Frequently in near-infrared reflectance spectroscopy, a calibration is developed using very restricted data sets, e.g. material from one season, a small area or of a limited type: consequently, the predictions may have limited validity. This paper describes the use of both restricted and wide calibration sets for the prediction of crude protein in grass, silage and hay. Results show that predictions from the wider calibration sets are often as good as or better than predictions from restricted calibration sets. Therefore the use of wide calibration sets should be considered much more frequently in near-infrared reflectance.

Two experiments were conducted to examine the effects of supplementation of a control diet of barley grain and barley straw containing 4 mg copper (Cu)/kg dry matter (D.M.) either with 5 mg molybdenum (Mo)/kg D.M. or with 500 or 800 mg iron (Fe)/kg D.M. on puberty, fertility and oestrous cycles of cattle. Puberty occurred normally in control, Fesupplemented and control animals on a restricted intake whereas it was delayed by 12 and 8 weeks respectively by Mo supplementation. This effect of Mo was not due to the low Cu status since this was equally low in the Fe-supplemented animals, nor was it due to the reduced growth rate since puberty occurred normally in control animals that had a similar live-weight gain. A significant reduction in the pulsatile release of luteinizing hormone was observed within 11 weeks of the Mo supplementation and before any of the other clinical signs were evident, suggesting that Mo may be affecting puberty by altering the release of luteinizing hormone either directly or indirectly.

Mo supplementation significantly reduced the percentage conception rate to 12–33% compared with 57–80% in control and Fe-supplemented animals. This effect was not dependent on the rate of live-weight gain which was standardized across the different treatments at approximately 0·6 kg/day. Within 12 weeks of the replacement of dietary Fe by Mo a lower conception rate occurred; replacing dietary Mo by Fe led to a normal conception rate within 12 weeks without any accompanying changes in Cu status or in the rate of live-weight gain. The plasma Mo concentrations, however, changed significantly during these alterations in dietary supplementation. The pre-ovulatory peak height of luteinizing hormone was significantly lower in animals on the Mo-supplemented diet compared with control and Fe-supplemented animals, but the administration of LHRH did not alter the conception rate.

More Mo-supplemented animals failed to ovulate following prostaglandin induced synchronization in comparison with the other treatments, and by the 84th week a significantly greater number of Mo-supplemented animals (12/18) had become anoestrous compared with the other groups (2/30). Cu repletion of these anoestrous Mo animals for a period of 20 weeks did not result in resumption of normal oestrous cycles, but ovulation and oestrus were induced by progesterone and LHRH treatment. Results in the latter part of the study indicated that Mo caused superovulation.

These data show that Mo supplementation delayed the onset of puberty, decreased the conception rate and caused anovulation and anoestrus in cattle without accompanying changes in Cu status or in live-weight gain. It is suggested that these effects of Mo are associated with a decreased release of luteinizing hormone that might be due to an altered ovarian steroid secretion.

Ninety-nine lactating British Friesian cows, in their second and subsequent lactations, were divided into two groups receiving ad libitum grass silage. The cows in group A were offered silage individually through Calan Broadbent gates while those in group B were given 24 h/day access to a self-feed silage clamp. The crude protein (CP), acid-detergent fibre (ADF), neutral-detergent fibre (NDF), in vitro digestible organic matter and estimated metabolizable energy (ME) values in the D.M. of the silage were 140, 373, 584, 610g/kg D.M. and 9·7 MJ/kg D.M. respectively. The cows in each group received either 11kg/day fresh weight of concentrate S in which the principal energy source was cereal starch or 9 kg/day fresh weight of concentrate F containing mixed high quality fibre sources plus 2 kg/day fresh weight of concentrate S. The CP, ADF, NDF, starch and sugar concentrations in concentrates S and F were 203, 71, 204, 444, and 44 and 202, 147, 329, 116 and 149g/kg D.M., respectively.

The only significant difference produced by method of silage allocation was that groupfed cows gained more live weight (P < 0·01) than those fed individually.

Over the whole experimental period the mean silage D.M. intake of cows fed concentrate F in group A was 0–5 kg/day higher than those fed concentrate S. However, in early lactation the benefit to silage intake in favour of concentrate F was 1 kg D.M./day.

In group A estimated intakes derived from calculated ME inputs and outputs were markedly lower than the observed intakes, but were similar to correspondingly derived estimates for group B.

Although concentrate type did not affect milk yield, concentrate F was associated with a higher fat concentration and yield (P <0–05) but a lower protein concentration (P < 0·001) and a slightly lower yield.

The in vivo digestibility coefficients for D.M. and organic matter determined in lactating dairy cows given concentrates S and F respectively were similar (0·724 v. 0·716 and 0·749 v. 0·742) but that for ADF was significantly (P < 0·001) higher (0·565 v. 0·673) in the ration containing concentrate F than in that containing concentrate S.

Estimates of the mean efficiency of utilization of the production ME made on a weekly basis were unaffected by concentrate type.

All combinations of four harvesting treatments and two sowing dates were comparedin each of 3 years in field experiments near Nicosia. In two of the years, different levels of applied nitrogen were compared. The harvesting treatments were: (1) a milk-stage cut,(2) a boot-stage cut and a regrowth cut, (3) a grazing-stage cut and a regrowth cut, and(4) three grazing-stage cuts and a regrowth cut. The sowing dates were (1) normal (12 November) and (2) early (15 October), with sufficient irrigation to ensure establishment. There was almost no response to applied nitrogen, evidently because of high reserves of available nitrogen in the soil. Early sowing reduced by 20 days the time taken to reach the stage at which the first node appeared, but increased, by 36 days, the length of time between the first node and the milk stage. Early sowing increased yield when the firstcut was at the grazing stage, but reduced yield when the first cut was at the boot or milk stage. At the grazing stage, the proportion of green leaf blade and the concentrations of nitrogen and nitrate-N were relatively high (means 74, 4·0 and 0·13% respectively) and digestibility was moderate (mean D-value 61). By the boot stage, green leaf, nitrogen and nitrate-N, but not digestibility, had declined (means 21, 1·8, 0·02 and 63 respectively). By the milk stage, green leaf, nitrogen and digestibility, but not nitrate-N, had declined further (means 3, 1·5, 48 and 0·05 respectively). Fully dead leaf blades were 28 units less digestible and 3·2 percentage units lower in nitrogen concentration than emerging leaves. Half dead leaf blades were higher in nitrate-N than fully expanded, fully green blades. 'Stem’ declined greatly in nitrogen concentration, but relatively little in digestibility, between the grazing and the boot stage.

Crops cut once or three times at a grazing stage recovered well, although theretended to be some reduction in the number of tillers compared with an undefoliated crop.

The suitability of stained food and [103Ru]phenanthroline-treated food as markers for measuring dietary retention time was examined in sheep. The digestibility of oat husk, assessed by incubation of samples contained in nylon bags in the rumen, was greatly depressed by staining with brilliant green, and slightly depressed by basic fuschsin. The latter effect appeared largely to be due to boiling rather than the stain itself.

When milled hay treated with both basic fuschsin and [103Ru]phenanthroline was introduced into the rumen, the mean retention time estimated with the stain was about 40% longer than that estimated with [103Ru]phenanthroline.

The digestion and retention times of milled hay diets were first studied in three sheep fitted with rumen and abomasal cannulae. The hay was milled through 5, 20 or 40 mm screens. Solubility of dry matter (D.M.) in water and rate of loss of D.M. from hay in nylon bags suspended in the rumen increased with fineness of grinding. Apparent digestibility of D.M. within the gut tended to fall with fineness of grinding, as did mean retention time (MRT) of a particulate marker, [103Ru]phenanthroline ([103Ru]phen), within the gut and of [103Ru]phen and a liquid marker, [51Cr]EDTA, within the rumen. Dry-sieved abomasal particles were much smaller than particles of any of the hay preparations, most passing through a 0·6 mm screen.

A second experiment was made on four other sheep, fitted with rumen cannulae and given a similar hay, chopped to 50 mm or milled through 2, 5 or 20 mm screens. Apparent digestibility of D.M. within the gut fell significantly with fineness of grinding, from 55% with the chopped hay to 46% with the 2 mm hay, as did MRT, from 45 to 40 h respectively. Rumen pH at 1 h after feeding decreased with fineness of grinding.

In a third experiment, using the four sheep given chopped or 2 mm hay, retention times of particulate matter and of fluid in the reticulo-rumen and in the caecum-colon were measured. [103Ru]phen and [51Cr]EDTA were dosed together into either the rumen or the terminal ileum, and the rates of decline of marker concentration in rumen samples or faeces respectively were used to calculate the appropriate half times(t½). In the rumen particulate and fluid markers showed significantly shorter t½ values with the 2 mm hay (20·8 and 12·4 h, respectively) than with the chopped hay (25·9 and 15·6 h, respectively) without a significant difference in rumen outflow (0·437 and 0·492 l/h, respectively). In the caecum-colon the corresponding t½ values were 6·4 and 6·3 h, and 7·1 and 6·6 h, there being no significant difference either between markers or between diets.

