1: \begin{abstract}

2: We present improved numerical approximations to the exact Poissonian

3: confidence limits for small numbers $n$ of observed events following

4: the approach of Gehrels (1986). Analytic descriptions of all

5: parameters used in the approximations are provided to allow their

6: straightforward inclusion in computer algorithms for processing of

7: large data sets. Our estimates of the upper (lower) Poisson confidence

8: limits are accurate to better than 1\% for $n\le 100$ and values of

9: $S$, the derived significance in units of Gaussian standard

10: deviations, of up to 7 (5). In view of the slow convergence of the

11: commonly used Gaussian approximations toward the correct Poissonian

12: values, in particular for higher values of $S$, we argue that, for

13: $n\le 40$, Poissonian statistics should be used in most applications,

14: unless errors of the order of, or exceeding, 10\% are acceptable.

15: \end{abstract}

16: