In glasshouse studies the degree of control of Avena fatua increased as the period between application of difenzoquat and the onset of simulated rain was prolonged. 0.5 mm of 'rain' removed 29% of the herbicide deposit without adversely affecting performance at the recommended dose of 1 kg/ha. A further 30% was removed by 2.0 mm of 'rain', resulting in a marked reduction in acrivity. With lower amounts of 'rain' (0.16 mm), some of the spray deposit was redistributed from the leaf lamina to the leaf base/ligule area. The rate of penetration of 14C-difenzoquat was much greater when applied to the inner surface of the leaf sheath than when the leaf blade and outer sheath areas were treated. Translocation from the 'inner sheath' to other parts of the plant was up to 100 times greater than from other areas. It is suggested that the performance of difenzoquat is not reduced by low amounts of rain because: (1) the spray deposit is removed principally from the leaf blade, whilst in the more responsive liguleheaf sheath area the herbicide remains in solution, (2) the recommended dose of 1 kg/ha allows for some loss of active ingredient without reduction in performance. The practical implications of the work are discussed and further topics for research are outlined.
I N T R O D U C T I O NDifenzoquat is a water-soluble herbicide widely used for the control of Avena fatua in barley and wheat (Wingfield & Caldicott, 1975). In common with other foliage-applied compounds, its performance can be diminished by rainfall (Baldwin & Livingston, 1976) but quantitative information has hitherto been lacking. The work reported here consisted of two groups of experiments. The first dealt with the effects on spray retention and distribution and herbicide performance of different amounts of simulated rain applied at several time intervals after spraying. In the second group, the consequences of redistribution on penetration and movement of the herbicide were followed using 14C labelled difenzoquat.
M A T E R I A L S A N D M E T H O D S
Plant materialSeed of Avena fatua (L.) was obtained in autumn 1972 from the field. For consistency of plant material between experiments only large (20-25 mm) brown-husked seeds were selected. These were sown on moist filter paper and germinated at 15 OC. When the radicles had just emerged, the young seedlings were planted in a soil mix (one seedling/9 cm diameter pot, at a depth of 2.5 cm) and put in an unheated glasshouse where the temperature was 16 8 "C and the humidity 60 f 20%. The soil mix consisted of 1 part peat, 1 part sand and 4 parts of Begbroke sandy loam. The seedlings were not given any additional nutrients until growth stage 12 (Zadoks, Chang & Konzak, 1974), when weekly feeding was commenced using a proprietary