Fourteen Friesian and 13 Angus steers, grown at pasture, were selected so that their carcass weights fell evenly throughout the range 200–300 kg. The right half of each carcass was dissected into muscle, fat, bone and fascia and tendon, and the left half boned-out and fat trimmed into retail cuts. A step-wise multiple regression procedure, including a pseudovariable for breeds, was used to compare compositional components on a common weight basis.

Angus, at the lower end of the live-weight range, had heavier empty bodies than Friesians; at the heavier end of the live-weight range, this was reversed. This relationship between live weight and empty body weight was due to variation between breeds in the weight of contents in the fore stomachs but not the intestines.

When compared at either the same live weight or the same empty body weight, Angus had more hot carcass than the Friesians (8·0 and 8·4 kg, respectively). There was no difference between breeds in loss of carcass weight in the 24 h post-slaughter.

There was no breed difference in weight of blood, head, kidney and channel fat, kidneys, liver, diaphragm, heart, lungs, tail or fore-stomachs, when compared at the same offal weight. The feet and intestines were, respectively, 0·55 and 2·43 kg heavier for Friesians than for Angus at the same offal weight, but the pizzle was 0·11 kg lighter. Hide weight was greater in the Angus at all offal weights, with the difference between breeds being 0·016% of (offal weight).

There was no difference between breeds in the weight of muscle or the weight of fascia and tendon when compared at the same dissected side weight; however, the Angus had 4·8 kg more fat and 3·0 kg less bone than the Friesians at the same dissected side weight.

When compared at the same muscle weight the Friesians had 1·04 kg more proximal hind-limb muscles, 0·30 kg more proximal fore limb muscles, but 0·74 kg less abdominal muscles than the Angus. At all dissected muscle weights the Angus had a greater weight of muscles of the neck and thorax, and this difference increased with increasing weight of dissected muscle. The Friesians also had 1·52% more of their muscle as ‘expensive muscle’. There were no breed differences in the distribution of any other muscle groups.

There was no breed difference in the distribution of dissected fat between subcutaneous and intermuscular depots when these were the only fat depots considered. However, when kidney and channel fat was included in the total dissectable fat of the carcass, Friesians had 22·4 % more kidney and channel fat, the same weight of intermuscular fat and less subcutaneous fat than the Angus at the same total dissected fat weight.

Friesians tended to have more of their bone weight in their legs (humerus, femur, tibiar–tarsus, radius–ulnar–carpus) and Angus more in their thoracic region (thoracic vertebrae and ribs, scapular and sternum-costal cartilages).

At the same retail side weights there was no difference between breeds in the weight of fat-trimmed, boned-out, retail cuts; however, the Friesians had 3·3 kg more retail bone than Angus, but 2·3 kg less fat trim and 0·46 kg less sausage mince.

There was no difference between breeds in the distribution of retail bone or of fattrimmed, boned-out cuts between the forequarter and hindquarter. However, the Friesians had 13·2 % more fat trim in the hindquarter. Friesians had 0·27 kg more retail cuts located in the rump but 0·80 kg less in the loin, when compared at the same weight of retail cuts. Also, the Friesians had more of their retail cuts as topside, thick flank and foreshin. There were no other differences between breeds in the distribution of retail cuts.

Two trials with 82 and 54 Chios ewes and one with 44 Damascus goats were conducted during 1973–4 to study the effects of partial suckling on lactation performance and the growth rate of the lambs and kids. The treatments studied were: (a) continuous (24 h a day) suckling; (b) partial (8 h a day) suckling and (c) partial with decreasing duration of suckling from 8 h daily to residual suckling, i.e. 20 min after each milking.

No significant differences were found in either the 35-day milk yields (commercial milk yield+milk sucked) of ewes or the 70-day milk yields of goats between animals suckling continuously and those suckling partially. However, the 35-day and 70-day commercial milk yields of ewes and goats on the partial suckling treatments were significantly higher than those of ewes and goats on the continuous suckling treatment. Residual suckling of ewes from 36 to 120 days after lambing resulted in an increase in the total milk yield but not in the milk sold. Significant differences in milk yields were found between ewes nursing one and those nursing two lambs. The corresponding mean values for the 35-day milk yield were 78 and 94 kg and those for the 120-day milk yield 205 and 238 kg.

