Having actually studied evolutionary systematics in college (even though that was a long time ago in a galaxy far far away)... I can actually babble about the subject some.
Population diversity:
A diverse population is required to survive environmental changes. This is what Darwin observed in the Finch population of the Gallapagos. The finches had 3 different beak types, thin long beak, medium beak, and large blunt beak. A drought hit. Those with the thin long beaks could get at insects inside plants (think, woodpecker). The large blunt beaks could crack open shell fish. The medium beaks died out from lack of food. That study happened while Darwin was on the island observing it -- not over a few hundred millions of years.
That is why there is concern over a genetic "bottleneck" -- where the diversity of a species is not sufficient enough to survive an environmental catastrophe. This likely helped with the creation of the "endangered species list".
Within the past few years, there was a discovery that a naturally-enough occuring protein acted as a hormone and caused the same mutation across fish, mammal, and birds. From the context of small changes over millions of years evolution, that means the diversity/adaptability/new-trait was given on a "whim" -- "overnight". That particular enzyme is used to increase jaw bone growth size -- and, as I stated, does so across a wide species line (fish, birds, & mammals). They still do not know what else is out there to be discovered that could cause other viable "mutation" effects.
What really makes one think, though....
Chromosomal mutations are typically fatal. About the only one people are familiar with in humans is Down's syndrome -- and I think it was the only non-fatal one. The genomic pairing of the 23 (female) + 23 (male) for 46 chromosomes does not offer genetic (chromosomal) variability. Even if one did get an extra chromosome, the resultant birth is typically fatal or does not find a mate within the species to procreate (if procreation is actually possible -- sterility is almost a guarantee). As far as I know, and I have been waiting to observe any results to counter this, evolution does not allow for chromosomal counts to change.
To put it a different way. Chromosomal changes are not driven by normal evolutionary means.
That means there had to be an "Ancestor" with the 46 chromosomes we currently have. Likewise, for any organism, there had to be an "ancestor" that has just as many chromosomes as the finished, current/modern, organism. Now the species and organisms may have been different throughout history -- but the number of chromosomes does not offer a viable scheme for "change".... and chromosomal changes had to have happened. "How" has been a big question for a long time.
Even if a chromosomal change happened, that new organism (there is only one) would have to be compatible enough to mate and reproduce with the previous (parent-species) organism. If the chromosomal change is not dominant, it would be bred out. If the enough of a population diversity was not created, it would become extinct (flu/drought/etc). There also would have to be some mechanism to prevent "common" birth defects of mating within the new species.
Extensively studying evolution, evolutionary systematics, and how chromosomal changes would not be viable under those "rules" -- leads one to think there was something more than evolution involved with the species of the world. I may sound kind of off; but consider, Darwin -was- an agnostic that believed in a "personal God".
As was obvious to me in my studies, there is something "missing" in the theories to describe how chromosomal shifts can occur and produce a viable offspring-species.
Scott offers a viable theory, external dominant (or recessive) chromosomes introduced into the mix by breeding with other similar species. External sources probably have a higher chance of producing a viable chromosomal shift in organisms than random chromosomal mutations.
Scott's theory with the chromosomes is even demonstrated in horse + donkey = mule. Donkey's have 62 chromosomes, horses have 64 chromosomes. Mules (the offspring) are almost always sterile and have 63 chromosomes (my point, the chromosomal shifts do not produce enough diversity & enough viability).
Regardless, there is a BIG gap of knowledge in how chromosomal shifts happen in species and in evolution. Or at least, there was -- I usually look for articles on it and am waiting to find one.
Oh, and Scott's theory for Ancestor is one of my favorites.
-Pookie Chang says to buy stock in Depends, believe him!