CONTENTS ***** ***** *****

Transcription

1

2 CONTENTS ***** ***** ***** Implications of Selecting Poultry for High and Low Adrenal Response to Stress, by Keith I. Brown and Karl E. Nestor.... Effects of Stress on Rate of Gain and Feed Conversion of Ohio Low and High Adrenal Response Turkey Lines, by Keith I. Brown and Karl E. Nestor... 4 Effects of Sperm Number on Onset and Duration of Fertility in Turkeys, by Keith I. Brown Acrosomal Proteolytic Enzymes and Seminal Plasma Inhibitors: Their Implications in Semen Preservation, by B. A. Less ley and K. I. Brown Selection for Semen Yield in the Turkey, by Karl E. Nestor The Influence of Genetic Changes in Several Traits of Economic Importance on Semen Production in the Turkey, by Karl E. Nestor Effects of Genetic Changes in Egg Production Traits, Growth Rate, and Response to Cold Stress on Reproduction in the Turkey, by Karl E. Nestor Influence of the Initial Stimulatory Lighting Program on Fighting and Semen Production of Turkey Males, by K. E. Nestor, Y. M. Saif, and P. A. Renner Intermittent Lighting for the Production of Turkey Hatching Eggs, by Karl E, Nestor and Wayne L. Bacon Attempts to Reduce Broodiness of Turkey Hens by Special Light Treatments, by Karl E. Nestor and Wayne Bacon The Genetic Association of Growth and Egg Production in the Turkey, by Karl E. Nestor Yolk Production in Turkeys Differing in Egg Production Level and Growth Rate, by K. E. Nestor, W. L. Bacon, P. A. Renner, and D. A. Ehlhardt Influence of Strain and Feed Restriction on Egg Production, Egg Quality, and Hatchability of Cage Hens, by K. I. Brown, K. E. Nestor, S. P. Touchburn, and D. A. Ehlhardt., Egg Shell Quality According to Clutch Size and Position in Clutch and Its Relationship to Hatchability, by Karl E. Nestor and D. A. Ehlhardt Association of Certain Body Measurements on Live Birds with the Amount of Leg Muscle in the Turkey, by K. E. Nestor, P. A. Renner, and D. A. Ehlhardt Influence of Plumage Color Pattern Genes on Growth and Reproduction in the Turkey, by Karl E. Nestor Inheritance of Livability During the Early Growing Period of Turkeys, by K. E. Nestor, P. A. Renner, and K. I. Brown Influence of Genetic Changes in Several Traits on Poult Mortality During the Early Growing Period, by K. E. Nestor, K. I. Brown, and P. A. Renner The Influence of Dipping Eggs in Gentamicin on Growth and Leg Weakness of Turkeys, by Karl E. Nestor and Y. M. Saif

3 Changes in Free Fatty Acid and Neutral Lipid Concentrations with Reproductive State in Turkey Hens, by Wayne L. Bacon and Margery A, Musser 90 Estrogens and Neutral Lipids in Turkey Hens at Different Reproductive States, by Wayne L. Bacon, Keith L Brown, Margery A. Musser, and David W. Long 93 Calcium and Phosphorus Concentrations in the Plasma of Turkey Hens at Various Reproductive States, by Wayne Bacon and Margery Musser 97 Variation in Plasma Calcium During Egg Formation in the Turkey, by Margery A. Musser, Wayne L. Bacon, and Sherman P. Touchburn 101 Withdrawal of Plasma Calcium by the Shell Gland of the Turkey, by Margery A. Musser, Wayne L. Bacon, Sherman P. Touchburn, and Karl E. Nestor Calcium-Binding Protein in the Turkey Hen, by Margery A. Musser, Karl E. Nestor, and Wayne L. Bacon 106 A Study of Normal and Abnormal Shell Formers Among Caged Breeders, by Margery Musser, Wayne Bacon, and David Long 109 Selenium Supplements for Turkey Rations, by Austin H. Cantor Cage Confinement for Growing, Finishing, and Breeding Turkeys, by Austin H. Cantor, Keith I. Brown, and Karl E. Nestor 113 Predator Control on the OARDC Turkey Ranges, by Philip A. Renner and Karl E. Nestor Effects of Relative Humidity on Aspergillus Infection in Turkey Poults, by Y. M. Saif. 117 Aeromonas and Salmonella Infections in Turkey Poults, by Y. M. Saif and W. F. Busch Isolation and Characterization of the Turkey Immunoglobulins M and G, by Y, M. Saif and J. E. Dohms 121 Observations on the Serologic Response of the Turkey to Mycoplasma me1eagrfdjs_ Antigens, by Y. M. Saif and J. E. Dohms 123 Comparative Protein-Amino Acid Level Response of Broad White and Medium White Turkeys, by Paul E. Waibel and B. R. Behrends 126 Influence of Ambient Environment and Nutrition on Development of Growing Turkeys and Their Subsequent Reproductive Performance, by M. E. El Halawani, P. E. Waibel, W. H. Burke, and R. N. Shoffner 134 Comparative Performance of Turkey Breeder Hens in Cages and Floor Pens, by P. E. Waibel, M. E. El Halawani, and W. H. Burke 139 In 1974, the Ohio Agricultural Research and Development Center and the University of Minnesota entered into an informal agreement to hold Turkey Days in alternate years. Each institution agreed to provide speakers for the other's Turkey Days. The articles on pages 126, 134, and 139 are a result of this agreement. Ill M

4 IMPLICATIONS OF SELECTING POULTRY FOR HIGH AND LOW ADRENAL RESPONSE TO STRESS Keith I. Brown and Karl E, Nestor Department of Poultry Science Some years ago it was postulated that animals (turkeys) which can adapt to stress conditions without initiating the General Adaptation Syndrome (G.A.S.) or at lease respond to a low release of adaptation should grow faster, convert feed more efficiently, reproduce more efficiently, and be more resistant to a wide variety of stress conditions (Brown, World's Poultry Sci. J., 15: , 1959). To test the above postulate, two lines were developed from the Ohio randombred control (RBC) population on the basis of their plasma corticosterone level after cold stress. One line (HL) was selected for increased plasma corticosterone level after cold stress, while the other line (LL) was selected for low corticosterone level. A subline of the control population was also maintained separately as a randombred control (RBC) through six generations of selection. All offspring of all three lines from four weekly hatches were cold stressed at 4 weeks of age. At the end of 4 hours in a 4 C. room, 5 ml. of heparinized blood were drawn from each bird by cardiac puncture and analyzed for plasma corticosterone. The plasma corticosterone levels by year and by line are shown in Table 1. By the sixth ^eneration^ the mean corticosterone levels after cold stress of the HL were nearly double that of the LL. Measurement of CorrelatedResponses Space will not allow detailed presentation of all data. However, as predicted, the LL birds were significantly heavier at 4 weeks of age in all generations and by the seventh generation they were also significantly heavier at 16 and 24 weeks of age (Brown and Nestor, Poultry Sci., 53: , 1974). Similarly, the LL hens produced significantly more eggs after four generations of selection than either the RBC or HL birds. When significant differences in fertility and hatchability occurred between lines, the LL was superior to the HL. This, coupled with significantly higher egg production in the LL, resulted in a highly significant difference in poults per hen. Mortality to 4 weeks of age for generations 3 through 9 is shown in Table 2. Both natural mortality and mortality due to stress and blood collection were significantly lower in the LL than in the RBC and the HL for all seven generations. Because response to stress and the G.A.S, are thought to be nonspecific, it was decided to measure the response of the selected lines to heat stress. After 4 hours of heat stress, the plasma corticosterone levels were 6.9, 8.5, and 11.6 ug% for the LL, RBC, and HL, respectively. Implications The similar adrenal response (via corticosterone release) to heat and cold stress demonstrates that this is a nonspecific systemic response to stress. This would indicate that other agents which are potential stressors, such as handling, food or water deprivation, debeaking, etc., would elicit the same adrenal response. All these have been documented in scientific literature as stressors. Similarly, agents which might be termed psychological (high population densities, low social ranking, introducing birds to an existing social group, and keeping birds in groups rather than

5

6 single isolation] are all proven strcssors in chickens. In addition, the data presented in labie 1 convincingly show that it is possible to rapidly select lines which exhibit a high and low adrenal response to stress. A legitimate question, then, is what are the implications of this finding to avian researchers and to the poultry industry? In other words, what are the consequences of adrenal hyperactivity, or conversely, what are the benefits of avoiding adrenal hyperactivity? Recent studies indicate that epinephrine secretion is controlled by cortlcosterone. If this is true, then selecting birds for low adrenal response may indeed by extremely beneficial. For example, elevated epinephrine levels have been shown to result in loss of appetite, suspension of egg laying, high blood pressure, increased heart rate, etc. It has been shown that the posterior aorta of hypertensive turkeys have greater epinephrine concentrations than low blood pressure ones. Thus the relation of aortic rupture to hyperadrenal activity may become an important field of study. In addition, it can be postulated that along with faster growth, the LL birds should convert feed better than HL birds. A careful study now pnderway indicates that LL birds do exhibit a lower gain/feed ratio than HL birds. This difference is accentuated when the birds are alternately subjected to cold and hot environments. A review of the literature presents no clearcut picture of the effects of stress on disease susceptibility. However, it is clear that stressed birds are certainly more susceptible to certain diseases (particularly viral diseases), although they are not more susceptible to all (bacterial) diseases. Practical Applications for the Poultry Industry The increased growth rate, reduced excitability, superior reproductive performance, and increased resistance to viral diseases make selection of birds with a low adrenal response to stress attractive. On the other hand, the geneticist can not afford to ignore body conformation and dressing percent. The authors suggest that a selection index placing about equal weight on plasma cortlcosterone levels following stress and on body conformation should be worth trying in turkeys and broiler chickens. Currently, the Ohio LL turkey which is a medium white turkey arid already has a low adrenal response is being maintained in cages and is being selected for improved body conformation. Just enough selection will be made for corticosterone levels after stress to maintain the current low levels. Summary The implications of turkey lines selected for high and low adrenal response to stress were discussed. There is increased growth rate, reduced excitability, superior reproductive performance, and increased resistance to viral diseases in the low line. The converse is true of the high line. Based on these data, the Ohio low line turkey is being maintained in cages and is being selected for increased breast width at a constant body weight when hens are placed in cages. Offspring are selected only from the high and persistent egg producers. A line thus selected from the Ohio low line could be an excellent medium turkey. Genetic breakthroughs of the future may come from selection for appropriate physiological traits as demonstrated in this study.

7 EFFECTS OF STRESS ON RATE OF GAIN AND FEED CONVERSION OF OHIO LOW AND HIGH ADRENAL RESPONSE TURKEY LINES Keith I. Brown and Karl E. Nestor Department of Poultry Science It has been well documented that glucocorticoids have a catabolic effect (Brown et al., Am. J. Physiol., 192: 43-50, 1958). Eight-week-old chickens injected with cortisone exhibit a negative nitrogen balance and lose weight, whereas controls remain in a positive nitrogen balance and continue to gain weight. For this reason, it is postulated that turkeys selected for a high response to stress (HL) should not only grow more slowly, but should also convert feed less efficiently than turkeys selected for low response to stress (LL) (p.l). To test the above hypothesis, 18 HL hens and 18 LL hens were grown to 4 weeks of age in a Petersime battery under normal brooding conditions. At 4 weeks, the birds from each line were distributed into cages (two birds per cage) and placed in an environment chamber. From the 4th to the 5th week the chamber was maintained at 75 F. and 50% relative humidity (7 days). From the beginning of the 5th week, the chamber was maintained at 32 C. (90 F.) for 4 days and 4 C. (39 F.) for 3 days each week. The experi- The birds were weighed and the feed consumed was measured each week. ment was terminated at the end of 11 weeks of age. The weekly gains and feed conversion for both lines are shown in Table 1 and Figure 1. The weights and weight gained each week were significantly (P<0.01) higher for the LL than for the HL. As predicted, the feed/gain ratio for the entire period studied was significantly lower for the LL (2.71) than for the HL (3.05). Thus it has been convincingly demonstrated that under stress conditions LL turkeys grow faster and convert feed more efficiently than HL turkeys. This study will be repeated in 1975 with the addition of a control group for each line being maintained under normal growing conditions. It is probable the differences between the lines will be smaller under normal growing conditions. Summary Turkeys selected for high adrenal response (HL) to stress and low adrenal response (LL) to stress were placed in an environment chamber. They were subjected to 90 F. 4 days and to 39 F. 3 days each week from 5 weeks through 11 weeks of age. The LL turkeys were significantly heavier and gained significantly more weight each week. The feed per gain ratio was significantly lower (P<0.05) for the LL than for the HL.

8 TABLE 1.--Effects of Stress (Alternate High Temperatures and Low Temperatures) on Rate of Gain and Feed Conversion of Ohio Low and High Adrenal Response Lines (Hens). Age, Weeks Low Adrenal Response Line (LL) Live Weight Feed Required Gain for for per Period Period Gain Wt. (gin.) High Live Wt. (gm.) Adrenal Response Line (HL) Weight Feed Required Gain for for per Period Period Gain 4 528** ** ** ** ** ** ** ** wk. period 2169** * *Significantly different from HL (P<0.05), **Significantly different from HL (P<0.01), -5-

9

10 EFFECTS OF SPERM NUMBER ON ONSET AND DURATION OF FERTILITY IN TURKEYS Keith I. Brown Department of Poultry Science Observations on the relations of frequency of insemination by artificial means and of semen dosage have been made by a number of researchers. However,, the most critical information on the pattern of fertility subsequent to a single insemination was reported by Loreriz in 1950 (Poultry Sci. 3 29: 20-26). He performed his studies using Broad Breasted Bronze virgin hens. It was decided to repeat this type of study on large white hens after they had been in egg production 16 weeks. Late in the reproductive season, poorer fertility and shorter duration of fertility would be expected. In addition, the modern turkey is a far different bird than the 1950 Bronze hen. The hens used in this study represent the six large white strains Dr. Nestor is selecting. They are RBC, 16 week, I, C, B, and F as described (p.46) in this issue. At 16 weeks of egg production and 7 weeks following the last insemination, the hens were given a single insemination. The treatments were 0.02 ml., 0.01 ml., and ml. of undiluted semen, and 0.06 ml., 0.03 ml., and ml. of semen diluted threefold in Ohio IV extender (Table 1). Thus approximately 140 million, 70 million, or 35 million sperm were inseminated both as undiluted and diluted semen. TABLE 1.~0h1o Extender IV.* Chemical Grams/Liter Glucose Inositol Citric Acid Sodium Salt:2H Citric Acid CaCl MgCl 2 :H Na Glutamate Tes Raffinose Gentamycin *Add 600 ml. H 2 0. Titrate to ph 7.4 with equal volumes of KOH and NaOH (650 MQsm). Adjust O.P. to 325 MOsm by the addition of distilled H 2 0. Pass through a millipore filter and package in a sterile serum vial.

11

12 A pooled sample of semen was collected and the six treatments were made up from the single pool. Two pens of hens (all six strains distributed in each pen) were inseminated with each treatment. All inseminations were completed within 1 hour of semen collection. The hens were trapnested and the eggs were pedigreed and dated. Eggs were set daily the day following their production. All eggs were candled at 7 days, and eggs removed from the tray were examined macroscopically to determine the very early dead germs. Typical fertility patterns after insemination are illustrated in Figures 1 and 2. Failures are omitted because at 16 weeks a number of the hens were broody,, going broody, or coming out of a broody period. Thus it is not surprising that a number of hens laid few or no fertile eggs following a single insemination. Although it is obvious that when ml. (35 million) sperm were inseminated, the average duration of fertility was much shorter than when 0.02 ml. (140 million) sperm were inseminated. Two hens inseminated with 35 million sperm laid fertile eggs for 40 days or more. This suggests that hens might be selected for this trait. All data for this experiment are summarized in Table 2 by average fertility for 5-day intervals. When 140 million sperm were inseminated via extended semen (i.e., 0.06 ml. of semen diluted 1:2), the fertility was less than when the same number of sperm were inseminated via 0.02 ml. of undiluted semen. Since the other treatments o diluted semen were not different from undiluted semen, it is concluded that the larger volume results in mechanical losses of semen from the vagina during or following the insemination process. It is interesting to note that 0.03 ml. of diluted semen gave fertility results not significantly different from 0.02 ml. of undiluted semen. TABLE 2. --Effects of Number of Sperm Inseminated on Fertility (Omitting Failures). Days After Insemination Undiluted Semen 0.02 ml ml. 140 million 70 million Percent Fertility ml. 35 million 0.06 ml. 140 mil lion Diluted Semen 0.03 ml. 70 million ml. 35 mil 1 ion (- 1.4)* 78.9 (-11.4) (- 7.4) 71.7 (-10.1) (-10.3) 81.1 (-10.3) (- 1.7) 70.2 (- 8.6) (- 8.9) 67.6 (-12.4) (+ 2.1) 51.2 (-23.8) (-24.0) 41.0 (-44.7) (- 0.3) 52.5 (-20.8) (- 6.2) 31.0 (-29.5) (+22.5) 35.0 (-22.5) (-18.6) 35.0 (-40.8) (+13.0) 47.1 (+ 2.4) (-18.6) 22.7 (-24.5) (+ 4.9) 28.6 (- 3.5) (-19.9) 20.0 (-20.5) (+ 9.9) 8.6 (- 3.2) (-13.5) 14.3 (- 6.2) (+ 0.6) 11.1 (+ 2.0) ( + 3.3) 00.0 (- 3.3) (+ 4.0) 3.0 (+ 3.0) Means'" 58. 8a 47.5 be 38. 2c 46. 4bc 51.2ab 37. 9c *( ) difference from 140 million sperm/insemination. "Means followed by different letters are significantly different.

13

14 The fertility patterns obtained following a single dose of undiluted semen of 140 million, 70 million,, and 30 million sperm are shown in Figure 3. Up to approximately 14 days after Insemination, there were no significant differences between the treatments. However, from 14 days on, the fertility differences became highly significant, the persistence of fertility being greatly reduced when 35 million sperm were inseminated. These data would Indicate that, even 16 weeks into the reproductive season, inseminations at 2-week intervals should result in acceptable fertility. However, it must be emphasized that failures were omitted from these data and most of the failures (approximately 25 o of the hensj were hens either broody, going broody, or just coming out of a broody period at the time the inseminations were made. Therefore, if an insemination was a failure, that particular hen would in effect not be inseminated for 4 weeks. Many studies substantiate that 14-day inseminations later in the season when broodiness becomes a problem are not sufficient. Another study was conducted in which 144 large white hens maintained in cages were inseminated with six different treatments (24 hens) per treatment. The treatments were: Treatment Insemination Schedule 12 Days After Lighting to 14-HourDay 1 Undiluted 0.01 ml. 14, 17, 21, 28, 35, 42, 40 2 Undiluted 0.01 ml. 14, 17, 21, 35, 49, 63, 75 3 Undiluted 0.01 ml. 14, 17, 21, 42, 63, Diluted (1:2) 0.03 ml. 14, 17, 21, 28, 35, 42, 49 5 Diluted (1:2) 0.03 ml. 14, 17, 21, 35, 49, 63, 75 6 Diluted (1:2) 0.03 ml. 14, 17, 21, 42, 63, 84 - Thus it is seen that all hens were inseminated three times in 7 days with either 0.01 ml. of neat semen or 0.03 ml. of semen diluted 1:2 in Ohio Extender IV. Thereafter, the insemination intervals became 7, 14, or 21 days. The fertility results are shown in Table 3. When all hens were inseminated three times within 7 days at the beginning of egg production, there were no significant differences in fertility between subsequent insemination intervals of 7 and 14 days to 16 weeks of egg production. However, 21-day insemination intervals resulted in a significant reduction in fertility. The semen extended in Ohio Extender IV yielded fertility results not significantly different from undiluted semen. The effects of Insemination interval and extender on hatch of fertile eggs are shown in Table 4. Undiluted semen inseminated at 7-day intervals resulted in significantly lower hatch of fertile eggs than diluted semen inseminated at 7 and 14-day intervals. It is postulated that undiluted semen without gentainycin probably resulted in a high percentage of hens being infected with Mycoplasma meleagridis, Since the eggs were not dipped in this study, this may account for the lower hatchability in this treatment. It is interesting to note that diluted semen containing gentamycin inseminated at 7 and 14-day intervals tended to improve hatchability late in the reproductive season. This was not true of the 21-day insemination interval. In the latter instance, the effect of older sperm in the sperm glands fertilizing eggs may account for lower hatchability as the season progressed. The possible beneficial effect of gentamycin on hatchability needs further study. -11-

15

16 Summary It appears that if hens are inseminated three times with 70 million sperm each time at the onset of egg production and at 14-day intervals with 70 million sperm each time thereafter, adequate fertility can be maintained. Further, the use of an extender such as Ohio IV containing gentamycin may be beneficial to hatchability of fertile eggs and does not depress fertility significantly. However, the use of the extender will not improve fertility and at this time should be thought of only as another available tool which may help reduce the spread of disease from semen to the hen. Because the use of an extender reduces the number of sperm per insemination, it should be used only by well-trained insemination crews. Poor insemination crews will not obtain maximum fertility with either an extender or a 14-day insemination interval, Good insemination crews can probably use both.

17 ACROSOMAL PROTEOLYTIC ENZYMES AND SEMINAL PLASMA INHIBITORS: THEIR IMPLICATIONS IN SEMEN PRESERVATION B. A, Lessley and K. I. Brown Department of Poultry Science For fertilization to occur, spermatozoa must be able to perform two separate functions. First, the sperm cells must have the motility necessary to transport the genetic material they contain to the site of fertilization. Second, once in the vicinity of the ovum, the spermatozoa must penetrate the membranous barriers surrounding the yolk prior to fusion with the blastodisc. Although the motility of spermatozoa has been studied for nuny years, interest in the mechanisms involved in the penetration of the ovum is relatively new. Recent research in mammals has indicated that a portion of the spermatozoa known as the acrosomc is responsible for entiy into the ovum. The acrosomc is a cap-like structure located at the anterior end of the sperm head which apparently serves as a storage site for enzymes capable of digesting a pathway through the membranes surrounding the ovum. Normally these acrosomal enzymes are stored in an inactive state until the sperm comes in contact with the ovum. At this time, specific changes occur in the structure of the acrosome, allowing release or exposure of the enzyme activity. Although several types of enzymatic activity have been identified in the acrosome, proteolytic enzymes capable of breaking down the primarily protein-containing structure of the ovum membranes are probably most important in the penetration process. The most abundant and thoroughly characterized acrosomal proteolytic enzyme is the trypsin-like enzyme (TLE). This particular enzyme is present in especially large quantities in avian spermatozoa The main reason for initiating a study of avian acrosomal enzymes was the possibility that the resulting information would be helpful in resolving certain problems involved in low temperature preservation of avian semen. Evidence from studies of spermatozoa with the electron microscope indicate that the acrosome is severely damaged during the freeze-thaw process in many species Since the amount of freeze damage to the acrosome sustained by spermatozoa of a particular species seems to be related to the Level of fertility which can be expected from inseminations of frozen semen from that species, destruction of the acrosome during freezing may be one of the major causes of poor fertility. The basic hypothesis is that damage to the acrosome may result in the premature release of the acrosomal enzymes, rendering the spermatozoa incapable of penetrating the ovum even if they retain enough motility to reach the site of fertilization. The vast majority of the research on acrosomal enzymes has been performed on mammals, and it was not known how many of the previously mentioned research results and hypotheses were applicable to avian species. In order to investigate similar areas of interest in avian species, a series of studies was conducted using chickens as the experimental animals. The first experiment undertaken was designed to determine whether acrosomal enzymes were released from chicken spermatozoa after freezing. Undiluted semen was frozen without any cryophylactic., thawed, and allowed to incubate at various temperatures for periods up to 3 hours. The trypsin-like enzyme activity was used as an indication of acrosomal damage. Normally the TLE activity in the seminal plasma is very low and any increase in activity was interpreted as release from damaged aerosomes. The results of this experiment are summarized in Table

18 TABLE 1.--Trypsin-like Enzyme Activity* in Seminal Plasma at Various Times After Thawing. Minutes After Thawing Treatment Not frozen - incubated at 40 C Frozen - incubated at 40 C Frozen - incubated at 20 C Frozen - incubated at 5 C ^Activity expressed as um of BAEE hydrolyzed/min./ml. of seminal plasma. Immediately after thawing, there was little difference between frozen and unfrozen samples. As time progressed, however, the amount of activity in the frozen samples rose dramatically, indicating severe acrosomal damage, while the activity in the unfrozen control sample increased only slightly. The TLE release was most rapid at 40 C., a temperature similar to that which the spermatozoa would be exposed after insemination into the hen's oviduct. Enzyme release at this temperature had essentially reached a maximum after 1 hour of incubation, possible indicating total depletion of the acrosomes. Although the rate of release was slower at lower temperatures, enzyme release was still considerable even at 5 C. Since this experiment has established that TLE activity was indeed released from freeze-damaged spermatozoa, it was decided to investigate the possibility that the amount of enzyme activity released from the acrosome could be used as a quantitative assay for cell damage. It is well established that the addition of certain cryophylactic agents to the semen will improve the survival of spermatozoa after freezing. If the amount of the TLE released is a good indication of acrosomal damage, use of such protective agents should result in reduced enzyme release after freezing. To test this hypothesis, semen was diluted fourfold in extenders containing no cryophylactic, or 1.0 molar levels of ethylene glycol, dimethyl sulfoxide, or glycerol. The samples were frozen, thawed, and incubated at 40 C. for 3 hours. A sample diluted with the basic extender, but not frozen, was provided as a control. The results of this experiment are summarized in Table 2. The pattern of release was very similar to the first experiment in that differences among treatments were minimal at the beginning of the incubation and became more pronounced at later intervals. Compared to the previous experiment, the overall levels of enzyme activity were lower because of the dilution. All samples which contained cryophylactic released significantly less TLE activity than the unprotected sample. There seemed to be no real difference in the amount of protection offered by ethylene glycol and dimethyl sulfoxide, but glycerol was superior to both. The protection offered by the cryophylactics was not complete, however, since all treatments released more TLE activity than the unfrozen control. Apparently this technique can provide an indication of the relative amount of protection offered by a particular cryophylactic, as well as differentiate between protected and unprotected samples. -15-

19 Similar experiments have been performed using TLE release to examine the effects of several other variables generally considered in the design of a freezing procedure, The same technique has also proved useful in investigating the basic mechanisms of cellular damage induced by freezing. On the basis of these results, it appears TLE release into the seminal plasma provides an excellent quantitative assay for acrosomal damage to frozen-thawed spermatozoa. Since the integrity of the acrosome is intimately related to the process of fertilization, modifications of freezing procedures which result in improved preservation of the acrosome could well lead to improvements in the fertility obtained with frozen semen. Unfortunately, at the time of this writing this assumption has not yet been tested with fertility trials. In addition to the studies concerned with acrosomal enzymes,, concurrent research has also lead to the discovery of a component of chicken seminal plasma which is capable of specifically inhibiting the proteolytic action of TLE* This inhibitor has been purified and partially characterized and is a small polypeptide with a molecular weight of about 5,500. Further investigation has revealed the presence in seminal plasma of an inactive complex consisting of TLE in combination with the seminal plasma inhibitor. The presence of the inhibitor-tle complex implies that the inhibitor serves as a protective mechanism to prevent possible damage to the spermatozoa or lining of the sperm ducts by TLE released from dead, disintegrating spermatozoa. This information may lead to a practical application of such inhibitors in the formulation of media used in the storage of spermatozoa in the frozen or liquid state. It is unrealistic to expect that any freezing procedure will result in preservation of 100% of the original number of spermatozoa, and the damaged cells will release considerable amounts of the TLE activity. The TLE release is usually sufficient to overcome the limited amount of inhibitor normally present in the seminal plasma very quickly. Although spermatozoa stored in the liquid state do not suffer the extreme TABLE 2. --Effects of Various Cryophy lactic Agents on TLE Activity* in Pur i ng 3-jjqur I ncuba 1 1 on at 40 C. Minutes After Thawing Treatments Control - not frozen 0,21 0, , Frozen with 1.0 Molar Glycerol (10,1%) ,18 Frozen with 1.0 Molar Dimethyl Sulfoxide (9,4%) Frozen with 1.0 Molar Ethylene Glycol (7A%) , Frozen without cryophylactic *Activity expressed as um of BAEE hydrolyzed/min./ml. of diluted seminal plasma. -16-

20 trauma of freezing, steady attrition through the death of senescent sperm should result in enough TLE release to seriously deplete and eventually overcome the available inhibitor. In both types of storage, the inclusion of extra inhibitor in the extender to insure that any TLE released is rendered inactive may improve the survival of the remaining viable spermatozoa. Summary It has been demonstrated that the trypsin-like enzyme is present in turkey spermatozoa and is released when the sperm are stored in liquid state and when frozen and thawed. The release of TLE after thawing is an excellent indicator of cell damage. In addition, a specific TLE inhibitor is found in avian seminal plasma. This inhibitor forms a TLE-inhibitor complex which is inactive. The presence of this inhibitor-tle complex implies that the inhibitor serves as a protective mechanism to prevent possible damage to the spermatozoa by TLE released from dead, disintegrating spermatozoa. This information may lead to a practical application of such inhibitors in the formation of media used in the storage of spermatozoa in the frozen or liquid state. -17-

21 SELECTION FOR SEMEN YIELD IN THE TURKEY Karl E s Nestor Department of Poultry Science The production by turkey males of a large quantity of high quality semen Is important to turkey hatching egg producers since almost all turkey hens are artificially inseminated. An experiment was started to study the inheritance of semen yield and to find the genetic relationship of semen yield to other traits of economic importance in the turkey. Two approaches were used* First, a strain was developed from a randombred control strain on the basis of selection only for increased semen yield and observing changes occurring in this and other traits. This aspect of the study will be discussed In this article. The other approach consisted of surveying the semen production of several lines which had been developed by selecting each for different traits of economic importance* This will be discussed in the following article. The randombrcd control line used in this experiment started from a cross of four large type commercial turkeys. The line was closed in At the present time, this line is considered medium weight since no intentional selection was practiced and theoretically, under the random breeding method used to maintain the line, there should have been little genetic change since The randombred control line was maintained along with the semen line which was selected for increased semen volume to measure any environmental differences from year to year. The birds of both the randombred control and semen lines were grown in confinement until they were 8 weeks of age. At this time, they were transferred to ranges. At 24 weeks of age in September and October, the breeder males were housed in a pole shelter and exposed to natural light conditions until Dec. 1. Fourteen hours of artificial light per day were given beginning Dec. 1. Semen measurements (volume, concentration, and bent sperm) were made the last week in January. Selection in the semen line was based on semen volume at this time. The males were selected on an individual basis, while the females were selected on the basis of their brothers' performance. The results of five generations of selection for increased semen yield in the semen line are presented in Table 1. The semen production of semen line males consistently gained each generation. The heritability and its corresponding standard error of semen yield based on response to selection was 0.35 ± 0.20, which is relatively high. The semen production of the randombred control line from which the semen line was developed was relatively poor in semen production. Five generations of selection for increased semen yield more than doubled the semen yield of the semen line. The increase in semen yield in the semen line did not result in any consistent change in sperm concentration or in the frequency of bent sperm. Semen measurements were also made on selected males of the semen and control lines during March and June in the first four generations of selection of the semen line. This was done to see if the increased yield of semen line males was maintained throughout the reproductive period. For purpose of analysis, only males which produced semen at all three collections were included. The results are presented in Table 2. Although the semen yield of semen line males was greater than that of randombred control line males in March and June, the magnitude of the difference was not as great as in January.