Up to weaning (35 days for sheep and 70 days for goats) the growth rate of lambs and kids on the partial suckling treatments tended to be poorer than that of those on continuous suckling. After weaning, however, the growth rate of the partially suckled lambs and kids as well as that of the lambs on residual suckling from 36 to 120 days were as good as that of those suckling continuously so that the differences in the final weight were not significant.

It is concluded that partial suckling in sheep and goats is profitable in the early stages of lactation provided that the milk left to the young and the intake of solid feed are sufficient to sustain satisfactory growth.

Fairfield Smith's law relating the error of a field experiment to the size of its plots is well established by observation and in general use, but it appears to have no theoretical basis. It is here shown by computer simulation that it could arise in many ways as a good approximation, provided the range of plot sizes considered is not unreasonably large.

Two cocksfoot varieties, Be 6394 and Be 6392, selected for high in vitro digestibility (IVD) were compared with S. 37 under sward conditions. The pattern of extension growth in reproductive tillers, the associated changes in N and water-soluble carbohydrates content (WSC), and IVD of individual plant organs were studied.

Ten days after the mean date of head emergence, the most digestible plant organs were the leaf blades and the inflorescences; the lower internodes had the lowest IVD. At the mature hay stage the ranking was leaf blades > upper sheaths > inflorescence > lower internodes > upper internodes. At each stage N content was highest in the upper portions of the reproductive tiller, whilst the highest WSC content was found in the basal internodes.

After completion of extension growth in the lower three of the five or six internodes which showed appreciable extension at flowering, there were markedly greater increments in weight per unit length in Be 6394 and Be 6392 than in S. 37. The WSC content of the stem was higher in Be 6394 and Be 6392 than in S. 37. The IVD of the lower internodes declined more rapidly in S. 37 than in Be 6394 and Be 6392.

Results are discussed in relation to formulating a model for improved nutritive quality in conservation crops, and in relation to the height of harvesting such crops.

Early stages in the development of reproductive tillers of cocksfoot were studied in uncut swards, and in regrowth after cuts taken at a critical stage.

In uncut swards, initial stages of stem extension (up to 10 cm) occurred over 3 weeks. During this period the last three or four leaves initiated on the tiller became visible externally. Tillers with stems of less than 2 cm generally had one or two partially emerged blades. Those with stems 2–6 cm long carried two, three or, occasionally, four partially emerged blades; the sheaths of two, and sometimes three of these leaves extended beyond the stem apex. Hence, for tillers with 2–6 cm stems, even lenient defoliation involving the removal of only one expanded blade would result in the severance of a further two, three or four developing leaves; at an earlier stage only one or two developing leaves would be severed.

In the regrowth studies it was found that defoliation of reproductive tillers whose apices were below cutting height and which, at the time of cutting, had stems of up to 8 cm, led to contrasting effects: some made only negligible regrowth, whilst others made rapid extension growth. Some of the most rapidly extending tillers retained no leaf blades, and therefore the alternative effects of defoliation cannot be completely explained in terms of leaf retention. Other possible mechanisms contributing to these effects are discussed.

Five potato varieties were grown at four sites over 2 years, and assessed for yield, specific gravity and for sloughing, texture and discoloration following steaming. Variation in discoloration was attributed to differences between sites. Genetic differences were established for specific gravity, sloughing and texture, all of which were positively correlated. Genotype × environment interactions and years × sites interactions contributed to the variation for these and the remaining variables.

The effect of six intervals between harvests and three levels of N application on the number of tillers was studied in four ryegrass (Lolium) varieties in the first harvest year and in two varieties in the second and part of the third harvest years in a field experiment. Weight/tiller was recorded in the first 2 years and weight/green leaf and number of green leaves harvested were recorded in the first year.

There was a large negative interaction between level of N and interval between harvests in respect of number of tillers, the number being increased by N with short intervals and tending to be reduced by N with long intervals. The interaction was larger in the second and third years than in the first. The negative interaction in respect of number of tillers and number of green leaves harvested was associated with a reduction in the percentage increase in yields of dry matter, digestible organic matter and N due to the application of N as the interval was increased. It is suggested that one should in general avoid a combination of high N and long intervals between harvests with present varieties of perennial ryegrass.

