Observations of the behaviour of confined sows provide information about how difficult it is for each individual to cope with the conditions in which it lives. Behaviour which has been described as occurring more frequently in confined sows than in sows with space to walk around includes stereotypies and apathetic behaviour. A stereotypy is a relatively invariate sequence of movements occurring so frequently in a particular context that it could not be considered to form part of one of the normal functional systems of the animal, feeding, body maintenance, reproduction, etc. (Broom 1983). The term apathetic behaviour has generally implied inactivity and lack of responsiveness. In this study, carried out at the Animal and Grassland Research Institute, the occurrence of stereotypies and the degree of responsiveness of stall-housed sows have been quantified.

Calculations based on conventional feeding standards suggest that cows grazing well-managed pasture are frequently consuming excessive quantities of protein in relation to their requirements and to the starch equivalent consumed. They also suggest that milk production from potentially high-yielding cows is limited primarily by the supply of dietary energy. This paper describes two experiments on the effects of low-protein supplements on the performance of grazing dairy cows and reports the main results on yield and composition of milk and herbage consumption. Full details and results, and some other aspects of the experiments, are being reported elsewhere.

Seventy-eight lambs, 9 weeks of age, the offspring of sires of the Scottish Blackface, East Friesland, Finnish Landrace, Suffolk and Texel breeds mated to Scottish Blackface females, as a common maternal breed, were given a barley-fish meal diet containing 1-6 g magnesium and 5·9 g phosphorus per kg dry matter for 13 weeks. Urinary excretion of Mg and P was estimated from the ratio of Mg or P to creatinine in spot samples of urine. A relationship between creatinine excretion and live weight was obtained using total urine excretion from a sample of 20 lambs, two of each sex from each sire breed group. Good agreement was found between the actual and predicted urinary excretion of minerals by the 20 lambs on three occasions.

The effect of sire breed on urinary excretion of P was significant (P < 0·05) and the breed types ranked, in descending order of excretion rate, Texel, Blackface, East Friesland, Finnish Landrace and Suffolk, with a four-fold difference in excretion rate between Texel and Suffolk. This difference in urinary excretion of P was taken as evidence of breed differences in efficiency of absorption of dietary P.

Urinary excretion of Mg was correlated with the intake of Mg. The mean estimated efficiency of absorption of dietary Mg varied, over the three occasions, from 0·42 to 0·50 and from 0·39 to 0·45 for total collections and for spot samples of urine respectively. Individual variation in the estimated efficiency of absorption within sires was not significant whereas that between sires was significant (P < 0·05). Heritability was estimated to be 0·62 (s.e. 0·30).

Progesterone, oestradiol-I7β and luleinizing hormone were measured by radioimmunoassay in plasma samples collected daily or on alternate days following insemination in six ‘repeat breeder’ buffaloes. The endocrine profiles in these animals demonstrated that a variety of temporary disturbances in behaviour of these circulating hormones resulted in reproductive failure.

The birth weight and growth rate of sucking calves reared by three groups of cows grazing on veld and receiving different amounts of concentrates during winter and the first 8 weeks of lactation were studied over four successive calving seasons. Calves were weaned at either 150 or 240 days. The efficiency of use of concentrates for weaned calf production was calculated.

Calves from cows given most concentrates were heavier (P<0·01) than othei calves at birth, 150 days and 240 days. Milk consumption during the first 98 days of life accounted for most of the variation in pre-weaning growth. Time of weaning had little effect (P<0·05) on birth weight of the next year's calf and although cows in the early- weaning groups produced calves which were the heaviest at 150 days (P>0·05) these calves weighed less than other calves at 240 days (P<0·001). Simple and multiple regression analyses were used to show that birth weight was affected significantly by cow-weight change during winter and these differences were in part independent of treatment and year. Calf birth weight increased as age of dam increased but this was due to increase in weight of cow with age. Within classes, calf birth weight had a significant and positive effect on weight gain after accounting for milk consumption and this was shown to be an effect of sex. Cows given concentrates produced more weight of 240-days-old weaned calf per year than those given none (P<0·01), and late weaning led to the production of more weaned calf per annum than early weaning (P< 0·001) even when allowance was made for differences in concentrate consumption.

