The Journal of Agricultural Science

Recently developed regressional techniques suggest that some of the problems posed by genotype-environment interactions may be solved. These techniques have been applied, therefore, to data from an extensive trial involving perennial ryegrass (Lolium perenne) varieties grown under various treatments at several locations throughout Great Britain. Analysis reveals that the relative performance of the varieties depends mainly upon whether they are grown as spaced plants or swards. Even though the genotype-environment interactions are large, much of their effects could be reduced to an essentially linear scale. These results are considered in the light of the known characteristics of the varieties concerned, whilst the wider implications are also discussed.

Twenty-seven sheep feeding trials were carried out on freshly cut Andropogon gayanus, fed daily during three wet seasons, 1966–8.

In each year, dry-matter voluntary intake and digestibility were highest during the first part of the growing season and declined with advancing maturity of the herbage. However, at the time of ear emergence in early October, there were small increases in voluntary intake of dry-matter and crude protein digestibility, although further decreases followed after mid-October.

Determinations of in vitro digestibility indicated that elongating stems during early October were at least as digestible as leaves.

Seasonal changes in crude protein digestibility were much greater than changes in dry-matter digestibility; seasonal changes in feed intake were intermediate. Voluntary feed intake was related to dry-matter digestibility.

Voluntary feed intake and digestible crude protein were both closely related to the crude protein content of the forage. It was concluded that crude protein is a useful criterion for predicting nutritive value.

Data from the experiment described in the first paper of this series were subjected to regression analysis. Lean meat and live-weight gain response to both lysine and crude protein intake was almost linear in period 1 but in later periods little response was evident. Response of live-weight gain to increased energy was continuous though diminishing in the later periods. In periods 1 and 2 lean gains showed positive though diminishing response to energy.

In a multivariate regression analysis it was clear especially for younger pigs that the response to dietary lysine for both live-weight and lean meat gains is influenced by the dietary level of crude protein and the energy intake. Separate multivariate analysis for each sex indicated that females are more efficient converters of lysine and energy into live weight and lean meat. Also an interaction between sex and dietary lysine level was suggested.

A methodology for an economic analysis taking the joint objectives of total cost of production, speed of grain and quality of gain is established as a basis for design of high profit system of production.

The live-weight gains of male and female lambs were measured during growth from 16 to 40 kg live weight when five cereal-based diets varying in dietary crude protein concentration (from 10 to 20%) were given at three levels in a 5×3×2 factorial design.

Males grew faster than females by about 15% (P < 0·001); this difference increased with age and with the amount of feed given (P < 0·05). Growth rate responded linearly to increase in feeding level (P < 0·001) and curvilinearly to increase in dietary protein concentration (P < 0·01). A significant interaction (P < 0·01) occurred whereby growth increased with higher protein concentrations as feeding level increased.

At the highest feeding level (near ad libitum) the results suggest that the optimum dietary crude protein concentration for growth was about 17·5, 15·0, 12·5 and 12·5% at body weights of 20, 25, 30 and 35 kg respectively. The overall optimum dietary crude protein concentration for growth between 16 and 40 kg body weight was about 17·0, 15·0 and 11·0% when the mean digestible energy intake was 3·0, 2·6 and 2·1 Mcal/day.

The results agree reasonably well with estimates of protein requirements for lambs given by the Agricultural Research Council (1965).

The efficiency of the utilization of the energy of casein in adult fowls at different feeding levels both for maintenance and for lipogenesis, was studied. The casein was utilized for maintenance with an efficiency of 66·6 ± 1·3% of the gross energy and of 83·7 ± 1·4% of the metabolizable energy. For lipogenesis efficiencies were 51·6 ± 2·15%, and 61·3 ± 2·6% respectively.

Two varieties of irrigated rice N.P. 130 (a tall variety) and Taichung Native-1 (a dwarf variety of indica rice) were compared in 1966 and 1967 when grown with several amounts of sodium nitrate and ammonium sulphate. Comparisons of ammonium sulphate treated with ‘N-Serve’ or with ‘AM’ against untreated ammonium sulphate were made.

