A protocol of infrequent, but strategically timed milk-sampling was established for predicting the time of ovulation, and thus the optimum time for insemination, in lactating dairy cows. In the experimental group (no. = 49), the time of ovulation was predicted on the basis of a fall in milk progesterone concentrations, which was identified by the use of ‘on-farm’ progesterone assay kits. Reproductive performance in these animals was compared with that in control cows (no. = 45) that were inseminated solely on the basis of oestrous detection by stockmen. The accuracy of ovulation prediction, and of oestrous detection, was assessed from progesterone profiles based on milk samples collected three times weekly from all 94 cows. These milk samples were analysed by laboratory enzyme-immunoassay after the end of the trial.

Over the period during which milk samples were monitored with ‘on-farm’ progesterone kits, a total of 88 ovulations occurred in the experimental group. The progesterone-testing protocol accurately predicted 87 (99%) of these. Over the same period, there was a total of 81 ovulations in the control group and 63 (78%) of these were associated with correct oestrous detection. Conception rates to correctly timed insemination did not differ significantly between groups, but by the end of the three-cycle experimental period, significantly more of the cows in the experimental group (21·8%) than in the control group (4·8%; P < 0·05). In contrast, the use of ‘on-farm’ progesterone results to confirm oestrus avoided any mistimed inseminations (13% of inseminations in the control group). By using the milk-sampling and ‘on-farm’ progesterone-testing protocol, only 1% of ovulations were not accompanied by a correctly timed insemination. This compared with 22% of ovulations in the control group not associated with an insemination because oestrus was not detected. By ensuring that all ovulations are associated with a correctly timed insemination, herd reproductive performance can be significantly improved.

An experiment was carried out to investigate the effects of compound composition and silage characteristics on silage intake, feeding behaviour and productive performance of dairy cows during the first 25 weeks of lactation. Over a period of 3 years, 36 lactating British Friesian cows (12 per year), in their third or later lactations, were divided into two groups. The cows in each group received either compound S, in which the principal energy source was cereal starch, or compound F containing a mixture of high quality digestible fibre. Cows given compound F ate 2·2 kg more silage dry matter per day (P < 0·05). The type of compound had no effect on the frequency of silage feeding and the time spent eating was significantly different only over weeks 10 to 25 of lactation (P < 0·05), with cows on compound F spending on average 20 min longer feeding per day. Cows on compound F produced 1·7 kg more milk per day than cows on compound S. Although there were no significant differences in the concentration of milk constituents, compound F was associated with higher yields of milk constituents. Significant differences were found between the years of experiment in the frequency of feeding (P < 0·05) and also in the concentration of milk protein and milk fat. There were no significant differences in the magnitude of live-weight change between treatments or year of experiment.

The effect of increased female reproductive rates on selection response, on efficiency of progeny testing and on the openness of the nucleus was investigated in open nucleus breeding plans. Conventional progeny testing plans and closed nucleus plans are special classes of open nucleus plans. In the open nucleus plans, generation intervals and selection across tiers were optimized. The number of offspring per elite dam was varied from 1 to 41, progeny testing of young bulls in the female base population was varied from 0 to 100 test records and the size of the nucleus was varied from 250 to 2000 young bulls born per year. Also efficiency of selection was varied: efficient selection in T(heoretical)-schemes and less efficient selection in P(ractical)-schemes. Especially, selection of base parents was less efficient i n P-schemes.

The deterministic prediction model took account of variance reduction due to selection and reduction of selection differentials due to correlations between estimated breeding values of relatives (order statistics). For closed nucleus plans, the results of the model were verified with Monte Carlo simulation results.

By increasing female reproductive rates, genetic gain increased by a factor 0·08 and 0·16 for the T- and P-schemes respectively. The nuclei in P-schemes were less open, due to the less efficient selection in the female base population. Schemes that were less open benefited more from increased female reproductive rates because selection differentials in small nuclei increased more than those in large base populations. The optimal open nucleus plan became less open with increasing female reproduction. Generally, progeny testing of bulls reduced genetic gain (by up to a factor 0·1) but it also reduced inbreeding rates. Progeny testing was more efficient in schemes that were less open: in P-schemes with 41 offspring per dam, progeny testing increased genetic gain. With many offspring per dam there were fewer full-sib families, causing lower selection differentials due to order statistics effects. This effect could be prevented by increasing the size of the nucleus.