The effects of interval between harvests, N application, variety and time of year on number of tillers and weight/tiller are discussed in relation to their effects on yield to show how tiller information can contribute to a better understanding of the ways in which yields are achieved. Rather low yields in the late summer and early autumn of the second harvest year were in many instances attributable to a rather low number of tillers rather than to low weight/tiller.

There was a considerable increase in number of tillers between the end of one growing season and the early part of the next, suggesting that the period in which this increase is achieved may be an important one in relation to the continued vigour of the sown species.

A reduction in annual yield of green leaf with extension of interval from 6 to 10 weeks with N applied resulted from a reduction in the number of green leaves harvested/harvest and in the number of harvests, which outweighed an increase in number of green leaves harvested/tiller and in weight/green leaf. Increasing the level of N with intervals of 3–6 weeks increased the number of tillers, the number of green leaves harvested per tiller and per unit area of ground and weight/green leaf and hence consistently increased green leaf yield.

The effect of six intervals between harvests and three levels of N application on the dry-matter yield of total herbage, the proportion and yield of green leaf, dead leaf, ‘stem’ and inflorescence of the sown species, the proportion and yield of unsown species, the digestibility and digestible yield and N content and yield was studied in S. 23 and S. 321 perennial ryegrass during a 30-week period in the second harvest year in a field experiment. In S. 23 the ‘stem’ was divided into true stem, leaf sheath, unemerged leaf and unemerged inflorescence.

The results supported the main findings from the first harvest year (Wilman et al. 1976a, b).

There was marked ingress of unsown species in the second harvest year with 3-, 4- and 5-week intervals in S. 321. The combination of 262–5 or 525 kg N/ha/year and 8- and particularly 10-week intervals over 2 years was too severe a treatment for the satisfactory survival of S. 23. With 525 kg N and 10-week intervals, S. 321 was equally badly affeoted.

The application of 525 kg N/ha/year compared with nil reduced the proportion of green leaf in total herbage of the sown species by 11 percentage units, on average, and increased the proportion of ‘stem’ by 12 percentage units, in the second harvest year. The effect of N application on the proportion of crop fractions was found in both varieties and in all months of harvest. The effect was much greater than in the first harvest year. In S. 23 the application of 525 kg N compared with nil in the second harvest year increased the proportions of both true stem and leaf sheath (in total herbage of the sown species), true stem being the more important of the two, in this context, with the longer intervals and leaf sheath being the more important with the short intervals.

Digestibility was not in general affected by N application despite the higher proportion of stem and leaf sheath and the lower proportion of green leaf blade resulting from N application. N did, however, tend to reduce digestibility at the harvests at which the proportion of stem was highest.

Digestibility varied from one time of the year to another with a constant interval between harvests, but not as much as in the previous year. Lower digestibility of leafy crops in summer and autumn than in April and early May in both years may have been partly due to a higher proportion of dead material.

Three periods were distinguished approximately in both years: May-June, July-August, and September-October. Only in the first of these periods was there a substantial increase in yield of digestible organic matter as a result of doubling the interval between harvests. Doubling the interval reduced digestibility in all three periods, but especially at harvests within the second period. Yield response to N was large in the first period, intermediate in the second, and low in the third. Apparent recovery of N was low and N content of herbage unduly high in the third period. N content of herbage was low with the long intervals between harvests at harvests in the first two periods. Applied N increased N content at these harvests and at all other times.

Changes in the amounts of the trace elements iron, zinc, manganese and copper, present in intact grain, endosperm, testa pericarp and embryo throughout the developmental period were recorded. Accumulation of iron and zinc in intact grain was extremely rapid, rising in the case of iron from 40 ng/grain to 1600 ng/grain between 5 and 40 days after anthesis. Manganese accumulated steadily throughout, reaching a final value of 475 ng/grain. Significant amounts of copper were present throughout and a small peak was observed at 38 days after anthesis. Some variation from this overall pattern was found in the different morphological parts as development proceeded. This was notable in the case of embryo manganese, which was undetectable until 45 days after anthesis, when levels rose dramatically to reach a steady maximum of around 100 ng/embryo. The results are discussed in relation to the biochemical events accompanying maturation.

In capacity studies on adult White Leghorn hens during exercise it was found that the maximum velocity of the bird was 2·4 km/h.