Applying more than 60–80 kg N/ha did not increase the number of panicles per m with N.P. 130 but lowered the number of spikelets and filled grain percentage. Up to 120 kg N/ha, panicle number and the number of spikelets of Taichung Native-1 increased significantly but the filled grain percentage was not increased; yield of grain responded significantly up to 120 kg N/ha. Taichung Native-1 yielded more grain and less straw than N.P. 130. Increasing the nitrogen applied significantly increased the yield of straw from both varieties.

Ammonium sulphate increased panicle number and reduced the percentage of filled grain compared to sodium nitrate but the increase in panicle number was higher with Taichung Native-1 than with N.P. 130; as a result ammonium sulphate consistently increased the grain yield of Taichung Native-1. When the ammonium sulphate was treated with either inhibitor it gave more grain of Taichung Native-1 than untreated fertilizer; treated fertilizer gave more straw of both varieties.

1. In an experiment with ninety-nine lambs the effects on the body composition of male and female lambs were examined when five diets containing different concentrations of crude protein (in the range 10–20%) were given at three levels of feeding and lambs were slaughtered at two live weights (27·5 and 40 kg).

2. With lambs slaughtered at 27·5 kg there were significant increases in the rate of both nitrogen and fat retention with increases in levels of feeding. There were also linear increases in the rate of protein deposition and decreases in fat deposition with increases in the concentration of crude protein. This effect was particularly marked at the high level of feeding.

3. With lambs slaughtered at 40 kg live weight there were also linear increases in fat and in nitrogen deposition with increasing feeding level but the effect of increasing the protein concentration on increases in nitrogen retention departed from linearity.

4. While at 27·5 kg there were no significant effects of feeding level on nitrogen and ether-extract content of the bodies at slaughter, with animals slaughtered at 40 kg there was a significant linear decrease in ether-extract content with increasing feeding level and a corresponding linear increase in nitrogen content with increased level of feeding.

5. Male lambs deposited more nitrogen and less fat than females. This was true of both rate of deposition and of carcass composition at 40 kg live weight.

Bulk harvests of primary growth were made at intervals during the spring and summer of 2 consecutive years from an ungrazed area of lucerne (Medicago sativa) and the digestibility by sheep was measured. At the same times, shoots of known length representative of the sward were collected, divided into 7·5 cm lengths, and separated into leaf and stem for the determination of ash, nitrogen, cellulose, and in vitro organic matter digestibility.

Digestibility and leaf: stem ratio of the bulk harvests decreased and dry matter increased as the season advanced. Leaf composition changed little, but stem increased in lignin and decreased in ash and in vitro digestibility. In vitro digestibility and leaf: stem ratio were highly correlated with sheep digestibility.

Chemical composition of shoots of known heights was similar to that of the bulk harvests. Leaf weights/shoot tended to increase linearly with height of shoot, whereas stem increased nearly as the square of the height. Dry matter of leaf and stem were inversely related to leaf:stem ratio. Stem digestibility decreased with leaf:stem ratio and with increase in shoot height.

Chemical components of shoot fractions were associated with each other and were largely dependent on the distance from the top of the shoot at which they had been taken, irrespective of shoot height or time of year. Bottom stems had lower ash, nitrogen and digestibility but higher cellulose, dry weight and dry matter than top stems, whose composition approached that of leaves. Leaf fractions showed little consistent trend with position on the shoot.

Some of the difficulties in studying changes in plant composition under field conditions are discussed.

In two feeding experiments the effects of different sources of protein on nitrogen utilization were studied. Three protein sources were examined in both experiments; (a) soya-bean oil meal; (b) lucerne hay or vetch oat hay, and (c) a mixture of soyabean oil meal and hay. The first experiment was carried out with nine adult rams and the second with twelve growing rams.