The importance of family sizes in adult multiple ovulation and embryo transfer (MOET) nucleus schemes with discrete generations of single trait selection was examined using Monte Carlo simulation. Two areas were investigated. Firstly, the number of sons and daughters per dam was varied in schemes using hierarchical mating designs. With four or eight sires and 32 dams selected, increasing the number of sons per dam from one up to four achieved 1 to 8% higher rates of response but at the expense of increased variation in response and 10 to 56% higher rates of inbreeding. With four or eight sires and 16, 32 or 64 dams selected, the number of daughters was set to two, four or eight (with one son per dam in each case). For schemes transferring equal numbers of embryos, responses were lower with two daughters per dam but were fairly similar with four or eight daughters per dam while inbreeding rates increased as fewer sires and dams were selected. Secondly, the effects of variation in family sizes due to biological factors and chance were investigated with eight sires and 32 dams selected and with hierarchical or factorial (two or four sires per dam) mating designs. When all selected cows yielded embryos, changes in family sizes due to differences in sex ratios, in survival rates of embryos to selection and to variation in the number of embryos per donor reduced response by 1 to 4%. However, when 20% or 33% of the superovulated females yielded no embryos, thus requiring the use of genetically inferior replacements, response was reduced by a further 9 to 13%

Dairy cattle were injected at intervals of 4 weeks with recombinantly derived bovine somatotropin (bST) i n a slow-release form (somidobove, Lilly Research Centre Ltd, Windlesham, Surrey), to examine the effect on milk production. In the first experiment with 38 autumn-calving animals in a randomized-block design, treatment lasted for 28 weeks and a control and three doses of bST: 320 (low), 640 (medium) and 960 (high) mg were compared. Over the 28 weeks the mean daily milk yields were: control 23·4, low 23·4, medium 25·4 and high 25·4 (s.e.d. 1·17) kg. Milk composition was not affected by bST. In experiment 2 with 40 autumn-calving animals, the medium dose of bST was used for 24 weeks in animals given two levels of concentrate in a factorial design. On average there was a non-significant increase in daily milk yield of 1·3 kg in the bST-treated cattle; there was no interaction between bST and level of concentrate. There were no effects of bST on milk composition. Changes in live weight and body condition of the cattle and their food intake were measured. It was concluded that the trends toward higher milk yields were supported initially at the expense of live weignt and possibly later by increased food intake. In a third experiment 10 grazing cows were given three consecutive doses of 640 mg bST and 10 similar cows acted as controls. There were no significant effects of bST on milk production and food intake. Reasons for the low responses in milk production compared with several recent American and European experiments may possibly be the pattern of use of the slow-release form of bST and the low content of dry matter in the maize silages which restricted forage intake and response in milk production.

Resting heat production (H), respiratory rate (RR) and rectal temperature (Tr) were measured at different controlled temperatures (Tt) in Hariana (Bos indicus) and its F, crosses with Jersey (JH), Brown Swiss (BH) and Holstein Friesian (FH) (Bos taurus) breeds and the values obtained were used to assess their relative thermosensitivity.

The lowest Tt at which H significantly decreased from that at 17°c was 32°c for Hariana, JH and BH and 27°c for FH after exposure for 8 days. The corresponding values after exposure for 18 days were 37°c for Hariana and 32°c for all the three crossbred groups. Differences between the genetic groups were also significant. The lowest Tt at which metabolizable energy (ME) decreased significantly in comparison with those at 17CC was 32°c in all the genetic groups. The differences in ME intake between genetic groups were significant only at 32°c Tt. The lowest Tt at which RR significantly increased from those at 17°c were 32°c in Hariana, 27°c in JH, BH and FH for both 5 to 7 and 15 to 17 days of exposure. The corresponding Tt for increase in Tr was 37°c in Hariana, 32°c in JH and 27°c in BH and FH at both 5 to 7 and 15 to 17 days of exposure.

The ambient temperature at which H would have significantly decreased and RR and Tr increased from the respective values at 17CC Tt were calculated curvilinearly for different genetic groups. There were differences in these values of calculated ambient temperatures between genetic groups and between exposure durations in respect of H, RR, and Tr, indicating differences in thermosensitivity.

Three randomized-block experiments involving 87 British Friesian, 18 Simmental × Friesian and 18 Hereford × Friesian, artificially reared calves (mean initial live weight 47 kg and age 9 days), were carried out to examine the effects of the quantity of milk substitute consumed on the lifetime performance of beef cattle. The three treatments were (1) 400 g milk powder per head daily, weaned at 42 days of age (2) ad libitum intake of milk powder, weaned at 42 days and (3) ad libitum intake of milk powder, weaned at 60 days. All calves were individually penned and fed until after weaning; those on treatment 1 were bucket fed twice daily, while those on treatments 2 and 3 sucked the milk from 25·1 containers through artificial teats. They were offered a pelleted, barley/soya-bean meal concentrate (197 g crude protein per kg dry matter (DM)) ad libitum from purchase until intake reached a maximum of 2·7 kg per head daily, and well preserved grass silage ad libitum from 6 weeks of age. Three commercial, skimmed milk-based, acidified milk substitutes (242 and 187 g crude protein and oil per kg respectively) were used. Intakes of milk substitute, concentrate DM and silage DM per calf to 12 weeks of age for treatments 1 to 3 respectively were as follows: 13·0, 39·7 and 54·8 (s.e. 1·67) kg; 88, 73 and 58 (s.e. 1·5) kg and 9·4, 10·6 and 11·2 (s.e. 0·23) kg. Live-weight gains to 12 weeks of age and carcass weights adjusted to 19 months of age were 0·71, 0·82 and 0·84 (s.e. 0·026) kg/day and 335, 340 and 341 (s.e. 3·8) kg. It is concluded that increasing the input of milk substitute above 400 g per head daily until 6 weeks of age produced only a small and uneconomic increase in the lifetime performance of beef cattle.

An analysis of cow reproductive traits was carried out on 2057 lactation records of Egyptian buffaloes. A total of 651 daughters (paternal half-sisters) of 82 sires were available for the analysis. Reproductive traits analysed were age at first, second and third calving (AC), number of services per conception (NSC), gestation length (GL) and calving interval (CI). Year of calving affected most reproductive traits (P < 0·001) while no important differences were detected among months of calving. Means of NSC, GL and CI decreased linearly (P < 0·01) as parity advanced. Age of cow at calving exerted a pronounced effect (P < 0·001) on NSC and CI. There was no systematic change in the estimates of heritability and sire components of variance for reproductive measures over the first three lactations. Heritability estimates for NSC and GL in the different parities were generally low, ranging between 0·04 and 0·18, while moderate estimates for AC and CI were obtained. Phenotypic correlations between reproductive traits were positive and relatively moderate or low. For most traits, positive and relatively high or moderate estimates of genetic correlations were observed. Mating bull was found to be an important source of variation for reproductive performance of Egyptian buffaloes. The highly significant effect of mating bull on reproductive traits could be considered as evidence for the importance of evaluating bull semen quality before the mating of cows

A series of change-over experiments examined the intake and digestibility of straw-based diets in yearling dairy heifers. Voluntary intakes of straws given as the sole food to yearling dairy heifers were 1·28, 1·20 and 0·90 kg dry matter (DM) per 100 kg live weight for barley, wheat and oat straws respectively, and the corresponding digestibility of organic matter was 0·47, 0·53 and 0·48. When supplemented with 2 kg rolled barley the intakes of straw did not alter, nor were they affected by replacing part of the rolled barley with an equal amount of soya-bean meal and thus increasing the dietary concentration of protein from about 50 to 110 g/kg DM. The use of molassed sugar-beet pulp in place of rolled barley reduced the intake of straw. Flaked maize promoted slightly higher intakes of straw than an equal amount of coarsely milled maize. Replacing rolled barley by lucerne hay reduced straw intake. It is estimated that, to provide sufficient energy and protein for a daily gain of 0·5 kg, heifer straw-based diets would require a supplement of 1·2 kg rolled barley and 0·8 kg soya-bean meal.

