POSTPARTUM REPRODUCTION AND METABOLITES IN THE COW AS AFFECTED BY ENERGY AND PHOSPHORUS INTAKE Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY JANE ALLISON CARSTAIRS _ 1975 / (N /7 ABSTRACT POSTPARTUM REPRODUCTION AND METABOLITES IN THE COW AS AFFECTED BY ENERGY AND PHOSPHORUS INTAKE By Jane Allison Carstairs A study was designed to examine the influence of energy and phosphorus intake on postpartum reproductive function and metabolite levels of heifers. High and low energy and phosphorus refer to 100 and 75% of requirements. Primiparous Holsteins were assigned at parturition in a 2x2 factorial design in groups of six to: (1) high energy, high phosphorus (HEHP); (2) high energy, low phosphorus (HELP); (3) low energy, high phosphorus (LEHP); (4) low energy, low phosphorus (LELP). Treatment extended through 84 days postpartum. At this time, heifers were returned to a standard herd ration and observed for a further 21 days, the experiment ending at 105 days postpartum. Heifers were bled twice weekly and serum was assayed for several hormones and metabolites. Reproductive status was monitored via serum pro- gesterone and rectal palpation. Mean energy statuses (meal/day) for high and low energy groups were 5.4 and -3.4, (P < .0005). Mean phosphorus statuses (g/day) for high and low phosphorus groups were 16.9 and -5.0 (P < .0005). Mean milk yields (kg/day) for high and low energy groups were 17.4 and 20.1 Jane Allison Carstairs (P = .l2). High energy groups had higher (P = .19) incidence of ill health compared to low energy groups. Body weight change, serum urea nitrogen, glucose, non-esterified fatty acids (NEFA) and insulin were highly related to energy status. Mean weight changes (kg/week) for high and low energy groups were 1.75 and -0.14 (P = .12). Urea nitrogen levels (mg/d1) were higher (P < .0005) in low energy groups than high energy groups, means were 15.8 and 10.4. Mean glucose levels (mg/d1) for high and low energy groups were 75.5 and 71.6 (P = .06). Mean NEFA levels (nmoles/ml) for high and low energy groups were 221.3 and 258.9 (P = .3). Mean NEFA levels (nmoles/ml) for high and low phosphorus groups were 221.8 and 258.5 and were also different (P = .3). Mean insulin levels (uU/ml) were higher (P < .0005) in high energy groups than in low energy groups, means were 16.5 and 6.0. Both insulin and glucose were negatively related to milk yield. Serum phosphorus was a good indicator of phosphorus status. Mean serum phosphorus levels (mg/d1) for high and low phosphorus groups were 7.1 and 5.9 (P = .02). Serum levels for the low phos- phorus groups reflected marginal phosphorus deficiency. Serum calcium, cholesterol, creatine phosphokinase, aspartate aminotransferase, and alkaline phosphatase were influenced by energy and phosphorus intake but were not useful as indicators of energy or phOSphorus status. Cholesterol levels increased (P < .001) with time postpartum (r = 0.9), as did alanine aminotransferase (r = 0.88, P < .001). Regression analysis on these data on energy and phos- phorus status tended to support the analysis of variance. After Jane Allison Carstairs the end of the treatment period, values returned to normal by 105 days. Groups imbalanced for energy and phosphorus took longer (P = .12), postpartum to reach 3 ng/ml progesterone than balanced groups. Means were 41 and 31 days. Imbalanced groups also took longer (P = .27), to experience first ovulation than balanced groups. Means were 27 and 21 days. For second and subsequent estrus', the tendency was for high energy to lengthen and high phosphorus to shorten estrous cycle length. Mean incidence (%) of undetected heats for imbalanced and balanced groups was 40.6 and 56.4 (P = .14). Mean number of days non-pregnant for balanced and imbalanced groups were 145 and 117 (P = .14). High phosphorus groups needed more (P = .08) services/ conception than low phosphorus groups, 3.3 and 2.2, respectively. Cystic follicles were equally distributed among groups. This study demonstrates that postpartum energy and phosphorus intake influence metabolites and reproductive function in primiparous Holstein cattle. POSTPARTUM REPRODUCTION AND METABOLITES IN THE COW AS AFFECTED BY ENERGY AND PHOSPHORUS INTAKE By Jane Allison Carstairs A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Dairy Science 1975 ACKNOWLEDGMENTS I would like to acknowledge the following people for their help in executing and writing this thesis: My Major Professor, Dr. Roy Emery, for his enthusiasm, encouragement, stimulating discussion and mostly for his excellent guidance throughout my graduate career. The members of my committee, Dr. Allen Tucker, for his guidance and Dr. Loran Bieber, for his advice with the non-esterified fatty acid assay and his ever positive attitude. Dr. David Morrow, for the reproductive examination of the cows in this experiment and also for his excellent advice and dis- cussion of the reproductive data. Dr. Charles Lassiter for making the facilities of the Depart- ment of Dairy Science available for the execution of this research. Dr. John Gill and Dr. Roger Neitzel for statistical advice and computer programming. James Liesman for his invaluable help in sample collection. Finally, my thanks to Wilson Cunningham and all my friends who made my graduate experience easier and more fulfilling. ii TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES . INTRODUCTION . . . . . . LITERATURE REVIEW . Level of Energy . . . . Phosphorus and Reproductive Function MATERIALS AND METHODS . Experimental Design . Blood Serum Hormones . Metabolites and Enzymes in Blood Serum RESULTS . Feed Intake . . . . . . . . . . . . . Energy and Phosphorus Status Body Weight . Milk Yield Milk Fat Percent Milk Protein Percent Serum Calcium . . . . . . . . . . . . . . . . . . . Serum Phosphorus Creatinine . . . . . . . . . . . . . . . . . . . Urea Nitrogen . Total Cholesterol . Glucose . Alkaline Phosphatase Alanine Aminotransferase Aspartate Aminotransferase Creatine Phosphokinase Insulin . Non- Esterified Fatty Acids Reproduction Data . Multiple Regression and Simple Correlation Analysis . iii Page vii 19 19 23 24 29 29 29 35 36 39 39 39 4O 43 43 44 45 45 48 48 51 51 S6 57 64 Page Health . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Post Experimental Responses in Energy and Phosphorus Status, Milk Yield and Serum Parameters . . . . . . . . . . 67 DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 APPENDIX . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . 98 iv Table 10. 11. 12. 13. 