GUTOWSKA et al. (1943) reported that

Effect on Performance of Selection for Level of Alka Phosphatase in Serum 1 F. H. WILCOX Dept. of Poultry Science, University of Maryland, College Park, Md. GUTOWSKA et al. (1943) reported that high producing hens had a higher level of alka phosphatase in serum than low producers. This finding prompted a study to obtain further information on this relationship and to determine if egg production would be increased by selection solely for increased alka phosphatase level in serum. Some of the results have already been reported (Wilcox et al, 1962, 1963; Wilcox, 1963). A positive correlation between alka phosphatase level of adult hens and egg production was observed, as well as a positive genetic correlation but no phenotpic correlation between phosphatase level at six weeks of age and egg production. Progress was rapid in development of a high phosphatase by selective breeding for enzyme level at six weeks of age. Initial data indicated that egg production of the high was also increased when compared to the random bred control from which the high originated. The purpose of this communication is to report more extensive data on egg production of the high as well as the relative performance of this in respect to other productive traits. MATERIALS AND METHODS Selection for high phosphatase level was practiced in the high for five generations (Si-S-,) after which the was reproduced by random mating for an addi- 1 Scientific Article No. A1246 Contribution No. 3757 of the Agricultural Experiment Station. (Received for publication December 22, 1965) 776 tional two generations (RiHR 2 ). Procedural details for the five generations of selection have previously been presented along with the methods used for handling and analyzing serum (Wilcox, 1963). In all generations matings of full sibs were not employed and only rarely were first cousin matings used. A control was used throughout; this consisted of Cornell Random Bred White Leghorns shipped as pedigreed hatching eggs each year, either from Cornell University or the North Central Regional Poultry Breeding Laboratory, Lafayette, Indiana. During the first four generations (Si-S*) chicks hatched from these eggs served as controls. During the remaining three generations (S 5, Ri-R 2 ) the parents of the control were shipped as ha + ching eggs; the parents were selected and mated at random with the restriction that no full sibs were mated together. Number of parents used each year for controls for the S 5, Ri and R 2 generations and Ri and R 2 generations of the high were IS males and 37-45 females per (only 20 for control for S 5 ). In the high each sire-family was represented in males used in matings. Chicks of the control and high s were raised intermingled throughout. During the first five generations 60 pullets were selected per, and each housed in a separate pen. During the first four generations of the high birds selected were those with high phosphatase level at six weeks of age. Birds of the fifth selected generation were chosen at random. Data based on extremes rather than the entire

SERUM PHOSPHATASE AND PERFORMANCE 777 TABLE 1. Serum alka phosphatase levels at different ages* Sex Females Males Generation Sot S, S 2 S 3 S 4 S 6 Ri R 2 Ri 66.2 46.5 40.6 35.1 26.5 44.6 75.2 45.9 91.2 6 weeks 64.6 67.3 72.4 74.5 145.7 381.9 261.8 503.5 7.9 6.1 9.9 3.6 4.5 14.1 9.5 5.2 8 months 10.4% 15.9 10.2 10.6 25.5 26.8 27.4 27.2 3.0 4.3 27 months 6.1 4.6 9.7 9.5 14.6 * mm nitrophenol/l./hr., antilogs. Data based on following No. birds per figure: for control at 6 weeks, 30-60, except 158 for S 0 generation; for high at 6 weeks, 95-365 for S1-S4, 60 for S 5, and 30 for Ri and R2 generations; for both s at 8 months, 28-30 except 117 for So and 49 for Si generations of control and 19 for males of both s; for both s at 27 months, 18-30. Certain of these data have been published elsewhere (Wilcox et al., 1962; Wilcox, 1963) and are included for purposes of completeness. f So = random bred population from which original selection was made, Si = progeny obtained from matings of birds selected from original population, S2-S5 = subsequent generations whose parents were selected for high phosphatase level, and R1-R2 = generations whose parents were selected randomly. J Figures in italics based on extremes selected for high phosphatase level at 6 weeks of age. are listed in italics. During the Ri and R 2 generations all pullets were placed in individual cages in a randomized order. Blood serum for alka phosphatase analysis of adult hens was collected during the afternoon from hens which had laid an egg earlier in the day. Egg production