A model of the Australian orange growing industry to explain changes in plantings, removals, the number and age composition of trees and orange production is developed and estimated. Most of the variation in plantings is explained by the expected profitability of growing oranges, the current stocks of bearing and nonbearing trees, and removals of trees last year. Estimates of the elasticities of response of plantings and production to price changes are low and there are long time lags. An illustrative application of the model projects future developments in the industry for alternative assumptions about the profitability of growing oranges.