22 It is interesting to note that for the January measurements, selected males of the semen line had consistently greater sperm concentration than the randombred control line. Since this was not true for nonselected males (Table 1), the effect is not genetic and must be the result of environmental Influences. There was no consistent influence in the March or June measurements. There were no consistent trends in percent bent sperm for January, March^ or June measurements. TABLE 1. The Influence of Selection for Increased Semen Yield on Semen Traits Measured in January. Generation of Selection Semen Yield (cc.) Control Semen* Sperm Concentration (billions/cc.) Control Semen* Bent Sperm (*) Control Semen* Expressed as deviation from the randombred control line, TABLE 2. Semen Production of Selected Males of the Semen and Randombred Control Lines in January, March, and June. Line Generation of Selection Semen Yield (cc.) Jan. March June Sperm Concentration (billions/cc. ) Jan. March June Bent Sperm (X) Jan. March June Control Semen* Expressed as deviation from the randombred control -19- line.

23 Body weight was measured on the two lines at 8, 16, and 24 weeks of age and at sexual maturity (when the females first attained an egg production level of approximately 50%). The average values are in Table 5. Body weight of the semen Ijne increased greatly during the first generation of selection for increased semen yield, but this gain was lost in the next few generations. After six generations of selection, there was little difference between the two lines in body weight during the growing period. Body weight of semen line females was reduced at sexual maturity. Egg production of semen line females was greatly increased during the first generation of selection and this gain was generally maintained during the six generations of selection (Table 4), In the fifth generation of selection, egg production of the control line was slightly higher than normal, while that of the semen line females was depressed, resulting in a large negative line difference. Since this was the only generation in which this occurred, it is possible that an environmental factor influenced the semen line hens adversely while not influencing the randombrcd control line hens, With the exception of generation five, the number of broody periods and total days lost from broodiness was reduced in the semen line, with little influence of average length of the broody period (Table 4). For purpose of analysis, a broody period was considered as a period of nonproduction for 5 or more consecutive days. TABLE 3, Body Weight (Lb.) of Semen and Control Lines. Line Generation of Selection 8 Wk. Males 16 Wk. 24 Wk. At Sexual Maturity 8 Wk. Females 16 Wk. 24 Wk. At Sexual Maturity Control Semen* Expressed as deviation from randombred control line. -20-

24

25 The effective days in the period were obtained by taking the total days in the laying period (180) and subtracting days at the end of this period, whenever there was a pause in production of 5 or more days immediately prior to the end of the 180-day production period. There was no consistent influence of selection for increased semen yield on this measure. Intensity of lay as measured by average clutch length, maximum length of clutch, and rate of lay increased in the semen line (Table 4). A clutch is the number of eggs produced on consecutive days before skipping a day. Rate of lay was measured by dividing the total number of eggs laid by the total days in the period (180) minus the total days broody. The results from other selected lines indicated that whenever clutch length increases, body weight at older ages decreases. The early large gain in body weight in the semen line after one generation of selection (Table 3) occurred when average clutch length was greatly reduced in the semen line (Table 4). In the following generations of selection when average clutch length was greater in the semen line, the gain in body weight of the semen line during the growing period was lost and there was a loss in body weight at sexual maturity. TABLE 5. Reproduction in Semen and Control Lines for a 12-Week Period Beginning when the Hens First Attained a Production Level of 50%. Line Generation of Selection Percent Fertility Percent Hatchability of Fertile Eggs No. of Poults/Hen, Egg wt. i (gm.) Control Semen* Stressed as deviation from the randombred control line. -22-

26 The hens of both lines were inseminated twice weekly at the beginning of the 12-week reproduction period (when the first egg was laid) and biweekly thereafter. The semen of one male was used to inseminate one female. In almost all cases, the volume of semen inseminated was greater than that recommended and the volume inseminated per bird varied from hen to hen. There was no consistent change in percent fertility, percent hatchability of fertile eggs, or number of poults produced per hen (Table 5). Egg weight of the semen line was slightly higher than that of the randombred control line but there was no trend with generations of selection. Summary Selection was effective in increasing semen yield in the turkey. After five generations of selection, semen yield in the line selected for increased semen yield was more than double that of the randombred control line from which the semen line originated. There was no consistent change in sperm concentration or percent bent sperm associated with the increased yield. Egg production, clutch length, maximum clutch size, and rate of lay were increased in the semen line. There was no consistent change in percent fertility, percent hatchability of fertile eggs, or number of poults produced per hen during a 12-week hatching period. Body weight during the growing period exhibited an initial increase in the first generation of selection in the semen line, but declined to randombred control levels in later generations. Body weight at sexual maturity decreased in later generations in the semen line. -23-

27 THE INFLUENCE OF GLNtTlU CHANbtb in TRAITS OF ECONOMIC IMPORTANCE ON SEMEN PRODUCTION IN THE TURKEY Karl E. Nestor Department of Poultry Science The influence of selecting for increased semen yield on growth rate, egg production, and reproduction was discussed in the previous article. In general, genetic increases in semen yield were accompanied by an increase in egg production, but there were no significant changes in reproduction or semen quality. This article presents information on the influence on semen production of selecting for increased egg production, increased 16-week body weight, both increased body weight and increased egg production, increased clutch length, decreased broodincss, increased fertility, and high and low levels of blood corticosterone after cold stress. Selecting foreggproduction A line (egg) was initiated in 1960 by selection for high egg production. The selection criterion in early generations (one through three) was 34-day egg production. In later generations, selection was based on egg production for a 180-day period. Semen yield, sperm concentration, and percent bent sperm were measured early in the reproduction period (January), during the middle of the reproduction period (March), and near the end of the period (May-June). Semen yield of the egg line was consistently greater than that of the randombred control from which it was developed from the eighth through the fourteenth genenerations of selection (Table 1). In most cases, the differences were statistically significant. However, there was no evidence of a time trend from the eighth generation on, indicating that the major change may have occurred between the sixth and eighth generations of selection. Sperm concentration was greater in the egg line in all comparisons, but the differences were significant in only one comparison. There was no consistent difference between lines in percent bent sperm. Selection for Increased BodyWeight Alone andincombination with Increased Egg Production Genetic increases in body weight in the 16-week line resulted in decreased semen yield and sperm concentration in later generations of selection (Table 2). A part of the decreased semen yield in the 16-week line was probably mechanical in nature. As the body weight increased in this line, the amount of tissue surrounding the pubic bones increased which made it difficult to collect semen from nonresponsive males. Selection for increased egg production in combination with increased body weight in the index line eliminated the depressing effect on semen yield observed in the 16-week line early in the reproductive period (January). Late in the reproduction period, a depression in semen yield was still evident in the seventh generation of selection in the index line. Selection for Increased Clutch Length, Decreased Broodiness, andincreased Pert11ity Since clutch length increased in the line selected for increased semen yield, semen production of a line selected only for increased clutch line was studied. There was no significant difference in semen yield or sperm concentration between the clutch line and its corresponding randombred control population during the first

28

29 TABLE 2. The Effects of Selection for Increased 16-Week Body Weight Alone (16-Week Line) and in Combination with Increased Egg Production (Index Line) on Seiren Production in the Turkey. Line Generation of Selection Yield (cc.) January Sperm Concentration (Billions/cc.) Yield (cc.) May-June Sperm Concentration (Billions/cc.) Control 16 Uk.t o L. 6 7 a ** , , Index * "^Expressed as a deviation from the randombred control line. *Difference from randombred control was significant (P<.05). **Difference from randonbred control was highly significant (P<,Q1), TABLE 3. Semen Production in Lines Selected for Increased Clutch Length, Decreased Broodiness, and Increased Fertility. Generation of Yield Line Selection (cc.) January Sperm Concentration (Billions/cc.) Yield (cc.) May- June Sperm Concentration (Billions/cc.) Control Clutch* Broody* Fertility* , *72-0, *Expressed as a deviation from the randombred control. -26-

30 TABLE 4.--The Influence of Selection for High and Low Plasma Corticosterone Level After Cold Stress on Semen Production. Line Generation of Selection Yield (cc.) January May-June Sperm Sperm Concentration Yield Concentration (Billions/cc.) (cc.) (Billions/cc.) Control Low Line ** High Line * -0.80* *P<.05. **Expressed as a deviation from the randombred control. three generations of selection (Table 3). had no significant effect on semen yield. Selection for decreased broodiness also Selection in the fertility line was based on fertility of hens over a 12-week hatching period beginning when the hens first attained an egg production level of approximately 50%. Each hen in this line was mated to a different male and offspring from dams with the highest fertility were selected. It was hoped that selection would increase the fertility of females with little change in semen production, After three generations of selection, this appears to be the case (Table 3). Selection for High and Low Blood Levels of Corticosterone After Cold Stress Two lines developed on the basis of a high (high line) and low (low line) level of plasma Corticosterone after cold stress at 4 weeks of age have exhibited changes in several traits of economic importance. Therefore it was decided to study semen production in these lines. Early in the reproductive period (January), there was no apparent difference in semen yield, but there was a tendency for the low line to have lower sperm concentration and the high line to have greater sperm concentration than the corresponding randombred control (Table 4). Late in the reproduction period, both lines tended to have both reduced yield and reduced sperm concentration. Summary. Selection for increased egg production resulted in increased semen yield and increased sperm concentration. Genetic increases in body weight resulted in decreased semen yield and sperm concentration. Selection for increased clutch length, decreased broodiness, and increased fertility had no significant effects on semen production during the first three generations of selection. -27-

31 EFFECTS OF GENETIC CHANGES IN EGG PRODUCTION TRAITS, GROWTH RATE, AND RESPONSE TO COLD STRESS ON REPRODUCTION IN THE TURKEY Karl E. Nestor Department of Poultry Science The turkey is a relatively poor reproducer. For good reproduction, there must first of all be production of high quality semen and a large number of eggs. The effects of selection for various traits on these aspects of production are examined on pages 14 and 46. Fertility and hatchability are also two important components of reproduction. This article examines the effects of genetic increases in egg production and growth rate and of genetic changes in response to cold stress on fertility and hatchability of fertile eggs, as well as the total number of poults produced during a 12-week hatching period beginning when the hens first obtained an egg production level of approximately 50%. A paired mating system was used to reproduce the lines. With this system, each female was artificially inseminated with the semen from one male. In almost all cases, the volume of semen inseminated was larger than that generally recommended so that sperm number per insemination was probably not a serious factor The hens were inseminated when the first egg was laid, followed by an insemination 1 week later. Biweekly inseminations were used throughout the remainder of the breeding season. Genetic increases in egg production in the egg line generally resulted in improved fertility in later generations of selection (Table 1). This was expected since semen production was improved in this line. There was no consistent change in TABLE l.-~the Influence of Genetic Increases in Egg Production on Fertility, Hatchability of Fertile Eggs, and Number of Poults Produced per Hen During a 12-Week Reproduction Period. Line Generation of Selection Fertility (%) Hatchability Fertile Eggs (*) Poults per Hen (No.) Control Egg* ^Expressed as a deviation from the randombred control. -28-

32 TABLE 2.--The Effects of Selection for Increased Body Weight Alone (16~Week nri a^ i9 n u f i n f i n W n th Increased E 99 Production (Index Line) on Reproduction During a 12-Week Hatching Period Line Control 16-Week* Index* Generation of Selection Fertility U) (+ 5)t - 4 (- 4) 0 (+12) + 2 (+ 7) - 6 (- 3) + 1 (+ 2) + 5 (+8) Hatchatn lity Fertile Eggs (+10) 1 " - 9 (-13) + 4 (+ 8) - 6 (- 6) -4(0) - 4 (+ 2) + 2 (+ 8) Poults per Hen (No.) i (+ 8)t (- 2) (+ 8) (+ 5) (+ 6) (+ 5) (+10) Expressed as a deviation from randombred control. f Number in parentheses is index line - 16-week line. TABLE 3.--Reproduction in Lines of Turkeys Selected for High and Low Plasma Levels of Corticosterone After Cold Stress at 4 Weeks of Age During a 12-Week Hatching Period. Line High Low* Generation of Selection Fertility (%) ^Expressed as a deviation from the high line. Hatchability Fertile Eggs (X) Poults per Hen (No.)

33 hatchability of fertile eggs in the egg line. The number of poults produced per hen, which is influenced by number of eggs produced, fertility, and hatchability of fertile eggs, was increased greatly in the egg line. Selection for increased 16-week body weight in the 16-week line resulted in decreased fertility, hatchability of fertile eggs, and number of poults produced per hen in later generations of selection (Table 2). Both semen production and egg production were reduced by selection for increased growth. Selection for increased egg production in addition to increased growth rate in the index line alleviated most of the harmful effects of genetic increases in growth rate fertility, hatchability, and number of poults produced per hen. The selection index gave three times the emphasis on 16-week body weight as on egg production. Selection for a low blood plasma corticosterone level after cold stress at 4 weeks of age generally improved reproduction in comparison to a line selected for a high level after cold stress (Table 3). The differences were not always consistent, but if a major difference occurred between the two lines, the low line was usually superior. The low line is also superior to the high line in egg production, early growth rate, and livabllity during the early growing period. Summary Genetic increases in egg production increased fertility and number of poults produced per hen during a 12-week hatching period, while genetic increases in body weight decreased fertility, hatchability of fertile eggs, and number of poults produced per hen. Selecting for both increased growth rate and increased egg production by means of a selection index resulted in little change in these traits. A line selected for a low level of an adrenal gland hormone (corticosterone) after cold stress generally had better reproduction than a line selected for a high level of the adrenal gland hormone.

34 INFLUENCE OF THE INITIAL STIMULATORY LIGHTING PROGRAM ON FIGHTING AND SEMEN PRODUCTION OF TURKEY MALES K. E. Nestor, Y. M. Saif 9 and P. A. Renner Department of Poultry Science A large loss of turkey males sometimes occurs In floor-reared turkey toms during the period when they are being stimulated into semen production by Increasing the length of the light day. This loss of males, which may be as large as 25%, is usually due to fighting. It was believed that a lighting program which would more gradually bring the males into semen production might result in less fighting and, perhaps, a smaller loss of males. Two experiments were conducted to study the influence of two lighting programs. Males from 10 lines of turkeys (six heavy and four medium weight) were divided into two groups in both experiments. In the first experiment, both groups of males were exposed to natural declining daylight conditions from housing (in mid-september to mid-october) until Oct. 30. The males of both groups were exposed to artificial light for an 8-hour period per day at this time. The intensity of light in the house fluctuated widely, depending on the brightness of the day. The natural day length on Oct. 30 was approximately 10-1/2 hours, so the 8-hour period of artificial light was less than the natural day length. However, on dark days the effective day length within the breeder house probably did not differ greatly from the 8-hour period in which males were given artificial light. The length of the light day was gradually increased on one group of males once each week. This group was given an additional hour of light per day. By Nov. 13 (second week of increasing light), the period of artificial light for this group exceeded that of natural day length. On Dec. 11 when this group was given 14 hours of light per day, the length of the light day on the other group was abruptly increased from 3 to 14 hours per day. Semen measurements were made on Jan. 29 (approximately 7 weeks after giving 14 hours of light), Feb. 26 (about 11 weeks after giving 14 hours of light), and May 8 (after about 16 weeks of semen production). Each male was given a head score at the Jan. 29 collection in order to estimate the relative amount of fighting. The score ranged from 1 (no scabs on head) to 5 (head almost completely covered with scabs). The presence or absence of crooked necks was also recorded at this time. Based on autopsies of some crooked neck males, the condition appeared to be the result of muscle damage, presumably from fighting. Gradually increasing the length of the light day reduced the mortality of all lines of heavy males by below that of the group which was given an abrupt increase in length of the light day (Table 1). The mortality of the two heaviest lines was reduced from 26.1 to 15.2%, a reduction of 10.9%. The mortality level in the medium weight strains was low and did not differ between light treatments. The severity of fighting, as Indicated by percent of males with crooked necks and head rating, was less in the group of heavy males given a gradually increasing light day (Table 1). These measures of fighting were similar and low for both groups of medium weight males. The two lighting programs had no significant effects on semen yield or total sperm yield (Table 1). The average sperm concentration at the Jan. 29 collection

35

36 TABLE 2.--Design of Experiment 2. Date Gradual Increaseof Light Length of Light Day (hours) Size of Light Bulbs (watts) Abrupt Increase of Light Length of Light Day (hours) Size of Light Bulbs (watts) Oct Nov Nov Nov Nov Dec Dec was significantly higher in the group given the gradually increasing day length. No other differences in average sperm concentration were significant. The design of Experiment 2 is given in Table 2. The increase in length of light day of the gradual increase group was similar to that of the first experiment. The major difference in this experiment was starting the males on low intense light and increasing the intensity as the length of light day increased. Since the toms were housed in a pole barn, the intensity at any one size of light bulb varied from day to day depending on the brightness of the natural daylight. The group of heavy line males given the lighting program which gradually increased both length of light day and intensity of light had higher mortality than those given an abrupt increase in light and held under relatively high intensity light (Table 3). This result was not expected since gradually increasing the length of the light day reduced mortality in Experiment 1, especially for the heavier lines. The mortality level of Experiment 2 was not as high as in Experiment 1. This along with the light intensity factor might explain the differences in results obtained. There was no larger difference in mortality between the two groups of medium weight males. The frequency of crooked neck males was reduced in the group given a gradual increase in length of light day. However, the frequency of crooked neck males was reduced more in Experiment 1. Although the head rating of heavy males was slightly less in the group given the gradually increasing light day length, the differences were not statistically significant as was the case in Experiment 1. Semen yield and sperm concentration were similar for the heavy males on both light treatments. In the case of the medium weight males, semen volume and sperm concentration were greater in the group given the gradually increasing length of light day. This did not occur in Experiment 1. It is possible that increasing both length of light day and intensity of light results in a higher level of male hormone than just gradually increasing length of light day. This higher level of male hor- -33-

37 TABLE 3.--The Influence of Gradually increasing Both Length of Light Day and Intensity of Light vs. Abruptly Increasing the Length of Light Day During the Initial Stimulation on Mortality, Frequency of Crooked Necks, and Semen Production in February, Treatment Mortality ( 0 Crooked Necks Head (") Rating Yield Sperm (cc.) Concentration 1 " Total Sperm Yield (Billions/Ejac.) Heavy Lines Abrupt Gradual Difference Medium Lines Abrupt Gradual Difference ** ** +0.31** **P<.01. tspectrophotometer units. TABLE 4. Semen Production of Crooked Neck Males vs. Males with Normal Necks in the Large Body Weight Lines. _ ^^^-im? 1 " 1 * 11160^ ^ ^^ ^L2 men^ 2 Semen Yield Sperm Yield Sperm No, (cc.) Concentration" 1 " No, (cc.) Concentration* 1 " Normal Necks Crooked Necks 77 0,292 0,157 Difference * *P<.05. fspectrophotometer units. -34-

38 mone might result in more fighting and be responsible for the difference in results obtained between the two experiments. More work needs to be done on this aspect. Most hatching egg producers discard their crooked neck males since they believe that they are poor semen producers. A summary of the semen production of normal and crooked neck males from the heavy lines is presented in Table 4. In Experiment 1, there was no significant difference in semen production between the two groups of males, while in Experiment 2 crooked necked males produced significantly less semen. There were no significant differences in sperm concentration in either experiment. These results indicate that semen produced from crooked neck males is not drastically reduced. However, if an adequate number of males are available, it would be desirable to cull the males with the most extreme crooked necks since these males struggle more than normal during semen collection. Summary Gradually increasing the length of light day from 8 to 14 hours at the rate of 1 hour per week, in comparison to abruptly increasing light from 8 to 14 hours, reduced fighting as indicated by reduced mortality, decreased frequency of crooked necks, and decreased head rating. When gradual increase in light intensity was employed along with the gradual increase in length of light day, mortality was higher and the frequency of crooked neck was lower than the group given the abrupt increase. Semen production was not greatly influenced by gradual increasing only length of light day, but both semen yield and sperm concentration were increased in medium weight males by increasing both intensity and length of light day gradually. The semen production of crooked necked males was lower than that of males with normal necks in one of two experiments. -35-

39 INTERMITTENT LIGHTING FOR THE PRODUCTION OF TURKEY HATCHING EGGS Karl E. Nestor and Wayne L. Bacon Department of Poultry Science Egg production of turkeys Is controlled by the length of the light day. Usually turkey hens are restricted to a short day length of 6 to 8 hours for a period of at least 6 to 8 weeks prior to being given stimulatory lighting of 14 or 15 hours of continuous light per day. Since energy usage is now an important consideration, two experiments were designed to test the influence of intermittent light on egg production of turkeys. The hens in both experiments were from four medium weight strains and six large weight strains used in genetic studies. All birds were reared in confinement until they were 8 weeks old. They were then transferred to ranges. At 24 weeks of age in September and October, they were housed in a windowless breeder house, The length of light day was gradually dropped until it reached 6 hours at 8 weeks prior to giving the hens a stimulatory light regime. The length of light day was held at 6 hours during the following 8 weeks. On Feb. 1 when the hens were approximately 39 weeks old, they were placed on a stimulatory light regime. The light intensity used was 5 foot candles during the restricted lighting and stimulatory lighting periods. In Experiment 1, one group of hens was given continuous light for 14 hours per day. A second group was placed in a lighting regime consisting of 1 hour of light followed by 2-1/4 hours of darkness repeated four times, followed by a 1-hour light period (1L:2 1/4D, 4X + 1L). The length of the period from the beginning of the first light period to the end of the last light period was 14 hours, followed by 10 hours of darkness. An effort was made to care for the hens of this group during the light periods. However, on some occasions, work was done during the dark period with the aid of very dim hall lights. Egg production was recorded for a 180-day period beginning when the first egg was laid. For purposes of analyses, egg production was summarized by months and for periods of production of 84, 120, and 180 days* In order to more thoroughly study the effects of intermittent lighting on egg production, several measures of intensity of lay and broodiness were estimated. The measures of intensity of lay included number of clutches, average clutch length, maximum clutch length, and rate of lay. A clutch is the number of eggs laid on consecutive days before skipping a day. Rate of lay was obtained by dividing total number of eggs produced by total days in the period of lay minus total days lost from broodiness. Broodiness measures include number and average length of broody periods, total days broody, and effective laying period. For purpose of analysis, a broody period was considered as a period of nonproduction for 5 or more consecutive days. The effective laying period was 180 days minus the period of time exceeding 5 days in which the hen did not lay at the end of the reproduction period. Reproduction data (fertility, hatchability of fertile eggs, and number of poults per hen) were obtained for a 12-week period beginning when the hens first attained an egg production level of approximately 50%. Each hen was mated to a different male by artificial insemination using a paired mating system. Since a large number of males was used for each group of hens, the males used to inseminate the group of -36-

40

41 hens, given the continuous or intermittent light, should have similar average fertilizing ability. The egg production of hens subjected to intermittent and continuous lighting programs is presented in Table 1. The egg production of the two groups of medium weight hens was essentially the same in February. After this time, the egg production of the group subjected to intermittent lighting was consistently less than that of the group given the continuous lighting program. There were approximately ten eggs per hen difference between the two groups after 180 days of production. This difference, however, was not statistically significant. The four medium weight strains all tended to react the same to the lighting treatment since there were no significant interactions between lines and treatments. There were no consistent differences in the egg production of the large weight strains when given the two lighting treatments. The average egg production based on trapnested eggs was slightly lower in the group given the intermittent lighting, but this was primarily due to the increased number of floor eggs in this group (Table 2). Both medium and large weight strain lines laid more eggs on the floor under the intermittent lighting program than on the continuous lighting program early in the laying period (Table 2). Later in the laying period, the reverse was true for the medium weight strains and the total number of floor eggs was higher (268 vs. 226) for the group with continuous light. The number of floor eggs for the large weight strain hens under intermittent lighting was almost double those produced under continuous lighting. TABLE 2.--The Influence of Intermittent Lighting on the Production of Floor Eggs, Type of Lighting Feb. March Apri 1 Intermittent Continuous 6 20 Difference Total May Number June of Floor Eggs July Medium Weight Strains Aug. 84 Days 120 Days 180 Days Intermittent 6 40 Continuous Difference Large Weight Strains Previous research indicated that average clutch length, maximum clutch length, and rate of lay were the best measures of intensity of lay. Hens from the intermittent lighting group had lower intensity of lay based on these measures than the hens in the continuous lighting group (Table 3). This was evident for both the medium weight and large weight lines, although only one difference was statistically significant. The other measure of intensity of lay, number of clutches per hen, did not -38-

