Effects of insecticides on non-target organisms in African agroecosystems: a case for establishing regional testing programmes

Abstract

Field trials were simultaneously conducted in Algeria, Nigeria, Tanzania, Uganda and Zambia between 1992 and 1995 in order to determine the effects of organochlorine pesticides lindane (all countries) and endosulfan (two countries) on non-target arthropods in African maize agroecosystems. In addition, a laboratory screening test was conducted to determine the initial toxicity’s of several organochlorines and a synthetic pyrethroid to three species of non-target arthropods. Lindane significantly reduced the numbers of Collembola in over 80% of the field trials for an average of six weeks. Similarly, spiders were reduced in 53% of the trials for an average of 2.8 week, and ants were reduced in 64% of the trials for an average of 2.5 week. The lindane treatment significantly reduced organic matter breakdown in over 45% of the trials, whereas endosulfan had no effect. The latter had little or no effect on non-target arthropods. The lindane treatment significantly reduced plant damage in all countries. However, corresponding significant increases in yield were observed in less than 50%. There was an apparent consistent effect of lindane on different non-target groups as indicated by the positive correlation’s between these variables. Endosulfan was found to be harmless to all the non-target arthropod species included in the laboratory screening tests. The order of toxicity was lindane = deltamethrin ⪢ chlorpyrifos ⪢ endosulfan. It is concluded that African countries need to develop procedures for testing pesticides in Africa in order to arrive at the correct conclusions concerning adverse side effects from pesticide use.