Ten daylit, controlled-environment cabinets were used to investigate the possible impacts of global rises in atmospheric CO2 concentration and temperature on beetroot (Beta vulgaris L.), carrot (Daucus carota L.) and bulb onion (Allium cepa L.) plants. Their responses to CO2 concentrations of 350, 450, 550, 650 and 750 vpm and temperatures of 12, 13·5, 15, 16·5 and 18°C were examined by using a fractional factorial design for the two treatment factors. Use of the daylit cabinets allowed the plants to be grown in natural light, common atmospheric humidities (vpd 0·7 kPa) and non-limiting supplies of water and mineral nutrients.

Polynomial models were used to summarize the whole plant dry weight and fresh weight yield responses and to indicate the potential impact of climate change. Additionally, the models were used to generate predictions of the percentage change in whole plant dry weight and plant fresh weight yield for the years 2025 and 2050 relative to 1992. Baseline values of 350 vpm for CO2 and a mean temperature of 13·5°C for 1992 together with forecast CO2 values of 407 and 442 vpm and temperature increases of 0·7 and 1·1 C for 2025 and 2050 respectively were used. For 2025, fresh weight yield changes of +19%, +9% and +13% were obtained for beetroot, carrot and onion crops respectively, while for 2050 the respective changes were +32%, +13% and +21%.

Measurements of the ratio of the maximum diameter of the bulb to the minimum diameter of the neck for onions showed that there was little or no influence of CO2, whereas the effect of temperature was substantial. Bulbing was accelerated by high temperature and was greatly delayed at low temperature. At temperatures <15°C, the delays to bulbing resulted in the development of undesirable, thick-necked onions which tended to remain green with erect leaves. These results suggest, therefore, that a warmer climate will be advantageous for the commercial production of bulb onions in Britain.

Latin square and near-Latin square designs are valuable row-and-column designs for crop research but the practical size range of such designs is severely limited. Semi-Latin square designs extend this range but not all semi-Latin squares are suitable for experimental designs. Trojan square designs are a special class of optimal semi-Latin squares that generalizes the class of Latin square designs. The construction of Trojan squares both for unstructured and for factorial treatment sets is discussed and the utility of Trojan square designs for practical crop research is demonstrated. The corpus of available designs is further extended by a discussion of incomplete Trojan square designs obtained by omitting one main row or one main column from a complete Trojan square design. Some advantages of Trojan square and incomplete Trojan square designs for crop research are discussed and some suggestions for further design research are made.

Three white clover cultivars of differing geographical origin, namely AberHerald, Grasslands Huia and Sandra, were grown in binary mixtures with perennial ryegrass (cv. Preference) for a period of 3 years at Uppsala (59° 49′ N, 17° 39′ E) where the length of the main growing season is c. 200 days. Changes in growing point numbers, stolon length and stolon weight from autumn until late spring were followed by four to six destructive harvests each season. Changes in carbohydrate content (starch and water-soluble carbohydrates) were also determined on six or seven occasions during the same time. Cold hardening and dehardening were followed by artificial freezing tests on stolons removed from the plots on seven occasions during the last study season. The rate of node production was followed in the last spring. Temperatures in the stolon environment were recorded during the last two study seasons.

All cultivars suffered a loss of stolon length during the winters, with Huia being most affected. The extent of recovery during spring varied between years. Late spring frosts checked the growth of both AberHerald and Huia more than that of Sandra. The accumulation of starch appeared to depend on autumn temperatures and there was a decline in starch content during the winter. Sandra maintained a higher content of water-soluble carbohydrates during the winter and the spring than the other cultivars. Sandra also achieved a higher degree of freezing tolerance and dehardened later than both AberHerald and Huia. However, temperatures in the stolon environment always exceeded the degree of freezing tolerance attained irrespective of cultivar.

It is concluded that the accumulation and utilization of non-structural carbohydrates are important in determining the extent to which a white clover cultivar survives the winter and spring in a harsh climate. Freezing tolerance per se does not seem to be the most limiting trait, but in a climate where frequent spring frosts occur, a late dehardening seems to be valuable.

