Consider a species. Without some particular mutation, its members have quality A. Suppose that quality B is better for the survival of any individual member of the species. If one member possesses quality B, natural selection could take over.

What if B is preferable to A, but the intermediary stages are not preferable to either A or B? Can a species go from A to B?

Say an organism has 4 legs but having 6 would be advantageous. Could it get to 6 without some disadvantaged intermediate forms? Or is there a way for a single genetic mutation to result in a one step change? Also, if no organism has ever made such a transition then I suppose it can go without explanation. But then maybe something went from 6 legs to 4, or something.

I hope this isn't too rambling. Kindly,

  • $\begingroup$ Your general line of inquiry is good, but with the current state of the question, the answer is "maybe". There's not enough details to evaluate or scrutinize the scenario and give a really meaningful answer. There's so many variables that you're leaving open to us that it's unfortunately more of an exercise in imagination than a request for mechanical clarification. Perhaps for the legs scenario I could point you to Hox genes, which, if duplicated, could potentially duplicate a whole leg-pair segment. Try looking up pictures of "fly mutant hox genes". $\endgroup$ – Harris Aug 22 '16 at 16:59
  • $\begingroup$ Thanks for the help. I realize my question was too vague but I'll look into Hox genes, because it seems, after a quick glance, like it will give me some of the insight I was looking for. $\endgroup$ – dactyrafficle Aug 22 '16 at 18:12
  • $\begingroup$ Unfortunately, at least to me, the issue isn't four vs. six legs, to borrow your example. The issue is what other number of variables contributed to the emergence of the 4-legged archetype besides the fitness of # of legs? $\endgroup$ – CKM Aug 22 '16 at 21:16
  • $\begingroup$ Maybe also take a look at my answer to this question. The concepts of gradual evolution and punctuated equilibrium might be of interest for gaining a deeper understanding of macroevolution. $\endgroup$ – AlexDeLarge Aug 23 '16 at 20:57
  • $\begingroup$ I think it'd be great to make a community post that describe concept of fitness landscape and fitness valley and then to talk about shifting balance theory and the role of phenotypic plasticity and of large effect mutations to explain the transition over a potential fitness valley. There are so many layman that have question on this subject. $\endgroup$ – Remi.b Aug 24 '16 at 21:57

The question is a very deep one, and the answer is 'it depends'. Organisms can in general only evolve by taking rather short steps through genome space, where each step is better than the last. This could mean one or two nearby basepairs change, or one or two segments of DNA get duplicated, inverted, deleted, etc. It's the DNA that evolution acts on, not the organism's 'qualities', or as we call them, it's phenotype.

In your example, there might be a single piece of DNA which when duplicated, causes the organism's number of legs to go from 4 to 6. In this case the mutation can happen and be selected for and extra legs can evolve. If however the evolution of new legs requires a lot of mutations, non of which are individually useful, then the trait won't arise before the intermediate mutations are selected away.

What may happen instead is something more complicated. E.g. the many mutations for the new legs could initially arise because they are doing something different, and then gradually get selected for other purposes.

In general though the problem of how organisms' move from one peak in the 'fitness landscape' to another is an ongoing topic of research.

  • $\begingroup$ Part of the challenge of the legs analogy was that it seemed so complex that it would've required a symphony of mutations to produce in one generation to avoid disadvantaged intermediate forms. I thought it very unlikely, more still if the advantage was only slight, or if it required such a mutation in not just one but a subset of the original population. I guess my initial question is impossible to answer and that I need specific states A and B, for which evidence of the transition from A to B exists, otherwise it is disingenuous to demand an explanation from evolution. $\endgroup$ – dactyrafficle Aug 23 '16 at 17:48

I think the example is actually confusing the issue. As others have pointed out, there might very well be a single mutation separating the four-legged type from the six-legged type, so there would be no intermediate type.

But the basic question stands:

What if B is preferable to A, but the intermediary stages are not preferable to either A or B? Can a species go from A to B?

I did my thesis on this! The answer is a qualified "yes" under certain conditions:

  • There are no more than a few necessary intermediate steps.
  • The intermediate types are only slightly less fit than the original type.
  • The population is large.
  • There is not too much recombination among the underlying genes. (A little recombination actually makes it easier.)

For details, see here and here. (Sorry, they're pretty technical. I've been "almost done" with a review non-specialists for years now...)

  • $\begingroup$ This is interesting. Is your work a model based on the assumptions you've stated and using simulated data? And given those assumptions, a species can go from its current state A to some preferable state B? $\endgroup$ – dactyrafficle Aug 25 '16 at 17:50

Not the answer you're looking for? Browse other questions tagged or ask your own question.