The statistical significance of close pairs of QSOs

Burbidge, Burbidge and O'Dell have proposed a method of statistical analysis of close pairs of quasistellar objects (QSOs) to test the null hypothesis that the redshifts of QSOs are of cosmological origin. Only two close pairs of QSOs were then known, therefore the analysis was inconclusive. Since then a number of close pairs of QSOs have been discovered and the same statistical test can be rediscussed with a greater measure of confidence. Indeed, as we show here, the probability of finding so many close pairs by chance, as required by the cosmological hypothesis, is as small as ≤10<SUP>−4</SUP>, depending on observational uncertainties