Evolution is traditionally spoken of as an inherited change over generations.

Does evolution happen one change at a time - or are there multiple changes occurring between two successive generations? Is there a known timeline for the frequency of such changes in any species?

Something like ...

20 Million years ago - Species X showed A, B, C.. Z characteristics

19 Million years ago - Species X following characteristics changed A'

17 Million years ago - Species X following characteristics changed B', C', Q', M'

16 Million years ago - Species X following characteristics changed A''

  • 1
    $\begingroup$ Your question presumes that species X is recognizable as species X for 4 billion years. And, the universe is only ~13.75 billion years old. $\endgroup$
    – kmm
    Nov 12, 2012 at 19:43
  • $\begingroup$ If not X, then it's most probable derivative. Homo Habilis, Homo Rudolfensis ... so on. $\endgroup$
    – Everyone
    Nov 12, 2012 at 19:51
  • 1
    $\begingroup$ I think you want some kind of character mapping. e.g., mesquiteproject.org/mesquite_folder/docs/mesquite/… $\endgroup$
    – kmm
    Nov 12, 2012 at 19:53
  • $\begingroup$ It's unlikely that multiple independent, fitness-relevant changes would occur at exactly the same time. I mean, if getting a beneficial random mutation is like rolling a 1 on a 1000-sided die, then getting two of them is like rolling 1s on two dice simultaneously. However, you could have multiple changes in phenotype if they both trace to the same genetic change (e.g., an increase in metabolic efficiency leads to increased size and also changes in foraging behavior). $\endgroup$
    – octern
    Nov 12, 2012 at 22:13

2 Answers 2


There can be multiple changes at the DNA level generation after generation, however most of these changes are insignificant (doesn't contribute for any specific character). But rarely, there can be a significant trait/change gained by an individual because of such changes in DNA level. Unfortunately such changes gets 'lost', unless it is fixed in the population (i.e., the gained individual should be successful in breeding and passing the trait/change to large number of progenies. Those progenies should further continue to pass on the trait till about every one (or most of them) in the population has those trait/change). Only such changes contribute for the species development (continuous accumulation of such changes, over time will result in the novel population that is very different from the original population).

Most of the timelines that we know are reconstructed from the DNA/protein sequences. Usually, DNA or protein sequences from the closely related species will be compared and looking at the differences they have, using suitable models, ancestral state of those sequences are determined. Once the sequence for the last common ancestor is known, mutational rate per unit time can be used to estimate the evolutionary time for those species.

A simple example would be the blue eye color in humans: it occurred about 10,000 years ago and it is still not fixed in the population.


Blount ZD, Barrick JE, Davidson CJ, Lenski RE. (2012) Genomic analysis of a key innovation in an experimental Escherichia coli population. Nature 489:513-8. doi: 10.1038/nature11514

Richard Lenski is conducting an experiment called the long-term evolution experiment (LTEE) in which twelve populations of E. coli are grown in a glucose-limited minimal medium. This experiment has now been running for over 40,000 generations (>20 years). If we take a human generation as being 25 years, this is equivalent to one million years of Homo evolution.

One of the diagnostic traits of the species E. coli is the inability to grow aerobically on citrate. The medium that is used in the LTEE does however contain citrate as a chelating agent (probably to solubilise ferric ions, although I haven't been able to confirm this). The Lenski group have recently reported on the appearance of a novel citrate-utilising mutant in one of the populations. The Cit+ variant appeared at around 31000 generations, and became dominant after 33000 generations. Because the LTEE involves storing samples of each population every 500 generations it has been possible to analyse in detail the emergence of this new phenotypic trait, and indeed to replay it.

The process that they describe has three phases:

1) Potentiation. This stage involved the appearance of mutations that were a necessary prerequisite for subsequent events. In this case it is thought that there were at least two potentiating mutations. These are as yet unidentified but their existence is inferred from replay experiments.

2) Actualization. This is the emergence of the new trait, and in this case is the event that took place at 31,000 generations. This involved a tandem duplication event at the citT locus, bringing citT under control of a new promoter, belonging to the rnk gene. The citT protein is a citrate transporter that is only expressed under anaerobic conditions in wild-type E. coli. The gene duplication event led to the expression of the transporter under standard aerobic conditions, opening up the possibility of taking up citrate and metabolising it. Replay experiments indicate that this event originally conferred a very weak Cit+ phenotype.

3) Refinement. This involves changes that strengthen the Cit+ phenotype. In this case these seem to have been further gene duplications of the rnk-citT module (with up to eight copies in one replay experiment). This eventually stabilised with the presence of four copies.

This may be the nearest that we can hope to get to defining a timeline of evolution of a new characteristic.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .