Made available in DSpace on 2014-06-11T19:32:57Z (GMT). No. of bitstreams: 0 Previous issue date: 2007-06-11Bitstream added on 2014-06-13T19:22:50Z : No. of bitstreams: 1 sirol_mlfg_dr_botfmvz.pdf: 1072564 bytes, checksum: 6ed08e0d4f835b1a9e0ed7d60ba42a2d (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Universidade Estadual Paulista (UNESP) / Foram utilizados dados de 116406 bovinos da raça Nelore, participantes do Programa de Melhoramento Genético da Raça Nelore (PMGRN), nascidos entre 1995 e 2005, com o objetivo de estimar parâmetros genéticos para os pesos padronizados aos 120(P120), 210(P210), 365(P365), 450(P450) e 730(P730) dias de idade e para os ganhos em peso do nascimento aos 120(GP120), dos 120 aos 210 (GP210), dos 210 aos 365(GP365), dos 365 aos 450(GP450) e dos 450 aos 730(G730), além de avaliar a tendência genética das características citadas, tanto para efeito direto como materno. Os componentes de variância foram estimados pelo método de máxima verossimilhança restrita utilizando o programa AIREMLF90, sob modelo animal, o qual incluiu como efeitos fixos, os grupos de contemporâneos e idade da vaca ao parto e como aleatórios, efeito genético aditivo direto para todas as características estudadas e efeito genético materno para P120, P210, P365, GP120, GP210 e GP365. As estimativas de herdabilidade direta foram 0,21, 0,22, 0,22, 0,34, 0,28, para cada peso, respectivamente, e 0,19, 0,20, 0,18, 0,18 e 0,20, para os respectivos ganhos em peso. As herdabilidades maternas foram 0,26, 0,25 e 0,12, para P120, P210 e P365 e 0,23, 0,17, 0,12 para GP120, GP210 e GP365. As correlações direto-maternas foram todas negativas, exceto para P365 (0,06). As tendências genéticas diretas foram todas positivas. As tendências maternas foram quase nulas para todas as características. As estimativas de herdabilidade para os pesos padronizados e para os ganhos em peso indicam que a seleção pode promover mudanças genéticas. As herdabilidades maternas para P120, P210, P365 e GP365 indicam que a seleção nestas características pode contribuir para melhorar a habilidade materna do rebanho. Os ganhos genéticos diretos observados, para todas as características estão aquém dos ganhos potenciais da raça Nelore... / Data of 116406 bovines of the Nellore beef cattle, participants of the Programa de Melhoramento Genético da Raça Nelore (PMGRN), been born between 1995 and 2005 were used with the objective to estimate genetic parameters for the 120-days weight (P120), 210-days weight (P210), 365-days weight (P365), 450-days weight (P450) and 730-days weight (P730) and for the weight gain from birth to 120(GP120), from 120 to 210(GP210), from 210 to 365(GP365), from 365 to 450(GP450) and from 450 to 730 days weight(GP730), besides evaluating the genetic trends of the traits, so much for the direct effect as for the maternal. The variance components were estimated by the restricted maximum likelihood method using the program AIREMLF90. The animal model included fixed effects for contemporary groups and age of the dam at calving, and also included random effects for genetic direct effects for all the studied traits and genetic maternal effect for P120, P210, P365, GP120, GP210 and GP365. The estimative of direct heritability were 0,21, 0,22, 0,22, 0,34, 0,28, for each weight, respectively and 0,19, 0,20, 0,18, 0,18 and 0,20, for the respective weight gains. The maternal heritability were 0,26, 0,25 and 0,12, for P120, P210 and P365 and 0,23, 0,17, 0,12 for GP120, GP210 and GP365. The direct-maternal correlations were all negatives except for P365 (0,06). The direct genetic trends were all positive ones. The maternal trends were almost null for all the traits. The estimative of heritability for the adjusted weights and for the weight gains indicated that the selection could promote genetic changes. The maternal heritability for P120, P210, P365 and GP365 indicated that the selection in these traits could contribute to improve the maternal ability of the herd. The direct genetic gain observed, for all the traits were on this side of the potential gain of the Nellore beef cattle... (Complete abstract click electronic access below)