14. LIST OF TABLES Composition of grain mixes (%) Dry matter, net energy and phosphorus values used to calculate NE intake and phosphorus intake Methodology of assays used by the Hycel Mark K clinical autoanalyzer . . . . . . . . . . . Mean intakes (kg/day) of corn silage, hay and grain for the first 12 weeks of lactation . Mean energy status (meal/day) for the first 12 weeks of lactation . . . . . . . . . . . Mean phosphorus status (g/day) for the first 12 weeks of lactation . . . . . . . . . . Mean serum calcium levels (mg/d1) for the first 12 weeks of lactation . . . . . . . . . . Mean serum phosphorus (mg/d1) for the first 12 weeks of lactation . . . . . . . . . . . Mean serum total cholesterol (mg/d1) for the first 12 weeks of lactation . . . . . . Mean serum glucose levels (mg/d1) for the first 12 weeks of lactation . . . . . . Mean serum aspartate aminotransferase levels (U/l) for the first 12 weeks of lactation . Mean serum creatine phosphokinase levels (U/l) for the first 12 weeks of lactation . Mean insulin levels (uU/ml) for the first 12 weeks of lactation . . Means of glucosezinsulin ratio for the first 12 weeks of lactation . Page 22 29 36 40 43 44 48 52 55 Table Page 15. Means of insulinzglucose ratio for the first 12 weeks of lactation . . . . . . . . . . . . . . . . . . . 55 16. Standard partial regression coefficients for milk yield, glucose and insulin . . . . . . . . . . . . . 56 17. Mean NEFA levels (nmoles/ml) for the first 12 weeks of lactation . . . . . . . . . . . . . . . . . . . 57 18. Number of days postpartum for serum progesterone to reach 3 ng/ml . . . . . . . . . . . . . . . . . . . . 59 19. Number of days postpartum to first ovulation . . . . . . . 59 20. Mean overall interestrual interval . . . . . . . . . . . . 62 21. Treatment means for number of undetected heats, cystic follicles, days non-pregnant and services per conception . . . . . . . . . . . . . . . . . . . . . 63 22. Standard partial regression coefficients . . . . . . . . . 65 23. Simple correlations . . . . . . . . . . . . . . . . . . . . 66 24. Mean incidence of disease for the first 12 weeks postpartum . . . . . . . . . . . . . . . . . . . . 66 25. Mean energy and phosphorus status and milk pro- duction for the two week period immediately following the end of the experimental treatments . . . . 68 26. Mean intakes (kg/day) of grain, hay, haylage and corn silage for the two week period immediately following the end of the experimental treatments . . . . 69 27. Mean insulin levels (uU/ml) for the two week period following the end of treatment . . . . . . . . . . . . . 73 A-l. Composition of scintillation fluid . . . . . . . . . . . . 95 A-2. Mean interval to second, third and fourth estrus in the first 13 weeks of lactation . . . . . . . . . . . 96 A-3. Mean interval to second, third and fourth estrus in the first 15 weeks of lactation . . . . . . . . . . . 97 vi LIST OF FIGURES Figure Page 1. Mean energy status (meal/day) for the first 12 weeks of lactation . . . . . . . . . . . . . . . . . . . 32 2. Mean phosphorus status (g/day) for the first 12 weeks of lactation . . . . . . . . . . . . . . . . . . 34 3. Mean milk yield (kg/day) for the first 12 weeks of lactation . . . . . . . . . . . . . . . . . . . 38 4. Mean serum calcium (mg/d1) for the first 12 weeks of lactation . . . . . . . . . . . . . . . . . . . 42 5. Mean serum cholesterol (mg/d1) for the first 12 weeks of lactation . . . . . . . . . . . . . . . . . . 47 6. Mean serum alanine aminotransferase (U/l) for the first 12 weeks of lactation . . . . . . . . . . . 50 7. Mean serum insulin (uU/ml) for the first 12 weeks of lactation . . . . . . . . . . . . . . . . . . . 54 vii INTRODUCTION The relationship between nutrition and reproductive efficiency has been studied for many years. This efficiency of production is of even more importance today in view of the growing need for food production in a world with an expanding population. Infertility may be the result of a single, or a combination of nutrient deficiencies. Combined deficiencies have caused difficulties in assessing the effect of dietary intake on reproduction. Deficiencies in energy and phosphorus have been implicated in delayed onset of puberty, postpartum estrus and other reproductive failures. These effects of energy and phosphorus are discussed in the Review of Literature. Previous studies have tended to be oriented towards a specific aspect of the deficiency. Some workers have emphasized the hormonal changes resulting from various deficiencies. Others have examined mainly metabolite changes and only general reproductive differences, such as conception rate. This study was designed so that both hormonal and metabolic responses could be integrated and related to the nutrient deficiency. In order to do this, parameters were measured which would indicate the general status of the cow over the postpartum period. Primiparous Holstein heifers were used since these animals were still growing and consequently the nutrient deficiency should have caused more of a stress on reproduction and lactation. Energy and phos— phorus balances were calculated so that a precise description of the status of each animal would be available. In addition several serum metabolites and enzymes were measured so that responses to the diets could be monitored. Serum progesterone and rectal palpation were used to monitor reproductive status. Hopefully, this investigation will describe, more fully than before, the relationship between energy, phosphorus, metabolites and reproduction. Hopefully, the parameters measured can be related to differences in performance of the heifers. Perhaps also, certain of these parameters can be used in the future as indicators of energy and phosphorus status and enable marginal deficiencies to be detected. LITERATURE REVIEW Many theories have been postulated as to the cause of sterility, anestrus, cessation of ovulation and other reproductive failures. Nutrition has been one of the factors implicated in infer- tility and much work has been devoted to this area. Rank deficiencies of certain nutrients are known to impair reproductive function. However, the effects of excesses, marginal levels of nutrients or nutritional imbalances have not been fully elucidated. Reproductive efficiency at this point in time is of prime importance to the pro- ducer. The characterization and application of the relationship between nutrition and infertility will greatly aid this efficiency. In this review the influence of energy and phosphorus on postpartum estrus and fertility will be discussed. Possible mechan- isms for the action of these nutrients on reproductive function will also be examined. Level of Energy The cow divides her food, excluding maintenance needs, between milk production and liveweight gain (Broster, 1972). This division of nutrients is related to the milk yield potential of the cow, the stage of lactation and the level of food intake. It has been suggested the liveweight change is a measure of plane of nutrition and evidence has been reported that there is a positive relationship between plane 3 of nutrition and fertility (Broster, 1973). The hypothesis that low planes of nutrition can lead to anestrus and reduced conception rates is widely accepted. The evidence in the literature, however, is not quite as clear