Samples of chopped hay and of 20, 5 and 2 mm hays were suspended in bags in the rumen of sheep given either chopped or 2 mm hay to eat. The fine hays lost D.M. more rapidly than the coarse hays but D.M. loss was slower when the sheep's diet was 2 mm hay that when it was chopped hay.

It is concluded that three factors accounted for the effect of milling on digestibility: retention time in the reticulo-rumen, D.M. degradation rate in the rumen, and the potential digestibility of the hay preparation (asymptotic D.M. loss as measured in the rumen bag studies). It appeared that fine grinding reduced the digestibility of the hay because the effects of shortened MRT in reducing duration of digestion and of lowered rumen pH in depressing cellulolysis outweighed the effects of grinding in accelerating digestion and increasing potential digestibility within the rumen.

An experiment to measure the variation in the phenological and apical development ofwinter wheat (cv. Avalon) in England and Scotland is described. Ten sites which ranged from Aberdeen (57·2° N), the most northerly, to Newton Abbot (50·6° N), the most southerly, were included in the survey, and at each site seed was hand-sown in mid-September, October and November 1983. Developmental stages and sampling procedures were precisely defined to ensure uniformity in scoring by the observers at each site. Temperatures during the growing season were in line with the long-term means, though spring was cooler at all sites and summer warmer at most. The range of monthly-mean temperatures between sites was about the same as the difference between consecutive months. The method of analysis of development rates and durations was in terms of thermal time, modified by sensitivity to photoperiod and a vernalization requirement that slowed early development until a number of days of low temperatures had been experienced. In general, crops at northern sites developed more slowly than those in the south and particularly the south-west of England. There was less variation in the timing of apical stages for later sowings. Developmental rates responded linearly to temperature and photoperiod, with the base temperature increasing for later phases of development. The effect of photoperiod in modifying the rate of development was apparent for all developmental phases from emergence to anthesis, longer days accelerating development, but there was no effect on the duration of the grain-filling period. Vernalization exerted its effect solely within the phase from emergence to double ridge, and had a major influence on the variation between sites only for the first sowing.

Seventy-one small-framed Angus and seventy-four Holstein steers were chosen to represent small and large mature types in a study designed to provide quantitative information on the effect of breed, diet and housing type on composition of carcasses from cattle slaughtered serially over a wide weight range. Within breed, steers were individually fed ad libitum either a maize-grain or a maize-silage diet and assigned randomly to one of two housing types. The effects of treatments on response variables with increasing carcass weight were tested by covariance analysis.

The rate of change of soft tissue chemical composition with increasing carcass weight was not influenced by diet, breed mature size or housing type except for ash which decreased at a faster rate in Angus than in Holstein steers. Percentages of water, protein and ash decreased while lipid increased with increasing carcass weight. At any carcass weight, Angus had more lipid and less water in soft tissue. Percentages of soft tissue protein and ash were affected by diet within breed. The energy content of soft tissue increased with weight and was affected only by breed. Angus soft tissue had more energy than that of Holsteins. Similar results were obtained when the entire carcass was analysed.

Cattle fed grain had slightly more water and less ash in bone than cattle fed silage. Angus steers had more protein and ash in bone than Holsteins. Although there was a diet × breed × housing type interaction for lipid content of bone, Holstein bone had higher lipid content than Angus bone for either diet or type of housing. All percentages of chemical components in bone, except water, increased with increasing carcass weight.

The experiment showed that large mature-size breeds have the ability to transform a high-energy diet into protein for human consumption. The percentage changes in ash, protein and lipid with increasing carcass weight appeared to follow a pattern of development similar to bone, muscle and fat, respectively, suggesting that Hammond's theory of differential growth holds also for chemical components of bovine carcasses. Bone chemical components, however, had a slightly different developmental pattern. The quantitative information on bone presented herein could be used to exploit bone as a source of human food, dietary supplements and products for bone remodelling surgery.

Kinetic studies of Ca metabolism, using 45CaCl32, were carried out on two mature red deer stags during the period of maximum rate of Ca deposition in the antlers. They were offered green-feed oats to provide energy for maintenance; the diet provided approximately 42mg Ca/kg W per day.

Ca appeared to be irreversibly lost from the circulation into the antlers and could be treated for kinetic purposes in the same way as loss of Ca from the body in milk. The size of the rapidly exchangeable Ca pool in the body, excluding the antlers, was 0·21 g/kg W, similar to estimates for lactating cattle. Rates of Ca deposition in the antlers of the two stags, calculated from the model of Ca metabolism, were 58·4 and 38·6 mg Ca/kg W per day. Net endogenous loss was 6–7 mg Ca/kg W per day, much lower than estimates available for other ruminant species. Only 25–40% of Ca requirement or 11–24 mg Ca/kg W per day was derived from the diet, suggesting that the availability of Ca in greenfeed oats is much lower than current estimates for forages. Bones removed on completion of the experiment showed evidence of considerable skeletal demineralization.

In other stags subjected to stable Ca balances at the same stage of antler growth, while consuming ryegrass-white clover forage, 60–80% of Ca requirement was derived from the diet, with calculated rates of true absorption of 32–46 mg Ca/kg W per day. These rates of Ca absorption are low compared with values observed in other ruminants at times of high Ca demand such as during lactation.

A simple mathematical model of the distribution of potato tuber yield between size grades is presented. It has two parameters, μ, a measure of crop tuber size and σ, a measure of the spread of yield across size grades. The model is shown to be useful for the analysis of field experiments, the prediction and physiological study of tuber size distribution, and the economic analysis of management decisions. The model is applied to a variety of dataandthe practical control of tuber size discussed.

The initiation of leaf and spikelet primordia was studied at sites ranging in latitude from Newton Abbot (50·6°N) to Aberdeen (57·2°N) in crops sown in the middle of September, October and November 1983. The rate of primordium initiation tended to decrease from south to north but there were also marked differences between quite close sites.

The rate of leaf initiation increased with temperature but photoperiod had little effect; the rate of spikelet initiation was affected both by temperature and by photoperiod. There were differences in the total number of leaves initiated which were only partlyexplained by differences in vernalization.

Expressing leaf and spikelet initiation rates in terms of thermal and photo-thermal time respectively showed a constant rate of leaf initiation and a constant and more rapid rate of spikelet initiation.

The influences of the temperature during plant raising and the age of plants at transplanting on the time of hearting and head weight at maturity of the crisp lettuce variety Saladin R100 were determined in seven experiments between 1984 and 1986.

Plants raised at ambient temperatures and transplanted before the end of May produced heads which were heavier and matured later than those from plants raised at higher temperatures under glass. The influence of transplant age on head weight was much smaller than that of raising conditions and there were inconsistent effects on the time of crop maturity. When the head weights at maturity of all treatments were considered there were highly significant positive correlations with total solar radiation in the periods 7 and 10 days before 50% hearting occurred. Further examination of this effect revealed that heads heavier than 800 g were strongly associated with mean daily solar radiation exceeding 15MJ/m2 in the periods 7 and 14 days before 50% hearting. This suggests that breeding for stability of head weight from crop to crop is likely to be aided by selection for consistency of head weight under a range of light levels in the prehearting phase.

The coefficient of variation of the period from transplanting to maturity, measured in terms of various environmental characters, was lowest using effective day-degrees, which take account of different levels of solar radiation at the same mean temperature. This suggests that effective day-degrees may be the best indicator of the duration of crop growth and therefore of benefit in developing techniques to predict the time of crop maturity.

The effects of sowing date on leaf appearance, spikelet initiation and mainstem apical development in spring barley cv. Triumph were investigated in field experiments involving 13 sowings made during the period 22 February to 1 July over the years 1982–1985. Delaying sowing was associated with faster rates and shorter durations of leaf appearance and spikelet initiation and earlier attainment of all stages of apex development. The rates of leaf appearance and spikelet initiation were correlated with daylength at crop emergence although there was evidence that the latter process was also influenced by temperature. The derivation of three models relating the duration of developmental phases to temperature and/or daylength is described. In these models the attainment of successive stages of apical development was assumed to require the perception by the crop of a ‘threshold amount’ (THR) of accumulated temperature (THR(ΣT)) and/or daylength (THR(ΣPT), THR(ΣP)) above certain base values (TbPb). The base values of temperature and/or daylength for each phase were derived as those values which minimized the coefficient of variation of the amounts of accumulated temperature and/or daylength experienced by all 13 sowings. For various developmental phases the model based on temperature gave base values between 0 and 3 °C. The model based on daylength gave base values between 11 and 13 h. In both these models there was a highly significant correlation between observed and expected dates of attainment of various stages of apex development. It is suggested that the derived relationships between temperature, daylength and apical development could be used as an aid in the forward planning of crop management. The model based on temperature and daylength (photothermal time) gave no useful base values and it is concluded that more precise methods of relating development to these factors must be sought.

