Six alternative lactation curves were derived by considering the proportional rates of change in milk yield during lactation. Experimental data showed that rates of change do not necessarily fall to a constant value in mid and late lactation. This accounts for the systematic lack of fit reported by several authors using the derivative of the incomplete gamma function, whose second derivative has a horizontal asymptote. Curves whose derivatives are not restricted in this way described patterns of milk yield in first to third lactation animals more closely. Some statistical properties of their parameters have been examined. The curve loge (y) = a - bt'(1 + kt') + ct'2 + d/t, where t is days since calving, t' = t – 150, A: is a constnat and a, b, c and d are parameters to be estimated, is recommended. This curve was also fitted to yields of milk fat, protein and lactose.

Repeated cloning of bovine embryos by nuclear transfer, producing large clones of monozygous animals, may be possible in the future. Initially, clones could be tested and the best one selected and spread over the commercial population by embryo transfer. Further genetic improvement could be obtained by rebreeding a number of the best clones to produce a new set of clones. However, the testing and selection systems to pick the best clone (for short-term clonal response) and to pick clones with the best breeding values (for long-term genetic response) are different. The objective of this study was to derive a system which achieves both high clonal and high genetic responses. An adult MOET (multiple ovulation and embryo transfer) scheme with 40 breeding males and 40 breeding females per generation (cycle) was used to maintain adequate genetic variation for continued genetic response. For a fixed set of testing facilities and a given family structure initial clonal response is maximized by testing several members per clone. Long-term genetic response is usually greatest when testing one member per clone. Compromises to obtain both high clonal response and high genetic responses were from 95 to 100% efficient.

The generation of profit in dairy production can be approximated by a generalized profit equation, which is a function of the genotype of the animals used. In the absence of legislated quotas on production, the economic weights for traits contributing to profit, for use in a selection index, have been shown to be simple functions of the partial derivatives of profit with respect to output of the traits. These functions reflect the fact that output in most agricultural industries will already be maximized, either because of saturated markets or limitations on total inputs. When a single quota applies, different functions result, which reflect the downward rescaling of enterprise size as output per animal of a trait under quota is increased. Difficulties arise when multiple non-independent quotas apply, such as in the United Kingdom (UK) milk market where quotas are triggered by both milk volume and fat concentration. The functions describing the economic weights are then dependent on the form of the dependency between the quota criteria and on the genetic change resulting from the applied selection index. Economic weights for milk volume, fat, protein and lactose yield applicable to Holstein'Friesian cattle in the UK were found to be –1·6 p/1, 76·6 p/kg, 170·0 p/kg and 7·0 p/kg, scaled to 1986 prices. These weights would not change much if the quota were changed to fat yield only. Use of appropriate selection indexes should result in genetic increases of milk volume, fat, protein and lactose yield, with gradual increases in fat and protein concentrations and the fat to protein ratio. In most situations, selecting on the combined evaluation for fat plus protein yield would be a simple procedure with high efficiency (0·995 of maximum efficiency).

Four groups of six cows were fed from 12 weeks before calving to achieve condition scores at calving of 3·23 (F) and 1·98 (T). For the first 20 weeks of lactation all cows were given 10 kg/day of isonitrogenous (180 g crude protein per kg dry matter (DM)) compounds containing either a high (13·0 MJ/kg DM; HE) or a low (9·8 MJ/kg DM; LE) concentration of metabolizable energy, together with 3 kg/day molassed sugar-beet pulp and hay offered ad libitum.

The mean milk yield of cows receiving compound HE (27·7 kg/day) was higher (P > 0·05) than that of cows receiving compound LE (25·6 kg/day). Cows in group FHE yielded 27·0 kg/day, compared with 25·5 kg/day for FLE (P > 0·05); cows in group THE yielded 28·4 kg/day compared with 25·17 kg/day for TLE (P <0·05). After peak lactation, milk yields were maintained better in group THE but declined at a faster rate in group TLE than in groups FHE and FLE. Neither dietary energy concentration nor condition score at calving significantly affected milk composition (butterfat 43·5, protein 29·1, lactose 50·2 g/kg). Cows receiving compound HE consumed significantly (P < 0·001) more energy (208 MJ/day) than cows receiving compound LE (188 MJ/day). Over the first 10 weeks of lactation, changes in condition were –0·83, –0·88 +0·08 and –0·25 (s.e.d. 0·22, P < 0·05) condition score units for cows in groups FHE, FLE, THE and TLE respectively.

