Genetic variation in milk urea nitrogen concentration of dairy cattle and its implications for reducing urinary nitrogen excretion.

Nitrogen (N) leached into groundwater from urine patches of cattle grazing in situ is an environmental problem in pasture-based dairy industries. One potential mitigation is to breed cattle for lower urinary nitrogen (UN) excretion. Urinary nitrogen is difficult to measure, while milk urea nitrogen concentration (MUN) is relatively easy to measure. For animals fed diets of differing N content in confinement, MUN is moderately heritable and is positively related to UN. However, there is little information on the heritability of MUN, and its relationship with other traits such as milk yield and composition, for animals grazing fresh pasture. Milk urea nitrogen concentration data together with milk yield, fat, protein and lactose composition and somatic cell count was collected from 133 624 Holstein-Friesian (HF), Jersey (J) and HF×J (XBd) cows fed predominantly pasture over three full lactations and one part lactation. Mean MUN was 14.0; and 14.4, 13.2 and 13.9 mg/dl for HF, J and XBd cows, respectively. Estimates of heritability of MUN were 0.22 using a repeatability model that fitted year-of-lactation by month-of-lactation by cow-age with days-in-milk within month-of-lactation and cow-age, and 0.28 using a test-day model analysis with Gibbs sampling methods. Sire breeding values (BVs) ranged from -2.8 to +3.2 indicating that MUN could be changed by selection. The genetic correlation between MUN and percent true protein in milk was -0.22; -0.29 for J cows and -0.16 for HF cows. Should the relationship between MUN and UN observed in dietary manipulation studies hold similarly when MUN is manipulated by genetic selection, UN excretion could be reduced by 6.6 kg/cow per year in one generation of selection using sires with low MUN BVs. Although J cows had lower MUN than HF, total herd UN excretion may be similar for the same fixed feed supply because more J cows are required to utilise the available feed. The close relationship between blood plasma urea N concentration and MUN may enable early selection of bulls to breed progeny that excrete less UN.

Heritability of ram mating success in multi-sire breeding situations.

Multi-sire mating of a mob of ewes is commonly used in commercial sheep production systems. However, ram mating success (defined as the number of lambs sired by an individual) can vary between rams in the mating group. If this trait was repeatable and heritable, selection of rams capable of siring larger numbers of lambs could reduce the number of rams required for mating and ultimately lead to increased genetic gain. However, genetic correlations with other productive traits, such as growth and female fertility, could influence the potential for ram mating success to be used as a selection trait. In order to investigate this trait, parentage records (including accuracy of sire assignment) from 15 commercial ram breeding flocks of various breeds were utilised to examine the repeatability and heritability of ram mating success in multi-sire mating groups. In addition, genetic and phenotypic correlations with growth and female fertility traits were estimated using ASReml. The final model used for the ram mating success traits included age of the ram and mating group as fixed effects. Older rams (3+years old) had 15% to 20% greater mating success than younger rams (1 or 2 years of age). Increasing the stringency of the criteria for inclusion of both an individual lamb, based on accuracy of sire assignment, or a whole mating group, based on how many lambs had an assigned sire, increased repeatability and heritability estimates of the ram mating success traits examined. With the most stringent criteria employed, where assignment of sire accuracy was >0.95 and the total number of lambs in the progeny group that failed to have a sire assigned was<0.05, repeatability and heritability for loge(number of lambs) was 0.40±0.09 and 0.26±0.12, respectively. For proportion of lambs sired, repeatability and heritability were both 0.30±0.09. The two ram mating traits (loge(nlamb) and proportion) were highly correlated, both phenotypically and genetically (0.88±0.01 and 0.94±0.06, respectively). Both phenotypic and genetic correlations between ram mating success and growth and other female fertility traits were low and non-significant. In conclusion, there is scope to select rams capable of producing high numbers of progeny and thus increase selection pressure on rams to increase genetic gain.

Genetic analysis and genomic selection of stayability and productive life in New Zealand ewes.

