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

In describing the dynamics of fixation of mutations, it is necessary to consider not only the probability with which a mutation will become fixed in a population of a certain size, but also the time required on an average for fixation of a mutation. The probability that a newly formed mutation will be fixed is equal to 1/2N. Similarly, the average time required for fixation of one mutation is proportional to the size of the population. However, this is a case of direct proportionality. M. Kimura derived that the average time for fixation of a mutation by genetic drift is equal to 4N_e generations, where N_eis the effective size of the population (the effective size of the population is a term that will be explained in Section V.3.2.1) For a population with an effective size of 30,  fixation of a neutral mutation will thus require an average of 120 generation periods.

The graph describing the shape of the time distribution required for fixation of a mutation by genetic drift is highly asymmetric. The asymmetry of the graph reflects the fact that it is highly improbable that a mutation will become fixed sooner than in 0.8N_e generation periods and a great many mutations require substantially more time than the average 4N_egeneration periods.