cut and varying reasons as to how this effect is mediated have been suggested. A ten percent fall in liveweight from calving to the time of mating has been associated with infertility (McClure, 1970). Simi- larly, a one percent change in conception to first service per one percent change in liveweight has been noted by King (1968). In that study the cows that were gaining weight had a much higher conception rate at first service than those that were losing weight. Attempts to repeat and confirm such statements have not been as successful. Although Boyd (1972) did find evidence for increased conception when cows were gaining weight, fertility differences between weight change groups were not statistically different and he concluded that weight loss had no adverse effect on fertility. Similarly, Broster (1973) quotes work by Munro (1970, unpublished) in which he could find no conclusive relationship between fertility and liveweight change. There appears to be a limitation to the relationship of liveweight and fertility and the factors that may influence it. The question of the greater vulnerability of the cow with a higher yield potential to the adverse effects of undernutrition com- pared with the lower yielding cow is controversial. King (1968), reviewing work in this area found no evidence to suggest that cows yielding more milk showed reduced fertility compared to lower yielding cows. Similarly, returns to service were reported to be equally distributed amongst cows of differing milk yield potential in New Zealand herds by Simpson (1972). From here, however, the evidence is not as clear. Simpson's work was strongly criticized by Dawson (1972) who suggested that Simpson analyzed his data in such a way that it was biased towards finding no differences. Dawson (1972) quoted recent European evidence disagreeing with Simpson's (1972) results. In the United States a positive correlation between milk production and the interval from calving to conception as well as services per conception, especially for animals producing more than 7272 kg of milk per lactation, has been reported (Morrow, Roberts, McEntee and Gray, 1966). Hewett (1968) showed in Swedish herds that repeat breeders averaged 86.4 kg more milk during the first 120 days of lactation when compared with contemporary controls. He classified a repeat breeder as a cow which was declared non-pregnant after at least three inseminations or a cow which became pregnant after at least four inseminations. Each repeat breeder cow had a strictly selected control cow. This control had to belong to the same herd, be born within the same four months, calve within 35 days of the repeat breeder cow and have conceived within 100 days of calving after not more than three inseminations. This study may have some limita- tions due to the fact that two—thirds of the herds examined had less than 15 cows and in some cases a control could not be found and thus some repeat breeders must have been excluded from the analysis. More recently 393 calving intervals were used to compare the effects of breeding at the first postpartum estrus after 74 days as modified by two different levels of nutrition and tWo different genetic levels for production (Whitmore, Tyler and Casida, 1974). The incidence of silent heats for first postpartum ovulation was greater in the high nutrition groups. The interval to first postpartum estrus was longer in cows with superior genetic potential for milk production than for those which were genetically inferior producers and longer for cows on high compared to average nutrition. Finally, Simpson (1972), who did not report any infertility associated with increased milk yield did add a cautionary note in his report which although it does not explain the results of the scientific research in this area does sug- gest why the relationship between milk yield and infertility is commonly accepted. He notes that this relationship is observed, particularly by veterinarians, because poor producers are less likely to be presented for treatment, thus, the proportion of infertile cows could be biased toward the cow producing more milk. The situation is obviously in need of clarification. Turning now to the effects of energy intake alone, Dunn, Ingalls, Zimmerman and Wiltbank (1969) studied the effects of energy intake on reproductive performance of two year old heifers from 140 days prepartum to 120 days postpartum. Two levels of digestible energy, low and high, were fed prepartum. At parturition the low group was divided further in half: one half were fed a moderate diet and one half were fed a high diet. The high group was divided into thirds, one third were fed a low diet, one third moderate and one third high. It should be noted that these designations were for convenience only, all were below the levels recommended by the National Research Council. Pregnancy rate 120 days after calving was directly related to the postpartum energy level. Eighty seven percent of the cows fed the high level after calving were pregnant 120 days postpartum com— pared with 72% of those fed the moderate level and 64% of those fed the low energy level. Estrus was delayed in the heifers receiving the low level of energy prepartum. The detrimental effects of feeding low levels of energy just prior to parturition were partially overcome by feeding higher levels postpartum. Evidence to date would suggest that adequate energy intake is more critical than protein intake for maintaining reproductive function (Wiltbank gt 31., 1962; Wiltbank, Rowden, Ingalls and Zimmerman, 1964; and McClure, 1968a). Assuming that reproductive function is impaired with decreased energy intake, several hypotheses as to how this effect is mediated have been suggested. Some years ago it was hypothesized that restricted energy intake caused a hypo- glycemia and that this in turn caused reproductive hypofunction. This suggestion was made after fertility was significantly improved in an Australian herd when randomly selected cows were fed an additional 5.5-6.4 kg of hay per cow per day from calving to 3 weeks after mating (McClure, 1965). The involvement of hypoglycemia was noted because the blood levels of glucose of the fertile cows at the time of mating were rising and averaged 28 mg/dl whereas the levels in the infertile cows were falling and averaged 22 mg/dl. Body weight followed the same pattern. This hypothesis was reiterated a few years later (McClure, 1968a) when low fertility syndrome herds in Australia and New Zealand were characterized as returning to estrus after irregular but often lengthened intervals. The affected cows lost 5-10% of their body weight between parturition and mating and also had low blood glucose levels of between 20 and 30 mg/dl. In addition, supplementation with energy rich concentrates to the rations of these infertile hypoglycemic herds resulted in significant increases in both blood glucose levels and