From trials with cotton (variety Stoneville 7A) grown on the Ord River Irrigation Area of Western Australia, the results of delaying insecticide protection of crops against Heliothis attack by four and eight weeks were re-examined. With each delay the crop set later when temperatures were cooler, the plants older and larger, and the accumulated insect damage greater. Yet there was no significant difference in the final number of bolls produced in each crop. The growth of cotton and the compensatory mechanisms of the plant indicated by the data were simulated by a simple computer model. A modification of the model to simulate the response of a cotton plant to attack by Heliothis larvae was also developed. Output from the model suggests that, under the conditions of the experiment, there is a theoretical maximum number of bolls produced by this cotton variety. To produce this maximum crop requires complete protection from Heliothis larvae, but the period of protection can be minimised.