42 TABLE 3.--Intensity of Lay and Broodiness of Hens Under Different Lighting Programs, Clutches No. Av. Type of per Length Lighting Hen (Eggs) Broody Periods Max. Kate NQ_ Av> Length of ' f Length (Eggs) Lay Hen (Days) Total Days Broody Effective Laying Period (Days) Intermittent Continuous Difference Intermittent Continuous Difference *P<.05. Medium Weight Strains Large Weight Strains * * TABLE 4. Influence of 12-Week Hatching Period. Lighting Program Egg Wt. (g.) Intermittent 84.4 Continuous 85.4 Difference Intermittent 90.8 Continuous 90.8 Difference 0 Lighting Program on Reproduction of Percent Fertility Medium Weight Strains Percent Hatchability Fertile Eggs Large Weight Strains Turkey Hens During a No. Poults per Hen

43

44 exhibit consistent differences. This measure is also influenced by the amount of broodiness. There was no interaction between light treatment and lines. The number of broody periods and average length of these periods were less in the group of medium weight turkeys exposed to intermittent light (Table 5). However, the total days lost from broodiness were greater in this group since more of the hens had a long broody period at the end of the laying period. This is reflected in the significantly lower effective days in period for the medium weight lines. The heavy weight strains given the intermittent light lost more days from broodiness primarily because of a greater number of broody periods. All of the strains within each group reacted similarly to the treatments since there were no significant line by treatment interactions. The effects of the lighting program on reproductive traits were inconsistent (Table 4). The intermittent lighting regime appeared to increase fertility, hatchability of fertile eggs, and number of poults produced per hen in the medium weight lines, but decrease these traits in the large weight strains. A second experiment was recently completed. This experiment involved exposing hens from the ten strains used in Experiment 1 to similar light treatment as in the first experiment, plus the following additional treatment. The hens were placed on the intermittent lighting program for the first 8 weeks of production and then the length of the light periods was increased by 15 minutes every 2 weeks from the 8th through the 18th weeks of production. After the 18th week of production, the dark periods were completely eliminated, resulting in 14 hours of continuous light per day. All hens in this experiment were grown similar to those in Experiment 1. The means for egg production are presented in Table 5. For the medium weight strains, the group of hens exposed to the intermittent lighting program throughout the laying period had greater egg production earlier in the laying period, but lower egg production later in the period than the group exposed to 14 hours of continuous light per day. The mean egg production based only on eggs laid in the trapnest averaged 2.4 eggs per hen higher in the intermittent light group for the entire period of lay. When the eggs laid on the floor were taken into consideration, the difference increased to 3.9 eggs per hen. Increasing the length of the light period in the one intermittent light group did not greatly alter the results obtained in comparison to the group given the intermittent light for the entire period. Hens from the large weight strains, which were exposed to the intermittent light for the entire laying period, did not exhibit the decline late in the laying period and were generally superior in production to those exposed to continuous light. Increasing the length of the light period in the intermittent light program did not increase egg production above that obtained with having a constant light period in the intermittent light program during the entire laying period. The number of floor eggs obtained was greater with both intermittent light groups (Table 6). Return of the hens to 14 hours of continuous light per day resulted in a reduction in the number of floor eggs obtained in comparison to control (continuous light) levels during the last 2 or 3 months of lay. The results of the two experiments were inconsistent. Intermittent lighting was detrimental to egg production of medium weight strains in Experiment 1, but slightly beneficial in Experiment 2. For the large weight strains, intermittent lighting had a slight beneficial effect in Experiment 1 and a large beneficial effect in Experiment 2. Another experiment will be conducted next year to study these incon- -41-

45 TABLE 6.--The Effects of Light Regime on Production of Floor Eggs in Experiment 2. Lighting Regime Feb. March Total Floor Eggs April May June July Aug. 84 Days Days Days Continuous Intermittent 5 21 Intermittent with Increase in Light Periods 5 29 Continuous Intermittent Intermittent with Increase in Light Periods Medium Weight Strains Large Weight Strains TABLE 7. --Yolk Production Based on Autopsy Data After 180 Days Body Lighting No. Wt. Regime Hens (Lb.) Continuous Intermittent Intermittent with Increase in Length of Light Period 22 Continuous 28 Intermittent 31 Intermittent with Increase in Length of Light Period 18 of Egg Production* Hens Hens Hens Yolks in with with Yolk with Yolks Raj^JDeyelgpment Atretic in Body Developing Total Wt. Follicle Cavity as Pairs No. /Hen (g.) (%) (%) Medium Weight Strains Large Weight Strains (%}

46 sistencies. floor eggs. In both experiments, the intermittent lighting program resulted in more A sample o hens laying after 180 days of egg production were sacrificed and surveyed for yolk production in Experiment 2. The results are in Table 7. Body weight of hens exposed to intermittent light was larger than those given continuous light for both medium and large weight strains. The number of yolks observed in rapid development (0.6 g. and up) in the ovary was slightly less for the groups given intermittent lighting throughout the laying period. Summary The reproductive efficiency of turkey hens exposed to either 14 hours of continuous light or five intermittent 1-hour periods of light during the same 14-hour period was compared in medium and heavy weight strains. Intensity during the light periods was 5 foot candles. In medium weight strains, a decreased rate of egg production was observed with the intermittent light in one experiment while an increased rate was observed in a second experiment. In heavy weight strains, intermittent light was slightly superior in the first experiment, but was greatly beneficial in the second experiment. A greater number of floor eggs was associated with the intermittent light regimen in both experiments. Body weight of the hens under intermittent light was also slightly increased with the intermittent light regimen..43.

47 ATTEMPTS TO REDUCE BROODINESS OF TURKEY HENS BY SPECIAL LIGHT TREATMENTS Karl E. Nestor and Wayne Bacon Department of Poultry Science Several methods have been used to reduce the amount of broodiness in turkey hens. Many of these methods are successful, but time consuming. It was believed that if it was possible to treat only broody hens with an additional light period (beyond the normal 14-hour light day) without removing the hens from their laying pens, a successful broody hen management system with reduced labor would be the result. Four similar pens containing 13 hens carrying various plumage color genes were used in this study. The genetic makeup of each pen was similar. Two pens were given the routinely used broody hen management system. This system consisted of palpating the hens just before the lights went out on Monday, Wednesday, and Friday afternoons. Any hen in the nest which did not have a hard-shelied egg in her uterus or had not laid that day was considered broody and removed to the broody pen, which had a wire floor. While in the broody pen, the hens were exposed to 24 hours of high intensity light. Hens removed from the laying pen on Monday afternoon were returned to the laying pen on Wednesday morning, those removed on Wednesday afternoon were returned on Friday morning, and those removed on Friday afternoon were returned on Monday morning. A broody hen was placed in the treatment pen only two consecutive times. This prevented hens from laying a large number of unidentifiable eggs while in the broody pen. After skipping, if a hen which had been treated for two consecutive times still showed signs of being broody, the process was repeated. The remaining two pens were given the following treatment. A special nest was designed so that when the lights went off in the laying pens, a solid metal, lighttight door dropped down in front of the nests. When this door dropped, small lights inside the nests came on, exposing hens on the nests to 24 hours of light per day. A record was made of the hens which were closed off into these nests each day. The only hens removed to the broody pen were hens which would not JbreaJc during the weekly insemination. Egg production, the incidence and duration of broodiness, and total days broody were obtained for a period of 70 days beginning with the production of the first egg. For purpose of analysis, a broody period was considered a pause in production for 5 or more consecutive days. TABLE 1. The Influence of Special Light Treatment on Egg Production and Broodiness During a 70-Day Production Period. Group Days From Lighting to First Egg Egg Production (No. /Hen) Number of Broody Periods Av. Length of Broody Period (Days) Total Days Broody per Hen Control Treated

48 The hens given the special light treatment had lower egg production and a greater number of days lost from broodiness than the control group (Table 1). The average number of broody periods per hen was similar for the two groups of hens. It was readily apparent from examining egg production records of the treated hens that the special light treatment was not effective in breaking up broody hens, even though several special light treatments were given. In the treated groups, the average number of times a hen was in the nest was 15,3, with a range from 0 to 38. The lack of response to the special light treatment is believed to be due to a lack of change in environment other than light. Previous research (Nestor, Bacon, and Renner, 1971, Poultry Sci., 50: ) has shown that moving hens from the floor to cages in the same pen with the same lighting program was almost as effective in reducing broodiness as moving hens to a different building and using various light treatments. Summary Exposing hens which went into specially designed nests shortly before the pen lights went off in the afternoon, providing 24 hours of light while in the nest, was not effective in controlling broodiness. -45-

49 THE GENETIC ASSOCIATION OF GROWTH AND EGG PRODUCTION IN THE TURKEY Karl E. Nestor Department of Poultry Science Introduction Turkey breeders have made great strides in Improving growth rate and conformation of the modern turkey. A study at the Ohio Agricultural Research and Development Center (Nestor, McCartney, and Bachev, 1969; Poultry Science, 58: ) indicated that commercial turkey breeders had gained totals of 6.8 and 4.1 Ib., respectively, for 24-week body weights of large white males and females for the 10-year period from 1957 through Similar gains in 16-week body weights were 4 and 5 Ib., respectively. Gains attributable to improvement in the environment (feeding, management, etc.) amounted to 2.8 and 1.1 Ib, for 24-week body weights and 2.0 and 1.0 Ib. for 15-week body weights, respectively, for males and females during this period* Thus, the genetic improvement was greater than the environmental improvement during the period from 1957 through The modern medium weight white turkey was characterized by growth rate similar to the 1957 large white turkey until 16 weeks of age, after which time the growth rate of the modern medium weight white turkey declined relative to the 1957 large white turkey. The amount of white meat (breast) was also increased by selection between 1957 and The increase was greater for the modern medium white than for the large white after the influence of body weight was removed. Very little environmental variance in breast width was noted. Reproduction in the large white turkey was improved slightly between 1957 and The number of poults produced during a 12-week hatching period was increased as the result of selection by a total of 1*8 per hen during this 10-year period. The number of eggs produced per hen was increased at the rate of 1.6 eggs per hen per year for a 180-day production period, but both fertility and hatchability of fertile eggs were slightly reduced. From the results o the above study, it is apparent that the major emphasis of the large white turkey breeders has been on improving meat production with little regard for egg production. The modern turkey still remains a poor reproducer, which results in high poult costs. Egg production and number of poults produced per hen are closely related and anything which improves egg production will usually result in a greater number of poults being produced. There are several alternatives which turkey breeders might use to improve egg production and number of poults produced per hen. Among these are: 1) selection for increased egg production alone; 2) selection for increased egg production for a few generations and then selecting for increased growth rate (tandem selection); 5) selecting a female line for increased egg production and a male line for increased growth and crossing the two lines to obtain a desirable type offspring; 4) selecting for increased growth rate and increased egg production simultaneously; and 5) selecting for a component of egg production which is highly correlated with the total number of eggs produced. Genetic studies at OARDC have been devoted to comparing the different methods of improving reproduction of the turkey. In these studies, randombred control popu- -46-

50 lations were maintained without conscious selection. These were used to remove the influence of environmental variation from year to year on the results obtained. Theoretically, these control populations do not change genetically from year to year and any yearly differences observed are the result of environmental differences. Most of the values presented in this paper are expressed as the deviation from the corresponding randombred control from which the selected line was developed. This method should remove most of the influence of the environmental variation. Rejection for Egg Production Alone Dr. M. McCartney initiated a line in 1960 by selecting offspring from the best dams on the basis of the number of eggs produced for 84 days beginning with the first egg. Later (1964 to present), the selection was based on egg production for 180 days. Since egg production is influenced by both broodiness and intensity of lay, these components of total egg production were measured. Based on data reported in 1972 (OARDC Research Summary 64), total days lost from broodiness was selected as the best measure of broodiness, while average clutch length was selected as the best measure of intensity of lay. TABLE 1.--Results of 13 Generations of Selection for Increased Egg Production, Year Generation of Selection Egg Production (No. /Hen) Days Days Av. Clutch Length (No. of Eggs) Days Days Total Da^s Broody Days Days Body Weight Females (Lb.) Weeks Weeks ""Expressed as a deviation from corresponding randombred control -47-

51 The results of 13 generations of selection for increased egg production are presented in Table 1. The egg production of the egg line increased through about the seventh generation of selection. The apparent large increase in 1965 was primarily due to the egg production for the randombred control being abnormally low in this particular year. From the seventh through the ninth generations, the egg line lost egg production relative to the randombred control. In the environment was drastically changed. All lines were moved from a conventional turkey laying house to a windowless house and the system of management of broody hens was drastically changed. The broody hen management system used in 1966 and thereafter was more efficient than the system used previously. Since broodiness was greatly reduced by selection in the egg line, this change affected the egg production of the randombred control line more than it did the egg line, thus resulting in the loss of egg production observed for this line. In 1970 and later, no broody hen management system was used for the egg line hens, although the randombred control population was treated as before. This change in management resulted in a five and ten-egg decline, respectively, for egg production of the egg line for 84 and 180 days. However, three generations of selection in this environment have resulted in egg production levels similar to those observed in The rate of gain in number of eggs laid for 84 days during 1970 through 1972 was three eggs per hen. A similar rate was noted in the first three generations of selection ( ), indicating that genetic variance has not declined in the egg line. Genetic increases in egg production in the egg line resulted in very little change in body weight at 16 and 24 weeks of age during early generations (Table 1). However, body weight declined greatly during the period of declining egg production ( ). When the component parts of total egg production were analyzed, it was quite apparent that there was a close association in the egg line between average clutch length and body weight at 16 and 24 weeks of age. The largest loss of body weight occurred in the seventh generation. A major gain in average clutch length was also observed during this generation. The average phenotypic correlations between adjusted generation means for 84-day clutch length and body weight at 16 and 24 weeks of age were -.79 and -.86, respectively. The respective correlations between 180-day clutch length and body weight at 16 and 24 weeks of age were -.78 and All of these correlations were significant statistically. No significant correlations were observed between generation means for total egg production and body weight and total days broody and body weight. These results suggest that changes in body weight observed in the egg line were due primarily to changes in intensity of lay as measured by average clutch length. During the period of declining egg production ( ), clutch length increased in the egg line but some of the gains in broodiness were lost, resulting in a slight loss in e gg production and a major loss in body weight. The gains in egg production observed during the first few generations in the egg line were primarily due to a reduction in broodiness. Thus, little change in body weight was observed. Average clutch length and total days broody were both closely related to the total number of eggs laid. Based on the generation means in the egg line, the correlation between average clutch length and total egg production for 84 days was.65 (P<*GS). The regression coefficient was 7,50, which indicates that for every addition of one egg to a clutch, total eggs produced would increase by 7.5 eggs (P<.05). The correlation between 84-day production and total days broody was -.91 (P<.01). The regression coefficient of -.69 (P<.01) indicates a loss of.69 egg for each day increase in the total days broody. There was no significant association between clutch length and total days broody for the 84-day production period. However, based -48-

52 on the 180-day production period, the correlation coefficient was.77 (P<.05), which suggests that as the clutch length increases, broodiness also increases. Selection for egg production alone for a few generations would improve reproduction with little absolute loss in body weight. However, there would be a loss in potential gains in body weight when a comparison is made with selection only for increased body weight. Tandem Selection A subline (egg growth) of the egg line was started in 1968 by selecting for increased 16-week body weight. The results are presented in Table 2. All of the loss in 16-week body weight in the egg line was gained back in one generation of selection for increased body weight. This was accomplished with a loss of seven eggs per hen. Two generations of selection for 16-week body weight were required to gain all the lost weight at 24 weeks of age. This was accomplished with a loss of 12 eggs per hen. However, the egg growth line still laid an average of 24.8 more eggs per hen than the corresponding randombred control population. Thus, it appears that the reduction in body weight in the egg line did not reduce genetic variation for this trait. The tandem method might be a possible approach for improving reproduction of the turkey. However, the gains in body weight with this method would probably be less than gains observed by selection for growth alone over all generations. If growth traits are at a desirable level, the tandem method might be useful in maintaining the growth and improving reproduction. TABLE 2.--Results of Selection for Increased 16-Week Body Weight in the Egg Line. 180-Day Egg Production (No.) Body Weight (Lb.) 16 Weeks 24 Weeks Year Genera- Egg tion Egg* RBC Growth* EJJCJ Growth Egg* RBC Dev.* Gaint Egg* RBC Egg Growth Dev.* Gaint Females Males ^Expressed as deviation from randombred control (RBC). f Gain in egg growth line was obtained by: deviation of egg growth line minus 1968 deviation of egg line from control..49-

53 Crossing Growth and Egg Lines This method would be useful in improving reproduction in the turkey if the offspring of such a cross exhibited desirable growth characteristics. To study the feasibility of such a cross, the egg line was crossed with two commercial large white strains in In the case of commercial strain A when the egg line was used as the female parent, growth rates of both male and female offspring were less than the average of the two parental lines (Table 3). The same was evident for males but not for females when strain B was crossed with egg line females. These results suggest TABLE 3.--Effects of Crossing Growth Strains on the Egg Line on Growth and Conformation. Commercial A Commercial B Egg AXE Av. of A and E B X E Av. of B and E A X B ( R2)* A X B ( F)t E R2 X E Av. of R2 and E F X E Av. of F and E Body Wt. 16 Wk Males Ob.) 24 Wk Breast Width (cm.) Body Wt. 16 Wk Females (Ib.) 24 Wk *A random sample of the two reciprocals of A X B designated as R2. T A sample of the two reciprocals of A X B which was selected for increased 16-week body weight designated as F Breast Width (cm.)

54 that there was negative heurosi^ fjr growth rate in these particular crosses. The amount of breast muscle (manured \ breast width at 24 weeks of age) was about intermediate between the two parental strains for both crosses. The offspring of the threeway crosses in 1967 were about intermediate between the two parental strains in both growth and amount of breast muscle. Ihc results of this study indicate that there is little nonadditive genetic variance which is important in heterosis for growth traits in the strains used. It is possible that certain lines selected for increased egg production (having larger body weight than the one used in this study) would nick well with certain growth lines to produce a desirable type offspring. However, this study and results given in the literature show that this does not generally occur. Commercial turkey breeders usually use two or three strain crosses for production of poults for sale. In many cases, an extremely large male line is crossed with a female line which is high in egg production. The question arises as to why commercial breeders cross strains when there may be no heterosis in growth rate or conformation. First of all, certain crosses may exhibit heterosis in either growth rate or conformation or both. In three-way crosses, the crossbred female might have superior reproduction. Results in the literature suggest that heterosis is exhibited in fertility and hatchability, with little or no heterosis in egg production. Use of strain crosses makes it more difficult for competitors to obtain genetic material so they can accurately reproduce a strain. All of these factors, as well as others, are probably reasons why crossing is so prevalent on a commercial basis. Simultaneous Selection for Increased Growth Rate and Egg Production One possible way of improving growth rate and egg production is by selecting for both traits by means of a selection index which gives appropriate weight to both traits. To test this possibility, a strain was started from a large-bodied randombred control population which was selected for both traits while giving three times the emphasis on 16-week body weight as on 180-day egg production (index line). The results in this line were compared with those of a strain initiated from the same source and selected only for increased 16-week body weight (F line). Selection for three generations for both increased 16-week body weight and increased 180-day egg production in the index line resulted in a significant gain in 180-day egg production (16 eggs per hen) and in 16-week body weight (0.8 Ib. per bird). However, the gain in 16-week body weight of the index line was only about 40% of the F line, even though three-fourths as much emphasis was placed on growth rate in the index line as in the 16-week line. No further gains in egg production were made in the fourth generation of selection in the index line. Major losses in egg production occurred after the fourth generation of selection. After seven generations of selection, the index line had gained approximately half as much in 16-week body weight as the F line. Selecting only for increased 16-week body weight resulted in minor changes in egg production for 84 days for the first four generations and for 180 days for the first three generations of selection (Table 4). After this, a major decrease in egg production occurred in the F line. Sixteen-week body weights of the F line increased until the sixth generation, with a slight decline in the seventh generation. The early gains in egg production of the index line were the result of a large decrease in the amount of broodiness and an increase in intensity of lay (clutch length) (Table 4). The decline in egg production in the F line was associated with a decline in intensity of lay, with little change in the amount of broodiness.

55 Although the index line lost egg production relative to the randorabred control population, it was still superior to egg production of the V line. For a 180-day production period, the differences between the two lines (Index - 1') were 13.3, 15.9, 20.2, 25.6, 30.1, 17.6, and 17.6 for generations 1 through 7, respectively. The results indicate that gains in both growth rate and egg production can be observed for a few generations. However, continued selection will likely result in a reduction in egg production, but the reduction is not as great as in a line selected solely for increased 16-week body weight. TABLE 4.--Influence of Selection for 16-Week Body Weight Alone (F Line) and in Combination With Egg Production (Index Line). Egg Production Generation (No. /Hen)* of Year Selection Days Days Av. Clutch Length* Days Days Total Days Broody* Days Days Body Weight (9 (Lb.)* V Weeks Weeks F Line "*" * Index Line Expressed as a deviation from corresponding randombred control. -52-

56 Selection for Components of Egg Production Under certain circumstances, selection for a correlated trait results in greater response in the desired trait than selection directly for the desired trait. In ordei to obtain more information on the relationship between total egg production, intensit) of lay, and broodiness, and to estimate the feasibility of using clutch length and total days of broody as selection criteria for improvement of egg production, two selected sublines were drawn from the large-bodied randombred control population. One of these lines (clutch length) was selected for increased clutch length, while the other (broody) was selected for decreased broodiness. The results are in Table 5. Although selection was effective in both lines, total egg production has not changed significantly in either line. This may have resulted because of a positive genetic correlation between clutch length and broodiness. More generations are required for more definite conclusions on this aspect of the study. Summary Several alternative means were tested to increase reproduction capability of the turkey. Among these were: (1) selection for increased egg production alone; (2) tandem selection for egg production and growth; (3) crossing growth and egg lines; (4) simultaneous selection for both growth rate and egg production; and (5) selection for a component part of total egg production. In general, genetic changes in egg production which are primarily due to changes in amount of broodiness are not closely associated with growth rate, and increases in egg production can be obtained without large losses in growth rate. However, changes in clutch length (intensity of lay) are negatively associated with body weight at 16 and 24 weeks of age so that changes in egg production resulting from changes in clutch length result in appropriate changes in growth rate. TABLE 5. The Influence of Selection for Increased Clutch Length and Decreased Broodiness. Egg Production Generation CNo./Hen)* of Year Selection Days Days Av. Clutch Length* Days Days Total Days Broody* Days Days Body Weight (2) (Lb.)* Weeks Weeks Clutch Length Line Broody Line Expressed as a deviation from corresponding randombred control. -53-

57 YOLK PRODUCTION IN TURKEYS DIFFERING IN EGG PRODUCTION LEVEL AND GROWTH RATE K. E* Nestor, W. L. Bacon* P. A. Renner, and D. A, Ehlhardt 1 Department of Poultry Science A previous comparison of an egg and a meat strain of turkeys Indicated that the meat strain had more follicles (yolksj in rapid development in the ovary than the egg strain early in the laying period (Nestor, Bacon, and Renner, 1970, Poultry Sci,, 49: ). Later in the laying period, the number of developing yolks was similar in the two strains, even though prior egg production was greater in the egg strain (Bacon, Nestor, and Renner, 1972, Poultry Sei., 51: ). Nestor and Bacon (1972, Poultry Sci., 51: 13ol-13e>5) reported that the frequency of broken eggs, membrane eggs, and soft-shelled eggs was higher in the meat-type turkeys than in the egg-type turkeys. The meat strain also had a higher frequency of two eggs laid per day than the line of turkeys selected for high egg production, In d recent study, Bacon, Nestor, and Musser (.1973, Poultry Sei., S2: ) found that the concentration of low density fraction of lipo-protein (yolk precursor) in the blood plasma was similar in the egg and growth strains, However, the total amount of circulating lipoprotein was greater in the large-bodied strains than in the small-bodied egg line. The above experiments indicate that meat-type turkeys produce more yolk material relative to their egg production rate than egg-type turkeys. Many of the yolks produced by the meat-type turkeys are either lost by regression of the developing follicles or in production of abnormal nonsettable eggs, In the above comparisons, the strains differed in both body weight and egg production. Either one of these differences could have been responsible for the results obtained, A series of experiments were conducted in order to separate the influence of body weight and egg production level on yolk production. In order to study the influence of selection for increased egg production on yolk production, a comparison was made of the yolk production of hens from a line (egg) selected for 13 generations for increased egg production and hens from the randombred control line from which the egg line originated. Hie randombred control line theoretically does not change genetically from year to year. Ten randombred control and 23 egg line hens which had been in egg production for 180 days were weighed, sacrificed* and the number and weight of the developing yolks were recorded. The number of regressing yolks was also noted. Since egg production was increased and body weight was decreased in the egg line (Table 1)* it was impossible to separate the influence of these two factors on the results obtained. Genetic increases in egg production level resulted in decreases in growth rate, mature body weight, and the number of developing yolks, with little influence on total weight of developing yolks (Table 1). The percentage of hens with yolks in a state of regression also was larger in the control line than in the egg line. Within lines, the hens used in this study laid at a greater rate than the average of all hens. This was the result of selecting only hens in production after a relatively long production period (180 days). Many of the poorer layers of each line were not laying at the time of the experiment. However, the differences between the average production Present address: The Netherlands. Spelderholt Institute for Poultry Research, Beekbergen, -54-

58 TABLE l.--the Influence of Geretic Increases in Egg Production en Ovur Production Based on Autopsy Data, Line No. Hens Prior Hens Used Egg Prod.* Entire Line Body Wt. (Kg.) Ova in No. Rapid Devel. Total Wt. (Gin.) Atretic Poll icles Per Hen (No.) Ova As Pairs (No. /Hen) No. Hens With Egg in Oviduct Egg 23 Control e SC Significance of Difference P<.01 P<-01 P<.01 P<.01 N.S. *18Q~day period. of the hens used and that of the entire line were similar for both lines. From these results, it is apparent that the selection for increased egg production, which resulted in reduced body weight and higher egg production in the egg line, at the same time reduced the number of developing yolks at any one time and probably the frequency of regressing yolks (atretic follicles) after initiation of rapid development. Three comparisons were made to study the influence of difference in growth rate on yolk development in the absence of large genetic differences in egg production. In the first comparison, heavy and medium weight varieties of colored turkeys having similar egg production were compared with the egg line hens. Hens of the three groups were fed fat soluble dye according to the method of Bacon and Cherms (1968, Poultry Sci., 47: ) beginning the day the hens were given stimulatory lighting (14 hours of light per day). After the hens were in production for 27 days, beginning when the first egg was laid, the hens were autopsied. The number of yolks in rapid development was obtained by both dye feeding and autopsy. The number of days required for rapid development and total weight of yolks in rapid development were also recorded. The heavy varieties of colored turkeys had more yolks in rapid development based on autopsy data than medium weight varieties, even though the level of egg production TABLE 2. Influence of Growth Rate and Level of Egg Production on Ovum Production Based on Dye Feeding and Autopsy Data. Dye Feeding Data Autopsy Data Total Wt. Ova in Days in of Ova Previous Rapid Rapid No. in in Rapid No. of No. Ova Body Egg Develop- Develop- Rest Egg Yolk Rapid Develop- Atretic Developing No. Wt. Prod.* ment ment Period Wt. Wt. % Develop- ment Eggs in Follicles as Pairs Hens (Kg.) (No./Hen) (No./Hen) (No.) (Days) (Gm.) (Gm.) Yolk ment (Cm.) Oviduct per Hen per Hen Significance of Difference: P<.01 N.S. P< days. Egg Line Heavy Varieties of Colored Turkeys Medium Weight Varieties of Colored Turkeys N.S. N.S. N.S. N.S. P<.01 P<.01 P<.01 N.S N.S N.S.

59 was slightly less in the heavy varieties (Table 2). However, since no significant differences existed between these two groups in number of yolks in rapid development based on interpretation of dye bands of laid eggs, it appears that the heavy varieties lost more yolks after start of rapid development. The number of regressing yolks (atretic follicles) was higher in the heavy varieties than in the medium weight varieties (1.7 vs. 0.4 per hen) but the difference was not statistically significant. The number of yolks (ova) developing as pairs tended to be higher in the heavy weight than in the medium weight varieties. The egg line hens had similar body weight and higher egg production than the medium weight varieties (Table 2). The number of yolks in rapid development^ was similar in the two groups based on autopsy data. However, based on dye" feeding data, the egg line hens had significantly more yolks in rapid development than the medium weight varieties, indicating that the medium weight varieties lost more yolks after rapid development started. The egg line hens produced more yolk material in laid eggs (previous egg production x yolk weight) than the heavy varieties of colored turkeys (432 grams vs. 300 grams), even though the heavy varieties had a larger weight of yolks in rapid development upon autopsy (Table 2). In a second comparison, large and medium weight varieties of colored turkeys were autopsied after 79 days of egg production and the number and weight of developing yolks were recorded. The heavy varieties had more yolks in rapid development and a larger number of atretic ova and yolks developing as pairs (Table 3). However, the differences were not statistically significant. The results also suggest that the over-production of yolk material in heavy turkeys is not responsible per se for the low level of production, since the prior egg production was higher in the heavy than in the medium weight varieties while the yolk production was larger in the heavy varieties. A third comparison involved three groups of hens from the same commercial strain of large white turkeys. One group was full-fed during the growing period and served as a control. The other two groups were given restricted feeding programs during the growing period resulting in 7% and 9% reductions in mature body weight. The number of yolks in rapid development and total weight of these yolks were obtained after autopsy of the three groups. Reducing body weight by restriction of feed during the growing period did not consistently reduce the number of ova in rapid development or the total weight of TABLE 3. Influence of Genetic Differences in Growth Rate on Ovum Production Based on Autopsy Data. Group No. Hens Body Wt. (Kg.) Prior Egg Production* (No. /Hen) Ova in Rapid Development No. /Hen Total Wt. (6m.) Atretic Follicles Per Hen (No.) Ova as Pairs (No./Hen) No. of Hens with Eggs in Oviduct Heavy Varieties Medium Wt. Varieties , Significance of difference P<.01 N.S. N.S. N.S. N.S. N.S. *79 days. -56-

60 TABLE 4.--The Influence of Environmental Difference in Growth Rate on Ovum Production Based on Autopsy Data. Group No. Hens Body Wt. (Kg.) Prior Egg Production* (No. /Hen) Ova in Rapid Development No. /Hen Total Wt. (Gm.) Atretic Follicles Per Hen (No.) Ova Developing As Pairs (No. /Hen) No. of Hens with Eggs in Oviduct Control 5 Restricted 1 11 Restricted 2 9 Significance of Difference P< N.S N.S. N.S N.S N.S *31 days. these ova (Table 4). However, the number of atretic follicles and ova developing as pairs was less in the groups grown on limited feed intake. The frequency of occurrence of atretic or regressing yolks and yolks developing as pairs was consistently greater in the faster growing turkeys than in the more slowly growing turkeys, although no significant difference existed in any of these comparisons. Since earlier work at OARDC observed similar results, it is suggested that the difference observed is real and that an increase in body weight is accompanied by an increase in atretic yolks and an increase in the number of yolks developing as pairs. Summary After 13 generations of selection for increased egg production, decreases in body weight, number of yolks in rapid development in the ovary, and number of regressing yolks were observed in the selected line when compared to the corresponding randombred control. Large body size strains when compared with medium size strains had more yolks in rapid development, more yolks developing as pairs, and more regressing yolks. When body weight was reduced by feed restriction, fewer yolks became atretic and fewer developed as pairs. -57-

61 INFLUENCE OF STRAIN AND FEED RESTRICTION ON EGG PRODUCTION, EGG QUALITY, AND HATCHABILITY OF CAGE HENS K. I. Brown, K. E. Nestor, S. P. Touchburn 1 and D. A. Ehlhardt 2 Department of Poultry Science There has been considerable interest recently in the possibility of maintaining turkey hens in cages during the breeding period, LVJdeuce obtained at this Research Center indicated that hens from a medium weight strain selected for high egg production laid as many eggs in cages as similar hens housed in floor pens. However, the hens in cages had more broken and abnormal eggs. Although no strict comparisons have been made, large bodied hens do not appear to perform as well in cages as in floor pens In order to study the effect of strain on performance of hens in cages, three commercial strains were obtained. Strain A had been selected for 11 generations on wire floors, hmplusis in this strain was placed on egg production traits. Strain B had been trapnested onl> on wire for 4 years and partially on wire for about 8 years. The selection criteria for this strain were unknown. Strain C had not been maintained in cages or wire floors. One group of the females of each strain was full fed during the growing period when maintained in floor pens. The other group of hens was placed on a restricted feeding program beginning at 3 weeks of age. This program initially restricted the amount of feed to 90% of the average feed consumption for the full-fed group the previous week. The amount of restriction was modified to try to obtain a reduction in body weight to 8Q o of the full-fed group at 52 weeks of age* when the birds were given a stimulatory light regime of 14 hours of light per day. All hens were restricted to a 6-hour light day at 20 weeks of age. The body weights obtained at various ages are presented in Table 1. The goal of restriction (80 o of the body weight of the full-fed group) was almost reached at 32 weeks of age. The restricted group of hens when put on full feed at 52 weeks of age gained more weight than the group which was full fed during the growing period. By 47 weeks of age (after approximately 12 weeks of egg production), the restricted groups weighed 93 to 97 o as much as the full-fed females. There was a highly significant effect of strain on body weight. Strain C was largest, followed by strains B and A. The feed restriction was similar on all strains since there was no interaction between strain and type of feeding. Restriction of feeding during the growing period resulted in a significant reduction in the length of the shank and keel and in body depth (Table 1). The effect on body depth was the greatest. There was no significant effect on breast width. Thus, the restriction influenced bone development, with no effect on the amount of fleshing. The strains differed significantly in shank length, keel length, and body depth. 1 Present address: Canada, Macdonald College, Ste. Anne de Bellevue, Province of Quebec, 2 Present address: The Netherlands. Spelderholt Institute for Poultry Research, Beekbergen, -58-

62

63

64 Egg production was recorded for a 16-week period from August 16 until Dec. 6. Eggs were recorded on the egg record sheets as normal, broken but with an apparently normal shell, weak-shelled (partial shell formation but breaks easily), shell-less (membrane eggs), and slab-sided (one side flattened due to two eggs in uterus or shell gland at the same time). Strain A, which had been selected for production on wire floors, had significant! higher egg production than the other two strains (Table 2). Strains B and C had similar egg production, even though strain C had not been selected on wire floors and strain B had been selected for four or more generations on wire floors. However, it is unknown what the selection criteria were for strains B and C. Restricted feeding had no statistically significant effect on egg production, even though the hens on the restricted program laid 4.8 eggs more per hen than those on full feed. There was no significant strain by type of feeding interaction, even though the restricted group of strains A and C appeared to have improved egg production while that of strain B was depressed. Strain or type of feeding (restricted vs. full fed) had no significant effect on the production of abnormal eggs (weak-shelled, slab-sided, and shell-less). A recent study indicated that eggs classified as broken but with apparent normal shells on visual inspection actually had reduced shell thickness. Egg weight, specific gravity, and percent weight loss after 7 days of incubation were obtained on individual eggs laid by the hens from Sept. 16 through Sept. 30. These measurements were made on eggs which were apparently normal upon visual inspection. Specific gravity was obtained by weighing the eggs in the air and under water and using the formula: weight in air divided by (weight in air minus weight in water). Specific gravity measures the amount of shell. The strains differed significantly in egg weight (Table 2), varying in the same order as their body weights (Table 1). There were no statistically significant differences between full or restricted fed hens in their average egg weight during this period. The average egg weight for restricted hens was actually slightly greater, even though their body weights were significantly lower. The specific gravity of eggs laid by strain A was lower than those laid by the other two strains. This was probably due to the higher egg production of strain A. There was no effect of feed restriction on specific gravity. Neither strain nor type of feeding influenced percent hatchability of fertile eggs during a 4-week hatching period from Oct. 11 to Nov. 1. Summary A survey of the performance of three commercial strains in laying cages revealed a significant difference in total number of eggs produced, with no influence on the frequency of broken, weak-shelled, slab-sided, or shell-less eggs. A strain known to have been selected for several generations for high egg production when housed on wire floors had superior egg production. Restricted feeding during the growing period did not significantly influence egg production. Strain differences in egg weight and specific gravity were observed. There was no strain difference in hatchability of fertile eggs. -61-

65 EGG SHELL QUALITY ACCORDING TO CLUTCH SIZE AND POSITION IN CLUTCH AND ITS RELATIONSHIP TO HATACHABILITY Karl E. Nestor and D. A. Ehlhardt* Department of Poultry Science Turkeys lay eggs In clutches and then skip 1 or more days between clutches. The clutch is therefore defined as the number of consecutive days a hen lays an egg without skipping a day. Clutch length can vary from a single egg to 20 or more. One purpose of this study was to estimate the influence of clutch length and position of the egg within the clutch on egg weight, specific gravity, and percent weight loss after 7 days of Incubation. The data used were from three commercial strains, with each subjected to either full feeding or restricted feeding during the growing period. The results of the strain and feeding program on egg traits are reported on p. 58. For this analysis, the effects of strain and type of feeding were removed by statistical analysis. The hens were given 14 hours of light per day when they were 52 weeks of age on July 26. They had been restricted to a 6-hour light day for 12 weeks prior to this time. The egg measurements were made on normal eggs laid between Sept. 16 and Sept. 30. The egg measurements included egg weight, specific gravity, and percent weight loss after 7 days of incubation. Specific gravity was measured by determining the weight of eggs in air and submersed in water and dividing the weight of the eggs in air by (weight of eggs in air minus weight of eggs in water). Clutch lengths were as follows: 1 egg, 2 eggs, 3 eggs, and 4 or more eggs. The grouping of the longer clutches was necessary since each of the longer clutches was not represented in each strain-type of feeding subgroup. The position in the clutch was the actual position for clutch lengths 1 through 3. For clutch length 4 (all of the longer clutches), clutch position 1 was the first egg and clutch position 3 was the last egg of the clutch, Clutch position 2 represented the average of all the eggs for the intermediate positions between the first and last eggs. Clutch size and position in clutch had no significant effects on egg weight and specific gravity of the eggs (Table 1). Since the specific gravity of the shell is much greater than that of the contents of the egg, variation in specific gravity of the whole egg reflects variation in specific gravity of the egg shell. Position in the clutch had a significant effect on weight loss after incubation, which is a measure of porosity of the egg. The first egg of all clutches had similar weight loss. In two egg clutches, there was no significant difference between the weight loss of the first and second eggs of the clutch. In clutches of three or more eggs, the first egg of the clutch was more porous than the last egg of the clutch, while the eggs in the intermediate positions had less porosity than either the first or last eggs of the clutches. The difference in porosity was not due to differences in amount of shell since specific gravity did not exhibit a similar difference. The frequency of broken eggs (eggs with apparently normal shell upon visual examination), weak-shelled eggs (eggs with some shell but which break easily), shellless eggs (membrane eggs), and slab-sided eggs (eggs with a depression resulting from *T Present address: Spelderhold Institute for Poultry Research, Beekbergen, The Netherlands. -62-

66 two eggs being in the uterus or shell gland at the same time) were recorded during the entire 16-week laying period. The ones classified as broken, but apparently normal, had reduced shell thickness in comparison to nonbroken eggs in a later study. Correlation coefficients among these measures and egg traits were estimated and are given in Table 2. Percent hatchability of fertile eggs was based on four weekly hatches from Oct. 11 to Nov. 1. The total number of eggs laid per hen was positively correlated with percent hatchability of fertile eggs. Ihis indicates that better layers also have better hatchability. As expected, percent normal eggs was negatively correlated with the percent of abnormal eggs. The hens which lay a larger proportion of normal eggs also have higher hatchability as indicated by the correlation coefficient of 0.36 between percent normal eggs and hatchability. Percent broken eggs was positively correlated with percent weak-shelled eggs, but with no other frequency of abnormal eggs. Percent weak-shelled eggs also was significantly correlated with percent shellless eggs, but not with percent slab-sided eggs. Only percent shell-less eggs was significantly negatively correlated with specific gravity of normal eggs. Percent weak-shelled eggs and percent slab-sided eggs were negatively correlated with hatchability of fertile eggs. When the frequency of these abnormal eggs increases, hatchability declines even though only apparently normal eggs were used for hatching. Specific gravity was negatively correlated with percent incubation weight loss, indicating that as the amount of shell increases, the porosity declines. However, neither measure of shell quality was significantly correlated with hatchability of fertile eggs. TABLE "L The Effects of Clutch Length and Clutch Position on Egg Weight, Specific Gravity, and Percent Incubation Weight Loss* Clutch Length Position in Clutch Egg Wt. Specific Gravity (gm./cc.) Percent Incubation Wt. Loss b * ab a b d b e d *Means not followed by the same letters are significantly different (P<.05). -63-

67

68 TABLE 3. Egg Quality of the "Normal" and Slab- Sided Members of Pairs and Normal Eggs Not Laid in Pairs. Normal 84.8 Slab-sided Eggs 84.6 Egg Specific Percent Weight Gravity Incubation (gm-) (gm./cc.) Weight Loss Normal Eggs Not in Pairs In order to find the reason for the strong negative correlation between percent slab-sided eggs and percent hatchability of fertile eggs, a study of eggs laid in pairs in which one of the members was slab-sided was made. Slab-sided eggs are caused by a second egg entering the shell gland before the previous egg is laid. Lgg weight of the normal member and slab-sided member of the pairs was similar (Table 3). However, the slab-sided egg had less specific gravity and higher incubation weight loss (higher porosity) than the normal egg of the pair. The normal egg of the pair had similar egg weight and specific gravity but less incubation weight loss than other normal eggs not laid in pairs. This suggests that the normal egg of the pair is held without calcification, but some change occurs which results in decreased porosity. Previous studies have shown a significant negative correlation between percent weight loss after 7 days of incubation and hatchability. The correlation in this study was positive, but not significant. The reason for this inconsistency can be obtained from the data presented in Table 4. In both studies, eggs with a loss of TABLE 4.--Percent Hatchability of Fertile Eggs Summarized According to Percent Injcub^tl^qn Weight Loss in the Present and Former Study. Percent Incubation Weight Loss Present Study(Autumn) Former Study (Summer) Percent Percent Hatchabi1i ty Hatchabi1i ty No. Hens of Fertile Eggs No. Hens of Fertile Eggs Greater than

69 4% or less after 7 days of incubation had reduced hatchability. Presumably a certain amount of porosity is required to enable the proper gaseous exchange. Eggs with weight losses of 4 to 7% had similar hatchability. Those eggs with very porous shells (weight loss of greater than 7%) have reduced hatchability and are probably responsible for the significant negative correlation obtained in the former study. Since no eggs had weight losses greater than 6% in the present study, the correlation was slightly positive. The negative correlation between frequency of slabsided eggs and percent hatchability can be explained as follows. As the frequency of slab-sided eggs increases, the frequency of normal eggs of the pairs also increases. Since these eggs appear normal, they are set. But since they have less porosity than required for high hatchability, the average hatchability for the hen is reduced. Summary Clutch size and position in the clutch had no significant influence on egg weight or specific gravity of the egg. The first egg of all clutch sizes had similar percent weight loss after 7 days of incubation. However, in clutches of three or more eggs, weight loss after incubation was reduced in eggs laid later in the clutch. The total number of eggs laid and percent normal eggs were significantly correlated positively with hatchability of fertile eggs. The hatchability of eggs with either a small or large percentage weight loss after 7 days of incubation was reduced in comparison to those eggs with an intermediate weight loss. The normal egg laid as a pair with a slab-sided egg had reduced weight loss which resulted in a significant negative correlation between the frequency of slab-sided eggs and percent hatchability. -66-

70 ASSOCIATION OF CERTAIN BODY MEASUREMENTS ON LIVE BIRDS WITH THE AMOUNT OF LEG MUSCLE IN THE TURKEY K. E. Nestor, P. A. Renner, and D. A. Ehlhardt 1 Department of Poultry Science Leg weakness Is a major problem of male turkeys both during the growing and breeding periods. I his leg weakness has been attributed to many factors including nutrition,, disease, genetics, etc. Most of the emphasis of breeders of large white turkeys has been on increasing body weight and the amount of white meat (breast). Although the actual amount of leg muscle increases with increases in body weight, the relative amount based on percent of body weight actually decreases (Marsden, 1940, Poultry Sci., 19: 23-28; Miller, 1958, Poultry Sci., 47: ). Thus, as body weight is increased by selection, the amount of leg muscle also increases, but at a less rapid rate. This may place a stress on growing turkeys as their body weight increases which might result in leg weakness. Several body measurements were made on live 24-week old male turkeys, and these measurements were correlated with the amount of leg muscle. Among the measurements made were: body weight, body depth, breast width, length of shank, width of shank, length of drumstick, circumference of drumstick, and two width measurements of the drumstick. The latter two measurements were made at the largest end of the drumstick. The widths from head-to-tail or anterior to posterior (measurement A) and from sideto-side or laterally (measurement B) were measured with calipers. In order to estimate how repeatable the various measurements were, two people made four measurements each on every male. The male was returned to the pen after each series of measurements so that the person doing the measuring did not know the idenity of the bird. Twenty males were used in this experiment. After all of the measurements were completed on the live birds, they were killed and total leg weight, thigh weight, and drumstick weight were measured. Total leg muscle, thigh muscle, drumstick muscle, thigh bone, and drumstick bone were also weighed. The percent thigh muscle, percent drumstick muscle, and percent total leg muscle were calculated. The average values for the various measurements on the live birds, according to observer, are presented in Table 1. The averages for the two observers were similar but not identical for each trait. These slight differences occurred even though both observers were instructed in detail on how to make the various measurements. Repeatability estimates were made for the measurements on the live birds. Repeatability values can range from 0 to 1.0, with values close to 1 indicating very small variation among the different measurements of the same bird, while those close to 0 indicate large variation among measurements on the same bird. These values are presented in Table 2. The effects of differences between observers were removed prior to obtaining the repeatability values. Repeatabilities for body weight and measurements of skeletal size (body depth, shank length, shank width, and drumstick length) were large. Drumstick circumference, which measures both muscle and bone, 1 Present address: Spelderholt Institute for Poultry Research, Beekbergen, The Netherlands. -67-

71 TABLE 1. Average Values for the Various Measurements to Observer.* According Trait Observer Average Body Weight (Ib.) A B Body Depth (cm.) A B Breast Width (cm.) A B Shank Length (cm. ) A B Shank Width (cm.) A B Drumstick Width A (cm.) A B Drumstick Width B (cm.) A B Drumstick Length (cm.) A B Drumstick Circumference (cm.) A B *Based on four measurements on each male by each observer. TABLE 2. Repeatability of the Various Body Measurements. Measurement Repeatabi1i ty Body Weight.98 Body Depth.84 Breast Width.32 Shank Length.79 Shank Width.68 Drumstick Width A.08 Drumstick Width B.31 Drumstick Length.60 Drumstick Circumference

72 was moderate in repeatability. Breast width and drumstick width B have similar repeatability. The repeatability of drumstick width A was very low, probably because of the large amount of muscle measured. Correlation coefficients were calculated between measurements made on the live birds and measurements of leg muscle and bone. These measure the relation between the two traits in question. The coefficient can range from to A negative correlation coefficient indicates that as one trait increases in magnitude, the other trait declines. A positive coefficient indicates that both traits tend to move in the same direction. The closeness of the relationship is indicated by the size of the correlation coefficient. Those approaching ±1.00 indicate a strong association, while those close to 0 indicate that there is no association or that the association is weak. TABLE 3.--Correlation Coefficients Between Measurements Made on Live Birds and Amount of Leg Muscle and Bune*. Total Leg Wt. Total Thigh Wt. Thigh Muscle Wt. Total Drumstick Wt. Drumstick Muscle Wt. Total Leg Muscle Wt. Percent Thigh Muscle Percent Drumstick. Muscle Percent Total Leg Muscle Tnigh Bone Wt. Drumstick Boiit Wt. Body Wt..92 Body Depth.55 Breast Width ,22 Shank length ' Shank Width "Drumstick Width A Drumstick Width B.65, Drumstick Length Drumstick Circumference,92 Total Leg Wt., Total Thigh Wt ,74 Tota3 Drumstick Wt..94,97." M»? o Thigh Muscle Wt ;< Drumstick Muscle Wt «Total Leg Muscle Wt. Percent Thigh Muscle , H./.,.,1 Percent jrumstick Mu s c 1 e. 7/ *A correlation coefficient of.44 is si jnificanct at D.01.»J., f. nr\ 4 t*k ic map.tprt T it -69-

73 The correlation coefficients are presented in Table 5. Body weight was closely correlated with actual weights for the entire leg and its component parts. However, there was no significant association between body weight and percent muscle in the leg. Body depth, breast width, drumstick length, shank length, and shank width were negatively correlated with percent leg muscle, although most of these coefficients were not statistically significant or different from 0. Drumstick width A exhibited no consistent relation with percent leg muscle, while the relationship was positive and significant in two out of three correlations for drumstick width B. Drumstick circumference was significantly correlated onjy with percent thigh muscle. Many of the correlation coefficients shown in Fable 5 were the result of a common association of the two traits in question with body weight, r lo study the association of traits in the absence of the influence of body weight, correlation coefficients were obtained after the influence of body weight was removed from each trait by statistical analysis. These correlation coefficients are given in Table 4. TABLE 4.--Correlation Coefficient* Between Measurements Made on Live Birds und Amount of Leg Muscle After the Influence of Body Weight Was Removed, Total fotdl Total Drum- Leg Thigh stick Wt. Wt. Wt. Thigh Muscle Wt. Of urnstick Hustle Wt. Total leg Muscle Wt. Percent Thigh Muscle Percent Drumstick Muscle Percent Total Leg Muscle Thigh Bone Wt. Or urnstick Bone Wt. Body Depth Breast Width lb Shank Length ,41 Shank Width Drumstick Width A ,07 Drumstick Width B Drumstick Circumference.61,48.70 Drumstick Length , Total Leg. Wt Total Thigh Wt Total Drumstick Wt Thigh Muscle Wt Drumstick Muscle Wt Total Leg Muscle Wt Percent Thigh Muscle Percent Drumstick Muscle *A correlation coefficient of.46 is needed for significance at P*-.Q5 and a coefficient of.58 at P<

74 Most of the correlation coefficients between measures of leg muscle and measures of skeletal size (body depth, shank length, shank width, and drumstick length) were negative or very slightly positive after the Influence of body weight was removed. Drumstick width B and drumstick circumference appeared to be closely associated positively with the percent of leg muscle. Since these measurements also had moderate ropeatabilities (Table 2), they appear to be the best measurements for estimating percent leg muscle. A selected line will be Initiated in which selection is based on increased percent leg muscle at a constant body weight. The presence of leg weakness will be closely followed in this line, Summary Body weight, body depth, breast width, shank length, shank width, drumstick length, drumstick circumference, and drumstick width were measured on live birds and correlated with the percent of leg muscle. Drumstick circumference and drumstick width, measured from side-to-side, were strongly associated positively with percent leg muscle and appeared to be the best indicators of percent leg muscle. -71-

75 INFLUENCE OF PLUMAGE COLOR PATTERN GENES ON GROWTH AND REPRODUCTION IN THE TURKEY Karl E. Nestor Department of Poultry Science An attempt was made to obtain turkeys carrying all known plumage color pattern genes. These colored turkeys, which were quite common in the past, are now almost extinct. Since evidence obtained at Pennsylvania State University by Dr. L. Buss and at OARDC (Turkey Research Summary 64, 1972) indicated that these genes might have beneficial effects on growth and mortality, a program was initiated to test the possible value of these genes on traits of economic importance. Since most of the genes obtained were from very small, exhibition-type turkeys with poor growth, individuals carrying the plumage color genes were crossed with either large white turkeys maintained at the Research Center or large bronze turkeys obtained from a commercial breeder. This was done in order to place the genes in a growth background which was closer to that of the modern large white turkey. Body weights of the pure varieties and their crossbreds for three crosses are given in Table 1. The two reciprocal crosses between the large whites and reds (Wo x R9 vs. RCT x w9) differed greatly in body weight to 16 weeks of age. This may have been a maternal effect due to the small egg size of the red females. When positive heterosis occurs, the average of the crosses is greater than the average of the parent lines and approaches that of the larger parent line. The average of the male TABLE 1,~The Influence of Crossing Different Varieties on Body Weight. Body Weight (Ib.) Pure Variety or Cross 8 Wk. Males 16 Wk. 24 Wk. 8 Wk. Females 16 Wk. 24 Wk, Large White (W) Bourbon Red (R) Rcfx W^ Av. of Parents Av. of Crosses Light Gray (G) GC? x W Wff x G$ Av. of Parents Av. of Crosses Bronze (Br) Dark Brown (DB) DBtf? x Br^ Av. of Parents « »

76 offspring of the crosses of red and white parents was similar to that of the average of the parents, indicating no heterosis. In the case of the female offspring, the average of the crosses was lower than the average of the parents, suggesting negative heterosis. Although body weights were not recorded at 24 weeks of age, the reciprocal crosses of whites and reds appeared to be only slightly larger than that of the pure reds. The weight of the pure light gray group more closely approached the weight of the large whites than did the body weight of the pure reds. Some heterosis appeared to be obtained in female body weight, with no heterosis in the male body weights. The reverse was true for the crosses of bronze and dark brown varieties. Generally,, in the crosses given in Table 1 and others not given between varieties differing greatly in body weight, there was very little heterosis in body weight. Since the averages in Table 1 were based on from 25 to 250 individuals per group, some of the inconsistencies between sexes may have been due to chance sampling. However, the general lack of heterosis in these crosses was evident. Most of the pure varieties are maintained with four to six parental pairs in which one male is mated to one female. Under these conditions, there does not appear to be any deterioration in growth rate or egg production of the pure strains. The bronze variety has been maintained with 15 parental pairs, with some selection for increased growth rate since this variety is involved in several crosses. TABLE 2.--The Influence of the Gray Gene (c9) on Body Weight, Egg Production, and Semen Production in Offspring of Reciprocal Crosses of Light Gray (c9c) and White (cc). Light Gray (c9c) White (cc) Male Body Wt. (15.) No. 8 Wk. 16 Wk. 24 Wk. Female Body Wt. (Ib.) No. 8 Wk. 16 Wk. 24 Wk. 70 Day Egg Production No. Hens Av. per Hen Semen Yield (cc.) No. Av. Sperm Cone, (billion/cc. ) Total Sperm/Ejaculate (billion) *P<.05. **P< * * 4.09*.75**

77 Reciprocal crosses were made between heterozygous light gray individuals (e*e) and homozygous white Individuals (ccj. The offspring from these crosses were divided about equally between light gray (c^e) and white individuals tec). When comparisons were made between the performance of light gray and white individuals from the same family and same hatch, the effect of the c 8 gene could be evaluated. The results of these test crosses are presented In Table 2. The light gray gene had no effect on body weight to 24 weeks of age. The light gray gene did improve egg production, semen yield, sperm concentration, and total sperm per ejaculate. The egg production period began with the production of the first egg. The semen traits were based on four or five collections from the males early in the reproduction period. These collections were made no closer than 2 days apart. Sperm concentration was estimated by use of a spectrophotometer. Test crosses to evaluate the effect of genes responsible for prohibiting the expression of the red color resulting in the bronze plumage pattern were made by taking the offspring from the reciprocal crosses of large whites and reds (Table 1) and mating these males and females (Rr) to pure red Individuals (rr). After this initial test cross using pure red males and females, other test crosses were made utilizing bronze and red individuals of the test crosses for the previous year. The R gene, based on TABLE 3. The Effects of the Red Gene (r) on Body Weight, Egg Production, and Semen Production in Reciprocal Crosses of Bronze (Rr) and Red (rr) _ ^ ^m r ^ n r^^^^ Bronze (Rr) Red (rr) Male Body Wt. (Ib.) No Wk ** 16 Wk ,4** 24 Wk ** Female Body Wt, (Ib.) No, Wk. 2, Wk Wk, Day Egg Production No. Hens Av. per Hen Semen Yield (cc.) No. Av. Sperm Cone, (billion/cc.) Total Sperm/Ejaculate (billion) **P<

