The remainder of the main body of the book concerns observations and the ways in which they can be used to constrain the theoretical development we have made so far. The level of technical detail will be considerably less, because the observational situation doubtless will change. Roughly speaking, we shall be working from the largest scales to the smallest, beginning here with the cosmic microwave background (cmb).

Large angles and the COBE satellite

When we study cmb anisotropies on large angular scales, we are as close as one can get to directly studying the initial perturbations. In Section 2.4, we found that the Hubble length at the time of last scattering subtends an angle of around 1 deg on the last-scattering surface itself. On scales significantly larger than this, we are directly studying the effects of perturbations on scales greater than the Hubble length at the time of decoupling, which therefore have retained their primordial form.

The crucial observations in this regime are those of the Cosmic Background Explorer (COBE) satellite, taken over four years, which rightly can be said to have revolutionized cosmology. These observations provided, for the first time, an estimate of the spectrum of inhomogeneities in the Universe on very large scales, of order thousands of megaparsecs.

The COBE satellite carried three separate experiments. The Far Infra-Red Absolute Spectrometer (FIRAS; Mather et al. 1990) provided what is by far the most accurate measurement of the frequency spectrum of the microwave background, confirming it as a blackbody within experimental limits.