To test the hypothesis that the ingestion of Mo-rich soils by ruminants might adversely affect Cu metabolism, four groups of initially hypocupraemic ewes were repleted with a Cu-supplemented diet either soil-free or containing 10% of one of three soils of low, medium and high Mo content. The three soils were equally effective in inhibiting the response in plasma Cu and probably reduced the availability of Cu by more than 50%. The ingestion of certain soils produced increases in urinary Mo and Zn concentrations, indicating that the Mo and Zn ingested in soil can be biologically available: however, these Cu antagonists were probably released in insufficient quantities to have contributed greatly to the observed interference with Cu metabolism. Soil ingestion may be involved in the aetiology of hypocuprosis in cattle and swayback in sheep, since it commonly constitutes 10% of the D.M. intake of animals grazing winter pasture.

Genotype by environment (G×E) interaction effects are of special interest for breeding programmes to identify adaptation targets and test locations. Their assessment by additive main effect and multiplicative interaction (AMMI) model analysis is currently defined for this situation. A combined analysis of two former parametric measures and seven AMMI stability statistics was undertaken to assess G×E interactions and stability analysis to identify stable genotypes of 11 lentil genotypes across 20 environments. G×E interaction introduces inconsistency in the relative rating of genotypes across environments and plays a key role in formulating strategies for crop improvement. The combined analysis of variance for environments (E), genotypes (G) and G×E interaction was highly significant (P<0·01), suggesting differential responses of the genotypes and the need for stability analysis. The parametric stability measures of environmental variance showed that genotype ILL 6037 was the most stable genotype, whereas the priority index measure indicated genotype FLIP 82-1L to be the most stable genotype. The first seven principal component (PC) axes (PC1–PC7) were significant (P<0·01), but the first two PC axes cumulatively accounted for 71% of the total G×E interaction. In contrast, the AMMI stability statistics suggested different genotypes to be the most stable. Most of the AMMI stability statistics showed biological stability, but the SIPCF statistics of AMMI model had agronomical concept stability. The AMMI stability value (ASV) identified genotype FLIP 92-12L as a more stable genotype, which also had high mean performance. Such an outcome could be regularly employed in the future to delineate predictive, more rigorous recommendation strategies as well as to help define stability concepts for recommendations for lentil and other crops in the Middle East and other areas of the world.

The current study was designed to determine the effect of roughage to concentrate ratio (R : C) on rumen pH, fermentation and bacterial population in dairy steers. Four rumen fistulated dairy steers (170±20 kg) were randomly assigned according to a 4×4 Latin square design, in which the steers were fed with four dietary treatments with different R : C ratios of 0·8 : 0·2, 0·6 : 0·4, 0·4 : 0·6 and 0·2 : 0·8, respectively. All animals were kept in individual pens and received feed according to the respective R : C ratios at 0·025 body weight (BW)/d; urea-treated rice straw (prepared using 3·5 kg urea+100 kg water sprayed onto 100 kg of rice straw) was used as a roughage source. The experiment was conducted for four periods of 21 days each. During the first 14 days, feed intake was measured and the animals were then moved to metabolism crates for total urine and faecal collection for 7 days. Total dry matter intake (DMI) was similar among treatments. Energy intake increased as the proportion of concentrate in the diet increased. Apparent digestibilities of dry matter (DM), organic matter (OM) and crude protein (CP) were improved, while neutral detergent fibre (NDF) and acid detergent fibre (ADF) were reduced when the levels of concentrate increased. A decreasing ratio of R : C reduced rumen pH linearly, from 6·4 to 5·9 at 0·2 : 0·8. High levels of concentrate impacted on volatile fatty acids (VFA) molar proportions and decreased acetate (C2) linearly, while propionate (C3) was increased, leading to decreased C2 : C3 ratio. Numbers of protozoa, fungi and proteolytic bacteria were not affected by R : C ratio. Cellulolytic bacteria decreased linearly while amylolytic bacteria increased linearly with 0·60 and 0·80 concentrates. Quantitative polymerase chain reaction (qPCR) based on 16S RNA revealed that Fibrobacter succinogenes numbers were increased when steers were fed with R : C ratio of 0·8 : 0·2. Conjugated linoleic acid (CLA)-producing bacteria, especially those of Butyrivibrio fibrisolvens, increased linearly with R : C ratios of 0·8 : 0·2 and 0·6 : 0·4, while Megasphaera elsdenii, a lactate-utilizing bacterium and reported producer of trans-10, cis-12 CLA increased linearly with R : C ratio of 0·8 : 0·2. In addition, microbial CP synthesis increased quadratically when steers were fed high levels of concentrate. However, the efficiency microbial N synthesis (EMNS) based on OM, truly digested in the rumen, was not affected by different R : C ratios. From the current study, it can be concluded that roughage to concentrate ratio of 0·4 : 0·6 had positive effects for the creation of healthy rumen (rumen pH and ecology), and improved energy intake and rumen fermentation, particularly propionic acid and microbial protein synthesis, in dairy steers fed urea-treated rice straw as a roughage source.

Sheep prepared either with a rumen cannula, or with a rumen cannula plus re-entrant cannulae in the duodenum and ileum were fed fresh ‘Ruanui’ perennial ryegrass (R), ‘Manawa’ short-rotation ryegrass (M) and white clover (C) at dry-matter intakes ranging from 450 to 1000 g/24 h.

The re-entrant cannulated sheep were given paper impregnated with chromic oxide (Cr2O3) once daily via the rumen fistula as a marker; recoveries of Cr2O3 at the duodenum, ileum and faeces were approximately 75, 69 and 100% respectively. Regressions of Cr2O3-corrected digesta constituent flows (g/24 h) on intake were constructed and from these were calculated the sites of digestion of OM, gross energy, readily fermentable carbohydrate, structural carbohydrate and lipid between the stomach, small intestine and large intestine.

There were differences between pasture species in partition of OM and energy digestion. With M, digestion in the stomach was significantly less (P < 0·01) and digestion in the small intestine was significantly more (P < 0·05) than with either R or C. These partitions of OM and energy digestion changed with level of feed intake. A substantial proportion (9–18%) of the digestible OM and energy was apparently digested in the large intestine with all three herbages.

With all three herbages approximately 93% of the digestible readily fermentable carbohydrate and 90% of the digestible structural carbohydrate was digested in the stomach region.

Reasons for differences in gross efficiency of utilization between the pasture species in the order C > M > R were examined.

The fraction (f) of surface-applied nitrate leached below any depth h cm in a uniform soil profile may be calculated from the equation

where P is the quantity of water draining through the soil (in cm) and Vm is the percentage volumetric field capacity. The fraction of nitrate retained is then (1—f).

This equation has been tested using published data. Values of h corresponding to the mean displacement (f = 0·5) were calculated for a wide range of soil and weather conditions and the results compared with mean displacements measured in the field. Similar comparisons were made with the leaching equation of Rousselle (1913) and Levin (1964). The new equation gives good agreement with the observed data, whereas the Rousselle-Levin equation generally overestimates the mean displacement of nitrate. Methods of applying the equations to field situations are discussed.

Four extra-short-duration (ESD) lines in 1991 and eight ESD lines in 1992 were grown with adequate soil moisture throughout their growth or subjected to drought coinciding with the vegetative, flowering and pod-filling stages under rainout shelters. In both years, drought stress treatments significantly reduced dry matter accumulation and grain yield. The extent of reduction in grain yield varied with the line and stage of stress imposition. Drought stress at the flowering stage caused greater reduction in total dry matter and grain yield than the stress imposed during the pre-flowering and pod-filling stages. Drought stress coinciding with the flowering stage reduced grain yield by 40–55% in 1991 and 15–40% in 1992 in different lines. ESD genotypes could extract moisture from up to a metre depth during pre-flowering and flowering stage stress but less so during the pod-filling stage stress. Genotype ICPL 88039, followed by ICPL 89021, showed consistently lowest sensitivity to drought stress at flowering. Protracted drought stress commencing from the pre-flowering to flowering or from the flowering to pod-filling stages was more harmful than stress at the individual stages. The reduction in yield under drought stress could be attributed mainly to less total dry matter accumulation, but also increased abscission of plant parts. The results suggest variation in sensitivity of ESD lines in relation to timing of stress, which should facilitate targeted screening for different intermittent moisture stress environments.