78 preliminary data (lablc 3), appeared to improve body weight and egg production, with little influence on semen yield. More data need to be collected on this test cross. Test crosses are being made to evaluate the effects of the genes responsible for other plumage patterns (black, narragansett, dark brown, and slatej. Evidence already obtained indicates that the plumage color genes may have an influence on traits of economic importance, and this effect should be evaluated before these genes become extinct. Summary Crosses between various varieties of turkeys carrying different plumage color genes and either large white or large bronze turkeys did not exhibit any significant heterosis in body weight. Data from backcrosses indicated that the gene responsible for gray color (cs) improved egg production and semen production, with no effect on growth rate, in comparison to the gene responsible for white color (c). Preliminary evidence suggests that bronze (Rr) individuals from test crosses of bronze (RR) and red (rr) may have better growth and egg production than red (rr) individuals from the same cross. -75-

79 INHERITANCE OF LIVABILITY DURING THE EARLY GROWING PERIOD OF TURKEYS K. E. Nestor., P. A. Renner, and K. I. Brown Department of Poultry Science A large portion of the total mortality prior to market age in turkeys normally occurs during the first few weeks after hatching. Since the cost of poults is high, this represents a large economic loss to turkey growers. One method of evaluating the inheritance of a trait is to estimate the heritability of the trait. Heritability estimates the amount of genetic variation relative to the total variation of the trait resulting from all causes. A search of the literature failed to reveal any studies on the inheritance of viability during the early growing period of turkeys. Kondra and Shoffner (1955, Poultry Sci., 34: ) obtained heritability estimates varying from -.98 to.32 for livability to 24 weeks of age in four strains of turkeys. Heritability estimates theoretically range from 0 to +1.0, with the size of the estimate indicating the proportion of the variation resulting from genetic differences. However, due to sampling errors, estimates may be negative or greater than +1.0, although the true heritability can only be from 0 to Livability is an all-or~none character. Either an individual turkey lives or dies, Therefore, a special method of Robertson and Lerner (1949, Genetics, 34: ) and Dempster and Lerner (1950, Genetics, 35: ) was used to estimate the heritability of livability based on variation between full-sib families. Heritability estimates were obtained within lines and years. To obtain an estimate based on several years data, the values were polled to give the final estimate. Heritability of Livability from 0 to 4 WeeksofAge Three lines of turkeys were used to obtain estimates of heritability during this period. These lines consisted of a randombred control (RBC3) and two lines selected on the basis of their blood level of corticosterone after cold stress at 4 weeks of age. One of these lines was selected for a high level (high line) and the other for a low level (low line). Considerable mortality occurred as the result of cold stressing and collecting a 5-ml. blood sample. The majority of these birds died during collection of blood by heart puncture or a few hours thereafter. A few birds died in the cold room each year. Heritability estimates of the livability during cold stress were obtained. The heritability estimates of livability to 4 weeks of age are in Table 1. The unadjusted estimates in the randombred control 3 line averaged slightly greater than those of the low and high lines, although none of the estimates were large. The regression of heritability estimates on mortality level was highly significant (b =.0085), indicating that as mortality increased, the heritability estimates increased, After adjustment for differences in mortality level by regression analysis, the average estimate of the RBC3 and high lines was similar, while that of the low line was lower. Selection for low response to cold stress has reduced mortality to 4 weeks of age. Therefore, it is possible that some of the genetic variation has been dissipated. The heritability of livability during cold stress was low, indicating that genetic differences were not an important source of variation. -76-

80

81 A brief summary of lines used to obtain estimates for livability to 8 weeks of age Is presented in fable 2. The egg and semen lines were subjines of the medium weight randombred control, while lines F, 1, C, B, and R were sublines of the large bodied randombred control line. The heritability estimates of livability to 8 weeks of age are given in Tables 3 and 4. In general, these estimates for the medium weight lines were a little higher than those for the large bodied lines. In order to study the influence of mortality level on the magnitude of the heritability estimates, the estimates from the various lines were grouped according to mortality level and a pooled estimate was obtained. Also, linear regression coefficients of heritability estimates on mortality level were calculated from the yearly means. Table 2. Lines Used to Obtain Estimates of Heritability of Livability to 8 Weeks of Age. Line Weight Classification Selection Criteria Reference Randombred Control 1 Egg Medium Medium None Increased Egg Production McCartney (1964, Poultry Science, 43; ) Nestor (1971, Poultry Science, 50: ) Semen Medium Increased Semen Yield Nestor and Brown (1970, Poultry Science, 49: 1421) Randombred Control 2 Large None Nestor, McCartney, and Bachev (1969, Poultry Science, 48: ) F Large Increased 16-Wk. Body Weight I Large Increased 16-Wk. Body Weight and Increased Egg Production C Large Increased Clutch Length B Large Decreased Total Days Lost from Broodiness R Large Increased Fertility -78-

82

83 TABLE 5. Summary of Heritability Estimates (h 2 ) According to Mortality Level Mortality Level (*) No. of h 2 Estimates Average Mortality (%} h 2 Standard Error Average No. per Fami ly Total No. of Families The size of the heritability estimates increased with increases in the mortality level (Table 5). The estimates based on levels of mortality greater than 10 0 o were higher than corresponding estimates reported for chickens in the literature. The linear regression coefficients of heritability estimates on mortality level were.011 and.020 for the medium weight and large weight strains, respectively. Both of these regression coefficients were highly significant. Thus, for every l o increase in mortality, the heritability estimates of livability would increase by.011 and.020, respectively, for the medium and large weight strains. The results obtained in this study indicate that the heritability of livability during the early growing period of turkeys is low. Therefore, genetic progress from selection would be slow and selection should be on a family basis. Summary Heritability estimates of livability of turkeys from 0 to 4 weeks averaged,080±.014. Estimates of livability during cold stress and collection of blood at 4 weeks of age averaged.002±.012. The heritability estimate of livability from 0 to 8 weeks of age in several lines of medium weight and large weight turkeys was.117±.008. Heritability estimates of the medium weight strains (.130+,010) averaged slightly higher than that of large weight strains (.G96±.Q12). The magnitude of the heritability estimates increased with increases in mortality. -80-

84 INFLUENCE OF GENETIC CHANGES IN SEVERAL TRAITS ON POULT MORTALITY DURING THE EARLY GROWING PERIOD K. E. Nestor, K. I. Brown, and P. A. Renner Department of Poultry Science A large portion of the total mortality of the growing period normally occurs during the first 2 weeks after hatching. Losses from dehydration occur during the 3rd and 4th days after removal from the hatcher (Enos, Kienholz, and Moreng, 1971, Poultry Sci., 50: 1574). Starvation results in mortality from 6 to 12 days of age (Harper and Babcock, 1953, Poultry Sci., 32: ; Moreng, Settle, Kienholz, and Enos, 1970, Poultry Sci., 49: ; and Enos, Kienholz, and Moreng, 1971, Poultry Sci., 50: 1574). Research at QAJRJDC has indicated that genetic differences contribute to the variation in livability during the first 8 weeks of age, but the heritability estimate (0.12) was low and increased with increases in mortality level. A search of the literature failed to reveal any studies on the influence of genetic changes in traits of economic importance on mortality during the early growing period. Mortality levels of all lines at OARDC were analyzed to evaluate the effects of genetic changes in these traits on poultry mortality during the early growing period of the turkey. The lines analyzed were selected for increased egg production (E), increased 16-week body weight (F), both increased egg production and increased 16-week body weight (I), increased semen yield (S), and high (H) and low (L) blood corticosterone levels after cold stress (held without food or water in a cold room at a temperature of 40 C. for 4 hours) at 4 weeks of age. Their corresponding randombred control populations (RBC1, RBC2, and RBC3) were analyzed to evaluate the effects of genetic changes in these traits on poult mortality during the early growing period of the turkey. A description of these lines and management procedures used was given previously by Nestor, Renner, and Brown (1973, Poultry Sci., 52: ). An analysis of variance based on full-sib family averages was used to estimate the influence of lines, years, and the lines x years interaction. Years and generations were confounded since only one generation was obtained each year. Mortality data on all lines except H, L, and RBC3 were summarized according to size of family. The correlation coefficient of the dam! s egg weight and mortality of her offspring to 8 weeks of age was estimated for 1970 and 1971 data of these lines. Average egg weight was estimated by periodically weighing all eggs laid by each hen during a 12-week hatching period beginning when the hens first attained an egg production level of approximately 50%» In order to estimate the influence of hatching time on early poult mortality of the E line, the eggs from a flock of E line hens were divided into two groups in each of five weekly hatches. One group served as a control in which the poults were removed from the hatcher after 28 days of incubation. Poults from the other group of eggs were removed after the 26th, 27th, 28th, and 29th days of incubation. All poults were placed on feed and water immediately after wingbanding on the day of removal from the hatcher. All groups were reared intermingled in battery brooders. Mortality was recorded to 3 weeks of age. Genetic increases in egg production in a medium weight line (E) resulted in a highly significant increase in mortality of poults to 8 weeks of age (Table 1) based on analysis of data from the fifth (1965) through the twelfth (1972) generations of selection. Mortality occurring in the E line was consistently higher than the cor-

85 TABLE 1. The Influence of Genetic Increases in Egg Production in the E Line on Mortality to 8 Weeks of Age. Mortality (%) Line Average RBC E f Average for Both Lines f *Expressed as deviation from the control line (RBC1). "'"Differences between lines, years, and their interaction were highly significant. responding randombred control line (RBC1) in all generations of selection studied. However, there was no trend with years based on linear regression for the deviation of the E line from the RBC1, indicating that no further change occurred after the fifth generation of selection. Year differences and the interaction of lines and years were highly significant. The interaction resulted from the difference between lines being highly variable from year to year (range was 0.9 to 12.2% with an average of 2.9%). The line differences were usually larger in those years in which mortality in the RBC1 line was greater, although in 1967 the mortality level was highest and the line difference was the smallest. TABLE 2.--The Effects of Genetic Increases in Both 16-Week Body Weight and Egg Production (I Line) and in 16-Week Body Weight Alone (F Line) on Mortality to 8 Weeks of Age. Line Mortality Average RBC , I* F* Average of All Lines f "^Expressed as deviation from the corresponding randombred control line (RBC2), "*"Year differences and the interaction of lines and years were significant, Genetic increases in both 16-week body weight and 180-day egg production in a heavy weight line (I) did not result in increased mortality during the first (1967) through the sixth (1972) generations of selection (Table 2). Likewise, no signifi- -82-

86 TABLE 3.--The Effects of Genetic Increases in Semen Yield in the S Line on Mortality of Poults to 8 Weeks of Age, Line RBC1 S* Mortality (%) Average Average for Both Lines , ,6 f ^Expressed as the deviation from the corresponding randombred control (RBC1). [Differences between years and interaction between years and lines were highly significant cant changes in early poult mortality were observed in a line (F) selected for, and exhibit ing, increased body weight at 16 weeks of age. The effects of years and the interaction between lines and years were statistically significant. The two selected lines (I and F) did not exhibit consistent differences from their corresponding randombred control line. However, there was no trend in the differences with generations (years). The average yearly differences from the randombred control for the I and F lines were +1,0 and +1.5, respectively. Mortality during the early growing period was not influenced by genetic increases in semen yield over six generations of selection in a medium weight line (S) (Table 3). However, tbc^ie was a highly significant interaction of lines and years which was not the result of a time trend in the S line. Genetic changes in the blood corticosterone level after cold stress at 4 weeks of age resulted in differences in mortality to 4 weeks of age and during the stressing procedure and blood collection at 4 weeks of age (Tables 4 and 5). Table 4 was based on analysis of data of the RBC3, H, and L lines over four generations, while the data in Table 5 were based on six generations of selection in the H and L lines. Mortality to 4 weeks of age was consistently reduced in the L line relative to the corresponding randombred control, but there was no time trend from the third (1966) through the sixth (1969) generations of selection. Mortality in the H line and randombred control was similar during this time (Table 4). Mortality resulting from the stress and bleeding procedures at 4 weeks of age was decreased slightly in the L line and increase slightly Ln the H line, with no time trends in either line. There was no significant time trend for the difference between the H and L lines over six generations of selection (Table 5). The interaction between lines and years was significant for mortality to 4 weeks of age and mortality during the cold stress and blood collection at 4 weeks of age* Genetic increases in egg production and decreases in blood corticosterone level after cold stress apparently resulted in significant changes in mortality in early generations of selection when data were not available for analysis. Both of these lines exhibiting significant changes in mortality were also superior to their respective randoinred control lines in egg production. However, since mortality increased in one line (E) and decreased in the other (L), it is unlikely that genetic changes in egg production were the cause of the differences in mortality for both lines. -83-

87 TABLE 4.--The Influence of Selection for High (H Line) and Low (L Line) Response to Cold Stress on Mortality to 4 Weeks of Age and on Mortality During Stress Procedures at 4 Weeks of Age. Line Average Mortal ity to 4 Weeks of Age (%) RBC3* L* Hi- Average of All Lines t Mortality During Stress Procedures (%) RBC3* L f Ht Average of All Lines f *Randombred control strain. ^Expressed as deviation from the randombred control. ^Differences between lines and years and their interaction were highly significant. TABLE 5. The Influence of Selection for High and Low Response to Cold Stress on Mortality to 4 Weeks of Age and on Mortality Due to Stress Procedures at 4 Weeks of Age. Line Average L 6.0 Mortal ity to Weeks of Age ill H Average for Both Lines * I 11.4 Mortality Due to Stress Procedures (%} H Average for Both Lines * ^Differences between lines and years and their interactions were highly significant. -84-

88 Females with higher egg production usually produce more poults within a given period of time than lower producing dams. Since egg production in the E line was greatly increased by selection, E line dams produce a larger number of poults than RBC1 hens. The association between family size and poult mortality is presented in Table 7. Those families having only a few poults (1 to 5) or a large number (16-20) of poults in a 4-week hatching period had higher mortality than those families with an intermediate number. This was very pronounced for the E line, but similar trends were evident for RBC1 and all lines combined. The E line had a larger percentage of high producing dams than RBC1. This would also contribute to the line differences in mortality observed. Nevertheless, the mortality level of families of comparable size was higher in the E line than in the RBC1 line. TABLE 6.--Correlation Coefficients Between Egg Weight of Dams and Mortality of Their Offspring. Medium Weight Lines RBC1 E S Within lines and years Large Weight Lines RBC2 F I C B Within lines and years Both Groups Within lines and years Both Years (34) f (69) (35) (31) (33) (34) 0.06 (70) (35) (35) (34) (33) (35) 0.14 (32) (242) (268) (510) P<.05. tnumbers in parentheses was based. = number of dams on which the correlation coefficient Since mortality level did not change in the I line (Table 2) which was selected for both increased egg production and increased body weight,, it was believed that some factor other than total egg production was changing in the E line which was resulting in increased mortality, McCartney, Nestor, and Harvey (1968, Poultry Sci., 47: ) and Nestor (1971, Poultry Sci., 50: ) observed that average egg weight declined in the E line relative to the RBC1 line. Correlation coefficients between dam's egg weight and mortality of her offspring are presented in Table 6. The correlation coefficients within lines and years were small and nonsignificant, but the majority was negative. The correlation coefficient (-0.10) within types (medium or large weight hens), lines, and years was significant but small in magnitude. This indicates that as the dam f s egg weight declines, an increase in mortality of her offspring is expected. However, only 1% of the variation in early poult mortality was accounted for by variation in dam's egg weight. -85-

89 TABLE /.--The Association of Family Size and Poult Mortality to 8 Weeks of Age. Family Size (No.) No. of Poults RBCl Percent Mortality No. of Poults Egg Percent Mortality All No. of Poults Lines Percent Mortality Small family size can be caused by poor egg production, fertility, or hatchability of fertile eggs, or a combination of these. The increased mortality occurring in these families may be an expression after hatching of decreased viability which was evident prior to hatching. However, Funk (1950, Poultry Sci», 29: 64-66) failed to observe i significant correlation between percent hatch of all eggs set and percent mortality to 8 weeks of age, Hens producing a larger number of poults during the 4-week hatching period have smaller egg weight. The average egg weights for the t line hens producing 1 to 5, 6 to 10, 11 to 15, and 16 to 20 poults were 77.5, 76.8, 76.5, and 75.6 grams, respectively. It is unlikely that differences in egg weight account for all of the increasec mortality in the offspring of high producing dams. Since Nestor, Brown, and Weaver (1972, Poultry Sci. s 51: J observed a negative correlation coefficient between egg production and albumen height in turkeys, and since Skala and Swanson (1952, Poultry Sci., 41: ) observed differences in percent solids, percent total nitrogen, and sodium content between high and low quality chicken eggs, it is possible that nutrient content of the eggs might influence the livability after hatching and that higher producing dams do not produce eggs with as great a nutrient supply as lower producing dams. The length of the Incubation period of turkey eggs is highly variable relative to other species of birds. The length of the incubation period of turkey eggs increases with increases in egg weight. Since egg weight was reduced in the E line, the length of the incubation period for this line would be expected to be reduced. Thus, when poults were removed from the hatcher after 28 days of incubation, some poults may have remained in the hatcher for 5 or more days after hatching, and this might be a cause of pouit starvoouts. Removal of 13 line poults from the hatcher and placing on feed and water prior to the normal time of removal (28 days) did not reduce mortality even though a majority of the poults hatched on the 26th and 27th days of incubation (Table 8). Mortality was higher in the group removed after 29 days of incubation than in the other groups, but the differences were not statistically significant. Genetically reducing the blood level of corticosterone after cold stress in the L line may have modified this line's susceptibility to certain types of infections, thus reducing mortality. Gross and Colmano (1971, Poultry Sci., 50: ) observed that a line of chickens selected for a low corticosterone level after social stress was more resistant to Marek's disease and Mycoplasma gallisepticum infection, but more susceptible to E. coli infection than a line selected for high plasma corticosterone after social stress. -86-

90 TABLE 8.--The Influence of Time of Removal from the Incubator on Poult Mortality in the E Line. Group Day of Removal from Incubator No. of Poults Percent Mortality to 3 Weeks of Age Control 28th day Treated 26th day th day th day th day th day The frequent occurrence of line by year (generation) interaction without time trends occurring in various comparisons with line differences (Tables 1, 4, and 5) and without line differences (Tables 2 and 3) indicates that, perhaps under certain environmental conditions, genetic differences in resistance to death are expressed to a greater degree, thus resulting in genotype-environmental interactions. Also, the mortality may have resulted from different causes in the various years. Lush, Lamoreux, and Hazel (1948, Poultry Sci., 27: ), Robertson and Lerner (1949, Genetics, 34: ), and Hale (1959, Poultry Sci., 38: ) observed that the heritability of viability in chickens differed slightly when the cause of mortality was identified. In the present experiment, no attempt was made to identify the cause of mortality. No specific disease outbreak resulting in high mortality prior to 8 weeks of age occurred during this study. Summary Mortality during the early growing period of turkeys was increased by genetic increases in egg production in a medium weight line (E), but genetic increases in both 16-week body weight and egg production in a large weight line did not influence poult mortality. Since mortality declined and egg production increased in a line selected for low plasma corticosterone after cold stress at 4 weeks of age, it was believed that changes in some factor other than total egg production resulted in the increased mortality of the E line. A part of this increased mortality was attributed to decreased egg weight (1% of the variation). Hens producing a small (1 to 5) or large (16-20) number of poults in a 4-week hatching period had increased mortality. Since the E line had a higher percentage of high producing dams than the corresponding randombred control line, this factor contributed to the difference in mortality observed between the two lines. Genetic increases in 16-week body weight and semen yield did not influence mortality. -87-

91 THE INFLUENCE OF DIPPING EGGS IN GENTAMICIN ON GROWTH AND LEG WEAKNESS OF TURKEYS Karl E. Nestor and Y. M. Saif Departments of Poultry Science and Veterinary Science Treatment of hatching turkey eggs by dipping in antibiotic solutions to control certain egg-transmitted pathogens is a widely used practice. It has been established (Saif and Nestor, Poultry Sci», 52: , 1973) that dipping eggs in a solution of gentamicin sulfate (1,000 p.p.m.) was not harmful to fertility or hatchability of fertile eggs. There has been little evidence published as to the effect of dipping hatching eggs on the subsequent performance during the growing period. Eggs from a group of colored turkeys were dipped on alternate days in a solution containing 1,000 p.p.m. of gentamicin sulfate. The eggs were heated for 4 hours at approximately 100 F. followed by dipping in the gentamicin solution maintained at 40 F. The eggs from these hens on the remaining days were not dipped and served as a control. The dipped and nondipped eggs were incubated and hatched separately. The offspring from the dipped eggs were wingbanded first and placed in scpar.itc pens from the offspring from nondipped control eggs. However, all birds were housed in the same building. Body weights were recorded at 8, 16, and 24 weeks of age. The frequency of bad legs was recorded. The results are presented in Table 1. The percent mortality to 24-weeks of age of birds with normal legs was similar in offspring from dipped and nondipped eggs, The frequency of bad legs was nine times greater in the nondipped controls than in offspring from the dipped eggs. The frequency of bad legs was 5.5 o in the controls, vs. 0.6% in the offspring from dipped eggs. Thus, if the individuals with weak legs which were removed for autopsy were included in mortality, the percent mortality was 8.9 and 13.5%, respectively, for the dipped and control groups. The male offspring from dipped eggs grew slightly better than those from control eggs. However, only the difference at 8 weeks of age was statistically significant. The body weight of female offspring from dipped eggs was slightly heavier at 8 and 16 weeks of age than offspring fron nondipped eggs. At 24 weeks of age, these two groups had similar body weights. It is noteworthy that the variance in 24-week body weight of females was much greater in the control group than in the dipped group. Since the comparisons were made within hatch and family, this difference becomes important. No differences were noted for 24-week body weight in males. However, the vari ance in the control group was probably reduced since many of the weak-legged individuals were removed prior to 24 weeks of age for autopsy. Since the frequency of leg weakness was higher in males than in females and there was more leg weakness in the control group, this removal would bias the comparison of the variation in body weight for the males. Summary A comparison of the performance of offspring from eggs dipped in a gentamicin solution containing 1,000 p.p.m. and from nondipped control eggs was made. Dipping eggs in gentamicin resulted in less leg weakness and slightly larger body weight at 8 and 16 weeks of age, with no effect on mortality of individuals with normal legs during the growing period. There was evidence that the variance in body weight relative to the mean was greater in offspring from nondipped eggs..88-

92

93 CHANGES IN FREE FATTY ACID AND NEUTRAL LIPID CONCENTRATIONS WITH REPRODUCTIVE STATE IN TURKEY HENS Wayne L. Bacon and Margery A, Musser Department of Poultry Science In chickens, the concentration of plasma-free fatty acids (FFAj was reported to increase from about 0.3 u equiv./ml. in immature hens to about 1*0 to 2.0 p equiv./ml. in hens at time of first egg. Similar increases were also reported for plasma total lipids (Heald and Badman, 1963, Biophys, Biochirn. Acta., 70: ). A similar report for turkey hens by Bajpayee and Brown has recently appeared (Poultry Sci., 51: , 1972). They reported increases in plasma FFA concentrations from 1,14 ± 0.20 y equiv./ml. in immature hens to 2.15 ± 0.11 p equiv./ml. in hens at the time of first egg. Neutral lipid (NL) levels were not reported in this study. In both of these studies, phospholipids (PL) were not removed from the extracts prior to determination of FFA concentration* It is known that large concentrations of PL may be a source of error in overestimating FFA concentrations. Bacon and Musser (OARDC, Turkey Research--1972, Research Summary 64, Dec. 1972) reported that the concentration of FFA in laying turkey hens was much less after PL removal. Subsequently, FFA and NL levels have been followed in female turkeys which were immature, stimulated but not laying, laying, and broody. These data are presented in this report. All sampling times are referred to as days from first egg. The procedure used to determine plasma FFA concentration is described by Bacon and Musser (OARDC, Research Summary 64, 1972). The concentration of plasma NL was determined by weighing the extracted lipids after PL removal as described In the above reference. Radioactive palmitic acid was added to the plasma samples and its recovery was used to correct the FFA concentrations for losses during analysis. The NL concentrations were not corrected for losses. Figure 1 shows the results of the first trial in which six hens were sampled between January and March. The hens were about 32 weeks of age arid had been maintained under 3 hours of light per day for 8 weeks prior to being exposed to stimulatory lighting of 14 hours of light per day (14L;IOD). Egg production began about 20 days after the change in lighting. The concentration of FFA was significantly greater on day -6 than on day 15. The other FFA levels were not significantly different. Standard errors of the means were not plotted, but were between 10 and 15% of the means. On day -7, the hens were moved from a floor pen to a room with individual bird cages, The relatively high concentration of FFA at day -6 might be associated with the stress of this change in housing. Figure 1 also shows NL levels during the initiation of egg production. Significant increases in NL began about 15 days prior to first egg, or about 5 days after exposure to 14L:10D. After first egg, no significant changes in NL concentration occurred as long as egg production was maintained. A correlation coefficient between FFA and log^q NL concentration was calculated with sampling day effects absorbed. The log conversion was used to normalize sample mean variance. This correlation was and was not significant. Figure 2 shows the results of a second trial in which six other birds were sampled. Free fatty acid concentration did not change significantly in relation to reproductive state. However, there were significant differences between hens. Neutral lipids did increase significantly between the start of stimulatory lighting and laying -90-

94

95 TJTjlh^^ TABLE 1. FFA and Neutral Lipid Concentrations Experiment 1 Experiment 2 Neutral Neutral ji^^ Laying 24.19± H * } 1.2 Broody 18.23± ± «^ 2^^ *No. of observations. ^y equiv./ml. ± s.e. **mg./ml. i s.e. of the first egg. In this trial, the hens were housed in individual laying cages during the entire experiment. A significant correlation of.431 between log^q NL and FFA concentration was observed when the sampling day effects were absorbed, and.301 when the individual hen effects were absorbed. The partial correlation was calculated removing the effect of feed intake, which was significantly correlated with I FA concentrations. This coefficient was.268 and was significant (P<.01). This trial was conducted during the warm weather months of July to September, which may ha\e affected the FFA concentrations. Means of NL and FFA concentrations in hens of broody and laying states are shown in Table 1. A broody period is defined as a period of 5 or more days during which no eggs are laid. Free fatty acid concentrations were not significantly different between reproductive states within the experiment. However, NL concentrations dropped drastically in the hens when they became broody. The concentrations of NL in the broody hens are comparable to those found in these same hens in the immature state. If the partial correlation coefficient of.268 mentioned above was meaningful, a drop in FFA as well as in NL would have occurred. The data in Table 1 thus cast doubt on the biological significance of this partial correlation coefficient, Assuming a mean value of 0.22 y equiv, FFA per ml., the total plasma pool of FFA would be 77 y equiv./hen. The turnover time of FFA in turkey hens is about 0.85 minute (unpublished data). The amount of FFA cleared as stearate would then be about 35 g./day. Using the concentration of 16.5 mg. NL per ml. and a turnover time of 0.5 day (OARDC Research Summary 64, 1972), 13.7 g, per day would be cleared. At 7Q l «production, which was the level observed in these two trials (excluding the broody periods), about 6.0 g. of NL are produced per day as egg yolk. Thus, all of the yolk NL could be accounted for by only 17% of the plasma FFA cleared per day, or by only 43% of the plasma NL cleared per day. This indicates that the laying turkey hen is producing more NL than is needed for egg yolk formation, and that some of this NL is entering metabolic pathways other than the one terminating in egg yolk production. This excess NL could be available for maintenance energy or for fat deposition. Summary 1. The concentrations of plasma free fatty acids (FFA) and neutral lipids (NL) were determined in immature, laying, and broody turkey hens. No sustained differences in FFA levels were associated with the different reproductive states. A mean value of y equiv./ml. plasma was found for most hens, 2. Neutral lipid concentrations increased from mg./ml. to mg./ml. during the transition from the sexually immature to the laying state. During broody periods, the NL concentration dropped to a level comparable to that of immature hens. -92-

