differences in specific gravity in three White Leghorn strains were reported by Nagai and Gowe (1969). Difference between strains in tensile strength were observed by Carter (1970). Significant difference in percent broken eggs in White Plymouth Rock lines selected for divergence in body size were reported by Siegel et al. (1978). A desert inhabiting breed (Sinai) were found to have shells that were less porous and thicker than that of commercial White Leghorns (Arad and Marder, 1982). In contrast to these studies using non-commercial genetic groups, there are few published studies on differences between genetic groups which would be considered as commercial layers. Significant differences were shown in breaking strength, thickness and percent cracks in eight commercial strains (Bowman and Challender, 1963). Significant differences in shell strength among three white and three brown egg strains were reported by Potts and Washburn (1974). In this study shell quality was assessed by a number of methods and the deterioration in shell strength started earlier in brown egg strains. Differences were observed between 10 commercial 138

4 strains and crosses in shell strength measured by percent shell, shell weight, and relative changes with age (Hamilton, 1979). Heritability estimates (h2): A summary of h2 estimates for various genetic groups and methods of measuring is presented in Table 2. These estimates show a wide range in values. It appears that the h2 estimates obtained from brown egg populations is higher than that obtained from white egg populations. Many of these values were obtained on pure breeds and may not be representative of the amount of genetic variation in commercial layers. In the study of Potts and Washburn (1985) in which a single commercial leghorn type parent line Wasused, the h2 ranged from.18 to.31. The values of.39 and.31 were obtained as a composite value of many h2 estimates. It is probable that more emphasis should be placed on these estimates. Another study of importance is the relatively recent one by Grunder (1986) in which the h2of intact eggs were studied. These h2 estimates were relatively low (.10 to.26). Several of these studies measured shell strength in the same population at different ages. The data presented in Table 3 indicate that h2 of shell strength decreases with age. This may have an impact on the age at which shell strength is measured for selection studies. The reasons for this decline needs to be studied. Four conclusions can be obtained from the h2 estimates: 1. The h2 of shell strength should be considered moderate (about.3). 2. The h2 estimates obtained from white shelled eggs tend to be lower than those obtained from brown shelled populations. 139

5 3. The h2 of shell strength appears to decrease with age. 4. h2 estimates for current commercial laying strains are needed. In evaluating genetic variation, the most important consideration is progress in selection experiments. A summary of selection experiments for shell strength is presented in Table 4. These studies used a variety of genetic populations and methods of measuring shell strength. Most of these studies show that genetic progress can be made in improving shell strength. However, none of these studies utilized populations comparable to the current commercial lines. The genetic variation in shell strength can be summarized by the following points: 1. There is sufficient genetic variation to develop an effective selection program. 2. Genetic variation differs between populations. 3. The amount of measurable variation decreases with age. 4. Information is needed on heritabilities, selection progress, and correlated changes in current commercial laying strains. Basis for genetic variation in shell strength: There are a number of areas that provide the basis for genetic variation in shell strength. Two of these will be discussed: 1) Egg characteristics and 2) Physiological response. Selection for these traits could interfere with selection progress for shell strength if there were negative correlations between them and shell strength. 140

6 Possible changes in egg characteristic in response to selection for shell strength include 1) shape, 2) shell color, 3) membrane mass, and 4) egg weight. Some of the older data using pure breeds have shown a degree of correlation of egg shape and shell strength, the eggs with rounder shape having better shell strength. However, in commercial strains we have found little correlation between shape and strength (Table 5). The data in Table 5 is a composite of three brown egg and three white egg strains in which the shell strength was determined by a number of methods. Only for breaking strength of brown eggs does the correlation approach a magnitude of importance. Some of the older data, mostly using pure breeds, have also shown a relationship between egg shell tint and strength. However, in commercial brown egg strains tint does not appear to be an important factor in shell strength (Table 6). In this study, shell tint of eggs from two commercial brown egg strains was visually scored on a scale of 1 to 5 and these values correlated with shell strength measured by deformation, thickness, specific gravity, and breaking strength. One egg characteristic often overlooked does have an effect on shell strength. That characteristic is the shell membrane. In the study presented in Table 7, three commercial white and three commercial brown egg strains were divided into low deformation (high shell strength) and high deformation (low shell strength) groups. After determination of shell strength, the weight of the shell membrane of these eggs were then determined. This study shows that the shell membrane weight was associated with shell strength both in groups selected for differences in deformation and between brown and white shell groups which differed in shell strength. 141

