A new edge—corrected kernel—based estimator is proposed for the density function of the nearest neighbour distance of a stationary and isotropic point process. The performances of the new estimator and the existing estimators are compared in a simulation study. The results of the simulation study suggest that the new estimator is preferable to the existing alternatives. The use of the estimators for testing whether a spatial point pattern is consistent with the hypothesis of a Poisson process is demonstrated by a geographical example