Frozen Evolution. Or, that’s not the way it is, Mr. Darwin. A Farewell to Selfish Gene.

Sexual reproduction or, to be more exact, the genetic recombination that accompanies it, is currently considered to be an important mechanism through which the population gets rid of most slightly negative mutations in the DNA. In asexually reproducing organisms, mutations accumulate in the genome that, in most cases, worsen the properties of the encoded proteins. Mutations that reduce the viability of an organism in a drastic manner can be removed together with their carriers by natural selection. However, a large proportion of the mutations are almost neutral in their effects, so that there is very limited potential for natural selection here. Mathematical models indicate that gradual worsening of the average viability of individuals in the population must necessarily occur through the effect of slightly negative mutations occurring randomly in all the genes of the individual members of the population. This irreversible process is called Muller's ratchet (Muller 1964). A ratchet is a toothed wheel with a pawl that ensures that the wheel can rotate in only one direction (for example, in the equipment for stretching a volleyball net). The existence and importance of this process in the individual types of organisms is still a subject of discussion at the present time.

            It should simultaneously be realized that the accumulation of slightly negative mutations can also have an accelerating trend. It can happen that mutations can also occur in the genes for enzymes active in DNA replication and reparation. Thus, the potential for positive feedback also arises: deterioration of the quality of the DNA-polymerase – lower precision of replication – more mutations (amongst others also in the gene for DNA-polymerase) – deterioration of the quality of the DNA polymerase, etc. According to some theories, the ageing of multicellular organisms is caused by just this process (Orgel 1973; Kirkwood 1977).  In asexual reproduction, it cannot happen that the daughter individual would carry fewer mutations than the parent individual. However, in sexual reproduction, genetic recombination leads to individuals that can have more or fewer mutations than their parents. Individuals with a large number of mutations can be eliminated by natural selection, while individuals with a smaller number of mutations can be placed at an advantage. This process can be especially effective under the conditions of positive epistasis, i.e. when the detrimental effects of the individual negative mechanisms are not simply added up, but multiplied together, i.e. when, for example, the concurrent occurrence of four negative mutations in the genome leads to more than twice the reduction in the fitness of the particular individual compared to the occurrence of two mutations (Fig. XIII.7). In this case, even the elimination of a relatively small number of unfit individuals from the population is accompanied by the elimination of a large number of negative mutations from the particular gene pool (Kondrashov 1988).

            The existence of sexual selection leads to a further increase in the effectiveness of removal of negative mutations from the population (Agrawal 2001a; Siller 2001). Because males have a greater variability in fitness compared to females (see Chap. XIV), females can preferentially choose, as their sexual partners, very fit males that apparently have only a few negative mutations in their genome. The results of mathematical modelling indicate that the existence of males and the associated existence of sexual selection substantially reduce the average mutation load on members of the population.

            It is thus apparent that sexual reproduction at the level of populations or species, but not at the level of the individual, can be a form of adaptation towards stopping or at least retarding Muller’s ratchet, i.e. to stopping or even reversing the otherwise irreversible process of accumulation of slightly negative mutations (Kondrashov 1988; Howard 1994).