7 _. The characteristic of the egg that is most often associated with shell strength is egg weight. As the hen becomes older the size of the egg increases and the shell strength decreases, however, the amount of shell deposited remains relatively constant or is only slightly increased, thus there is a relative decrease in the amount of shell. This relationship, utilizing deformation as the measure of shell strength, is shown in Figure 1. It is clear that the egg weight increased throughout the 77 week period and shell strength decreased (higher deformation = lower strength) with age after 37 weeks. This type of relationship has resulted in an association of increased egg weight with poorer shell strength. In the study shown in Table 8, the association of egg weight "andshell strength was studied in different strains utilizing four different methods to measure shell strength within specific age periods. In general, correlations between weight and strength within ages were low:" However, in strain 1 there was a moderate a correlation regardless of the method used to assess shell strength. Egg weight was also measured in the deformation studyutilizing a commercial grand-parent leghorn line to obtain information about the association of shell strength and egg weight. Phenotypic correlations between deformation and egg weight were near 0 and genetic correlations of these variables ranged from to (Table 9). These correlations indicate that selection for increased shell strength should not have a deleterious effect on egg weight. However, none of these studies used the current commercial layer which is smaller in size and probably lays an egg whose size is proportionally larger than those on which these data were obtained. 142

8 P_h_hysioloNeal respo_nscs: There are a number of possible physiological responses to selection for shell strength. These include: 1. Length of time egg is in shell gland 2. Rate of shell deposition 3. Uterine environment 4. Changes in acid-base balance *5. Efficiency in metabolizing calcium These are all important, but I would like to focus on the efficiency in metabolizing Ca. In discussing the efficiency of metabolizable Ca, I will focus on two points: 1. Feeding excess Ca to improve shell strength 2. Effects of selection emphasis on Ca efficiency The importance of Ca metabolism in shell formation is well documented and increasing Ca level to a certain level may result in improved shell strength. Thus, one might hypothesize that differences in response to dietary Ca may explain genetic differences in shell strength. In the study shown in Table 10, commercial layers were divided into high and low deformation groups. Each deformation group was divided into two treatment groups. The control group received a diet containing 2.9% Ca while the Ca treatment group received this diet plus free access to oyster shells. After a 3 week period, shell strength was assessed by deformation, breaking strength, and shell thickness. There were no consistent trends to suggest that the strength of eggs from 143

9 low shell strength groups receiving additional Ca was improved any more than that of the high strength groups. Some of our current problems with shell strength may be the result of the selection pressure to produce a layer which provides the opportunity for maximum economic returns. There is selection pressure to decrease body size to improve efficiency and increased egg numbers without changing egg size. In the pullet, the formation of new bone stays active only to the beginning of lay at which time it appears to shut down. Activity then shifts from the conical bone to the medullary bone. The hen uses dietary Ca and Ca from the medullary bone for shell formation. As the body size and bone mass is genetically decreased with expectations of more eggs and no decrease in egg size it may become more difficult to maintain sufficient Ca deposition to maintain good shell strength. This may become especially critical with combinations of aging and hot environmental temperature. Studies need to be conducted using the modern layer and measuring shell strength over the production period and in different environmental conditions. Individual Variation Considerable variation in shell strength is observed between individuals within a flock. This may or may not be all genetic variation but it is important from two aspects: 1) the differences seem to be predetermined at an early age, 2) the differences remain consistent at least over short periods. These differences observed when the birds first come into lay continue and are amplified throughout the period of lay (Figure 2). 144