Genetic parameters for ewes were estimated for stayability (STAY) and productive life (Prodlife). Records from 658,871 animals from 241 seed stock or ram breeder flocks and one research flock were used to estimate parameters for ram breeder flock STAY and Prodlife. These flocks would have a culling policy based on criteria typical of a commercial enterprise, but including culls based on knowledge such as estimated breeding values. A subset of 35,688 of these animals from 4 ram breeder and one research flock was used to infer the animals that might be culled based on the perspective of a commercial flock as these flocks recorded reasons for culling. These data were used to estimate parameters for commercial flock STAY and Prodlife. Heritability for ram breeder flock STAY until age 3 to 6 yr ranged from 0.07 to 0.09. The heritability of commercial flock STAY was similar in yr 5 and 6, but lower at 0.05 (age 3) and 0.07 (age 4). Heritability of Prodlife for ram breeder flock and commercial flock was respectively 0.10 and 0.13. The genetic correlations between ram breeder flock STAY and commercial flock STAY ranged from 0.20 to 0.99. Both STAY and Prodlife were most correlated genetically with the trait number of lambs born, at about 0.4, and to a lesser degree, live weight up to the age of 12 mo for the traits analyzed with lower genetic correlations estimated for the commercial flocks. Live weight at 18 mo and fleece weight had little or no correlation with STAY and Prodlife. From the data set used to estimate genetic parameters from a ram breeder flock, 4,330 animals were genotyped with the Illumina OvineSNP50 BeadChip and the genotypes used to develop genomic predictions via genomic best linear unbiased prediction. The animals used in the training data set were multibreed, but were heavily Romney breed based. The genomic prediction accuracy for ram breeder flock STAY, according to breed, was dependent on sample size and ranged from 0.4 to 0.44 for Romney and 0.26 to 0.4 for Coopworth. Accuracies for Perendale and Composites, for which sample sizes were less, were more variable. This study has derived genetic parameters for STAY and Prodlife and investigated genomic predictions for these traits. As these traits are of low heritability, sex limited and expressed late in life genomic selection will be useful.

The effects of age, weight, and sire on pregnancy rate in cattle.

The goal was to estimate the heritabilities and genetic variances for pregnancy rate (PR) and calving date (CD) in Angus cattle along with the effect of weight, age, and sire on PR and CD. The data consisted of 4,999 records on PR and CD. Statistical models included year as a fixed effect; premating/postmating weight and age as covariates; and sire of embryo, maternal grandsire (MGS), and permanent maternal environmental effects as random effects. The models also included the interactions between herd and weight (weight change). Direct and maternal effects on PR and CD were estimated using sire MGS and animal models in REML. Pregnancy rate increased from age 2 to 6 and decreased from age 7 to 11 (P < 0.01) and this effect was independent of the culling strategy. There was a quadratic effect of premating cow weight independent of age on PR, with lower PR for low weights (P < 0.01). Overall, cows with a premating weight of 550 kg had the greatest PR. Cows that lost weight during mating had lower PR (P < 0.01). The maternal additive heritability for PR was 0.001 ± 0.012 and the direct additive heritability was 0.024 ± 0.020. The ratio of permanent maternal environmental variance to phenotypic variance was significant (0.048 ± 0.017; P < 0.01). This demonstrates that permanent maternal environmental effects play a major role in the repeatability of PR (0.049 ± 0.015; P < 0.01). The maternal additive heritability for CD was 0.040 ± 0.022, and the direct additive heritability was 0.076 ± 0.045. The ratio of permanent maternal environmental variance to phenotypic variance was low (0.014 ± 0.017) and the repeatability for CD was significant (0.0544 ± 0.0180; P < 0.01). This suggests that maternal genetic effects are as important as direct genetic effects on CD. There was a positive quadratic relationship between premating cow weight and CD with delayed calving for low/high weights (P < 0.01). Cows that lost weight over mating also had a later CD (P < 0.01). Comparisons of a weight-selected herd to the control herd showed differences (P < 0.05) in the optimal premating weight for early calving (control, 480 kg, and weight selected, 615 kg). Calving date was also more sensitive to changes in weight over mating in the weight selection herd (P < 0.05). Therefore, the set point in the weight-fertility axis and the sensitivity of fertility to changes in weight both changed in the weight selection herd.

Effect of age, weight, and sire on embryo and fetal survival in sheep.

