The effects of conventional and novel methods for the manipulation of growth in meat animals are reviewed within the context of the fundamental laws that determine the biological efficiency of energy conversion. Interspecies comparisons reveal large differences in the energetic efficiency of growth between mammals and birds. The similarity between mammals of different sizes is remarkable, both between and within species, which suggests that manipulation of growth rate per se has little effect on efficiency. The best way to improve the efficiency of growth is to maximize the conversion of metabolizable energy (ME) to lean tissue at all stages of maturation. The principal destination of ME is heat, however thermogenesis linked to essential metabolic functions is resistant to manipulation. It is more profitable to manipulate the partition of retained energy between protein and fat. Whether this is achieved by nutrition, conventional breeding or bioengineering, it is necessary to ensure that it does not compromise the normal health and vigour of the growing animal.

The relationship between carcass and meat quality research and industry requirements is examined with reference to changes in consumer demand.

It is argued that much applied research has not focused sharply enough on commercial requirements and that technology transfer has been slow. But there are signs that the pace of future development will be faster, associated with a reorientation of research objectives. The three key factors stimulating change are as follows.

(1) Increasing demand by retailers for a consistent product (a demand enforced by the buying power of the major multiple grocers).

(2) The reappraisal by breeders of selection objectives.

(3) Recent developments in sensor technology and the exploitation of information technology at the producer-processor interface.

The implications of each of these are discussed.

Driven by these factors research will be targeted increasingly on integrated systems from production to consumption, aimed at specific markets setting different balances between production costs and quality. These will be blueprints for best operating practices and the ‘state of the art’ against which new research developments will be evaluated.

Throughout the review, emphasis is placed on the importance of good communication between research workers and industry to confront change and realize the opportunities created.

A full lactation (300 days) experiment on 32 crossbred milch cows, separated into two groups was carried out. Group 1 (control) was given a concentrate mixture consisting of 400 g groundnut cake, 300 g crushed maize, 270 g wheat bran, 20 g mineral mixture and 10 g common salt per kg; while in group 2 (experimental) the groundnut cake was replaced by water-washed neem seed kernel cake (WWNSKC). Roughage was common in both the groups. Digestion and balance study on nine cows in group 1 and eight in group 2 was carried out after 3 months experimental feeding. Milk yield was recorded twice daily and butter fat, protein and total solids were determined every month in the milk of each animal. The results showed that there was no significant difference (P > 0·05) in the milk yield, butter fat content, organoleptic evaluation of milk, dry-matter intake, digestibility of nutrients, haemoglobin, SGOT, SGPT, acid phosphatase and alkaline phosphatase in blood and reproductive ability of the cows in the two groups. The nitrogen balance was higher in the WWNSKC group (P < 0·05) due mainly to less excretion of urinary nitrogen and a concomitent decrease in blood urea nitrogen. It is concluded that water washing of neem seed kernel cake, whose potential availability in India alone is about 0·9 Mt annually, can convert this cake, hitherto going waste, into an excellent high protein animal food and can be used for feeding milch animals without any adverse effect. It is recommended that this technology be adopted by all neem seed-cake producing countries, specially those developing countries having chronic shortage of foods and fodders for animal feeding.

A trial was performed to investigate the interaction between energy and protein source for cows in negative energy balance. Six cows were assigned to each of four types of concentrate (metabolizable energy (ME) 12 MJ/kg dry matter (DM), crude protein (CP) 180 g/kg DM) with low or high fibre contents (LF or HF) and protein degradability (LD or HD). For groups HDLF, LDLF, HDHF and LDHF respectively, acid-detergent fibre contents (g/kg diet DM) were 69·2, 66·1, 117·5 and 113·3; protein degradability values were determined as 0·78, 0·61, 0·72 and 0·66. Low-fibre diets were given at the rate of 11 kg/day concentrates with 6 kg/day hay (ME 8 MJ/kg DM, CP 84·5 g/kg DM) and high-fibre diets at the rate of 10 kg/day concentrates with 7·5 kg/day hay from weeks 4 to 13 of lactation. Cows were given a standard diet over the first 3 weeks of lactation and performance in week 3 was used as a covariate.

Milk yield was not affected by treatment but the butterfat content of milk from cows given the high-fibre diets (44·9 g/kg) was higher than the low-fibre diets (36·4 g/kg; P < 0·01), which resulted in significant differences in fat-corrected milk yield (HDHF: 25·1, LDHF: 26·2, HDLF: 22·7, LDLF: 21·5, s.e.d. 1·5 kg/day). ME balance was lower for the LDHF group (−4·1 MJ/day) than for groups LDLF and HDLF (-15·4 and -16·8 MJ/day; P < 0·05) but was not significantly different from the HDHF group (-29·4 MJ/day). Undegradable protein (UDP) balances were 53·4, 93·8, -21·2 and 193·8 (s.e.d. 45, P < 005) g/day for groups HDLF, LDLF, HDHF and LDHF respectively. Differences from zero in ME and UDP balance were accounted for in all groups except LDHF by changes in live weight plus a systematic correction of approximately 10 MJ/day. It is considered that the cows given high-fibre diets responded to high UDP by increasing fat-corrected milk production but mobilization of body fat to support this increase was accompanied by retention of protein with associated water, resulting in very little change in live weight. It is also probable that the high UDP levels improved the digestion of fibre i n the rumen, but this is unlikely to have been sufficient to have accounted for all of the response.

Developments, both recent and potential, in procedures for manipulating embryos are described. The procedures considered include: embryo transfer, multiple ovulation and embryo recovery, recovery of oocytes, in vitro maturation (IVM) and fertilization (IVF) of oocytes, in vitro culture of zygotes, embryo splitting and nuclear transfer, embryo storage, embryo sexing, gene transfer and embryo stem cells. The impact of these procedures on breeding strategies such as multiple ovulation and embryo transfer (MOET) nucleus breeding schemes and progeny testing are discussed for both dairy and beef cattle.

For MOET nucleus schemes all these procedures have potential applications in producing maximal rates of genetic progress for a fixed rate of inbreeding. With the current effectiveness of the procedures, embryo sexing and nuclear transfer would have the most impact. The potential for increasing genetic progress through progeny testing is enhanced using multiple ovulation, embryo recovery and transfer in cows to breed bulls, but no other procedures appeared to offer major benefits. The efficiency of beef production from the dairy herd could be increased either by using IVM and IVF to produce more beef-type calves or, potentially, by cloning and embryo transfer, to produce pure beef calves. Procedures leading to the production of clone families would make an impact on the evaluation of genotypes and environments. Gene transfer may be used to modify the composition of milk including the production of pharmaceutical proteins, and to increase milk yield or the efficiency of lean meat production.

It is concluded that, although much further research is required, the procedures discussed will have major implications for the structure and organization of dairy and beef cattle herds over the next decade.

A study was made of the effect of fibre level in forages on eating and rumination behaviour. Eight forage diets were prepared from the leaf and stem fractions of two grasses and two legumes and were given at hourly intervals to four steers under steady-state conditions. Eating and rumination behaviour were measured automatically by recording changes in intra-oesophageal pressure.

Mean voluntary intake of leaf was higher than that of the stem fractions (9·9 v. 5·6 kg/day; P < 0·001). This was associated with a shorter mean retention time in the rumen of the leaf than that of the stem fractions (21·4 v. 30·6 h; P < 0·001) and a lower concentration (g/100 g dry matter (DM)) of fibre (52-0 neutral-detergent fibre (NDF) and 30·5 acid-detergent fibre (ADF) v. 68·2 NDF and 45·3 ADF). Similar values (P > 0·05) between diets were found for both the water and DM contents of the rumen (60·1 kg, 7·8 kg). Voluntary intake was not related to either.

No difference was found between forage diets in the mean time (132 min) and number (18·7) of periods spent eating each day (P > 0·05). Legume leaf fractions were eaten at a faster rate (g/min) than either the grass leaf or the stem fractions. Voluntary intake was related to the rate at which food was eaten (r = 0·89; P < 0·01) but no relationship was found with the time taken to eat food (r = –0·14; P>0·05). Eating rate was related to the level of both NDF (r = –0·91; P < 0·01) and ADF (r = –0·96; P < 0·001).

