Performance of ratio‐based,soil‐adjusted and atmospherically corrected multispectral vegetation indices in predicting herbaceous aboveground biomass in a Colophospermum mopane tree–shrub savanna

Accurate and near‐real‐time estimation of herbaceous aboveground biomass (AGB) at farm level is crucial for monitoring pasture production and proactive management of stock in semiarid rangelands. Despite its importance, remote sensing has been rarely used by range ecologists and managers in Zimbabwe. This study aimed at assessing the performance of classical multispectral vegetation indices (MVIs) when either singly regressed with measured herbaceous AGB or combined with other visible spectral bands in predicting herbaceous AGB in a Colophospermum mopane savannah. Field herbaceous AGB and corresponding Landsat 8 Operational Land Imager visible spectral data were collected during the 2016–2017 rainy season. Relationships between measured AGB and classical MVIs and extended models of MVIs combined with other visible bands were analysed using bootstrapped simple and stepwise multiple linear regression functions. When MVIs were singly regressed with measured AGB, ratio‐based indices yielded the highest r² value of 0.64, followed by soil‐adjusted indices (r² = 0.61), while atmospherically corrected MVIs showed the lowest r² of 0.58 (p = 0.00). A significant improvement in herbaceous AGB estimation was obtained using a combination of MVIs and other visible bands. Soil‐adjusted MVIs showed the greatest increase (44–46%) in r², while atmospherically corrected and ratio‐based MVIs poorly improved (<5%). The findings demonstrate that combining MVIs with Landsat 8 optical bands, especially green band, provides the best models for estimating AGB in C. mopane savanna rangelands. These findings emphasize the importance of testing band‐MVI combinations when developing models for estimating herbaceous AGB.     

Hyperspectral remote sensing of yellow mosaic severity and associated pigment losses in Vigna mungo using multinomial logistic regression models

Yellow mosaic disease (YMD) has been a serious threat to blackgram cultivation especially during post-monsoon season. Visual assessment of disease severity is qualitative and time consuming. Rapid and non-destructive estimation of YMD by hyperspectral remote sensing has not been attempted so far on any of its hosts. Field studies were conducted for two seasons with eight blackgram genotypes having differential response to YMD. Comparison of mean reflectance spectra of the healthy and YMD infested leaves showed changes in all the broad band regions. However, reflectance sensitivity analysis of the narrow-band hyperspectral data revealed a sharp increase in reflectance from the diseased leaves compare to healthy at 669 (red), 505 and 510 nm (blue). ANOVA showed a significant decrease in leaf chlorophyll (p < 0.0001) with increase in disease severity, while no such relationship was observed for relative water content. By plotting coefficients of determination (R²) between leaf chlorophyll and percent reflectance at one nm wavelength interval, two individual bands (R₅₇₁; R₇₀₅) and two band ratios (R₅₇₁/R₇₂₁; R₇₀₅/R₅₉₃) with highest R² values were selected. These bands showed a significant linear relationship with SPAD chlorophyll readings (R² range 0.781–0.814) and spectrometric estimates of total chlorophyll content (R² range 0.477–0.565). Further, the relationship was stronger for band ratios compared to single bands. With optimal spectral reflectance ratios as inputs, disease prediction models were built using multinomial logistic regression (MLR) technique. Based on model fit statistics, reflectance ratios R₅₇₁/R₇₂₁ and R₇₀₅/R₅₉₃ were found better than the individual bands R₅₇₁ and R₇₀₅. Validation of MLR models using an independent test data set showed that the overall percentage of correct classification of the plant into one of the diseased categories was essentially same for both the ratios (68.75%). However, the MLR model using R₇₀₅/R₅₉₃ as dependent variable was of greater accuracy as it gave lower values of standard errors for slopes (β_G range 9.79–36.73) and highly significant estimates of intercept and slope (p < 0.05). Thus the models developed in this study have potential use for rapid and non-destructive estimation of leaf chlorophyll and yellow mosaic disease severity in blackgram.     

