I n field experiments on a sandy loam at Wellesbourne, England in 1972 and on a silt loam at Agassiz, British Columbia in 1973, combinations of herbicides and insecticides were applied at sowing to determine their effects on weeds and invertebrate populations and on the growth and yield of cauliflowers grown at high density There was good agreement between the results from the two locations. The two combinations of herbicides, 0.6 kg trifluralin/ha incorporated pre-drilling plus 2.2 kg propachlor/ha pre-emergence and 2.2 kg nitrofen/ha plus 2.2 kg propachlor/ha both applied pre-emergence, gave good weed control, their relative effectiveness depending on the species composition of the weed population. The insecticides isophenphos, carbofuran, chlorfenvinphos and fensulfothion were applied as bow-wave treatments. None of them, whether in combination with herbicides or not, adversely affected crop stand or yield. Yield was reduced when either weeds or root-fly maggots (Hylemya brassicae (BouchC)) were not controlled.Only in one experiment was there any evidence of any herbicideinsecticide interactions. Where trifluralin and carbofuran were applied together at Agassiz, the control of both weeds and maggots was less than that with the other combinations. None of the treatments affected the populations of predatory beetles, but the numbers of earthworms were greatly reduced by carbofuran and to a lesser extent by chlorfenvinphos. Except for carbofuran in one experiment, the treatments had no effects on the numbers of aphids, lepidopterous larvae or leaf miners present at harvest.