fertility (McClure, 1972). Further support was given by a study in which cows were made hypoglycemic with insulin. The cows mated 0—2 days after daily insulin treatment for 3 or 4 days and those treated with insulin daily during the first four days after mating were significantly less fertile than either cows treated at other times with insulin or control untreated cows (McClure, 1968b). These results are similar to those in mice in which hypoglycemia was induced by fasting or insulin or glucose metabolism was inhibited by 2-deoxy-D-glucose (McClure, 1967). The hypothesis was that the hypothalamus was failing to control the adenohypophysis as a direct consequence of failure of the hypothalamus to be supplied with, or to utilize glucose (McClure, 1967). It should be noted that the nature of the malnutrition described by McClure (1968a) was not well characterized. He observed infertility under two different conditions: (a) when the pasture was dry, of low digestibility and deficient in protein, phosphorus and carotene, or (b) when the pasture was green, lush and young (McClure, 1968a). Thus, his underfeeding is not just a simple case of energy deficiency but appears to be complicated by other nutrients. It should also be noted that the nutritional and hypoglycemic conditions he describes are extreme when compared with a concentrate fed dairy cow in North America. Support, however, to McClure's hypothesis is lent by Howland, Kirkpatrick, Pope and Casida (1966) who also suggest that the hypoglycemia exerted its influence through depression of hypothalamic function thus resulting in loss of ovarian activity. Additionally, a significant negative correlation was found between plasma glucose level and postpartum interval to occurrence of a 10 mm follicle and ovulation in a study of primiparous cows (Oxenreider and Wagner, 1971). Both energy and lactation had a significant effect on plasma glucose levels during the first 8 weeks postpartum. Lactation and low energy significantly delayed postpartum follicular growth and ovulation. Throughout these studies the independence of energy level, lactation and glycemia can be questioned. It is diffi- cult to say that one or the other is the main cause or effect. One of the limitations in interpreting these data is that hormones involved in reproduction were not examined, thus, conclusions must to some degree be questionable. At a more physiological level, the effects of energy intake on postpartum ovarian activity and changes in serum hormone levels pre and postpartum have been studied. Unfortunately, metabolite levels were not concurrently measured with alterations in reproductive function. Normal ovarian and hormonal changes have been reported by numerous workers such as Echternkamp and Hansel (1973). The effects of different energy intakes prepartum on'post- partum progesterone and estradiol concentrations in beef heifers have been examined (Corah, Quealy, Dunn and Kaltenbach, 1974). The rations were approximately equal in crude protein content and postpartum rations were adequate in energy. Blood was collected from 14 days 10 prepartum to 20 days after the first postpartum estrus. There was no significant effect of nutrition on peripheral concentrations of progesterone or estradiol either prior to, or following, parturition. Reducing energy intake for 100 days prior to calving markedly reduced body weight and fat cover but in contrast to the work of Dunn 23 El. (1969) and Bellows, Varner, Short and Pahnish (1972), the interval from parturition to first estrus was not influenced by the reduced energy intake. Of particular interest in this study was the marked elevation of progesterone only in those cows conceiving, suggesting that a period of elevated progesterone may be necessary for conception at first estrus. This effect has also been noted in dairy cows main- tained on high and low levels of nutrition (Folman, Rosenberg, Herz and Davidson, 1973). Cows that conceived after one insemination had significantly higher progesterone concentrations during the estrous cycle preceding insemination than did cows that did not conceive. The concentration of serum progesterone required appeared to be around 3 ng/ml. Level of nutrition had a profound effect in cows that needed more inseminations for conception. During the luteal phase preceding insemination, cows that conceived after the first insemination gained weight whereas cows that did not conceive lost weight, the difference approached significance. In contrast to the work in which level of nutrition did not appear to affect peripheral progesterone levels per_§e, other studies have shown an effect. Plasma luteinizing hormone (LH) and progesterone concentrations in groups of six Holstein heifers fed 100 or 62% of energy requirements have been examined during three estrous cycles 11 (Gombe and Hansel, 1973). Plasma LH increased progressively from the first to the third estrous cycle in heifers fed the low energy ration. This increase was first seen in the maximum LH of the cycle but by the third cycle this increase was seen throughout the cycle and was also higher than that of the control heifers. During the first cycle progesterone was slightly higher in the cows fed 62% compared to 100% of their energy requirements but became progressively lower in the subsequent cycles. When the corpora lutea were examined on the tenth day of the third cycle it was found that total progesterone and pro- gesterone concentration in the corpora lutea were lower in the heifers fed 62% of their energy requirements than in their normal counterparts. Evidently ovarian hypofunction in cases of energy deficiency is not due to reduced circulating LH as was previously thought by such workers as Wiltbank, Rowden, Ingalls, Gregory and Koch (1962). The first effect may be a reduced ability of the ovarian tissue to respond to LH. This effect of decreasing progesterone has been noted before in heifers receiving 25% of the total feed consumed by the controls (Donaldson, Basset and Thorburn, 1970). Declines in plasma progesterone have occurred within even 5 days of the reduction in feed intake (Hill, Lamond, Dickey and Niswender, 1970). The undernutrition which was about 85% of maintenance, also temporarily reduced the number of medium sized follicles, altered the length of the estrous cycle and reduced the proportion of heifers with normal fertilized ova. Weights of corpora lutea formed in undernourished heifers were about 70% of control values. It should be noted that Donaldson §£_al, (1970) and Hill §t_§l, (1970) varied both energy and protein content of the 12 ration whereas Gombe and Hansel (1973) reduced only energy intake. Evidence suggests that the observed reduction in plasma progesterone is a reflection of a smaller corpus luteum, containing less total and concentration of progesterone (Gombe and Hansel, 1973). Further, the simplest suggestion is that the first effect of this restricted energy intake is at some step in steroidogenesis within the corpus luteum, causing the observed reduction in plasma progesterone levels. Several additional suggestions have been made as to how this effect is mediated (Gombe and Hansel, 1973). Accumulating evidence suggests there is an interrelationship of energy intake and reproductive function in the cow. The character- istics of this relationship are still not clearly defined and the mechanism of action is still not fully elucidated. The evidence does suggest that restricted energy intake can cause reproductive failures and that this relationship is not simple but must be considered con- currently with body weight, stage of lactation and other physiological factors. Phosphorus and Reproductive Function Variations in blood phosphorus levels in the bovine have been partially explained by pregnancy, parturition, lactation and age (Lane, Campbell and Krause, 1968). These changes in blood phosphorus levels with physiological state are not surprising in view of the fact that phosphorus functions in energy metabolism, skeletal growth _ and milk production. The National Research Council (1971) quotes normal values for bovine plasma phosphorus as 4-6 mg/dl for cows and 6-8 mg/dl for l3 calves under one year of age and states that unlike some other nutrients, blood concentrations will decline before any clinical signs develop. A relationship between phosphorus and fertility was postu- lated many years ago and appears to be widely accepted. The evidence for this relationship is often questionable since the phosphorus deficiency required to impair fertility is usually extremely severe. This does not tell how marginal the condition has to be in order to show reproductive effects. Clinical signs of phosphorus deficiency are easily noticeable and easily treated. A sub-clinical or marginal deficiency may be even more important as it may indeed cause production losses and yet not be overtly detectable. Research in the severely deficient animal is quite well docu- mented while research in the marginally deficient animal is indeed lacking. The situation is complicated when it is realized that most reports are confounded, knowingly or not, by deficiency of another nutrient. The widespread occurrence of a natural deficiency of phosphorus affecting cattle has been described by Tuff (1923), Theiler, Green and du Toit (1924), Eckles, Becker and Palmer (1926), Hart and Guilbert (1928) and numerous others. As early as 1906 however, in a study of the effect of phosphorus compounds in the diet of milking cows, it was noted that phosphorus deficiency was accompanied at times by the cessation of estrus (Jordon, Hart and Patten, 1906). South African workers were among the first to report repro- ductive hypofunction in cattle maintained on veld deficient in phosphorus. In one study, it was shown that reproductive efficiency was increased from a 51% calf crop in the control group to an 80% 14 calf crop in the experimental group, simply by supplementing the naturally deficient animals with bone meal (Theiler, Green and du Toit, 1928). It was also noted that blood phosphorus was around 2.3 mg/dl in phosphorus deficient pastures whereas with bone meal supplementation the concentrations doubled (Malan, Green and du Toit, 1928). Hypo- phosphatemia was noted in animals being fed on a hay-oats phosphorus deficient diet also but the effects on reproduction were not mentioned (Palmer and Eckles, 1927). Years later in Ireland both clinical and sub-clinical aphosphorosis of cattle was noted in phosphorus deficient pasture. The cattle developed all the symptoms of phosphorus defi- ciency plus "temporary sterility" (Sheehy, 1946) or "anestrus or estrus with repeated failures to conceive after service" (O'Moore, 1950). Hypophosphatemia was noted in all these cases and phosphorus supplementation seemed to solve the problem. In the United States, Eckles, Palmer, Gullickson, Fitch, Boyd, Bishop and Nelson (1935) were some of the first to try and obtain experimental data regarding reproductive function in cattle on controlled uncomplicated phosphorus deficient rations. They were following the example of workers such as du Toit, Malan and Greenewald (1934). Eckles §£_al, (1935) did not have a definite feeding level of phosphorus. They tried to supply each cow so as to maintain the plasma phosphorus at approximately 2.5 mg/dl, about one-half normal. This concentration was selected as being representative of cattle in a severely deficient area. These concentrations were accomplished and indicated severe phosphorus deficiency. The daily milk yields in these cows were between 3.2 and 11.1 kg, somewhat low by today's 15 standards, and probably did not cause a large strain on the metabolic system. Even though this experiment was run for about three years no evidence was gained to show that the phosphorus deficiency influenced the estrous cycles of the cows. All of the cows came into heat quite regularly, it did appear, however, that breeding efficiency was reduced. It was suggested from this study that the disturbances in estrus and reproductive function in naturally deficient areas was probably due to the complication of other nutrient deficiencies. Concurrently with the study on cows, rats were fed a diet containing 0.1% P and the following types of reproductive failures were noted: complete cessation of estrus sometimes following a successful pregnancy or successful lactation; regular estrus cycles but no normal breeding; fetal resorptions in 33233; irregular cycles; undersized weak litters (Eckles g£_al., 1935). More recently a herd with heifer infertility problems was described (Morrow, 1969). These problems were attributed to phos- phorus deficiency presumably resulting from depleted phosphorus in the soil and consequently in the crops. Calculations of intake and requirements for protein, energy, calcium and phosphorus showed that the phosphorus intake was deficient. Clinical signs consisted of rough coat, depraved appetite and infertility. Low blood phosphorus was also observed. The condition was treated with the feeding of dicalcium phosphate. Blood levels returned to normal and fertility was restored. The phosphorus deficiency did not appear to affect the length of the estrous cycles or the frequency of silent estrus, but the problem appeared to be in conception. C3 16 The effects of a combined phosphorus and protein deficiency have also been examined. Cows were observed for at least 24 months and up to 59 months on a protein and phosphorus deficient ration (Palmer, Gullickson, Boyd, Fitch and Nelson, 1941). It was felt that this was more analogous to a real deficiency in the field. The cows showed delayed sexual maturity, silent heat but normal, regular ovulation and conception was not interfered with. Breeding efficiency was not reduced. Feeding trials with cows on phosphorus deficient pastures in the Northern territory of Australia showed positive effects regarding fertility (Hart and Mitchell, 1965). Supplementation of range cows with 8 g P/head/day improved body weight and fertility in the lactating cows. A pregnancy rate of 60% in the treated group compared with 41% in the controls was observed. The authors did com- ment, however, that the provision of protein in their opinion was of equal, if not more importance, than phosphorus. Phosphorus and calcium have also been linked in causing infertility. Hignett (1950) suggested from the results of a herd survey in England that breeding efficiency of cattle might be related to the Ca:P ratio of the feed consumed, always assuming that the amount of each element was sufficient in itself. It was suggested that a 1:1 ratio was ideal for fertility. The risk of herd fertility being impaired was great when the ratio was 2:1 or more, especially when the phosphorus intake was only slightly in excess of minimum requirements. The feeding of too much phosphorus was also warned against since Hignett (1950) associated infertility with excess levels of phosphorus. After work was completed on another 802 cows, it was 17 concluded that the generally accepted recommendations for phosphorus (23 g/day for maintenance plus 19 g for each 4.5 kg of milk) were not enough for high fertility in dairy cattle (Hignett and Hignett, 1951). It was observed later that fertility was decreased in rapidly growing heifers which were fed rations deficient in manganese and unbalanced in calcium and phosphorus (Hignett, 1959). Hignett and Hignett (1951) warned against the exaggeration of the importance of the influence of calcium and phosphorus intakes on fertility and their caution is justified. Most of their work was of the field survey type and only feed analyses and breeding data were used. No blood parameters were measured. As support to this caution, a large scale controlled experiment was carried out at Weybridge in 1958 and 1959 and this failed to demonstrate any signifi- cant relationship between the Ca:P ratio of the diet and fertility in dairy heifers, whether the herd fertility was high or low (Little- john and Lewis, 1960). A high Ca:P ratio depressed growth rate. These data are probably more reliable than the survey work reported by Hignett (1950) and Hignett and Hignett (1951). In contrast, it has been suggested that phosphorus deficiency acts at the anterior pituitary level and causes "cessation of estrus, lack of sexual libido, testicular and accessory organ atrOphy" (Guilbert, 1942). Also the "ovary in phosphorus deficiency becomes quiescent and infantile" (Guilbert and Hart, 1930) and (Kleiber, 6055 and Guilbert, 1936). Finally, it has been reported that phosphorus deficiency lowers the total efficiency of energy utilization mainly by lowering 18 the appetite and secondly by lowering the partial efficiency of energy utilization. It does not seem to influence fasting catabolism (Kleiber et 31., 1936). From the evidence to date, it does appear that there is some link between severe phosphorus deficiency and infertility. The extent to which this effect is related to deficiencies in other nutrients such as protein, other minerals, or energy is not yet entirely clear. The mechanism for its action has not been fully described. More work needs to be done, not only in the more marginally deficient animal, but also on how phosphorus interacts with other nutrients and what influence this may have. It is to be hoped that the following study will answer some of these questions and help further the understanding of nutrient deficiencies and their relationship to fertility. MATERIALS AND METHODS Experimental Design Two levels of energy and phosphorus (100 and 75% of N.R.C., 1971), requirements were fed to 24 primiparous Holstein heifers from the Michigan State University herd using a 2x2 factorial design. These heifers weighed between 437 and 625 kg at parturition. Heifers were assigned at parturition in groups of six to one of four treatment groups: (1) high energy, high phosphorus; (2) high energy, low phosphorus; (3) low energy, high phosphorus; (4) low energy, low phosphorus. Heifers were assigned so as to represent equivalent genetic potential within each treatment group. Treatment began on the day of parturition and extended to 84 days postpartum. At that time, heifers were returned to the herd ration and measurements were taken for a further 21 days (the experiment ending at 105 days postpartum). The high energy rations were corn silage treated with non protein nitrogen fed until feed refusals were about 10% of intake and grain fed to the maximum which the heifers would consume. Grain was fed daily at 0800 and 1300 hours. Grain intake was restricted to 2.3 kg at 0800 hours for the low energy rations. All cows were fed 2.3 kg of the alfalfa brome hay and the corn silage once daily. Phosphorus was restricted by replacing the dicalcium phosphate in the grain mix with corn and CaCO3 (Table l). 19 20 Table 1.-—Composition of grain mixes (%). High Energy Low Energy component High P Low P High P Low P Shelled Corn 73.9 74.3 —- 1.8 Soybean Meal 18.8 18.8 86.1 86.1 Molasses 5.0 5.0 5.0 5.0 Dicalcium Phosphate 1.2 -- 6.0 -- Limestone 0.5 1.3 0.2 4.4 Trace Mineral Salt 0.6 0.6 2.7 2.7 Total 100.0 100.0 100.0 100.0 Daily records of disease, feed intake and milk production were kept. Composite morning and evening milk samples were taken every two weeks and analyzed for fat and protein. Body weights were mea- sured at parturition and once a week thereafter. This was done as far as possible at the same time each week. Blood was collected at approximately 1000 hr twice weekly throughout the experiment via the tail vein using a 20 gauge 25 mm vacutainer needle and 20 ml vacutainer tubes (Becton Dickinson Inc., Rutherford, N.J.). The blood was allowed to stand at room temperature for one hour after collection and was then placed at 5 C for 5 hours. It was then centrifuged at 1000 x g for 20 minutes and the serum was removed. Serum was stored at -15 C until assayed. Reproductive status was determined by the concurrent examination of serum proges- terone and weekly rectal palpation data. 21 Feed samples were taken and analyzed for dry matter, protein and phosphorus (analysis was done in the analytical labOratory of the Department of Biochemistry, M.S.U. or in the analytical laboratory at Wooster, Ohio). From these composition measurements of the feed and N.R.C. feed composition tables a value for Net Energy (NE) and phos— phorus was calculated for each feed used in the experiment (Table 2). From these values and the feed intake data, milk production and com- position measurements, a calculated energy balance was determined for each heifer once every two weeks over the first 14 weeks of lactation. First, energy requirements (mcals NE lactation/day) were cal- culated from N.R.C. (1971) tables using the two week average for body weight, milk production and milk fat percent. Then energy intake was calculated (mcal NE lactation/day) using feed intake data and the energy values of the feeds. Thus, energy balance was calculated from the difference between requirements and actual intake. Phos— phorus balance was calculated in the same way. Energy balance indi- cates the difference between input and output of energy. A require- ment is estimated according to N.R.C. standards to account for the energy needed for maintenance and milk production. If the energy consumed is not enough to meet these requirements then the heifer would have to supply this energy difference from endogenous sources or decrease production. The heifer would under these circumstances be in negative energy balance or status. If the amount consumed is equivalent to the amount required then the animal is at zero balance (or 100% N.R.C.). 