Differences between diets were found in rumination times (mean 425 min; P < 001), the number of boluses regurgitated during each period (27·6; P < 0·05) and during each day (485; P < 0·001), and in the weight of boluses (455 g; P < 0·05). No differences (P > 0·05) were found between diets in the mean number of rumination periods each day (17·6), the mean time spent ruminating during each period (24·3 min), the mean rate at which boluses were regurgitated (53·2 s per bolus), the interval between boluses (5·1 s), and the DM in a bolus (27·5 g). Rumination time and the number of boluses regurgitated either per period or per day were not related to the fibre content of the diet (P > 0·05).

The regurgitated boluses from leaf fractions were chewed less than the stem fractions (43·7 v. 54·7 chews per bolus). The regurgitated boluses of lucerne leaf were chewed at a faster rate (1·13 chews per s; P > 0·05) than regurgitated digesta of the other diets which were chewed at similar rates (0·97 chews per s; P > 0·05). The total number of rumination chews made each day by animals given lucerne leaf (12 300) was much lower (P < 0·001) than that by animals given the other fractions (25 300). The number of chews made on each bolus was related to fibre levels in the diets (NDF, r = 0·78, P < 0·05; ADF, r = 0·91, P < 0·01).

It is concluded that the voluntary intake of high-fibre diets is not always restricted by rumen fill or rumination. The ease with which forage is eaten should be investigated as a factor influencing intake of fibrous forages.

The value of cloning in MOET nucleus breeding schemes has to be considered within the constraint of fixed resources. Under this constraint cloning was found to add to genetic progress only when (i) the heritability is low and (ii) it is used at the expense of a reduction in the number of bull families. This course would exacerbate inbreeding and other potential problems with MOET. All other options for using clones lead to a reduction in genetic progress due to a loss of selection intensity that is not made up for by gains in selection accuracy.

With a view to incorporating identifiable environmental factors such as month of calving as random effects in models for dairy sire and cow breeding value estimation, an analysis was undertaken of environmental components of variance for fat yield in lactations 1 to 5 of pedigree British Friesian-Holstein cattle. Records of fat yield and of log fat yield on daughters of widely used sires were first corrected for age and month of calving and for sire in each lactation on a national basis. Variance components for herd-year, herd-year × lactation, herd-year × month of calving, herd-year × month of calving × lactation and error were then estimated for pairs of lactations.

The results were most consistent for log fat yield, where heterogeneity of variance was less. Correlations were, approximately, 0·45 and 0·42 for records in the same herd-year and lactation for lactation 1 or a later lactation, respectively, and approximately 0·40 for records in the same herd-year and different lactations. Within herd-year and lactation, correlations of records in the same month were approximately 0·10 for lactation 1 and 0·08 for later lactations, and for different lactations the correlations were 0·06 if lactation 1 was included and 0·08 otherwise. Thus, the effects appear to be similar for later lactations, which can be pooled, but effects for month and month × lactation 1 v. later should be fitted.

Two pedigree indexes were calculated for 754 Friesian/Holstein bulls born between 1974 and 1980 and tested by the Milk Marketing Board. Correlations between the indexes and the average progeny performance were used to compare the indexes calculated using information from different ancestors. The first index, or estimated predicted difference (EPD) was calculated as one-half the sire progeny test, known as an Improved Contemporary Comparison (ICC), plus one-quarter the maternal grandsire ICC. The second index or estimated ICC (EICC) was calculated from one-half the sire ICC plus one-half the dam cow genetic index (CGI). The correlations between the deregressed bull evaluations (DICC) and EPD, calculated from 310 bulls, were 0·40 for milk yield, 0·40 for fat yield and 0·39 for protein yield. The corresponding correlations between the DICC and EICC, calculated from 314 bulls, were 0·43, 0·40 and 0·41. The regression coefficients for predicting average progeny performance from the EPD index were 1·00 (s.e. 0·13), 0·97 (s.e. 0·12), 0·96 (s.e. 0·13), 1·09 (s.e. 0·12) and 1·08 (s.e. 0·10) for milk, fat and protein yield, fat and protein percentage respectively. Those for predicting progeny performance from the EICC index were 0·95 (s.e. 0·11), 0·81 (s.e. 0·10) and 0·84 (s.e. 0·10) for milk, fat and protein yield respectively. Although the correlations were lower than the expected values of 0·50 (EPD) and 0·55 (EICC) the regressions were near the expected value of 1. The results indicate that the indexes are useful as a preliminary means of selecting bulls prior to progeny testing.

