How could they get that wrong? It should be very easy to tell if a species has been transplanted because its genetic code will be very different from the native life on the planet.
Not in the Trek universe, though. Everybody has the same genetic roots there, after all. For all we know, we here on Earth are a native species but dogs were inserted from Sirius by a sinister alien race intent on turning our brains to porridge with excessive cuddliness.
And of course, it could be that Vulcan has no native life. There could be competing belief systems: one says that the Vulcan sapient humanoids (and perhaps some of their domesticated animals or pets) survived some sort of an ancient extinction event thanks to being technologically advanced while no other, lesser forms of native life did, while another says that the presence of Vulcans on a planet that cannot naturally harbor life is proof positive of interstellar migration. Neither camp has any explicit historical facts on its side, though, because life on the brutal planet has reduced historical records to dust. So the argument continues.
In any case, if Vulcans learned about genetics before they did about starflight, then naturally their science would have been formulated to accommodate the facts of their genetic inheritance. Their first or second hypothesis probably would not be that they themselves were alien to their own world. So there'd be likely to exist several highly logical scientific doctrines that explain the Vulcan humanoids as a native lifeform quite regardless of the observable genetic facts of the matter.
On the issue of how long a "generation" is for a Romulan "generation ship", the current definition only pertains to the average age difference between a parent and a child. How long the parent lives after the child is born is irrelevant to the definition. The age difference between Spock and Sarek would be the important thing here, then - as long as we remember that Spock was only the second child of Sarek (as far as we know).
We know that this time difference is 65 years. We don't know Sybok's age, though, and with Vulcans it's very difficult to tell by looks alone. Perhaps the Vulcan generation is 50-60 years long. But perhaps it's much shorter than that? Tuvok was born in 2264 and had children no sooner than his marriage at 2304 as far as we know; by 2374, his firstborn had his first child. Again, the average might hover a bit over the half-century mark, then. T'Pol, born in 2088, hadn't bred by 2154, but may have been about to.
Any other interesting datapoints out there?
Timo Saloniemi