96 ESTROGENS AND NEUTRAL LIP1DS IN TURKEY HENS AT DIFFERENT REPRODUCTIVE STATES Wayne L. Bacon, Keith I. Brown, Margery A e Musser, and David W. Long Department of Poultry Science Introduction There is a need to determine more completely the physiological changes occurring at sexual maturity. These changes are initiated by the pituitary gland, which in turn affects the ovary f s hormone output, which in turn affects several organs concerned with reproduction. Some of these changes, which are influenced by the ovarian estrogenic hormone output, include oviduct development and changes in blood plasma lipoproteins. The present study was undertaken to follow changes in plasma levels of the three major estrogens and neutral lipids (NL) in female turkeys. Samples were analyzed from hens when they were sexually immature, when they were stimulated but pre-laying, when they were laying, and when broody. Methods Neutral Lipids: Neutral lipids were determined gravimetrically on lipid extracts of 1 ml. of plasma after phospholipids were removed. Lipids were extracted using iso-propanol and iso-octane under acid conditions. The iso-octane phase was taken to dryness, and the phospholipids removed by silica gel column chromatography with hexane: diethyl ether, 1:1. This procedure has been verified in this laboratory to give greater than 95% recovery of triglycerides, cholesterol, cholesterol ester, and free fatty acids. Estrogens: Estrogens were determined using a radioimmunoassay technique. Labeled estrone, estradiol, and estriol were added prior to extraction with diethyl ether. This allowed corrections for losses during analysis. The extracts were dried and then separated chromatographically into neutral lipids, estrone, estradiol, and estriol fractions using Sephadex LH The eluting solvent was iso-octane: benzene:methanol, 62:20:18. After drying, assay buffer was added to each vial, duplicate samples were withdrawn for radioimmunoassay, and a single aliquot was counted for percent recovery calculation. Recoveries ranged from 75 to 85% for the various assays. After addition of antiserum and overnight incubation at 4 C., charcoal suspension was added to separate free from bound estrogen. The supernatant was decanted and counted in a liquid scintillation counter. The picogram values of the unknowns were found from standard curves of CPM bound vs. picograms of each estrogen, and were corrected for percent recovery and converted to pg./ml. plasma. Birds: An experiment was conducted in which neutral lipids and estrogens were followed in six large white turkey hens housed in cages. The six hens were about 24 weeks old when housed and were maintained under 6 hours of light per 24 hours (6L:18D) until 30 weeks old. They were then exposed to 14 hours of light per 24 hours (14L:10D) Light intensity was about 100 lumens. Blood samples were drawn at various times prior to changing the lights from 6 hours/day to 14 hours and during the first 72 days of egg production. All blood samples were drawn either 1 hour into the light period (6L:18D) or 6 hours into the light period (14L:10D). -93-

97

98

99 TABLE 2. Half Lives * Concentrations, Clearance Times, and Clearance Rates of Estrone, Estradiol, and Estriol in Laying Turkey Hens (Preliminary Data). Half Lives Cage Estrone Estradiol Estriol I ± S.D. 38 Turnover Time (min.) 60 Concentration (min. ) 27 Clearance Rate (pg./ml./min.) ± ± two nonlaying states mentioned above. Estrone, estradiol, and estriol levels were about the same in the immature and broody states, but the highest levels of all three occurred about the time of the first egg (Figure 1). During the laying period, estriol was maintained at a level about 4.8-fold greater than that found in the immature hens, while estrone was about 0.8-fold greater and estradiol about 0.3-fold greater. It is apparent that more than the concentrations of the estrogens and NL need to be known to understand the biological significance of changes in their concentrations. Table 2 shows preliminary data on half lives and calculated clearance rates for the three estrogens. Such data for NL have been reported (OARDC Research Summary 64, 1972). Notice that the turnover time for estriol is 31 minutes in comparison to 60 and 54 minutes for estrone and estradiol, respectively. Clearance rates are 3.58 pg./ml./min. for estriol vs..45 and.37 pg./ml./min. for estrone and estradiol. Thus, only about 25% of the estriol being cleared could be accounted for as if it were derived only from circulating estrone and estradiol. Clearly, estriol must be directly synthesized by the turkey ovary or have other metabolic sources besides estradiol and estrone. Summary 1. In laying hens, estriol levels increased more than those of estradiol or estrone. 2. Neutral lipid levels are most highly correlated with estriol. 3. The clearance rate of estriol is not accounted for by direct conversion of circulating estrone and estradiol. -96-

100 CALCIUM AND PHOSPHORUS CONCENTRATIONS IN THE PLASMA OF TURKEY HENS AT VARIOUS REPRODUCTIVE STATES Wayne Bacon and Margery Musser Department of Poultry Science The mineral composition of egg shells is almost entirely crystalline calcium carbonate. The calcium (Ca) is removed from the blood plasma by the shell gland during shell formation. It has been known for some time that there is an increased level of Ca and phosphorus (P) in the plasma of the laying chicken and turkey hen. The levels of Ca and P in the blood plasma of turkey hens at various reproductive states are reported here. The total Ca (Cax) was divided into two subclasses: ultrafilterable (Cay) and bound (Cag); while the total P (P^) was divided into three subclasses: lipid (PJ), protein (P p ), and inorganic (P^). Samples were collected at various times in six hens. The first collections were when the hens were sexually immature. The hens were then given stimulatory light* and samples were collected when they were stimulated but prelaying. Samples were also collected when the hens were laying and during subsequent broody periods, A broody period is defined as a period of 5 or more days during which no eggs are laid. The hens were the same as those used in the determination of levels of estrogens and neutral lipids (p. 93), and the samples were aliquots of the samples collected for the above determinations. The husbandry of the birds is given in the above report was determined after wet ashing a small sample of the plasma. Cay was determined on ultrafiltrates of plasma. Ultrafiltration was conducted as described by Musser et al. (Gen. Comp. Biochem. and Physiol., in press). All Ca measurements on these samples were by atomic absorption spectrometry. Cag was calculated by difference. Phosphorus determinations were conducted by the colorimeteric method described by Bacon and Musser (Poultry Sci., 53: 1167, 1974) for P t, PI, and P^, with P p being calculated by difference. The results for the Ca determinations during the immature state, stimulated but prelaying state, and the initial 23 days of egg production are given in Figure 1. Notice that Caj started to increase about 9 days before the first egg, and that the increase occurred only in the Cag fraction. The increase in Cag was from about 2-3 mg./loo ml. plasma to mg./loo ml. plasma just prior to first egg. This increased level was maintained in all hens during the first 17 days of egg laying. Standard errors are not given, but were about 5% of the mean values. The Cay analysis for day 23 was not successful, and so values for Cay and Cag for this day are not available. The results of the P analyses are given in Figure 2. Total P increased significantly during the stimulated but prelaying period. This increase is attributed to increases in Pj and Pp. No increase occurred in the level of PI. Egg yolk, rich in phospholipids and phosphoproteins, is not formed in the ovary, but in the liver. These phospholipids and phosphoproteins are synthesized only when estrogens are present in adequate amounts. The increases in Pj, as well as neutral lipids, are closely correlated with the increases in estrogens (p. 93). The increased levels of Cag in the stimulated and laying hens can be explained by the increase in Pp. Phosphorus is bound to protein and at physiological ph has an ionic charge. This ionic phosphate then binds calcium. -97-

101

102 TABLE l.--the Distribution of the Various Forms of Phosphorus and Calcium in Immature, Laying, andbroody Turkey Hens. Reproductive State Phosphorus (ug./ml. plasma) Total Inorganic Lipid Protein Immature ± 4* ± 5 63±3 Laying ±20 52±2 232±12 198±8 Broody ±13 45±3 55± 9 78±5 Total Calcium (mg./iooml. plasma] Ultrafilterable Bound Immature ±.l 6.4±.l 2.1±.l Laying ±.8 5.9±.l 18.0±.8 Broody ±.3 6.7±.5 ± standard error. The levels of the various forms of plasma calcium and phosphorus in the birds when they were sexually immature, laying, and broody is summarized in Table 1. The levels of Pj_ and Cay were about the same regardless of the reproductive state of the birds. The level of P p was lowest in the immature bird and highest in the birds when they were laying. This was also the case for Gag. For Pj, the lowest levels occurred in the broody birds. In the case of phosphorus, the determination of P t and P-^ can be accurately made. However, for Pp and P^, problems may arise in the analysis. P] is measured directly after extraction with organic solvents (ethanolrdiethyl ether, 3:1). If the extraction is not complete, Pj will be underestimated and Pp will be overestimated. Also, the types and quantities of lipoproteins with which the plasma PI is associated change with the reproductive state of the bird (Bacon and Musser, 1974 Poultry Sci., 53: 1167). The relative efficiency of extraction of the PI from the various lipoproteins may be variable. The data for Pp and PI must therefore be considered as preliminary in nature and in need of confirmation. Notice that both Cag and P p were greater in the broody birds than in the immature birds. When a bird becomes broody, the ova developing in the ovary will be reabsorbed and might be a source of the increased plasma Pp. Also, when broody birds are given a treatment to terminate broodiness, they will begin laying about 8 to 10 days later (OARDC Research Summary 47, 1970), while immature hens given stimulatory light normally require at least 3 weeks before they begin to lay eggs. The most drastic changes in the levels of CaB, PI, and P p occur with the longterm changes in reproductive state of the female turkey. The levels of Gag in the plasma in the immature vs. broody bird are not identical as shown by the increased levels of Gag in the latter. The level of Pp was also greater in the broody than in the immature bird, while Pj is lower in the broody than the immature bird. Changes in relative amounts of Pj and P p may indicate changes in types and amounts of lipoproteins and phosphoproteins in immature vs. broody hens. ^99.

103 1. The levels of plasma inorganic phosphorus and ultrafilterable calcium did not change with the reproductive state. 2. The level of bound calcium was increased in the laying and broody states in comparison to the immature state. The level in the laying state was greater than in the broody state. 3. The levels of protein and lipid phosphorus were greater in the laying than the immature state. The level of protein phosphorus was greater in the broody than in the immature state, while the level of lipid phosphorus was greater in the immature than the broody state

104 VARIATION IN PLASMA CALCIUM DURING EGG FORMATION IN THE TURKEY Margery A. Musser, Wayne L. Bacon 9 and Sherman P. Touchburn 1 Department of Poultry Science It has been determined that the calcium for shell calcification is obtained from the plasma calcium rather than being stored in the shell gland. Since the rate of deposition during rapid calcification is about 200 mg. Ca/hour in the turkey, there must be just as high a replacement rate during calcification in order to maintain the level of plasma Ca present in a hen prior to calcification. In order to determine if the plasma calcium changed significantly during the process of shell formation, the following experiment was carried out. Twelve caged hens 44 ± 1 weeks of age, which had laid an egg in the morning, were selected for sampling. At 4-hour intervals, 2.5 ml. of blood were removed from a brachial vein of each hen and placed in a heparinized tube. Samples were taken only at the first and last of the six collection times on four additional hens and showed no difference which could be attributed to the handling of the experimental birds. Lights were on 14 hours per day. Only collections four and five were taken during the light period. Palpatation was carried out at each sampling time until a hardshelled egg was detected. Time of lay was recorded hourly. All but one of the 12 birds laid an egg by 5 p.m. on the second day. Plasma was directly wet ashed for determination of total calcium. Plasma for determination of diffusible calcium was equilibrated under 5% C0 2 in N 2 and centrifuged through a filter to obtain an ultrafiltrate which was then wet ashed. Duplicate for each determination were diluted and read by atomic absorption spectrophotometry against suitable standards. Protein-bound calcium was obtained by difference. There were significant (P<Q.Q5) differences due to time of collection (Table 2). At the time of collection 1, eggs were in the upper oviduct or had recently entered the shell gland. By the time of collection 4 when the eggs had been in the shell gland for 10 hours or more, bound calcium had decreased significantly compared with collections 1 and 2. The lowest values for the bound form were at collection 5, at which time four of the eight birds had laid and the rest had nearly completed calcification. Total calcium followed the same pattern due to these changes in bound calcium. Diffusible calcium showed little variation, with no consistent pattern. If the values from samples taken during calcification are grouped by presumed state during the 20-hour period, there appear to be no significant differences. These groups were established by use of both palpation and time of lay data. The value for total calcium at collection 1 was, in every case, higher than the value at collection 5, which was around the mean time of lay for all of the birds. Summary 1. Hens which were on egg laying cycles no more than 5 hours apart were sampled for plasma calcium analysis every 4 hours for a 24-hour period. 2. Although the plasma calcium data analyzed by collection time appear to show a lowering of bound calcium as shell calcification proceeds, the data when analyzed by actual stage of calcification show no significant changes. Canada. 1 Present address: Macdonald College, Ste. Anne de Bellevue, Province of Quebec, -101-

105 TABLE 1.--Plasma Calcium Concentration of Laying Turkey Hens Throughout a 24-hour Period. Collection Number and Time p.m. 9 p.m. 1 a.m. 9 a.m. 5 1 p.m. 6 5 p.m. Diffusible Calcium* 5.3±0.1 bt 5.4±0.2 ab 5.6±0.2 ab 5.2±0.2 b 5.4±0.1 b 5.8±0. l a Bound Calcium 19.7±0.8 a 19.7±1.0 a 17.3±l.l ab.u.c.x..u i 15.0±0.8 b 15.4iO. u Total k k k k Calcium* 25.0±Q.8 a 25,0±l.Q a 22.9±l.l ab 21.4±l.l b 2Q.4±CK8 b 21.2±Q.8 b Diffusible Calcium as Percent of,,, Total 21.3±Q.9 b 21.9±1.2 b 24.3±1.3 abc 24,6±1.3 abc 26.7±1.0 abc 27.9±Q.9 a *Least squares mean in mg% ± standard error. tmeans with different superscript letters are significantly (P<Q,Q5) different compared by D test (Snedecor, 1956). TABLE 2.--Plasma Calcium Concentrations of Laying Turkey Hens in Different Stages of Shell Formation. Condition of Hen n Calcium Level Diffusible in mg./loo ml. Plasma Bound Total No Egg in Shell Gland ±0.1* 17.6± ±0.7 Egg in Shell Gland ± ± Stage of Shell Formation (Hr. Before Lay) Before Rapid Calcification (18-20) ± ± ±0.6 Early Calcification (14-16) Mid-calcification (6-12) 11 n 5.3± ± U ± ± ±0.8 Late Calcification (2-4) ± ± ±0.7 Least squares means ± standard error

106 WITHDRAWAL OF PLASMA CALCIUM BY THE SHELL GLAND OF THE TURKEY Margery A. Musser, Wayne L. Bacon* Sherman P. Touchburn, 1 Karl E. Nestor Department of Poultry Science In order to determine what form of blood calcium (Ca) is removed from the plasma for shell calcification, it was desirable to estimate protein-bound and diffusible (mainly ionic) Ca in the blood vessels supplying and draining the shell gland. The blood supplying the shell gland has essentially the same Ca content as that in the general circulation of the bird. The blood leaving the shell gland in the uterine vein would not yet have passed an organ from which it could gain additional Ca. The sampling sites chosen to estimate Ca removal by the shell gland were the wing vein and the uterine vein. A series of 27 hens of 52 to 56 weeks of age were sampled from the two sites (the brachial or wing vein and the uterine vein) within 10 minutes of each other. A local anesthetic was injected In the muscles which were cut to expose the shell gland before either sample. After sampling, hens were killed so the entire oviduct could be inspected. Six hens had no egg in the oviduct (condition 1), six had an egg in the upper part of the oviduct (condition 2), and 15 had an egg in the shell gland (condition 3). Several determinations showed no significant difference in Ca levels between plasma obtained from the brachial vein and the heart. Hodges (Comp. Biochein. Physiol, 28: , 1969) found calcium levels to be the same in the systemic arterial and venous plasma taken at nearly all points of the egg laying cycle. The fresh plasma samples were protected under a 5% C0 2 atmosphere while separatioi of diffusible calcium was accomplished (see p. 101). Portions of ultrafiltrate and of whole plasma were wet ashed and analyzed by atomic absorption. Bound calcium was calculated by difference. This fraction Is bound to plasma proteins. In all cases, levels of calcium were lower in uterine vein samples (Table 1). The overall means of diffusible and of total calcium from the two sites of sampling were significantly different (1% level). Diffusible calcium expressed as a percentage of total calcium was also significantly different. This had previously been reported as relatively unchanging during calcification. Although that appears to be true for brachial vein samples (see p. 104, Table 1), it is not true for uterine vein samples. Hens forming an egg showed a significant difference in withdrawal of diffusible Ca, while those with no egg in the reproductive tract did not. There was a significant difference in total Ca levels between sampling sites only in those hens with an egg in the shell gland. The level of uterine vein plasma Ca is nearly 23% below that of the brachial vein. There was the same absolute difference in both diffusible and bound fractions in hens of all three conditions, but the percent difference for diffusible was always much greater. and Canada. 1 Present address: Macdonald College, Ste. Anne de Bellevue, Province of Quebec, The data from hens with an egg in the shell gland were analyzed according to condition of the egg present at the time of sampling as determined by visual inspec

107 TABLE 1.--Mean Plasma Calcium Levels in Brachial and Uterine Veins of Laying Turkey Hens in Three Different tages of the Egg Formation Cycle. 3rtn of alcium Condition I 2 3 Hen with Egg in Hen with Lgg in Over All Conditions Hen with No EJCL U J 1T Oviduct ^htll Gland Brachial Uterine Brachial Uterine Brachial Uterine brachidl Uterine iffusible 1 " oundr otalt iffusible as of Total 5.9± ** I 0.3 b.9^ U j* *U 2* 15.2± ± * * ± ** , * * 28.7± ±1.3* * * * "^"Least squares mean in mg.% f standard error. *Brachial-uterine difference significant (P^O.05). **Brachial-uterine difference significant (P^O.Ol). TABLE 2.--Mean Differences Between Levels of Calcium in Brachial and Uterine Samples When an Egg is Present in the Shell Gland. Condition of Egg in Shell Gland n Diffusible Bound Total Unpl umped Plumped Rapid Calcification Shell Nearly Completed ±0.4* 3.6± ± ±0.5 *Mean difference in mg% ± standard error. 0.6± ± ± ± ± ± ±1.3 TABLE 3.--Correlation Coefficients Between Plasma Calcium Measures in the Laying Turkey. Measures of Calcium Diffusible Bound Total Bound Total NS Diffusible as Percent of Total Significant (P<0.01) unless marked NS. -104*

108 tion. Table 2 shows the mean difference between sampling sites. The largest differences in total Ca were in hens with plumped and rapidly calcifying eggs. The largest difference in bound Ca and percent difference in bound Ca (difference in bound/difference in total x 100) was in the hens rapidly calcifying. This is consistent with the theory that in this period of a highest rate of Ca withdrawal from the blood^ Ca from the protein carrier is released, while under conditions of lower rate of transport the diffusible fraction is almost solely drawn from. Various workers have found uterine venous blood to have a lower ph than peripheral blood during calcification. This would facilitate the release of Ca from the protein carrier. Correlation coefficients between the measures of plasma Ca in this experiment were all highly significant except between total and diffusible Ca (Table 3). The highest positive correlations were between bound and total Ca and between diffusible and diffusible as a percent of total Ca. Summary 1. Maximum withdrawal of plasma calcium by the shell gland occurred in hens with plumped and rapidly calcifying eggs in the shell gland. There was some withdrawal even by laying hens with no egg in the reproductive tract. 2. The largest percentage withdrawal was of diffusible calcium, although this made up a smaller percent of the total withdrawal during rapid calcification than did bound calcium. 3. Bound calcium had a high correlation with total calcium and a small but negative correlation with diffusible calcium

109 CALCIUM-BINDING PROTEIN IN THE TURKEY HEN Margery A, Musser, Karl E. Nestor, and Wayne L. Bacon Department of Poultry Science It has been found that in the chicken a metabolite of vitamin D 3 causes the formation of a specific calcium binding protein (CaBPj in the mucosa of the duodenum and in the shell gland (uterus) of the laying bird. This protein, identified by its ability to bind calcium, is also found in the turkey. The hens used for this study were about 44 weeks of age when sampling was started. Data are presented here from samples of two hens under stimulatory lighting, eight hens under stimulatory lighting but not yet in egg production, four hens calcifying their first egg, and two hens which were in production for 1 week. The duodenum and the shell gland from each bird were removed and placed in an ice bath. The cells lining the interior of each organ were scraped off with a glass slide, weighed, diluted fivefold with 4 C. Tris buffer of ph 7.4, and homogenized. The yield was about 0.5 g. from a uterus and 3-4 g. from a small intestine. The homogenate was centrifuged and the supernatant fluid was heat treated at 60 C. for 10 minutes. This was centrifuged and the supernatant was assayed for Ca binding activity according to the method of Wasserman et al. (J. Biol. Chem., 14: , 1968) in which the protein competes with a resin for radioactive Ca ( 45 Ca). All comparisons were of percentages of 45 Ca bound in the supernatant. In each assay, 0.2 ml. supernatant containing about 2 mg. of protein was used. Molecular weight estimation of turkey duodenal CaBP was carried out on a Sephadex G-100 gel filtration column at 4 C. The buffer was the same as that used in the assays with the addition of 1 mm 2-mercaptoethanol. To eliminate some of the other proteins in the preparation and concentrate the CaBP, the supernatant fluid as prepared above was treated with 0.48 g. of ammonium sulfate per ml., allowed to stand at 4 C., and then centrifuged. The supernatant was dialyzed against buffer and then concentrated in an Amicon stirred cell with a Diaflo membrane type UM 10. This concentrated mixture of proteins was loaded on the column. Fractions of 4 ml. were collected. Assay of Ca binding activity was done on alternate fractions which indicated the presence of protein by absorbance of UV light (280 nm). Comparative values for duodenal and uterine calcium binding activity are shown in Figure 1. The level of activity in the duodenum is greatly increased by 9 days before first egg in comparison with a non-stimulated hen. However, there is virtually no activity in the uterine material until the first egg is being calcified. Uterine activity continues to rise in the first week of lay. Uterine samples were not obtained from hens with an undeveloped uterus. This pattern of change in calcium binding activity agrees with that found by Bar and Hurwitz (Comp. Biochem. Physiol., 45A: , 1973) in the chicken. The elution volume/void volume ratios of CaBP and globular proteins used as standards are presented in Table 1. All of these were from the same column. The molecular weight of turkey CaBP calculated from its elution volume is 31,000. This is based on the regression equation calculated from the data presented in Table 1. The molecular weight with 95% confidence lies between 28,400 and 33,900. The molecular weight reported by Wasserman et al. (J. Biol. Chem., 243: , 1968) for chicken intestinal CaBP is 28,

110

111 TABLE 1. Gel Filtration of Sephadex G-100 Column. Protein MW Log 1Q MW Ve/Vo Value 1 Value 2 Bovine Serum Albumin 65, Ovalbumin 45, e-lactoglobulin 37, Trypsin 24, a-lactalbumin 15, Cytochrome c 12, CaBP 31,000* Calculated from the regression equation of all the Ve/Vo values vs. log MW of the above standards

112 A STUDY OF NORMAL AND ABNORMAL SHELL FORMERS AMONG CAGED BREEDERS Margery Musser, Wayne Bacon, and David Long Department of Poultry Science A search was made for physiological parameters correlated with production of abnormally shelled eggs (defined by visual inspection as cracked, weak, shell-less, or slab-sided). Plasma levels of the three estrogens, of total and diffusible calcium (Ca), and of total and Inorganic phosphorus (P) were measured, When a hen Is coming into production, the estrogens, total Ca, and total P undergo an increase of twofold or more, and these high levels are maintained as long as egg production continues. The estrogens are necessary for the formation and maintenance of medullary bone which is a supply for a portion of the shell Ca. The blood Ca Is the immediate source for the shell gland, and must constantly be replenished during calcification since the total plasma Ca in the hen would be not more than 4.0% of that normally found In one egg shell. The rise in total P is due to the estrogenstimulated liver synthesis of specific phosphoproteins and phospholipids which are incorporated In the yolk. It was because of their importance in the process of egg formation, as well as shell formation, that these parameters were chosen. Blood samples of 8 ml. were collected between 9 and 10:50 a.m. on two consecutive days from eight normal she11-forming caged turkeys and eight abnormal shell formers. Equal numbers of each type of bird were selected from a line low in its corticosteronc response to cold stress, an index line selected for both Increased egg production increased 16-week body weight, an egg line selected only for Increased egg production* All were under fluorescent lighting (5 a.m. - 5 p.m.) 14 hours out of every 21 hours. Hens had been in production about 11 weeks. Those which had produced an avei age of 45 o abnormal eggs (range o) by this time in the season were considered to be abnormal shell formers and those which had produced a maximum of 8 o abnormal eggs were sampled as good shell formers. TABLE 1.--Blood Plasma Levels of Estrogens, Calcium and Phosphorus in Good and Poor She'll Forming Turkey Hens, Classification of Hens Good (<8% poor shells) Poor (21 to 56% poor shells) Good (<8% poor shells) Poor (21 to 56% poor shells) Mean ± standard error. No. of - Hens Estrone ±2* ±2.1 Calcium (mg./loo ml.) Estrogens (pg./ml.) Estradiol Estriol 1S.CH3J 106±10 18.U1.7 94i7.4 Phosphorus Total Diffusible Total Inorganic 21.5± ± ± ±24 468± ± ±

113 Plasma for all but the Ca determinations was frozen after collection. Duplicate analyses were carried out on each plasma sample for total and inorganic phosphate by the colorimetric vanadomolybdate method, and for total and diffusible Ca by atomic absorption spectrometry after wet ashing. Plasma for diffusible Ca determination was equalibrated under 5 o C0 2 :95% N 2 and ultrafiltered under this atmosphere on the day of collection in order to prevent differences due to pll change after sampling. The estrogens were analyzed by radioimmunoassay. Levels were slightly, but not significantly, lower from samples on the second day and so data from both days were pooled. The results are presented in Table 1. In no case was there a significant difference between the normal and abnormal shell formers. Measurement of these parameters gives no indication of the quality of egg shell which will be formed by turkey hens. Summary Levels of plasma estrogens, total and diffusible calcium, and total and inorganic phosphorus were measured in hens which formed good quality shells and hens which formed poor quality shells. None of the levels were significantly different between these groups. -no-

