When we look at the microwave background radiation it looks the same in every direction, even from opposite sides of the sky, to a precision of 0.1%. Since they are so nicely correlated one would naturally assume that at some time all points in the observable universe were in close contact with each other, for otherwise it would be an unbelievable coincidence for all of them to look so much the same (at least through a microwave detector).
Now, a perfectly reasonable question is whether the Big Bang model has this property: will the Big Bang model predict not only the existence of the microwave background radiation, but also its exquisite uniformity? The answer is ``yes'' but only with additional assumptions.
This seems confusing: is the Big Bang theory to be modified and tuned every time a new piece of data comes along which does not agree with its predictions? Isn't this cheating? Doesn't this sound like Ptolemy adding epicycles every time things weren't quite accurate?
Fortunately this is not the case. The Big Bang theory determines the evolution of the universe provided the matter and energy content is known, and their behavior at very extreme conditions is well understood. The fact is, however, that we are not certain of all the matter and energy in the universe, nor do we know, for example, how they behave at temperatures above 1015 oK. Hence these ``modifications'' of the Big Bang theory correspond to different hypothesis of the behavior of matter at very high temperatures and densities, not of the general description provided by the General Theory of Relativity.
The simplest version of the Big Bang model which predicts a very uniform microwave backgound goes by the way of Inflation. The idea is the following: the simplest way of getting uniform background radiation is if all the observable universe was in very close contact at an early time. Granted that, inflation provides a mechanism for increasing the size of this initially tiny region to the very large universe we see. Though mathematically involved what is assumed is that at a very time (about 10-35s after the Big Bang) a new force comes into play which forces an exponential increase in the size of the universe (hence the name `inflation'). After a fraction of a second this force is balanced by other interactions and the universe resumes a more dignified, if ponderous, expansion (see Fig. 8.23).
One tantalizing conclusion derived from the inflationary hypothesis is that there are regions in the universe which we have not yet seen and which might look very different. Since no light has reached us from those regions we are currently unaware of their existence, only our inheritors will see the light coming from these distant reaches of the universe.
It is a challenge for current researchers to produce models that generate the intergalactic voids, yet with the same amount of dark matter required to understand the rotation of stars (Sect. 8.5.1) and using the inflation hypothesis such models actually exist. The corresponding computer simulations produce results such as the one shown in Fig. 8.24 which should be compared to the observations (Fig. 8.22).