Metabolic rate increased linearly with running speed and, between the speeds of 1 and 2 km/h, was 53–65% higher than the metabolic rate at rest.

The bird moved faster largely by increasing its stride.

The cost of bipedal locomotion for hens, in contrast to men and rhea, is lower than the cost of the quadrupedal locomotion of mammals.

The influence of standing, spontaneous activity and eating on heat production was determined.

The extra heat production of standing is negatively correlated with the length of standing period. In a short standing period of 30 min the associated activity, pecking against the respirometer wall and fluffing the feathers, was high and the heat production was increased by 25% compared with that during sitting. After standing for 1½ h spontaneous activity was very low and the difference in heat production between the standing and sitting bird was reduced by 9%.

During eating the heat production increased by an average of 37% (range 11–68%); this was due mainly to the act of eating per se and not to the work of digestion.

The mean energy cost of eating was calculated to be 143 J/kg0·75/min spent eating.

The growth and development of two varieties of Phaseolus vulgaris (Purley King and Limelight) were compared in two experiments in 1973 and 1974 at a range of planting density from 20 to 100 seeds/m2. Within this range, the relationship between seed yield and density in Purley King was asymptotic, although there was a suggestion that if even higher densities had been tested, a downward trend in yield might have occurred. The optimum density of planting for Purley King in Expt 1 was 50 seeds/m2 while that for Limelight was 40 seeds/m2. The corresponding densities in Expt 2 were 75 and 50 seeds/m2 respectively. Although number of branches per plant generally decreased with increasing density, there was no significant density effect on the number of nodes per plant. Thus stabilization of seed yield occurred even at quite low densities. Although in the low-density treatments, less vegetative tissue was produced, the peak of dry-matter yield occurred later after flowering and the slower subsequent senescence ensured the presence of active photosynthetic tissue throughout the pod-fill stage. Less pod retention occurred at high density which, combined with the ability of widely spaced plants to produce pods over a longer period, resulted in a similar number of pods per unit area over a wide range of density.

Although the variety Purley King produced more than double the number of mature pods from its extra nodes and branches, it was outyielded by Limelight by 35% from the combined effect of more seeds per pod and a higher mean seed weight. Limelight also produced this high yield with less vegetative tissue. In both varieties it appeared that pod photosynthesis could take place, in Purley King because the pods were borne on higher nodes above the canopy and in Limelight due to the earlier senescence of its smaller leaf area. However, in spite of the apparent physiological advantages of Limelight, the pods are not borne high enough on this plant to enable satisfactory mechanical harvesting.

A tightly autumn-calving herd of British Friesian cattle, running throughout the year as a single group, received two different amounts of mineralized barley supplement per day during the winter of 1973/4. As they calved, alternate cows were allocated to one of two rates of supplementary feeding differing in the approximate ratio 2:3 which was applied according to the calendar and not specifically related to milk yield.

Average herd size for the year was 140 cows; average milk sales were 4680 kg/cow at a total supplementary concentrate use of 650 kg/cow. All cows had free access to selffeed silage reinforced with brewers' grains. Cows which received 50% more supplement from 8 September to 31 January (an extra 235 kg/cow) showed an improvement in total lactation milk yield of 248 kg/cow. Standard lactation curves were used to demonstrate a significant interaction between quantity of concentrates fed and season of production and showed that most of the improvement took place during the winter period.

Twenty experiments between 1970 and 1974 tested the effect of spring applications of all combinations of five amounts of potassium (0–333 kg K/ha) and two amounts of sodium (0, 150 kg Na/ha) on sugar-beet yield, profitability and the amount of K and Na in the crop in mid-summer. The experiments were in commercial sugar-beet crops where soils contained little exchangeable potassium, most having less than 100 mg K/l in top soil. On average, sugar yield was greatest when 150 kg Na/ha plus 167 kg K was used, the fertilizer increasing yield by 0·71 t/ha. Response to this combination of K and Na fertilizers was 1·45, 0·57 and 0·22 t sugar/ha when the soil contained less than 60, 61–120 and more than 120 mg K/l respectively.