The goal was to estimate the heritabilities and genetic variances for embryo and fetal survival (ES) in sheep along with the effect of premating ewe weight, age, and bilateral or unilateral ovulation on ES. The data consisted of 11,369 records on ovulation rate and litter size. Statistical models for ES included year and ovulation rate as fixed effects, premating ewe weight, and age as covariates, and sire of embryo, maternal grandsire (MGS), and permanent maternal environmental effects of the ewe as random effects. The variance components were estimated using REML. In ewes that survived to yr 6, the mean litter size was 1.87, 2.05, 2.01, 2.07, and 1.91 ± 0.04 in ewes of age 2, 3, 4, 5, and 6 yr, respectively. Litter size was less in ewes of age 2 and 6 yr compared to ewes of age 3, 4, and 5 yr (P < 0.01). Ovulation rate was lower at age 2 yr and increased from age 2 to 6 yr (P < 0.05). Two-year-old ewes had lower ES than 3-yr-old ewes (P < 0.01) and the probability of ES decreased after age 3 yr (P < 0.01). Thus, ES contributes significantly to lower fertility in 2-yr-old ewes. In ewes with high ovulation rates (i.e., 5 corpora lutea, CL), more balanced ovulations (i.e., 2 or 3 CL on each ovary) tended (P = 0.06) to be associated with increased ES. A quadratic relationship was observed between ewe weight and litter size (P < 0.01) and a positive linear relationship between premating ewe weight and ovulation rate (P < 0.01). A quadratic effect of ewe weight on ES was observed, with decreased ES for low and high ewe weights (P < 0.01). The optimal ewe weight for ES increased with ovulation rate, which is consistent with the requirement of greater body reserves for maintaining a larger number of fetuses during gestation. A quadratic relationship between ewe weight and the probability that a ewe is able to maintain a pregnancy was also observed (P < 0.05). Pregnancy loss is due to failure of the embryo or fetus or failure of the dam to maintain the pregnancy. The sire of the embryo only influences the embryo, whereas the MGS influences both the ewe and the embryo. The heritability for the direct additive effect on ES in ewes that lambed was 0.0081 ± 0.0139, and the heritability for the maternal additive effect was 0.0447 ± 0.0242. The permanent maternal environmental variance component was significant and explained 8.5% of the phenotypic variance. Thus, genetically, the dam's ability to maintain a pregnancy has 5.5 times the effect on pregnancy loss than the embryo's ability to survive, and this, in turn, was only half the size of the permanent environmental effect. Therefore, selection among dams based on the mean embryonic survival of their embryos will provide an effective way to improve embryonic survival.

Effects of quantitative trait loci and the myostatin locus on trace and macro elements (minerals) in bovine liver, muscle and kidney.

A quantitative trait locus (QTL) study of the concentrations of 14 trace and macro elements (minerals) in tissues of beef cattle was conducted in New Zealand. Back-cross calves with Jersey and Limousin ancestry (202 heifers and 211 steers) were generated using first-cross sires. This paper reports on testing for effects of QTL on the concentrations of minerals in liver, kidney and muscle in cattle at slaughter, following a growth phase during which rearing and finishing stages were on pasture. Fifteen QTL were identified (P < 0.05) on a genome-wide basis in combined-sire and within-sire analyses. In addition, the possible effect of the Limousin myostatin F94L allele was tested by fitting each calf's myostatin genotype, and 16 QTL were identified. Twelve were in common with those QTL identified previously, comprising six affecting the liver (copper and zinc, on two chromosomes each; plus iron and molybdenum), three affecting the kidney (calcium, copper and iron), and three affecting muscle (iron, strontium and zinc).

Genetic mapping of quantitative trait loci for meat quality and muscle metabolic traits in cattle.

A whole-genome scan was carried out in New Zealand and Australia to detect quantitative trait loci (QTL) for live animal and carcass composition traits and meat quality attributes in cattle. Backcross calves (385 heifers and 398 steers) were generated, with Jersey and Limousin backgrounds. The New Zealand cattle were reared and finished on pasture, whilst Australian cattle were reared on grass and finished on grain for at least 180 days. This paper reports on meat quality traits (tenderness measured as shear force at 4-5 ages on two muscles as well as associated traits of meat colour, pH and cooking loss) and a number of metabolic traits. For meat quality traits, 18 significant QTL (P < 0.05), located in nine linkage groups, were detected on a genome-wise basis, in combined-sire (seven QTL) or within-sire analyses (11 QTL). For metabolic traits, 11 significant QTL (P < 0.05), located in eight linkage groups, were detected on a genome-wise basis, in combined-sire (five QTL) or within-sire analyses (six QTL). BTA2 and BTA3 had QTL for both metabolic traits and meat quality traits. Six significant QTL for meat quality and metabolic traits were found at the proximal end of chromosome 2. BTA2 and BTA29 were the most common chromosomes harbouring QTL for meat quality traits; QTL for improved tenderness were associated with Limousin-derived and Jersey-derived alleles on these two chromosomes, respectively.