The results of N balance experiments showed clearly the superiority of the hay ration over the soya-bean meal ration. N retention when the mixed ration was given was greater than on hay alone. No parallel trend was found between the protein digestibility of any of the diets and its efficiency for N retention.

With adult sheep a negative correlation between the extent of N retention and blood urea concentration was found; this agreed with our earlier findings. No such negative correlation was found with growing animals.

The highest concentrations of the various forms of soluble N compounds such as soluble a-amino N, diffusible N and diffusible peptide N were found in the rumen liquor taken from the sheep kept on the hay diet. The lowest amounts were observed when the soya-bean oil meal diet was given and intermediate amounts with the mixed diet. These trend sappeared to be related to the NPN (non-protein nitrogen) content of the diet, but the amounts of soluble N compounds present in rumen liquor were not directly correlated with the extent of N retention.

Concentrations of VFA present in rumen liquor depended primarily on the carbohydrate composition in the diet. The type of protein ingested exerted some influence, particularly on the concentrations of caproic and iso-caproic acid. The ratios of concentrations of the individual fatty acids in blood to the concentrations of the same acids in rumen liquor were found to increase with lengthening of the fatty acid chain.

Maturation studies were carried out on 241 cattle oocytes. Media used were bovine follicular fluid and growth medium. Both media supported a resumption of meiosis in ca. 80 % of cases, but growth medium proved superior in terms of final maturation.

Fertilization attempts were carried out on 564 oocytes following culture (27–32 h) in growth medium. In vitro fertilization attempts (107 oocytes) and in vivo fertilization attempts (82 oocytes) in the rabbit yielded negative results. Following in vivo fertilization attempts in the ewe, ova containing pronuclei as well as apparently normal cleaved ova were recovered.

Modifications to the New Zealand meter (Campbell, Phillips & O'Reilly, 1962) are briefly described and the results obtained with the modified meters presented.

Water was shown to have the dominant influence on meter reading—dessication of the plant material reduced readings, application of water increased readings.

The relation between yield of water (W) and meter reading (MR) depended on the circuitry of the meter used and was similar to that between capacitance change induced by a variable capacitor and meter reading. With the new meters this relation was strongly linear. For individual species the simple correlation of W on MR was generally superior to that for weight of dry matter. Regression coefficients for weight of water on meter reading differed between species and/or days but were correlated with the relative water content of the species (RW). Inclusion of RW in addition to that of W increased the variation accounted for (V%) when all species were included from 76·0 to 86·9%. However, the multiple regression did not eliminate bias. For individual species, there was no benefit in including the RW term.

The inclusion of height of herbage (H) in a multiple regression equation with weight of water greatly improved the V% when compared with the regression of meter reading on weight of water alone for oats dressed with three levels of nitrogen. The V% increased from 71·3% for W alone to 90·0% when W, H and H2 were included in the regression. Further testing using height of herbage over a wide range of species will be required to confirm this improvement for use in an overall regression.

Emphasis is placed on the very strong linear relation between yield (of water or dry matter) and meter reading within any one pasture at any one time. Use is made of this relation in the presentation of a new method, which overcomes the bias associated with prediction equations, for estimating pasture yields. This entails establishing the mean reading for any pasture and cutting for yield estimation three samples having such a mean. The limitations of the method for use with mixtures is discussed.

A field experiment in each year 1963 to 1967 in Suffolk, England, tested the effect on yield of sugar beet of sowing dates ranging from 13 March to 11 May and harvest dates ranging from 20 September to 8 December. Sowings in March or early April gave similar yields of sugar but with later sowings yield decreased progressively faster. The relationship is represented by the equation

y = 106·3 (±2·24) + 0·212 (±0·184)x-0·009 (±0·003)x2,

y = yield as percentage of mean (60·1 cwt/acre of sugar), x = number of days after 12 March.

Delayed harvest increased sugar yield; the relationship is represented by the equation

y = 80·22 (±1·51) + 0·836 (±0·09)z-0·006 (±0·001)z2,

z = number of days after 19 September. This represents an increase of 0·247 cwt/acre/day of sugar in October and 0·083 cwt/acre/day in November, equivalent to about 2½ tons/acre of roots in October and 1 ton/acre in November.

The effect on sugar yield of different lengths of growing period, which ranged from 138 to 271 days, is represented by the equation

y = 38·7 (±26·6) + 1·045 (±0·267)a-0·0017 (±0·0007)a2,

a = number of days between sowing and harvest.

On average, sowing date had negligible effect on sugar percentage at harvest. In 3 years sugar percentage increased after the first harvest to a maximum and then decreased; in 2 years it decreased with consecutive harvests.

The rumen was found to be the main site of metabolism of biuret in sheep adapted to this compound. Some of the biuret which passed into the lower gut could be broken down there but this could account for the breakdown of less than 1 g/day when 15 g was fed. Biuret injected into the jugular vein or entering the bloodstream from the rumen or intestine was excreted unchanged in the urine. Sheep excreted 9–21% of biuret fed in a solid form and 37–39% when it was dosed as an aqueous solution. Sheep not adapted to biuret failed to retain any introduced into the rumen, duodenum or jugular vein.

Nine experiments in the East of Scotland were completed in the years 1966, 1967 and 1968 to test the response of potatoes grown for ware to five levels of a 15½–15½–21 fertilizer applied at rates of 0, 4, 8, 12, 16 and 20 cwt per acre.

1. Mean yield of ware tubers increased up to the highest fertilizer level.

2. Mean yield of small tubers increased only up to the 8 cwt per acre level and then declined.

3. Dry matter content of ware tubers decreased with increasing levels of fertilizer.

4. Increasing fertilizer levels lowered soil pH. At one site this was associated with a significant reduction of common scab on the tubers.

Sweating rate and the electrolyte content of secretions from the skin of cattle have been measured on five Bos taurus and five B. indicus cross-bred cows by absorbing secretions into filter papers under polythene disks applied to shaved skin areas on the shoulder, sacral and lumbar regions for 5 min. The increase in weight of the filter paper was taken as a measure of sweating rate and the distilled water eluate from the filter paper was analysed for sodium and potassium. Animals were exposed for 4 h to air temperatures of 20–45 °C at 30% r.h., and for 5–7 h to air temperatures of 40 and 45 °C at 40% r.h. Estimated sweating rates were low by comparison with previously reported values, probably due to rising levels of humidity under the polythene disks during exposure to the skin. B. indicus cross-bred cows had higher sweating rates than B. taurus cows at high air temperatures but the difference between the groups was not significant statistically. Sweating rates were generally highest on the shoulder and lowest on the lumbar region.

The amounts of sodium and potassium recovered from filter papers were small and very variable at low air temperatures but increased significantly with air temperature (P < 0·01). No significant differences in the amounts of electrolyte recovered from filter papers were recorded between the species groups or between different sites of collection. The secretions from cattle skin at high ambient temperatures contained at least four to five times as much potassium as sodium. Total sodium and potassium loss through the skin of these experimental animals at the highest ambient temperatures was estimated to be no more than 1–3 % of the sodium and potassium intake in the feed. Absorbing sweat on to filter paper as a method of measuring sweating rate and sweat composition is rather less satisfactory for use with cattle than with man.

Differences between Sahiwals and Jerseys and between summer and winter in the quantities of lipids on the skin surface and on hair were studied.

Sahiwals had significantly greater quantities of lipids on both the skin surface and hair in summer and in winter. In both breeds, there were significantly greater amounts of lipids on the skin surface and on hair in winter. In both breeds a significantly greater quantity of lipids was present on hair than on skin in winter but not in summer.

Jerseys had greater weight of hair/unit area of skin than Sahiwals in both summer and winter. There was no significant correlation between the amount of lipids on hair and the weight of hair.