Two groups of weaned sheep weighing about 30 kg, one born in March and the other in September, were each divided into two and given ad libitum one of two pelleted diets, ruminant diet A or ruminant diet AA6. The March-born sheep commenced experiment in July and those born in September in January. The experiment continued for 4½ years. At intervals sheep were killed and the fat protein and ash contents of their digesta-free bodies determined.

The voluntary intake of feed showed a seasonal periodicity with minima in the winter and maxima in summer. The amplitude was 30% in the 1st year and in subsequent years averaged 13%. Those given the higher quality diet (AA6) consumed slightly less than those given the poorer one.

Mean daily feed intake averaged over 6-month periods from January to July and July to January was invariant with age during 4 years of observation. In this time the sheep increased in weight from about 30 kg to about 130 kg. There were, however, considerable differences between individual sheep in the amount of feed they habitually consumed.

The body weight of the sheep increased and eventually plateaued. The asymptotic weight defined as A in the equation W = A–Be–kt, where W is weight at time t and B and k are constants, was related to the mean daily feed intake averaged over 6 months; mean daily feed intake measured over 6 months was proportional to a fractional power of body weight indistinguishable from 0·75, the interspecies power to which metabolism is proportional. Growth of wool during successive 6-month periods did not vary with age of animal but differed significantly between animals.

Fasting metabolism of the sheep was 316 kJ/kg W0·75 for wethers and 336 kJ for rams.

Analysis of the bodily composition data showed that over a range of digesta-free body weight from 46 to 130 kg it was not possible, on statistical grounds, to distinguish between linear relationships between body weight and its fat, protein, ash and water content, and allometric ones. The linear relationships had marginally smaller residual standard deviations and the regression coefficients show that the gain of the empty body in these sheep consisted of 68% lipid, 8% protein, 1% ash and 24% water. The lipid in the carcass accounted for 88% of the total lipid gain and half the accretion of protein and ash was in the carcass. These results confirm those of Searle, Graham & O'Callaghan (1972) based on tritium dilution which showed that post-puberal growth in sheep is of constant composition.

The results of the metabolic studies are shown to be consistent with the growth studies.

Growth to maturity, as affected by different hypotheses related to the determinants of maintenance energy expenditure and the regulation of appetite, was examined algebraically. It is shown that mature weight is the rate of feed intake divided by the rate of maintenance metabolism per unit weight, and for defined feeding systems the rate constant for the approach to mature weight is the rate of maintenance metabolism divided by the feed equivalent of unit gain.

The implications of the results in terms of the measurement and prediction of feed intake and the use of metabolic body size as a scaling factor in comparative and genetic studies of growth are briefly discussed.

1. The resting heart rate of cocks and hens was measured in varying degrees of heat stress, and during acclimatization to heat.

2. When the rectal temperature was below 110°F. (43·3° C), the heart rate varied inversely with the severity of the climate up to an air temperature around 99° F. (37·2° C).

3. At air temperatures of 99° F. (37·2° C.) and above the heart rate was nearly constant over a wide range of rectal temperatures below 110° F. (43·3° C). It averaged about 50 beats/min. less than in an equable climate.

4. Above 110° F. (43·3° C.) there was an increase of about 30 beats/min./° F. (17 beats/min./° C.) rise in rectal temperature. The rate equalled that found in an equable climate, when the rectal temperature reached 111–112° F. (43·9–44·4° C). Over this zone of body temperature the fowls were often agitated.

5. In severe heat stress there was a slight fall in the heart rate on acclimatization, but this was small compared with the effect of climate.

6. In severe heat stress the heart rate during the day was slightly higher than that at night.

7. The relation of these findings to the mechanism of temperature regulation is discussed.

An adapted animal is one in perfect harmony with its environment. Of all the interrelated factors influencing the environment, climate is probably the most important, affecting not only the vegetation and parasitical fauna but likewise the density of human population, its needs and culture.

A study of the climatological influence on animals must be comprehensive and include species, breeds and types within breeds, land as well as aquatic animals, both steppe and mountain animals, and existing animals as well as those of historical importance. The climatology with which the animal breeder is concerned aims at establishing the influence of climate, with its constantly recurring phases, upon the animal (in this case the bovine). The reaction of any animal to a particular external environmental stimulus is closely correlated with the animal's efficiency of production.

From the foregoing account the following summary may be given:

1. Examination of Rothamsted soils on different occasions has revealed the presence of an organism capable of breaking down cellulose with comparative ease.

2. This organism presents a number of features of morphological and physiological interest. Morphologically, the organism appears to possess greater affinities with the Spirochaetoideae than with the bacteria and the name Spirochaeta cytophaga is, therefore, suggested.

3. While the spirochaet is capable of considerable vegetative growth as a sinuous filamentous cell, it also appears to pass through a number of phases which terminate in the production of a spherical body (sporoid) which differs in a number of respects from the true spores of the bacteria. Germination of the sporoid again gives rise to the filamentous form, which possesses perfect flexibility and is feebly motile. The latter does not apparently possess flagella.

4. Spirochaeta cytophaga is essentially aerobic; its optimum temperature is in the region of 30°. Both the thread and sporoid stages are killed by exposure to a temperature of 60° for ten minutes.

5. The nitrogen requirements of the organism may be met by a number of the simpler nitrogen compounds—ammonium salts, nitrates, amides and amino-acids. Peptone is also suitable in concentrations up to 0.025 per cent. Stronger solutions, e.g. 0.25 per cent., lead to marked inhibition of growth. The organism fails to grow on the conventional nutrient gelatine or agar.

Two methods for calculating the flow of digesta and its constituents from the concentrations of two markers, 51Cr-labelled chromium EDTA (51Cr-EDTA) and 103Ru-labelled ruthenium phenanthroline (103Ru-P), and digesta composition were compared in fistulated sheep fed once daily and in other fistulated sheep fed continuously. One method uses the marker concentrations to calculate the flow of whole digesta directly (double-marker) whereas the other uses the markers to calculate separately the flow of fluid and non-filtrable particles, the sum of which is digesta flow (two-marker).

When feed was given continuously, flow values calculated using the double-marker method and the two-marker method, which requires a correction to be made for nonideal behaviour of 51Cr-EDTA, did not differ. When feed was given once daily, the double-marker method consistently overestimated flow relative to the corrected twomarker method but the differences were extremely small (less than 0-8% for digesta flow and 0-5% for nitrogen flow). 51Cr-EDTA as a single marker slightly overestimated digesta flow (up to 2·3%) but grossly overestimated nitrogen flow (up to 22·5%); 103Ru-P as a single marker grossly underestimated both digesta flow (up to 30·5 %) and nitrogen flow (up to 23·1%).

It was concluded that the double-marker method is to be preferred when the diet is given continuously or frequently at fixed intervals. When the diet is given once daily, the two- and double-marker methods can be expected to give similar values. Single markers should not be used for estimating the flow of digesta and its constituents when samples are taken from a simple cannula.

