In January 2003 reports were received from the Breede Valley (about 100km north east of Cape Town) of glyphosate failing to control Conyza bonariensis at registered dosage rates. Seed from plants growing in a deciduous fruit orchard in the area were collected and germinated in pots. The seedlings were subsequently transplanted into pots and a month later the plants were subjected to a dose response trial with glyphosate. A control population of the C. bonariensis derived from seed collected from a population of the weed growing at Stellenbosh University’s experimental farm, Welgevallen, was similarly treated. At the time of spraying, the plants had 10-12 leaves and were still in the rosette stage. The weeds were subjected to a range of glyphosate dosage rates viz. 0.36, 0.72, 1.08, 1.44, 2.16, 2.88, and 3.60 kg a.e. glyphosate ha^-1. Four replicates per treatment were used.

From the results of this dose response trial (Table 1), it is clear that the Breede Valley population was not controlled at all at the registered dosage rate (1.08kg a.e. glyphosate ha^-1) and only 17% of the weeds were controlled at twice the registered dose of glyphosate. The control (Welgevallen) population was fully controlled at all dosage rates above 0.36 kg a.e glyphosate ha^-1. The trial was rated 56 days after treatment.

Table 1. Dose response of two Western Cape populations of Conyza bonariensis to glyphosate.

Subsequent screening has identified 3 more glyphosate-resistant populations of C. bonariensis (all from the Breede Valley). Work is in progress to map the extent of glyphosate resistant populations of C. bonariensis in the Western Cape and to identify cost effective control strategies.

A.L.P. Cairns
Department of Agronomy
University of Stellenbosch