114 SELENIUM SUPPLEMENTS FOR TURKEY RATIONS Austin H. Cantor Department of Poultry Science Selenium deficiency has been recognized as an important nutritional problem in poultry production. It has been shown that a number of areas in the United States including the Pacific Northwest, the Northeast, and the Southeastern Seaboard states have soils with a low selenium content. Consequently, rations composed of feedstuffs produced in these areas are generally selenium deficient. A combined deficiency of selenium and vitamin E in turkeys is generally characterized by degeneration of the gizzard wall musculature. Heart muscle degeneration is also observed. These symptoms are often accompanied by skeletal muscular dystrophy (Scott et rjl., J. Nutr., 91: , 1967; Walter and Jensen, J. Nutr., 80: , I9b3). Lxudalive diathesis has also been observed in turkeys deficient In selenium and vitamin L {( r^t/ch et al.. J. Nutr» 5 62: 83-95, 1957J. Studies by Waller and Jensen (J. Nutr., 30: , 1963) showed that low levels of the antloxidant ethoxyqujfi (0.025%) or selenium (0.01 p.p.m. or 0.1 p.p.m.); cysfine, mid m^thioniue did not prevent muscular dystrophy In poults fed diets deficient in selenium and vitamin t. Partial protection was obtained with 0.5% ethoxyquin and complete protection was provided by 1 p.pan. selenium or 20 IU vitamin E per kg. jf diet. Other studies (Scott et al., J. Nutr,, 91: , 1967) showed that 0.2 p.p.m. selenium as sodium selenite was required to completely prevent gizzard myopathy in poults fed a practical turkey starter containing only 0.08 p.p.m. of natur'sil/ occurring selenium. However 0*05 p.p.m. selenium was required to obtaiii maximal growth. When II IU of vitamin E per kg e of diet was added to the proctica^ turkey starter, only 0.1 p.p.m. selenium was required for prevention of gi:zit*d myjpathy. These studies indicated that the total requirement for selenium for growth and protection against tnyopathies was 0.18 p.p.m. and 0.28 p.p.m. in the presence and absence of vitamin E, respectively. The food additive regulations of the Food and Drug Administration now permit the addition of selenium to feeds for growing chickens, turkeys, and swine in the form of sodium selenite or sodium selenate. The amounts of selenium added to the complete turkey feeds can not exceed 0.2 p.p.m. Selenium must be added to turkey feeds in the form of a premlx formulated so that a minimum of 1 lb. of premix is added per ton of feed. High levels of selenium can be toxic to animals and humans. However, the addition of selenium to turkey rations appears to be a safe practice since it does not lead to the accumulation of selenium in the tissues of the turkey. Using a low selenium practical diet, Scott and Thompson (Poultry Sci., 50: , 1971) showed that adding as much as 0.8 p.p.m. of selenium as sodium selenite resulted in a concentration of 0.10 p.p.m. selenium (wet basis) in the muscles of poults after 4 weeks of treatment. Cantor and Scott (Poultry Sci., 51: 1790, 1972) studied the effect of administering selenium for a period of 20 weeks to turkeys fed basal diets containing 0.13 to 0.2 p.p.m. of naturally occurring selenium. As shown in Table 1, supplementation of either 0.1 or 0.2 p.p.m. failed to Increase the concentration of selenium in blood, liver, breast muscle, and leg muscle. Without selenium supplementation, feeds produced in areas having low selenium concentrations in the soil are generally either deficient or marginal with respect to the turkey r s requirement for that mineral. Studies with chicks have shown that -111-

115 Tissue TABLE 1. Effects of Selenium Supplementation on Selenium Concentration of ^ Breast Muscle Leg Muscle Liver Blood Tissue Selenium Concentrat1onj[ug_Se/a>_ Fresh Tissue ) + Basal Basal p. p.m. Sef Basal + 0.,2 p. p.m. Sef + ± ± ± ± ± ± *From Cantor and Scott (1972). ^Mean ± SEM for four male and four female turkeys. *As sodium selenite ± ± ± ± selenium In Ingredients of plant origin is generally only two-thirds available and selenium in products of animal origin, e.g., fishmeal, is less than 25 o available. Furthermore, certain environmental contaminants, e.g., polychlorinated biphenyls, can interfere with the utilization of selenium, thereby increasing the requirement for dietary selenium (Combs et al., Poultry Sci., 52: 2013, 1973). Under conditions which cause feed to become rancid, vitamin b is destroyed, further increasing the requirements for selenium in the turkey. Therefore, selenium supplementation of turkey feeds, even those containing as much as 0.2 p.p.in. of naturally occurring selenium, is highly recommended. Summary Selenium deficiency in turkeys is characterized primarily by gizzard and heart muscle degeneration. The deficiency can be prevented by supplementing rations with 0.2 p.p-m. selenium as sodium selenite or sodium selenate. Such supplementation is a safe practice since it does not lead to an accumulation of selenium in tissues of the turkey

116 CAGE CONFINEMENT FOR GROWING, m Austin H. Cantor, Keith 1 Brown, and Karl E. Department of Poultry Science The use of cage confinement has greatly increased in the poultry industry during the past decade. Cages have been mostly used for growing broilers and for laying hens. There has been increasing interest in using cage systems J for turkeys J in recent years. Cage confinement offers a number of advantages to the poultry producer, of the possible benefits include increased labor efficiency, increased housing density, better control of diseases, and easier handling of manure. The cage situation is extremely useful for keeping track of individual egg production and pedigreeing eggs. Wabeck (Poultry Sci., 53: , 1974} found that 8-week-old broilers raised in cages had significantly more tender breast muscle (measured by a shear press) than their counterparts raised in floor pens. The primary disadvantage of caged confinement is that in many cases it is associated with a high incidence of breast blisters, leg weakness, and foot deformities* Performances of both growing birds as well as breeders in the cage is affected by a number of factors. These include size of the cage f type of cage floor Material, and sex, weight, and genetic background of the turkey, Olsen and Lucas (Poultry Sci., 42: 43-45, 1963) studied the effect of floor design upon egg breakage in turkey laying cages, using Beltsville Snail Miite Tht use of a plastic-coated floor with longitudinal flat members in place of tional wire-type floors which were not plastic- coated reduced egg breakage a level of 10% to 0.6%. Woodard and co-workers (Poultry Sci«, 40: , 1961) studied the effect of cage management on male and female Broad Breasted Bronze as well as Beltsville White turkeys. In three assays, egg production for cage-managed Broad Breasted hens was higher than for similar hens under floor management. The rate of soft shell or broken eggs for the cage-managed hens ranged from 12% to 28%. Five percent of these hens had severely swollen feet which were associated with reduced egg production* There was no difference in semen volume for Beltsville Small White and iroad Breasted Bronze males housed in either cages or floor pens. Semen volume was aot affected by the severity of foot swelling in caged males. The performance of growing male and female large white turkeys in cages and in floor pens was compared by Carson and co-workers (Poultry Sci., 52: 2008, 1973). These workers found that at 18 weeks of age hens grown in cages had an average weight of 7.5 kg*, while those raised in floor pens had an average weight of 7.0 kg* Only four of 59 caged hens had breast blisters. In contrast, the weights of 22-week-old toms grown in cages and floor pens were 12.5 and 13.5 kg., respectively. Seven of 52 caged toms showed fluid- filled breast blisters. In addition, caged birds showed significantly poorer feed efficiency and higher incidence of porosis than birds raised in floor pens. Thomason and co-workers (Poultry Sci., 51: , 1972) used large white turkey hens to evaluate the cage and floor environments on reproductive performance, In this study, caged birds had a higher rate of egg production than birds in floor pens. The difference between these treatments was significant at 12 but not at 24 weeks. However, the percentage of settable eggs from hens in floor pens was significantly higher than from Consequently, a larger of settable

117 eggs was obtained from hens in floor pens than from caged hens. Fertility but not hatchability of eggs from hens in floor pens was significantly greater than for those hens in cages. During the 24-week experimental period, the average feed consumption per bird was kg. and kg, for the floor and caged treatments, respectively. The authors indicated that broodiness in floor pens may have been a factor in the differences found in feed consumption between floor environments. The growth performance of Broad Breasted White hens grown in cages with four different types of flooring or in floor pens was compared in a study by Manley and Muller (Poultry Sci., 52: , 1975), The materials used for cage flooring were Neoprene matting, hail screen, tire link matting, and Bressler polyethylene self supporting poultry floor. The hens were grown in the cages from 6 to 24 weeks of age«at the end of the growth trial, the turkey hens in all four types of cages were significantly heavier than hens grown in fjoor pens. The hens in cages with the Bressler plastic floor had an average foot debility score significantly lower than that for the other caged hens. However, this score was not significantly different from hens from floor pens. Wolford and Fadika (Feedstuffs, 46 (16): f 1974) used large white turkeys to compare the effect of cage vs. floor rearing and growing. This study used cage floor materials made of Bressler plastic, welded wire, and plastic slats. No significant differences in body weights for either 17-week-old females or 21-week-old males were observed* The incidence of breast blisters ranged from 8 o to 17% for female turkeys grown in cages with wire floors. No breast blisters were observed in females grown either in floor pens or in cages with the Bressler plastic flooring. A high percentage of males grown in cages had breast blisters. The lowest incidence was observed in cages utilizing the Bressler plastic floor* The feed-to-gain ratio for male and female turkeys housed under the different confinement systems was quite variable. In another study (Fadika and Wolford, Feedstuffs, 46 (33): 20-21, 1974) cagebrooded large white turkeys had heavier body weights than those brooded on litter in floor pens at 6 weeks of age. Seventeen-week-old females grown in cages with the floors made of either Bressler plastic or plastic slats were significantly heavier than their counterparts grown in floor pens on litter when there was a bird density of square meters per bird. When a smaller cage (0.141 square meters per bird) was used, no significant differences were observed between cage and floor systems. Breast blisters were observed in 17-week-old females housed in either floor pens or cages with wire bottoms. No breast blisters were observed in hens in cages with the Bressler plastic or plastic slat bottoms. There did not appear to be much difference in feed-to-gain ratio between the caged and floor-confined turkeys. Twentyone-week-old males grown on a litter floor (0.319 square meters per bird) were heavier than birds grown on wire-floored cages at a density of or 0,319 square meters per bird. The males grown on a litter floor were also significantly heavier than those grown in cages with plastic (Bressler) bottoms providing square meters per bird, but no difference was observed between the birds grown on a litter floor and those grown in a cage with a plastic floor providing square meters per bird. There did not seem to be a difference in the feed-to-gain ratio for males grown under the various confinement systems. Studies at the OARDC (p* 58) indicated considerable differences in the egg production of different genetic lines. Three commercial hen lines were caged and given different feeding regimes (control or restricted). During the trial period, two lines produced an average of 41 and 44 eggs while the third line produced an average of 68 eggs. The percentage of normal eggs was approximately the same for all three lines

118 Another study (Nestor and Bacon, Poultry Sci,, 51: , 1972) used caged females from a medium weight line selected for increased egg production and two large bodied growth lines. The medium weight egg producing line laid an average of 81 eggs during the trial period while the two large bodied growth lines produced averages of 62 and 52 eggs. Besides producing a greater number of eggs, the line selected for egg production had a greater percentage of normal eggs (81% vs. 73% and 63% for the two heavy lines). There are a number of interesting problems which poultry scientists must investigate concerning the effects of cage confinement. For example, how does cage confinement affect the turkey's energy expenditure and ability to regulate body heat, and how do these factors affect the turkey's requirement for dietary energy? Confinement has been known to affect calcium metabolism in a number of species. How does cage confinement affect the calcium metabolism of turkeys and their requirements for calcium, phosphorus, and vitamin D? A number of problems concerning the physiology, nutrition, genetics, and management remain to be solved in order to maximize the advantages of cage confinement for turkeys. Summary Cage confinement of turkeys offers several advantages, including increased labor efficiency, increased housing density, better control of disease^ and easier manure handling. Primary disadvantages include breast blisters, leg weakness, and foot deformities* Performance in cages is affected by size of cage* cage floor material, and sex, weight, and genetic background of the turkey* -115-

119 PREDATOR CONTROL ON THE TURKEY Philip A. Renner and Karl E. Nestor Department of Poultry Science Predator control is a problem for turkey producers who grow turkeys on ranges. This had also been a problem requiring considerable control efforts at the OARDC turkey farm prior to Predators (foxes* raccoons, dogs* and owls) can be a serious economic threat to growers who range turkeys. Until 1971, poison could be used to control foxes and raccoons, but this method was declared illegal by the federal government. Good success had been obtained at the OARDC turkey farm with poison baits, even though a large amount of labor was involved in making the baits, placing them in the proper places* and inspecting them periodically. In 1972, several hundred turkeys were lost to predators, mostly foxes, even though a total of 26 foxes were killed by trapping and hunting. Since local trappers were involved, the actual number of foxes removed from the area around the turkey ranges may have been higher. Charles McGriff of the Fish and Game Service helped in trapping foxes and hunting them at night by using a recorded rabbit-in-distress call. In previous years, predator problems primarily occurred between the ages of 8 weeks, when the birds were placed on range, and about 12 weeks. In 1972, foxes killed turkeys throughout the summer ranging period. After the 1972 growing season, at the suggestion of the Fish and Game Service, an electric fence was placed around the range to aid in prevention of turkey losses to foxes. A pulsating current Bulldozer Electric Fencer of 4,000 volts with less than 1 amp was used in an effort to scare the foxes away without doing permanent injury. This fencer works on a 110-volt current. The electric fencing was placed outside of the fences surrounding the ranges about 6 inches off the ground and about 4 inches out from each post. The wire encircled the entire area of the ranges. The wire was attached to the same posts as the 6-foot poultry fence with 60-penny nails which were driven through plastic fence insulators. Problems with this fencing included build-up of leaves and growth of vegetation. This fencer was strong enough so that some shorting out could occur and the shock would still be effective. Chemical control of all vegetation underneath the electric fence wire was required. In spots where leaves accumulated, they had to be removed manually. Since the electric fence was installed in 1972, there have been no losses due to foxes out of approximately 8,000 turkeys grown on range. The OARDC night watchman has observed foxes in the area both years. Thus, the electric fence is an effective and a relatively inexpensive method of controlling losses from predators. Summary An electric fence was placed just outside of the regular range fences by driving 60-penny nails through plastic fence insulators and into the posts used for the range fence. The wire was placed about 6 inches from the ground and 4 inches out from the post. Chemical control of the vegetation under the wire along with occasional removal of leaf build-up was the only maintenance required. Following a year in which a large number of turkeys were killed by foxes, no losses were observed in about 8,000 turkeys grown on range in 1973 and 1974 after the electric fence was installed, even though foxes had been seen in the immediate area around the ranges

120 EFFECTS OF RELATIVE HUMIDITY ON ASPERGILLUS INFECTION IN TURKEY POULTS Y. M. Saif Departments of Veterinary Science and Poultry Science The effects of relative humidity (RH) on the development of respiratory diseases have been studied very little. Studies in this area were initiated as a result of some observations made 3 years ago when attempts were made to infect turkey poults with Aspergillus fuimgatus (Af). It was noticed then that early in February it was easy to reproduce the disease caused by Af with high morbidity, mortality, and a short incubation period. On the other hand, in late June when birds of the same age were exposed to the same strain of Af, using the same number of spores under presumably very similar conditions, it was not possible to reproduce the severe disease described earlier. Nevertheless, Af was isolated from the lungs and airsacs of birds exposed at both times. After a review of the different variables in both experiments, it was concluded that the RH was probably the main changeable variable. Consequently, a series of experiments was conducted to study the effects of RH on Af infection in turkey poults. Seven experiments were conducted using 3 or 4-day old poults maintained in small isolation units to control the conditions of the experiments, and to control the spread of the spores. The plan was to have two or three isolators for each experiment with high, medium, and low RH. Reducing the RH proved to a difficult task. Therefore, the experiments were conducted with little attempt to reduce the RH in the so-called low RH isolators. Increasing the RH was an easier job and was accomplished in most of the experiments. The temperature was the same in all isolators in a given experiment and was maintained at 95 and 90 F. during the first and second weeks after hatching, respectively. The temperature and RH were monitored during the experiments. The experiments lasted from 1 to 2 weeks. All birds surviving until the end of the experiments were killed and examined. The birds which died during the experiment were also examined for lesions and samples were collected for cultural work. A score of 0 to 4 was used to describe the lesions. A score of 4 meant extensive lesions of the airsacs and lungs, whereas a score of 1 described cases where only one or two abcesses were found in either the lungs or airsacs. The total score from a given group was divided by the number of birds in that group to obtain a lesion and score average. A crude procedure was used to gain some information on the number of spores in the atmosphere of the isolators. This was done by exposing culture media plates for 1 minute at different intervals followed by incubation, then the colonies were counted, The results of the different experiments are summarized in Table 1. The results indicate the existence of an inverse correlation between RH and concentration of airborne spores, and a positive correlation between the number of spores in the atmosphere and the extent of lung and airsac lesions. There is also an inverse relationship between RH and the extent of lung and airsac lesions. It is interesting to speculate on the reasons for these phenomena in terms of host and parasite. It is known that high RH can reduce the number of particles generated from surface areas into the atmosphere. It is also known that static charges on particles are greater at low RH. This might be a factor involved in the disper

121 TABLE 1. Effects of Relative Humidity on Mortality and Lesions of Poults Experimentally Exposed to Asperg i11 us furn1gatus Spores. No. of Birds Percent Humidity Range Mortality Aspergillus Isolations Lesion Score Plate Counts v * *0n several occasions, colonies were too many to count. sion of the spores in the atmosphere. In a recent publication by Anderson and colleagues from Sweden who were studying outdoor airborne bacteria, they indicated that whenever the RH rises above 80%, there is a reduction in bacterial numbers. There is a lack of information on the effects of the RH on the host's defense mechanisms against respiratory diseases and the rate of clearance of particles from the respiratory tract at different RH levels. Studies on this subject would certainly be valuable. Currently, procedures are being improved to continue these studies. Suiwnary Seven experiments were conducted to study the effects of relative humidity (RH) on Aspergillus fumigatus infection in turkey poults. The results indicated an inverse correlation between RH and the extent of lung and airsac lesions. Numbers of spores in the atmosphere of isolators were higher at low RH than at high RH.

122 AEgQMONAS AND SALMONELLA INFECTIONS IN TURKEY POULTS Y. M. Saif and W. F. Busch Department of Veterinary Science Salmonellas are pathogens of man and animals which have a defined role as a cause of disease. Aeromonas hydiophlla is isolated occasionally from man and animals. Other than causing disease in amphibians and fish, no definite role in disease in man or animals has been attached to them. The authors 1 interest in the role of A. hydrophila as a cause of disease in turkeys came as a result of detection in this laboratory of a natural infection in a group of S~week-old turkey poults with a combination of A. hydrophila and Salmonella infantis. The condition was investigated because of increased mortality and the main lesion detected was congestion of the intestines. Two experiments were conducted in an attempt to define the role of the two organisms in the disease process. Four groups of birds were used in each experiment according to the following schedule: Group 1. inoculated orally with A. hydrophilia Group 2» inoculated orally with S. infantis Group 5. Group 4. inoculated orally with a combination of the two organisms non-inoculated controls. The birds were 4 days and 1 day old at the time of inoculation in experiments 1 and 2, respectively. The birds were maintained in isolation units and observed daily for 2 weeks post-inoculation, when the experiments were terminated. Birds which died during the 2-week period were necropsied and attempts were made to isolate the organisms from the livers and intestines. Birds killed at the end of the experiments were examined macroscopically for lesions. Tables 1 and 2 show the mortality data in both experiments. Hie results indicate that the highest mortality was in birds inoculated with both organisms. TABLE 1.--Inocula and Mortality in Experiment 1. Inoculum Dose per Bird No. of Dead Birds A. S. A. S. hydrophila infantis hydrophila infantis None (controls) X X X X None /6* 0/7 2/6 0/7 *Number of birds which died per number of birds in group

123 TABLE 2. Inocula and Mortality in Experiment 2. Inoculum Dose per Bird No. of Dead Birds A. S. A. S. hydrophila infantis hydrophila infantis None (controls) X X X X None /8* 1/8 4/8 1/8 *Number of birds which died per number of birds in group. Mortalities started 1 day post inoculation and the last mortality was 1 week after the start o the experiment. There was no mortality in birds inoculated with A. hydrophila alone and 4 of the 14 birds killed had mild enteritis. Of the birds inoculated with S. infantis, one bird died and was autolytic at the time of examination, and 3 of the 14 birds killed had enteritis. Six of the 14 birds inoculated with both organisms died and exhibited various degrees of intestional and coccal congestion, and some birds had congested livers, of the birds killed from the same group, one had liver congestion, another had enterisis, and the third exhibited a congested caecum. None of the birds in the control group had lesions. The results indicate that in turkey poults a combined infection with A. hydrophila and S. Infantis produces higher mortality than in birds inoculated with either organism alone. More experiments are planned to confirm these observations. Summary A hydrophila and S. Infantis were isolated from naturally infected 3-week-old poults. High mortality and congestion of the intestines were associated with the infection. Two experiments were conducted in an attempt to define the role of both organisms in causing a disease in turkey poults. Birds inoculated with both organisms had higher mortality than those inoculated with either organism alone

124 ISOLATION AND CHARACTERIZATION OF THE TURKEY IMMUNOGLOBULINS M AND G Y. M. Saif and J. E. Dohms Department of Veterinary Science Immunoglobulins (Igs) are proteins which function mostly as antibodies. They are present in the serum and body fluids and some of them are important in protecting animals from infectious diseases. The isolation and characterization of Igs and the study of their relation to antibody activity provides a better understanding of the immune response of a particular species. In man, five classes of Igs (designated G, M, A, D, and E) have been isolated and characterized. The chicken is the most thoroughly studied aviari species, and so far three Ig classes (G, M, and A) have been identified. There is little information in the literature on Igs of the turkey, and there aren't any detailed studies on the ontogeny of its immune response. The paucity of knowledge in these areas encouraged a tendency to lump the turkey with the chicken in discussions of pathogenesis, immune response, and vaccination recommendations. This study was undertaken to isolate and characterize the Igs of the turkey. The main emphasis was on the isolation and characterization of Ig G and Ig M. Sera were delipidated, and the immunoglobulins were precipitated by adding sodium sulfate. The Igs were separated by column chromatography and tested for purity by immunodiffusion and immunoelectrophoresis. It was possible through the use of these procedures to obtain pure preparations of Ig G and Ig M which were subsequently used for antisera production and characterization. Antisera produced in rabbits against Ig M and Ig G were used to detect the presence of Ig M and Ig G in different body secretions. Sucrose density ultracentrifugation studies indicated that serum Ig G has an S value of 7.4 and Ig M an S value of To test for antibody activity in Ig preparations, selected fractions of serum were absorbed either with anti Ig G or anti Ig M sera. The pure fractions (either Ig G or Ig M) were then tested for Mycoplasma meleagridis antibodies, using a tube agglutination and a growth inhibitition test. Both the agglutinating and growth inhibiting activities were detected in the Ig G and Ig M fractions. The results of testing for the presence of Igs in different body secretions indicated that serum, bile, saliva, lacrimal secretions, tracheal washings, uterine fluids, oviduct scrapings, small intestine washings, and small intestine scrapings contained both Ig G and Ig M. Albumin and yolk contained only Ig G, whereas the large intestine contained neither. There are indications that the turkey has a third Ig similar to Ig A of mammals and chickens, but the authors have not purified it yet. Anti turkey Ig G serum crossreacts with chicken serum and bile, but anti turkey Ig M does not, indicating similarities and differences between turkey and chicken Igs. Immediate objectives from these studies were the identification and characterization of the turkey Igs. Information and reagents obtained from these studies will be utilized for more detailed studies of the turkey immune response to different pathogens -121-

125 Turkey iinmunoglobulins G and M were Identified and purified from turkey serum and monospecific antisera were produced in rabbits. Serum Ig G had an S value of 7.4 and Ig M had an S value of 16.9, Agglutinating and growth inhibiting antibodies against Mycoplasma meleagridis were detected in both the purified Ig G and Ig M. The presence or absence of Ig G and Ig M in different body secretions is reported. The antigenic relations of turkey Igs with chicken Igs also was investigated

126 OBSERVATIONS ON THE SEROLOGIC RESPONSE OF THE TURKEY TO MYCOPLASMA HELEAGRIDIS ANTIGENS Y. M. Saif and J. E. Dohms Department of Veterinary Science Information on the serologlc response of a species to stimulation with antigens of different pathogens is helpful In the study of diseases and their control. In the study reported here, the objective was to investigate the serologic response of the turkey to different preparations of M. meleagrldis introduced via different routes. Three preparations were used: live, heat killed, and formalin inactivated organisms. Each preparation was inoculated via one of the following routes: intravenous (IV), intramuscular (IM), intratracheal (IT), and subcutaneous (SC). There were eight birds in each group inoculated with the live and formalin inactivated organisms. Two birds were used for each route of inoculation. The group inoculated with the heat inactivated organisms was made up of two birds inoculated IV, and one bird for each of the other three routes of inoculation. All birds were free of M. meleagridis infection before inoculation. The birds were bled and the tracheas were swabbed on the following days after inoculation: 1, 2, 5, 4, S, 6, 7, 11, and 14, and thereafter once every week until the 16th week post inoculation. At that time, all birds were challenged IV with a live culture, then bled and swabbed at intervals up to 5 weeks post challenge when all the birds were killed and several tissues were examined for M. meleagridis. The serum collected was tested for M. meleagridis antibodies, using tube agglutination (TA) and growth Inhibition (Gl) tests. The swabs were streaked on Mycoplasma culture media. Live Organisms ResuUs Agglutinating (agl.) antibodies (ab.) were detected 2 days post inoculation, the titer increased and remained high until the birds were challenged and thereafter until the end of the experiment, 2. Growth inhibiting (GI) ab. were detected 5 weeks post inoculation, increased and remained high until the end. The titer reached a maximum at a time which corresponded with the clearance of organisms from tracheas. 3. M. meleagridis was isolated from the tracheas up to 13 weeks post inoculation, for 2 days following challenge, then disappeared and no organisms were found in the respiratory tract, reproductive tract, or bursa of babricus at the end of the experiment. IM 1. No detectable agl. or GI ab. were detected after primary inoculation. 2. Agl. ab. were detected 4 days post challenge, but no GI ab. were present. 3. Organisms were isolated continuously from the tracheas before and after challenge and at the time of killing

127 IT 1. Low liters of agl. ab. were detected only between 2 and 4 weeks post inoculation. 2. No GI ab. were detected after primary inoculation. 3. Good agl. and Gl at. responses were detected 4 days post challenge. 4. Organisms were cleared from the tracheas 1 week after primary inoculation and no organisms were detected in the tracheas following challenge. SC 1. There was a low level of agl. ab. titers between 2 and 6 weeks following inoculation. 2. No GI ab. were detected after inoculation. 3. Good agl. and GI ab. were detected 4 days post challenge. 4. Organisms were detected occasionally up to 9 weeks, but no organisms were detected following challenge or at the time of killing when several sites were examined. Formalin Inactivated Organisms 1. Agglutinating ab. were detected in all birds (inoculated via different routes) 4 days following inoculation. The highest titers were detected in IV inoculated birds. Titers were relatively low and disappeared fast (2 weeks in IV, I days maximum in other groups), 2. No GI ab. were detected after primary inoculation in any of the birds in the different groups. 3. Agglutinating ab. were detected 4 days post challenge and Gl ab. were detected for the first time 5 weeks following challenge. 4. Organisms were isolated from tracheas of all birds following challenge and from different tissues examined at the end of the experiment. Heat Inactivated Organisms 1. Agglutinating ab. were detected at a low level in all birds except for the SC inoculated birds, which had none. Otherwise, the pattern of development, duration, and titers was similar to those in birds inoculated with formalin inactivated organisms 2. No. GI ab. were detected following primary inocualtion. 3. The agl, ab. were detected 4 days post challenge and the GI ab. were detected 2 weeks following challenge. 4. Organisms were isolated following challenge from tracheas of all birds from different sites at the end of the experiment

128 Discusslpn These results illustrate the effect of treatment of the organism and the route of inoculation on the serologic response of the turkey,, Some of the interesting aspects of this study include: 1. The clearance of the organism from the tracheas and other tissues at the time when the GI ab. titer reached its maximum, which suggests a possible protective role. 2. The failure of the live organism when inoculated IM, IT, or SC to elicit a GI ab. response, although it did elicit an agl. ab. response when introduced by the same route. 3. Although no GI ab. were detected following inoculations of the live organisms IM, IT, and SC, on challenge this ab. was produced in the birds inoculated via IT and SC routes but not in those birds inoculated IM. This ab. was detected 4 days following challenge in birds which had no GI ab. after primary challenge. 4. The heat and formalin inactivated organisms were capable of inducing a low titer agl. ab., but not a GI ab. response. This might be due to: (a) the small size of the Inoculum compared to live organisms which multiply resulting in a large antigen mass or (b) the loss of certain antigens during the inactivation process. It would be interesting to study the possibilities of eliminating M. meleagrid^s from a certain population of male and female turkeys by the use of some of the procedures described earlier before these birds came into production. Summary An experiment was designed to study the serologic response of turkeys to M. meleagrldjls. Three preparations were used: live, heat inactivated, and formalin inactivated organisms. Four different routes of inoculation were employed. All preparations elicited an agglutinating antibody response which varied in intensity, but only the live organisms inoculated IV were capable of eliciting a GI response after primary inoculation. On challenge with live organisms, all birds except those inoculated primarily IM with live organisms developed GI ab. titers. Nevertheless, the time interval between challenge and first detection of the GI ab. varied significantly. The results indicated that the GI might have an important protective role against M. meleagridis. This speculation is based on the failure to isolate the organism at the time when this ab. was detected in high titers

