Your whole genome comparison comparison should make it fairly evident what sort of genetic changes have occurred. If all you detect are point mutations then it is very unlikely that any sort of genomic re-arrangement was involved.
I get the impression that nearly ANYTIME you add an antibiotic to any bacterial population, you soon wind up with a resistant population (due to extermination of all non-resistant types). This just seems a little quaint to me if RANDOM MUTATIONS are the key factor.
Your impression is incorrect. If it was then molecular biologists would have a much harder time of it. The usual method for producing large amounts of DNA is using an antibiotic resistant plasmid carrier in a bacterial line. If any bacterial population would evolve resistance when exposed to an antibiotic then we would not be able to screen for plasmid containing bacteria. It may be true when you have multiple exposures for very large populations, but it certainly isn't true for all populations.
A comparison of antibiotic resistance based on initial concentration of the bacterial culture should answer the question of how the resistance arises. If the mechanism is along the pre-programmed variability spectrum you suggest then a much smaller initial concentration should have at least one resistant individual than if the resistance arises from random variation throughout the genome.
I can't give you the right sort of figures off hand, but I'm sure that both systems could be modelled sufficiently to get some reasonable ball park figures, or at least the straight forward traditional random mutation could be, your model seems a little sketchy at the moment and therefore it may be rather hard to predict what the values would be, but the expected frequency of resistant bacteria should certainly be higher than in a traditional model.
(Edited to add) Actually in retrospect there are some already recognised chromosome rearranging or mutation rate altering strategies which have been observed in bacteria under stress, the problem is that these are very rarely directional in anyway. If all you are doing is creating a new source of random variation I don't see how you are changing the situation in regards to the origin of the trait. Survival strategies relying in increased rates of random mutation still have random mutation as the key factor.
TTFN,
WK
This message has been edited by Wounded King, 01-31-2005 06:45 AM