The addition of a categorical variable to the lactation model of Wood (1967) has been shown to be useful in the explanation of seasonality of milk production using data from an experimental herd. The technique is further investigated using monthly milk-yield data from 43 commercial herds in Northern Ireland in the years 1978 to 1982. Coefficients, d, of the categorical variable were estimated for each month of calving. The maximum value of d was 0·256 for the month of October and the minimum value was 0·022 for the month of March.

In ruminants and humans, the majority of whole body glucose utilization is not mediated by insulin. However, while in man most non-insulin-mediated glucose utilization (NIMGU) occurs in the brain, in ruminants the locations of NIMGU remain less well defined. As fasting would be expected to limit NIMGU to what would be regarded as an essential minimum, two studies were performed to establish the contribution of NIMGU to total glucose metabolism in fed and fasted sheep. Each study used four adult castrated male sheep. In study 1, a primed continuous infusion of U- [13C] glucose was begun at time 0 and continued for 7 h. After 3 h of isotope infusion (basal period) somatostatin (0•417 µg/kg per min; SS) was administered for 4 h, either alone (SS-only) or together with insulin (1•0 mU/kg per min; SS + insulin) with normal glucose to maintain euglycaemia for 2 h. Normal glucose was then infused for both the SS-only and SS + insulin treatments to induce and maintain hyperglycaemia over the final 2 h of the experiment. In study 2, fed or 72-h fasted sheep were infused with 6-[3H] glucose from time 0 for 8 h, with SS infusion starting at 3 h and continuing for 5 h. After 3 h of SS infusion, glucose was infused to induce and maintain hyperglycaemia. In both studies SS infusion inhibited insulin secretion, however in study 2, SS in fed sheep caused hyperglycaemia; this effect was not significant in the fasted animals. The rate of glucose utilization was reduced by SS-only as it eliminated insulin mediated glucose uptake (IMGU); under such conditions whole body glucose disposal should be NIMGU. In fed sheep, average NIMGU levels represented between proportionately 0•61 and 0•67 of the basal glucose metabolic clearance rate. During the infusion of SS + insulin in fed sheep, NIMGU fell to 0•34 during euglycaemia and 0•33 during hyperglycaemia, as the infused insulin caused IMGU to predominate. In fasted sheep the absolute rates of both IMGU and NIMGU were reduced, though NIMGU as a proportion of total turn-over (IMGU + NIMGU) increased to 0•88 of glucose metabolic clearance. Calculations suggest that, in contrast to man, only a minor proportion of NIMGU is utilized by the brain and central nervous system in fed or fasted sheep. It is suggested that skeletal muscle and the gastro-intestinal tract may make a major contribution to NIMGU, even in fasted sheep.

Four silages were prepared from a sward of perennial ryegrass (Lolium perenne cv. Melle). Two were ensiled directly either without (UWN) or with formic acid at 31 1/t fresh material (UWA). The remaining herbage was left undisturbed in the swath for 43 h (WL) or was spread and harvested after 19 h (WS). The dry matter (DM) concentrations at harvest for the unwilted herbage were 229 (UWN) and 238 g DM per kg (UWA) and wilting increased DM concentration to a similar extent for both treatments (WS, 339 v. WL, 346 g DM per kg).

Forty British Friesian steers initially 3·5 months of age and 105 kg live weight were used for the trial. Eight steers were allocated to an initial slaughter group and the remainder were offered the four silages at a restricted level to give 940 kJ digestible energy per kg M0·75. These animals were slaughtered after 112 days to estimate the incremental increases in the components of the body.

Loss of DM in the field by respiration increased from 22·2 to 40·6 g/kg DM cut with increased length of wilting. In silo losses were greater with unwilted (UWN, 222, UWA 183 g/kg DM ensiled) than with wilted silage (WS, 133 WL, 122 g/kg DM ensiled). All silages were well preserved. Wilting was more effective than formic acid in reducing total fermentation acids but a lower proportion of these acids was in the form of lactic acid.

Live-weight gain was lower for the steers given the control (UWN) silage (398 g/day) than for those given silages UWA (463 g/day), WS (476 g/day) and WL (470 g/day). However, a lower proportion of the gain by steers given UWN was in the form of gut contents so that gains in fat, protein and energy were not significantly affected by treatment. In conclusion, when steers were given similar amounts of digestible energy and unwilted silage was well preserved, neither the use of formic acid nor wilting influenced gains in protein and energy. Further, the duration of wilting, in the absence of any differences in DM content, had no effect on animal performance.

In a previous experiment the provision of straw to group-housed pigs increased the number of daily feeder visits. This effect of straw is comparable with conditions of reduced social competition for access to the feeder. The aim of this experiment was to investigate if changing straw provision within a pen resulted in similar and reversible effects in this behaviour. One hundred and twenty male pigs (41·1 (s.e. 0·43) kg) were used in three blocks each of four pens of 10 pigs. In each block of 42 days duration one pen had no straw throughout (treatment 1); one pen had no straw for 21 days (period 1) followed by straw for 21 days (period 2) (2); one pen had straw followed by no straw (3) and one pen had straw throughout (4). The pens were situated in a naturally ventilated building with light from 06.00 to 20.00 h. Each pen had a kennel, a lying area and a single-space computerized feeder for recording feeding behaviour. The pigs’ activity was recorded on video in the 2nd and 5th weeks of each block. The number of feeder visits was slightly higher for pigs with straw (1: 9·69 v. 4: 10·46) but this was not significant (P > 0·05). The effect was more marked for treatment 2 where the visits increased (P < 0·01) from period 1 (8·52) to period 2 (10·64) and for treatment 3 where they decreased (11·20 v. 10·10). Pigs without straw (treatment 1) spent more time in the kennel than those with straw (treatment 4) (P < 0·05). For treatment 2 the proportion of time spent in the kennel fell from 0·607 in period 1 to 0·288 in period 2 (P < 0·01). There was a general trend for more activity and more lying outside the kennel when straw was provided. The effect of straw provision was not as marked as that seen previously; nevertheless changing straw provision (treatments 2 and 3) did result in an increase in feeder visits with straw, probably related to the increase in general activity and lying outside in the area adjacent to the feeder.

Two changeover treatments were studied over a 5-week period in eight early-weaned concentrate-fed Holstein calves of initial weight 160 kg. Fresh forage was given ad libitum for one week (A) or two weeks (B); calves on the latter treatment also had 500 g daily of maize meal. All calves received 300 g/100 kg live weight of a protein supplement and had free access to sodium and phosphorus-rich minerals. A rumen inoculation was given after seven days of forage feeding, at which point a liquid molasses/urea mixture was introduced, first in small amounts and then after a further seven days ad libitum. The forage was then restricted to 3 kg/100 kg live weight. Both treatments were effective in raising rumen pH and buffering capacity to the point of allowing the establishment and maintenance of protozoa, chiefly Entodinium and Isotricha spp. Protein N as a percentage of the total N of rumen contents was about 90 with the all-concentrate diet, fell to 35 with the roughage diets and then rose again to 80–90 by the end of the experiment.

Information on the chemical composition of Suffolk × Blackface lambs was collected and used to predict empty body (EB) component weights from empty body weight (EBW) and non-carcass (NC) component weights. EBW accounted proportionately for 0·94, 0·89 and 0·95 of the variation in EB crude protein, fat and energy content. A combination of NC component weights accounted proportionately for 0·94, 0·95 and 0·96 of the variation in EB crude protein, fat and energy content. EBW and a combination of NC component weights together accounted proportionately for 0·97, 0·97 and 0·98 of the variation in EB crude protein, fat and energy content. Chemical analysis of NC components is cheaper and easier than whole body analysis and provides a viable alternative to sample joint or specific gravity analysis.