In experiment 1, 12 cows (treatment L) received natural daylight (mean 8 h), 10 h supplementary light (mean intensity 481 lux) and dark (mean 6 h), and 12 cows (treatment N) received natural daylight (mean 8 h) and dark (mean 16 h) for 8 weeks. Supplementary light increased milk yield and tended to increase food intake and loss of live weight. There were no significant effects on milk composition although milk fat concentration tended to be reduced for cows in treatment L. Daily feeding times were not affected by treatment although more feeding took place during the supplementary light in treatment L. Supplementary light increased the time spent lying down per day and reduced the distance walked per day, although cows in treatment L spent less time lying during the light supplementation period. In both treatments, the activity rate was increased on the day of oestrus, but in treatment L the incidence of oestrus-specific behaviour was significantly reduced compared with treatment N.

In experiment 2, 16 cows received natural daylight (mean 9 h) and 10 h supplementary light of mean intensity 0, 101, 191 or 529 lux in a change-over design. There were no significant effects of supplementary light or its intensity on the food intake, milk yield or live-weight change of the cows, but milk fat concentration was reduced for cows receiving supplementary light at 101 or 529 lux and milk protein concentration was reduced for cows receiving light at 529 lux.

The use of cottonseed (CS) as an energy and protein supplement to high wheat-straw diets was studied in a digestion and nitrogen balance trial, and as a component in fattening diets in a feeding trial. The proportions of CS studied were 0, 60, 120, 180 and 240 g/kg diet dry matter (DM). DM intake per kg M0·75 was 33, 31, 40, 31 and 29 g, respectively. The digestibility coefficient of organic matter was 415, 463, 417, 441, 350 g/kg DM and of neutral-detergent fibre was 350, 436, 411, 309, 334 g/kg DM. Nitrogen balance was –2·52, 1·1, 5·38, 5·63, 7·60 g/day respectively, for the same order of treatments. The effect of high proportions of CS in the diet in restricting DM intake and reducing its digestibility was evident. The results of the feeding trial were in agreement with those of the digestion trial and indicate the proportion of 120 g CS per kg DM as optimal and maximal.

This study was designed to examine the influence of pre-wilting and season of harvesting of silage on milk production and food utilization by dairy cattle and also the response to protein and fat concentration of the supplement given with silage. Twenty-four lactating British Friesian cows were used i n a four-period, partially balanced, change-over design experiment to evaluate 12 treatments consisting of three silage types (spring harvest unwilted (lUnW), autumn harvest unwilted (3UnW) and autumn harvest wilted (3W), each offered in addition to supplements containing both two crude protein levels (160 and 210 g/kg fresh weight) and two levels of inclusion of a fat supplement (0 and 100 g/kg fresh weight) in a 3 × 2 × 2 factorial arrangement. All supplements were offered at the same rate of 6·8 kg/day. Total diet digestibility and food utilization studies were carried out on all animals at the end of the third and fourth experimental periods.

The wilted silage (3W) had a higher digestibility than the unwilted (3UnW) material and animals offered 3W silage consumed proportionately 0·18 more silage DM and produced 1·6 MJ/day more milk energy output than those offered 3UnW. The total ration digesibility and food utilization data showed the animals given the 3W silage consumed 13·5 MJ more metabolizable energy (ME) per day than those offered the 3UnW silage and were less efficient at converting the ME available for production into milk energy output than those given the 3UnW silage (efficiencies were 0·58 and 0·53 for 3UnW and 3W silages respectively).