The chemical composition of the muscle and fatty carcass tissues of Hereford × Friesian (HE), Friesian (FR) and Charolais × Friesian (CH) steers finished on high (H) (12·6 MJ/kg dry matter) and medium (M) (10·4 MJ/kg dry matter) metabolizable energy (ME) concentration diets and slaughtered at low (L), normal (N) and heavy (W) carcass weights was determined in a 3 × 2 × 3 (no. = 9) factorial experiment. There were also pre-finishing slaughter groups of nine animals of each breed type. Target L, N and W carcass weights were 260, 300 and 340 kg, for HE and FR and 260, 320 and 380 kg, for CH, respectively.

Main effect side soft tissue weights of the finished groups were 128, 125 and 134 (s.e.d. 1·3) kg for HE, FR and CH, 133 and 125 (s.e.d. 1·1) kg for H and M and 108, 129 and 150 (s.e.d. 1·3) kg for L, N and W, respectively. Soft tissue chemical composition was 540, 562 and 590 (s.e.d. 5·0) g/kg moisture, 162, 174 and 181 (s.e.d. 1·7) g/kg protein and 288, 254 and 220 (s.e.d. 6·6) g/kg lipid for HE, FR and CH and 558 and 570 (s.e.d. 4·1) g/kg moisture, 170 and 175 (s.e.d. 1·4) g/kg protein and 262 and 246 (s.e.d. 5·4) g/kg lipid for H and M, respectively. The allometric regression coefficients for moisture, protein and lipid weights on soft tissue weight were 0·65, 0·69 and 206, respectively. The coefficients for moisture and protein weights on muscle and fat weights were < 1·0, whereas for lipid, they were >1·0. The breed differences in soft tissue chemical composition paralleled differences in physical composition. Dietary ME concentration had little effect on the chemical composition of individual tissues. For the M diet, HE, FR and CH were calculated to have similar soft tissue lipid concentrations (250 g/kg) at side soft tissue weights of 113, 125 and 157 kg, respectively. Corresponding side muscle weights at similar muscle lipid concentration (70 g/kg) would be 77, 90 and 108 kg.

Forty-two weaned suckled Charolais-cross steers were used to measure changes in body composition during compensatory growth in growing cattle. Six cattle were slaughtered initially and the remaining 36 allocated to either a low level of feeding to 350 kg live weight followed by a high level (LH) or a high level of feeding throughout (HH). Above 350 kg live weight, food intake on both treatments was the same at any given live weight. Six cattle were slaughtered from each treatment at 350, 400 and 450 kg live weight. From initial live weight (259 kg) to 350 kg, live-weight gains were 0·45 and 0·78 kg/day for the LH and HH treatments respectively (P < 0·001). From 350 to 400 kg live weight, live-weight gains were 1·35 and 0·98 kg/day (P < 0·01) for the LH and HH cattle respectively, while from 400 to 450 kg live weight there was no significant difference (1·38 v. 1·20 kg/day). The LH cattle contained less fat in the empty body than the HH cattle at 350 kg (118 v. 153 g/kg; P < 0·05) and 400 kg live weight (117 v. 169 g/kg; P < 0·01), but at 450 kg there was no significant difference between treatments. From 350 to 400 kg live weight the composition of the empty body-weight gain was 663 g water, ' 108 g fat and 216 g protein per kg in the LH cattle and 422 kg water, 311 g fat and 173 g protein in the HH cattle. From 400 to 450 kg live weight the equivalent figures were 491, 291, 156 g/kg for the LH cattle and 744, 67 and 203 g/kg for the HH cattle. The results demonstrate that following a period of food restriction the empty body-weight gain of cattle initially comprises increased proportions of protein and water and a reduced proportion of fat compared with unrestricted cattle when both are given the same amount of food and compared at the same weight. There then follows a second phase in which the proportion of fat increases and the proportions of protein and water decrease.

In two experiments involving ovariectomized Scottish Blackface ewes and steroid replacement therapy, the effects of level of food intake and of body condition on hypothalamic and pituitary sensitivity to progesterone and oestradiol feedback were investigated. In experiment 1, groups of 16 ewes in similar levels of body condition were given dried grass pellets ad libitum (mean dry-matter (DM) intake of 2481 (s.e. 71) g per head per day; group H) or 1105 g DM per head per day (group M). In experiment 2, groups of 16 ewes were fed so that they achieved body condition scores of 2·70 (s.e. 0040; group HBC) or 1·67 (s.e. 0·036; group LBC) prior to the start of the experiment. All ewes were then fed to maintain their respective levels of body condition. In both experiments, all ewes were ovariectomized and intravaginal progesterone pessaries and subcutaneous oestradiol implants were inserted; these were designed to induce circulating concentrations of these steroids within the normal range for intact ewes. Approximately 1 week later, blood samples were collected at 15-min intervals for 10 h, and at 10-min intervals for 8 h before and after pessary withdrawal respectively. All samples were assayed for LH and selected samples were assayed for FSH and progesterone. Gonadotropin profiles were also recorded during 2-h periods following injection (i.v.) of 0·25 μg and 5 μg GnRH. When circulating progesterone concentrations were elevated, due to the presence of a progesterone pessary, there was no effect of level of food intake on mean LH or FSH concentrations or mean LH pulse frequency but H ewes had a higher mean LH pulse amplitude than M ewes (P < 005). When the progesterone pessaries were absent, H ewes had a higher mean LH concentration (P < 001), mean LH pulse frequency (P < 001) and LH pulse amplitude (P < 0·01) than M ewes but mean FSH concentrations were similar in the two groups. Irrespective of the presence or absence of elevated progesterone concentrations, the pituitary responses to GnRH injection were either not significantly different or were lower in H compared with M ewes. In experiment 2, HBC ewes had a higher mean LH concentration (P < 001), LH pulse frequency (P < 0·001), pulse amplitude (P < 0·001) and FSH concentration (P < 001) than LBC ewes, irrespective of circulating progesterone concentrations. The mean heights and ‘areas’ of the LH peak induced by injection of 0·25 μg GnRH were larger in HBC than LBC ewes (P < 005) irrespective of circulating progesterone concentrations. When mean concentrations before injection were taken into account, the corresponding increases in FSH concentration were similar for the two groups. Following injection of 5 u.g GnRH, the differences in peak size were generally not significant. It is concluded that during the follicular phase of the cycle, a low level of intake is associated with increased hypothalamic sensitivity, but not pituitary sensitivity to oestradiol feedback. Body condition affects both hypothalamic and pituitary activity but the relative importance of direct and indirect effects of body condition on the sensitivity of these organs to steroid feedback, remains to be elucidated.

A total of 1754 records for litter traits of 458 D'man ewes sired by more than 45 rams were analysed to estimate repeatability, heritability and genetic and phenotypic correlations. Repeatability estimated by the intraclass correlation method was 0·11, 0·11, 0·12 and 0·15 for litter sizes at birth and at weaning and litter weights at birth and at weaning, respectively. Paternal half-sib estimates of heritability were 0·09 (s.e. 0·06), 0·04 (s.e. 0·05), 0·15 (s.e. 0·07) and 0·08 (s.e. 0·05), respectively. Genetic correlations among the traits were all positive and varied from 0·17 to 0·72, whereas phenotypic correlations ranged from 0·68 to 0·86.

Eleven sun-cured hays with crude protein (CP) and neutral-detergent fibre (NDF) concentrations ranging from 57 to 207 and 428 to 744 g/kg dry matter (DM) respectively, were used to study the relationship between the DM and NDF degradation characteristics and the voluntary DM intake (VDMI) by sheep.

The VDMI was measured in the last 10 days of a 21-day period using 50 mature ewes in five groups of 10 ewes per group, with a mean live weight (LAY) of 55 kg and ranged from 41·2 to 69·8 g/day per kg LW075. Disappearanc e rates of DM and NDF were measured by incubating samples in nylon bags in the rumens of three hay-fed rumen-cannulated ewes for 3, 6, 9, 15, 24, 48 and 72 h and by fitting the exponential model Y = a + b (1 — e-ct) to the results. Potential degradabilities (defined as a + b) for DM and NDF ranged from 679 to 868 and from 445 to 808 g/kg respectively.

