Understanding the relationships between food intake, milk output and body condition in high-yielding dairy cows is crucial in determining suitable management strategies. During two winter feeding periods 38 and 37 cows were individually fed, to appetite, complete diets which on average contained 11·7 MJ metabolizable energy per kg dry matter and comprised grass silage, concentrate meal and brewers' grains (draff). The groups' mean 305-day yield was 7 240 kg (s.d. 1 281) with 42 g (s.d. 4·3) fat per kg. Regression analysis was carried out to describe dry-matter intake both for 26 weeks post calving and for four successive 6-week periods from calving. The final equations, which had a residual s.d. of 0·07 to 0·10 of the observed intake, included milk yield, cow size and a measure of body-condition change. The cows were divided into three groups (high, medium and low) on two criteria: (1) mean milk yield (MJ/day) during the first 26 weeks of lactation and (2) post-calving backfat index determined ultrasonically. Differences were found between milk-yield groups from gross efficiency (milk yield (MJ)/energy intake (MJ metabolizable energy)) (P < 0·001), mean metabolizable energy intake (MJ/day) (P < 0·01), dry-matter intake as a proportion of live weight (P < 0·05), and post calving live weight (kg) (P < 0·05). Differences were found between backfat-index groups for maximum backfat loss and loss to day 42 (P < 0·001); also for mean live weight during the 26 weeks and post calving live weight (P < 0·001), dry-matter intake as a proportion of live weight (P < 0·05) and lactation number (P < 0·05). Interactions were found between the milk yield groups and backfat groups for milk yield (P < 0·01) and gross efficiency (P < 0·05) with the fattest group containing the highest and lowest yields and efficiencies.

A survey was carried out of 1 640 British Friesian heifers calving predominantly in the autumn of 1979. The monthly samples of 1 055 animals showing no reported evidence of udder infection were used to evaluate the parameters of a lactation curve in milk cell count. The model was

C = 190 n−0.4880exp(0·178n)

where C is the monthly cell count in millions per 1 during the nth month of lactation. The cell count varied from 230 × 106 in week 1 and 190 × 106 in week 11 to 400 × 106 in week 44 of lactation.

On applying the model to the whole sample, milk sampled within a month before or after antibiotic treatment for clinical mastitis contained more than 200 × 106 cells per 1 above the level suggested by the lactation curve. Lactation mean cell counts of treated cows were 400 × 106 cells per 1 higher than those of untreated cows. It was not possible to identify periods in which cows required treatment, or those with high cell counts, by reference to the lactose concentration in the milk samples. Among the untreated cows, the cell count at the third monthly test-day was lower than at any other time, and was more highly correlated with the lactation mean cell count.

Possibilities for increased rates of genetic change in dairy cattle through embryo transfer and embryo splitting are examined, using the multiple ovulation and embryo transfer systems previously proposed. These involve embryo transfer from 1-year-old females (juvenile scheme, generation interval 1·8 years) and from females after 1 lactation (adult scheme, generation interval 3·7 years), with use of males at similar ages. Though selection is less accurate than in conventional progeny testing, the annual rate of genetic improvement can be increased, and even doubled. If the number of transfers is restricted andm the inbreeding rate is limiting, the adult scheme for both sexes is preferred. A scheme with 1 024 transfers per year and 512 females milk-recorded per year will sustain a rate of genetic improvement some 30% above that possible by a conventional national progeny-testing programme. Because of the relatively small number of animals involved, it is argued that greater control over recording, breeding and selection should be possible, leading to a larger proportion of the possible genetic gains being realized in practice. Other advantages, and disadvantages of these systems, and their integration in dairy cattle improvement are discussed.

Correlation between circulating thyroid hormones and live-weight gain during the first year of life were observed in four trials on cattle of the two main Belgian breeds: Belgian Blue and Friesian. Thyroid hormones assayed and considered as potential predictors of growth rate were: thyroxine (T4), triiodothyronine (T3) and free thyroxine (FT4). Triiodothyronine uptake (T,U), thyroid stimulating hormone (TSH) and free thyroxine index (FTI) were also estimated.

No significant correlation was found between live-weight gain and T4 or T3 in 13 Belgian Blue heifers (trial 1). In 49 bulls submitted to a performance test, T4 and FTI increased from 3 to 12 months whilst T3U decreased (trial 2). Significant correlations between live-weight gain and T4 and between live-weight gain and FTI were obtained from samples taken between 66 and 95 days. Live-weight gain was correlated with T3U when animals were between 96 and 125 days, and 126 to 155 days of age. It is possible that the interaction between total T4 and T3U is important in controlling growth. In 13 young calves, beef merit, expressed as live-weight gain in 1 year, was highly correlated with T4 concentration at 8 to 10 days of age (trial 3). T3 and FT4 did not seem to be directly associated with the growth processes. In trial 4, live-weight gains of 42 young animals housed on a commercial farm were negatively correlated with T4 and with FTI. The conflicting results encountered may perhaps be explained by the lack of information on the use of anabolic preparations in field conditions. T4 concentration and FTI may be considered as potential parameters for the estimation of beef merit but the results are insufficient to draw definite conclusions.

One hundred and fourteen castrated male cattle of various breeds were measured ultrasonically immediately prior to slaughter and the data were examined for correlations with the proportion of adipose tissue in the carcasses. Two flaw detectors and a simple scanner, the Scanogram, were used to make ultrasonic pulse-echo measurements of tissue thicknesses and areas. These were used to compute volumes of adipose tissue and muscle in a region of the back of each animal and the ratio of the volumes of the two tissues was examined for correlation with carcass fatness. Three individuals interpreted the Scanogram photographs and two made measurements using flaw detectors.

In general, pulse-echo measurements made by different individuals differed in magnitude and predictive value, even when identical scans were measured. Equally there were substantial differences between measurements when the same animals were measured by the same individual using different ultrasonic instruments.

In a parallel experiment, the speed of ultrasound transmission was measured at various locations on the living animal and the data were examined for correlations with carcass fatness. Unlike the pulse-echo technique, the transmission method was not prone to errors of subjective interpretation and did not require skill to interpret. Predictions based on speed measurements at two sites in the hind limb were as well related to carcass fatness as were those made on the basis of the best estimate of tissue volumes in the back. The use of two speed measurements in a multiple regression with the best Scanogram data significantly improved the accuracy of predictions of side fatness made on the basis of speed or pulseecho measurements alone.

Two types of analysis were used to examine the effectiveness of Ultrasonic selection of lean animals from a group of 50 Hereford castrated males of similar age and weight.

Primary growths of perennial ryegrass and red clover were ensiled with formic acid (2 1/t fresh crop), o an equal quantity of formic acid together with formaldehyde (46 or 49 g/kg crude protein in the ryegrass or red clover respectively). The four silages were offered ad libitum to 12 groups, each of five British Friesian steer calves, either alone or with urea or maize starch supplements at 18·4 or 185·2 g/kg total dry-matter intake respectively.

Digestible organic matter and digestible energy intake, live-weight gain, carcass weight, nitrogen retention and all digestibility measurements were higher on the ryegrass silages than on the red clover silages. However, dry-matter intake was higher on the red clover silages.

Formaldehyde treatment reduced lactic and total acid content, and protein degradation in the silages. It also increased intake, live-weight gain and nitrogen retention on ryegrass but not on red clover, the effect being greater when the urea supplement was given. Digestibility measurements were depressed by formaldehyde treatment, although cellulose digestibility was only depressed in the ryegrass silage.

Supplementation with starch depressed silage intake and nitrogen and cellulose digestibility, but increased dry matter, organic matter and energy digestibilities, digestible organic matter and digestible energy intakes, live-weight gain and carcass weight. The positive intake, live-weight gain and carcass weight responses were greater on the silages treated with formic acid, while the digestibility and the live-weight gain responses were greater on the red clover silages. Starch supplementation did not improve nitrogen retention.

The different animal production responses to formaldehyde treatment on the ryegrass and red clover silages are discussed.