The effects of adding a culture extract from Aspergillus oryzae (AO) to the fermentation of a basal ration were investigated using the rumen simulation technique (Rusitec). The ration consisted of hay, barley, molasses, fishmeal and a minerals/vitamins mixture, at 500, 299–5, 100, 91 and 9–5 g/kg dry matter respectively. AO eliminated the transient fall in pH that occurred following the addition of substrate to Rusitec, but had no influence on the buffering capacity of the medium. Increases in the acetate: propionate ratio and in the proportion of butyrate in the fermentation products were observed in vessels receiving AO. Ammonia concentrations increased by over 30%. The number of total viable bacteria almost doubled and the cellulolytic population increased three-fold in vessels receiving AO. Protozoal numbers were reduced by almost 45% by AO. The proportion of methane in the headspace gas collected from the vessels during the day was significantly (P < 0·01) lower in vessels receiving AO. These results demonstrate that AO has a marked effect on the microbial population and the stoichiometry of the rumen fermentation, and that these changes can be seen in an in vitro system.

Experiments were conducted over the period 1987–94 at Cambridge University Farm and two other sites to examine the effect of various husbandry factors, particularly variety and irrigation regime, on rate, depth and density of rooting in potatoes. Maximum rooting depth ranged from 59 to 140 cm, indicating that potatoes can root to considerable depths and thereby have access to large volumes of water to satisfy the potential demand for water created by the atmospheric conditions and the size of the canopy. Root extension vertically through the soil profile was best described as a three-phase process: an initial rapid period lasting 3–5 weeks with growth rates c. 1·2 cm/day, a second period of slower growth (c. 0·8 cm/day), followed by cessation of root extension for the rest of the life of the crop. Variety had a major influence on the ultimate depth of rooting, primarily owing to variations in the length of the different periods of rooting rather than the rate in each period. It was observed that changes in the rate, or the cessation of root penetration were always preceded 4–9 days earlier by a change in the rate, or cessation, of leaf appearance. This feature should make it possible to characterize the duration of rooting of varieties through measurement of leaf emergence. Varieties which ceased leaf production early, such as Atlantic, were found to have a duration of root growth of c. 60 days, with Cara rooting for c. 30 days longer. Maximal total root length (TRL) and root length density (RLD) in the experiments reported were 16·9 km/m2 and 5·5 cm/cm3, respectively, similar to those found previously in potatoes and crops such as sugar beet, but considerably greater than many other vegetables. Rooting density decreased with depth, but the root systems were not as surface-oriented as many other studies have shown. When TRL was close to its maximum, the vertical distribution of RLD showed that between 40 and 73% was confined to the upper 30 cm, with irrigated crops possessing a greater proportion of their roots in the plough layer. Despite being planted in rows 70–91 cm apart, rooting systems were homogeneously distributed in a horizontal direction by c. 35 days after emergence, at which time the roots had reached a depth of c. 50 cm. Therefore, apart from a short period after emergence, the potato crop is capable of accessing considerable volumes of soil from which to extract water and nutrients. Ensuring that soil conditions are conducive to maximal rates of root growth should be the target for growers, since this will lead to a more efficient use of soil water and irrigation.

In a study of the effect of animal species on the threshold particle size leaving the rumen, two grasses cut at two stages of growth and one mature legume were separated into leaf and stem fractions and fed to cattle and sheep. Samples of rumen digesta and faeces were used to determine the validity of using a 1·18 mm porosity screen to separate the rumen particles into large and small pools when studying escape of particles from the rumen. Samples of rumen digesta and faeces were collected for the determination of particle size by wet sieving and the calculation of resistance of particles to passage from the rumen relative to small particles retained on a 0·15 mm sieve.

Particles < 1·18 mm but > 0·5 mm had a mean relative resistance to passage of 2·0 and 2·6 for cattle and sheep respectively, compared with resistance values of between 10·9 and 31·2 for particles between 1·18 and 2·36 mm. It is suggested that there is no justification for using different threshold particle sizes for sheep and cattle and that a 1·18 mm sieve may be used to divide the rumen contents of both cattle and sheep into two pools of particles with high and low relative resistance to passage from the rumen.

1. The reflex ejaculation responses of cocks to lumbar massage have been examined, and some basic reactions described. A list of the crude fluid components obtained by the massage technique of semen collection is given.

2. From these observations, and a chemical test on the contents of the hen's vagina after natural and artificial insemination, it was deduced that the normal ejaculate of the cock is composed mainly of contents from the vasa deferentia, with some secretions from the mucosal glands of the vascular bodies and lymph folds. The blood plasma transudate sometimes expelled into the semen is obtained as a result of compression of the terminal organs during the collection procedure.

3. Points emerging from the observations on ejaculation, and the search for the normal constituents of semen are discussed.

Chaffed lucerne hay of 64% apparent organic matter (OM) digestibility was fed to wether sheep under four feeding regimens: two levels of dry matter (D.M.) intake (700 (L) and 1050 (H) g/day) and within each level two feeding frequencies (once a day (daily) and once an hour (hourly)). Three separate groups of sheep were used concurrently: a slaughter group was used to obtain feeding behaviour data and to measure pool sizes and obtain samples from the reticulo-rumen; a digestion group, in which each sheep was prepared with a rumen and a duodenal cannula, was used to measure duodenal digesta flow, rumen microbial growth and reticulo-rumen motility; a balance group was used to measure digestibility and nutrient balances.

High D.M. intake increased reticulo-rumen pool sizes and flow rates but it did not affect apparent digestibilities or the proportions of OM, fibre, cellulose, hemicellulose, lipid and nitrogen digested in the stomach and intestines. Increased feeding frequency had a major effect on reticulo-rumen pool sizes but did not affect apparent digestibilities or partition of digestion of non-nitrogenous constituents. Daily feeding resulted in increased total-N flow to the duodenum; however, N retention was significantly greater with frequent feeding. It is suggested that this was due to a more efficient tissue utilization of N.

The kinetics of digesta flow within the reticulo-rumen, expressed as fractional flow rates, were studied with data from sheep fed hourly. The fractional inflow, outflow and disappearance rates for OM, fibre, cellulose and hemicellulose did not change with an increase in intake because of an equivalent increase in reticulo-rumen volume. Increasing D.M. intake by 50% resulted in a 24% increase in water intake, a 19% increase in reticulo-rumen water volume, and a 49% increase in water outflow rate. The data suggest that the increased outflow of water was achieved by increasing the net flow of water across the mucosa into the reticulo-rumen, rather than by increasing salivation.

There was no difference between treatments in the frequency of reticulo-rumen contractions. It was calculated that for each A sequence contraction, OM flow was 0·26 and 0·37 g and water flow was 4·38 and 0·36 g on L and H intakes respectively. A 50% increase in intake resulted in a 42% increase in OM passage per A sequence contraction. This increased passage with intake was not accompanied by an increase in reticulorumen contraction frequency.

A total of 21 rumen-cannulated wether sheep and six rumen-cannulated Ayrshire cows were used in a series of experiments to investigate fermentation in the rumen in animals given silage diets and the nominal metabolism of silage lactic acid. Dietary experiments with sheep (three experiments) and cows (two experiments) given diets consisting of formic acid–preserved perennial ryegrass (Lolium perenne) silages, in some cases with supplements of rolled barley, were conducted to investigate the ruminal concentrations of lactic acid and of volatile fatty acids (VFA) after feeding. Ingestion of silage was followed by a short-lived peak in lactic acid concentration in the rumen and by a subsequent increase in the concentration of propionate or propionate and butyrate. Butyrate production was more evident in cattle than in sheep. Taken as an average over a 6 h period of sampling the proportions (m-mol/mol total VFA) of acetate, propionate and butyrate in the rumen in animals given silage alone were typically 590–600, 210–250 and 70–150 respectively. When silage was supplemented with barley, in both sheep and cattle, the proportion of propionate was reduced and the proportion of butyrate was increased. In one experiment with sheep these changes in fermentation pattern were shown to be accompanied by an increase in the number of rumen protozoa from 2·78×105/ml to 18·66·105/ml (P <0·01). However, in a subsequent experiment it was shown that defaunation led to an increase in the proportion of butyrate in the rumen.