22 .mfimmn we: m :o mspocmmocm mom mosHm> nocuo HH< .mwmmn xpw m :o commopmxo mosfim> & mu paw mz HH< .owco .aoumooz um mmMHOpmuoan Hmofiuxamcm on» an pmnxflmem .mfimmn poppme Aye m so commonmxo acoohom manozmmozma .mvm.o -- mo.~ oo.om .mxo u:o-eA\~N\m owmaam cuou .omm.o -- mo.H oo.mm ea\H~\m-45\eH\m mwaaam choc .omm.o H.NH mo.~ oo.me VA\mH\m-eA\AN\N ommaam caou .oom.o o.NH mo.H 00.44 4A\©N\N-¢A\Hm\a omaaam cpou moH.o o.ma me.H oo.ov 4A\Hm\a-mk\w\ofl mwaaam cuou .oom.o m.m~ $3.2 oa.ww .mxo eco-VA\aN\N aw: 4mm.o o.ka mH.H om.ow VA\SN\~-mk\w\oH km: .oom.o m.ma 4H.H oo.oo “socmsopne mamasa: .omv.a H.wH mm.a oo.ww usogmsopgs cameo ego: omm.o A.me em.a oo.aw ozogmsoune cameo mama oom.~ m.ke mo.H oo.mw psocwsogze cameo arms 34m.o 4.wH om.“ 00.0w usonmsopee :aacu mam: mmm.o 4.8a om.H 00.5w psozmsouge cameo azmz monomwwogm :fiouomwvovapu xwmwmwmmww Houummvxha pom: 63Hm> moumo poem .oxmucfi msuozmmocm pew mxmucfl m2 oumfisoamo ou new: mosaw> manozmmonm paw anoco no: .uoupme xhnuu.~ oHan 23 Energy and phosphorus balance or status and their interaction were used as the independent variables in a multiple regression anal- ysis. The regression equation is described below: -< II a + BX1 + 8X2 + BXIX2 where: Y = dependent variable a = constant X = energy balance X2 = phosphorus balance X X = interaction of X and X l 2 1 2 Each heifer had a calculated X1 and X2 for each two week period of the experiment. Thus, there were six values for each heifer and 144 observations were used in the regression analysis. Blood Serum Hormones Progesterone. Progesterone was measured by radioimmunoassay as described by Louis, Hafs and Seguin (1973). Progesterone was measured twice weekly for all heifers throughout the experiment. Insulin. Serum insulin was measured by radioimmunoassay courtesy of Dr. E. Veenhuizen of Eli Lilly and Co., Indianapolis, IN. Serum insulin was measured in all heifers once every two weeks throughout the experiment. 24 Metabolites and Enzymes in Blood Serum Nonesterified Fatty Acids (NEFA). NEFA were measured using the methods of Ho (1970) as modified by Bieber (1974). NEFA were extracted from 0.2 ml serum with 1.0 ml of Dole extraction mixture (isopropanolzheptane:1NH2S04, (Dole, 1956). Samples were vortexed and placed in ice. After 10 40:10:l) minutes of standing, 0.2 ml of heptane and 0.2 ml of water were added, vortexed, and placed on ice to allow the separation of organic and aqueous phases. An aliquot of the heptane layer (0.2 ml) was removed and placed in a 1.5 ml polypropylene centrifuge tube with an attached cap (Brinkman Instruments Inc., Westbury, N.Y.). Chloroform (0.8 ml) was added to each tube, the tube capped and placed on ice. Nickel reagent was made by dissolving 0.4050 g of NiC12°6H20 in 100 ml of 1 M triethanolamine giving a final Ni concentration of 1 mg/ml of reagent. Radioactive 63 Ni (1.17 mCi/lOO pgNi; Amersham Searle Corp., Arlington Heights, IL.) was added to give approximately 106 cpm per 0.1 m1 of Ni reagent. Nickel reagent (0.1 ml) was added to each tube, vortexed vigorously for 45 seconds and placed in an ice bath. The organic and aqueous phases were separated by centrifugation at 500 x g for 5 minutes. The aqueous phase was removed by aspiration and a 0.5 ml aliquot of the organic phase transferred to a scintillation vial and evaporated to dryness. Ten m1 of scintillation fluid (Appendix Table l) were added and 63Ni counted in a Nuclear Chicago liquid scintillation counter. 25 Palmitic acid was used as standard for calculation of unknown NEFA concentration. The concentration of NEFA in serum was obtained from an equation derived from regression of palmitic acid (Y) in con- centrations from 0 to 100 nmoles on counts for blank activity (X). Standards were carried through the same manipulations as serum. Serum NEFA's were measured once weekly in all heifers throughout the experi- ment. Serum Parameters Measured by Autoanalytical Techniques Serum glucose, calcium, phosphorus, urea nitrogen, creatinine, aspartate aminotransferase, alanine aminotransferase, creatine phos— phokinase and alkaline phosphatase were measured on a Hycel Mark X clinical autoanalyzer (Hycel Inc., Houston, TX.). These parameters were measured once a week in all heifers throughout the experiment. The methodology is described on the following pages (Table 3). 26 .xofimsoo xcfim Show on ovfinumowEomofizu can men: spa: momcovcoo maw\mev v.0Hum.o ocoflpuocmusnum.m can: cowonuwz Anomfiv aoxoouu A.Hd AHHmofinuoEMnoHoo wonsmmoz .po>fifi ea cmufimocucxm «my: .pfiom owcoanSmocoE mm xHHmowuuoEfluoHoo ousmmmz .ocofivmumofio o mHu\mEv o.nmmno.moH -m.m Show wow I .mnoumo fichoumofioco can Honeymoflonu AmmmHv yommx o.cm~ + Hououmofiogo vmuomnuxm Honeymofiono mm asuom :H Hmuoe .oflmusm xofimeoo manna omen neumm xHHmofluuoefiHoHoo ouswmoz ”momfiv .ocoonmEoo :fiofimnumfiomono .mCOa oficmmpo spa: :mEHouHu so woxoflmeoo mu Esaom pmonmEoo vcm wagon Avomfiv seawaoz m.HH-m.w Hmpoe .eaom an emmaoHou :aououm .emaae0a Aae\wev new Hoflmmox m.m mu canon :fiasnflm esnom .wooHn :H .mocoo :H Esfiofimu .a aficmwpocfi .xHHmofinuoEfipoHoo mm can muflmflfiocmmocm enammmz .oumuanoEocm mm 559mm :H .mHHoo mmmmfiv ngocm m.n-fi.m umocm Esflcossm msuom non :fi sacwme .wooHn man\msv can xxmmsmh v.0 oumwxnfioe 53H:OEEm + a“ :H .mocon a“ wow monogamocm mfippmu pom oocoammom osam> Hmenoz ofimwocfiua ummb :ofiumooq oflnmfinm> paw omcmm .nouxfimemousm Hmuwcflfio w xumz Hoe»: ecu xn pom: mxmmmm mo smeaovonuoz-n.m oanmh 27 .opmzmmocm mzoeomoweo new vmpoouuou .xHHmoflnuoEfiaoHoo venommos an .em.m an on ma + ao< fla\=v 8mm Afiomav o.vnuo.fim + oumnmmogm ocfiummno db< .unmo: paw :fimun ucfixocmmozm amm.mm mxmcflxo o.mv + ocfipmono memxfimumo xmu .oHomse vopmfinum ocfiumopu .xHHmofinuoEfluoHoo nonfiEuouoo .cooamm u mcsfi .mw.a :8 pm oumnamoem u Ao>aa A anon ma\sc o.mm-o.o noses cfiofimcuzmaoexzu u socvflx A wucoomfim ommumcmmonm moomav cmEoHou o.mH enamocmme mo mfimxfionmx: u mmooze HmcflpmoucH ocwamxfl< .oxp ocoon spa: mafiamsoo Amomfiv mmfififlfizm noumm xHHmofluumeflaoHoo mesa .coofimm PAH\:V can mamfiafifiz ousmaoz .opmuoomonxo .mmouocwm A socwwx ommuom .oafimmcm o.om~-o.