Abstract
We show that a good fitting to the first two years of {\it COBE}-DMR observations of the two-point angular correlation function of CBR temperature is given by models with a non-zero infrared cutoff $k_{min}$ in the spectrum of the primordial density perturbations. If this cutoff comes from the finiteness of the universe, say, a topological T3 model, we find $k_{min} \sim (0.3 - 1.1)\pi$H$_0/c$ with confidence level 95\%. Such a non-zero $k_{min}$ universe would also give a better match to the observations both of the RMS quadrupole anisotropy of CBR and of galaxy clustering.

This publication has 0 references indexed in Scilit: