Selection effects or high opacity? Understanding the surface brightness distribution of inclined disc galaxies

In a recent much-quoted paper, Valentijn (1990) has used the ESO-Uppsala galaxy catalog to carry out the classical surface brightness against inclination test to try to measure the opacity of spiral galaxies. From his results he infers, contrary to previous work, that galaxies are optically thick out to large radii. In this paper we show that the methods and selection criteria used by Valentijn will constrain his galaxies to have a constant surface brightness independent of their inclination, whatever their actual optical depth. We demonstrate, first, the practical efficacy of such selection biases by showing that the observed ranges of surface brightness for ESO galaxies at high and low Galactic latitude are the same, despite the obvious foreground extinction of each individual galaxy in the latter group. We then show, from simulations, that galaxies of a particular preferred observed surface brightness are always more likely to be included in the sample regardless of their inclination