The effects of increasing amounts of insecticides added to the soil on the growth of cotton plants were studied. This was an attempt to verify whether or not a possible accumulation of insecticides in the soil might become detrimental to plant growth. The experiment was carried out in Mitscherlich pots with two types of soil: one sandy (arenito de Bauru) and the other clayey (terra-roxa). The amounts of insecticides added to the pots were calculated on an area basis to correspond to those applied to cotton plantings at the end of 1, 3, and 7 crops, and are called in this paper doses 1, 3, and 7, respectively. The amounts of insecticides as recommended for cotton in the state of São Paulo are as follows: BHC, 108 kg; Toxaphene, 14.5 k; Lindane, 0.4 kg and DDT, 3.6 kg. In the sandy soil, doses 1 and 3 of BHC caused 20% and 56% reduction in yield, respectively; dose 7 inhibited the production completely. Dose 7 of Toxaphene (72.24 kg/ha) when applied in emulsion form showed some toxicity. The same insecticide applied at the rate of 101.5 kg/ha in powdered form did not have any toxic effect nor did the emulsifier (15 ml of tritton 177 + xilol q.s.p. 100 ml) when applied alone. Lindane, emulsified Lindane, and DDT did not have any adverse effect on the cotton plant0 growth and yield. After the first crop the soil in each pot was passed through a sieve, returned to the pot and left undisturbed for one year. Cotton was then planted again. The development of the plants in this second planting was very irregular. This was due to aluminum toxicity, because the protective painting of the aluminum trays of the Mitscherlich pots came off in many spots and the metal wos attocked by the salts used as a fertilizer, forming aluminum salts that were detrimental to the plants. It wos possible, however, to see that toxic effects of the BHC added to the soil decreased fast enough after two years to allow even the plants in pots which received dose 7 of the insecticide to come to bearing. In the experiment with clayey soil only BHC was used since it was the only insecticide that showed a dangerous level of toxicity. The doses employed were 1, 3, and 5. Here too these numbers mean that the omounts of BHC added to the soil correspond to the omounts applied to cotton fields after 1, 3, and 5 years of cultivation, respectively. In the clayey soil, dose 1 of BHC did not decreose the yield nor caused any other toxicity symptom. Doses 3 and 5 decreased the yield in 14 % ond 27%, respectively, but the analysis of variance did nat show a significant difference between the two results. The growth and yield of the plants sown offer the soil had been left undisturbed far one year did not show any difference between the control ond treated plants, thus indicating that the toxic effects of the BHC had completely disappeared.