Extrapolating the present motion of expansion of the universe backward in time, we conclude that the early universe must have been very dense. And extrapolating the (adiabatic) expansion of the cosmic background radiation backward in time, we conclude that the early universe must have been very hot. Thus, at an early time, the universe must have been very different from what it is now. There were no stars and no galaxies, but only a uniform hot plasma, consisting of free electrons and free nuclei. The chemical composition of the early universe must also have been different. The heavy elements (that is, elements other than hydrogen, deuterium, helium, and lithium) in our immediate environment were formed by nuclear reactions in the cores of stars, so these elements did not exist in the early universe. At very early times, the violent thermal collisions would have prevented the existence of any kind of nuclei, and the matter in the universe must have been in the form of free electrons, protons, and neutrons. At the earliest times, even the protons and neutrons would have been disrupted, and the universe must have contained a mix of quarks, gluons, and other elementary particles.

The observed expansion of the universe and the observed cosmic background radiation provide the empirical basis for a Friedmann-Lemaître model of the universe with a Big Bang, sometimes called the Standard Model. Further evidence supporting this model is provided by calculations of the synthesis of helium in the universe. Although stars make helium by the thermonuclear burning of hydrogen, most of the helium in the universe must be primordial, since it is found even in stars that have not yet burned long enough to accumulate a significant amount of helium. This primordial helium was formed by nuclear reactions in the early universe at about 100 s, and the abundance of this helium (relative to hydrogen) can be calculated by examining the thermal equilibrium attained by protons and neutrons in reactions in the early, hot universe. The numbers obtained by such calculations of the helium abundance are in excellent agreement with the observational data. The abundances of other light elements formed in the early universe can be calculated similarly.