This paper deals with the calculation of Hazardous Concentrations of toxic substances from small sets of laboratory toxicity data, e.g., NOECs. A procedure due to Van Straalen and Denneman, as adapted from Kooijman (case n = 1), in which one seeks a concentration that protects 95% of the biological species is modified to account for the uncertainty in the estimates. New constants are obtained by simulation. These allow the calculation of the one-sided 95% left confidence limit of the Hazardous Concentration, from the mean and standard deviation of a sample of (laboratory) toxicity data. This 95% confidence limit is always lower than the 95% certainty value calculated with the Kooijman (n = 1)/Van Straalen method. The authors also derive constants to calculate a one-sided 50% confidence value, that overpredicts as often as it underpredicts. This value may be used as a median guess of the Hazardous Concentration. It will always be higher than the 95% certainty value of the Kooijman (n = 1)/Van Straalen method. However, by using the 50% value, one runs the risk of protecting substantially less than 95% of the biological species.