ANTIPODAL MICROWAVE

We examine inflectional Friedmann-Lemaître universes and find that current observations (gravitational lenses and QSO absorption lines) cannot rule out an antipode at za > 5. For za ≫ 1, the structure on the recombination shell is magnified, and the linear size observed at a given angular scale is much smaller than in an Einstein-de Sitter universe. In the extreme case of an antipode near or inside the shell, | za - zr | ≲ Δzr, the (linear) magnification is Ma ∼ 800Ωm01/2; thus δT/T is reduced by photon diffusion and projection effects on scales ≲90° (rather than ≲8′ in an Einstein-de Sitter universe). If 250 < za < ∞, we observe the microwave background radiation (MBR) emitted within a causally connected region; thus anisotropies from the Sachs-Wolfe effect are reduced. A high-redshift antipode produces anomalous ratios between the coefficients of the Taylor expansion of the MBR autocorrelation function. It can thus be ruled out or discovered when two or more such coefficients are measured. A simple test is the decrease of multibeam anisotropy measurements with the number of beams, on angular scales within some range above 8′ × Ωm0-1/2 (in contrast to predictions of various theories in a flat universe). In addition, gravitational lensing by galaxies and clusters reduces MBR brightness contrasts on scales ≲1′-1°. To our knowledge, this effect has not been correctly appreciated before. We point out that it is analogous to the reflection of the Sun in an undulating water surface. However, on these scales, the antipodal magnification and the damping already reduce δT/T below the present-day sensitivity. Thus lensing by galaxies and clusters is irrelevant for current bounds on the MBR anisotropy (as it is in conventional cosmologies). An antipode at za ∼ 5-10 may produce a departure of the MBR from the blackbody spectrum, if it lies inside an infrared source. An obvious prediction, then, is large dipole and higher multipole moments of the departure, with axes uncorrelated with the known MBR dipole.