It is concluded that with diets of high energy concentration intake is mainly limited by physiological mechanisms so that thin cows eat more than fat cows and produce similar amounts of milk. With diets of lower energy concentration, intake is limited by rumen capacity and thin cows eat the same as fat cows. This results in increased fat mobilization and a slight decrease in milk yield by cows which are fat at calving but the limited fat reserves of cows which are thin at calving are insufficient to compensate for reduced energy intake so large reductions in milk yield are observed.

A deterministic model was developed to examine the optimization of open nucleus breeding schemes in order to maximize the rate of genetic response in dairy cattle. By changing the parameters, the model was able to simulate both a closed nucleus and a progeny testing scheme. The model implicitly optimized the generation interval and the selection across tiers by means of truncation across age classes and tiers respectively. The effects of size of the progeny test group and the nucleus size were assessed by comparing alternative plans. It is possible to optimize a breeding plan given the reproduction rates of the animals, the availability of different sources of information, the age distribution of the animals (survival rates) and the phenotypic and genetic parameters of the trait.

The steady state selection response was assessed by calculating the genetic progress year after year until it stabilized. The genetic gain was corrected for the effects of reduced variances due to previous selections and increased variances due to genetic differences between parental age classes.

In an example, the model was used to predict the improvement in milk yield in a closed artificial insemination breeding scheme. The genetic gain of a conventional progeny testing scheme was about one-third lower than the genetic gain of the optimized breeding plan. The variance reduction due to selection decreased the steady state genetic gain by a factor 0·3

Two alternative ways of estimating cumulative selection differentials are compared. An alternative method is developed that includes the advantages of both methods. This alternative method is shown to give rise to less biased estimates of realized heritability

This paper evaluates the consequences of using the wrong genetic model when predicting the merit of previously untested crossbred genotypes. Eight models are considered, seven including different biological interpretations of two-locus epistatic interaction, plus one excluding epistatic effects. Published results from 13 genotypes generated from Hereford and Angus parental breeds were analysed, and predictions of a further seven genotypes made using different models. Under a dominance model, the predicted superiority in pregnancy rate (%) of a ¾ Hereford: ¼ Angus composite over a two-breed rotation was +1·2%, yet under all other models this was a negative value ranging from –1·9% to –3·7%. However, few such cases were found in which significant decision errors could conceivably be made. It is concluded that decisions on the choice of crossbred genotypes are generally quite robust to differences in the genetic model of the type studied here

The reduction in additive genetic variance due to selection is investigated when index selection using family records is practised. A population of infinite size with no accumulation of inbreeding, an infinitesimal model and discrete generations are assumed. After several generations of selection, the additive genetic variance and the rate of response to selection reach an asymptote. A prediction of the asymptotic rate of response is considered to be more appropriate for comparing response from alternative breeding programmes and for comparing predicted and realized response than the response following the first generation of selection that is classically used. Algorithms to calculate asymptotic response rate are presented for selection based on indices which include some or all of the records of an individual, its full- and half-sibs and its parental estimated breeding values. An index using all this information is used to predict response when selection is based on breeding values estimated by using a Best Linear Unbiased Prediction (BLUP) animal model, and predictions agree well with simulation results. The predictions are extended to multiple trait selection.