Estudos foram conduzidos com o objetivo de avaliar os efeitos da relação proteína metabolizável: energia metabolizável (PM:EM) sobre o crescimento acelerado de novilhas pré-púberes, assim como, estimar o peso corporal por meio de equações de regressão que utilizam medidas indiretas de desenvolvimento corporal. Inicialmente doze novilhas Holandês-Jersey (idade média de 90 dias) foram alocadas em baias individuais, com livre acesso a água e sombra, até o início da puberdade (aproximadamente 300 kg). Os animais foram distribuídos em delineamento inteiramente casualizado em dois tratamentos, de acordo com a relação proteína metabolizável:energia metabolizável: controle (PM:EM = 39 g/Mcal); e alta relação (PM:EM = 44,5 g/Mcal). Quinzenalmente, os animais foram pesados e as medidas de desenvolvimento corporal avaliadas. Mensalmente, foram medidos o tamanho e comprimento dos tetos, como método indireto de aferição do desenvolvimento mamário, e colhidas amostras de sangue para avaliação das concentrações plasmáticas de glicose, nitrogênio uréico e progesterona. As relações de PM:EM observadas foram superiores as inicialmente preditas pelo NRC (2001) 44,39 vs 39 g/Mcal e 52,98 vs 44,5 g/Mcal, respectivamente para os tratamentos controle e alta relação. Não foram observadas diferenças estatísticas entre os tratamentos para consumo de matéria seca, peso corporal (PC), perímetro torácico (PT), altura na cernelha (AC), largura da garupa (LG), largura e comprimento dos tetos (P< 0,05). Posteriormente, os dados de PC, AC, LG e PT monitorados até a obtenção de 300 kg de PC, foram avaliados em análises de regressão, sendo o peso a variável dependente e as medidas de desenvolvimento corporal as variáveis independentes. Foram considerados os efeitos lineares, quadráticos e cúbicos destas variáveis. Não foram observados efeitos significativos (P<0,05) do método de estimação do PC sobre esta variável. No entanto efeitos significativos são observados para idade e interação ente idade e método de estimação (P< 0,0001). As equações lineares de AC e LG subestimaram (P<0,05) PC na faixa de idade de 11-13 meses. Entre 6-7,9 meses todas as equações lineares de desenvolvimento corporal superestimam (P< 0,05) os valores de PC. Na faixa etária de 3-5,9 meses apenas as equações lineares de AC superestimaram os valores de peso (P< 0,05). A equação proposta por Heinrich; Rogers; Cooper (1992) subestimou os valores de peso em relação à aferição em balança mecânica na faixa etária de 6-7,9 meses (P<0,05). Em conclusão, medidas de crescimento e desenvolvimento corporal não foram influenciadas pela relação PM:EM da dieta. O uso de medidas indiretas de estimação do peso corporal mostrou-se eficiente na estimação do PC de fêmeas mestiças, assim como a estimação por meio da equação proposta por Heinrich; Rogers; Cooper (1992).; Cooper (1992). / Trials were conducted in order to evaluate the effects of metabolizable protein: metabolizable energy (MP:ME) ratio on accelerated growth of prepubertal dairy heifers, and also the use of indirect methods of corporal growth for body weight estimation. First, twelve Holstein-Jersey heifers (90 days of average age) were housed in individual pens, with free access to water and shadow, until beginning of puberty (around 300 kg). Heifers were allocated on one of the two treatments, control (MP:ME = 39 g/Mcal) and high MP:ME ratio (MP:ME = 44,5 g/Mcal), according to a completely randomized statistical design. Biweekly, heifers were weighed and growth parameters were measured. Monthly, size and length of teats were taken, as indirect method for mammary growth assessment. Blood samples were also collected monthly to evaluate the plasmatic concentration of glucose, and urea nitrogen and progesterone. The MP:ME ratio observed were higher than those firstly predict by NRC (2001): 44.39 vs 39 g/Mcal and 52.98 vs 44.5 g/Mcal, respectively for control and high MP:ME treatment. There were no statistical effects of diet MP:ME ratio on dry matter intake, body weight (BW), hearth girth (HG), withers height (WH), hip width (HW), size and length of teats (P< 0.0001). With the same data a second study was conducted. Data of BW, WH, HW and HG monitored until 300 kg of BW, were used for regression analysis. BW was considered the dependent variable and growth parameters the independent one. Linear, quadratic and cubic effects of the independent variable were considered. There was no statistical effect of method for estimation BW (P<0.05). However, statistical effects were found to age and interaction of age and method for estimation BW (P< 0.0001). Linear regression of WH and HW underestimated the BW on age range from 11-13 months. For 6-7.9 months all linear regression of corporal parameters overestimate the values of BW (P<0.05). On age range from 3-5.9 months only linear regression of WH overestimate the values of weight (P<0,05). Between 6-7.9 months, the regression equation proposed by Heinrich; Rogers; Cooper (1992) underestimated the BW compared with mechanic scale. In conclusion, measures of growth and corporal parameters were not influenced by diet MP:ME ratio. Use of indirect methods of body weight estimation is efficient to estimate the BW of crossbreed heifers, at the same manner as the Heinrich; Rogers; Cooper (1992) proposed equation.

This study, being part of a larger project - "Canadian/New Zealand GxE Interaction Trial" - is comparing Canadian and New Zealand sired heifers for growth traits within Canada, since differences for growth traits were found in the Polish strain comparison (Jasiorowski et al., 1987) and due to selection programs in the two countries. Twenty Canadian Holstein and twenty New Zealand Friesian progeny tested, A.I. bulls were randomly mated to over 1,000 cows in 10 Canadian herds. 3,539 records of weight and wither height from 475 heifers (i.e. 241 Canadian and 234 New Zealand sired) were generated. Subsets of the data for different stages of heifer maturity were analyzed separately. Herd and strain effects least squares means were estimated using analysis of variance. Genetic and phenotypic and correlations and heritability for weight and wither height were estimated by a Derivative-Free Restricted Maximum Likelihood (DFREML) algorithm and an animal model (AM). No differences were found between sire strains for weight except at 15 and 18 months where sib groups of Canadian (CN) sires were heavier than their New Zealand (NZ) contemporaries (393 vs 386 kg and 447 vs 445 kg, respectively). CN sired heifers were taller at all ages except at birth, 3 and 9 months of age. At 24 months CN heifers were 136 cm while NZ heifers were 133 cm. Heritability estimates for weight at birth was 0.62 for the CN strain and 0.59 for the NZ strain. CN estimates (3 to 6 months) and NZ estimates (3 to 9 months) were close to zero. Between 9 to 24 months CN strain estimates ranged from 0.44 to 0.69 while NZ estimates were 0.17 to 0.51. The joint estimates ranged from 0.10 to 0.66. Heritability estimates for wither height for CN strain at birth and between 9 to 21 months were between 0.34 to 0.66 and close to zero between 3 to 6 and at 24 months. The NZ estimates at birth, 18, 21 and 24 months were between 0.36 to 0.93 but close to zero between 3 to 15 months. The joint estimates ranged from 0.32 to 0.75 between 12 to 24 months. Genetic correlations between weight and wither height ranged from 0.62 to 1.0 for CN strain and from -0.04 to 0.91 for NZ strain between 4.5 to 21 months. At six months of age the genetic correlation for CN strain was -0.01 and NZ strain was 0.54. At birth, both sire groups had a genetic correlation of 1.0. At 24 months NZ strain had a genetic correlation of 0.84 while that of the CN strain was 0. Genetic correlations for the joint analysis ranged from 0.61 to 1.0 for all ages except at 6 months (0.18). Phenotypic correlations between weight and wither height were between 0.33 to 0.60 for CN group and 0.33 to 0.62 for NZ group. The joint estimates were 0.36 to 0.61. There were no differences in the phenotypic variances except at 9, 12 and 15 months. Genetic variances were different at all ages except at birth for weight. / Land and Food Systems, Faculty of / Graduate

This study compares the energy utilization of two systems for producing cattle of desirable slaughter weight and grade from weanlings. Both systems produce beef cattle as a primary output; various types of baled hay are produced as a secondary output. One system uses generally accepted, "best management practices" while the other uses experimental, sustainable agriculture techniques. Since the adoption of new practices in agriculture often hinges on economics, an economic comparison is also presented. Beef produced in the sustainable system required 32% less energy per kilogram than that produced in the conventional system. However, baled alfalfa produced in the sustainable system required 8% more energy per kilogram than the alfalfa grown in the conventional