Pigs living in commercial husbandry systems may experience both acute stress due to standard management procedures and chronic stress through limitations in their barren housing environment. This might influence their immune status, including antibody responses to neural and danger autoantigens. Levels of natural autoantibody (NAAb)-binding phosphorylcholine-conjugated bovine serum albumin (PC-BSA) and myelin basic protein (MBP) were measured over time in pigs that were kept in environmental enriched v. barren housing, and that underwent a regrouping test. In total, 480 pigs were housed in 80 pens in either barren or straw-enriched pens from 4 through 23 weeks of age. Blood samples were taken from pigs before (week 8), and 3 days after a 24 h regrouping test (week 9), and at 22 weeks of age. Phosphorylcholine-conjugated bovine serum albumin (PC-BSA) and MBP antibody titres in serum were measured using ELISA. Enriched-housed pigs had higher levels of IgM-binding MBP, and tended to have higher levels of IgG-binding MBP and IgA-binding PC-BSA than barren-housed pigs. Each NAAb measured in this study was affected by gender and litter. These results suggest that enriched housing conditions, as well as acute regrouping stress, have an influence on levels of serum NAAb-binding danger and neural antigens in pigs.

Single concentrate feeds are mixed together forming compound feeds for cattle. However, knowledge regarding the potential interactions (associative effects) between the feeding values of single feeds in compound feeds is lacking. The main objective of the present study was to evaluate ruminal fermentation characteristics and feeding values of eight industrially produced compound feeds in mash form from their constituent single feeds for dairy cows through in vitro assays. Additivity was given for gas production (GP), digestibility of organic matter (dOM) and utilisable CP at the duodenum (uCP). Additivity of CP fractions (determined using the Cornell Net Carbohydrate and Protein System (CNCPS)) was dependent on the fraction and compound feed type; however, the effective degradation calculated from CP fractions (EDCNCPS) showed additivity. Additivity was not given for intestinal digestibility of rumen-undegraded protein (IDRUP) for five out of eight compound feeds. Precise calculation of metabolisable energy (ME) of compound feeds from ME of single feeds was possible when using the same ME equations for all single and compound feeds. Compound feeds are often provided in pellet form; therefore, our second objective was to evaluate the effects of pelleting on ruminal fermentation characteristics and feeding values of compound feeds. Pelleting affected GP at 24 h (GP24; up to 2.4 ml/200 mg DM), dOM (up to 2.3 percentage point (pp)) and ME (up to 0.3 MJ/kg DM), but these differences were overall small. More considerable effects of pelleting were observed for uCP, which was increased in all compound feeds except the two with the highest CP concentrations. The IDRUP was lower in most compound feeds following pelleting (up to 15 pp). Pelleting also affected CP fractions in a non-systematic way. Overall, the effects of pelleting were not considerable, which could be because pelleting conditions were mild. Our third objective was to compare in situ ruminal CP degradation (EDIN_SITU) of compound feeds with ED using two prediction methods based on CP fractions. EDIN_SITU reference data were obtained from a companion study using the same feeds. Prediction accuracy of EDIN_SITU and EDCNCPS was variable and depended on the compound feed and prediction method. However, future studies are needed as to date not enough data are published to draw overall conclusions for the prediction of EDIN_SITU from CP fractions.

Chemical and physical compositions were respectively determined on right and left sides of 23 Scottish Blackface ewe carcasses. Relationships were derived whereby physical carcass composition, in terms of muscular tissue plus associated fatty tissue, subcutaneous plus perirenal fatty tissues, and bone, could be estimated from weights of chemically determined water, fat, ash, and fat-free dry matter. The use of these relationships is discussed, and it is suggested that in certain circumstances the estimation of physical carcass composition from chemically determined components could result in considerable economies of time and labour.

The use of certain readily determined physical components as indices of the chemical composition of the whole body is considered, and relevant prediction equations presented.

An investigation was undertaken to evaluate the crossbred boar per se, and also to assess the value of including the American Hampshire into a first-cross sire. Groups of three boars were placed on each of 20 farms. Each group contained a purebred white (Landrace or Large White), a crossbred white and a crossbred Hampshire boar. Co-operators were asked to obtain at least six litters by each boar, record their performance to weaning, and provide a sample of the progeny of each boar to be reared under standard conditions to slaughter at 90 kg live weight. Breeding of the boars did not affect either the level or variability of litter performance. Differences in performance and carcass traits between the progeny of purebred and whitecross boars were found only for ‘eye-muscle’ area and fat depth ‘C’. Pigs sired by Hampshire-cross boars had carcasses which were 2% shorter than those from white boars.

The effect of recombinant bovine FSH (rbFSH) administration on serum FSH and follicular growth was determined in 16 GnRH-immunized heifers (primary day 0 and booster day 28 using 0·4 mg human serum albumin (HSA) – Cys-Gly-GnRH in DEAE-dextran). All heifers produced GnRH antibody titres (mean 58·8 (s.e. 2·8)% binding at 1: 640 serum dilution on day 42), follicular growth ceased (follicles ≤ 4 mm) between days 33 and 49, progesterone concentrations were ≤ 0·5 ng/ml in all heifers by day 53, and pulsatile LH secretion was markedly reduced (v. luteal phase control heifers) on day 63. Heifers were blocked by antibody titre and LH concentrations on day 63 and assigned (no. = five or six per treatment) on day 78 to: (1) control (saline given i.m. four times per day for 4 days), (2) 0·5 mg equivalent (USDA bFSH BP 1) of rbFSH (Granada Biosciences, TX) given i.m. four times per day for 6 days (12 mg in total) or 3) 1·5 mg rbFSH given i.m. four times per day for 4 days (24 mg in total). Ovaries of heifers were examined by ultrasound and blood samples were collected at 3- to 6-h intervals from days –1 to 10 and then twice a day until follicles ≥ 5 mm were no longer detectable. The increase in serum FSH concentrations was different (P < 0·05) between treatments (mean peak FSH above pre-treatment baseline concentrations at 102 h after first rbFSH were 1·4 (s.e. 0·8), 14·5 (s.e. 3·3) and 33·6 (s.e. 5·3) ng/ml for treatments 1 to 3, respectively). The rate of decline in FSH from 102 to 302 h was not different between treatments 2 and 3. There was no detectable follicular growth in heifers given saline injections. Following either rbFSH treatment, medium (MF; 5 to 9 mm) and large (LF; ≥ 10 mm) follicles grew. For heifers in treatments 2 and 3, there was no difference in the interval from start of rbFSH until the first appearance of MF (4·2 (s.e. 0·7) v. 3·3 (s.e. 0·3) days) or LF (5·3 (s.e. 0·3) v. 4·8 (s.e. 0·8) days) or in the duration of the presence of MF and/or LF (13·2 (s.e. 4·5) v. 16·6 (s.e. 2·6) days). Increasing dose of rbFSH increased (P < 0·05) the mean maximum number of MF (4·3 (s.e. 1·1) and 9·2 (s.e.0·9) for treatments 2 and 3) and LF (2·0 (s.e. 1·1) and 8·4 (s.e. 1·2)). In summary, low doses of rbFSH induced a small cohort of follicles to grow and develop without selection or development of a single dominant follicle, whereas high doses of rbFSH resulted in large numbers of both MF and LF.

The use of herbal plants as traditional medicines has a century long history. Plantain (Plantago lanceolata L.) is a perennial herb containing bioactive components with free radical scavenging activities. An isotope dilution technique using [U-13C]glucose was conducted to determine the effect of plantain on the responses of plasma glucose metabolism to exogenous insulin infusion in sheep. Six crossbred sheep (three wethers and three ewes; mean initial BW=40±2 kg) were fed either a mixed hay of orchardgrass (Dactylis glomerata) and reed canarygrass (Phalaris arundinacea) (MH-diet) or mixed hay and fresh plantain (1 : 1 ratio, dry matter basis, PL-diet) and exposed to a thermoneutral (TN, 20°C; 70% relative humidity (RH)) environment or a heat exposure (HE, 30°C; 70% RH) for 5 days using a crossover design for two 23-day periods. The isotope dilution was conducted on days 18 and 23 of the experimental period during TN and HE, respectively. Plasma concentration of α-tocopherol was greater (P<0.0001) for the PL-diet than the MH-diet and remained comparable between environmental treatments. Plasma glucose concentration before isotope dilution technique was reduced for sheep (P=0.05) during HE compared with TN and remained comparable between diets. Plasma glucose turnover rate during the preinfusion period of insulin did not differ (P=0.10) between dietary treatments and between environments (P=0.65). The response of plasma glucose utilization to exogenous insulin administration was lower (P=0.04) for the PL-diet than the MH-diet. Under present experimental conditions, the plantain group was found to be resistant to the effects of insulin infusion.

An experiment was conducted to study the effect of feeding level and semen collection frequency on semen quantity and quality of 62 Yorkshire boars at two artificial insemination centres during 10 successive periods of 2 weeks. The experiment had a 2 × 2 factorial design with two feeding levels (2·75 and 3·5 kg/day) and two semen collection frequencies (three and five times in 2 weeks).

In the first 2-week period boars on the high frequency semen collection scheme produced 94 × 109 sperm cells more than boars on the low scheme. In periods 2 to 10 the mean difference was only 18 × 109 sperm cells. Boars on the high feeding level produced 32 × 109 sperm cells more per 2-week period than the boars on the low feeding level (P < 0·05). Boars at the high collection frequency and the high feeding level tended to produce a higher number of sperm cells during periods 2 to 10 than the boars on the high semen collection frequency and the low feeding level (P > 0·05). No effects of food intake or semen collection frequency on semen quality were found.

The effect of three moisture concentrations (10, 50 and 75%) and of four fermentation conditions (none, 24 hr aerobic, and 24 and 48 hr anaerobic) on the nutritive value of two complete fattening diets containing 25 and 45% poor roughage, was examined in three digestion and nitrogen balance trials, each of Latin square design, in cattle.

Fermentation increased the lactic acid and volatile fatty acid concentrations of the feed, decreased its pH and the crude fibre content, and caused losses of dry matter.

There were non-significant differences in voluntary intake between the dry and soaked diets. The apparent digestibility of dry and organic matter, crude protein, nitrogen-free extract and crude fibre was non-significantly improved by added moisture. There were no consistent differences in nitrogen retention.

The average metabolizable energy values (ME) observed for both diets were 6% higher than those calculated. There were no significant differences in the apparent digestibility of gross energy or in ME value between treatments of different moisture content.

The fluctuations in pH of rumen liquor during the day differed markedly when the animals received different levels of roughage. These differences were greater when the feed was allowed to ferment. Soaking of feed resulted in a significant increase in the concentration of propionic acid in the rumen liquor, and a non-significant decrease in that of acetic acid. Consequently the C2/C3 ratio was significantly lower on the moist feed treatments.

There were no consistent differences in concentration of ammonia in the rumen liquor, or of urea in the blood.

Field data of the Austrian Simmental population were analysed using restricted maximum likelihood (REML) with an animal model where additive direct, additive maternal and cytoplasmic effects were treated as random and the effect of the year of first calving as fixed. Traits analysed were milk yield, fat and protein content, persistency, days open and herd life. All dairy traits were pre-adjusted for best linear unbiased prediction (BLUP) herd-year effects, milk yield additionally for season, age at first calving and days open. After applying specific data restrictions, the number of records for the various traits ranged from 3360 to 51889. Identification of cow lineages was based on pedigree information from the official milk recording scheme, with a span of at least four and up to 16 generations. The number of lineages per trait varied and ranged from 484 to 3195, with an average size of 15 members (for herd life 7). Evaluations of the relevant variance components for the dairy and fitness-related traits investigated were separate for the first three lactations.

The estimated variance components for cytoplasmic effects were close to zero for all dairy traits with the exception of first lactation milk yield, where a significant value of 2.0% of the total phenotypic variance was found. Significant contributions of cytoplasmic lineages to total variance in all lactations, however, were estimated for persistency (2·6 to 3.8%), days open (1·8 to 2.9%) and for both true and functional herd life (4.6% each). The portions of additive maternal variance and covariance between additive direct and additive maternal effects on total variance were very close to zero for all traits investigated. The maximum differences between BLUP lineage effects were 373 kg for first lactation milk yield, 44 days for days open (first lactation), 1·6 and 2·8 years for true and functional herd life and, on average, 1·0 kg for standard deviation of test day milk yields (persistency) of the first three lactations.

Removing the cytoplasmic effect from the model led to increased estimates of the additive direct heritability. Further model aspects such as interaction between additive and cytoplasmic gene effects and possible confounding between cytoplasmic and herd effects are discussed.

A series of laboratory and animal studies examined the use of chemical and biological agents to enhance the digestibility of Rhodes grass (grass) cut at 60 (young) and 100 (mature) days of regrowth and ensiled as big round bales. The treatments included an untreated control (C), a microbial inoculant (I), NaOH, CaO and NaOH plus inoculant (NaOH + I). Inoculant was grown anaerobically, using a starter culture of rumen fluid from cattle given Rhodes grass. Treatments C, I, NaOH, NaOH + I, were offered separately to twelve dairy heifers, in a 3 ✕ 4 randomized complete block design, repeated twice for each grass silage. C and I had substantial mould growth, compared with no visible mould in NaOH or NaOH + I. CaO treatment was effective in preventing mould growth, but had little effect on the chemical composition and in sacco digestibility of mature grass silage. NaOH reduced NDF content and increased in sacco digestibility (P < 0·05) but not the in vivo digestibility (P > 0·05) of both mature- and young-grass silage. The effects of other treatments on nutritive value were non-significant at both stages of maturity. NaOH increased the intake of mature-grass silage by 24-26% (P < 0·05), but had little effect on the intake of young-grass silage (P > 0·05). Treatment I consistently reduced grass silage intake (P < 0·05) for young-grass silage. The findings of these studies show that treating mature Rhodes grass with NaOH will improve its nutritive value and reduce mould growth in conserved herbage. However none of the treatments in this study had any consistently positive effects on the in vivo nutritive value or storage quality of young-grass silage.

Plasma progesterone concentrations were measured over two collection periods from two groups of mature non-pregnant cows involved in a twin-breeding experiment. One group (T) consisted of cows obtained from private herds after having produced at least two sets of twins each; the other was a control group of similar ages and breeds. Yearling heifers from the same breeding experiment were also included. The mean progesterone concentration from cows in the T group was proportionally 0·83 of the value in control cows and the cow group effects were consistent in direction for samples obtained after single or double ovulations. Results were also consistent in early oestrous cycle (days 3 to 7, oestrus = day 0), mid cycle (days 8 to 13) but not late cycle (days 14 to 18). Concentrations were significantly higher in yearling heifers than in control cows. There was a trend for higher progesterone concentrations after double than after single ovulations assessed by rectal palpation, but differences were not significant.

The sensitivity to dietary lysine: energy content of pigs from lines divergently selected over seven generations for components of efficient lean growth rate was quantified by the within-selection line regression of performance test traits on diet class. Large White pigs were performance tested over three 14-day test-periods starting at 30, 50 and 75 kg and offered, to appetite, isoenergetic diets differing in dietary total lysine: energy (0•59, 0•69, 0•81, 0•91, 1•01, 1•12 and 1•23 g/MJ digestible energy (DE)). Within each litter, full-sibs were performance tested on different diets. Pigs were also performance tested on a diet-choice procedure using diets with total lysine: energy of 0•69 and 1•12 g/MJ DE to determine the correlation between performance on phase-feeding and diet-choice.

The study consisted of 578 animals with 459 pigs tested with phase-feeding and 119 tested on diet-choice procedures. The study detected significant selection line and dietary effects on performance test traits, but no significant between-selection line differences in sensitivity to dietary total lysine: energy. When combinations of performance test traits were transformed into predicted lysine and energy utilization traits there were significant selection line effects on predicted nutrient allocation, but not in responses to increasing dietary total lysine: energy. The lack of between-selection line differences in sensitivity to dietary total lysine: energy indicated that the consequences of changing dietary total lysine: energy will be broadly similar across a range of genotypes, as spanned by the selection lines of the study.

Genetic correlations between performance traits of phase-fed and diet-choice pigs suggested that predictions of genetic merit will be similar with animals tested on either procedure between 30 and 75 kg but post 75 kg predicted genetic merit for growth rate and lysine intake should be estimated separately for performance on diet-choice or for performance on a given diet. In contrast, predicted genetic merit for fat deposition with performance testing on diet-choice will be highly correlated with predicted genetic merit with testing on a single diet.

The results of earlier work by Breirem have been reanalysed, and some recent results are presented. Breirem's results indicated that for pigs of 16 to 96 kg live weight and fasted for 4 to 9 days, heat production (HP) was related to live weight (LW) according to:

HP = 229 LW0·512

which could be converted to:

HP = 185 LW0·569

This equation gives values for fasting heat production which are about 20% higher than those usually estimated from Breirem's results on pigs of 16 to 196 kg. The recent results, for pigs fasted for 24 hr, were slightly higher than would be indicated by the latter equation. If the values derived from the reanalysis of Breirem's data are related to several estimates of the energy required for maintenance by pigs, values for km% between 87 to 103% are obtained.

A Piétrain boar was placed on each of six farms where the terminal sire was a Large White and the breeding females were either Large Whites or Large White × Landrace. Cooperators were requested to obtain 10 litters by each boar and record their performance to slaughter at 64 kg live weight. Farrowing rate to first service was relatively higher for sows mated with Large White boars but not significantly so. Pietrain-cross piglets were heavier at weaning (7·5%) but breed of boar did not influence other litter traits or age of progeny at slaughter. Pietrain crosses had higher carcass yields (2·4%), shorter carcasses (2·5%) and larger eye muscles in cross-section (10·6%) and in commercial evaluations relatively more were placed in the highest grade.

In a 15-year period (1956-1970) data were available from 192 Brahman cows exposed to bulls on 1409 occasions (1199 calvings) and from 155 Santa Gertrudis cows exposed 907 times (847 calvings). All were mated by natural service in the same herd in Cuba. Calving percentage and net calf crop significantly favoured the Santa Gertrudis with overall means of 92·7 and 85·3% for the two traits respectively. Parity, lactation status and year of calving significantly affected calving percentage with significant year × breed and status × parity interactions. Lactation status, year of calving, breed × year and status × parity interactions also affected net calf crop. Calf mortality was 6·41% in Brahmans and 5·64% in Santa Gertrudis but more of the latter breed were stillborn (41·7 v. 16·9% of all deaths). There were no breed effects on age at first or last calving, age at culling or calves per cow which were 38·5 months, 117·1 months, 130·1 months and 6·62 calves in Brahmans and 39·3, 111·5, 123·2 and 6·58 respectively in Santa Gertrudis. Twinning percentage was low (under 0·4%) and male births high (53·54%) in both breeds. Some 30% of cows lost at least one calf prior to weaning during their lifetime.

The contribution of the leucine/valine substitution at amino acid position 127 in the bovine growth hormone (GH) protein to variation in carcass traits was studied. The data included 109 Polish Friesian bulls slaughtered at 15 months of age. The traits measured were carcass gain, weights of meat, bones, intermuscular and subcutaneous fat in the carcass and meat, bones and fat in valuable cuts (fore and best ribs, sirloin, round of beef and shoulder). The bulls’ GH genotype was determined using the PCR-RFLP technique. The frequencies of leucine (Leu) and valine (Val) alleles were 0·64 and 0·36, respectively. The GH concentration was determined in serial blood plasma samples collected every 15 min starting from 15 min before to 135 min after intravenous administration of 0·15 µg thyrotropin releasing hormone (TRH) per kg live weight. Response GH variables were: baseline (the mean of samples collected at –15 and 0 min), peak (the sample taken at 15 min post injection of TRH) and disappearance rate (calculated as peak minus the sample at 60 min, divided by time interval 45 min). Mixed animal models were used for the statistical analysis. Differences were found between the Leu/Leu and the Val/Val genotypes for carcass gain and weight of meat in the carcass (P ≤ 0·05). Moreover, differences in the size of the GH peak between the two homozygotes approached significance (P ≤ 0·10). The effect of GH genotype accounted for a moderate part of the phenotypic variance in the carcass traits, corresponding to a reduction in the residual variance of ≤ 5·25% when included in the model, whereas the corresponding value for the effect of GH genotype on the variation in GH release was lower, ≤ 1·77%. In conclusion, the Leu/Val polymorphism seems to be associated with carcass traits in dairy bulls, although the effect was relatively small when compared with the effects of season and background genome.

The objective was to assess the effects of inclusion rate and chop length of lucerne silage, when fed in a total mixed ration (TMR), on milk yield, dry matter (DM) intake (DMI) and digestion in dairy cows. Diets were formulated to contain a 50 : 50 ratio of forage : concentrate (DM basis) and to be isonitrogenous (170 g/kg CP). The forage portion of the offered diets was comprised of maize and lucerne silage in proportions (DM basis) of either 25 : 75 (high Lucerne (HL)) or 75 : 25 (low lucerne (LL)). Lucerne was harvested and conserved as silage at either a long (L) or short (S) chop length. These variables were combined in a 2×2 factorial arrangement to give four treatments (HLL, HLS, LLL, LLS) which were fed in a Latin square design study to Holstein dairy cows in two separate experiments. In total, 16 and 8 multiparous, mid-lactation cows were used in experiments 1 and 2, respectively. To ensure sufficient silage for both experiments, different cuts of lucerne silage (taken from the same sward) were used for each experiment: first cut for experiment 1 (which was of poorer quality) and second cut for experiment 2. Dry matter intake, milk yield and milk composition where measured in both experiments, and total tract digestibility and nitrogen (N) balance were assessed using four cows in experiment 2. In experiment 1, cows fed LL had increased DMI (+3.2 kg/day), compared with those fed HL. In contrast, there was no difference in DMI due to lucerne silage inclusion rate in experiment 2. A reduction in milk yield was observed with the HL treatment in both experiment 1 and 2 (−3.0 and −2.9 kg/day, respectively). The HL diet had reduced digestibility of DM and organic matter (OM) (−3% and −4%, respectively), and also reduced the efficiency of intake N conversion into milk N (−4%). The S chop length increased total tract digestibility of DM and OM (both +4%), regardless of inclusion rate. Inclusion of lucerne silage at 25% of forage DM increased milk yield relative to 75% inclusion, but a S chop length partially mitigated adverse effects of HL on DMI and milk yield in experiment 1 and on DM digestibility in experiment 2.

Low and variable egg quality remains a major issue in aquaculture impeding a reliable and continuous supply of larvae, particularly in emerging species, such as pikeperch, Sander lucioperca. We assessed the influence of batch-specific egg parameters (fatty acid (FA) profiles, cortisol content) on embryo life-stages until hatching (survival at 2, 24, 48, 72 h post fertilization (hpf), hatching rate) in an integrated study under commercial hatchery conditions (44 egg batches). Embryo mortality was elevated until 48 hpf (average 9.8% mortality between 2 and 48 hpf). Embryos surviving until 48 hpf were very likely (98.5%) to hatch successfully. The inherent egg FA composition was variable in-between batches. Total FA content ranged form 66.1 to 171.7 µg/mg (dry matter) total FA. Whereas specific FA ,18 : 0 and 20 : 5(n-3) (eicosapentaenoic acid) of the polar fraction and the ratio of 22 : 6(n-3) (docosahexaenoic acid) to 20 : 5(n-3) within the neutral fraction, were significantly correlated with early embryo development, contents of the respective FA did not differ between high (>90% hatching rate), mid (70% to 90% hatching rate) and low (<70% hatching rate) quality egg batches. Late embryo development and hatching were relatively independent of the FA profiles highlighting stage-dependent influences especially during early embryogenesis. Cortisol levels ranged from 22.7 to 293.2 ng/ml and did not directly explain for mortalities. However, high cortisol was associated with a lower content of specific FA, in particular highly unsaturated FA. These results demonstrate the magnitude of inter-individual differences in the batch-specific biochemical egg composition under stable hatchery conditions and suggest a stress-mediated lack of essential FA, which in turn affects early embryo survival. Surprisingly, embryos are able to cope well with a broad range of inherent egg parameters, which limits their predictive potential for egg quality in general. Still, specific FA profiles of high quality egg batches have potential for formulating species-specific broodstock diets and improving reproductive management in pikeperch.

The lactation model of Wood, y = Anbe−cn was fitted, using a logarithmic transformation, to two sets of lactation data obtained from the Agricultural Research Institute of Northern Ireland. The first set was from animals calving in the autumn and the second set from animals calving in winter and early spring. A categorical variable was added to the transformed model to investigate the effect on the parameters of the animals going out to grass. The technique proved more useful with the autumn-calving herd.

One hundred and sixty-four Scottish Blackface ewes from two different farm sources were brought into a uniform moderately-poor level of body condition by early November. Five groups of approximately 33 ewes were then fed different amounts of pelleted dried grass and hay on pasture to achieve either high (H) or live-weight maintenance (M) levels of intake for different durations and at different times in the 36 days prior to mating at a synchronized oestrus. One group of ewes was fed the M level throughout, two groups were fed the H level for 18 days either preceded or followed by 18 days of the M level and two groups were fed the H level for 27 days either preceded or followed by 9 days of the M level. After mating, all ewes were fed the M level until slaughtered on return to service or at 28 (± 8) days after mating for counts of corpora lutea and viable embryos relating to first mating.

The gain in body condition and live weight after 18 days of H feeding was about half that achieved after 27 days. Time of H feeding had no differential effect on gain in condition or live weight. H feeding increased ovulation rate only if fed immediately prior to mating. When a period of M feeding preceded mating, most of the advantage was lost, particularly with 18 days of M feeding after 18 days of H feeding. Embryo mortality measured as ova loss was not significantly affected by either the duration or time of H feeding, although there was a trend for loss to decrease with increasing duration. Potential lambing rate to first mating was therefore significantly increased by H feeding for 27 days even when followed by 9 days of M feeding prior to mating but was not significantly increased by H feeding for only 18 days even when fed immediately prior to mating.

Differences in response were obtained from the ewes from the two sources. H feeding only produced a significant improvement in reproductive performance when potential was high and this was partly related to greater size. In populations with low potential, the provision of H feeding in the pre-mating period is therefore of doubtful economic advantage.