records were collected on a pen basis for hens in pens and on an individual basis, 5 days per week, for hens in cages. Egg quality data are based on three consecutively-laid eggs per bird. Shell thickness measurements are based on three determinations and include the wet shell membranes. Analysis of variance was used in comparing all traits with the exception that Chi square was used in comparing pen egg production, mortality, and sex ratio. Variates were transformed to logarithms for alka phosphatase level and to angles for fertility and hatchability. RESULTS The effect of selection for high alka phosphatase level on the enzyme level at three different ages is shown in Table 1. The level in the high was significantly greater (P <.01) than the control throughout. It is of interest that the difference was maintained to at least the end of the second laying year, even though the criterion of selection was level at six weeks of age. This was the case whether or not the values for adults of the high represented extremes selected at six weeks of age or a sample of the entire population. A marked variation from year to year is evident, especially at the two youngest ages. This variation occurred in spite of the fact that the birds were fed the same sequences of rations each generation and that age of blood sampling and analytical procedures were carefully standardized. The existence of such variation increases the importance of comparing each value of the high with its control. With one exception phosphatase level decreased with age. Level was decreased at eight months to 14 percent of the six week value, and at 27 months to 10 percent of the six week value. Individual variation is shown graphically for a typical generation (S 5 ) in Figure 1. The values for the high are spread out more than those of the control ; this

778 F. H. WILCOX CONTROL LINE, HIGH LINE 20-1 YEARLINGS L PULLETS CHICKS I II 50 100 50 100 150 mm nitrophenol ///hr. 200 250 300 350 FIG. 1. Histogram of phosphatase level of females of random bred control (stripes) and high phosphatase (solid) at three ages for the fifth selected generation (30 birds per group, except 28 for yearlings of the control ). effect is especially evident at the youngest age. Data on alka phosphatase level of males were obtained for only one generation (Table 1). The level in the high was significantly greater (P <.01) than in the control. Differences due to sex can be noted by comparing these data to those of females of the same generation. Level was higher in males than females for both s at six weeks of age (P <.05) and in the control was lower in males than females at 8 months (P <.01). However, no significant difference due to sex was observed in adults of the high. The effect of age on phosphatase level and the higher level in adult females as compared to males (control ) are in agreement with the literature (see McDaniel et al., 1964). A higher level in young males as compared to females has not been found consistently. Such a difference is present in data presented by Bell (I960), Tanabe (1962), and Tanabe and Wilcox (1960), but others (Correll and Wise, 1938; McDaniel and Dempsey, 1962) observed no sex difference at this age. Egg production data from pens are presented in Table 2. Production during the first laying year was greater in the high than the control during all generations except one (S 4 ). No reason for this discrepancy is apparent. All differences are statistically significant (P <.01). There was considerable fluctuation in the relative difference from one generation to another, possibly because only one pen was involved per datum. Therefore, data were subsequently obtained from larger numbers of birds kept in cages and are listed in Table

SERUM PHOSPHATASE AND PERFORMANCE 779 TABLE 2. Percent egg production of control Une and high phosphatase ; birds kept in pens* Generation Sit S 2 S3 S4 s 5 60.6 66.0 68.2 65.9 6 71.0% 71.5 71.9 58.3 66.8 1st laying year (Oct.--Sept.) 2nd laying year (Oct.--Jun.) 42.7 43.8 40.7 49.6. 48.5 40.0 46.2 47.1 37.6 43.6 * Each figure based on 60 birds at start of laying year. t Terminology is given in Table 1. I Figures in italics based on extremes selected for high phosphatase level at 6 weeks of age. 3; individual records were obtained for these birds. Results are comparable whether computed as survivors or hen-day egg production. Statistical analysis was performed for the former since data of individual birds could be utilized for an analysis of variance. For both generations the production of the high was over ten percent greater than the control; these differences were highly significant (P <.01). The difference between the high and the control was maintained throughout the year, as illustrated in Figure 2. Some data are also available on egg production during the second year (Table 2). There were no consistent differences between the high and the control. The performance for the high and control for other productive traits is TABLE 3. Percent egg production during first egg laying year {Oct.-June) of control and high phosphatase ; birds kept in cages* Generation - Survivors Egg Production Rit 59.7 R 2 53.7 65.2 59.7 Hen-day Egg Production 58.9 53.1 * Each figure based on 168-265 birds. t Terminology is given in Table 1. 65.0 58.7 shown in Tables 4 and 5. The only consistent, significant differences were slightly increased egg weight, slightly decreased shell thickness, lowered fertility and possibly fewer blood spots in the high than the control. DISCUSSION 0 There is evidence that the higher egg production of the high is not merely the result of selection of females in the starting generation which were superior egg producers, with subsequent maintenance of this superiority. Trap-nest records at 7-9 months of age show that the egg production of the control population from which the high originated was 66.1 percent as compared to 68.1 percent for the hens within this population which were used to start the high. This difference is smaller than the average difference obtained after selection. Furthermore, the original difference would be exeo 70 60 O 50 r> 0 0 OL 40 O Ul 80 t- 2 LU " 70 UJ 50 40 ^T\^ v ^ L ^ R, >» ^^. > x ^^^ V v ^^ ^ ^ \ R e - ^ ^ \ y ^ ^ x / \ ^ \ ^ 30 1 1 1 1 1 1 1 1 OCT NOV. DEC. JAN FEB. MAR. APR. MAY JUN. FIG. 2. Survivors' egg production during the first laying year of random bred control and high phosphatase for Ri and R s generations.

780 F. H. WILCOX TABLE 4. Performance of females of control and of high phosphatase reproduced by random mating* Ri Generationf R 2 Generation Body wt., g: 6 wk. 1 yr. Mortality, %: 0-6 wk. (both sexes) 6 wk.-6 mo. 6 mo.-15 mo. Age at first egg, days Egg wt. 1 yr., g. Haugh units, 1 yr. Shell thickness, 1 vr., ju. Blood spots, 1 yr., % Sex ratio at hatching, %c? 422 2,200 4.1 2.4 11.6 58.5 77.1 400 13.0 49.5 420 2,091** 4.7 4.1 6.7 60.1** 80.4** 390** 9.6 46.4 428 2,126 2.6 9.1 162.2 58.7 77.4 341 13.9 46.4 449** 2,144 1.5 5.0 6.8 162.4 61.0** 77.4 330** 7.5** 47.1 * Data based on following No. birds per figure: 6 wk. body wt., 30-35; 1 yr. body wt. and egg quality traits, 150/204; 6 wk.-6 mo. mortality, 6 mo.-15 mo. mortality and age first egg, 180-279; and 0-6 wk. mortality and sex ratio, 384-539. t Terminology is given in Table 1. ** Significantly different from control at 1% level. pected to be responsible for only a small portion of the subsequent higher production of the high because the female contributes only part of the genome of her daughters and also because of the moderate heritability estimated for hen-day egg production of this control (King, 1961). Another possibility that can also be ruled out is that the increased egg production of the high is the result of a relationship whereby egg production influences phosphatase level rather than the opposite. Such a relationship appears to exist since phosphatase level has been shown to increase at onset of egg production (Common, 1936; Tanabe, 1962). The basis for ruling this out as a factor in the present experiment is that selection was based on phosphatase level at an age (6 weeks) when egg production could exert no influence. Presumably alka phosphatase either itself influences egg production or its level is a reflection of other biochemical or physiological conditions which are responsible for genetic variation in egg production. The physiological mechanism whereby phosphatase level is related to egg production is not known. One explanation might be that the differences in phosphatase level are the result of differences in thyroid secretion rate or in differences in response to thyroxin. Production rate has been shown to be positively related to thyroxin secretion rate (Booker and Sturkie, 1950); and thyroxin has been found to increase alka phosphatase level with sufficient consistency that the enzyme level can be successfully used in place of thyroid weight in the goitrogen assay (Tanabe, 1964). HOW- TABLE 5. Percent fertility and haichability of Generation of parents S 6 f Ri R 2 88 87 90 eggs laid by control and high phosphatase * Fertility 74** 79 76** Hatchability (fertile eggs) 78 82 79 69 81 77 * Each figure based on 15 males mated to 37-45 females. No. eggs set was 637-856. t Terminology is given in Table 1. ** Significantly different from control at 1% level.