129 AND ACID LEVEL WHITE Paul t. Waihel and B. R. Department of Animal Science University of Minnesota It is a challenge of considerable magnitude to determine tl'.r ib aiging minimum protein and am'mo acid needs uf a turkoy js it grows from a snail poult weighing grams to a market bird ot perhaps 12 kilogrjms. iliere are a variety of strains and growth rates to consider. Ammo acid requirements depend upon, tho balance of other amino acidb; hence the type of diet liab relevance. Studio with compensatory growth indicate that early rapid growth may not be necessary. Flic influence of ambient temperature and exercise on feed and nutrient intake i* important. Increasing costs of protein may make it desirable to aim for somewhat" blower growth in order to achieve maximum profitability. 1 The approach in the latter study was to iormulate a ratjon series based upon four protein-aid no acid levels and to observe growth responsiveness to these diets. Corn and soybean meal type diets were formulated to selected methioninc and lysine levels. The studies utilized Broad White male turkeys. Any relevance of the data to female Broad White turkeys and to Medium White turkeys was derived through assumptions regarding expected protein needs of the latter turkeys as compared to the Broad White male turkeys* The present study was designed to compare directly responses of Broad White male and female turkeys (Nicholas strain) and Medium White male and female turkeys (Wrolstad strain) to the same series of protein levels. Six treatments were applied to each segment of the experiment. A, B, C, and D protein-amino acid levels represented optimum, marginal, suboptimum, and deficient conditions, respectively. In addition, two treatments representing level cross-overs were employed; these are indicated as, for example, ABC, meaning that A diet was fed during periods 1 and 2, B during periods 3 and 4, and C during period 5. The experimental diets are given in Table 1. The concept of lower requirements for the females and smaller turkeys was applied to the experiment by providing the latter turkeys with an accelerated ration change schedule, as follows; Feeding Period I Male 0-4 weeks 5-8 weeks 9-12 weeks weeks weeks Broad White Female 0-4 weeks 5-8 weeks 9-11 weeks weeks weeks 0-4 OF Medium White Male "TiiaTe 5-8 weeks 9-11 weeks weeks weeks 0-4 weeks 5-7 weeks 8-10 weeks weeks weeks 1 Waibel > P.E., M.E. El Halawani, B.R. Behrends, and R.E, Elsabee Nutrition and economy of turkey production. Proc., Minn. Nutr. Conf., pp

130 TABLE 1 --Diet Composition in Experiment TG-734 Ingredients (%) and Composition 1 2 Feeding Period Diet A Corn, Ground \ellow Soybean Flea], Dehulled Dl-methionine (98») Constant Ingredients (below) ,7 0, « , Protein, * Metabohzable Energy, Kcal./kg. Methiomne, * flethionine, /meal. Lys i ne, % Lysine, */mcal , , Diet E Corn, Ground Yellow Soybean Heal, Dehulled DL-methionine (96?.) constant Ingredients (below) , Protein, A Metabol liable Energy, Kcal./kg, Methiomne, > Hethionine, %/mcal. Lysine, Lysine, /meal , , <J Diet C Corn, Ground Yellow Soybean Meal, Dehulled DL-methionine (98%) Constant Ingredients (below) , , ProtPin, Metabolizable Energy, Kcal./kg. Methionine, c^ Methiomne, Vmcal. L>sine, \ Lysine, «/mca , Diet D Corn, Ground Yellow Soybean Meal, Dehulled DL-rnethiomne (98 ) Constant Ingredients (below) Protein, / Metabolizable Energy, Kcal./kg Methiomne,.. Methionine,./meal. Lysine, c e Lysine, 1/mcal Constant Ingredients Animal Fat Fish Solubles, 100% Dried on SBM Fermentation Residue Product Dicalcium Phosphate Calcium Carbonate Salt (Iodized) Trace Mineral Mixture* Vitamin Mixture TSR " , Vitamin Mixture TGR-701** *The trace mineral mixture contained: 25% calcium. 6% manganese, 0.12% iodine, 2% iron, 0.2% copper, 6% zinc, 0.2% cobalt, and f Vitamin mixture TSR-701 contained the following per gram of mixture: vitamin A paimitate 4,000 III, vitamin D 1,000 ICU, vitamin E acetate 5.5 IU, vitamin K (MPB) 0.82 mg., riboflavin 2.58 mg., Oj-calcium pantothenate 3,53 mg., niacin 23.3 mg., choline chloride mg., vitamin B meg., folic acid 0.3 mg., biotin 33.3 meg. **Vitamm mixture TGR-701 is the same as TSR-701, except that folic acid and biotin are omitted

131 The study was conducted in the nutrition-growth building* Turkey Research Unit, Experiment Station* Rosemount. Each strain-sex of turkey was placed in an identical quarter of the building and the two strains of turkeys were started within a few days of each other. There were four replicate pens (6 x 8-foot or 1.83 m. x 2.44 m.) per treatment. Wood shavings were used for litter* Bird and feed weighings were accomplished at 3 or 4-week intervals according to the ration change schedule. The secondary objective of the experiment was to gain further information on how protein-amino acid levels may be altered in recognition of economic conditions* particularly prices of corn and soybean meal. The economic calculation method was described earlier (Waibel, et al., 1973). It requires a determination of the number of days to produce a turkey of a given market size and the feed efficiency at that time. These points are established with the aid of regression analysis. Using these values, with fixed and variable costs, the net flock and yearly returns are calculated in dollars per bird. In the present comparisons, the sale price of turkeys was assumed to be 40 cents per pound on the farm. This price enabled most of the profitabilities to be on the positive side when the calculation model considered corn prices of $5 and $8 per 100 Ib. and soybean meal prices of $5, $10, and $20 per 100 Ib. Other ingredient costs per 100 Ib. were DL-methionine, $165; animal fat, $16; fish solubles product, $15; fermentation residue product, $20; dicalcium phosphate, $6.50; calcium carbonate, $1; salt, $1.50; trace mineral mixture, $10; vitamin mixture TSR- 701, $30; and vitamin mixture TGR-7Q1, $25. Results The growth data for Broad White males (Table 2) show an increasing responsiveness from D to A protein levels. The growth response from the B to the A level, while positive, was not statistically significant at the 5% level of probability except at 4 weeks of age. It is believed, however, that the responses from the B to the A level were meaningful. The turkeys fed the BCC combination or the CBB combination performed between the B and C levels of the straight series. Observing the return/bird/year data, with the low soybean meal price, turkeys fed the A or B level returned greatest and approximately equal profitability; with the medium soybean meal price, those fed the B or the combination BCC or CBB levels were more profitable; and with the high soybean meal price, greatest profitability was achieved with the BCC combination. The data for Broad White females are presented in Table 3. The response to the A through D levels was graded. The B level performed nearly as well as the A level and by the end of the experiment (17 weeks} there was no difference in body weight. The BCD and CCD combinations performed similarly to the C level. Thus it would appear that the B level was nearly adequate for maximum gains with this type of bird. Maximum return/bird/year was achieved at the A and B levels with $5 soybean meal, at the B and C levels with $10 soybean meal, and at the C level with $20 soybean meal. The data for Medium White males are presented in Table 4. In this case, there were steep growth responses from lower to higher protein levels, including the step from B to A. It was surprising to note the apparently greater need of Medium White males for protein (amino acids). Turkeys fed the BCD combination and the CCD combination performed approximately as expected (C range). The Medium White males were evaluated economically at two stages of growth: the normal roaster-fryer stage which was assumed to be 5.3 kg. (11.68 Ib.) and the large turkey hen stage which was assumed to be 6.5 kg. (14.33 Ib.). The returns/bird/year were the greatest at the A level with low priced and medium priced soybean meal. When the soybean meal price was highest, there were not large differences in profitability at any of the dietary treatment levels

132 TABLE 2.--Influence of Protein Level on Performance of Broad White Male Turkeys and on Net Economic Returns. Measure Body Weight, kg. 4 weeks 8 weeks 12 weeks 16 weeks 20 weeks A B Ration Formulation Series C D BCC CBB Least Significant r\-i ~ -f*\ v*/"!mrtr»ft U Ci d IOC P< Feed/Gain 0-4 weeks 0-8 weeks 0-12 weeks 0-16 weeks 0-20 weeks N.S..10 N.S. N.S. N.S. Days of Age Feed/Gain No. Flocks per Year At 11 kg. ( Ib.) Body Weight Return per Bird per Corn SBM $5.00 $ Flock* $ $ $ " $ $ $ Return per Bird per Corn SBM $5.00 $ Year* $ $ $ $ $ $ See text for method

133

134 TABLE 4. Influence of Protein Level on Performance of Medium White Males and on Net Economic Returns, Measure Body Weight, kg. 4 weeks 8 weeks 11 weeks 14 weeks 17 weeks Feed/Gain 0-4 weeks 0-8 weeks 0-11 weeks 0-14 weeks 0-17 weeks Days of Age Feed/Gain No. Flocks per Year Return per Bird per Flock* Corn SBM $5.00 $ Return per Bird per Year* Corn SBM $5.00 $ * Days of Age Feed/Gain No. Flocks per Year Return per Bird per Flock* Corn SBM $5.00 $ , Return per Bird per Year* Corn SBM $5.00 $ A , $ $ $ $ , B At $ $ ,52 At $ $ Ration Formulation Series C kg. ( $1, $3.93 1, $1.12 ff $ D BCD lb.} Body Weight $ $ kg. ( Ib.) Body , $ $ $ $ Weight $ $ CCD $1.47 0, $ , $ $ Least Significant Difference P<, N.S. N.S. N.S. N,S. N.S.

135 The data with the Medium White females are presented in Table 5. The females performed similarly to the Medium White males in that the responsiveness to increasing protein was significant and of considerable magnitude, not only with the lower levels but with the A and B levels as well. The economic returns on a per bird/year/ basis again showed greatest profitability with the A level at low soybean meal prices, At medium soybean meal prices, the A level was numerically superior but the B and C levels were fairly good. At the high priced soybean meal, the C protein-amino acid level series was somewhat more profitable. TABLE 5. Influence of Protein Level on Performance of Medium White Females and on Net Economic Returns. Measure Body Weight, kg. 4 weeks 7 weeks 10 weeks 13 weeks 16 weeks Feed/Gain 0-4 weeks 0-7 weeks 0-10 weeks 0-13 weeks 0-16 weeks Days of Age Feed/Gain No. Flocks per Year Return per Bird per Corn SBM $5.00 $ A Return per Bird per Flock* Corn SBM $5.00 $ 5.00 $ , Year* $ See text for method. Ration Formulation Series B At $ $ C kg. ( $ $ D Ib.) Body $ $ CDD Weight $ $ DCC $ $ Least Significant pifforonro P< N.S

136 The following performance values were calculated to assist In the Interpretation of the responsiveness of the four types of turkeys to the dietary protein- amino acid levels. Broad_jjji1 te... _... Medium White MaH[e Female Male Female For, all ^5...penodsj " ~~ ~ ~ Average daily gain (A treatment), g Response in growth (A-D)/A, % Response in growth (A-B)/A, % ds : Response in growth (A-B)/A, % The average daily gains of turkeys fed the A protein level sequence for the entire study confirm the well known growth rate characteristics of the four types of turkey. If required percentage dietary protein is related positively to growth rate, as frequently assumed, the Broad White male would have the highest requirement, the Broad White female and Medium White male would have the intermediate requirement, the Medium White female would have the lowest requirement* The latter portion of the feeding schedule allows somewhat for these differences; however, for the first two periods the schedule Is the same for all turkeys. As shown in the tabular data above, the 5-period growth response from D to A protein w^s substantially greater for the mediam as compared to the large turkey. The 5-period response from B to A protele level, more indicative of the requirement for maximum growth, shows a greater need of the medium turkey for protein. The response in growth of B to A protein aftei two periods also shows the more sensitive needs of the medium turkey; during tills time all of the turkeys were fed the same A* B, C, and D diets. When these four protein levels were set, before the inception of the experiment, it was expected that the Broad White male turkeys would show maximum response between the B and A levels,, and the Medium White females might show maximum response at about the C level, especially during the first two feeding periods. As it turned out^ the Broad White turkeys were on target and the Medium White turkeys may actually have had a higher requirement than the A level for maximum growth. The experimental comparison was considered unbiased in that the turkeys were in the same facilities at the same time and were fed the same diets. This study does not rule out growth rate as a possible important factor modifying protein requirements, but genetic and/or other factors may also have relevance. This area is In need of investigation. The economic analysis study illustrated, for each type of bird studied, how net returns on a per flock or yearly basis may be affected by dietary protein level, considering variable corn and soybean meal prices. The concept was supported that higher protein levels are more profitable when soybean meal prices are low, and that when soybean prices are high, it may be desirable to feed for less than maximum growth. The break points vary somewhat according to the type of bird and in relation to the specific formulation, as indicated In the results. The price of corn influenced profitability, of course, but its influence on formulation decisions relating to protein level was relatively minor as compared to that involving price of soybean meal

137 INFLUENCE OF AND NUTRITION ON 01- AN!. 1 THEIR M. E. Cl Halawani, P- t. Wdibel, W. H. Burke, and R. N. Shoffner Department of Animal Science University of Minnesota A description of the experimenta1 plan of this study, together with a synopsis of pertinent literature, was presented by LI Halawani, <*t ^1. Light intensity and length of dav are important la regulating reproduction. Turkeys raised on long days arc normally n ftitetor) unless given a special pretreatment involving a short day length. It 3*> not known whether light intensity during the prebreeder period has any effects which can be carried over into the reproduction period. The question of required protein Irvel of growing turkeys to he used for reproductive purposes is also pertinent. There have been conflicting reports as to whether feed or growth restriction results in improved reproductive performance. In the study reported here, the authors attempted to determine the influence of certain variables during the development period on reproductive performance in turkeys, Objectives were: 1) to study the separate and combined effects of light manipulation and diet during the starting, growing, and holding periods on subsequent reproductive performance of turkoy hens; 2) to evaluate fertility and hutchabiiity o± hens inseminated with semen from turkey males raised with control arid low protein diets. Broad White male and female lines of the Nicholas si rain were used. The females were raised in four separate environmental rooms (each containing 16 pens, 2.44 m* x 2.44 m, in size] at the rate of 18 birds per pen, arid subjected to the following light treatments: A) 14 hours of light (50 lux) daily to 21 weeks of age, day length reduced to 6 hours during weeks of age; B) similar to A, except that light intensity was at 0.1 lux; C) continuous light of 0.1 lux; D) birds subjected to intermittent light (50 lux) with 2 hours of light: 4 hours of dark (3 x a day) to 30 weeks of age. The 16 pens per room were divided into four nutritional treatments (four replicates per treatment) including: 1) control, starting, growing, and holding diets of corn-soybean meal type with 27.7% protein during the first 4 weeks, decreasing to 25.5, 22.9, 20.4, and 17.6% at 5, 9, 13, and 17 weeks of age, and finishing at 14.8% protein during weeks of age; 2) control diet series fed ad libitum during the first 2 weeks of life and thereafter fed skip~a~day until 30 weeks of age; 3) as 2 except that skip-a-day feeding started when the birds were 12 weeks of age; 4) low protein corn-soybean series starting with 20% protein until 4 weeks of age, then reducing protein by 2 o every 4 weeks and finishing at 10.3% protein during weeks of age. At 30 weeks of age, the hens were placed on the same breeder diet (corn-soybean meal type, 16% protein, 2,900 kcal. metabolizable energy per kg.) and subjected to *Ei Halawani, M. E., P. E. Waibel, W, H. Burke, and R. N. Shoffner Minnesota Turkey Research Report, Minn. Agr. Exp. Sta., Misc. Rept. 121, pp , -134-

138 14 hours of light per day at 50 lux. The same pens were utilized, with 12 hens per pen. Lfg production records commenced at 53 weeks of age (March 27) and terminated Julv 17 after a short 16-week season due to low production resulting from extreme heat in early July The male line poults for the study were housed in confinement pens at an average light intensity of 22 lux. Duplicate groups were fed a control protein series (as female treatment 1) and duplicate groups a low protein series (as female treatment 4). The diets of control male turkeys were reduced to 13.3% protein and the low protein males were kept on the I0.3 o protein treatment during the breeder season* Of the four replicate female pens* two were inseminated by semen from the replicate pens of control protein and two from replicates of low protein. Resu.1ts Only main effects are reported in this paper. Light Treatment Effects (Table 1)_ Rearing period light treatments resulted in a significant effect on body weight at 30 weeks of age. Birds on intermittent light (Treatment D) were significantly heavier fp<0.01) than those receiving 14 hours of light at an intensity of 30 lux (treatment A). Turkeys receiving continuous low intensity light (0.1 lux) (treatment C) were significantly lighter than those of the other three treatments. The depressing effect of continuous low intensity light on body weight is hard to explain. It has not been determined whether a lack of physical exercise and/or a disturbance of metabolism are contributing factors. It appears that low intensity light should be interrupted by periods of darkness in order to obtain favorable growth. Hie best egg production was obtained from hens raised under the conventional (A) and inteimitterit (D) light treatments. Hens which received 14 hours day length with 0.1 lux light intensity (treatment B) during the prebreeder period laid significantly (P<0.01) fewer eggs. Exposure of turkeys to continuous low intensity light during the growing and holding periods resulted in a dramatic reduction in egg production, with the majority of the hens not coming into production. Thus not only the day length but also the intensity of light during the development period appear to influence the rate of lay (compare treatments A and B). Furthei investigation is needed to determine whether even higher intensity light during the growing and holding periods affects egg production of turkey breeders. ^ Fertility and hatchability were estimated from a 1-week collection of eggs which was repeated every 4 weeks. There was no adverse effect of light treatments on eithei fertility or hatchability; however, fertility was somewhat lower (P<0.05 only) with light treatment B. Egg weight was significantly greater with light treatment C as compared to either of the other three light treatments. This may be attributed to the lower level of production observed

139 TABLE 1.--Effects of Prebreeder Light Treatments on Production Characteristics of Female Turkeys. Body Weight at 30 Weeks (kg.) Feed/ Gain Efficiency Egg Production, Fertility, % Hatchability of Fertile Eggs, % Egg Weight (g.) Body Weight at 49 Weeks (kg.) Feed/Hen/Day (g.) Weeks *L1ghttreatments: A S.3 87, Light Treatment* B C , D Least Significant Difference P<0.05 A - 14 hours of light daily to 21 weeks of age; day length reduced to 6 hours during weeks of age. Light intensity 30 lux. B - Similar to A except that light intensity was 0.1 lux. C - Continuous light of 0.1 lux. D - Intermittent light, described in text NS 1.54 NS NS Least Significant Difference P< NS 4.46 MS NS 2.05 NS NS TABLE 2. Effects of Prebreeder Nutrition Treatments on Production Characteristics of Turkey Females. 1 Nutrition Treatment* Least Significant Difference P<0.05 Least Significant Difference P<0.01 Body Weight at 30 Weeks (kg.) 9,23 8, Feed Consuiepti on/bird (2-30 Weeks) (kg.) Feed/Gain Efficiency Egg Production, % , NS NS Fertility, % NS NS HatchabfUty of Fertile Eggs, % Egg Weight (g.) , NS 1,6 NS 2.1 Body Weight at 49 Weeks (kg.) , Feed/Hen/Day (gj Weeks ,6 NS NS *NutHtion treatments: 1 - Control. 2 - Skip-a-day at 2 weeks of age. 3 - SMp-a-<day at 12 weeks of age. 4 - Low protein

140 Nutrition^ Treatment Effects^ (Table 2) The consideration of nutrition main effects is taken across A, B, and D environments only. Production was so poor with hens raised in low light intensity that omission of this treatment was considered desirable, Limiting feed intake by skip~a~day programs (treatments 2 and 3) or feeding the low protein diet series (treatment 4) reduced body weight significantly at 30 weeks of age > as compared with ad libitum feeding (treatment 1), Body weight was less affected the *>lip-a-day program was started at 12 weeks of age (treatment 3J as ro!a Mr d to the skip-a-day started at 2 weeks of age. Feeding a low protein achieved runilar growth restriction as skip-a-day started at 2 weeks of age. The low body weight of birds receiving the skip-a-day program commencing at 2 week* of age was associated with the lowest feed/gain value. The most feed restriction was achieved by skip-a-day at 2 weeks (treatment 2) and low protein (treatment 4, The of feed consumed 18% and 15% less for treatments 2 and 4 as compared to the ad libitum control, The body weight reductions resulting from the prebreeder period skip-a-day and low protein treatments persisted throughout the laying period, although the differences tended to decline. There was no significant difference in the feed consumed during the breeder period. Rate of lay from weeks of age of birds fed every other day or fed the low protein diet was less than those full fed the control diet during rearing. However, these differences were not significant. There was no adverse effect of dietary regimen on either fertility or hatchability, Egg weights of birds fed low protein diets and those fed every other day starting at 2 weeks of age were significantly smaller than those of control and skipa-day at 12 weeks of age. The reduced egg weight may be related to the smaller body weight, 3)... Males fed the low protein series weighed less at all ages than males fed control series of diets, with the greatest weight differential occurring at 28 weeks of age, There was a small but significant reduction (P<0.01) in fertility of hens inseminated by semen from low protein males. The slight lowering of hatchability was not significant. The data indicate the severity of the low protein diet fed to the males, Further studies are required to determine the protein level which will result in reduced growth without an adverse effect on fertility.

141 TABLE 3. Influence of Dietary Control and Low Protein Levels on Certain Reproductive Characteristics, Body Weight, and Feed Consumption of Male Turkeys. Control Treatment Low Protein Body Height (kg.) 4 Weeks Weeks Weeks Feed/Bird/Day (g.) 0-4 Weeks Weeks Weeks Fertility, % Hatchability, %

142 COMPARATIVE PERFORMANCE OF TURKEY BREEDER HENS IN CAGES AND FLOOR PENS P. E. Waibel, M. E. El Halawani, and W. H. Burke Department of Animal Science University of Minnesota There has been considerable Interest in housing turkey breeder hens in cages, It is done commercially to some extent but in general the performance has not been satisfactory enough to create a mass movement to turkey hens in cages similar to that seen with egg-type chickens. In developing the facilities at the Agricultural Experiment Station, Rosemount, Minn., and recognizing that a house with floor pens was already in service, it seemed appropriate to build one breeder house for studies with turkeys in cages. One of the four identical rooms in the cage building was outfitted with floor pens so that floor-cage comparative studies could be conducted. There are a few reports comparing performance of hens in cages to controls in floor pens (Woodard, et al., 1961; Woodard, 1967; Berg and Shoffner, 1968; Thomason, et al., 1972). In general, total egg production was either slightly better or similar for hens in cages. However, the number of settable eggs was invariably reduced with cages due to increased incidence of broken and soft shelled eggs. Fertility was usually decreased and hatchability was invariably decreased with hens in cages. Cage birds appear to lose less weight during production. The present study was established to compare results of hens in floor pens (four densities) with hens in cages (one or two hens per cage, polyethylene 1 or simulated rubber 2 floors). The turkeys were Orlopp Broad White turkeys (the hens were No. 93 hen-line hens and the toms were No. 62 torn-line toms). They were raised on a full feeding program and fed diets 610, 620, 630, 640, 650, and 680 (Waibel, et al., 1967) during 0-4, 5-8, 9-12, 13-16, 17-20, and weeks of age. At lighting they were given the 16% protein diet shown in Table 2 of the publication by Waibel (1974). This diet contained 16.0% protein by calculation, 2,900 kilocalories of metabolizable energy per kilogram, 2.26% calcium, and 0.69% total phosphorus. Bird density was varied in the floor pens as shown in Table 1. In the cage portion of the experiment, both polyethylene floors and simulated rubber floors were employed; each floor type accommodated one^or two birds per cage. The cages were 45.7 cm. (18 in.) wide, 61.0 cm. (24 in.) deep, and 61.0 cm. (24 in) high, arid utilized egg roll-out to the front. There were four replicate pens or lots per treatment in either environment. The experimental number of birds per unit varied in the floor pens as indicated. A section of eight cages* comprised one replicate lot. The hens were 35 weeks of age at time of lighting rather than 30 weeks due to a delay in obtaining equipment. The egg production period was for 20 weeks. Tabulation of egg production commenced April 2, 1974, (3 weeks after light stimulation) and concluded on August 20. Settable eggs are defined as total eggs minus broken, 1 Bressler Plastic Poultry Flooring. A molded smooth polyethylene flooring of 1 x 2-inch mesh. Aspec Corp., P.O. Box 391, Nazareth, Pa Simulated rubber (Kraton) molded floors with egg catch roll-out. Limited Plastics, P. 0. Box 89, Lemoncove, Calif

143

144 soft shell, pee wee, double yolked, gpnd deformed eggs* Fertility and hatchability data were obtained at a hatchery 3 from eggs pen marked every fourth week. Results 1n Floor Pens The total egg production data suggest*a somewhat variable response to bird density, but in general there was a downward trend in production as bird density increased from 8 to 16 hens per pen. A technique was employed wherein hens were eliminated from the hen day totals if they were out of production for 8 or more days due to broodiness or other conditions. These data are indicated in the column entitled less pauses. There was a steady decrease of 2% in egg production as bird density increased by each interval. The small least significance difference required indicated that these effects were significant. The data show that virtually all of the eggs laid in floor pens were settable. There was a body weight loss of 692 g. (approximately 1.5 Ib.) per hen from time of inception through 20 weeks of production. There were no significant differences in feed required per hen per day, fertility, hatchability o fertile eggs, or egg weight. - 1 Results in Cages? Having two hens per cage resulted in reduced total egg production, although the levels of egg production were much less different when pauses-were omitted. With two hens per cage, one of the hens often went out of production. Tota^ egg production was similar to floor pen controls (13 or 16 hens per pen x ) when one hen was housed per cage. Egg breakage was excessive when the polyethylene floors were used, and was also somewhat greater than in floor pens when the simulated rubber was employed. Observing the treatment with one hen per cage and simulated rubber floors, there were 46% settable eggs. This compares to 55.4% settable eggs with the hens in floor pens and represents a significant difference. The hens in cages did not lose weight during production as did those in floor pens. Amount of feed per hen per day, egg fertility, and egg weight were not significantly different among the cage and floor treatments. There were significant differences in hatchability of fertile eggs; the cage treatment which yielded the greatest number of good eggs (one hen per cage, with soft floor) had a hatchability value of 67.5%, which was significantly lower than any floor treatment. Conclusions With turkey breeder hens in floor pens, there was a significant increase in egg production intensity as floor space per bird increased from to square meters. The decision on how much space to provide each hen will, of course, depend more on profitability than on small differences in production. The data on hens in cages showed that total egg production per hen was higher with one rather than two birds per cage. The hens on simulated rubber flooring had fewer broken eggs than those on polyethylene flooring. The production of settable eggs was less with hens in cages than with hens in floor pens. During production, hens in cages maintained body weight while those in floor pens lost body weight. There were no significant differences in feed consumption, fertility, or egg weights between cage and floor environments. However, hatchability of fertile eggs was significantly lower with cages as compared to floor pens. * Appreciation is expressed to Willmar Poultry Co., Willmar, Minn., for assistance in obtaining fertility and hatchability data.