A mapping population of 114 lines from Bala×Azucena was grown under drought stress at two field sites with contrasting soil physical properties. Drought was imposed between 35 and 65 days after sowing (DAS) and root density at 35cm depth was measured 70 DAS. Leaf rolling, leaf drying and relative water content were recorded as indicators of drought avoidance. Root density correlated with indicators of drought avoidance. Two significant and two putative quantitative trait loci (QTLs) for root density and 28 QTLs for drought avoidance were identified. Most QTLs did not agree between sites. There was also reasonable agreement between leaf-drying QTLs and previously reported root-growth QTLs detected under controlled conditions (in contrast to a previous screen on soil with a higher penetration resistance). These data also reveal QTL × environment interaction, which will need to be understood more clearly if progress towards breeding for drought resistance via alterations of root morphology is to be achieved.

In this study we designed and tested a methodology to minimize the variation of soil matric potential due to changes in soil water content, thus achieving a close control over the water regime in a microcosm, used to study the biocontrol of Pythium sp. infection of seeds by antagonistic bacteria. The variation of volumetric water content and matric potential were monitored at different depths during an average experimental period of 14 days in soil, contained in replicate 1 315 cm(3) microcosms, each sown with 16 sugar beet (Beta vulgaris) seeds and placed in a phytotron. Several experiments with target soil matric potentials of -10, -100 and -300 kPa were performed. It was found possible to maintain a matric potential of about -10 kPa in soil with minimum water content gradients between different parts of the microcosm by watering uniformly at each depth. With all appropriate watering and sampling procedure it was also possible to control psi(m) between -150 and -50 kPa, with a target value around -100 kPa; and to control psi(m) the range between -250 and -375 kPa for a target value of -300 kPa. In all experiments spatial and temporal variability of matric potential was larger at the top than in the Middle and in the bottom of microcosms. Variability in matric potential between replicate microcosms within each experiments was attributed to slight variation in packing, soil texture and drying rate, initial VWC.

Soil water and temperature regimes in the tropical moist forest on Barro Colorado Island, Panama, were simulated directly from meteorological data using the model SWEAT. Separate field observations from root-exclusion, litter-removal and control treatments in one small and one large forest gap were used for calibration and validation. After irrigating all treatments to field capacity, soil matric potential and temperature were measured over 17 days at four depths ≤50mm using the filter-paper technique and bead thermistors. Understorey environments were also simulated under the same initial conditions. The results suggest that three distinct scenarios, controlled by gap size, describe how the above- and below-ground processes controlling soil drying are coupled: (1) in the large gap, root water extraction by surrounding trees is negligible so soil drying is dominated by evaporation from the soil surface. Soil temperature is dominated by direct solar heating and cooling due to evaporation. (2) In the small gap, root water extraction dominates soil drying with soil evaporation playing a minor role. Soil temperature is still dominated by direct sunlight with some cooling due to evaporation. (3) In the understorey, root water extraction dominates soil drying. Soil temperature is dominated by heat conduction from deep soil layers with some evaporation and sensible heat transfer. The contrasting soil drying regimes imposed by variation in canopy structure enhance micro-environmental heterogeneity and the scope for differential germination and seedling establishment in coexisting tropical tree species.

Soil temperature is an important variable governing plant growth and development. Studies were conducted under laboratory conditions to determine the effect of soil temperature on root and shoot growth of cotton during emergence. Cotton seedlings were grown for 192 h at 20, 32 and 38 °C in soil packed in 300 mm long and 50 mm diameter cylinders. The data indicated that the longest roots (173 mm) as well as shoots (152 mm) were recorded at 32 °C followed by 20 (130 mm root and 82 mm shoot) and 38 °C (86 mm root and 50 mm shoot). Roots grown at 20 and 38 °C were 20% and 50% shorter, respectively, than those grown at 32 °C after 192 h. Roots and shoots exhibited the lowest length and dry biomass at 38 °C. Shoot lengths grown at 20 (74 mm) and 38 °C (51 mm) were 44% and 61% shorter than those grown at 32 °C (131 mm) after 180 h growth period, respectively. Growth at all three temperatures followed a similar pattern. Initially there was a linear growth phase followed by the reduction or cessation of growth. Time to cessation of growth varied with temperature and decreased faster at higher temperatures. Sowing of cotton should be accomplished before seedbed reaches a soil temperature (≥ 38 °C) detrimental for emergence. Further, the seedbeds should be capable of providing sufficient moisture and essential nutrients for emerging seedling before its seed reserves are exhausted to enhance seedling establishment in soil.

Altering root system architecture is considered a method of improving crop water and soil nutrient capture. The analysis of quantitative trait loci (QTLs) for root traits has revealed inconsistency in the same population evaluated in different environments. It must be clarified if this is due to genotype x environment interaction or considerations of statistics if the value of QTLs for marker-assisted breeding is to be estimated. A modified split-plot design was used where a main plot corresponded to a separate experiment. The main plot factor had four treatments (environments), which were completely randomized among eight trials, so that each treatment was replicated twice. The sub-plot factor consisted of 168 recombinant inbreed lines of the Bala x Azucena rice mapping population, randomly allocated to the seven soil-filled boxes. The aim of the trial was to quantify QTL x environment interaction. The treatments were chosen to alter partitioning to roots; consisting of a control treatment (high-soil nitrogen, high light and high-water content) and further treatments where light, soil nitrogen or soil water was reduced singly. After 4 weeks growth, maximum root length (MRL), maximum root thickness, root mass below 50 cm, total plant dry mass (%), root mass and shoot length were measured. The treatments affected plant growth as predicted; low nitrogen and drought increased relative root partitioning, low-light decreased it. The parental varieties Bala and Azucena differed significantly for all traits. Broad-sense heritability of most traits was high (57-86%). Variation due to treatment was the most important influence on the variance, while genotype was next. Genotype x environment interaction was detected for all traits except MRL, although the proportion of variation due to this interaction was generally small. It is concluded that genotype x environment interaction is present but less important than genotypic variation. A companion paper presents QTL x environment analysis of data.

The effect of initial inoculum density of the antagonistic bacterial strains Pseudomonas fluorescens B5 and Pseudomonas corrugata 2140 (103 to 108 CFU per seed pellet) on sugar beet seedling colonisation, in situ bioluminescence and antagonistic activity towards Pythium ultimum was investigated. Populations of the bacteria colonising sugar beet root systems approached an apparent carrying capacity of 105 to 106 CFU per plant after 12 d growth, irrespective of inoculum density. This meant an up to 320-fold population increase at low inoculum densities and a decrease at high densities. Population densities of both bacteria and their corresponding in situ bioluminescence (resulting from luciferase enzyme expression from the inserted luxAB genes) reached highest levels in the hypocotyl region and in the upper root region 0–20 mm below seed level (104–106 CFU/cm section, 101–103 RLU/cm section) and decreased with root depth. In situ bioluminescence, which indicates physiological activity, was measurable at lowest antagonist initial inoculum density (103 CFU per seed pellet) and did not increase significantly with increasing inoculum density. Bioluminescence was also significantly correlated with population density. For Pseudomonas fluorescens B5, the total population size per plant and downward colonisation of the root (below 40 mm depth) increased significantly with antagonist inoculum density applied to the seeds. For Pseudomonas corrugata 2140, no significant influence of initial inoculum density on root colonisation was observable. Survival and dry weight of sugar beet seedlings in Pythium infested soil increased significantly with increasing inoculum density of Pseudomonas fluorescens B5, whereas for Pseudomonas corrugata 2140, initial densities of 104 to 106 CFU per seed resulted in maximal survival of plants.

Two upland rice varieties, Azucena and Bala, were screened for root growth under droughted and irrigated treatments in two field sites at the West Africa Rice Development Association (WARDA) experimental farm, Côte d’Ivoire, during the dry season of 1999/2000. The sites were chosen to represent contrasting soil profile penetration resistance (PR) characteristics on upland sites, although both were relatively impeding. The number of nodal root axes per unit area passing through horizontal transects (root density) was counted at 35, 56, 77 and 98 days after sowing (DAS) at 10 cm depth intervals. Azucena consistently maintained a greater root density than Bala and a greater proportion of Azucena roots grew to 30 cm depth (22.7% vs. 8.4% at 77 DAS). There was little detectable effect of water regime on root distribution but evidence of lower root numbers at depths below 20 cm in the higher PR site was revealed. A site by variety by soil depth interaction suggests that Azucena roots are more strongly affected by very high PR than those of Bala. PR between 0–30 cm depth increased greatly with decreasing soil water content during the drought as the soil dried. This increase is likely to have prevented or greatly impaired further nodal root growth within this layer. At 40 cm depth, PR was high (3–4 MPa) but did not increase during the drought. At this depth root growth rate was likely to be greatly reduced despite the availability of water. These results demonstrate that varietal differences in root morphology characterised in the laboratory can be also detected in impeding field soils as differences in the density of roots at depth. Relatively poor root growth in these fields in the absence of drought was probably due to the high mechanical impedance and/or the physiological stress of the plants in the dry season. Our results indicate that high mechanical impedance was a more fundamental constraint on root growth than soil water availability during the drought. Thus, varietal differences in root penetration ability might be very important for drought avoidance in soils of this type.

The “ROOTMAP” model is designed to provide a realistic three-dimensional topological description of the growth of plant root systems. This paper describes and compares different techniques for obtaining ROOTMAP model input parameters from the measured properties of pea (Pisum sativum L.) root systems grown in aggregated soil at three values of penetration resistance (0.10, 0.25 and 0.51 MPa corresponding to bulk densities of 1.0, 1.2 and 1.4 Mg m-3 respectively). The elongation rate of the axis and the branching interval of laterals on the axis decreased with increasing penetration resistance. The resulting simulation gave a distribution of root contacts on the walls of a cylindrical container that was comparable to that observed. It was concluded that the model can simulate the growth and spatial distribution of the pea axis and first order laterals but simulation of the number and length of second order laterals is likely to be unreliable since their behaviour did not conform to the assumption of uniform elongation rate. The model now needs to be tested and validated under a range of experimental conditions.