Durbin'shtest for autocorrelated error terms is examined with respect to its small-sample power in addition to Durbin's other asymptotic test, thettest, and the Durbin-Watsonboundstest. Using sampling experiments for regression with lagged values of the dependent variable with first-order autocorrelated disturbances, we find that both thehandttests significantly outperform the Durbin-Watson bounds test and are acceptably powerful in small samples.