The density evolution of IRAS galaxies

We have applied a maximum likelihood test for density evolution to a full-sky redshift survey of 5297 galaxies selected from the data base of the Infrared Astronomical Satellite (IRAS) and flux-limited at f_v(60 μm) = 1.2 Jy. We find no evidence for evolution; if the comoving density of galaxies is characterized as evolving proportional to (1 + z)^α, we find α = 2 ± 3, where the quoted error includes both random and systematic components. The random errors in α, of order 2, arise primarily from counting statistics, and are comparable to those found by previous authors. We discuss in detail important systematic errors which decease the certainty with which we can measure evolution: limited knowledge of the cosmological model, the unknown intrinsic spectral energy distribution of IRAS galaxies from 16 to 140 μm, the effect of density fluctuations, a Malmquist-like bias arising from flux errors in the IRAS Point Source Catalog, and possible incompleteness of the sample at high redshifts and low Galactic latitudes. We show that the Malmquist bias could result in a significant overestimation of the evolution rate, especially if the catalog has a flux limit near the completion limit of the Point Source Catalog.