10 Figure 3 shows the deformation values over a 7 week period of two individuals differing in basal deformation values. These values were consistent over the 7 week period. However, it is interesting to note that the fluctuations in the poor shell strength individual (high deformation) was greater over the period than was that of the individual with better shell strength. Studies need to be conducted to determine if factors which result in changes in shell strength have a greater effect on those individuals with inherently poorer shell strength. Methods of Measuring Shell Strength Important aspects of measuring shell quality include: 1) physical method, 2) age, and 3) the effects of environmental conditions. Physical method: The strength of the shell has most often been measured by various lab estimates of shell strength. Many studies have been done to determine which of these methods are the best. The results of our studies (Table 11) show a very high phenotypic correlation between the most used different methods of measuring shell strength indicating that any of these methods would be acceptable in lab-type evaluations of shell strength. However, this relationship may not be so clear-cut when comparisons are made over time. In the study presented in Table 12, week deformation was used to classify sire families into low or high deformation groups. At 40, 60, 68, and 77 weeks of age the breaking strength and percent shell, in addition to deformation, was obtained. Differences between deformation groups remained significant throughout the period. Differences in breaking strength were 145

11 significant for most of the ages. However, percent shell differences were significant only for the 41 week period. Lab tests such as these would be of limited value unless they have a reasonable degree of relationship to the proportion of eggs cracked or broken under commercial conditions. A number of studies have attempted to determine the association of laboratory shell strength measures with shell breakage. Two points make it difficult to assess this: 1. Considering all the points at which shells can receive an insult sufficient to cause breakage, all eggs do not receive the same insult under commercial conditions so that the degree of insult a particular egg receives is a matter of chance. 2. If the degree of the insult is not controlled it may be so great that it exceeds the degree of variation in shell strength of the samples being tested. However, a number of studies have shown there is a good degree of association between shell strength measured under laboratory conditions and breakage under commercial conditions (these are summarized in Table 13). In addition, Grunder et al. (1989) reported that the phenotypic correlation between a number of shell quality traits and % intact shells ranged from.19 to.47. Genetic correlations ranged from.23 to However, genetic correlations between shell strength measures at 42 weeks with intact eggs at 68 weeks were low except for shell weight. 146

12 Effects of age: In addition to the physical method used for measuring, the age of the bird at which determinations are made is important. Since the major problem in shell strength is with older hens the question arises - should the genetic selection be based on data obtained on older hens? This has two obvious disadvantages: 1) h2 of shell strength decreases with age, 2) generation interval is increased if data for selection is obtained at later ages. In the study shown in Figure 1, sire families were categorized as high or low shell deformation families on the basis of values obtained from weeks of age. Deformation values were then obtained through 77 weeks of age. The differences between the groups were maintained throughout the 77 week period and became greater from 41 weeks on as the shell strength of the high deformation families deteriorated at a faster rate than the low deformation families. This indicates that selection based on determination made at an earlier age of the production cycle should be effective in improving shell strength at older ages when the shell strength is in general poorer. However, at very early age there appears to be considerable variation in shell strength of individual birds. Figure 2 shows an example of the shell deformation of a typical individual from beginning of lay through 44 weeks of age. The shell deformation values were quite variable during the earlier ages. This suggests that in this particular population egg strength data obtained before 30 weeks of age may not be reliable. 147

13 Environment: A number of non-genetic environmental factors contribute to the variability in shell strength. These include such known things as age, season, ambient temperature, diet and disease. In the assessment of shell strength for use in a breeding program, these factors Contributing to shell strength must be carefully separated from the genetic contribution to the variation. In addition to these known factors that contribute to variation in shell strength, unknown factors may result in temporarily altered shell strength. The disturbing thing about these changes in shell strength is that the temporarily altered values may not be representative of the true shell strength of those individuals (Table 14). Shell strength values obtained during week 1 (where the shell strength was altered) were not well correlated with values obtained after the shell strength had returned to normal (week 2). To summarize, I would like to focus on what I think the future genetic approach to solving the shell strength problem should be: 1. Determine the basis of the individual variation in shell strength and its relationship to genetic variation. 2. Conduct selection experiments using current commercial type layers as base populations and measure correlated changes. 3. Evaluate various combinations of methods of measuring shell strength, age, and temperature. 4. Evaluate the inter-relationships of selection pressures for decreased body weight, increased egg numbers, and maintaining egg weight on the birds ability to maintain shell strength. 148 i

14 REFERENCES Arad, A., and J. Marder, Differences in egg shell quality among the Sinai Bedouin fowl, the commercial White Leghorn and their crossbreds. Br. Poult. Sci. 23: Bowman, J. C., and H. I. Challender, Egg shell strength. A comparison of two laboratory tests and field results. Br. Poult. Sci. 4: Buss, E. G., Genetic differences in avian egg shell formation. Poultry Sci. 61: Buss, E. G., R. M. Leach, Jr., and J. T. Stout, Eggshell quality for chickens in selected lines, Fl's. and F2's. Poultry Sci. 56: Carter, T. C., The hen's egg: some factors affecting deformation in statically loaded shell. Br. Poult. Sci. 11: Engstr6m, G., C. Weyde, and L. E. Liljedahl, Genetic correlations and heritabilities for frequency of cracked eggs, egg number and egg weight in laying hens. Br. Poult. Sci. 27: Farnsworth, G. M., Jr., and A. W. Nordskog, Breeding for egg quality. 3. Genetic differences in shell characteristics and other egg quality factors. Poultry Sci. 34: Gowe, R. S., and R. W. Fairfield, Long term selection for egg production in chickens. 3rd World Congress on genetics applied to livestock production. XII. Biotechnology, selection experiments, parameter estimation, design of breeding system management of genetic resources. Long term selection for egg production in chickens, pp

15 Grunder, A.A., R. M. G. Hamilton, R. W. Fairfield, and B, K. Thompson, Genetic parameters of egg shell quality traits and percentage of eggs remaining intact between oviposition and grading. Poultry Sci. 68: Hamilton, R. M. G., K. G. Hollands, P. W. Voisey, and A. A. Grunder, Relationship between egg shell quality and shell breakage and factors that affect shell breakage in the field - a review. World's Poultry Sci. J. 35: Hunton P., Genetic factors affecting egg shell quality. World's Poultry Sci. J. 38: Jaff6, W. P., Egg production, body weight and egg quality characters, their heritability and the correlations between them. Br. Poult. Sci. 7: Johnson, A. S., and E. S. Merritt, Heritability of albumen weight and specific gravity of eggs from White Leghorns and Barred Rocks and the correlations of these traits with egg production. Poultry Sci. 34: McNally, E. H., The relationship of egg shell weight to cracked eggs. Poultry Sci. 44: Munro, S. S Effects of heredity on interior egg quality and shell composition. Poultry Sci. 17: Nagai, J., and R.S. Gowe, Genetic control of egg quality. 1. Source of variation. Br. Poult. Sci. 10: Perek, M., and N. Snapir, Interrelationships between shell quality and egg production and egg and shell weights in Wfiite Leghorn and White Rock hens. Br. Poult. Sci. 11:

16 Pevzner, I. Y., G. W. Friars, H. L Orr, and B. S. Reinhart, The use of selections and strain crossing to reduce egg shell breakage. Br. Poult. Sci. 17: Potts, P. L, and K. W. Washburn, Shell evaluation of white and brown egg strains by deformation, breaking strength, shell thickness and specific gravity. Poultry Sci. 53: Potts, P. L, Sr., and K. W. Washburn, Genetic variation in shell strength and its relationship to egg size. Poultry Sci. 64: Quinn, J. P., C. D. Gordon, and A. B. Godfrey, Breeding for egg shell quality as indicated by egg weight loss. Poultry Sci. 24: Rodda, D. D., Breeding for late egg shell quality in the domestic hens. Br. Poult. Sci. 13: Siegel, P. B., J. H. van Middelkoop, and P. R. K. Reddy, Comparisons of frequencies and egg shell characteristics of broken and intact eggs within diverse populations of chickens. Br. Poult. Sci. 19: Strong, C. F., Relationship between several measures of shell quality and egg breakage in commercial processing plants. Poultry Sci. 67:162. Taylor, L. W., and J. H. Martin, Factors influencing thickness of egg shell. Poultry Sci. 8: Taylor, L. W., and I. M. Lerner, Inheritance of eggshell thickness in White Leghorn Pullets. J. Agri. Res. 58: Taylor, L. W., and I. M. Lerner, Inheritance of shell finish in Single Comb White Leghorns. J. Hered. 32:

17 Tyler, C., and F. H. Geake, Studies on egg shells. IX. The influence of individuality, breed, and season on certain characteristics of egg shell from pullets. J. Sci. Food Agric. 9: Van Tijen, W. F., Shell quality in poultry asseen from the breeder's viewpoint. 3. Heritabilities: Expected versus accomplished response. Poultry Sci. 56: Van Tijen, W. F., and A. R. Kuit, The heritability of characteristics quality, their mutual correlation and the relationship with productivity. of egg Arch. Geflu gelk. 34:

18 TABLE 1. Important questions in evaluating genetic approach to improving shell strength 1. What is the status of the genetic variation? 2. What is the basis for the genetic variation and could selection for other traits interfere with selection progress? 3. What is the influence of individual bird variation? 4. Is the method of measuring appropriate? 5. Are non-genetic factors interfering with assessment of genetic potential? TABLE 2. A summary of heritability estimates for shell strength Stock Method h_ Reference 7 Pure breeds Specific gravity.09 TO.24 Farnsworth & Nordskog (1955) White Leghorn Specific gravity.32 Johnson & Merritt (1955) Barred Plymouth rock Specific gravity.56 Johnson & Merritt (1955) White Leghorn Specific gravity.28 TO.47 Jaffe (1966) Strains (Various) Specific gravity.22 TO.47 Nagai & Gowe (1969) Rhode Island Red Specific gravity.57 TO.33 Rodda (1972) White Leghorn Specific gravity.27 TO.05 Rodda (1972) Commercial White Leghorn Deformation.30 TO.18 Potts & Washburn (1982) Parent line White Leghorn Cracks.11 TO.43 Engstrom et al. (1986) Various Review Various.39 Van Tijen & Kuit (1970) Various Various.31 Buss (1982) Review TABLE 3. Effect of age on h2 estimates Age (wks) Stock Change Reference White Leghorn.44 TO.21 Johnson & Merritt (1955) Rhode Island Red.57 TO.33 Rodda (1972) White Leghorn.27 TO.05 Rodda (1972) Commercial White Leghorn.30 Potts & Washburn (1982) Commercial White Leghorn Commercial White Leghorn Commercial White Leghorn.36 TO.25 Grunder et al. (1989) 153

19 TABLE 4. Summary of selection experiments Genetic _oup _ Generations Results Reference White Leghorn Breed Percent Shell 5.67% Difference Taylor & Lerner (1939) from base population White Leghorn Breed Egg weight loss 8 3.6% Difference Qulnn et al. (1945) in lfigh & low lines White Leghorn Strains Deformation 3 Significant in 1 Pevzner et al. (1975) strain but not in another White Leghorn G shell/cm _ 8 Percent in thick Buss et al. (1977) & % shell shell line = 10.7, percent in thin shell line = 7.7 White Leghorn Index 4 Thick t from Van tijen (lg'f0.35 mm to.37 ram, specific gravity t from to CornelK BreakingStrength3 2.9kginhigh& Parsons& Combs (1978) 2.47 kg in low line Random-bred Breakage 1 37% breakage & Garwood (1979) 234mm thickness in high line, 29% breakage &.36 mm thickness in lowline - Specific gravity 4 Specific gravity McPhee et al. (1982) 1".04 Egg-type Specific gravity 9 Significant Gowe (1986) improvement in specific gravity in 3 or 4 strains 154