The consequences of sowing seed of burr medic (Medicago polymorpha) cultivars Circle Valley and Serena that was either free of or infected to different extents with alfalfa mosaic virus (AMV) was studied in mixed species pasture swards that regenerated annually. The swards were grazed by sheep and the predominant plant species in them were burr medic and capeweed (Arctotheca calendula). Serena matured earlier and was more tolerant of AMV infection than Circle Valley. Each year, seed-infected medic plants were the source for virus spread by aphids to healthy medic plants. When the extent of infection was determined in the fourth to sixth growing seasons, the amount of virus spread in the medic component of the swards varied between seasons. Final infection in plots originally sown with infected seed ranged from 47 to 93% for Serena and from 25 to 79% for Circle Valley. Viral seed transmission rates in medic seed produced each year by these plots ranged from 19 to 31% with Serena and from 3 to 7% with Circle Valley. Final percentage infection within swards originally sown with healthy seed (control plots) was smaller regardless of cultivar (0·1–7%) as were transmission rates in their harvested seed (0–0·6%).

AMV infection of the burr medic in regenerated plots originally sown with Circle Valley seed diminished medic seed yields, thereby decreasing the proportion of medic in the seed bank. This decreased germination of medic but increased germination of capeweed. In control plots, plant densities were up to 36% greater in the medic and 52% smaller in the capeweed components than in plots originally sown with infected Circle Valley seed. In plots containing Serena there was a smaller decrease in medic seed yields due to AMV infection, so the impact on germination was less. In the fourth to sixth years from sowing, when yields were determined at different times after grazing ceased, there was either a small decrease (up to 8%) or no significant decrease in overall herbage yields due to infection with AMV. However, in plots originally sown with infected Circle Valley seed, the balance of medic to capeweed was altered in favour of capeweed, sometimes dramatically so (e.g. capeweed content increased from 19 to 45% in the fourth year from sowing). In contrast, by the end of the growing season the balance of medic to capeweed was little altered by the presence of the virus in plots containing Serena. Thus, infection with this insidious virus disease substantially diminished the ability of Circle Valley but not Serena medic to compete with other species such as capeweed in self-regenerated, mixed species pasture swards. It did this both directly by decreasing the competitive ability of the medic plants that became infected during the growing season, and indirectly via seed production and the seed bank, by altering the proportions in which the species germinated.

In the context of EEC agricultural policy, the efficient production of sunflower (Helianthus annuus L.) requires the development of a wider range of crop management schemes than were used previously. Five different management options were applied to sunflower over a period of 10 years (1984–94) on a deep silty clay soil at Auzeville, near Toulouse, south-western France, combining different levels of crop density, N fertilization and supplementary irrigation. The effectiveness of these practices was evaluated on the basis of grain yield, oil production, amount of disease infection, N leaching, soil water deficit and economic return, using both experimental values and simulations with the epic-phase model.

Grain yield ranged from 1 t/ha (with severe disease incidence) to 4 t/ha (no diseases, full irrigation). There was no evidence of higher yield variability between years for low-input cropping systems managed without irrigation.

In the deep soils of Auzeville, c.100 mm of irrigation, split into two or three applications around flowering, appeared to be a good strategy for maximizing irrigation efficiency. Limiting leaf area expansion under rainfed management, by reducing crop density and N fertilization, limited stem infection frequency by Phomopsis helianthi and delayed soil water depletion, thus maximizing grain yield. Oil concentration was generally higher under full irrigation, but high values were also observed under rainfed conditions provided that a suitable genotype was used. The annual amount of N leaching, simulated by epic-phase, ranged from 17 to 40 kg/ha depending on N amount and irrigation volume. Under the current conditions of the Common Agricultural Policy, low-input and rainfed management appeared to be the most profitable strategy for the deep soils of south-western France.

The effects of four rates of fertilizer phosphorus (P) application (0, 9·8, 19·6 and 39·2 kg P/ha per year) on soil and crop P and cadmium (Cd) contents were measured in a field trial begun in 1968 and cropped each year with barley in south west England. In 1996, available and total soil P and Cd were measured in seven soil layers (0–20, 20–25, 25–30, 30–35, 35–40, 40–45 and 45–50 cm). Offtake of P in the crop was measured, or could be estimated, throughout the trial period. There was a linear relationship between P balance (total applied − total offtake) and P application rate with a balance of zero at a rate equivalent to 17 kg P/ha per year. The rate of P required for the economically optimum grain output was equivalent to 30 kg P/ha per year. No evidence was found for available P enrichment of soil layers below 25 cm. There was no evidence of Cd enrichment of either soil or crop after 29 years of P applications.