~n + oumumusfimouox .oHomse Hmpofimxm A -mcmuuocfie< .comnmm o.mHH .d + oumuhwmmmuq nm>fl~ .upmo: oumunmmm< Anomfiv panes .onmEoo pop o>flw ow .:0wuomuucoo -xumz use Homaon m.H-w.o “meam ca :ofl eumuofim oHomse mcflpsw booze me\mEV -Houm .covcoq H.H spa: mnemou mafiawumonu -oaa ocficflumopo some ocflcflumouu .coopm-o:Hn nonvopm cam .ommn wwwcom can hmomfig «aaxxaz madamasmousam Show ou whom one cocfipm>xm o.nnuo.nm uwuoom no: cw omoous mfin\mev Amomav axmzopso o.mo gun: muuaou ocaeasaou-o .eoon omoosau ofiupwu new monogamom 03Hm> Hmeuoz mamfiocflhm amok :ofipmoog ofinmfinm> cam omcmm .eoseapeou--.m «Hash .uuauzzw .HCOUII .Mw Fv~a~tnfi 28 .vmmfi ..o:H Hoax: .mwocpo: xuumfisonu Hook: a“ venom on emHm :wo xmoHovocuoE so Hflmuoc Hexagon .muwv wocmfifinsmcs Aenmflv cozoufix .: "sopm mosam> HmEHoz .op::HE Hem museumnsm mo mace: H mo cowuomon ecu ommeumo Ha“: scan: oexuco mo unseen mg» ma :H opens .uflcz Hmcofiumcpoucfi mm m“ :1 .xHHmofiuuoEfiuoHoo penummoz .oconwanonuwcfiw -v.m + meanwnwx: -chonmoguficfiv -v.m + oum>zhxm .opmemuzfimIA + AH\:V ommuom Ammmav Hoxcmum o.nw-o.om mum>spxm + opmumusfiw .oHomse u puma: -mcmhuocflEm cam swapfiom o.wm -ouox-a + ocwcmamuq A socvflx A 90>“; ocficma< oHuumu pom oucmuommm 03Hw> Hmauoz ofimfiocfiym amok :ofiumooa ofinmfinm> paw owcmm .emseapcoo--.m magma Feed Intake RESULTS The treatment means for intake of corn silage, hay and grain are shown in Table 4. The high energy groups consumed 8.2 kg more grain but 7 kg less corn silage daily than the low energy groups. Hay intake was approximately constant, the cows were all offered 2.3 kg/day. Table 4.——Mean intakes (kg/day) of corn silage, hay and grain for the first 12 weeks of lactation. High Energy High P Low P Low Energy HighTP Low P Corn silage (with NPN) (kg/day) Hay (kg/day) Grain* (kg/day) 18.64 17.83 1.90 1.89 10.25 10.70 25.57 24.25 2.05 2.03 2.27 2.27 *Dicalcium phosphate replaced by CaCO 3 (Intakes expressed on an as fed basis). Energy and Phosphorus Status and corn. The high energy groups were in positive energy status and thus, were receiving more than NRC (1971) requirements, while the low 29 30 energy cows were receiving less than requirements. Mean energy and phosphorus status expressed as a function of time are shown for each group (Figures 1 and 2). It can be seen from Figure 1 that the low energy groups were not as negative as the high energy groups were positive. The low energy high phosphorus group became marginally positive at eight weeks of lactation. Phosphorus status (Figure 2) was more difficult to maintain at the desired level. The high energy, high phosphorus group was in positive status but on the average was over twice as high as the other high phosphorus group. Negative phos- phorus status was achieved in the low energy, low phosphorus group but the other low phosphorus group was on the average marginally positive after about five weeks of lactation. It was difficult to restrict phosphorus intake by this group when coupled with high energy. There was considerable variation in energy and phosphorus status among cows within each group, mainly caused by the difficulty of adjusting intake along with rapidly changing milk production. When energy and phosphorus status were analyzed by a two way analysis of variance it was found that differences did exist among the four groups. The mean energy statuses of the high and low energy groups were 5.4 and -3.4 meal/day respectively (P < .0005, Table 5). Phos- phorus did not significantly reduce energy status and there was no interaction of dietary energy and phosphorus on energy status. A different phosphorus status was achieved between the high and low phosphorus groups. The mean phosphorus status for the high phosphorus group was 16.9 g/day as opposed to a mean of -S.0 g/day 31 .coflpmpomfi mo mxooz NH umnflm may now flxmp\amuev maumum xwuoco :moz--.H .mfim coflpwpowq mo mxmoz 3 w m s m Jr‘i‘. I- n m 38 m :8 U 1 m A swam m 38 I 1 : hww\amoE m :3 m swam O .. mzpmpm m swam m swam o .. m .. hwpmofl I AH/l’fi\r I j w 1‘ r S n ma 33 .cofiumpomfi mo mxooz NH umufiw may now Axmc\wv magnum mzaocmmocm :mozuu.m .mflm 34 a 30A . m sou m swam . m sea A 305 .m swam m swam .m swam copruowq mo mxmmB NH OH m m \1 IIJkIII Io 'l-"-'|"' man OHI OH ma Azme\mv mopmpm mSLOSCmocm om mm om .mm 35 Table 5.—-Mean energy status (mcal/day) for the first 12 weeks of lactation. Ener Phosphorus 8" High Low High 6.54 4.18 Low -2.67 —4.19 Energy P < .0005 for the low phosphorus groups (P < .0005, Table 6). The high energy, high phosphorus group and the low energy, low phosphorus group accounted for the most positive and the most negative phosphorus statuses. In addition, the mean phosphorus status for the high energy group was 10.5 g/day and for the low energy group 1.4 g/day (P = .03). This difference was not desired but does reflect the difficulty of maintaining or restricting intakes at the level needed. Hewever, means of the balanced and imbalanced groups were not different. Body Weight Body weight change generally followed the energy status, however, there was variation among cows in the same group. The high energy cows tended to gain or maintain body weight, whereas the low energy groups tended to either lose or maintain weight. The differ- ence in weight loss between the high and low energy groups approached significance (p = .12). The high energy groups gained an average of 1.75 kg/week against the low energy groups who lost 0.14 kg/week. The high energy, high phosphorus group gained more weight/week than 36 Table 6.-—Mean phosphorus status (g/day) for the first 12 weeks of lactation. Phosphorus Energy High Low High 22.34 —l.32 Low 11.53 -8.73 Energy P = .03 Phosphorus P < .0005 any other group (2.14 kg/week), the low energy, high phosphorus group lost the most/week (-0.32 kg/week). Neither phosphorus nor the inter- action term had any significant effect on weight change. Milk Yield The high energy, high phosphorus group had the lowest average milk yield. When considered together, the high energy groups started at 17.4 kg/day at 2 weeks of lactation as opposed to a mean of 20.1 kg/day for the low energy groups. By the sixth week of lactation the high energy group had reached an average of 24.2 kg/day, surpassing the low energy group who were giving an average of 23.6 kg/day (Figure 3). From this point the high energy group continued to produce more milk than the low energy groups, who were starting to decline in milk production. The high energy group peaked at 24.6 kg/ day about the eighth week of lactation. Milk yields increased with time (r = 0.76, P < .001) in the high energy groups, whereas low energy groups showed a nonsignificant negative correlation with time (R = -.09, NS). 37 .coHumuomH mo mxoo: NH pmHHm ogu How mxmc\mxv wHon xHHE :moZun.m .mHm 38 NH OH Hmzv 00.0: n :oH pm peg .Ho 9703» m :04 no N nmflm (H AHOO. VAH med n hv L.