The VDMI was significantly correlated with the soluble fraction of DM (aDM; P < 0·05) and the rate of degradation of DM (cDM; P < 0·01) and NDF (cNDF; P < 0·01). The inclusion of aDM and cDM ni a multiple regression model resulted in the equation: VDMI = 21·3 + 0·0733 (s.e. 0·01577) aDM + 138 (s.e. 26·0) cDM (R2 = 0·897; residual s.d. = 2·79), with a significant increase of the variance explained by one single degradation characteristic (P < 0·05). The prediction was not improved by adding the constant cNDF to the equation (P > 0·05). These results indicate the possibility of predicting the VDMI of hays by sheep from the rumen soluble fraction (aDM) and the rate of degradation (cDM) of DM.

Thirty-seven Cambridge (C) and Suffolk (S) × C ewes and their lambs were housed and penned individually in family groups for 35 days post partum and then grazed on pasture until the lambs reached slaughter weight. Lambs were reared as twins (TR-TW). The lambs weaned at 35 days were artificially reared and formed the TR-AR group.

Dry matter intake of the complete diet and hay offered ad libitum during the housing period was unaffected by rearing type. TR-TR and TR-TW ewes produced more milk at 10 days than TW-TW ewes, 4·4 and 4·2 v. 3·6 (s.e.d. 0·257) kg/day. From 35 days onwards the decline in milk yield tended o t be greater in the TW-TW and TR-TW ewes which also gained more condition than TR-TR ewes, 0·33 and 0·37 v. 0·01 (s.e.d. 0·139) of a score.

TW-TW lambs grew faster than TR-TW and TR-TR lambs during the housed period, 336 v. 295 and 292 (s.e.d. 9·8) g/day but after 35 days TR-TW, TR-AR and TW-TW lambs had similar growth rates, 348, 350, and 348 (s.e.d. 14·2) g/day respectively whereas TR-TR lambs grew less quickly at 310 g/day. Forty-two male lambs were slaughtered. TR-AR lambs had a higher killing-out proportion and carcass fat content than naturally reared animals. In the latter group carcass fat content was negatively related to time reared as triplets and the kidney knob and channel fat weight was significantly lower in the TR-TR group. Crossbred lambs grew faster, had a higher killing-out proportion and greater weight of m. longissimus dorsi and m. psoas major in the half carcass.

It is concluded that weaning one lamb at 35 days provides a suitable alternative system for rearing triplets and that it is possible to produce good quality carcasses with some desirable reduction in fat content from lambs reared as triplets throughout their life.

Over 4 years, a flock of mature North Country Cheviot ewes was grazed on different swards between weaning in August and mid October to achieve a range of condition scores at 5 weeks before mating (post-weaning period). During the subsequent 5 weeks before a synchronized mating (pre-mating period), all ewes were grazed on a perennial ryegrass/white clover sward, provided at high (H) or low (L) herbage masses (1600 to 2100 or 1200 to 1400 kg dry matter (DM) per ha, respectively). Ovulation rate was derived from the number of corpora lutea in 101 ewes slaughtered after first mating. Conception and lambing rates were derived from the number of lambs born to first mating of 406 ewes. In 2 out of the 4 years, 79 ewes were used to estimate herbage intake.

There was no effect of post-weaning management on reproductive performance. Herbage intake and reproductive performance were examined in relation to the body condition of the ewes at 5 weeks before mating. On an H herbage mass pre-mating, herbage intake in the period immediately prior to mating was greatest in ewes in low body condition and declined significantly with increasing condition. On an L herbage mass pre-mating, herbage intake did not vary significantly with condition. Ovulation rate was low in ewes in low body condition (scores =≤·25) at 5 weeks before mating, increased significantly with increasing condition up to scores 2·50/2·75 but did not change significantly above such intermediate scores. Conception and lambing rates increased significantly with increasing body condition up to scores 2·50/2·75 and then decreased significantly as condition increased above these levels. These effects of body condition were greater on an L herbage mass pre-mating than they were on an H herbage mass. Ewes in the intermediate body condition scores of 2·50 and 2·75 at 5 weeks before mating had a low sensitivity to contemporary nutrition pre-mating. Such condition levels are therefore considered to be above the range of condition in the North Country Cheviot breed within which reproductive response is affected by contemporary nutrition pre-mating.

