CAN EXTRA POWER EXPLAIN PERIODICITY ON LARGE SCALES

In a recent paper, Broadhurst et al. presented galaxy counts as a func tion of redshift in the direction of the north and south Galactic pole s. Their results indicated a surprising amount of power on large scale s and suggested a periodicity in the large-scale distribution of galax ies. Here we compute the one-dimensional power spectrum on large-scale s projected from the known small-scale clustering of galaxies, includi ng the effects of the coherent peculiar velocity due to gravity. We fi nd that the external estimate of the mean power is rather uncertain. O ur calculation shows the likelihood that the projected power as estima ted internally from the data set can account for the high peak on larg e scales is less than 1%. This likelihood goes up by a factor of 15, i f we restrict our internal power estimate to scales greater than 128 h -1 Mpc, indicating some extra power on very large scales. We also stud y the consequences of adding some intrinsic power on large scales. The periodicity in the pair counts may be consistent with extra power on those scales, but the sharply defined spike in the power spectrum rema ins unexplained, suggesting non-Gaussian statistics or some undiscover ed statistical problem with the sample.