Multivariate analysis of nitrogen content for rice at the heading stage using reflectance of airborne hyperspectral remote sensing

Airborne hyperspectral remote sensing was adapted to establish a general-purpose model for quantifying nitrogen content of rice plants at the heading stage using three years of data. There was a difference in dry mass and nitrogen concentration due to the difference in the accumulated daily radiation (ADR) and effective cumulative temperature (ECT). Because of these environmental differences, there was also a significant difference in nitrogen content among the three years. In the multiple linear regression (MLR) analysis, the accuracy (coefficient of determination: R², root mean square of error: RMSE and relative error: RE) of two-year models was better than that of single-year models as shown by R² ≥ 0.693, RMSE ≤ 1.405 g m⁻² and RE ≤ 9.136%. The accuracy of the three-year model was R² = 0.893, RMSE = 1.092 g m⁻² and RE = 8.550% with eight variables. When each model was verified using the other data, the range of RE for two-year models was similar or increased compared with that for single-year models. In the partial least square regression (PLSR) model for the validation, the accuracy of two-year models was also better than that of single-year models as R² ≥ 0.699, RMSE ≤ 1.611 g m⁻² and RE ≤ 13.36%. The accuracy of the three-year model was R² = 0.837, RMSE = 1.401 g m⁻² and RE = 11.23% with four latent variables. When each model was verified, the range of RE for two-year models was similar or decreased compared with that for single-year models. The similarities and differences of loading weights for each latent variable depending on hyperspectral reflectance might have affected the regression coefficients and the accuracy of each prediction model. The accuracy of the single-year MLR models was better than that of the single-year PLSR models. However, accuracy of the multi-year PLSR models was better than that of the multi-year MLR models. Therefore, PLSR model might be more suitable than MLR model to predict the nitrogen contents at the heading stage using the hyperspectral reflectance because PLSR models have more sensitive than MLR models for the inhomogeneous results. Although there were differences in the environmental variables (ADR and ECT), it is possible to establish a general-purpose prediction model for nitrogen content at the heading stage using airborne hyperspectral remote sensing.     

Cultivation offers effective management of subterranean clover damping‐off and root disease

Soil cultivation studies involving subterranean clover pastures were undertaken utilizing field cores from five farms and two in‐field trials. Tap and lateral root disease in cores was less (p < .001) severe and root and shoot weights greater (p < .001) following simulated cultivation. Germination and severity of root disease were both affected (p < .005) by three‐way interactions with cultivation, cultivar and field site. Cultivation in cores suppressed tap root disease for cultivars Meteora and Riverina across the five sites and Seaton Park for two sites. In‐field trials confirmed cultivation reduces root disease severity and increases germination and plant productivity. The best in‐field treatment was cultivation + fumigation that reduced (p < .05) tap and lateral root disease and increased nodulation and root and shoot weights for Riverina, Seaton Park and Woogenellup. Cultivation + fumigation also increased (p < .05) germination for Woogenellup and Seaton Park. There were negative correlations (all p < .001) between tap and lateral root disease with nodulation (R² = .85, R² = .58, respectively); tap root disease with root and shoot weight (R² = .58, R² = .854, respectively); and lateral root disease with root and shoot weight (R² = .83, R² = .64 respectively). This study highlights the close relationship between severe root disease and reduced nodulation, likely explaining much of the widespread poor nodulation in subterranean clover pastures. This study confirms that damping‐off and root disease can be mitigated by cultivation, offering producers flexibility in disease management, especially where autumn–winter feed shortages occur on a regular basis.     

Prediction of legume silage digestibility from various laboratory methods

The potential of different laboratory methods to predict legume silage organic matter digestibility (OMD) in vivo was evaluated by using data from thirty‐three pure legume silages in seven experiments. The samples were analysed for crude protein concentration, cell wall composition, in vitro digestibility by the methods of Tilley and Terry [Journal of the British Grassland Society, 18 (1963) , 104–111; OMD_T&T], pepsin‐cellulase solubility (OMS) and gas production (OMD_GAS), and for indigestible neutral‐detergent fibre concentration in situ (INDF). The relationships between the results obtained by the laboratory methods and in vivo OMD, all expressed as ratios, were studied using linear univariate regression models with experiment as a random variable (mixed model). Legume silage digestibility could be estimated with acceptable accuracy by different in vitro methods, but not from the chemical composition of the samples. The highest accuracy in OMD prediction was found with OMS (RMSE 0·0113; R² = 0·965) followed by OMD_GAS (RMSE 0·0149; R² = 0·944), OMD_{T& T} (RMSE 0·0149; R² = 0·940) and INDF (RMSE 0·0168; R² = 0·925). The relationships between the in vitro methods and in vivo digestibility are not universal, and should be determined separately for each laboratory and type of forage. Part of the error in OMD prediction can be attributed to errors in in vivo OMD determination.     

Effect of Quinoa Seeds (Chenopodium quinoa) in Diet on some Biochemical Parameters and Essential Elements in Blood of High Fructose-Fed Rats

The effect of Chenopodium quinoa seeds on lipid profile, glucose level, protein metabolism and selected essential elements (Na, K, Ca, Mg) level was determined in high—fructose fed male Wistar rats. Fructose decreased significantly LDL [42%, p < 0.01] and activity of alkaline phosphatase [20%, p < 0.05], and increased triglycerides level [86%, p < 0.01]. The analysis of blood of rats fed quinoa indicated, that these seeds effectively reduced serum total cholesterol [26%, p < 0.05], LDL [57%, p < 0.008] and triglycerides [11%, p < 0.05] when compared to the control group. Quinoa seeds also significantly reduced the level of glucose [10%, p < 0.01] and plasma total protein level [16%, p < 0.001]. Fructose significantly decreased HDL [15%, p < 0.05] level in control group but when the quinoa seeds were added into the diet the decrease of HDL level was inhibited. Quinoa seeds did not prevent any adverse effect of increasing triglyceride level caused by fructose. It was shown in this study that quinoa seeds can reduce most of the adverse effects exerted by fructose on lipid profile and glucose level.     

An SNP downstream of the OsBEIIb gene is significantly associated with amylose content and viscosity properties in rice (Oryza sativa L.)

Allele mining in starch synthesis-related genes (SSRGs) has facilitated the discovery of desired natural sequence variations for eating quality in rice. This study investigated the sequence variations from 10 SSRGs, and further evaluated their relationship with the amylose content (AC) and rapid viscosity analysis profiles in a global collection of rice accessions by association mapping (AM). In total, 83 sequence variations were found in 10 sequenced amplicons, including 73 single nucleotide polymorphisms (SNPs), eight insertion-deletions (InDels) and two polymorphic simple sequence repeats (SSRs). Four subpopulations were identified by population structure analysis based on 170 genome-wide SSR genotypes. AM revealed 11 significant associations between three phenotypic indices and three sequence variations. One SNP with a g/c transversion at the 63rd nucleotide downstream of the OsBEIIb gene termination codon on rice chromosome 2 was significantly associated with multiple trait indices in both the general linear and mixed linear models (GLM and MLM), including the final viscosity (p < 0.001, R² = 23.87%) in both 2009 and 2010, and AC (p < 0.01, R² = 11.25%) and trough viscosity (p < 0.01, R² = 20.43) in 2010. This study provides a new perspective of allele mining for breeding strategies based on marker-assisted selection.     

An SNP downstream of the OsBEIIb gene is significantly associated with amylose content and viscosity properties in rice (Oryza sativa L.)

Allele mining in starch synthesis-related genes (SSRGs) has facilitated the discovery of desired natural sequence variations for eating quality in rice. This study investigated the sequence variations from 10 SSRGs, and further evaluated their relationship with the amylose content (AC) and rapid viscosity analysis profiles in a global collection of rice accessions by association mapping (AM). In total, 83 sequence variations were found in 10 sequenced amplicons, including 73 single nucleotide polymorphisms (SNPs), eight insertion-deletions (InDels) and two polymorphic simple sequence repeats (SSRs). Four subpopulations were identified by population structure analysis based on 170 genome-wide SSR genotypes. AM revealed 11 significant associations between three phenotypic indices and three sequence variations. One SNP with a g/c transversion at the 63rd nucleotide downstream of the OsBEIIb gene termination codon on rice chromosome 2 was significantly associated with multiple trait indices in both the general linear and mixed linear models (GLM and MLM), including the final viscosity (p < 0.001, R² = 23.87%) in both 2009 and 2010, and AC (p < 0.01, R² = 11.25%) and trough viscosity (p < 0.01, R² = 20.43) in 2010. This study provides a new perspective of allele mining for breeding strategies based on marker-assisted selection.     

