A multivariate restricted maximum-likelihood procedure was used to estimate variance and covariance components between and within sires. This method, which considered all lactations simultaneously, accounted for the bias in later lactation records due to selection on dairy performance. Analysis was carried out for a mixed model with herd-year-seasons as fixed and sires as random effects, and fitting lactation length, calving age and month within season of calving as covariables. The data included 26 176 first, 19 978 second and 14 868 third lactation records for 679 test sires, and were analysed in 13 subsets. Additional records for proven sires, treated as fixed effects, were included to improve the data structure.

Estimates for all components were higher in later than in earlier data sets, probably, to a large extent, because of a scale effect. Pooled estimates of heritabilities for lactations 1 to 3 were 0·28, 0·19 and 0·24 for milk yield, and 0·27, 0·21 and 0·25 for fat yield. Genetic correlations of 0·91, 0·91 and 0·95 for milk yield, and 0·91, 0·91 and 0·99 for fat yield, were found between lactations 1 and 2, 1 and 3, and 2 and 3, respectively. Results suggest that dairy performance in all lactations is almost identical genetically.

A multibreed cattle experiment involving 25 British breeds was set up to study genetic variation between breeds and genetic inter-breed relationships for a wide spectrum of traits in order to examine the problems of between-breed testing and selection. The experimental design is described and results on between-breed variation are presented for four traits.

All animals were housed indoors and from 12 weeks of age were given a single complete pelleted diet ad libitum through a system of Calan-Broadbent electronic gates. Females were mated to produce one purebred and three crossbred calves, which were reared to slaughter in order to measure the efficiency of the cow-calf unit of production.

Results based on a total of 292 animals, with an average of 12 per breed, are presented for body weight, cumulated voluntary food intake, daily weight gain and daily food intake over the age range from 12 to 72 weeks. The 25 breed-mean curves for body weight and cumulated food intake displayed a remarkably uniform pattern of rankings at all ages and the rankings were very similar for both traits.

The multibreed design used was effective in estimating between-breed variation as a proportion of total variation for the four traits examined. After approximately 1 year of age, the proportion of variation between breeds was approximately 0·70 for body weight and 0·60 for cumulated voluntary food intake. Changes in these traits could therefore be brought about more effectively by selection between breeds rather than within breeds. For average daily weight gain measured over 12-week intervals, between-breed selection was estimated to be most effective in the period of maximum growth rate between 6 and 9 months of age, when between-breed variation was 0·52 of the total. For average daily food intake, measured over 12-week intervals, between-breed selection was likely to be effective beyond 6 months of age, when the proportion of between-breed variation plateaued at 0·48.

At all ages, the coefficient of genetic variation between breeds was approximately 0·14 for body weight and daily gain, and remarkably constant at approximately 0·12 for both daily and cumulated food intake. It is suggested that, for growth and intake traits, the genetic variances within and between breeds remain proportional to each other at all ages.

A mathematical model based on the energy requirements of the dairy cow and the lactation model of Wood was constructed to determine the effect on net returns over concentrates of different calving intervals for different weeks of the year under different feeding management conditions. The model was computerized, and this enabled an investigation of the effect on net returns of varying silage metabolizable energy levels and different nutritional management systems. The importance of the modelling approach in examining a range of interacting factors simultaneously is highlighted. To illustrate this approach, results are presented here with 1981 prices for milk, concentrates and calves kept constant over a period of 4 years and also, to validate this approach, using actual prices paid over the period 1976 to 1981.

1. A deterministic model and computer program of the energy system of lactating and pregnant cows have been developed.

2. Input data for cow, nutritional, management and environmental factors are required.

3. The output of milk production, live weight and heat production is a time series, with a weekly time-step.

4. Suckler and dairy cow examples are given to illustrate the use of the model to evaluate the effects of quality, quantity and pattern of food intake, and shelter.

1. The model of the energy system of lactating, and pregnant, cows proposed by Bruce, Broadbent and Topps (1984) has been evaluated using data relating to 166 lactations for dairy cows and 48 lactations for beef cows. The cows in both sets of data were subjected to various nutritional and managerial regimes.

2. The discrepancies between observed and predicted values of milk (energy) and live weight were small, both as a proportion of the measured values and total energy intake. For Danish Black and White dairy cows the discrepancies for milk were virtually zero on both bases; those for live weight represented 0·020 of the total energy intake, or an over-prediction by 0·054 in late lactation. For Hereford × British Friesian beef cows, suckled by one or two calves, the discrepancies as a proportion of the total energy intake ranged from 0·005 to 0·051 for milk and 0·001 to 0·063 for live-weight change across the nutritional and managerial regimes.

3. It was concluded that the model of Bruce et al. (1984), which considers genetic potential for milk production and growth, nutritional (energy), reproductive and environmental status, may be used to predict milk yield and live-weight change simultaneously from food (energy) input.

Physiological differences between cows in hormone and metabolite concentrations were sought that may account for genetic differences in milk production.

Two groups of New Zealand Friesian cows, differing in genetic merit for fat production (high breeding index (HBI) and low breeding index (LBI)) by approximately 23 breeding index units (20 kg fat on the United Kingdom improved contemporary comparison scale), were each divided into two further groups, one of which was given freshly cut pasture to appetite and in the other intake was restricted to 70% of appetite. Eighteen blood samples were collected over a 24-h period from each of 16 cows fitted with indwelling jugular catheters. Samples were analysed for growth hormone, insulin, free fatty acids and β-hydroxybutyrate. Mean concentration of insulin was significantly higher in the HBI underfed group than in any other group (P < 0·01). Glucose concentration was significantly lower in the LBI group (P < 0·05), whose intake was restricted. The concentration of (3-hydroxybutyrate was significantly higher in the cows whose intake was restricted (P < 0·01) and the concentration fell markedly after feeding. Growth hormone concentrations were higher in the cows on restricted intake (P < 0·05) and in HBI cows but the differences were not statistically significant.

It was concluded that different patterns of blood metabolites and hormones are associated with cows that differ in genetic merit. Aspects of insulin and glucose metabolism, particularly at low planes of feeding, appear to offer most promise as indicators of high milk-solids production.

Yearling steers, 39 Friesians on an 18-month beef production system (group 1) and 38 Hereford × Friesians on a 20-month beef production system (group 2), were weighed, blood sampled and then half of them were implanted subcutaneously with 300 mg trenbolone acetate and 30 mg hexoestrol. All steers were then weighed and blood sampled at 1-month intervals for a further 3 (group 1) or 4 (group 2) months. Serum was assayed for prolactin, insulin, growth hormone, glucose and blood urea nitrogen concentrations. Steroid treatment increased daily live-weight gain by 28 to 37% over the experimental period and the proportional response appeared to be correlated with the plane of nutrition. Neither prolactin nor glucose concentrations were affected by the treatment. However, in both groups 1 and 2, growth hormone concentrations were significantly higher in implanted steers, whilst urea-nitrogen concentrations were significantly lower. Insulin concentrations were consistently, although not significantly, lower in implanted than in control steers.

Data for a total of 3715 cattle from four sources were used to examine the relative importance of live weight at slaughter, and visual assessments of carcass conformation and subcutaneous fat cover, as predictors of killing out. The sources were the Meat and Livestock Commission's beef breed evaluation programme (1689 steers of different breeds and crosses), the beef demonstration unit at the National Agricultural Centre, Stoneleigh (814 commercial dairy-bred cattle), an East Anglian abattoir (405 commercial cattle) and the Milk Marketing Board's Warren Farm (807 dairy-bred steers by different sire breeds). An assessment of live body conformation was also examined as a predictor in the Warren Farm data.

Carcass conformation provided a more precise prediction of killing out than did either live weight or subcutaneous fat score. Regression coefficients in the different groups ranged from 5·4 to 12·0 g/kg for each increase in conformation class (six-point scale). The live conformation assessment was a less precise predictor (within sire breed residual s.d. = 17·7 g/kg) than the carcass assessment (15·6 g/kg).

A comparison was undertaken of 92 Canadian Holstein × British Friesian (HF) and 93 British Friesian (F), autumn born, male calves on three typical United Kingdom production systems. These were: intensive barley beef system (BB); a grass/cereal 18-month system (18m); and 24-month forage system (24m). Calves were brought in from a variety of farms. HF calves were heavier on arrival (43·1 v. 40.7 kg ***). The calves were the progeny of 27 Canadian Holstein and 43 British Friesian sires, and all out of British Friesian dams. They were castrated and reared conventionally to 100 kg before being allocated to treatment. Slaughter was undertaken when subjective handling suggested that the carcass would classify as fat-class 2 on the Meat and Livestock Commission classification scheme (equivalent to European Economic Community fat-class 3), provided that a minimum live weight (BB, 380 kg; 18m/24m, 420 kg) was achieved. There were no differences between breeds in daily live-weight gain on any system. Other results, HF first, follow. At slaughter, HF were both older (BB, 362 v. 342 days **; 18m, 591 v. 558 days ***; 24m, 716 v. 673 days ***) and heavier (BB, 413 v. 395 kg ***; 18m, 486 v. 469 kg NS; 24m, 531 v. 496 kg ***). Efficiency of food conversion, measured on the BB system only, was not significantly different (5·83 v. 5·84).

A comparison was undertaken of 85 Holstein × Friesian (HF) and 84 Friesian (F) steer carcasses that had been produced on three production systems. These were: intensive barley beef (BB); grass/cereal 18 month; and 24 month forage system. On each system, HF steers were heavier at slaughter (over all systems, 476·5 v. 453·3 kg ***) and had a lower killing-out proportion (512 v. 521 g/kg ***), which resulted in higher carcass weights (243·5 v. 235·9 kg **). Steers were slaughtered on reaching a standard degree of finish (equivalent to European Economic Community fat-class 3), provided that they had attained a minimum live weight. There were no significant differences in carcass fat class (3·23 v. 3·35) but HF had poorer carcass conformation classification (2·62 v. 3·21 ***). Of the HF carcasses, 60·4% were rejected on conformation for Beef Premium Scheme payments and only 29·0% of the F carcasses were rejected (P < 0·001): a considerable and unmerited financial penalty. Objective carcass measurements showed HF to have longer sides (1257 v. 1233 mm ***), longer legs (812 v. 790 mm ***) and deeper forequarters (420 v. 402 mm ***). These differences remained when dressed carcass weight and sample-joint fat proportion were taken as covariates. Comparison of the weight-adjusted carcasses showed the HF to have reduced rump (P < 0·001) and loin thickness (P < 0·05). No other consistent effects were seen for carcass measurements, subjective scores or colour attributes. Dissection of the 7th t o 9th rib joint, as a predictor of carcass composition, revealed no significant overall differences in proportions (g/kg) of muscle (571 v. 567), bone (219 v. 219) or fat (206 v. 212). The implications of carcass value of the above findings are discussed.

Rectal temperatures of cattle grazed near the tropic of Capricorn in central Queensland were studied. The cattle were of Bos indicus, B. taurus and crossbred lines, and were represented by over 200 growing heifers in each of 2 years and some supplementary groups.

Rectal temperatures were loge normally distributed when they were expressed as deviations from a basal temperature of 38°C. They were significantly repeatable, but were more highly repeatable when herd mean temperature was above 39·5°C. The heritability estimate was 0·33 (P < 0·01).

The mean phenotypic regression of growth rate on rectal temperature, within breed groups, was 0·04 (s.e. 0·006) kg/day per °C (r = 0·3, P < 0·01) over the entire growth period from birth to 18 months of age but greater during warmer seasons. The estimated genetic correlation was insignificant in one group of heifers but −0·86 (s.e. 0·17) in the other.

The evidence of favourable and possible unfavourable responses to selection of cattle for low rectal temperature in warm environments is discussed.

A study was made of weaning weights in a beef herd, using a rotational crossbreeding system including the Boran breed (zebu) and two temperate beef breeds (Devon and Limousin), and a seasonal calving period spanning the onset of the rains.

Significant sources of variation in calf weaning weights were crossbred type, age at weaning, sex, dam's age, season of birth, year and a year × crossbred type interaction (all P < 0·001).

The adjusted mean weaning weight was 203 kg. Weaners sired by a temperate bull were generally 1·08 times the weight of those sired by a zebu bull. Steers were 1·06 times the weight of heifers. Marked increases in weaning weights occurred as dams increased in age from 3 to 6 years and slight increases continued up to 9 years of age. Weaners born to 9-year-old cows were 1·09 times the weight of those from first calvers.

Alternative methods for investigating the effect of season of birth were examined and the final choice was a covariate for time from birth to the onset of the rains. Calves born in the earliest part of the calving period were approximately 1·14 times the weight of those born just after the onset of the rains.

Livestock food was contaminated with starling excrement by placing caged starlings (Sturnus vulgaris) over livestock food troughs and feeders, and allowing starling excreta to fall on food for periods ranging from 30 to 56 consecutive days. Several levels of contamination were tested by varying the number of caged starlings over troughs, from 1 to 18 birds per m2 of trough area. During the winters of 1979 and 1980, four trials were conducted involving 78 pigs of both sexes, 24 steers and 18 male calves. No significant differences were observed in daily rate of weight gain or efficiency of food conversion (food offered per unit gain), suggesting that neither cattle nor pigs appeared to be adversely affected by exposure to starling excreta in their food. Food rejection observed during these trials appeared to be a negligible problem overall. Although limited disease incidence did occur, it did not appear to be associated with food contamination.

Blood samples were taken from Scottish Blackface ewes of five lines (selected for long and short cannon bones, for hairy and fine fleeces, and an unselected control line) at ages ranging from 9 to 69 months. The samples were analysed for the concentrations of copper, glucose, calcium, magnesium, phosphorus, total protein, albumin and globulin in plasma.

The components of variance and covariance between sires and within sires were estimated and heritabilities were calculated for the eight blood constituents and live weight. The heritabilities ranged from 0·20 (s.e. 0·08) for the concentration of copper to 0·43 (s.e. 0·10) for total protein. Only one of the 34 genetic correlations was significantly different from zero, this being between the concentrations of calcium and phosphorus.

Ewes that had had a single lamb 9 months earlier had lower concentrations of copper, calcium, albumin (P < 0·001) and glucose (P < 0·01) than ewes that had not lambed. Concentrations of glucose (P < 0·05) and total protein (P < 0·001) were also decreased according to the number of lambs born to the ewe 3 months later.

Ewes with low red cell potassium concentrations had higher concentration of glucose in plasma than ewes with high red cell potassium concentrations (P < 0·05). Haemoglobin type affected the concentrations of copper (P < 0·01) and total protein (P < 0·05), with ewes of type B having higher and lower concentrations respectively. The constituents measured were not significantly affected by transferrin type.

Four experiments were carried out, with a total of 98 Suffolk-cross lambs fed at a restricted level, to study the effects of skeletal long photoperiod (7 h light: 10 h dark: 1 h light: 6 h dark) compared with short photoperiod (8 h light: 16 h dark) on growth and blood levels of hormones. The experiments started when the lambs were from 2 to 10 months of age and continued for 9 to 14 weeks.

In the first experiment, the weight gains of both females and castrated males were significantly increased by skeletal long photoperiod. Plasma prolactin was increased and cortisol decreased by skeletal long photoperiod.

Experiment 2 included females and castrated males; prolactin was stimulated by skeletal long photoperiod and there was a significant positive correlation between live-weight gain and prolactin.

The weight gains and carcass characteristics of the female lambs used in the third experiment were not affected by photoperiod. Live-weight gain was negatively related to plasma cortisol.

In experiment 4, there were 12 intact males and 12 castrated males. The intact animals grew significantly faster; skeletal long photoperiod caused a depression in testosterone levels in plasma even though it tended to increase live-weight gain.

When the data from all four experiments were pooled, it was found that skeletal long photoperiod gave a significant increase in live-weight gain in animals whose pineal gland was intact (40 per photoperiod; 138 v. 113 g/day); there was no effect on gut fill at slaughter and the increase in carcass weight (0·5 kg) was not significant.

Age at first parturition in this environment occurs at approximately 18 months. The biggest cause of variation is flock with neither species nor any other source being significant. Parturition intervals for both species are approximatley 10·5 months: flock and season are significant on this variable and there is a decreasing, although non-significant, interval with advancing age. Litter size in goats is 1·25 and in sheep is 1·04, with parity and flock having significant effects for both species, while season also has an effect on goats. Annual reproduction rate is 1·42 kids and 1·19 lambs per breeding female per year.

The effect of age on incidence of susceptibility to halothane anaesthesia in pigs was studied on animals that were repeatedly exposed to the anaesthetic. In three trials, 232 Piétrain/Hampshire and 201 Landrace pigs were given three 3-min halothane tests, at 19, 35 and 54 days of age. In a fourth trial 253 Landrace pigs received four 5-min tests, at 21, 35, 49 and 63 days. All pigs were offspring of positive × positive matings. Maximum-likelihood estimates of the probability of positive reactions were 0·47, 0·73 and 0·88 at the three ages in Pietrain/Hampshire; 0·20, 0·63 and 0·60 in the first Landrance trial; and 0·64, 0·86, 0·84 and 0·90 in the second Landrace trial. Thus, in all trials, the incidence of halothane positive reactions increased with age. In both breeds and at all ages, the incidence was lower in males than in females.

Twelve pregnant sows, 35 to 52 days pregnant initially and 89 to 103 days pregnant at the end of the experiment, were used to study the effect of ambient temperature on metabolic rate. Six sows were group-housed in two groups of three sows per group in one calorimeter and six sows were tethered individually in another calorimeter. Temperatures were changed by 3°C every 3 days in a stepwise fashion from 20 to 8°C and from 8 to 20°C. Sows were given a constant amount of food, which gave 540 kJ metabolic energy per kg live weight0·75 per day and which covered estimated needs that were proportionately 0·20 above maintenance.

Average daily live-weight gain was 531 g for the group-housed sows and 402 g for those housed individually. Heat loss increased below 14°C in group-housed and below 20°C in individually-housed sows. Extra daily thermoregulatory demand was 7·5 kJ/°C in group- and 13·7 kJ/°C in individually-housed sows.

Two experiments, involving 144 and 168 pigs weaned at 22 to 31 days, to investigate the response to inclusion of copper (CuSO4.5H2O), copper plus carbadox/sulphadimidine or copper plus avoparcin supplements in starter and grower diets were made. The carbadox and sulphadimidine levels were 50 and 100 mg/kg diet respectively in both diets. Avoparcin was included at 40 mg/kg in the starter diets and 20 mg/kg in the grower diets. In the first experiment the levels of copper were either 180 or 340 mg/kg in the starter diets and either 90 or 270 mg/kg in the grower diets. In the second experiment the levels were either 20 or 180 mg/kg in the starter diets and either 10 or 180 mg/kg in the grower diets.

In the starter period of the first experiment, supplementation with carbadox/sulphadimidine or avoparcin proportionately increased the rate of growth by 0·41 and 0·07 respectively in relation to supplementation with 340 mg copper per kg diet (P < 0·05). Neither supplement improved growth rate in the grower period. In the second experiment there were no significant differences in growth rate. However the rates of growth in both the starter and grower periods were higher in pigs receiving carbadox/sulphadimidine than in those receiving avoparcin, while these in turn grew slightly faster than those receiving copper only.

Taking into account the costs of the supplements the inclusion of the carbadox/sulphadimidine supplement produced an economic benefit compared with the other treatments.

A total of 142 pigs, born at four different times of the year, were slaughtered between 1 and 9 months of age to study the effects of season of birth and sex on heart weight and body weight in the equatorial climate of southern Nigeria.

Analysis of loge heart or body weight showed that season of birth influenced heart and body growth (P < 0·001). The differences occurred largely during the phase of rapid growth, with smaller differences at weaning and at 9 months. Pigs born during the high-rainfall period (May to September) showed stunted body and heart growth during the subsequent dry and warmer seasons. Those born during the dry seasons (November to March), but which passed their rapid-growth phase through the rainy, cooler seasons, were attaining market weights and adult heart weights earlier in life. However, all pigs with early growth reductions recovered, at least partially, later in life. The males had larger hearts at later ages than the females although the differences were not significant (P > 0·05). Allometric regression analyses for heart and body weight showed no seasonal effects but there was a small but significant sex effect (P < 0·001). The males had a smaller intercept and a larger slope when log, heart weight was fitted to log, body weight.

Pigs reared in this equatorial climate appeared to have higher heart weights relative to body weights than those reared in temperate regions. The highest recorded proportional value of 0·0032 is outside the range of 0·0023 to 00028 recorded in temperate regions. This suggests a probable physiological adaptation to increased work load on the heart under warm tropical conditions.