Here’s the latest from PhysOrg:
The process of “spatial sorting” relies on genes for speed accumulating at the increasingly fast-moving frontline. Unlike natural selection – a process first described by Charles Darwin, stating that traits which help an organism survive and reproduce will build up over time – spatial sorting does not require an animal’s survival or reproduction to be increased by it being quicker. The new process can only work within the limits set by natural selection, but may be an important cause of evolutionary change in species that are expanding their ranges into new territory.
There’s a name for this kind of process, isn’t there? I think you would term this a “sieve-like” process. But, IIRC, Richard Dawkins in the Blind Watchmaker says that a “sieve-like” can’t explain the complexity we see. So, here we have “changing gene frequencies” that have no connection to what NS is purported to be able to do. We simply have changing gene frequencies without any “increase in fitness”. It’s simply a matter of non-random mating patterns, patterns that are imposed simply by proximity of mates. I’m more than willing to concede that these kinds of “changing gene frequencies” happen all the time; but I’m not willing to say that this constitutes “evolution”. I think there is some “microevolution” that takes place, but that the most of what happens is right along this sieving process that is proposed by these experimenters. Share your thoughts.
BTW, the “limits set by natural selection” are, in my view, mostly conservative: i.e., NS works principally to root out genetic errors; not bring about significant change (“The Edge of Evolution”‘s view, for example).