Modelling photoperiod and temperature responses of flowering in quinoa (Chenopodium quinoa Willd.)

Two modelling approaches were used to quantify photoperiod and temperature responses of time from emergence to visible flower buds in nine quinoa (Chenopodium quinoa Willd.) cultivars. The first, non-interactive model, considers temperature and photoperiod responses as independent, and the threshold photoperiod, critical photoperiod, and base temperatures as constants. The second, interactive model, considers these attributes as variable, and allows for interaction between photoperiod and temperature responses. Controlled-environment experiments with a factorial combination of temperature and photoperiod provided information on responses, and data from field experiments were utilized in tests of the predictive capacity of the models.The two models were very similar in their goodness of fit and predictive capacity, but testing revealed that some assumptions about the interactive model were not fulfilled, whereas the non-interactive model is more consistent with the data. Both the models failed to predict dates of visible flower buds when average temperatures during the phase were >20°C; it is proposed that interaction between irradiance receipt and high temperature in controlled environments result in lower optimum temperatures there than in the field. Differences between field data and predicted values were eliminated when predictions were recalculated assuming no optimum for the temperature response.All nine cultivars examined are short-day plants. A juvenile sub-phase was observed in the six cultivars for which it was tested; and its duration was negatively associated with the latitude of origin of the lines (R² = 0.9, p < 0.05). Photoperiod sensitivity was negatively associated with the latitude of origin of the lines (R² = 0.55, p < 0.05) and positively associated with duration of the basic vegetative phase (minimal time between emergence and visible flower buds) (R² = 0.55, p < 0.05) using the non-interactive model. Photoperiod and temperature response parameters were not significantly associated with the latitude of origin for the interactive model (p > 0.05).     

A meta‐analysis examining lactic acid bacteria inoculants for maize silage: Effects on fermentation,aerobic stability,nutritive value and livestock production

A database containing 140 articles published in journals (731 treatment means evaluated) was used to examine the effect of different lactic acid bacteria (LAB) on fermentation, chemical composition and aerobic stability of maize (corn) silage. Compared with the control, dry matter (DM) loss increased by 8% and 50% (p < .01) due to inoculation of maize silage with either homolactic LAB (^hoLAB) or heterolactic LAB (^heLAB). In vitro DM digestibility of maize silage increased only with ^hoLAB inoculation (+2.22%; p < .01). The ^heLAB inoculation increased (p < .01) the aerobic stability of maize silage by 71.3 hr. To investigate the effect of silage inoculation on livestock production, a second database comprising 35 articles [99 treatment means evaluated based on results from 648 cattle (429 beef cattle and 219 dairy cows) and 298 sheep] was used. Inoculation of maize silage with either ^hoLAB or ^heLAB did not affect milk yield (p > .05), but their combination (^mixLAB) depressed milk yield (–2.5 kg/day; p < .01). Inoculation with ^hoLAB increased DM intake in sheep (+0.15 kg/day; p = .02), but decreased it in beef cattle (–0.26 kg/day; p = .01) without affecting average daily gain for both sheep and beef cattle (p ≥ .06). In conclusion, fermentative loss increased regardless of the bacterial inoculant used, while aerobic stability increased mainly by using ^heLAB. Benefits from ^hoLAB inoculation on animal performance were noted only for feed intake in sheep, while productive performances of dairy cows and beef cattle were not improved.     

Mixtures of grasses: An alternative to traditional pasture monocultures in the tropics

A two-year experiment assessed herbage production and above- and below-ground characteristics of a highly productive monoculture (‘BRS Zuri’ guineagrass [Panicum maximum Jacq.]) and two mixtures of three grasses (Mixture 1: ‘BRS Zuri’ guineagrass, ‘BRS Xaraés’ palisadegrass [Brachiaria brizantha Stapf.], and ‘Basilisk’ signalgrass [Brachiaria decumbens Stapf.]; Mixture 2: ‘BRS Quênia’ guineagrass [Panicum maximum Jacq.], ‘Marandu’ palisadegrass [Brachiaria brizantha Stapf.], and ‘BRS Paiaguás’ palisadegrass [Brachiaria brizantha Stapf.]), cultivated in the Brazilian tropical savanna. Mixtures 1 and 2 were subjected to two grazing intensities (removal of 40 or 60% of pre-grazing height) and ‘BRS Zuri’ guineagrass monoculture was defoliated to a single grazing intensity of 50%. Treatments were randomly assigned to fifteen 0.25-ha plots and managed under intermittent stocking by cattle. Herbage accumulation rate was similar among pastures and years (p > .1). The root mass in the tussocks did not differ (p > .1), with mean values ranging between 0.62 to 1.81 kg DM m⁻². Root density in the tussock interspaces was greater in the mixtures (p < .001), regardless of seasons (p = .405) and years (p = .292). The mixtures were dominated by guineagrass (70%) and palisadegrass (30%) at the end of the experiment, with the population of ‘Basilisk’ and ‘BRS Paiaguás’ being completely suppressed throughout the experimental period. Mixing guineagrass and brachiariagrasses can be an alternative to the traditional pastoral systems in the tropics, as it does not compromise herbage production and presents a capacity to produce more roots than a very productive monoculture of ‘BRS Zuri’ guineagrass.     

Cinnamon Shows Antidiabetic Properties that Are Species-Specific: Effects on Enzyme Activity Inhibition and Starch Digestion

Despite considerable research the evidence around the antidiabetic properties of cinnamon remains equivocal, and this may be due to varietal differences which is an aspect that is understudied. This study systematically compared the anti-hyperglycaemic properties of the four major commercial cinnamon types used around the world (Chinese; Cinnamomum cassia [CC], Indonesian; C. burmanii [IC], Vietnamese; C. loureirii [VC], and Ceylon; C. zeylanicum [SC]). LC-MS analysis showed distinct diffrences in the phytochemical profiles of cinnamon with SC showing the lowest coumarin concentration. CC and IC had the highest polyphenol levels and antioxidant potential, and all four types differed significantly in their content (P?<?0.001). All cinnamon types showed potent species-specific effects on starch digestion enzyme activity inhibition (P?<?0.001), CC was most effective against α-amylase and all four strongly inhibited α-glucosidase compared to acarbose. Cinnamon significantly reduced starch breakdown during oral (P?=?0.006) and gastric (P?=?0.029) phases of gastro-intestinal digestion with IC and SC showing consistent effects. No effects of cinnamon were seen in the intestinal phase. IC, VC and SC showed the greatest potential to inhibit formation of advanced glycation endproducts (AGEs) during digestion. In conclusion, cinnamon demonstrates anti-hyperglycaemic properties, however effects are species-specific with best overall properties seen for Ceylon cinnamon.

     

The effects of concentrate energy source and protein content on milk production in cows given grass silage ad libitum

Sixteen Friesian cows were given four dietary treatments in a 4 × 4 Latin square experiment with a 2 × 2 factorial arrangement of treatments. The diets consisted of grass silage ad libitum plus 2 kg of hay per day and two types of concentrates of either barley or a mixture of barley, oats and fibrous by-products [200,200 and 600 g per kg dry matter (DM) respectively], with two protein contents. For the low-protein diets, barley- (B) and fibre-based (F) concentrates were given without protein supplements, while for high-protein diets 1 kg of both concentrates was replaced with fish-meal (FM). The concentrates were given at the rate of 9kgd^-1 for the cows (n= 12) and 8kg d^-1 for the heifers (n= 4). The cows given the F diets tended (P < 0·10) to have a greater silage dry matter intake and produced 1·5kg d^-1 more (P < 0·05) milk with a lower (P < 0·05) protein content than those given the B diets. Increasing dietary crude protein concentration with FM had no effect on feed intake but resulted in significant increases in milk yield (P < 0·01), milk protein content (P < 0·05) and yields of milk constituents. The response in milk yield to FM tended to be greater with barley than with fibrous supplement (+2·5 vs+ 1·5kgd^-1). Compared with B diets, the greatest relative increase occurred in lactose yield (0·07) when the corresponding F diets were fed, while FM produced the greatest response in protein yield (0·12). The calculation of the utilization of metabolizable energy (ME) for milk production showed that both the feeding of a concentrate consisting of different carbohydrate sources and inclusion of fish meal improved the utilization of ME, the effects being partially additive. It is concluded that the nutrient supply to the cow's tissues can be modified by the source of carbohydrate and protein supplementation as indicated by different responses in the yield of milk constituents. The production response to protein supplementation may depend on the source of carbohydrate in the concentrate.     