Asymptotic responses are compared with one-generation responses for a variety of alternative breeding schemes differing in population structure, selection intensity and heritability of the trait. Asymptotic responses can be up to one-quarter less than one-generation responses, the difference increasing with selection intensity and accuracy of the index. Between family variance is reduced considerably by selection, perhaps to less than half its original value, so selection indices which do not account for this tend to place too much emphasis on family information. Asymptotic rates of response to selection, using indices including family information for traits not measurable on the individuals available for selection, such as sex limited or post-slaughter traits, are found to be as much as two-fifths less than their expected one-generation responses. Despite this, the ranking of the breeding schemes is not greatly altered when compared by one-generation rather than asymptotic responses, so the one-generation prediction is usually likely to be adequate for determining optimum breeding structure.

An enzyme-linked lectin sorbent assay (ELLSA) has been used to determine quantitatively the amount of soya-bean agglutinin (SBA) binding sites in and the sialic acid contents of small intestinal brush border membranes (BBM) of a group of 14 clinically healthy veal calves. Computer analysis of the results indicated 704 (s.e. 27) nmol SBA binding sites per mg of BBM protein with a Kj of 2·63 (s.e. 0·15) × 10-5mol/l; the sialic acid contents were 2·67 (s.e. 0·43) mg/g BBM protein. We could not demonstrate a correlation between the amount of binding sites and the sialic acid contents of the BBM.

Incubation with neuraminidase did not alter the number of SBA binding sites.

An experiment was carried out to compare silage-based and dried forage-based diets for finishing beef cattle and to examine the effects of protein supplementation and formaldehyde treatment of silage on performance and carcass composition. The four diets used consisted, on a dry matter basis, of (1) 700 g/kg formic acid-treated grass silage and 300 g/kg low protein cereal-based concentrate (LPC), (2) 720 g/kg formic acid-treated silage and 280 g/kg high protein cereal/fish meal concentrate, (3) 600 g/kg formaldehyde-treated silage and 400 g/kg LPC, and (4) 480 g/kg artificially dried grass and grass hay and 520 g/kg LPC. The silages were well preserved with 33 to 40 g ammonia N per kg total nitrogen. The diets were offered to 40 British Friesian castrated male cattle which were initially 387 (s.e. 2·9) kg live weight. For diets 1 to 4 respectively metabolizable energy intakes were 91·8, 94·7, 94·9 and 91·8 MJ/day; total nitrogen intakes 192, 234, 189 and 191 g/day; live-weight gains 0/89, 0/93, 0/86 and 0/82 (s.e. 0/045) kg/day; carcass gains 0·51, 0·52, 0·52, 0·52 and 0·50 (s.e. 0·33) kg/day; mean subcutaneous fat depths 5·6, 5·4, 5·3 and 6·0 (s.e. 0·41) mm; marbling scores 3·0, 3·3, 3·2 and 2·6 (s.e. 0·19) and areas of m. longissimus dorsi 65·7, 66·4, 64·7 and 64·3 (s.e. 1·66) cm2. It is concluded that performance, efficiency of energy utilization and carcass fatness were similar for a diet containing 700 g/kg well preserved grass silage and a dried forage-based diet with a similar intake of energy and protein. Protein supplementation of a diet based on formic acid-treated silage or formaldehyde treatment of silage did not affect performance or carcass fatness.

The phenotypie associations between scores for various types of non-fleece pigmentation and concentration isolated pigmented wool fibres in the fleece (experiment 1 and 2) or the piebald phenotype (experiment 3) were studied in Corriedale hogget ewes. These sheep were selected on the basis of either degree of nose skin pigmentation (experiment 1), presence or absence of pigmented fibres at the horn sites (experiment 2) and presence or absence of the piebald phenotype (experiment 3). Among affected fleeces the concentration of isolated pigmented fibres (PFC) had high correlations (0·96 and 0·97) with the proportion of staples affected, which indicates that when PFC is high the isolated pigmented fibres are widely distributed throughout the fleece. Three types of non-fleece pigmentation had significant (P < 0·05) positive correlations with PFC in experiments 1 and 2. These were pigmented skin on the nose lips (0·23 and 0·18; P < 0·05), pigmented fibres at the horn sites (0·34 and 0·30; P < 0·01) and pigmented fibres on the legs (0·36 and 0·29; P < 001). The pigmentation of skin or fibres in non-fleece areas was predominantly black/grey and there were usually significant (P < 0·05) positive correlations between the various types of non-fleece pigmentation. In experiment 3, piebald sheep were found to have higher scores for black/grey fibres on the ears (P < 0·001), face (P < 0·01) and horn sites (P < 0·05) and for black/grey skin under the tail (P < 0·01) than non-piebald sheep.

In six experiments carried out on individually penned Finn Dorset ewes estimates were made of the quantities of non-ammonia nitrogen (NAN) reaching the abomasum and truly digested in the small intestine. For experiments 1 and 2 which each involved 33 ewes given a complete diet of milled hay and concentrates supplemented with varying amounts of a good-quality fish meal during late pregnancy, the overall regression estimate for the daily amount of truly digested NAN required for zero N retention (ewe plus conceptus) was 438 mg/kg M0·75 (441 and 434 for experiments 1 and 2 respectively). The coefficients for the efficiency of utilization of increments of truly digested NAN and maternal tissue N for net N accretion in the conceptus were 0·48 (s.e. 0·039) and 0·84 (s.e. 0·127) respectively.

In experiment 3, 24 lactating ewes had their diet supplemented with either soya-bean meal or fish meal. There was no effect of protein source on the amounts of NAN reaching the abomasum in ewes fitted with an abomasal cannula or on milk yield and these observations confirmed the unusually high degradability (measured by the polyester bag technique) of the fish-meal protein in the rumen. For experiments 4, 5 and 6 a total of 36 ewes were used to test the efficiency with which the NAN truly digested in the small intestine was used for the synthesis of milk protein. In the absence of data on the possible contribution of body tissue N to milk N the coefficient for the apparent efficiency of utilization of truly digested NAN when the ewes were given a basal diet containing approximately 10 g crude protein per MJ metabolizable energy was 0·63. For those ewes receiving the basal diet supplemented with either soya-bean meal, fish meal or blood meal the coefficients for the efficiency of utilization for the production of milk N of the increments in truly digested NAN supplied by the three protein sources were 0·61, 0·54 and 0·29 respectively. It is suggested that the low coefficient for blood meal may be due to its low content of methionine.

Gonadotrophin profiles during the follicular phase of the cycle prior to a synchronized mating and LH and progesterone profiles at days 2, 6 and 10 post mating were investigated in three groups of 16 ewes. of one group were given 0·5 × estimated metabolizable energy requirements for maintenance (MEM) from 14 days before mating until slaughter 11 days after mating (LL). Ewes of a second group were given 1·5 × MEM during the 14 days prior to mating and then 0·5 × MEM thereafter (HL) and the remaining ewes (HH) were given 1·5 × maintenance throughout the experiment. During the day before mating, plasma samples were collected at 10-min intervals for 8 h and assayed for LH and FSH. Samples were collected at 15-min intervals for 8 h on days 2, 6 and 10 after mating. Each ewe was then injected intravenously with 10 μg LH and samples were collected for a further 3 h. All of these samples were assayed for LH and progesterone. Nutritional treatment did not affect mean follicular phase concentrations of FSH or LH or mean LH pulse amplitude but the mean LH pulse frequency (pulses per h) of LL ewes was lower than that of HH + HL ewes (0·37 v. 0·48; P < 0·05). After mating, mean LH concentrations were unaffected by sampling date or nutritional treatment but the mean LH pulse frequency was lower in HL + LL than HH ewes at day 2 (0·25 v. 0·40; P < 0·01) and day 10 (0·28 v. 0·38; P < 0·05). Mean progesterone concentrations (ng/1) were higher in HL + LL than HH ewes at day 10 (6·77 v. 4·80; P < 0·05) but pulse frequency was not significantly affected. Injection of 10 μg LH on days 2, 6 and 10 after mating was followed by a significant increase in progesterone concentrations. The progesterone response was greater (P < 0·05) in LL + HL ewes than in HH ewes. It is concluded that increased rates of embryo mortality often associated with undernutrition are unlikely to be related to lower circulating progesterone levels per se but may be attributable to reductions in mean LH pulse frequency and consequent changes in progesterone profiles.

The relationships between body fat depots and body condition score (BCS) were determined in 52 adult Rasa Aragonesa ewes aged 10 (s.d. 2) years and ranging in BCS from 1·5 to 4·5. BCS of each ewe was assessed by three people, the repeatability within individuals being 90% and between individuals 80%. The ewes were weighed before slaughter. After slaughter the omental, mesenteric, kidney and pelvic fat were separated and weighed. The fat of the left side of the carcass was separated into subcutaneous and intermuscular depots. The relationship between live weight and BCS was semilogarithmic and those between fat depots and BCS were logarithmic. Regression analysis was also used to describe the relationships between the various fat depots and BCS or live weight. Of the variation in total fat weight, proportionately 0·90 was accounted for by variations in BCS, while 0·84 was accounted for by variations in live weight. For individual fat depots proportionately 0·86 to 0·90 of the variation was accounted for by variation in BCS and 0·69 to 0·79 by variation in live weight. BCS was a better predictor than live weight of the weight of both total body fat and the individual fat depots.

A curvilinear regression between BCS and live weight showed that the increases in live weight for a unit change in BCS was 7, 10, 12 and 16 kg for each one point increase in BCS from 1 to 5 respectively.

The tail fat depot (tail fatness score) was assessed in the same ewes by score on a three-point scale. Of the variation in the weight of individual fat depots, proportionately 0·79 to 0·86 was accounted for by variation in tail fatness score. Thus the tail fatness score could be used as an additional method of assessing body condition in the Aragonesa breed.

Purebred lambs from three breeds (East Friesland, Oxford and Texel) were fed to appetite and treated with recombinant DNA-derived bovine somatotropin (BST) at 0·05, 0·10 or 0·20 mg/kg per day or a placebo between 19 and 27 weeks of age.

BST maintained growth over the later stages of treatment but did not affect average daily live-weight gain over the whole treatment period. Food intake was decreased and food conversion efficiency was improved by BST treatment. However, carcass weight and killing-out proportion were decreased by BST treatment in all breeds. Consistent breed differences were observed. Oxford lambs had higher live weights and carcass weights than the other breeds but no breed × treatment interactions were apparent. Ultrasonic backfat depth (mean of measurements at 13th rib and 3rd lumbar vertebra) was reduced by BST treatment in the Oxford (the fattest) and Texel lambs but not in the East Frieslands (the leanest). Ultrasonic muscle depth was not affected by BST; East Friesland lambs had considerably smaller muscle depth than the other breeds.

In the loin joint, BST decreased fat cover and subcutaneous fat proportion and increased longissimus dorsi proportion. There were also marked breed differences with Oxfords fatter than the other breeds. BST treatment did not affect any of the meat quality indicators but some differences between breeds were apparent in ultimate pH and colour of the muscle. Fatty acid composition of the subcutaneous fat was affected by breed but not by BST. East Friesland lambs had higher concentrations of unsaturated and branched chain fatty acids than the other breeds.

A mammogenic response (increased mammary development and some secretory activity) to BST was observed in some female East Friesland and Texel lambs but not in Oxford lambs.

Ninety-nine indigenous Greek goats (Capra prisca) on natural pasture were used to study reproductive performance as affected by method of breeding. At the beginning of the breeding season oestrus was induced with 45 mg fluorogestone acetate-impregnated vaginal sponges and 300 to 500 i.u. PMSG, given on day 15. The animals were separated into groups of 33 for mating: (i) by natural service (natural mating); (2) by artificial insemination with fresh semen (AI fresh); and (3) by insemination with frozen semen (AI frozen). Alpine bucks and Alpine semen were used for natural mating and AI of does.