The effect of insect pollination on plant development and seed yield of winter oil-seed rape, cultivar Jet Neuf, was investigated by comparing plots caged with a honeybee colony, plots caged to exclude insects and plots uncaged and 'open-pollinated' by naturally occurring insects. Plants in the bee-pollinated plots finished flowering earlier, showed more advanced pod growth, and were shorter than those in the plots without bees. Pods from the plots with honeybees contained more seed post-flowering than those from plots without honeybees but the proportion of them that grew into mature seeds for harvest was determined by water availability during seed growth.

The quantitative partitioning of digestion between the stomachs, small intestine and large intestine was compared in kids and lambs offered a high quality meadow hay (organic-matter digestibility, 0·74; N, 23·9 g/kg D.M.) in amounts between maintenance and ad libitum.

Nutrient flow past the abomasum and terminal ileum and in the faeces was measured by reference to 51Cr-EDTA and 103Ru-phenanthroline. Digestion of D.M., neutral detergent fibre (NDF) and non-ammonia N (NAN) in the rumen (R), small intestine (SI) and large intestine (LI) was estimated from the regression of nutrient flow on nutrient intake.

No difference was observed between the species in digestion of D.M., NDF and NAN at any site. Approximately 0·60, 0·29 and 0·13 of the total in vivo D.M. digestion (0·74) occurredin the R, SI and LI, respectively, and approximately 0·94 of the total in vivo NDF digestion (0·79) occurred in the rumen of both species.

There was a gain of total N and NAN of 22 and 15%, respectively, in the rumen of both species whilst net disappearance of NAN in the small intestine and of N in the large intestine was approximately 0·62 and 0·27, respectively. All these characteristics varied slightly with level of feeding.

It is concluded that there may be little difference in the partitioning of digestion of D.M., NDF and NAN in the digestive tract of kids and lambs when consuming high quality forages.

Twenty-four Friesian cows were allocated to one of four diets for weeks 3–14 of lactation following 2 weeks on a common diet. The diets (kg air-dry feed/day) were 7–2 kg hay and 10·8 kg either starchy or fibrous concentrates (60S and 60F) or 3·5 kg hay and 14·0 kg either starchy or fibrous concentrates (80S and 80F). Rumen samples were taken by stomach tube in weeks 10 and 12 of lactation and the digestibility of the diets was measured with four cows per treatment during weeks 13 and 14. The digestibility of the same feeds was also measured in sheep at maintenance.

The principal carbohydrate constituents of the concentrates were barley, wheat and cassava in the starchy concentrates and citrus pulp, sugar-beet pulp and wheat feed in the fibrous concentrates. The concentrates were designed to have similar concentrations of metabolizable energy (ME) and the diets were planned to provide similar intakes of digestible energy and crude protein.

Milk yield and composition were very similar for treatments 60S and 60F. With the higher proportion of starchy concentrates (80S), milk yield was about 20% greater than on 60S, fat concentration fell severely but protein and lactose concentrations were unaffected. With the higher proportion of fibrous concentrates (80F), milk yield and the protein and lactose concentration were similar to values on 60F but fat concentration was lower, though not nearly so low as on 80S. Milk energy yield was reduced by the higher proportion of concentrates but was unaffected by type of carbohydrates. Live-weight changes were small.

In both the sheep, consuming at maintenance, and the lactating cows consuming at about 3 times maintenance, digestibility of dry matter, organic matter and energy was higher with the higher concentrate diets but was unaffected by type of concentrate. The digestibility of fibre was greater with the fibrous concentrates but the effect of level of concentrate inclusion was inconsistent. Digestibility coefficients were consistently lower for the lactating cows than for the sheep.

The proportion of acetic acid in the rumen volatile fatty acids in the cows was higher and the proportion of propionic acid was lower with the fibrous concentrates. The differences were much greater with the higher proportion of concentrates.

During weeks 15–22 of lactation the cows were reallocated to concentrate treatments and given hay ad libitum. Hay intake was about 1 kg/day higher with the fibrous concentrates but the difference was not significant. Hay intake fell by about 0·6 kg/kg concentrate intake for both concentrate types. No significant differences in milk yield or composition were established, probably because of incomplete adaptation even after 8 weeks.