There were no differences between the digestibilities of the spring (lUnW) and autumn (3UnW) silages with the concentrations of digestible organic matter in the dry matter (DM) being 650 and 648 g/kg respectively. However, animals offered lUnW silage consumed proportionately 0·12 more DM and produced 1·05 kg/day more fat-corrected milk than those given 3UnW silage. The food utilization data indicated that this milk production difference could be accounted for by the higher ME intake on lUnW and that the efficiencies of conversion of ME available for production to milk energy output were similar with both seasons of harvests (efficiencies were 0·57 and 0·58 for lUnW and 3UnW respectively).

Increasing the crude protein concentration of the supplement from 160 to 210 g/kg increased silage DM intake, milk yield and milk energy output by 0·2 kg, 0·66 kg and 1·9 MJ/day respectively with the response being greatest with lUnW and least with 3UnW silage. Increasing the fat concentration of the supplement depressed silage DM intake but increased milk yield by 0·53 kg/day.

The effects of the treatments on total diet digestibility, energy and nitrogen utilization, and the volatile fatty acid concentrations of rumen fluid are also given.

The effects of mixed v. mono-grazing by steers and sheep on individual animal growth rate, pasture carrying capacity and live-weight output per ha were measured in a 4-year (1978 to 1981) experiment, after a preliminary familiarization year, 1977. Annual stocking rate treatments consisted of three monosteer, three mono-sheep and seven mixtures of steers and sheep. Annual average stocking rates were 2·11 steers † 8·1 ewes per ha under mixed grazing and, under mono-grazing, 4·44 steers and 15·2 ewes per ha. The range from low to high in stocking rate in mono- or mixed grazing was close to 40%. Over the 4 years a total of about 280 steers, 900 ewes and 1100 lambs were used.

Overall, mixed grazing increased average lamb daily live-weight gain (ADG) to weaning and to drafting from 246 to 265 g (P < 0001) and from 211 to 223 g (P < 0·001) respectively. Steer ADG for these periods was increased from 1·419 to 1·520 kg (P < 0·01) and from 0·950 to 1·094 kg (P < 0·001). The choice of reference mono-grazing stocking rates for comparisons of mixed v. mono-grazing ADG can affect these results. Average live-weight outputs (kg/ha) from grazing for the mono-steers, mono-sheep, and mixed grazing were 663, 690 and 714, the range in the latter over the seven mixed grazing treatments being from 605 to 805. Stocking rate was the main factor affecting output per ha. Because of the management rules used in this experiment mixed grazing effects are more appropriately assessed through ADG and measures based on it than on output per unit of area.

Models were fitted relating steer and lamb ADG to lamb weaning and lamb and steer drafting to stocking rates of steers and ewes. Mixed grazing benefits to steer and lamb ADG to drafting were greater as their proportion in the mix declined and increased with stocking rate. At the 50% proportion, lamb and steer ADG were improved by 5·2 and 3·4% respectively at low stocking rate and 9·4 and 6·6 at high stocking rate. Predicted steer ADG to lamb weaning for a given steer stocking rate increased with increases in ewe proportion up to five ewes per ha and decreased rapidly with further increments in ewe proportion.

Mixed grazing efficiency was also evaluated through the Relative Resource Total. This showed that under mono-grazing 10 to 13% more area was required to produce the same grazing season output as under mixed grazing. The 10% improvement in carrying capacity was exceeded for ewe: steer frequencies ranging from 1·5: 1 to 10: 1. Explanations for this greater efficiency in resource capture/use under mixed grazing are discussed.

Substitution rates for lamb ADG to weaning (2·35) and to drafting (2·86) and for steer ADG to drafting (0·21) were fairly constant over the 4 years 1978 to 1981.

Selection of mixed stocking rates to suit growth rate targets for different animal types and to match food supply with demand under varying soil/climatic/topographical conditions is discussed.

Fifty-four Charolais-cross weaned suckled calves were used in an experiment to investigate the effects of feeding level during the post-weaning winter on their subsequent performance, when grazing different vegetation types in summer. During winter they were given grass silage and barley at one of three levels (low, medium and high). The winter live-weight gains were 0·50, 0·75 and 0·96 (s.e. 0·019) kg/day (P < 0·001) for the low, medium and high treatments respectively. During summer they grazed either a sown ryegrass pasture (S), a hill reseed (R) or part of an unimproved hill (H).