Twelve estimates were also made with sheep of the metabolism of lactic acid infused into the rumen in a single dose of 20–50 g. These experiments showed that both (L+) and DL-lactio acid were metabolized equally rapidly with a half-life of approximately 25 min. In faunated animals the major products of fermentation were propionate with a small amount of butyrate. In defaunated animals lactate metabolism was slowed and the major product of fermentation was butyrate.

In vitro studies of lactate fermentation were carried out using strained liquor and isolated bacterial and protozoal fractions from the rumen of sheep receiving silage diets. The rates of lactate disappearance (μg/mg protein/h) were 50 and 22·2 for the rumen liquor and bacterial fraction, respectively, which were significantly (P <0·05, n = 5) less than the rate of 294·7 for the protozoal fraction. Rumen liquor, bacterial and protozoal fractions yielded a similar mixture of fermentation end-products, propionate accounting for approximately 60% of the VFA produced.

The results are discussed in relation to the metabolism of silage lactic acid in the rumen, the production of VFA and the yield of ATP for microbial cell synthesis. It is concluded that the pathways of lactate fermentation are regulated by the rumen microbial population and that protozoa may have an important role in the regulation of lactate fermentation.

Simulation of phasic development in wheat is necessary in constructing wheat growth and yield models. It is also useful for evaluating cultivar adaptation and scheduling cultural practices. This paper describes a conceptual model of wheat development based on phenological principles, as affected by vernalization, photoperiod, thermal response and intrinsic earliness, and also reports the results of sensitivity analysis and validation of the model.

The model predicts when the plant will reach double ridge, terminal spikelet and heading. In the model, the daily thermal sensitivity of development following emergence is determined by an interaction of ‘relative vernalization completion’ and ‘relative photoperiod effectiveness’ for that day. After complete vernalization is reached, the daily thermal sensitivity is determined only by relative photoperiod effectiveness, which gradually increases from terminal spikelet to heading. A multiplication between the daily thermal sensitivity and thermal effectiveness generated daily flowering time, which was accumulated to trigger a particular developmental stage. Genotypic differences were characterized as vernalization requirement, photoperiod sensitivity and intrinsic earliness.

The model showed a sensitive response to environmental variables of temperature and daylength, and to genetic parameters of vernalization requirement and photoperiod sensitivity. Evaluation of the model using multiple experimental data involving various cultivars and planting dates exhibited a marked goodness of fit between simulation and observation with a root mean square error <5 days. The results indicate that the model can be used as a predictor for the major flowering stages, as well as functioning as a knowledge base for understanding the characteristics of different development components in wheat.

Water status traits have been proposed as criteria for drought tolerance improvement in several species especially in bread wheat, oat and barley. In order to test this possibility in durum wheat, in 1995 and 1997, a collection of 144 genotypes from different geographical origins was cultivated under rainfed Mediterranean conditions in Montpellier (southern France). Relative water content (RWC), osmotic potential (OP), as well as leaf morphological traits, grain yield (GY), total above-ground biomass and harvest index (HI) were measured. Large genotypic variations were observed for all traits. Differences between the 2 years were also noticed for all the traits which could be attributed to differences in water availability between the 2 growing seasons. The correlation analysis has revealed significant positive associations between water status traits and both GY and HI within and across years. These results suggest that genotypes which can maintain higher water status and then higher transpiration rates during grain filling, are more efficient in dry matter partitioning to grain and hence higher yield. Heritabilities of water status characters were higher than those of productivity traits. Our results emphasized the value of RWC and OP as criteria for durum wheat grain yield improvement under rainfed Mediterranean conditions. Relationships between RWC, OP and flag leaf morphological traits are also discussed.

Use of generalized lattice designs instead of complete block designs in 244 cereal variety trials grown in the U.K. since 1975 has resulted in an average reduction of 30% in variances of varietal yield differences. The lattice designs were most effective when the number of varieties was more than 50 but worthwhile reductions in variance, averaging about 24%, were obtained in trials with fewer than 20 varieties. The data also provide guidance on the choice of block size and the potential for improvement by neighbour methods.