The lower reproductive performance of ewes in high body condition (scores 3=3·00) at 5 weeks before mating was related to an increase in estimated ova and embryo loss and may be associated with a reduced herbage intake.

The effects of high-low (HL) and low-high (LH) patterns of nutrition during the 5 weeks prior to a synchronized mating were studied on the reproductive performance of 115 Welsh Mountain and 114 Brecknock Cheviot ewes in a range of body condition at the start of treatment. For 2 months prior to treatment imposition, ewes were grazed in a systematic way at different stocking rates on different sward heights to obtain a range of body condition scores. Target score groupings were 2·00 to 2·25, 2·50 to 2·75 and 3·00 to 3·25 and although scores of 1·50 to 3·25 were obtained, most lay in the 2·00 to 2·75 range. Two treatment groups, balanced for live weight and body condition score, received high and low levels of nutrition for 16 days from mid October, achieved, respectively, by a low stocking rate on a sward with a high surface height plus ad libitum concentrate and a high stocking rate on a sward with a low surface height plus 200 g hay per head per day. Treatments were then reversed for the 17 days prior to mating. Live-weight and body condition-score changes were recorded and reproductive performance at first mating was measured from counts of corpora lutea and viable embryos at slaughter either at return to service or at 4 weeks after first mating. There were no differences due to nutritional pattern in live weight or body condition score at mating. The LH treatment significantly increased the rates of ovulation, conception, multiple ova survival and potential lambing per ewe pregnant and per ewe mated in the Welsh Mountain breed but significantly increased only the rates of ovulation and potential lambing per ewe pregnant in the Brecknock Cheviot breed compared with the HL treatment. The relatively poorer response in the latter breed was partially due to a lower ovulation rate potential coupled with non-significantly lower rates of conception and survival of single-shed ova associated with the LH treatment. Much of this relatively poorer reproductive performance in the Brecknock Cheviot breed was in ewes with body condition scores 3= 2·75. Reproductive performance increased with increasing body condition at the start of treatment over the range of scores =≤ 2·25 to 2·50 in both breeds and, while continuing to improve in scores above 2·50 in the Welsh Mountain breed, in the Brecknock Cheviot breed it started to decline. Ewes with a body condition score of 2·50 in both breeds showed the greatest response in potential lambing rate to the LH treatment.

An experiment was conducted to study the ability of the pig to recover from the effects of a period on a food deficient in crude protein (CP). Forty young pigs were given free and continuous access to foods with either 150 (L) or 252 (M) g CP per kg in period 1 of the experiment, from 6·3 kg to 13·4 and 12·3 kg live weight respectively. These live weights were expected to give equal lipid-free empty body weights. In period 2, four males and four females from each of the period 1 treatments were given access to either M or a food with 377 g CP per kg (H) to a live weight of 30 kg, when the 32 pigs were killed.

Pigs on L took 11 (s.e. 0·6) days longer to complete period 1, and had, at the end of this period, 0·20 (s.e. 0·03) kg less protein and 1·20 (s.e. 0·06) kg more lipid in their bodies than the M pigs, at a common ash weight. In period 2, pigs from L grew at a faster rate (750 v. 633 (s.e.d. 20) g/day), ate food at the same rate (1115 v. 1085 (s.e.d. 35) g/day) and converted food more efficiently (0·676 v. 0·585 (s.e.d. 0·016) g gain per g food) than those from M. At 30·3 kg live weight the pigs from L had corrected their protein deficit relative to ash and reduced their fatness, so that they had the same protein: ash ratio and only 0·47 (s.e. 0·12) kg more lipid in their bodies than those from M. This was the result of a higher rate of gain of protein and water, a lower rate of lipid gain and similar rate of ash gain by the pigs from L than those from M. In the first 7 days of period 2 the pigs from L gained weight at 1·4 times the rate of those from M. In the final 7 days there was no significant effect of period 1 treatment on growth rate. The pigs from L given food H in period 2 were more efficient than those given M in period 2 (food conversion efficiency (FCE) values of 0·884 and 0·791 respectively; s.e.d. 0·027), but this difference was reversed in the final 7 days (FCE values of 0·521 and 0·603 respectively). t I is concluded from these results that a period of eating a food of low protein content produces a reduced protein: ash and an increased lipid: ash ratio in the body and reduced growth rate and efficiency. When subsequently pigs are given a food of sufficiently high protein content, the protein: ash and lipid: ash ratios return to normal. The repletion of labile protein reserves, with their associated water, leads to a substantial increase in the rate of live-weight gain. The lower lipid content of the gain leads to a high efficiency. The duration of these effects depends on the protein content of the food given.

