However, if mutations, basically due to randomness, were the only cause of diversification, we would not see organized and coherent structures in the terrestrial biosphere. Instead, the organization of life is shaped by the process of natural selection, understood and analysed for the first time by Charles Darwin (1859). In fact, in the first place, because of the genome stability, mutations are anyway very rare (a typical rate of mutation of a pair of DNA bases per generation is in the order of ); also, the vast majority of mutations are deleterious. Among those not detrimental to the survival of organisms only a few manage to pass through the sieve of natural selection, which favours those mutants that have a demographic advantage. Selection is very strict, but the sieve operates continuously on myriads of organisms. It is therefore very effective in the long run and produces organisms adapted to their environment.
In very many cases, as pointed out in more recent times by Kimura and Crow (1964), may nevertheless appear and remain in a population even so-called neutral mutations - i.e. neither advantageous nor deleterious. This makes it possible for the same species and more in particular the same population to be generally characterized by high genetic diversity. Keeping rare genes within a population can not confer an immediate advantage but may be nevertheless useful in the future for the conservation of the species when new environmental conditions might change the selective pressures. The loss of genetic variability can limit the ability of a population to respond to external changes in the long term such as those due to pollution, climate change, new emerging diseases. Unfortunately, in small populations, which is typically the case of populations threatened by extinction, purely random processes can lead to the disappearance of genes thus further increasing the risk of extinction in the long run. In the following sections we will analyse these phenomena in greater detail.