system. When all types of hay were considered, the sustainable system used 7% more energy to produce one kilogram of baled hay. To compare the energetics of the two systems on a whole farm basis, the amount of energy required to produce one dollar of return was calculated. The sustainable system required 12.4 megajoules to produce one dollar of return, while the conventional system required 17.1 megajoules to produce the same return. Although economic returns on beef and alfalfa production were comparable in the two systems studied, the conventional system showed greater returns on the whole farm, due to a greater export of baled hay. / Master of Science

Masculinization of the female fetus by administration of androgens has been studied in domestic and many laboratory species. Research concerned with cattle has been based solely on neonatal examination of the genital structures; no studies have reported on postnatal growth and development of androgenized offspring. In the present study, pregnant cows were treated with l 7α-methyl-testosterone (MET, 250 mg/d, sq) from day 40 through 60 of gestation to induce virilization of female fetuses. Control cows received no treatment. At parturition, the phenotypic characteristics of each calf were recorded, including birth weight, ano-genital distance and the appearance of the external genitalia. Calves were weighed every 28 days, and the effects of sex and age on body weight were determined. Blood serum was obtained twice weekly from female calves beginning at ≃230 days of age. Serum was analyzed for progesterone (P₄) concentration by radioimmunoassay and the P₄ profiles were used to estimate the age at puberty and estrous cycle lengths. At 4, 8 and 12 months of age blood samples were collected every 15 min for 12 h from three male, female and androgenized female calves. Serum was analyzed for growth hormone (GH), prolactin (PRL) and thyroxine (T₄) concentrations by radioimmunoassay. Visual examination of female calves born after MET exposure revealed that no external vulval opening was present; a penis, prepuce and scrotum had developed and gonads were not palpable in the scrotum. Mean birth weights were similar among male, untreated female and androgenized female calves, whereas weanling and yearling weights were greater (P < .07) in males and androgenized females than in untreated females. Androgen exposure did not affect the initiation or length of estrous cycles in female calves, however, puberty occurred at an earlier age (P < .05) compared with untreated heifers. Concentrations of GH in serum from androgenized female calves were generally lower than both males and untreated females, whereas concentrations of PRL and T₄ were similar to those of control female calves. / Master of Science

Orientador: Henrique Nunes de Oliveira / Banca: Raysildo Barbosa Lobo / Banca: Maria Eugenia Zerlotti Mercadante / Banca: Marcílio Dias Silveira da Mota / Banca: Luiz Artur Loyola Chardulo / Resumo: Foram utilizados dados de 116406 bovinos da raça Nelore, participantes do Programa de Melhoramento Genético da Raça Nelore (PMGRN), nascidos entre 1995 e 2005, com o objetivo de estimar parâmetros genéticos para os pesos padronizados aos 120(P120), 210(P210), 365(P365), 450(P450) e 730(P730) dias de idade e para os ganhos em peso do nascimento aos 120(GP120), dos 120 aos 210 (GP210), dos 210 aos 365(GP365), dos 365 aos 450(GP450) e dos 450 aos 730(G730), além de avaliar a tendência genética das características citadas, tanto para efeito direto como materno. Os componentes de variância foram estimados pelo método de máxima verossimilhança restrita utilizando o programa AIREMLF90, sob modelo animal, o qual incluiu como efeitos fixos, os grupos de contemporâneos e idade da vaca ao parto e como aleatórios, efeito genético aditivo direto para todas as características estudadas e efeito genético materno para P120, P210, P365, GP120, GP210 e GP365. As estimativas de herdabilidade direta foram 0,21, 0,22, 0,22, 0,34, 0,28, para cada peso, respectivamente, e 0,19, 0,20, 0,18, 0,18 e 0,20, para os respectivos ganhos em peso. As herdabilidades maternas foram 0,26, 0,25 e 0,12, para P120, P210 e P365 e 0,23, 0,17, 0,12 para GP120, GP210 e GP365. As correlações direto-maternas foram todas negativas, exceto para P365 (0,06). As tendências genéticas diretas foram todas positivas. As tendências maternas foram quase nulas para todas as características. As estimativas de herdabilidade para os pesos padronizados e para os ganhos em peso indicam que a seleção pode promover mudanças genéticas. As herdabilidades maternas para P120, P210, P365 e GP365 indicam que a seleção nestas características pode contribuir para melhorar a habilidade materna do rebanho. Os ganhos genéticos diretos observados, para todas as características estão aquém dos ganhos potenciais da raça Nelore... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Data of 116406 bovines of the Nellore beef cattle, participants of the Programa de Melhoramento Genético da Raça Nelore (PMGRN), been born between 1995 and 2005 were used with the objective to estimate genetic parameters for the 120-days weight (P120), 210-days weight (P210), 365-days weight (P365), 450-days weight (P450) and 730-days weight (P730) and for the weight gain from birth to 120(GP120), from 120 to 210(GP210), from 210 to 365(GP365), from 365 to 450(GP450) and from 450 to 730 days weight(GP730), besides evaluating the genetic trends of the traits, so much for the direct effect as for the maternal. The variance components were estimated by the restricted maximum likelihood method using the program AIREMLF90. The animal model included fixed effects for contemporary groups and age of the dam at calving, and also included random effects for genetic direct effects for all the studied traits and genetic maternal effect for P120, P210, P365, GP120, GP210 and GP365. The estimative of direct heritability were 0,21, 0,22, 0,22, 0,34, 0,28, for each weight, respectively and 0,19, 0,20, 0,18, 0,18 and 0,20, for the respective weight gains. The maternal heritability were 0,26, 0,25 and 0,12, for P120, P210 and P365 and 0,23, 0,17, 0,12 for GP120, GP210 and GP365. The direct-maternal correlations were all negatives except for P365 (0,06). The direct genetic trends were all positive ones. The maternal trends were almost null for all the traits. The estimative of heritability for the adjusted weights and for the weight gains indicated that the selection could promote genetic changes. The maternal heritability for P120, P210, P365 and GP365 indicated that the selection in these traits could contribute to improve the maternal ability of the herd. The direct genetic gain observed, for all the traits were on this side of the potential gain of the Nellore beef cattle... (Complete abstract click electronic access below) / Doutor

Kikuyu pasture (Pennisetum clandestinum) is potentially the most important source of roughage used to feed dairy heifers in summer in KwaZulu-Natal. It is commonly believed that on kikuyu pasture beef breed females grow at a faster rate than those from dairy breeds when no supplementation is given. Little conclusive evidence is, however, available to support this. Explanations as to why such differences may exist are also limited. Eight Hereford and eight Holstein heifers of similar age and maturity stage were used in a trial. The trial was run over a twenty week period. For the first ten weeks all the animals in the trial grazed ad libitum kikuyu pasture with no supplementation except for a mineral lick. Over this (grass only) period the two breed groups formed the two treatments. During the second ten week period of the trial all of the Holsteins and four of the Herefords were fed a restricted but equivalent amount (1 .7 kg) of a maize meal based concentrate. The use of a computerized, mobile feeding system allowed concentrate intake of individual animals to be measured. Animal height, weight and condition score readings were taken weekly over the grass only and the concentrate (final seven weeks) periods of the trial. Herbage intake and digestibility were estimated using n-alkanes as indigestible markers in two experiments conducted during the grass only and concentrate periods. The Herefords had a significantly higher ADG than the Holsteins (0.82 vs. 0.04 kg/day; P < 0.01) over the grass only period. During the concentrate period the rate of mass gain of the Holstein treatment did not differ significantly (P >0.05) from the Hereford treatment receiving concentrate. The Herefords receiving concentrate were also not significantly different (P > 0.05) in rate of mass gain from the Herefords not receiving concentrate. Rate of height gain was not significantly different (P> 0.05) between