Lactating red deer (Cervus elaphus) hinds and their calves were rotationally grazed on Lotus corniculatus or perennial ryegrass/white clover pasture at an allowance of 12 kg DM/hind/day during summer 1996 in Palmerston North, New Zealand. Half the hinds suckled pure red deer calves and half suckled hybrid (0·25 elk[ratio ]0·75 red deer) calves. Measurements were made of the diet selected, voluntary feed intake of the hinds and liveweight changes of the hinds and calves.

Lotus corniculatus and perennial ryegrass constituted c. 90% of green material in the diet selected on the respective forages. Total nitrogen (N) content and organic matter digestibility (OMD) were higher for Lotus corniculatus than for perennial ryegrass/white clover pasture. Lotus corniculatus contained 21 g condensed tannin (CT)/kg dry matter (DM), whilst pasture contained only traces of CT (1·6 g/kg DM).

Hinds grazing Lotus corniculatus tended to have higher voluntary feed intake, and calf liveweight gain (485 v. 399 g/day) and weaning weight (52·6 v. 48·1 kg) were greater than for deer grazing perennial ryegrass/white clover pasture. Hybrid calves grew faster than pure red deer calves (P<0·01), with hybrid calves grazing lotus having very high liveweight gain (c. 520 g/day). Liveweight gain of hinds grazing Lotus corniculatus also tended to be higher (91 v. 20 g/day) than for hinds grazing perennial ryegrass/white clover pasture. CT was bound more strongly during chewing by red deer than had been found in comparable studies with sheep and the nutritional significance of this is discussed. Nutritional reasons for the superior performance of deer grazing Lotus corniculatus are discussed.

A grazing experiment was carried out using 18 Holstein/Friesian dairy cows at the Scottish Agricultural College, Crichton Royal Farm in 1991, to investigate the use of n-alkanes for estimating supplementary grass silage intake by grazing dairy cows. Two groups of animals grazed perennial ryegrass swards and were offered a supplement, consisting of a perennial ryegrass silage, twice daily after milking. Mean group silage intakes were measured daily during the last 5 days of a 12-day experimental period and individual silage intakes were estimated using n-alkanes. Animals were dosed twice daily with paper pellets containing dotriacontane (C32). The silage was marked with hexatriacontane (C36). The mean silage intake, estimated by weighing, was 6·8 kg DM per day. The mean estimated silage intakes were 6·9, 8·7 and 8·3 kg DM per day respectively using odd-chain n-alkanes in the C27−C35 range of naturally occurring alkanes, the odd-chain n-alkanes in the C27−C35 range with C36, and C36 by itself. No significant differences were found due to sampling routine (morning or afternoon) when using the naturally occurring n-alkanes but significant differences did exist when using the C36 alkane in the calculation of supplement intake. The results indicate that the n-alkane technique can be used to estimate the supplementary silage intake of grazing dairy cows using naturally occurring n-alkane patterns.

The effects of oversowing introduced grasses and legumes, tree killing and stocking rate on simulated diet quality and feed intake of cattle were studied in an open eucalypt woodland in the semi-arid tropics of north-eastern Australia in 1990. Rainfall during the experimental period was above average, producing good conditions for pasture growth, which resulted in little influence of stocking rate on pasture or dietary attributes. During the growing season, forage quality, as measured by nitrogen concentration and in vitro digestibility, was higher in sown pasture than in native pasture, and higher in plots with live trees than in plots with killed trees. Treatment differences in herbage quality disappeared later in the year when pastures senesced. In sown pasture treatments, there was no effect of tree killing on legume (Stylosanthes spp.) quality. However, Stylosanthes made a greater contribution to the diet of cattle grazing in wooded paddocks than in paddocks where trees had been killed. The proportion of Stylosanthes in the diet reached a peak in May and then declined. Oversowing increased both diet quality and feed intake while, to a lesser extent, killing trees resulted in generally lower diet quality and feed intake.

Estimates of liveweight change were made from the quantities of energy and protein consumed by the cattle. As expected, large increases in liveweight gain were predicted to occur in oversown pastures compared with native pastures. However, estimates indicated that animal performance may be better where trees are left intact rather than killed. While carrying capacity is greatly increased with the removal of trees, this may be at the expense of individual animal performance. Possible reasons for this significant negative response in dietary attributes and predicted liveweight gain to tree killing are discussed.