Quantitative trait loci for organ weights and adipose fat composition in Jersey and Limousin back-cross cattle finished on pasture or feedlot.

A QTL study of live animal and carcass traits in beef cattle was carried out in New Zealand and Australia. Back-cross calves (385 heifers and 398 steers) were generated, with Jersey and Limousin backgrounds. This paper reports on weights of eight organs (heart, liver, lungs, kidneys, spleen, gastro-intestinal tract, fat, and rumen contents) and 12 fat composition traits (fatty acid (FA) percentages, saturated and monounsaturated FA subtotals, and fat melting point). The New Zealand cattle were reared and finished on pasture, whilst Australian cattle were reared on grass and finished on grain for at least 180 days. For organ weights and fat composition traits, 10 and 12 significant QTL locations (P<0.05), respectively, were detected on a genome-wide basis, in combined-sire or within-sire analyses. Seven QTL significant for organ weights were found at the proximal end of chromosome 2. This chromosome carries a variant myostatin allele (F94L), segregating from the Limousin ancestry, and this is a positional candidate for the QTL. Ten significant QTL for fat composition were found on chromosomes 19 and 26. Fatty acid synthase and stearoyl-CoA desaturase (SCD1), respectively, are positional candidate genes for these QTL. Two FA QTL found to be common to sire groups in both populations were for percentages of C14:0 and C14:1 (relative to all FAs) on chromosome 26, near the SCD1 candidate gene.

Genetic variation in the beta, beta-carotene-9', 10'-dioxygenase gene and association with fat colour in bovine adipose tissue and milk.

beta, beta-carotene-9', 10'-dioxygenase (BCO2) plays a role in cleaving beta-carotene eccentrically, and may be involved in the control of adipose and milk colour in cattle. The bovine BCO2 gene was sequenced as a potential candidate gene for a beef fat colour QTL on chromosome (BTA) 15. A single nucleotide base change located in exon 3 causes the substitution of a stop codon (encoded by the A allele) for tryptophan(80) (encoded by the G allele) (c. 240G>A, p.Trp80stop, referred to herein as SNP W80X). Association analysis showed significant differences in subcutaneous fat colour and beta-carotene concentration amongst cattle with different BCO2 genotypes. Animals with the BCO2 AA genotype had more yellow beef fat and a higher beta-carotene concentration in adipose tissues than those with the GA or GG genotype. QTL mapping analysis with the BCO2 SNP W80X fitted as a fixed effect confirmed that this SNP is likely to represent the quantitative trait nucleotide (QTN) for the fat colour-related traits on BTA 15. Moreover, animals with the AA genotype had yellower milk colour and a higher concentration of beta-carotene in the milk.

Genetic parameters for ewe rearing performance.

This paper reports genetic parameters for ewe performance traits in sheep breeders' flocks in New Zealand. Animal performance records from the AgResearch Lamb Survival Database and from Sheep Improvement Limited were used to generate data sets from 3 lambing years (2003 to 2005) in 24 flocks, and involving 31,651 ewes and many breeds and breed compositions (predominantly Romney, Coopworth, and Texel). The heritabilities and repeatabilities for the litter survival traits were very low. Litter weight traits had heritabilities ranging from 0.12 for BW of lamb weaned to 0.28 for total triplet litter weight at birth and repeatabilities ranging from 0.18 to 0.29. The repeatabilities of BCS and maternal behavior score were low to moderate. This study showed that there is little to be gained from including litter survival in sheep selection programs because heritabilities and repeatabilities for the litter survival traits were very low. However, genetic gains in BCS, maternal behavior score, litter weight at birth, and litter weight weaned are possible in this population. Incorporating these traits into sheep selection programs warrants investigation to improve ewe and therefore flock performance.

Quantitative trait loci for live animal and carcass composition traits in Jersey and Limousin back-cross cattle finished on pasture or feedlot.

