Field trials were conducted to dertermine the effects of period of weed interference and evaluate various herbicides for weed control in sweet pepper (Capsicum annuum L.) and chilli piepper (C. frutescens L.), respectively at Samaru during the wet season in 1982 and at Samaru and Kadawa during 1982/83 dry season. In all the trials, weed infestation for 9 weeks after transplanting and later significantly depressed various crop growth parameters such as crop vigour score, number of leaves and branches per plant and yield and yield attributes, including fruit number and fresh weight, compared with the crop kept weed-free throughout its growth. Although weed association with the crop during the first 6 WAT depressed crop vigour score, this effect was not reflected in sweet pepper fruit yield when the weeds were subsequently removed. In order to obtain statistically comparable fruit yield to that of the weed-free control in the wet season at Samaru, it was necessary to keep the crop weed-free for 6 WAT and later. In the dry season, however, both at Samaru and Kadawa, only crops kept weed-free beyond 6 and 9 WAT, respectively resulted in comparable fruit yield to the weed-free control. The critical period of weed competition was observed to be between 6 and 9 WAT. V Herbicide treatments which consistently combined effective weed control with high chilli pepper fruit yields comparable with the hoe-weeded control, during the wet and dry seasons include mixtures of alachlor with linuron and chlorbromuron at 1.0 + 0.5 and 1.0 + 1.0 kg a.i./ha, respectively, mixtures of linuron with diphenamid and oxadiazon at 0.5 + 6.0 and 0.5 + 2.0, respectively and pendimethalin plus metobromuron at 2.0 + 1.0 kg a.i./ha. Although high pepper fruit yield was obtained with both rates of metolachlor plus metobromuron during the wet season, these treatments did not consistently result in effective weed control. In this study unchecked weed growth throughout the crop life cycle caused between 60 and 92% loss in potential pepper fruit yield.