The feeding behaviour of Mamber dairy goats grazing on Mediterranean woodland in the Upper Galilee mountains of Israel was studied throughout a year (1991/92). The percentages of crude protein (CP), neutral detergent fibre (NDF), acid detergent fibre (ADF), acid detergent lignin (ADL) and condensed tannins (CT) were determined in samples collected by hand to simulate the plant parts selected by the goats. Total feed intake was evaluated by using Cr-sesquioxide as an external marker to determine the amount of faeces excreted, and 48 h in sacco degradability to assess the digestibility of samples. Overall, more time was spent by goats in feeding on ligneous material (tree and shrub foliage) than herbaceous vegetation (60 and 40% of total feeding time, respectively, P<0·05). The main ligneous species consumed by the goats were Quercus calliprinos, Sarcopoterium spinosum and Calicotome villosa (20, 13 and 7% of total grazing time, respectively). There was considerable variability in the 48 h dry matter (DM) in sacco degradability, and in the concentration of CP, NDF, ADF, ADL and CT, within and between seasons and plant species. Although requirements for nutrients varied according to the physiological stage of the goats, 48 h in sacco DM degradability and the concentration of non-ADF linked (available) CP, NDF, ADF, ADL and CT in the diet did not vary greatly: respective ranges were 45·0–49·4; 9–12·5; 44–53; 33–39; 12–17 and 3·5–4·7%. It is concluded that Mamber goats (i) may not select the best quality diet available, but may avoid wide variations in nutrient content of their diets throughout the year; and (ii) may not select a diet consistent with maximization of milk yield, but rather with optimization of body condition at the onset of the mating season.

Granular calcined magnesites originating from Spain, China and Greece and magnesium hydroxide powder were evaluated in a series of balance experiments conducted with wether sheep (c. 42 kg liveweight) given a basal ration of dried grass at Glasgow University Veterinary School between 1989 and 1994. Four Spanish ‘Agma’ products and three Spanish ‘Navarras’ products contained less material (0–104 g/kg)>1000 μm diameter than four Chinese and four Greek materials 178–483 g/kg). The Spanish products had lower (18–45 g/kg) losses on ignition compared with the Chinese and Greek materials (53–112 g/kg). A fifth Greek product contained only 4 g/kg>1000 μm diameter. In three experiments comparing supplementation of the basal diet with 2·0 g Mg as either magnesium hydroxide or as one of the three separate purchases of the commercial calcined magnesite sources, the powdered magnesium hydroxide had the highest mean apparent availability coefficient (0·30) (faecal increase method). The corresponding values were 0·24 (Agma and Navarras), 0·20 (Chinese) and 0·15 (Greek) (pooled s.e. 0·014, d.f. 10). Powdered magnesium hydroxide also had the highest (0·18) mean availability coefficient (urinary increase method) compared with 0·12 (Agma), 0·13 (Navarras), 0·14 (Chinese), 0·10 (Greek) (pooled s.e. 0·005, d.f. 10). In two subsequent experiments, Agma had a superior apparent availability coefficient (0·26) (faecal method) than either the Chinese (0·14) or the Greek materials (0·19). A finer grade of Greek calcined magnesite (0·25) was equivalent to Agma. For the 12 calcined magnesites evaluated in the three main experiments, significant (P<0·05) correlations (r) were found between the magnesium availability coefficient (faecal method) and loss on ignition (−0·65); particle size proportions (<250 μm, 0·59; and >1000 μm, −0·65) and tended towards significance for rumen solubility in 24 μm nylon bags in hay-fed cows over two days (r=0·54, P=0·07) and rate of reaction with citric acid (0·57, P=0·055). No correlations were found with magnesium availability coefficients (urinary increase method). This confirms the limitations of both in vitro predictions and in vivo (faecal) estimates of magnesium availability for different calcined magnesites. The mean magnesium availability coefficient (faeces) of the 17 calcined magnesites was 0·21 (s.d. 0·05). If this were to be confirmed by other observations, it would have implications for the amounts of dietary calcined magnesite required to meet recommended dietary magnesium allowances for ruminants.

This group, which is concerned with the applications of mathematics to agricultural science, was formed in 1970 and has since met at approximately yearly intervals in London for one-day meetings. The thirtieth meeting of the group, chaired by G. J. S. Ross of the Institute of Arable Crops Research, Rothamsted, was held in the Wellcome Meeting Room at the Royal Society, 6 Carlton House Terrace, London on Friday, 3 April 1998, when the following papers were read.