treatments over either the concentrate or the grass only periods. During the grass only period the Holsteins lost condition (0.07 condition score units per week) whilst the Herefords gained condition at an equivalent rate. The voluntary intake of concentrates was not significantly different (P > 0.05) between the Herefords and Holsteins (19.19 vs. 16.40 g/kg/L.W(liveweight) (0.75)). Regression coefficients relating level of concentrate intake to rate of mass gain were also not significant (P > 0.05) for either of the treatments receiving concentrate. The use of n-alkanes as indigestible markers showed the intake of the Holstein treatment to have an intake 55% (P < 0.0 1) higher than the Herefords (185.4 vs. 120.5 g/kg L.W(0.75)) over the first experiment where both treatments were grazing ad lib. kikuyu alone (grass only period). During the concentrate period intake of the Herefords receiving concentrate exceeded that of the Holsteins (P < 0.01) by 23% (139.1 vs. 113.1 g/kg L.W(0.75)). Review of the literature, suggests that the double alkanes technique greatly over-estimated intake. Errors in herbage sampling (accentuated by pasture rotation in the first experiment), a low daily dose of the synthetic alkane (C(32)) and incorrect estimation of the C(32) content in the daily doses are identified as possible causes of the over-estimation of intake. Faecal recoveries of the herbage n-alkanes were demonstrated to increase with increasing chain length and hence C(35) was proposed as the most reliable herbage alkane for dry matter digestibility determination. Digestibility differences between treatments estimated using the C(35) alkane were not significantly different (P > 0.05) in either the first or second experiments. The mean digestibility estimates (using the C(35) alkane) for the first and second experiments were 64.9 and 56.61 %, respectively. In conclusion, higher growth rates of Herefords on kikuyu pasture would seem to be primarily due to differences in the dry matter intake of the grazed herbage. Further work using other breeds of dairy and beef animals is required. The underlying cause of differences in dry matter intake between breeds also requires investigation. / Thesis (M.Sc.Agric.)-University of Natal, Pietermaritzburg, 1995.

The objective of the current study was to compare tick loads, growth, carcass characteristics and meat quality of Nguni, Bonsmara and Angus steers raised on natural pasture. A total of 30, 7-month old steers each of Bonsmara and Angus, and 40 Nguni steers were kept at the University of Fort Hare Farm till slaughter at 18 months. Monthly weights of the steers were recorded. Carcasses were electrically stimulated. The m. longissimus thoracis et lumborum was sampled for the measurement of meat colour, pH, drip loss, sarcomere length (SL), water holding capacity (WHC), cooking losses, myofibrillar fragmentation length (MFL), Warner Bratzler shear force (WBSF), fatty acid profiles and sensory characteristics of the steers. Urine samples were collected at the slaughter line for the determination of stress hormone concentrations. The Nguni had the lowest tick load (P < 0.05) while the Angus had the highest tick load (P < 0.05). Tick load did not affect the growth rate and carcass characteristics of the steers. Bonsmara and Angus steers had higher (P < 0.05) carcass weight and dressing percentage than the Nguni steers. Meat quality characteristics were similar (P > 0.05) among all the breeds, except that Nguni meat was darker (L*) (P < 0.05) than meat from the other two breeds. The Bonsmara had the highest (P > 0.05) concentrations while the Nguni had the lowest (P > 0.05) concentrations of stress hormones. There were significant (P < 0.05) correlations between WB values of meat aged for two and 21 days in Nguni and Bonsmara, but not in Angus. The correlations among stress responsiveness hormones and meat quality were breed-dependent. Except monounsaturated fatty acids (MUFA) and the n-6/n-3 ratio, fatty acid profiles among the breeds were similar (P > 0.05). Cholesterol levels among the breeds were similar (P > 0.05). The Nguni had the best (P < 0.05) sensory characteristics, such as flavour and tenderness. It can be concluded that while the Nguni is a small framed breed, its meat quality is similar to that of Bonsmara and Angus and has the best meat taste when raised on natural pasture.