It is concluded that at concentrate intakes of about 10 kg/day, the source of carbohydrate in the concentrates has little effect on milk production when the concentrates are of similar ME concentration. However, at higher levels of concentrate inclusion, although the diets may have similar ME concentrations, important differences in the yields of fat, protein and lactose occur due to carbohydrate source and these can be related to differences in rumen fermentation.

The computer model described simulates changes in soil mineral nitrogen and crop uptake of nitrogen by computing on a daily basis the amounts of N leached, mineralized, nitrified and taken up by the crop. Denitrification is not included at present. The leaching submodel divides the soil into layers, each of which contains mobile and immobile water. It needs points from the soil moisture characteristic, measured directly or derived from soil survey data; it also needs daily rainfall and evaporation. The mineralization and nitrification submodel assumes pseudo-zero order kinetics and depends on the net mineralization rate in the topsoil and the daily soil temperature and moisture content, the latter being computed in the leaching submodel. The crop N uptake and dry-matter production submodel is a simple function driven by degree days of soil temperature and needs in addition only the sowing date and the date the soil returns to field capacity, the latter again being computed in the leaching submodel. A sensitivity analysis was made, showing the effects of 30% changes in the input variables on the simulated amounts of soil mineral N and crop N present in spring when decisions on N fertilizer rates have to be made. Soil mineral N was influenced most by changes in rainfall, soil water content, mineralization rate and soil temperature, whilst crop N was affected most by changes in soil temperature, rainfall and sowing date. The model has so far been applied only to winter wheat growing through autumn, winter and spring but it should be adaptable to other crops and to a full season.

The model was validated by comparing its simulations with measurements of soil mineral N, dry matter and the amounts of N taken up by winter wheat in experiments made at seven sites during 5 years. The simulations were assessed graphically and with the aid of several statistical summaries of the goodness of fit. The agreement was generally very good; over all years 72% of all simulations of soil mineral N to 90 cm depth were within 20 kg N/ha of the soil measurements; also 78% of the simulations of crop nitrogen uptake were within 15 kg N/ha and 63% of the simulated yields of dry matter were within 25 g/m2 of the amounts measured. All correlation coefficients were large, positive, and highly significant, and on average no statistically significant differences were found between simulation and measurement either for soil mineral N or for crop N uptake.

A study of cross-bred ewes sired by Cambridge, Border Leicester and Lleyn rams out of Welsh Mountain ewes, involving a total of 408 observations, was carried out partly under lowland and partly under upland conditions in North Wales. The results snowed that the Cambridge and Lleyn were more precocious than the Border Leicester (relative percentage of ewes put to the tup lambing at 1 year old was 51·9, 66·3 and 66·8 respectively for Border Leicester, Cambridge and Lleyn cross-breds). In terms of lambs born and lambs reared per ewe lambing the ranking of the breeds was (in descending order) Cambridge, Border Leicester and Lleyn respectively. The Cambridge was lower in terms of mean weight of lambs at birth and at 56 days and relative mean age at slaughter tended to be lower in the Border Leicester crosses. Taking into consideration the lower body weight, particularly of the Lleyn cross-breds, ranking on relative feed conversion efficiency was Cambridge, Border Leicester and Lleyn when ewe fertility and days to slaughter were excluded.

Short-duration pigeonpea can give up to three harvests in environments with mild winters (e.g. minimum temperature above 10 °C) such as those prevailing in peninsular India (Sharma, Saxena & Green. 1978; Chauhan, Venkataratnam & Sheldrake, 1984). This is mainly due to the short time (about 120 days) taken to produce the first flush, and the strong perennial character of pigeonpea. The seed yield of short-duration pigeonpea in this multiple-harvest system may reach 5·2 t/ha (Chauhan et al. 1984).

The results of four field experiments conducted for 2 or 3 years on two soils, loamy sand (Typic Ustipsamments) and silty clay loam (Natric Ustochrept) showed that in rice, application of N in three equal amounts was more efficient than one or two applications, irrespective of source of N and type of soil. The results further showed that application of the first dose of N 7 days after transplanting rice was more beneficial than its application at transplanting. Calcium ammonium nitrate was significantly inferior to urea and/or ammonium sulphate for rice in both soils. There were no significant differences in wheat yield, N concentration or uptake due to time and source of N application.