Patwary and Haley (1967) proposed a mathematical model for estimating the parameters of a tolerance distribution when the dose is subject to error of administration. Using an artificial data, they discussed a particular case when observation is quantal response and the errors in dose follow the Poisson law, and presented a computational routine. Fitting trial of the model to the quantal toxicity test data of Bacillus thuringiensis for larvae of the gypsy moth, Lymantria disper, was made, since the active ingredients of BT formulation are discrete values and error in dose may follow a Poisson distribution, with mean equal to the nominal dose. Iterative calculations gave the good estimates of parameters and the median lethal concentration. Similar good results were also obtained in the data tested toxicities of BT formulation on the pine caterpillar, Dendrolimus spectabilis, and the fall webworm, Hyphantria cunea.