Relationship between P and N concentrations in maize and wheat leaves

Simple plant-based diagnostic tools can be used to determine crop P status. Our objectives were to establish the relationships between P and N concentrations of the uppermost collared leaf (P_L and N_L) of spring wheat (Triticum aestivum L.) and maize (Zea mays L.) during the growing season and, in particular, to determine the critical leaf P concentrations required to diagnose P deficiencies. Various N applications were evaluated over six site-years for wheat and eight site-years for maize (2004-2006) with adequate soil P for growth. Phosphorus and N concentrations of the uppermost collared leaf were determined weekly and the relationships between leaf N and P concentrations were established using only the sampling dates from the stem elongation stage for wheat and from the V8 stage of development for maize. Leaf P concentration generally decreased with decreasing N fertilization. Relationships between P_L and N_L concentrations (mg g⁻¹ DM) using all site-years and sampling dates were described by significant linear-plateau functions in both maize (P_L = 0.82 + 0.089 N_L if N_L ≤ 32.1 and P_L = 3.7 if N_L > 32.1; R² = 0.41; P < 0.001) and wheat (P_L = 0.02 + 0.106 N_L if N_L ≤ 33.2 and P_L = 3.5 if N_L > 33.2; R² = 0.42; P < 0.001). Variation among sampling dates in the relationships were noted. By restricting the sampling dates [413-496 growing degree days (5 °C basis) in wheat (i.e., stem elongation) and 1494-1579 crop heat units in maize (i.e., silking), relationships for wheat (P_L = 0.29 + 0.073 N_L, R² = 0.66; P < 0.001) and maize (P_L = 1.04 + 0.084 N_L, R² = 0.66; P < 0.001) were improved. In maize, expressing P and N concentrations on a leaf area basis (P_LA and N_LA) at silking further improved the relationship (P_LA = 0.002 + 0.101 N_LA, R² = 0.80; P < 0.001). Predictive models of critical P concentration as a function of N concentration in the uppermost collared leaf of wheat and maize were established which could be used for diagnostic purposes.     

Assessment of Potato Dry Matter Concentration Using Short-Wave Near-Infrared Spectroscopy

The utility of short-wavelength near-infrared spectroscopy (over the wavelength region 750–950 nm), used in a partial transmittance optical geometry, was assessed as a means of estimating the dry matter concentration of potato tubers. The sampling optics did not involve contact with the sample, and could be used on a moving stream of product. A prediction accuracy of R ² (correlation coefficient of determination) of 0.85 with a root mean square error of prediction (RMSEP) of 1.52% for intact, whole tubers and R ² = 0.95 and RMSEP = 0.50% for sliced tubers was achieved. We conclude that short-wavelength near-infrared technology using a partial transmittance optical sampling geometry can be a useful tool for rapid assessment of tuber dry matter concentration prior to processing.     

Investigations into the thiamine and riboflavin content of malt and the effects of malting and roasting on their final content

A simple and effective method for the analysis of thiamine (B₁) and riboflavin (B₂) vitamers, in unmalted and malted grains, by high performance liquid chromatography (HPLC) has been developed. This method makes use of trichloroacetic acid (TCA) as an extraction medium to effectively clean up the sample and analyze the vitamer content with high accuracy (R² > 0.9992; %RSD < 5%). This method was employed in the routine analysis of a wide range of malted and unmalted grains, and it was found that lighter colored malts contain higher concentrations (2–5 times higher) of thiamine and riboflavin vitamers compared to darker colored malts. The malting process has no effect on the overall riboflavin content; however, both steeping and kilning processes cause increases in thiamine vitamer content. Roasting profiles show that thiamine and riboflavin vitamer concentrations are significantly effected when the temperature reaches 120 °C, which explains why roasted products have lower vitamer content than paler malts.     