No effect of method of breeding was found on gestation length and sex ratio. Reproductive performance was significantly affected by method of breeding; at 21 days after insemination, fertility of does bred with fresh semen (69·7%) was comparable with that obtained with natural mating (72·7%) but the fertility of does bred with frozen semen (48·5%) was much lower. This difference was statistically significant at kidding (P < 0·05). Litter size was 2·14 and 1·86 (P < 0·05) for the AI fresh and AI frozen groups, respectively, and was 1·96 for the natural mating group. More multiple litters were obtained with the AI fresh group (P < 0·05). It is concluded that, induction and synchronization of oestrus and ovulation at the beginning of the breeding season gave reproductive performance with fresh semen comparable to that obtained with natural service, but use of frozen semen gave unsatisfactory results.

Rumen microbial attachment to food particles had a significant negative correlation with cell wall contents, cellulose and acid-detergent fibre (ADF) (P < 0·01) and a non-significant correlation with lignin, total phenolic content, condensed tannins and protein precipitation capacity. Microbial colonization was significantly positively correlated with in sacco dry-matter loss (DML) at 48 h of rumen incubation (P < 0·01). The cell wall contents and ADF were good predictors of DML. The relationship of DML (Y) with cell wall contents (X) and ADF (X') were Y = 1·012 – 0·0011Z, r = –0·89 (P < 0·01, no. = 10) and Y = 0·977 – 0·0014X', r = –0·88 (P < 0·01, no. = 10), respectively.

Sixty crossbred gilts, comprising 15 groups of four littermates from large litters (> 12 pigs) were used. The experiment was designed as a 2 × 2 factorial, with two initial live weights (light (L) or heavy (H)) and either a low (lp) or high (hp) plane of feeding during the rearing period and post mating within each litter group. The hp gilts were significantly heavier (P < 0·001) and had thicker backfat (P < 0·001) than the lp gilts at first oestrus. The Hhp gilts reached puberty 26·4 days earlier than the Hip gilts (P < 0·05). Level of food intake significantly influenced ovulation rates and total embryo numbers, the hp gilts shedding on average 2·4 more ova and having 1·9 more embryos than the lp gilts 28 to 35 days post mating (P < 0·05). However, due to somewhat higher embryonic mortality in the hp gilts, the mean number of live embryos was not significantly higher than for the lp gilts (P < 0·05).

The effects of mixing on agonistic behaviour and performance were studied in 90 individually fed pigs, grouped at a mean live weight of 23 kg by one of the following three methods: (1) six unmixed littermates; (2) three pigs from each of two litters; and (3) one pig from each of six litters. Unmixed animals (treatment 1) did not fight and the mode of fighting differed between treatment 2 and 3. Mixing decreased daily weight gain in the period 23 to 100 kg for barrows, but not for gilts. For gilts, food conversion efficiency was negatively correlated with the number of received attacks in the observation period (80 min after mixing), and for the barrows to fighting injury scores.

The potential of using low glucosinolate full-fat rapeseed meal (FFR) in the diets of early-weaned piglets was investigated in this experiment. FFR (containing 17 (xmol total glucosinolates and 2·8 mg erucic acid per g of whole seed) was used on an isonitrogenous basis at either 200 g<kg or 400 g<kg to replace full-fat soya-bean meal. Four replicates of three contemporary pen groups of 10 piglets weaned between 3 and 4 weeks of age were randomly allocated at a mean initial weight of 8·41 (s.e. 1·105) kg to the three dietary treatments. The experimental diets were given for 3 weeks. Inclusion of FFR significantly depressed food intake, daily live-weight gain and food conversion. There were no significant differences between the performance of piglets given either 200 or 400 g FFR per kg. Intakes of the diets containing FFR per unit of metabolic body weight as a proportion of the control diet decreased from about 0·85 to 0·75 from the 1st to the last week of the feeding period. This suggested that the acceptability of diets containing FFR decreased with time, possibly due to the presence of unpalatable and perhaps antinutritive factors which may include glucosinolates although present at relatively low concentrations. It was concluded that the feeding value of FFR would need to be improved by development of oilseed rape varieties with much lower levels of glucosinolates.