From results of earlier experiments, it was thought that response to even a large amount of K without Na would be less than to Na without K because usually there is insufficient rainfall in the spring for maximum response to K. In fact, 333 kg K/ha increased yield by about the same amount as Na, probably because a large proportion of these experiments were made in wetter-than-average springs.

The concentration of K in dried tops and roots in mid-summer increased linearly with soils containing from 40 to 120 mg K/l but when there was more K in the soil the concentration of K in the plants did not increase further. By contrast, the concentration of Na in plants did not reach a similar plateau. Whole plants in mid-summer contained between 128 and 251 kg K/ha and 42–102 kg Na/ha, depending on the fertilizer treatment. Recovery of applied K varied from 90% when small amounts were used on fields containing least K to about 20% when large amounts were given on fields containing moderate amounts. Recovery of Na varied from about 50 to 23%.

The comparison of the financial return from using K alone or K plus Na showed that Na was essential for maximum profit. On average, the most profitable dressing of K gave a return of £14/ha whereas K plus Na gave up to £46/ha. On fields with least K, the corresponding results were £50 and £115/ha respectively. In relation to soil K, the most profitable application was 150 kg Na/ha plus about 150 kg K/ha on soils with 60 mg K/l or less, and 150 kg Na/ha plus about 75 kg K/ha on most other soils.

Experiments with wheat describing the effects of crop thinning on grain yield and its components are presented. These were carried out over 5 years in northwest Mexico, using a high-yielding dwarf spring-wheat variety (Triticum aestivum cv. ‘Yecora 70’) grown under irrigation and high fertility. It was shown that thinning largely relieved competition for light, thus increasing photosynthate levels in the plants remaining after thinning. The objective was to evaluate this simple technique as a guide to understanding when grain yield and its components were determined and, in particular, the extent to which post-anthesis photosynthate supply limited yield.

There were major responses in grain yield with thinning between about 50 and 100 days after seeding, and in number of spikes and of grains with thinning between 50 and 90 days (50% anthesis was reached at 87 days). Number of spikelets per spike showed small responses to early thinning (before 50 days). Number of grains/spikelet and kernel weight snowed positive responses to thinning between 65 and 90 days, and 90 and 115 days, respectively. These results agreed with adjacent shading and CO2 fertilization studies but, because of certain difficulties in interpretation of responses, pre-anthesis thinning was not considered a very useful technique.

Anthesis thinning was carried out on 21 separate crops: the kernel weight increase relative to the unthinned control ranged from 6 to 41%, averaging 20%. Anthesis thinning led to increases in stem weight during the first half of the grain filling period, followed by increases in grain growth rate in the latter half. The increase in final kernel weight was greater with higher temperature and lower radiation during grain filling; these variables explained 64% of the variation in kernel weight response. It is suggested that the kernel weight response does indicate the degree of photosynthate limitation during grain filling, showing reasonable agreement with adjacent shading and CO2 fertilization studies. It was concluded that anthesis thinning, because of its relative simplicity, is a useful technique. Implications for yield improvement in Yecora of the results provided by this technique are discussed.

Four annual experiments were carried out on the control of couch grass by undersowing barley and field beans with Italian ryegrass and broad red clover. Pieces of couch grass rhizome 15 cm long were planted in the plots in spring and lifted in late autumn. Couch grass grew more under beans than under barley. Undersowing lessened the growth by a factor of two under barley and by a larger factor under beans, although the total amount of couch lifted from under beans was greater than under barley. Undersowing spring cereals can appreciably retard the spread of couch grass and prevent its rapid spread after harvest if cultivation or spraying is delayed.

Administration of ACTH to Finnish Landrace × Dorset Horn ewes during days 13–16 of the pre-mating oestrous cycle resulted in a significant delay in the onset of oestrus (mean cycle length of 17·5 ± 0·39 days as compared with 16·4 ± 0·61 in a control group). Subsequent ovulation rate was significantly higher than in the control group (3·12 and 2·44 ova/ewe respectively).

In a second experiment with Scottish Blackface ewes the effects on time of onset of oestrus and on ovulation rate of environmental stress, ACTH, cortisone acetate or metyrapone were studied. Environmental stress and administration of ACTH through the oestrous period significantly reduced ovulation rate (1·54 and 1·39 ova/ewe respectively) compared with the control (1·96). ACTH administered during days 13–16, as in the first experiment, and cortisone acetate administered through oestrus had no effect on ovulation rate (1·88 and 1·86 respectively). Administration of metyrapone reduced ovulation rate to 1·70 but this was not significantly different from the control. Both forms of ACTH treatment resulted in a delay in onset of oestrus but environmental stress had no effect on this character.