A quantitative trait locus (QTL) study was carried out in two countries, recording live animal and carcass composition traits. Back-cross calves (385 heifers and 398 steers) were generated, with Jersey and Limousin breed backgrounds. The New Zealand cattle were reared on pasture to carcass weights averaging 229 kg, whilst the Australian cattle were reared on grass and finished on grain (for at least 180 days) to carcass weights averaging 335 kg. From 11 live animal traits and 31 carcass composition traits respectively, 5 and 22 QTL were detected in combined-sire analyses, which were significant (P < 0.05) on a genome-wise basis. Fourteen significant traits for carcass composition QTL were on chromosome 2 and these were traits associated with muscling and fatness. This chromosome carried a variant myostatin allele (F94L), segregating from the Limousin ancestry. Despite very different cattle management systems between the two countries, the two populations had a large number of QTL in common. Of the 18 traits which were common to both countries, and which had significant QTL at the genome-wise level, eight were significant in both countries.

Effects of the myostatin F94L substitution on beef traits.

This study investigated the effects of a SNP in the myostatin gene (MSTN or growth differentiation factor 8, GDF8) on birth, growth, carcass, and beef quality traits in Australia (Aust.) and New Zealand (NZ). The SNP is a cytosine to adenine transversion in exon 1, causing an amino acid substitution of leucine for phenylalanine(94) (F94L). The experiment used crosses between the Jersey and Limousin breeds, with the design being a backcross using first-cross bulls of Jersey x Limousin or Limousin x Jersey breeding, mated to Jersey and Limousin cows. Progeny were genotyped for the myostatin SNP and phenotyped in Aust., with finishing on feedlot (366 calves, over 3 birth years) and in NZ with finishing on pasture (416 calves, over 2 birth years). The effect of the F94L allele (A allele) on birth and growth traits was not significant. The F94L allele in Limousin backcross calves was associated with an increase in meat weight (7.3 and 5.9% of the trait mean in Aust. and NZ, respectively, P < 0.001), and a reduction in fat depth (-13.9 and -18.7% of the trait means on live calves (600 d) and carcasses, respectively, Aust. only, P < 0.001), intramuscular fat content (-8.2% of the trait mean in Aust., P < 0.05; -7.1% in NZ, not significant), total carcass fat weight (-16.5 and -8.1% of the trait mean, Aust. and NZ; P < 0.001 and P < 0.05, respectively). Meat tenderness, pH, and cooking loss of the M. longissimus dorsi were not affected by the F94L variant. In the Jersey backcross calves, additive and dominance effects were confounded because the F94L allele was not segregating in the Jersey dams. The combined effects, however, were significant on LM area (4.4% in both Aust., P < 0.05, and NZ, P < 0.01), channel fat (-11.7%, NZ only, P < 0.01), rib fat depth (-11.2%, NZ only, P < 0.05), and carcass fat weight (-7.1%, NZ only, P < 0.05). The results provide strong evidence that this myostatin F94L variant provides an intermediate and more useful phenotype than the more severe double-muscling phenotype caused by knockout mutations in the myostatin gene.

Effect of myostatin F94L on carcass yield in cattle.

In this study, a highly significant quantitative trait locus (QTL) for meat percentage, eye muscle area (EMA) and silverside percentage was found on cattle chromosome 2 at 0-15 cM, a region containing the positional candidate gene growth differentiation factor 8 (GDF8), which has the common alias myostatin (MSTN). Loss-of-function mutations in the MSTN gene are known to cause an extreme 'double muscling' phenotype in cattle. In this study, highly significant associations of MSTN with cattle carcass traits were found using maternally inherited MSTN haplotypes from outbred Limousin and Jersey cattle in a linkage disequilibrium analysis. A previously reported transversion in MSTN (AF320998.1:g.433C>A), resulting in the amino acid substitution of phenylalanine by leucine at position 94 of the protein sequence (F94L), was the only polymorphism consistently related to increased muscling. Overall, the size of the g.433C>A additive effect on carcass traits was moderately large, with the g.433A allele found to be associated with a 5.5% increase in silverside percentage and EMA and a 2.3% increase in total meat percentage relative to the g.433C allele. The phenotypic effects of the g.433A allele were partially recessive. This study provides strong evidence that a MSTN genotype can produce an intermediate, non-double muscling phenotype, which should be of significant value for beef cattle producers.

Genotypic effects of calpain 1 and calpastatin on the tenderness of cooked M. longissimus dorsi steaks from Jersey x Limousin, Angus and Hereford-cross cattle.