A high molecular weight protein (HMWP) was isolated and purified from sow milk, and some of its biochemical characteristics and biological functions were identified. The origin of HMWP was also investigated. The molecular weight of HMWP was determined to be about 115 000 and 114 800 by SDS-PAGE and gel filtration, respectively. The sequence of 10 amino acids in N-terminal of HMWP was Ala-Leu-Val-Gln-Ser-Cys-Leu-Asn-Leu-Val. The sequence was blasted against GenBank. No protein showed significant similarity with this sequence suggesting the HMWP may be novel. The result of liquid chromatography mass spectrometry (LC-MS) also proved HMWP could be a novel protein. By amino acid assay, HMWP was rich in glutamate (including glutamine), cysteine, glycine, aspartic acid (including asparagines) and proline. The content of hydrophobic amino acids (Ala, Val, Leu, Ile, Met, Phe and Pro) was lower at 18.59% of the total amino acids suggesting HMWP has high solubility in water. Western blots of lectins were used to identify the kinds of carbohydrate residues attached to HMWP qualitatively. The result showed that HMWP was a kind of glycoprotein containing N-acetylneuraminic acid (NeuNAc), mannose (Man) and/or N-acetylglucosamine (GlcNAc). By isoelectric focusing, HMWP pI was found to be 5.1. Compared with milk fat globule membrane protein (MFGMP) isolated from the sow milk in SDS-PAGE, MFGMP did not contain HMWP. HMWP was assumed to be a secretory milk protein. HMWP was not found in bovine, goat, rabbit or human milk in SDS-PAGE gel suggesting HMWP may be unique to sow milk. By Western blot, HMWP could be detected in sow milk, not in sow serum, which suggests it is synthesized and secreted by the mammary gland. HMWP concentrations in sows milk were the lowest in the first day of lactation, rose significantly during lactation 1 to 7 days. The HMWP content of sows milk remained relatively constant ((1.95±0.13) g/l) during lactation 7 to 20 days. HMWP significantly inhibited Escherichia coli in a dose related manner in vitro. Overall, HMWP could be a novel sow milk protein with implications for the mammary gland and the piglet.

Recent studies highlighted the multiple positive and negative contributions of livestock to society. Livestock production, through its direct and indirect impacts on land use, is an important driver of services provision. Although a few studies provide an account on the multiple services in different livestock systems, there is still an important knowledge gap on the drivers that contribute to the differentiation of services provisioning across areas. We investigated the hypothesis that the current level of services has derived from past intensification trajectories of livestock. The objective of this study was to understand the influences of past changes in livestock, land-use and socio-economic variables on the current provision of social, environmental and cultural services by the livestock sector in France. We combined a long-term country-wide database on livestock intensification between 1938 and 2010 and a database on services provisioning in 2010. We used a set of multivariate methods to simultaneously analyse the changes in livestock intensification from 1938 to 2010 and the current level of services provisioning. Our analysis focused on a set of 60 French departments where livestock play a significant economic role in agricultural production. Our study revealed that the provision of services was spatially structured and based on three groups of departments, characterised by different rates of change in intensification variables. In the first group, ‘Intensive livestock areas’, the high level of employment in the livestock sector was mainly associated with high rates of change in monogastric stocking rates (+1045%) and milk productivity (+451%). In the second group, ‘Extensive livestock areas’, the high levels of environmental and cultural services were mainly associated with moderate rates of change in herbivores stocking rate (+95%) and the stability of grassland area (+13%). In the third group, ‘Transition areas’, the low provision of all services was associated with the decline in livestock due to crop expansion. This study provides knowledge to understand how past changes determined the current contribution of livestock areas in providing differentiated bundles of services, which might help steer the development of the current livestock sector towards more sustainable trajectories.

The accurate estimation of protein requirements for beef cattle is a key factor in increasing livestock profitability and decreasing the environmental impacts of excessive N excretion due to mismatching between assumed requirements and diet formulation. A meta-analysis was conducted to evaluate and validate a new equation to predict the net protein requirements for growth (NPg) of Zebu beef cattle. For the development of the new approach, a database of 552 observations comprised of bulls, steers, and heifers of different genetic groups (Zebu, beef crossbreed, and dairy crossbreed) was assembled. The new approach was evaluated and compared to current models devised by the international nutrient requirements system committees (Agricultural Research Council, 1980; Beef Cattle Nutrient Requirements Model, 2016; BR-CORTE, 2016) to predict NPg. The model evaluation was performed through the model evaluation system (version 3.1.16) using an independent data set (n = 177 observations). An equation was considered the best estimator of NPg if the following conditions were met: (1) the intercept and slope of the regression between ordinary residues and/or predicted NPg values must have been equal to zero and one, respectively; and (2) the greatest concordance correlation coefficient (CCC) and determination coefficient (R), and lowest mean squared error of prediction (MSEP) were attained. Based on the regression models of the observed v. predicted NPg of Zebu beef cattle, both the new approach and that of the ARC (1980) correctly estimated NPg, since the intercept and slope were not different (P > 0.05) from zero and one, respectively. Additionally, the new approach’s determination coefficient was the greatest and the closest to one. The fact that the new model achieved a higher CCC and lower MSEP than the existing models indicated its superior reproducibility and accuracy. The equations proposed by BR-CORTE (2016) and the BCNRM (2016) did not correctly estimate NPg in that the intercept and slope were different (P < 0.01) from zero and one, respectively. Thus, the equations proposed by the new approach and the ARC (1980) accurately and precisely estimated NPg and are recommended for Zebu cattle. Furthermore, the inclusion of equivalent empty BW (EQEBW) in the new approach improves the estimation of NPg. We suggest the use of the following equation to calculate NPg for Zebu beef cattle: NPg = 176.01 × EBG – 0.381 × EQEBW0.75 × EBG1.035 (R = 0.80 and CCC = 0.75); where NPg = net protein requirements for growth, EBG = empty body gain, and EQEBW = equivalent empty BW.

Replacing dairy components from milk replacer (MR) with vegetable products has been previously associated with decreased protein and fat digestibility in milk-fed calves resulting in lower live weight gain. In this experiment, the major carbohydrate source in MR, lactose, was partly replaced with gelatinized corn starch (GCS) to determine the effect on protein and fat digestibility in milk-fed calves. In total, 16 male Holstein-Friesian calves received either MR with lactose as the carbohydrate source (control) or 18% GCS at the expense of lactose. In the adaptation period, calves were exposed to an increasing dose of GCS for 14 weeks. The indigestible marker cobalt ethylenediaminetetraacetic acid was incorporated into the MR for calculating apparent nutrient digestibility, whereas a pulse dose of chromium (Cr) chloride was fed with the last MR meal 4 h before slaughter as an indicator of passage rates. The calves were anesthetized and exsanguinated at 30 weeks of age. The small intestine was divided in three; small intestine 1 and 2 (SI1 and SI2, respectively) and the terminal ileum (last ~100 cm of small intestine) and samples of digesta were collected. Small intestinal digesta was analysed for α-amylase, lipase and trypsin activity. Digestibility of protein was determined for SI1, SI2, ileum and total tract, whereas digestibility of fat was determined for SI1, SI2 and total tract. Apparent protein digestibility in the small intestine did not differ between treatments but was higher in control calves at total tract level. Apparent crude fat digestibility tended to be increased in SI1 and SI2 for GCS calves, but no difference was found at total tract level. Activity of α-amylase in SI2 and lipase in both SI1 and SI2 was higher in GCS calves. Activity of trypsin tended to be higher in control calves and was higher in SI1 compared with SI2. A lower recovery of Cr in SI2 and a higher recovery of Cr in the large intestine suggest an increased rate of passage for GCS calves. Including 18% of GCS in a milk replacer at the expense of lactose increased passage rate and decreased apparent total tract protein digestibility. In the small intestine, protein digestion did not decrease when feeding GCS and fat digestion even tended to increase. Overall, effects on digestion might be levelled when partially replacing lactose with GCS, because starch digestion is lower than that of lactose but fat digestion may be slightly increased when feeding GCS.

