SOIL PHYSICAL FACTORS LIMITING ESTABLISHMENT OF SORGHUM AND COWPEA IN2 CONTRASTING SOIL TYPES IN THE SEMIARID TROPICS

Experiments were carried out to study which soil physical factors, if any, would restrict germination and emergence of sorghum (sorghum bico lor (L.) Moench.) and cowpea (Vigna unguiculata (L.) Walp.), under fie ld conditions in Tanzania. The study also aimed to consider the useful ness of mulching as a strategy to overcome such problems, and to contr ibute to the development of a strategy for modelling seedling emergenc e in seedbeds in the field. Sorghum and cowpea were planted at two sit es in experimental plots which simulated local farming practices, with both bare soil and mulched treatments. The soil types were a sandy cl ay loam (Paleustaulf) and a sandy loam (Typic Untorthent), both of whi ch were hardsetting. Sequential, daily harvests of the seedlings were related to measurements of soil matric potential and temperature at di fferent depths. Penetration resistances (PRs) were also estimated from matric potentials, and strength characteristics obtained from the sam e soils. Germination of both crops was high (> 85%) in all of the tria ls. Emergence of sorghum seedlings was poor (0-30%) in all cases. In t he sandy clay loam it was primarily the large mechanical impedance tha t prevented emergence, whereas a combination of high temperature, wate r stress and mechanical impedance all contributed to poor emergence in the sandy loam. Cowpea emergence was higher but still only 56% or les s in the sandy clay loam due to the increase in mechanical impedance t o shoot growth as the soil dried. In the sandy loam, emergence of cowp ea was very good (> 85%). The greater percentage emergence of cowpea w as related to its greater initial rates of root and shoot growth, whic h allowed it to avoid the greater stresses caused by temperature, drou ght and mechanical impedance, experienced by the sorghum. In compariso n to sorghum, the larger cowpea shoots were also able to crack a great er thickness of overlying soil when they approached the soil surface a nd hence overcome mechanical impedance that might otherwise have preve nted emergence. The mulched treatment reduced soil surface temperature s by up to 20 degrees C, and delayed the onset of water stress and hig h mechanical impedance. However, in comparison to the bare soil treatm ent, the overall effect of these factors varied between a decrease in the percentage emergence of cowpea in the sandy clay loam and an incre ase in emergence of sorghum in the sandy loam. It is suggested that a modelling approach is required to understand fully these interactions.