44 An introduction to the physics of cosmology
Figure 2.6.Confidence contours on thev–mplane, according to Riesset al(1998).
Open models of all but the lowest densities are apparently ruled out, and non-zerois
strongly preferred. If we restrict ourselves tok=0, thenm 0 .3 is required. The
constraints perpendicular to thek=0 line are not very tight, but CMB data can help here
in limiting the allowed degree of curvature.
thatcould have been much larger than its actual value without making observers
impossible. Efstathiou (1995) attempted to construct a probability distribution for
by taking this to be proportional to the number density of galaxies that result in
a given model. However, there is no general agreement on how to set a probability
measure for this problem.
It would be more satisfactory if we had some physical mechanism that
guaranteed the coincidence, and one possibility has been suggested. We already
have one coincidence, in that we live relatively close in time to the era of matter–
radiation equality (z∼ 103 , as opposed toz∼ 1080 for the GUT era). What
is required is a cosmological ‘constant’ that switches on around the equality era.
Zlatevet al(1999) have suggested how this might happen. The idea is to use
the vacuum properties of a homogeneous scalar field as the physical origin of
the negative-pressure term detected via SNe. This idea of a ‘rolling’was first
explored by Ratra and Peebles (1988), and there has recently been a tendency