An experiment was carried out in which each of thirty-six ewes was offered one of six dietary treatments from the 30th to the 140th day of gestation. The treatments supplied two levels of metabolizable energy (ME) intake (128 and 116 kcal/kg W0.75/day) each at three patterns of protein intake, namely equal increments at 22 day intervals, equal decrements at the same intervals and a constant daily intake throughout. The protein intake patterns supplied a mean daily intake of digestible crude protein of 3·74 g/kg W0.75/day. Nitrogen balances were carried out on four ewes from each dietary treatment during the last 10 days of each 22 day period.

Thirteen ewes randomly distributed on all dietary treatments were barren. Mean rate of live-weight gain in the pregnant ewes was 123 g/day compared with 66 g/day in the non-pregnant ewes and was not significantly affected by treatments. Lamb birth weights adjusted to the basis of twins were 3·83, 4·02 and 3·56 kg for the increasing, constant and decreasing patterns of protein intake respectively.

Within patterns of protein intake there were no significant differences in nitrogen retention between pregnant and non-pregnant ewes until after 90 days of gestation. Thereafter retention increased in the pregnant ewes offered the constant and increasing patterns of crude protein and remained constant in the non-pregnant ewes. During the same period the retention of both pregnant and non-pregnant ewes on the decreasing pattern decreased.

The effects of both energy and protein intake on nitrogen retention in early and late pregnancy were assessed by multiple regression analysis. The results are discussed in relation to other recent findings.

Pigs were fed either at 3, 6 or 12 hourly intervals to a scale based on live weight from 21 kg to slaughter at 79 kg live weight. There was no effect of treatment on daily gain, feed efficiency or killing out percentage. Backfat thickness at C and K was increased by four times-a-day feeding compared with feeding twice or eight times/day, but this increase in fatness was not confirmed by sample joint dissection.

The digestibility of some fats (beef tallow, HEF, herring oil, soyabean oil and maize oil) were studied in balance trials with sheep which had been maintained on the diets for a 35 day period of acclimatization to establish ‘steady state’ conditions. Corrected digestibility coefficients for the individual fats were calculated from a knowledge of the intake and output of lipid on a low-fat basal ration. The difference between the intake of fat and the corrected lipid output (total lipid output minus lipid output on basal ration) expressed as a percentage gave the corrected digestibility coefficients.

It was found that the corrected digestibility coefficients for beef tallow was 85%; HEF, 74%; herring oil, 84%; soyabean oil, 83%; and maize oil, 70%. The values for beef tallow and HEF were comparable to similar estimates in pigs. The digestibility of the highly unsaturated soyabean and maize oil were markedly lower in comparison with non-ruminant species. Extensive hydrogenation of dietary unsaturated fatty acids was confirmed. For soyabean and maize oil the corrected output of stearic acid exceeded the dietary intake.

Surface soils (0–15 cm) from a field experiment on a sandy podzol were used to evaluate the residues from basic slag, Gafsa rock phosphate and superphosphate (using ryegrass in pots), and to measure the solubility and buffer capacity during cropping, and the distribution of the residual phosphate in different particle-size fractions of the soil.

After applying P at about 3·6 g P/m2/year for 6 years, residues of the three P fertilizers were equivalent but after applying more P (9–11 g P/m2/year) for a further 8 years, basic slag and rock phosphate gave larger labile P values, were more effective for ryegrass and buffered soils more than residues of superphosphate. The percentage recoveries of P from the surface soils were approximately 7, 40 and 65 % of the total applied in 8 years as superphosphate, basic slag and rock phosphate respectively.

Residues from basic slag (mainly 75–20 μm particles) accumulated in the < 20 μm fraction of the soil. Rock phosphate (mainly 200–20 μm particles) dissolved more slowly and about one-half of the material > 20 μm remained in the sand fractions of the soil.

Residues from rock phosphate maintained values of ½pCa + pH2PO4 and pH–½pCa consistent with the solubility of fluorapatite, even when P was removed from the soil by exhaustive cropping with ryegrass. Residues from basic slag were more soluble than hydroxyapatite.