SERUM PHOSPHATASE AND PERFORMANCE 781 ever, the physiological mechanism does not appear to involve the thyroid. It has previously been observed (Wilcox, 1963) that the high does not differ from the control in the response to thyroxin as measured by phosphatase level. A higher thyroxin secretion rate of the high is unlikely since the relative superiority of enzyme level was unchanged when both s were fed thiouracil. This is shown by the following data based on 18-19 birds per figure, seven weeks of age, in which interaction was not significant: Alka Phosphatase Level (mm nitrophenol/l./hr., antilogs) Line No thiouracil 43.8 85.5 Thiouracil (0.2% in feed, 2wks.) 25.6 48.0 There is little basis for an action of phosphatase while in the vascular system. At this stage the enzyme presumably performs no physiological function, but is en route to the liver for excretion. The level in serum is relatively low compared to the level in certain other organs. The most plausible explanation at this juncture would appear to involve the duodenum and /or the ovarian follicle, which have higher phosphatase levels than any other organs of the hen, even though there is no difference between s in phosphatase level of these organs (Wilcox, 1963; Wilcox and Cloud, 1965). Previously, evidence has been obtained suggesting that the duodenum is responsible for much of the increased enzyme in serum of high chicks on the basis of borate inhibition studies (Wilson and Wilcox, 1963). A significant difference (P <.01) in borate inhibition also exists between adults of the high and control s as shown by the following data (30 hens per figure): Random Bred Line (Alka phosphatase level (mm nitrophenol/l./hr., antilogs) 8.6 15.9 Residual activity in presence of borate (% of control) 90 74 Chicks of the high also differ from those of the control in response to dietary fat as measured by serum phosphatase level. The higher egg production of the high might be due to increased absorption of fat resulting in more rapid growth of ova. The ovarian follicle might be involved since it also functions in transport of fat. The participation of the ovary is suggested because the main effect of selection for phosphatase level is on egg number. If bone phosphatase were involved one would expect a greater effect on shell quality than egg number, whereas the reverse was the case. This is not to suggest that bone phosphatase has no role in egg formation, only that it seems not to be a prime factor in the relationship observed regarding egg number. Peterson and Parrish (1939) and Taylor et al. (1965) observed changes in serum alka phosphatase level during the cycle of egg formation, and suggested the involvement of the bone; however, the changes they reported might also be due to an influence of the egg laying cycle in other organs with high phosphatase level (duodenum, ovarian follicle, kidney, liver). For example, the increased serum phosphatase level during periods of active shell formation might originate, at least in part, from the kidney during excretion of phosphorus removed from bone. The main purpose of this study was to determine if performance of a productive trait, i.e. egg production, could be improved by selection solely for level of a chemical in blood. The results indicate this to be the case, since egg production of a selected for high phosphatase level was

782 F. H. WILCOX higher than the control in all but one year. These findings are similar in certain respects to those of Hutt and Crawford (1960), who were successful in altering resistance to pullorum by selection for differences in body temperature within the first week after hatching. It is likely that other relationships exist between productive traits and chemical or physiological measurements. Further knowledge on this subject should be useful, especially if researchers can discover the precise chemical means by which genetic variation is able to express itself in phenotypic variation of productive traits. Measurement of alka phosphatase level might be useful in selection for egg production. Since a relationship exists in respect to level in the young chicken it should be possible to screen a large population for enzyme level and discard a portion of it with low level at an early age. This would reduce the number of females on which production records are kept. More important, some measure of the genetic potential for egg production of males could be obtained. This would be useful even if the relationship is not close, since only a small portion of the males can normally be kept, and a poor measure would be better than none at all. There is, however, a risk in suggesting such possibilities that some may conclude, or infer that the author has concluded, that selection for alka phosphatase is the answer to the breeders' problems in respect to egg production. Such a conclusion is clearly out of the question because there are doubtless many other chemicals involved in gene action which results in variation in egg production. Furthermore, there is no assurance that the relationship observed in a single study and a single stock holds for other strains. Therefore alka phosphatase level can at best be used only as an aid in a breeding program. Knowledge of the effect of selection for high phosphatase level on other productive traits besides egg production is important for anyone contemplating selection for this enzyme. A consistent change was observed in only three or possibly four of these traits; in only one of these, fertility, was the effect in a direction and of sufficient magnitude to be of consequence in selection for phosphatase level. The slight increase in egg weight and reduction in shell thickness may be a secondary effect caused by the increase in production. Such a shift is indicated by genetic correlations previously reported on the control stock used in the present study. King (1961) observed a positive genetic correlation although King et al. (1963) found essentially no correlation between egg production and egg weight. The latter workers also observed a negative genetic correlation between egg production and shell quality (specific gravity). King et al. (1963) found no genetic correlation between egg production and blood spot incidence. The explanation for the possible effect on blood spots may involve the ovary with its high level of phosphatase, but'there appears to be no explanation for the lowered fertility of the high. The increased egg production of the high phosphatase is in agreement with correlation data previously reported on the random bred control used in this study (Wilcox et al., 1962, 1963). Most of the other productive traits were not altered to the degree expected from previous estimates of their genetic correlation to alka phosphatase level. Thus body weight and egg weight were not reduced, albumen quality was not increased, nor did blood spot incidence remain unchanged. However, shell strength was reduced and age at first egg unchanged as predicted. No other estimates have been reported of genetic

SERUM PHOSPHATASE AND PERFORMANCE 783 correlations for the productive traits studied herein, except body weight; Matsumoto et al. (1960) reported a negative genetic correlation between phosphatase level and body weight, whereas Okada and Tsutsumi (1963) reported a positive one. These estimates may not be applicable to the present study because a different strain was used. They are also not strictly comparable with the correlation values of Wilcox et al. (1963) who transformed phosphatase level to logarithms, whereas this was not done by the Japanses workers. SUMMARY Performance of a selected for high level of alka phosphatase in serum at six weeks of age was compared to that of the random bred control from which the originated. Phosphatase level was significantly higher in the high at all ages studied (six weeks, eight months, 27 months) and during all five selected generations and two subsequent generations of random breeding. Egg production of the high was greater (P <.01) than the control in all generations but one (S 4 ); the relative difference fluctuated considerably from one generation to another, but was over ten percent greater for the high during both of the last two generations in which individual records and larger numbers were used. The high also exhibited slightly increased egg weight, slightly decreased shell thickness, lowered fertility, and possibly fewer blood spots. The possible physiological bases for the increased egg production were discussed; and the likest explanation would appear to involve either the duodenum and/or ovarian follicle. REFERENCES Bell, D. J., 1960. Tissue components of the domestic fowl. 4. Plasma-alka-phosphatase activity. Biochem. J. 75: 224-229. Booker, E. E., and P. D. Sturkie, 1950. Relationship of rate of thyroxine secretion to rate of egg production in the domestic fowl. Poultry Sci. 29: 240-243. Common, R. H., 1936. Serum phosphatase in the domestic fowl. J. Agr. Sci. 26: 492-508. Correll, J. T., and E. C. Wise, 1938. Studies on the relative efficiency of vitamin D from several sources. II. Influence of vitamin D of different origins on the serum phosphatase of the chicken. J. Biol. Chem. 126: 581-588. Gutowska, M. S., R. T. Parkhurst, E. M. Parrot and R. M. Verburg, 1943. Alka phosphatase and egg formation. Poultry Sci. 22: 195-204. Hutt, F. B., and R. D. Crawford, 1960. On breeding chicks resistant to pullorum disease without exposure thereto. Can. J. Genet. Cytol. 2: 357-370. King, S. C, 1961. Inheritance of economic traits in the regional Cornell control population. Poultry Sci. 40: 975-986. King, S. C, L. D. Van Vleck and D. P. Doolittle, 1963. Genetic stability of the Cornell randombred control population of White Leghorns. Genet. Res. Camb. 4: 290-304. McDaniel, L. S., and H. A. Dempsey, 1962. The effects of fasting upon plasma enzyme levels in chickens. Poultry Sci. 41: 994-998. McDaniel, L. S., H. A. Dempsey and H. L. Chute, 1964. Enzyme levels in birds. Maine Agri. Expt. Sta. Bull. T8. Matsumoto, K., T. Tonoue and I. Okada, 1960. Heritability of physiological characters of chickens. III. Serum alka phosphatase activity and its relation to growth. J. Fac. Agri. Hokkaido Univ. 51: 315-323. Okada, I., and Y. Tsutsumi, 1963. Heritability of physiological characters of chickens. IV; Further studies on body weight, gain, serum alka phosphatase, and their relationships, using a diallel mating. Jap. J. Zoot. Sci. 34: 114-120. Peterson, W. J., and D. B. Parrish, 1939. Fluctuations of phosphatase and inorganic phosphorus in the blood of the laying hen during the period of egg formation. Poultry Sci. 18: 54-58. Tanabe, Y., 1962. Influence of age upon the ability of thyroxine and estrogen to increase serum alka phosphatase of the chicken. Gen. Comp. Endocrinol. 2: 446-452. Tanabe, Y., 1964. Comparison of estimates for thyroxine secretion rate determined by the goiter prevention assay and by measuring the activity of serum alka phosphatase in the chicken. Endocrinol. Japon. 11: 260-264. Tanabe, Y., and F. H. Wilcox, 1960. Effects of

784 F. H. WILCOX age, sex and on serum alka phosphatase of the chicken. Proc. Soc. Exp. Biol. Med. 103 : 68-70. Taylor, T. G., A. Williams and J. Kirkley, 1965. Cyclic changes in the activities of plasma acid and alka phosphatase during eggshell calcification in the domestic fowl. Canad. J. Biochem. 43:451-457. Wilcox, F. H., 1963. Genetic control of serum alka phosphatase in the chicken. J. Exp. Zool. 152: 195-204. Wilcox, F. H., and W. S. Cloud, 1965. Alka EVIDENCE has been reported by several workers that heat tolerance in chickens is genetically controlled. Hutt (1938) observed that mortality during three successive hot days was affected by age and breed, and that White Leghorns had less mortality than Barred Plymouth Rocks and Rhode Island Reds. Yeates et al. (1941) found that White Leghorns showed less change in body temperature than did Australorps when subjected to increased temperatures. In a study of the response of six different breeds to 105 F. (40.56 C.) and 25% relative humidity (RH), Lee et al. (1945) found that White 'Florida Agr. Exp. Stations Journal Series No. 2275. 2 Poultry Science Department. 3 Present Address: Soccion de Zootecnic, Centro de Investigaciones Agronomices, Maracay, Venezuela. 4 Agricultural Engineering Department. 5 Present Address: Poultry Science Department, Auburn University, Auburn, Alabama. * Dairy Science Department. phosphatase in the reproductive system of the hen. J. Reprod. Fert. 10: 321-328. Wilcox, F. H., L. D. Van Vleck and W. R. Harvey, 1963. Estimates of correlations between serum alka phosphatase level and productive traits. Poultry Sci. 42 : 1457-1458. Wilcox, F. H., L. D. Van Vleck and C. S. Shaffner, 1962. Serum alka phosphatase and egg production. Proc. XII World's Poultry Cong., Sydney, pp. 19-22. Wilson, H. R., and F. H. Wilcox, 1963. Origin of an increased serum alka phosphatase in chicks. Proc. Soc. Exp. Biol. Med. 113: 413-415. Familial Differences of Single Comb White Leghorn Chickens in Tolerance to Ambient Temperature 1 H. R. WILSON, 2 A. E. ARMAS, 3 I. J. Ross, 4 R. W. DORMINEY 5 AND C. J. WlLCOX 6 Florida Agricultural Experiment Station, Gainesville, Florida 32603 (Received for publication December 22, 1965) Leghorns had the least rise of body temperature. However, when the RH was increased to 75%, Brown Leghorns showed the least reaction, White Leghorns were intermediate and Australorps and Rhode Island Reds showed the most reactions. Hillerman and Wilson (1955) tested the reaction of hens to sudden changes in temperature and compared this to their adapted performance. They found that the Red Jungle Fowl was the most efficient in regulation of body temperature and that White Leghorns were superior to heavy breeds. They noted that White Leghorns in production drank more water than did the other breeds but the reverse was true for non-layers. White Leghorns were less severely affected in production by high temperatures than were New Hampshire or White Plymouth Rocks (Huston et al., 1957). Clark and Amin (1965) reported strain differences in performance under various temperature regimens. Fox (1951) found that both White Plymouth Rocks