Bias in the Cepheid PL relation

We show that the Cepheid PL relation is affected by a Malmquist type bias, that of so-called population incompleteness. Its calculated slope appears s hallower than the true one because of the cut-off in apparent magnitude due to the instrumental limiting magnitude. Furthermore, the use of the PL rel ation, even with the comet slope, leads to significant underestimation of d istances which is not negligible. A correction for this bias is difficult a s long as the completeness of the sample is not perfectly well established. (C) Academie des sciences/Elsevier, Paris.