Combinations of two high-energy forage finishing diets and two initial body condition scores (BCSs) in a 2 × 2 factorial experimental design were evaluated on cull Holstein–Friesian (HF) cows to improve animal performance and carcass characteristics, aimed at achieving proper fatness and conformation scores (a minimum of ‘4’ and ‘O’, respectively) required for the marketing of high-value loin steaks. The two finishing diets were (i) conventional maize silage complemented with concentrate diet and (ii) wet maize ear silage (pastone) complemented with dry-herbage diet. The two initial body condition levels were (i) low BCS < 5 (LBCS) and (ii) high BCS > 5 (HBCS). The HBCS animals had the greatest potential to respond to the finishing diets. They needed a smaller total feed intake (TFI) and a shorter finishing period (FP) to meet the marketing requirements. The average feed budgets necessary to finish cull dairy cows and to achieve the minimum scores ‘4’ and ‘O’ of carcass classification were 2.31 and 3.61 t of dry matter (DM)/cow for HBCS and LBCS animals, respectively, while the FP lasted an average of 143 and 224 days for HBCS and LBCS animals, respectively. With regard to the two feeding diets, we found no differences for TFI, carcass characteristics and loin muscle features, such as weight, diameter and intramuscular fat.

In a stochastic simulation study the effect of simultaneously changing the model for prediction of breeding values and changing the breeding goal was studied. A population of 100 000 cows with registrations on seven traits was simulated in two steps. In the first step of 15 years the population was selected for production and mastitis occurrence using a univariate model for prediction of breeding values for production and a trivariate model using information on mastitis treatments, udder depth and somatic cell score for prediction of breeding values for mastitis occurrence. In the second step six different scenarios were set up and simulated for 15 years combining two different breeding goals and three different models for prediction of breeding values in 20 replicates. Breeding goal 1 had relative economic value per genetic standard deviation on production (19.4) and mastitis occurrence ( − 50) whereas breeding goal 2 had a economic value on production (19.4), udder depth (4.2), mastitis occurrence ( − 50), non return rate (13.0) and days open ( − 16.75). Model 1 was a model similar to the one used in the first 15 years. Model 2 was an approximate multitrait model where solutions for fixed effects from a model corresponding to model 1 were subtracted from the phenotypes and a multitrait model with an overall mean, a year effect, an additive genetic and a residual effect were applied. Model 3 was a full multitrait model. Average genetic trends for total merit and each individual trait over 20 replicates were compared for each scenario. With the number of replicates the genetic responses using model 2 and 3 were not significant different. With a broad breeding goal using, model 2 or model 3 gave a significantly higher response in total merit than using model 1. Using a narrow breeding goal there was no significant difference between models used for prediction of breeding values. Results showed that with a breeding goal with a lot of emphasis on low heritable traits with a high economic value using a multitrait methodology for prediction of breeding values will redistribute the genetic progress in the total merit index. More gain will come from the low heritable traits in the breeding goal and less from traits with higher heritability. With a broad breeding goal and exploiting the available information in the data the inbreeding coefficient increased though not significantly.

Estimated breeding values (EBVs) and genomic enhanced breeding values (GEBVs) for milk production of young genotyped Holstein bulls were predicted using a conventional BLUP – Animal Model, a method fitting regression coefficients for loci (RRBLUP), a method utilizing the realized genomic relationship matrix (GBLUP), by a single-step procedure (ssGBLUP) and by a one-step blending procedure. Information sources for prediction were the nation-wide database of domestic Czech production records in the first lactation combined with deregressed proofs (DRP) from Interbull files (August 2013) and domestic test-day (TD) records for the first three lactations. Data from 2627 genotyped bulls were used, of which 2189 were already proven under domestic conditions. Analyses were run that used Interbull values for genotyped bulls only or that used Interbull values for all available sires. Resultant predictions were compared with GEBV of 96 young foreign bulls evaluated abroad and whose proofs were from Interbull method GMACE (August 2013) on the Czech scale. Correlations of predictions with GMACE values of foreign bulls ranged from 0.33 to 0.75. Combining domestic data with Interbull EBVs improved prediction of both EBV and GEBV. Predictions by Animal Model (traditional EBV) using only domestic first lactation records and GMACE values were correlated by only 0.33. Combining the nation-wide domestic database with all available DRP for genotyped and un-genotyped sires from Interbull resulted in an EBV correlation of 0.60, compared with 0.47 when only Interbull data were used. In all cases, GEBVs had higher correlations than traditional EBVs, and the highest correlations were for predictions from the ssGBLUP procedure using combined data (0.75), or with all available DRP from Interbull records only (one-step blending approach, 0.69). The ssGBLUP predictions using the first three domestic lactation records in the TD model were correlated with GMACE predictions by 0.69, 0.64 and 0.61 for milk yield, protein yield and fat yield, respectively.

The effect of recombinantly derived bovine somatotropin (bST) administration on wool follicle populations, wool growth and body growth was examined in young lambs. Romney × (Border Leicester × Romney) lambs were treated with bST at doses of 0·1 or 0·3 mg/kg live weight per day, or with excipient, commencing on the day of birth and continuing at weekly intervals until the last injection was given at 11 weeks of age. Concentrations of immunoreactive somatotropin in plasma were 19·2 (s.e. 2·1), 65·2 (s.e. 2·8) and 74·7 (s.e. 4·8) μgll in the control, low dose bST and high dose bST groups, respectively. Somatotropin treatment did not alter the density of primary (P) or secondary (S) follicles in the skin, or the S/P ratio, as measured by histological examination of mid-side skin biopsy samples at 14 weeks of age. Mid-side greasy and clean wool growth during weeks 2 to 14 of the study were increased (P < 0·05) in bST-treated lambs but treatment had no effect on wool growth during the immediate post-treatment period (weeks 14 to 22), or on yearling fleece weight or wool quality characteristics. Body growth was not altered by bST administration. It is concluded that bST treatment stimulates wool growth even in very young lambs but, contrary to the results of a previous study, does not alter the development of wool follicle populations.

The effects of level of supplemental carbohydrate from turnips and level of rumen degradable nitrogen on bacterial cellulolysis, food intake, live-weight gain and diet digestibility were examined with 36 300-kg Friesian steers allocated, in a randomized-block design, to one of six dietary treatments: (i) sodium hydroxide treated straw plus a supplement of urea (30 g urea per kg straw dry matter (DM)) (uO); (ii) uO plus turnips (23·5 g DM per kg M0·75); (Hi) uO plus a high level of turnips (47 g DM per kg M0·75); (iv) (v) and (vi), the same three diets, but with additional urea (60 g urea per kg straw DM). In two separate experiments each lasting for 56 days, the turnip allowance was given in either two feeds, or one feed per day and the straw was given to appetite. Rumen fluid was obtained by stomach tube from all steers before feeding (07·30 h) and after feeding (11.00 h) on days 10 and 24 of each experiment.

Weight gains for all groups given straw plus turnips were high (> 0·90 kg/day). Increasing the quantity of turnips gave significantly increased DM intakes and weight gains but then turnips substituted for straw, and the proportion of straw consumed decreased from 0·8 to 0·6 of the dietary DM (P < 0·01) There were no significant interactions between frequency of feeding of turnips and level of urea on the variables food intake and weight gain. Fibre digestibility was not depressed by the addition of turnips to the diet. Rumen fluid pH decreased after the steers ate turnips but tended to remain above 6·7. There were no large changes in the total number of viable cellulolytic bacteria as a result of either addition of urea or turnips to the diet; counts remained in the order of 107 per ml. Steers consumed large quantities of sodium hydroxide treated straw; turnips were therefore only a small proportion of the overall DM intake with relatively little effect on bacterial cellulolysis.