Levels of plasma luteinizing hormone (LH) were estimated on representative animals. These showed that ACTH and environmental stress influenced the pattern of the ovulatory surge of LH in respect of time of onset of the surge relative to onset of oestrus, duration of surge and peak levels.

In two separate experiments, ewes of the Scottish Blackface breed were subjected to management stresses in the first or second 10-day period after mating or were treated with daily injections of 60 i.u. ACTH for the whole 20-day period. Embryonic loss is defined as the number of corpora lutea not represented by viable embryos at slaughter. Examinations were carried out between 26 and 32 days post-mating. In the two experiments combined, losses were 29·9, 29·0, 37·7 and 17·2%, respectively, for ewes subjected to environmental stress during days 1–10, during days 11–20, treated with ACTH or kept as unstressed control. Expressed as the percentage of ewes in which partial or complete loss of shed ova occurred, the results were 49·0, 46·0, 54·3 and 25·5%, respectively. Losses were significantly higher in all treated groups than in the unstressed control.

During four winter seasons eight spacing and density experiments were made under irrigated high fertility conditions in north-west Mexico (latitude 27° N). Experiments included various Triticum aestivum and T. durum genotypes of spring habit, short stature derived from Norin 10 genes, and contrasting plant type. Measurements included dry-matter production, photosynthetic area index, and light interception during one experiment, total dry matter at maturity in most others and grain yield and its numerical components in all experiments.

Grain yield and most other crop characters were unaffected by row spacings within the range 10–45 cm interrow width. The optimal seeding density for maximum grain yield was 40–100 kg/ha (80–200 plants/m2). Yield reductions at lower densities (20, 25 kg/ha) were slight and accompanied by reduced total dry-matter production. Yield reductions at higher densities (160–300 kg/ha) were also slight and were associated with more spikes/m2 but fewer grains/m2 and reduced harvest index. It is suggested that lower than normal preanthesis solar radiation or weather conditions leading to lodging can magnify these yield depressions at higher densities.

Measurements showed rapid approach of crops to 95% light interception, reached even at a density of 50 kg/ha within 50 days of seeding. It is suggested that provided this occurs before the beginning of substantial dry-matter accumulation in the growing spikes (60 days after seeding) there will be no loss of grain yield with reduced seeding density. Results point to a ceiling photosynthetic area index for maximum crop growth rate although there was a tendency for rates to fall at very high indices (> 9). This tendency was associated with very high density, high maximum numbers of shoots, poor survival of shoots to give spikes (< 30%) and reduced number of grains/m3;. The relatively low optimal densities seen here may be characteristic of genotypes derived from Norin 10.

Genotype × spacing, genotype × density and spacing × density interactions were generally non-significant and always small. There was a tendency for the presence of non-erect leaves or branched spikes to reduce the optimal density, but large differences in tillering capacity had no influence. Differences in lodging susceptibility can however lead to substantial genotype x density interactions.

Examples of response surfaces for pairs of nutrients and results of 41 multi-level experiments with N only were used to compare the goodness-of-fit of polynomial, inverse polynomial, exponential and intersecting-straight-lines models.

Whereas no one model fitted best at every site, many results were well represented by two intersecting straight lines and on average, this model had the least residual mean square. Of 17 experiments with spring barley in south western England the few results best represented by smooth curves were from crops much affected by leaf diseases.

Fertilizer response was poorly represented by models without a falling asymptote, like the simple exponential and inverse linear. Study of residuals after fitting the quadratic showed that this widely used model consistently over-estimated both the amount of fertilizer needed for maximum yield and the yield loss when too much fertilizer was given.

When fitted to the mean yields of each nitrogen treatment, most models had residual mean squares equal to or less than the error mean square, repeating a result obtained at Rothamsted as early as 1927. We question the validity of some well-known evidence for block and treatment additivity.

For 12 experiments in 1970, between-site differences in the parameter values of the two straight lines representing grain yield were related to leaf area at ear emergence.