20 TABLE 5. Correlation of shape (W/L) with shell strength. C0rr l_ti0n_ Specific Breaking Strain _W/L Thickness _ Deformation strength Brown egg White egg TABLE 6. Correlation of shell tint with shell strength in commercial strains Strain 1 Strain 2 Deformation Thickness Specific gravity Breaking strength

21 TABLE 7. Relationship of shell membrane to shell strength High vs low deformation _oups LOw Membrane High Membrane Deformation weight Deformation weight White egg a b Brown egg (P b Deformation White vs brown deformation groups Membrane weight White egg x Brown egg y TABLE 8. Correlation of egg weight with shell strength Straim Thickness Specific gravity Deformation Breaking strength

22 TABLE 9. Association of shell deformation and egg weight in a commercial leghorn-type grandparent population Correlation Age Phenotypic Genetic x x x x x x Correlations not obtained. TABLE 10. Effect of supplemental Ca on shell strength variation Ca treatment Control High Low High Low deformation deformation deformation deformation Deformation (X 10"3) Breaking strength(kg) Thickness (X 10.2 mm) High, low = groups selected for high or low shell deformation. 157

23 TABLE 11. Phenotypic correlations of different methods of measuring shell strength (31-37 weeks of age) Strains SG/DEF SG/TH _ DEF/TH DEF/B$ TH/B$ SG = specific gravity; DEF = deformation; TH = shell thickness; BS = breaking strength. TABLE 12. Deformation, breaking strength, and percent shell at later ages of families selected for high and low shell deformation at weeks of age Deformation Breaking strength Percent shell Age Low High Low High Low High a 2.84b 6.16a 5.66b 9.18a 8.76b _ 2.72b 6.05_ 5.98_ 8.94_ 8.68a _ 2.98b 5.65_ 5.21b 8.92_ 8.74_ & 3.42b 5.21Y 4.88b 8.34_ 7.98a 158

24 TABLE 13. Association of shell strength measures with shell breakage REFERENCE McNally (1965) CONCLUSION Logorithimic t in cracking with # shell weight. Wells (1967A) Almost perfect curvilinear relationship between percent cracks under commercial conditions and specific gravity measures. Wells (1967B) Shell strength of eggs that cracked during transit was less than for non-cracked. Wells (1967A) Percent cracks was correlated with all methods of measuring. Bowman & Challender (1963) Significant correlation between shell thickness and number of cracks produced on farm or intransit. Holder & Bradford (1979) As specific gravity t there was significant # in number of cracked eggs after passing through egg grading machine. Shrimpton &Hann (1967) Values were not good predictors of subsequent breakage. Strong (1988) Specific gravity and percent shell were correlated with percent cracks. Breaking strength, shell thickness, and shell weight were not. TABLE 14. Effects of disruptions on correlations of repeated measures Deformati0n/Deformation Percent shell/percent shell Week 1 and week Week 2 and week Week 3 and week

25 A _5,0_ v i_1_ mclh _3.0 - _ _ w Q I 1 t I I I I WEEKSON TRIAL B FIGURE 3. Individual variation in shell strength A = individual selected on the basis of high shell deformation before trial started. B = selected on basis of low shell deformation. 160

26 ,8 77 WEEKS OF AGE FIGURE 1. Shell deformation of high and low deformation families from weeks of age. 161

27 65m o-,,--,--4 W_IGM T I-...4 OFFOA _IATION,_4.0 so- " i. ',....:......,...,/:......:,_._ _ \." I(' O" : ql '.. "" 4o I I L I l L L I l , WEEKS OF AGE FIGURE 2. Shell deformation of an individual from 21 to 41 weeks of age. 162