CORRELATION-FUNCTION CONSTRAINTS ON LARGE-SCALE STRUCTURE IN THE DISTRIBUTION OF GALAXIES

We consider whether photometric errors may have produced the large-scale power in the APM galaxy correlation function. Using the CCD galaxy sequences that are available on 39 of the 185 Schmidt fields of the APM survey, we apply an analysis of variance to make an independent estimate of the size of the zero-point errors associated with the APM galaxy photometry. The method finds an rms dispersion for these errors of 0.084 mag, a factor of 2 larger than previously claimed. Furthermore, we present evidence that this error may also be spatially correlated over the large APM survey area and show that the tests carried out by the APM/Oxford group could be insensitive to such large-scale effects. Thus, if it is confirmed that there are spatially correlated APM zero-point errors of this size, then removing the excess power caused by such errors from the measured APM correlation function would reduce the spatial break-scale for the galaxy correlation function to approximately 10 h-1 Mpc, in better agreement with the original estimate from the Lick catalogue. Thus, with the present CCD data, the APM correlation function cannot as yet rule out such scenarios as that of standard CDM.