Live-weight gain during summer was inversely related to winter feeding level on all grazing treatments, the mean live-weight gains being 1·01, 0·79 and 0·65 (s.e. 0·027) kg/day (P < 0·001) for the low, medium and high winter feeding levels respectively. Summer treatment significantly affected performance, the live-weight gains being 0·88, 0·94 and 0·61 (s.e. 0·027) kg/day for the S, R and H treatments respectively. There was no interaction between winter and summer feeding treatment on live-weight gain. The organic matter intake was highest on the H treatment but the digestibility of herbage consumed was the lowest, resulting in the lowest digestible organic matter intake.

Similar and high levels of performance were obtained on sown ryegrass pastures and reseeded hill land, while unimproved hill vegetation supported only moderate levels of live-weight gain. Compensatory growth occurred when a wide range of vegetation types were grazed in summer.

The assumption that observed values for litter size result from a set of thresholds which impose a discontinuity to the visible expression of a continuous underlying variate has been tested using data from the French sheep recording scheme: 629 724 natural lambings for 32 different pure breeds and 66 379 litters obtained after synchronization treatment in 15 breeds. Frequencies of single, twin, triplet and higher-order births varied regularly with litter size within each fecundation type. The coefficient of variation of litter size was remarkably constant from breed to breed under both natural (0·35 to 0·40) and induced lambings for which it was higher (0·40 to 0·45). The regression of the difference between the first two thresholds of an underlying normally distributed variate on prolificacy was slowly negative for natural fecundation, but not significant for induced fecundation. A significant effect of fecundation type was found, resulting in a lower incidence of induced than of natural twin litters for the same level of prolificacy. Multivariate polynomial regression of frequencies on natural and induced prolificacy was used to predict expected frequencies of litter size and the proportion of lambs of each birth type.

Six groups of eight 2-day-old calves were placed successively in a controlled environment chamber. Three groups were exposed to air temperatures of 10° and 25°C. A liquid diet of skimmed-milk substitute and 4 I/day of drinking (free) water was offered to each calf. Within each group, calves were allocated to either a low (<0·2 m/s) or a high (>3 m/s) air velocity and to either a low (30 g dry matter (DM) per kg M0·75 per day) or a high (40 g DM per kg M0·75 per day) feeding level. At 8 days of age the apparent digestibilities of DM at air temperatures of 10° and 25°C were 0·77 (s.e. 0·126) and 0·82 (s.e. 0·126) respectively (P > 0·05). The apparent digestibilities of DM were greater at the low feeding level with low air velocity than for either this feeding level with high air velocity or the high feeding level at both air velocities (P < 0·05) between which there was no significant difference (P > 0·05). At 8 days of age there were significant air temperature × air velocity (P < 001) and air velocity × feeding level interactions in the intake of free water (P < 005). There was a significant air temperature × feeding level interaction for total water intake (P < 0·05). Urinary water loss relative to total water intake was significantly greater at the low air velocity than at the high air velocity (P < 0·05).

In a further two groups of eight calves given 30 g DM per kg M 75 per day at 8 days of age, the apparent digestibilities of DM at air temperatures of 10° and 25°C were 0·71 (s.e. 0·020) and 0·90 (s.e. 0·013) respectively (P < 0·01). In the same calves given 40 g DM per kg M0·75 per day at 20 days of age, the apparent digestibilities of DM at air temperatures of 10° and 25°C were 0·89 (s.e. 0·009) and 0·93 (s.e. 0·011) respectively (P < 0·05). The free and total water intakes, the ratios of (total water intake-faecal water loss): total water intake and the urinary losses of water were significantly greater at the air temperature of 25°C than at 10°C (P < 0·05). Live-weight gains were lower at 10°C than at 25°C (P < 0·01).

The results suggested that air temperature, air velocity and feeding level can affect the health and growth of calves less than 4 weeks of age.

Ninety primiparous and 150 multiparous Israeli-Holstein dairy cows were allotted at random into three groups. One group (LSBM) was given 170 g crude protein (CP) per kg dry matter (DM). The two other groups were given 210 g CP per kg DM with the extra 40 g CP per kg DM supplied either by soya-bean meal (HSBM) or by fish meal (HFM). Roughage supplied proportionately 0·20 of the dietary DM and was maize silage with groundnut hay in the winter and wheat silage with oat grass in the summer. The primiparous and multiparous cows were offered the experimental diets for 16 and 24 weeks, respectively.

Milk, milk fat and milk protein production (kg/day), for treatments LSBM, HSBM and HFM, were: 39·3, 40·0, 40·8 (P > 0·05); 1·12, 118, 1·06 (P < 0·05) and 1·21, 1·23, 1·26 (P > 0·05) for multiparous cows and 29·4, 31·2, 33·4 (P < 0·05); 0·89, 1·07, 0·93 (P < 0·01) and 0·92, 0·98, 1·05 (P < 0·05) for primiparous cows, respectively. Cows of high yield potential responded more to increasing dietary protein level than did cows of low yield potential. Average live-weight gains on treatments LSBM, HSBM and HFM were (g/day): 220, 160 and 310 (P < 0·05) in multiparous cows and 220, 170 and 230 (P < 0·05) i n primiparous cows, respectively. Pregnancy rates, 16 weeks after parturition, were 0·65, 0·52 and 0·72 (P < 0·05), in the LSBM, HSBM and HFM groups, respectively. The decrease in pregnancy rate in the HSBM cows, occurred in primiparous cows and cows in their fourth or later lactation but not in cows in their second or third lactation. The milk production of primiparous and multiparous cows in relation to the level and source of dietary CP and the utilization of energy and protein for milk production is discussed.

Four experiments were carried out to measure the effects of exposure to a cellulase preparation on the degradation characteristics of whole barley straw and its botanical fractions. The effect of chemical pre-treatment (NaOH and H2O2) on treatment with a cellulase preparation and the addition of propionic acid to inhibit loss of dry matter were also studied during the fermentation of ensiled straws. Samples of each preparation were incubated in polyester bags in the rumens of three sheep to estimate degradability. The data were described using the equationp = a + b (1—ec) where p is degradability at time t and a, b and c are constants.

The pH of ensiled straw was consistently decreased (P < 0·01) and the solubility increased (P < 0·01) by treatment with a cellulase preparation and by increasing the period of ensiling. The b values were decreased by increasing the length of the ensiling period (F < 0·01). The increase in the solubility of the treated botanical fractions was particularly apparent for the leaf blades, followed by leaf sheath, with the internodes being least affected (P < 0·01). However, treatment with a cellulase preparation had little or no effect in increasing the potential degradability (a + b) of any botanical fraction. The degradability of the whole plant and botanical fractions of straw increased (P < 0·01) with NaOH treatment and was further improved (P < 0·01) by alkaline H2O2 treatment. The increase was greater in internodes than in leaf sheath (P < 0·01). Treatment with a cellulase preparation and chemical pre-treatment had little or no effect on the 48-h dry-matter loss (DML) and the (a + b) values, but it increased (P < 0·01) the a values and solubility.

DML from straw treated with a cellulase preparation during fermentation decreased (P < 0·01) from about 60 to less than 10 g/kg with 30 g propionic acid added per kg straw to inhibit bacterial activity. The decreased fermentation loss was reflected in an increase in the 48-h DML and potential (a + b) values of straw treated with a cellulase preparation.

Lactation curves are fitted to data as a preliminary to estimating summary statistics. Two widely quoted curves are atbe-ct (Wood, 1967) and a(1 - e-bt) - ct (Cobby and Le Du, 1978), each of which has three parameters. Restriction to either of these curves imposes limitations on the fit to the data and can result in biased estimation of summary statistics. Alternatively, lactation curves can be generated by the use of a non-parametric method which requires only weak assumptions about the signs of derivatives of the curves. Because the non-parametric curves are more flexible, estimates of summary statistics are less likely to be biased than those based on parametric models. Use of the non-parametric curves is particularly advantageous around the time of peak yield, where the curves of Wood and Cobby and Le Du are known to fit data poorly.