Single nucleotide polymorphisms (SNPs) in the calpain 1 (CAPN1) and calpastatin (CAST) genes were studied to determine their effects on meat tenderness in Bos taurus cattle. Strip loins (M. longissimus dorsi) were removed from cattle in four resource populations after slaughter (n = 1042), aged under controlled conditions until fixed times after rigor mortis, cooked and measured using a tenderometer. Animals were genotyped for the CAPN1 SNP c.947C>G (p.Ala316Gly; AF252504) and for the CAST SNP c.2959A>G (AF159246). Frequencies of CAPN1 C alleles ranged from 23% to 68%, and CAST A alleles from 84% to 99.5%. From all data combined, the CAPN1 CC genotype (compared with the GG genotype) was associated with a 20.1 +/- 1.7% reduced average shear force at intermediate stages of ageing (P < 0.001) and with a 9.5 +/- 1.3% reduction near ultimate tenderness (P < 0.001). The heterozygote was intermediate. For CAST, corresponding values for AA compared with AG genotypes were reductions of 8.6 +/- 2.0% and 5.1 +/- 1.6% respectively (both P < 0.001), but there were too few GG genotypes for comparison. There were small interactions between the CAPN1 and CAST genotypes. For the CAPN1 and CAST genotypes combined, the maximal genotype effect in average shear force was 25.7 +/- 5.5% (P < 0.001) at intermediate stages and 15.2 +/- 4.8% near ultimate tenderness (P < 0.01).

Evaluation of single-nucleotide polymorphisms in CAPN1 for association with meat tenderness in cattle.

Micromolar calcium activated neutral protease (CAPN1) was evaluated as a candidate gene for a quantitative trait locus (QTL) on BTA29 affecting meat tenderness by characterization of nucleotide sequence variation in the gene. Single-nucleotide polymorphisms (SNP) were identified by sequencing all 22 exons and 19 of the 21 introns in two sires (Piedmontese x Angus located at the U.S. Meat Animal Research Center in Clay Center, NE; Jersey x Limousin located at AgResearch in New Zealand) of independent resource populations previously shown to be segregating meat tenderness QTL on BTA29. The majority of the 38 SNP were found in introns or were synonymous substitutions in the coding regions, with two exceptions. Exons 14 and 9 contained SNP that were predicted to alter the protein sequence by the substitution of isoleucine for valine in Domain III of the protein, and alanine for glycine in Domain II of the protein. The resource populations were genotyped for these two SNP in addition to six intronic polymorphisms and two silent substitutions. Analysis of genotypes and shear force values in both populations revealed a difference between paternal CAPN1 alleles in which the allele encoding isoleucine at position 530 and glycine at position 316 associated with decreased meat tenderness (increased shear force values) relative to the allele encoding valine at position 530 and alanine at position 316 (P < 0.05). The association of maternal alleles with meat tenderness phenotypes is consistent with the hypothesis of CAPN1 as the gene underlying the QTL effect in two independent resource populations and presents the possibility of using these markers for selective breeding to reduce the numbers of animals with unfavorable meat tenderness traits.

Evaluation of a liquid formulation of monensin to control bloat in pasture-fed milking cows.

AIM: To evaluate the effectiveness of a liquid formulation of monensin in reducing bloat score in milking cows grazing pasture. METHODS: A Friesian x Jersey crossbred herd which had been genetically selected for high bloat susceptibility since 1973 was used in this study. Two trials were conducted, each involving two groups of 15 cows, randomly allocated to Treated or Control groups. Trial 1 involved twice-daily administration of a novel liquid formulation of monensin given to Treated cows at morning and afternoon milkings. Trial 2 involved once-daily administration of the same product to Treated cows at the morning milking only. Control cows received no preventative treatment. The total daily dose of monensin in each trial was 300 mg per cow, given in a total volume of 100 ml of liquid. Animals were scored for bloat twice-daily whilst grazing on white-clover/ryegrass or red-clover pastures. The scoring system used a scale of 0, 1, 1.5, 2, 2.5, or 3, representing scores of increasing bloat severity as assessed by palpation of abdominal pressure on the left and right sides of the cow. RESULTS: In Trial 1, severe bloat was recorded at 9 out of 23 scoring sessions. Twelve Control cows required therapeutic treatment for bloat on at least one day, compared to 3 cows in the Treated group (p=0.003). In Trial 2, severe bloat in 3 of 9 scoring sessions led to 8/15 Control animals requiring therapeutic treatment compared to 1/14 cows in the Treated group (p=0.02). CONCLUSIONS: Oral drenching with the liquid formulation of monensin tested was effective in reducing bloat score in milking cows grazed on pasture.