Growth and body composition of 121 Holstein heifers (4.6 to 18 mo and 129 to 407 kg) reared on pasture, drylot and pasture-drylot systems were evaluated in 6 experiments. Heifers were switched from drylot to pasture or the reverse to study carry-over effects from previous treatments; both systems were satisfactory. Alternate 28-day periods of supplement feeding to grazing heifers did not affect final body weight nor wither height, but ADF and heart girth varied during 2 yr. When switching drylot heifers previously fed low and high TDN, to pasture, gains were best for heifers fed the lower TDN diet; also, gains on pasture were best for light heifers. Gains by all heifers grazing mainly orchardgrass-clover pasture were acceptable, but supplementing with a 19% CP concentrate or lasalocid, usually improved gains. Heifers with lowest BW during grazing made compensatory gains in drylot. Urea space estimation technique showed that compensatory gains were mainly fat. Lasalocid feeding increased daily gain and subcutaneous fat deposition but reduced feed intake and ribeye area. When moving grazing heifers to drylot a total mixed ration with fishmeal or soybean meal as protein sources gave similar responses. DM intake of grazing heifers ranged from 8.1 to 10.1 kg/d, vs 7.5 kg/d for arylot. Supplementing grazing heifers with degradable or undegradable protein gave similar responses in growth and body composition. Based on growth and body composition, seasonal grazing of Holstein heifers may reduce costs for rearing replacement heifers. A corn silage-alfalfa silage-orchardgrass hay mixed ration without concentrates when fed ad libitum to heifers in drylot resulted in gain of 934 g/d. Pasture alone heifers gained from 368 to 755 g/d depending on drought and heifer age. Calculations of costs of rearing Holstein replacement heifers were prepared accounting for observations of response to grazing, supplements to grazing and drylot diets. Well managed grazing reduces costs of rearing. / Ph. D.

Two trials using 32 heifers each evaluated response to undegraded intake protein (UIP) (30 or SO% CP), energy (supporting .S9 or .91 kg ADG), and source of UIP (blood meal or combination protein supplement). Trial one was a 2x2 factorial, with two levels of energy and UIP. High UIP was achieved with blood meal supplementation. From 6-13 mo of age (phase I), high energy increased ADG and DMI, and high UIP decreased DMI. DM efficiencies (kg DMIlkg BW gain) improved with high energy and high UIP, and roN efficiencies (kg IDN/kg BW gain) improved with high UIP. From 13 mo until calving (phase n), heifers were housed together and fed a common diet. Low energy, high UIP treatment had the highest ADG (1.01 kg/day) for phase I, but the lowest for phase n (.33 kg/day), and low energy, low UIP treatment had the lowest ADG (.62 kg/day) for phase I, but the highest for phase n (.S3 kg/day). Overall ADG from 6 mo until calving averaged .S9 kg/day, and was not affected by energy or UIP. In trial 2, two levels of energy and two sources ofUIP were compared, resulting in four treatments: low energy, high UIP with combination protein supplement; low energy, high UIP with blood meal; low energy, low VIP with soybean meal; and high energy, low UIP with soybean meal. Combination protein supplement contained blood meal, com gluten meal, and fish meal. Trial was 300 days long, and began at 6.5 mo. of age. Dry matter intake and ADO were increased with high energy, but not affected by VIP. Overall DM efficiency was not affected by VIP or energy level. Results of both trials indicate VIP may improve feed efficiency of growing Holstein heifers. / Master of Science