Irrigation effects on elephant grass morphology,and its genotype-dependent responses

This study hypothesised that different elephant grass genotypes respond differently in terms of their morphological development to irrigation. The objective of this study was to evaluate how water availability could affect the morphological development of different genotypes of elephant grass. The treatments consisted of four genotypes of elephant grass [Cenchrus purpureus (Schumach.) Morrone], managed under irrigation or not, during a two-year field trial. The experiment was arranged in a randomised block design in a split-plot, with four replicates per treatment (n = 4). Between the genotypes, two of them were classified as tall-sized (IRI 381 and Elephant B), and two as dwarfs (Taiwan A-146 2.37 and Mott). A three-way interaction between season, irrigation, and genotype affected the variables plant height, stem diameter, number of dead tillers, light interception (LI), and leaf area index (LAI) (p < .05). The morphological development of all genotypes was negatively impacted by the dry season. All genotypes grew taller during the rainy season (p < .05) and had a greater light interception (62%–80%) compared to the dry period (28%–59%). The genotype Mott, showed comparable LI and LAI to the tall-sized genotypes, whilst Taiwan A-146 2.37 was characterized by the lowest values, despite being the genotype with the greatest tiller density (60 tillers m⁻²) (p < .05). The use of irrigation mitigated the negative effects of the dry season on the plants, however, the type and size (tall vs. dwarf) of the elephant grass genotypes defined the extent of the responses to the use of the irrigation during the dry period.     

The influence of weight of ryegrass per unit area and treatment at and after mowing on rate of drying

The rates of drying of perennial ryegrass, subjected to different treatments at mowing and after mowing, were assessed in the field by weight change of grass fresh weight in wire-mesh trays over 3·5 d (76 h). In a 5 × 3 × 3 factorial experimental design, the effects of five weights of grass per unit area [1·5, 3, 6, 12 and 24 kg fresh material (FM) m^?2], three treatments at mowing (no treatment, mower-conditioned, flail-treated) and three treatments after mowing (no treatment, inverted, mixed) were examined. The experiment was replicated twice on 16 occasions in 1992 at the Agricultural Research Institute of Northern Ireland. This gave a total of thirty-two replicates per treatment. The trays were weighed at 2-h intervals from 09.00 to 17.00 h each day. Data sets were restricted to rain-free days and also to the first day after mowing (day 1). On day 1, grass weight per unit area was a major factor dictating drying; reducing the grass weight per unit area of unconditioned grass from 6 to 3 kg FM m^?2 increased grass drying rate by 47%. There was no significant (P > 0·05) benefit over the untreated grass on day 1 from mixing or turning mower-conditioned or the unconditioned grass. Mixing of the flail-treated grass improved drying rate significantly (P > 0·001) over the control. Over the whole 76-h period, the relative benefit from either mower conditioning or flail treatment over no treatment was dependent upon both grass weight per unit area and initial dry-matter (DM) concentration. At higher initial DM concentrations (>150 g kg^?1) and lower grass weights (<6 kg FM m^?2) both mower conditioning using a nylon brush type conditioner and intensive conditioning by flail treatment gave substantial increases in drying over no treatment. Moisture regain of grass exposed to overnight dew was small. Rain had a much greater effect than dew on subsequent moisture regain. Unconditioned grass at 12 kg m^?2 retained 82% less water following rainfall than unconditioned grass at 3 kg m^?2.     

Evaluating cultivars in unbalanced on-farm participatory trials

Data from on-farm participatory varietal trials are often highly unbalanced. Using statistically robust techniques we showed that such data are amenable to sophisticated analyses. Data from 38 varieties of wheat tested in 44 mother trials (all test varieties compared) and 663 baby trials (single entries compared with local checks) over 6 years in Lunawada, Gujarat, India, were analysed. Two combined analyses for grain yield were performed by using either a fixed effects model within general linear modelling (GLM) or a mixed effects model using the restricted maximum likelihood (REML) procedure. Both analyses made substantial adjustments over the raw means and the estimated mean values from the two analyses for the fixed set of tested varieties were highly correlated (R² = 0.89) as were the t-value that compared test cultivars with the check variety (R² = 0.96). The GLM detected 15 varieties and REML detected 12 varieties that were significantly superior to the check cultivar with 11 varieties in common. Both methods were similar in identifying the top-most promising varieties but REML was more conservative (i.e. more subject to Type II error) with larger standard errors (S.E.s) of variety means for n > 17. However, for both procedures the S.E.s increased sharply when the number of trials decreased below 17.