The present study estimates the mean fat and solids-not-fat concentrations, and 305-day yields of fat, solids-not-fat, total solids and milk for three zebu breeds (Tharparkar, Sahiwal and Red Sindhi) and for halfbred F1 and F2 Brown Swiss ♂ × Sahiwal ♀ and ¾ Brown Swiss ♂ × Sahiwal 9 cattle maintained at this Institute during the period 1972–75. Mean fat and solids-not-fat concentrations ranged from 46·4 to 47·8 g/kg and 90·3 to 90·9 g/kg respectively, which were similar for all breeds. The 305-day fat, solids-not-fat, total solids and milk yields differed significantly between different breeds (P < 0–01). These yields were similar for the three zebu breeds (86 to 96 kg fat, 164 to 186 kg solids-not-fat, 250 to 282 kg total solids and 1 814 to 2 054 kg milk), but crossbreds showed significantly higher yields (125 to 171 kg, 242 to 332 kg, 368 to 503 kg and 2 670 to 3 676 kg, respectively), the highest being for F1 halfbreds. Season of calving had no significant effect on mean fat and solids-not-fat concentrations whereas the lactation yield parameters were found to be significantly higher for Tharparkar, Sahiwal and F1 Brown Swiss × Sahiwal cattle calving in the winter, followed in order by those calving in the monsoon and those calving in the summer. Selection of animals based on milk production alone was also found to increase the yield of milk constituents.

The effects of selection for low backfat thickness on tissue deposition in different body sites has been investigated in pigs. Eight castrated male and eight female pigs from each of the selection and control lines maintained at the University of Newcastle upon Tyne were used. One castrated male and one female from each line was killed at 15, 30, 45, 60, 75, 90, 105 and 120 kg live weight following ad libitum food consumption and fully dissected.

There were only small effects of line on carcass composition, selected pigs having 120 g/kg more bone than controls at the mean side weight (P < 0·001) and 40 g/kg more lean (NS). At the same mean weight of subcutaneous fat, selected pigs had thinner backfat than controls (approximately 3 mm) over m. longissimus at the last rib and over the shoulder but not at the mid-rump or mid-back positions. There appears to have been a slight shift in the sites of fat deposition from above m. longissimus in the loin towards the mid-line and rump as a result of selection, even though mid-rump, C and K measurements were all included in the selection index. However, there was no difference between the lines in the weight distribution of subcutaneous fat between six regions and so these shifts must have been very localized. There was no difference between lines in the relationship between P2 fat thickness and proportion of lean in the side.

The effects of selection on the sites of deposition within tissues other than subcutaneous fat were small. In particular there was no evidence that selection has caused relocation of body fat from subcutaneous to the other sites.

Growth performance and body composition differences between Large White control (C) and 10th generation index-selected (S) pigs were evaluated on a fixed food, fixed time trial. This avoided the complicated interpretation associated with interacting variations in food intake, days on test and body composition. Thirty-nine C and 40 S boars were slaughtered, their carcasses dissected and whole empty bodies minced and chemically analysed after an 84-day trial starting at 26·6 ± 0·27 kg live weight. S boars were older at the start of test (95·2 v. 89·0 days, P < 0·001), were heavier at the end of test (96·2 v. 93·8 kg, P < 001) and converted food more efficiently (0·392 v. 0·378 kg gain per kg food, P < 0·01) than C boars. Bodies of S boars contained more water (48·3 v. 45·9 kg, P < 0·001), less lipid (17·4 v. 18·8 kg, P < 0·001) less gross energy (1 004 v. 1 045 MJ, P < 0·05), but similar amounts of protein (13·8 v. 13·4 kg, NS) compared with C boars. S carcasses contained more lean (34·1 v. 31·8 kg, P < 0·001), less subcutaneous fat (12·5 v. 13·7 kg, P < 0·001), and had lower P2 backfat depths (15·8 v. 18·3 mm, P < 0·001) than C carcasses. Although carcass leanness had been increased, 10 generations of index selection had also reduced the overall energetic efficiency of the selection line, apparently by increasing non-productive heat loss.

Food intake was individually measured for Large White control (C) and 10th generation index-selected (S) pigs fed twice daily to appetite on a fixed time trial. Thirty-nine C and 38 S boars were slaughtered, their carcasses dissected and whole empty bodies minced and chemically analysed after an 84-day trial starting at 28·6 ± 0·30 kg live weight. S boars ate proportionately 0·08 less than C boars in the first 42 days of the trial (79 v. 86 kg, P < 0·01), but S and C boars ate similar amounts (114 v. 117 kg, NS) in the last 42 days. S boars grew 0-04 faster than C boars (918 v. 886 g per day, NS), and had 0·09 better food conversion efficiency (0·404 v. 0·369 gain per kg food, P < 0·001). S boars contained more water than C boars (53·5 v. 48·9 kg, P < 0·001), more protein (15-4 v. 14-2 kg, P < 0·01), less lipid (20·8 v. 25·1 kg, P < 0·001), less gross energy (1 199 v. 1 357 MJ, P < 0·01) but a similar amount of ash. S carcasses had 3 kg more lean (36·5 v. 33·5 kg, P < 0·001) and more bone than C carcasses (7·14 v. 6·71 kg, P < 0·05), but contained 3 kg less total fat (19·6 v. 22·6 kg, P < 0·01). The lean tissue growth rate of S boars was 327 g per day which was 31 g per day faster than for C boars (P < 0·001). Appetite reduction in S boars enhanced fatness differences between the lines.

Forty-five piglets were used to study the effects of feeding diets containing 153, 239 or 321 g crude protein per kg dry matter between 1·8 and 6·5 kg live weight, and low and high protein diets subsequent to 6·5 kg live weight on growth performance, body composition and the cellularity of skeletal muscle to 45 kg live weight.

Reducing dietary crude protein between 1·8 and 6·5 kg live weight depressed growth performance and at 6·5 kg live weight increased body fat content but reduced body protein, body water and the weight and DNA content of the adductor muscle. The effects of dietary protein content before 6·5 kg live weight on body composition at the latter weight were still evident in pigs killed at 11·5 kg live weight whilst the differences in muscle DNA persisted to 45 kg live weight.

On the lower protein treatment subsequent to 6·5 kg live weight there was a tendency for pigs given the lowest protein diet before 6-5 kg live weight to exhibit better growth performance and deposit protein at a faster rate than those given the higher protein diets. However, these responses were reversed on the higher protein treatment subsequent to 6·5 kg live weight.

The effects of dietary protein content subsequent to 6·5 kg live weight on growth performance, body composition and the cellularity of muscle tissue were qualitatively the same as those for the live-weight phase 1·8 to 6·5 kg.

1. Fifty-seven piglets were used to study the effects of rearing pigs on the sow or artificially on a low or high level of feeding between 1·8 and 6·5 kg live weight and either a low or high level of feeding between 6·5 and 20 kg live weight on growth and development to 75 kg live weight.

2. The artificially reared pigs fed at the higher level grew more rapidly (P < 0·05) and were fatter (P < 0·05) at 6·5 kg live weight than their more restrictedly fed counterparts. The pigs reared on the sow grew at the same rate as the artificially reared pigs fed at the low level but at 6·5 kg live weight were as fat as the artificially reared pigs fed at the higher level.

3. Between 6·5 and 20 kg live weight the artificially reared pigs previously fed at the lower level grew more rapidly (P < 0·05) than their more liberally fed counterparts which in turn, grew at a faster rate (P < 0·05) than pigs previously reared on the sow. However, nutrition prior to 6·5 kg live weight had no influence on performance or body composition subsequent to 20 kg live weight.

4. Raising the level of feeding between 6·5 and 20 kg live weight increased growth rate (P < 0·05) and body fat content at the latter weight (P < 0·05). However, subsequent to 20 kg live weight, the pigs previously fed at the lower level exhibited more rapid and efficient growth (P < 0·05) and at 75 kg live weight were leaner (P < 0·05) than those fed more generously between 6·5 and 20 kg.

5. At 6·5 kg live weight the pigs reared on the sow contained less deoxyribonucleic acid (P < 0·05) in the m. adductor than pigs from either of the two artificially reared groups. Muscle deoxyribonucleic acid at 20 and 75 kg live weight was unaffected by nutrition before 6·5 kg live weight or level of feeding between 6·5 and 20 kg live weight.

1. Forty-two piglets were used to study the effects of a low, medium or high level of energy intake (1·0, 1·4 and 1·9 MJ gross energy per kg W0·75 per day respectively) from 1·8 to 10 kg live weight and a low or high level of energy intake (1·4 and 1·8 MJ digestible energy per kg W075 per day respectively) subsequent to 10 kg live weight on growth performance, body composition and the cellularity of muscle and subcutaneous adipose tissue to 30 kg live weight. During both live-weight periods all pigs received the same daily intake of crude protein.

2. Raising energy intake in the period prior to 10 kg live weight increased (P < 0·-05) growth rate, body fat content and fat cell size but reduced food conversion efficiency, body protein and water (P < 0·05) and muscle deoxyribonucleic acid. These effects on body composition and muscle deoxyribonucleic acid at 10 kg live weight were still evident at 30 kg live weight.

3. Subsequent to 10 kg live weight, pigs previously given the lowest energy intake deposited protein and fat at a faster rate and exhibited more rapid and efficient growth (P < 0·05) than pigs previously given the high energy intake.

4. At 30 kg live weight pigs given the two higher levels of energy intake before 10 kg live weight contained less deoxyribonucleic acid (P < 0·05) in subcutaneous adipose tissue and had larger (P < 0·05) fat cells than those given the lowest energy intake before 10 kg live weight.

5. The effects of energy intake subsequent to 10 kg live weight on growth performance, body composition and the cellularity of muscle and adipose tissue were qualitatively the same as those for the period 1·8 to 10 kg live weight.

Nursing frequency, suckling duration and teat order were recorded in 12 ‘hybrid’ sows and their litters from parturition to weaning at day 42 post partum. Sows and their litters were moved from the farrowing quarters at day 20; half were housed as a group (G) and fed ad libitum with a boar present and the other half were housed individually (S) with no boar present and ration-fed to scale.

Some differences occurred in behaviour from day 20. A significantly greater number of false nursing periods, significant synchronization of nursing periods, and a tendency for more long and short nursing intervals occurred in G than S sows and litters. The duration of the preliminary nosing phase and the total nursing period were significantly longer in S than G litters. Teat order scores, indicating consistency of piglet suckling position on the sow, tended to be greater in S than G litters, and there was a fall in the score for G litters in the weeks following grouping. This was associated with a check in live-weight gain.

All G sows showed oestrus during lactation, on average 15 days after grouping. There was no incidence of lactational oestrus in the S sows which showed oestrus 5 days after weaning.

Eighty multiparous Camborough hybrid female pigs (sows) and their litters, on average 15 days post partum, were housed in groups of five in multi-accommodation pens. The sows were fed ad libitum on a proprietary diet containing 160 g crude protein per kg from grouping to 30 days post partum, when they were fed to scale twice daily in individual feeders until weaning at 42 days post partum. One of four Large White male pigs (boars) was introduced to each group on the day following grouping, each boar being run with four groups.

Oestrus occurred in 62 sows, on average 10·2 ± 3·3 days after grouping. The differences between groups in the interval from grouping and boar introduction to the onset of oestrus were statistically significant (P < 0·01). Neither the incidence nor the timing of oestrus could be related to the identity of the boar.

Regression analysis revealed relationships between a number of independent variables and the incidence and timing of lactational oestrus. Sows which gained more live weight from farrowing to weaning and whose litters consumed more creep food were more likely to show lactational oestrus. Sows were slower to show lactational oestrus when they gained little weight from farrowing to weaning, had lower live weights at weaning and gained less weight from the previous weaning, despite eating comparatively more food during grouping. The sows which took longer to show lactational oestrus also had heavier piglets at grouping and more piglets in their litters at weaning. The difficulties arising in the interpretation of these results are discussed.

Maturing patterns are established for 26 body organs using data from 35 rams of two strains of Merinos of different mature size (97 and 120 kg).

The proportion of shorn full live weight represented by each structure at maturity was very similar for all organs except the head, hide and distal limbs which were relatively heavier in the small strain. and small intestine and blood which were relatively heavier in the large strain.

