The platinum microelectrode technique for measuring oxygen flux in soils has been reviewed. Shortcomings in the existing technique and instrumentation have been discussed. The new instrumentation, electrode standardization and measurement techniques developed enable the method to be used with confidence in unsaturated soil systems. Measurements of oxygen flux index in four soil samples showed a very highly significant regression relationship between oxygen flux index and air space within the range 3 - 15% air space on each soil. There was no significant difference in the regression relation between soils. A field penetrometer, designed to measure the presence and strength of subsurface pans in field soils has been described. Measurements with the penetrometer on three depth of ploughing treatments (100, 230 and 355 mm) on tillage trials at four sites with different clay contents showed that hard layers were present on all the treatments. Except on the shallowest ploughing depth treatment on the fine-textured soil, where the pan was 225 mm below the nominal ploughing depth, the hard layers were present between a few mm and 150 mm below the nominal ploughing depth, and had strengths of between 16 and 24 bars. The theory and factors affecting measurement of soil strength with needle penetrometers have been investigated. The design and operation of a laboratory penetrometer used to measure soil strength under closely controlled laboratory conditions has been discussed. Physical factors likely to affect root growth, viz. soil texture, air space, bulk density, soil strength and available moisture, have been measured in a comprehensive range of undisturbed cores taken from the four tillage trials. High soil strength is considered as being the soil physical factor most likely to restrict root growth in these soils. Physical factors affecting soil strength have been investigated. Soil strength is shown to be highly dependent upon bulk density, matric potential and soil texture. The hard pans shown to exist in all the tillage trial soils exhibit many of the characteristics of tillage pans, but their existence cannot be attributed exclusively to the ploughing depth treatments imposed in the tillage trials. Rather, the pans have resulted from a combination of interacting factors, including the previous history of the soils, the imposed tillage treatments, crop, and climatic factors. A study of some of the data from the literature on root growth and soil strength shows that root growth is severely restricted by soil strengths of the order of 20 to 30 bars. In order to determine whether root growth was being restricted in the tillage trials, root profiles were extracted from one of the trials. These showed that the pans severely restricted root growth. Analysis of maize yield data from the tillage trials showed that on the three coarse-textured sites yield increased with increased depth of ploughing, and that there was a marked seasonal effect, ploughing depth having a relatively greater effect on maize yield in dry seasons than in wet. On the fine-textured site, however, where no pan existed near the surface in the shallow ploughed treatment, the ploughing depth effect was not significant, nor was there any marked seasonal effect of ploughing depth on maize yield. Evidence presented shows that the pans, by restricting root growth are reducing the amount of water available to the plant. This effect is greater in dry seasons, and in soils with low available water . / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 1976.

The main effects of lime, aluminium, iron and manganese were studied in field and greenhouse grown tobacco; relations between soil and plant measurements were examined. Ground limestone, ground mixed lime, ground dolomite and slaked lime at rates equivalent to 1,000 and 2,000 lb. CaC0₃/acre increased yield and quality of flue-cured tobacco both on Triassic and granite sands, whether applied early (February/March) or late (September); the highest rate and late application were often best. Yields increased with 4,000 and 6,000 lb. dolomite/acre applied late, but quality decreased when the pH was about 6.0. Lime did not affect leaf maturity as reflected by nitrogen and reducing sugars concentration . Where leaf discolouration (slate) occurred, the best quality and least discoloured leaf had the lowest manganese concentration and was grown on limed soil. On a very acid and probably nitrogen deficient soil, lime, borax and nitrogen (nitrate only tested) reduced the discolouration and improved the quality, but potassium sulphate increased discolouration and decreased quality. Calcium concentration in the leaf was increased by lime, particularly calcitic materials, and magnesium concentration was increased by dolomite. Lime also increased the filling value and petroleum ether extract, but decreased manganese, boron, chloride and sometimes potassium, and had no effect on phosphorus, nitrogen, aluminium, iron, crude fibre, nicotine, reducing sugars and equilibrium moisture. The inorganic composition of greenhouse plants was similar; generally gypsum increased calcium concentration more than calcium carbonate but it did not affect manganese concentration, which was decreased by calcium carbonate. In the stem and roots of field grown plants (dolomite only tested), the concentration of magnesium was increased but the concentrations of calcium, potassium, aluminium and iron were unaffected. Although the concentration of nitrogen was increased and that of phosphorus was decreased in the stem, these were unaffected in the roots. Aluminium and iron behaved differently to other nutrient ions, being more concentrated in the roots than aerial plant parts. Boron and magnesium deficiencies were observed in a dry and wet year, respectively, suggesting that variable mineral deficiencies can affect responses to lime. Initially soil pH was affected more by source of lime, but later mostly by rates. Slaked lime increased the soil pH more than did ground limestone, mixed lime or dolomite. In a glasshouse experiment, pH was more important than calcium supply and in the field, the largest yields were often associated with the highest pH. In pot experiments, aluminium drastically reduced yields in nutrient solution but not in soils, whereas iron was more severe in soils; manganese had little effect on yield. Manganese was readily taken up and translocated to the tops, but aluminium and iron were mainly concentrated in the roots, as was found in field grown plants. Iron decreased manganese concentration in all plant parts and aluminium decreased calcium and manganese in nutrient solution only. Although aluminium and iron generally increased the concentration of phosphorus in the roots, they did not interfere with phosphorus transport in the plant. Manganese caused the leaf to become chlorotic and when no iron was present the upper leaves became yellow, and developed brown and white lesions. However, in soil grown plants, sufficient iron was present in the soil solution to prevent break down of tissue. Yellowing of the upper leaves also occurred when plants were grown in nutrient solution with aluminium and no iron; when both were present, the plants were darker in colour. Although aluminium damaged roots in nutrient solution, high rates of iron severely damaged leaves of plants grown in soil. Since the concentrations of aluminium, iron and manganese were decreased in the soil solution by liming, they were compared with plant growth and composition in 17 different soils, with and without lime. As was expected, lime increased soil pH. It also increased exchangeable calcium, but decreased exchangeable aluminium, iron and manganese; exchangeable magnesium and potassium and resin extractable phosphorus were not affected. As the Ratio Law does not hold for all Rhodesian soils, anion adsorption will be avoided if the soils are equilibrated with O.OOOSM CaC1₂; the concentrations of the cations in solution were affected in the same way as exchangeable cations, but phosphorus was increased. There was no relationship between yield of tobacco and its chemical composition. The correlations between soil solution data and plant composition were poor, except for manganese and phosphorus; the relation between Mn ppm. in plant vsa (superscript)Mn/a (superscript)Ca + Mg (enclosed in square root sign √) in solution, and P% vs pH₂ P0₄ or pH₂ P0₄ +½ pCa, were both curvilinear. On the other hand, all measurements of exchangeable cations were poorly correlated with plant composition. Finally yield was poorly correlated with soil solution data, and pH was as satisfactory as any other measurement tested. Manganese toxicity was observed on three soils, and a probable manganese deficiency on one. It was not possible to define a limit above which manganese toxicity occurred, but manganese deficiency developed at about 63 ppm. manganese. Variations in pH and the availability of aluminium, iron and manganese occurred when soils were incubated at about field capacity, generally the main effects having developed within seven days. In all soils, there was an initial increase in soil pH and a maximum value was reached in one to four days, decreasing by variable amounts with longer periods of incubation. Although the concentration of aluminium was larger than that of iron, the relation between both ions and soil pH was curvilinear, their concentrations increasing with decreasing pH. Increased temperature of incubation increased pH with a resultant decrease in the concentration of aluminium, but in one soil it appreciably increased the availability of iron in the early periods of incubation. Autumn and spring ploughing did not affect subsequent pH or the concentration of aluminium and iron in the soil solution. Manganese concentration varied from soil to soil and was not related to soil pH. In most soils there was a decrease in manganese concentration with length of incubation and it decreased more rapidly the lower the initial concentration. Temperature effects were variable and moisture affected the behaviour of manganese more than temperature. These findings and the distribution of aluminium, iron and manganese in the plant helped to explain the poor correlations.

The effects of artificially supplied perennial water on soil properties, vegetation dynamics and the distribution of large herbivores was investigated in southeastern Zimbabwe. Data collection took place between March 1997 and July 1998. Water points were situated primarily on three different soil types (clay-loam, sandy-clay-loam and sand), and in four different vegetation types (Hill communities, Colophospermum mopane veld, Acacia nigrescens woodland and Albizia petersiana woodland). One water point in C. mopane veld (Bandama) had been closed two years prior to data collection, while another, in the Hill community (Manyoka), had been introduced, two years prior to data collection. Changes in physical (infiltration) and chemical (organic carbon and nutrients) properties of soils around water points were largely restricted to within 100 m of water. Chemical enrichment of the soil occurred only at water points that had been in place for more than two years. Soil surface conditions were altered to distances beyond 100 m from water. Manyoka (the new water point) was an exception, with extreme changes limited to within 100 m of water. Herbaceous and woody species composition changed in response to distance from water with changes best described by asymptotic equations. Changes in species composition of the woody component appeared to be longer lasting than changes to the herbaceous component. Most perennial grass species declined close to water, but Urochloa mosambicensis increased close to water in areas outside of the Hills. Herbaceous species diversity was adversely affected by distance from water on sandy soils (Hill communities and A. petersiana woodland), but was largely unaffected on clay-loam (A. nigrescens woodland) and sandy-clay-loam soils (C. mopane veld). Woody species composition and density was altered out to 500 m from perennial water in Acacia nigrescens woodland on clay-loam soils. Results suggest that this vegetation type may be susceptible to bush encroachment close to water. Trends in woody canopy utilisation were generally similar to trends in woody species composition, and it is proposed that the former may be used to indicate future changes in the latter. Conversion of trees to shrubs was highest at Manyoka (the new water point) indicating that woody destruction by elephants is extreme during the initial years following water introduction. Large herbivore biomass was greatest close to water (< 1 km) during the dry season but not during the wet season. Herbivore species distributions appeared to be influenced by the position of perennial water, but since all range was within easy access of water, it is unlikely that animal distributions were constrained directly by the position of surface water. It is more likely that herbivores were spatially separated on the basis of habitat type. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 1999.