A long-term experiment was designed to study the effect of feeding different sources of crude fibre during pregnancy on the reproductive performance of sows. The four treatments imposed during pregnancy were: C, a control compound diet; CS, the same control diet with additional chopped straw (0·2 kg/day); CM, the same control diet in which 0·25 kg was replaced daily by 1·5 kg maize silage; and CFR, a compound diet with a higher crude fibre content. In total 567 sows were used. On the treatments C, CM, CS and CFR the number of sows was 147, 142, 147 and 133 respectively. The sows remained on the experiment for five successive parities on a treatment or until the moment of culling. Feeding levels were about 32 MJ digestible energy (DE) per day during the first months of pregnancy and about 40 MJ DE per day in the last month of pregnancy. During lactation the feeding level ranged between 75 and 90 MJ DE per day.

In sows giving birth to five litters, the effect of parity was calculated irrespective of the treatments. The optimum litter size was observed in the fourth litter and the percentage of successful services within 10 days post weaning stabilized after the third parity. The live weight of the sows increased gradually up to the fifth parity.

Compared with the sows on the control treatment the proportional changes in consumption of the sows on the other treatments were: 0·45 more crude fibre, up to 0·02 more digestible energy and up to 0·06 less digestible protein.

There were only slight differences in reproductive performance. The mean number of weaned piglets per sow per year were 19·7, 19·3, 20·1 and 20·2, respectively for the treatments C, CM, CS and CFR. Higher piglet birth weights were observed for sows on the CM treatment and a higher proportion of the culled sows had leg disorders. On the CS treatment a higher number of weaned piglets per litter was observed and this was due to the higher number of piglets born and lower piglet mortality during lactation. The sows on this treatment tended to have more difficulty in becoming pregnant after the third and fourth parities. The sows on the CFR treatment had the highest body weights and had a higher number of litters per year mainly due to lower numbers returning to service and less nonproductive days. The results from this experiment gave no clear evidence that either the inclusion of roughage or a higher crude fibre content in the diet improved reproductive performance.

An experiment was designed to test whether pigs can differentiate between two foods differing in their protein content. A control food (160 g protein per kg food) and six choice-feeding treatments, in which the diets differed only in their protein concentrations, were used in the experiment. Pigs were given the following choices of protein: 220 v. 180; 220 v. 140; 220 v. 100; 180 v. 140; 180 v. 100 and 140 v. 100 g protein per kg food. Group data were collected on 240 Landrace × Large White pigs during the growing phase (30 to 90 kg), all pigs used being of similar genetic background. A 2 × 7 factorial design was replicated twice, with 10 pigs per pen, and with barrows and gilts being penned separately. Food bins for the choice-feeding treatments were placed side-by-side and an 8-day training period, in which the pigs were allowed access to only one of the two foods on alternate days, was used at the beginning of the trial to allow them to learn the position, taste and physiological effect of each of the two foods. All pigs were weighed weekly, as was the amount of food consumed in each pen. The conclusion reached was that growing pigs were able to differentiate successfully between two foods on the basis of their protein content and to change the selected diet to match their changing requirement for dietary protein. As a result, there was a significant reduction in food intake (P < 0·001) and in the P2 backfat thickness (P < 0·01) and the food conversion ratio was significantly improved (P < 0·001) compared with the control treatment.