All organs were early maturing relative to shorn full live weight except the hide and the penis plus bladder, which matured at the same rate as live weight. The maturity coefficients differed between strains only for the head, thyroid and omasum.

Differences in the proportion of live weight comprised by each organ in the two strains were compared at the same shorn full live weight and at the same proportion of mature shorn full live weight. Differences in the various portions of the alimentary tract and other internal organs, with the exception of the kidneys, which were evident in comparisons at the same weight, were reduced when compared at the same proportion of maturity. The difference between strains for head, hide and distal limbs was greater at equal maturity than at equal weight suggesting a functional relationship to body weight rather than to ultimate mature body size.

It was concluded in genetic comparisons of animals of different mature size, the most appropriate basis of comparison of relative organ weight will be at the same proportion of mature live weight.

The pattern of food intake as a function of age, and live weight as a function of food consumed was examined from soon after weaning to maturity in groups of fine and strong wool Merino rams and Dorset Horn rams and wethers. Each group, which initially comprised 20 sheep and from which individuals were removed for slaughter at approximately 6-kg increments in live weight, was fed a pelleted ration ad libitum for 98 and 77 weeks for the Merino and Dorset Horn groups respectively.

The pattern of food intake was similar for all groups of sheep, in that food intake increased to a maximum at approximately 50 weeks of age and then declined with age with a regular oscillation superimposed on this curve. In all three groups of rams, but not in the Dorset Horn wethers, the oscillations in food intake were on an approximate annual cycle with a decrease in food intake in summer and an increase in winter.

Estimated mature live weights for the strong and fine wool Merino rams were 120·5 and 97·6 kg, and for the Dorset Horn rams and wethers 108·6 and 101·4 kg. Variation between breeds in the level of food intake, food efficiency and consequently the pattern of growth, was largely a function of mature size. The Dorset Horn rams were slightly more efficient at converting food to live weight than the wethers.

1. Two experiments were conducted to determine the feeding value of high moisture wheat straw treated with ammonia. The straw was chopped, water added to attain 350 g dry matter per kg treated straw, then treated with 84 or 90 g anhydrous ammonia per kg wheat straw dry matter and stored in a large plastic bag.

2. In Experiment 1, the treated straw was provided ad libitum to lambs in individual pens with either a control supplement (molasses and minerals), or 200 or 400 g of a soya bean meal- or maize gluten meal-based supplement.

3. Lambs fed the control supplement consumed 24·8 g of treated straw dry matter per kg live weight and lost 32 g live weight per day during the 28-day study. Source of supplemental protein did not affect live-weight gain or treated straw consumption. Level of supplement increased live-weight gains (P < 0·01) but did not affect treated straw consumption.

4. In Experiment 2, the lambs were fed one of three levels of supplement that each supplied 58 or 116 g of crude protein per day. Supplemental protein level increased cellulose digestibility (P < 0·01) with no effect on hemicellulose digestibility. Protein level did not affect treated straw consumption or live-weight gain.

5. Supplement level decreased cellulose digestibility (P < 0·01), hemicellulose digestibility (P < 0·01), and treated straw intake (P < 0·01) but increased live-weight gain.

6. The results are discussed in relation to the effects of amount and type of supplemental protein and level of supplementation on utilization of the treated straw.

A discrete-time population model is presented which is specific to the characteristics of a breeding colony. It is intended to be a rigorous yet an easily applied model. The model is based on a female-dominated demographic system with constraints on colony size. Fecundity and survival probabilities are incorporated into a net reproductive rate which is age-specific and time and population size invariant. The model is applied to solve an optimization problem for a breeding colony of African Green monkeys and to examine a set of external constraints imposed on the colony.

A simple criterion is described for selecting which of a number of sample joints estimates most precisely for a given cost (T) the mean lean percentage of a population of carcasses, using double sampling. It is shown that the best joint is the one giving the largest value of:

where ρ is the correlation between the lean content of the sample joint and the lean content of a carcass side (y), and k is the cost of the sample joint dissection as a proportion of the cost of side dissection. The optimum proportion of carcasses to side dissect is:

and the standard error of the estimated mean is then:

where σyis the standard deviation of y and C is the cost of measuring y.