For a specific example of experimental evolution of antibiotic resistance I would direct you to this paper which I previously referenced in the
'Natural Limitation to Evolutionary Processes' thread.
Evolution of drug resistance in experimental populations of Candida albicans.Cowen LE, Sanglard D, Calabrese D, Sirjusingh C, Anderson JB, Kohn LM.
J Bacteriol. 2000 Mar;182(6):1515-22.
Adaptation to inhibitory concentrations of the antifungal agent fluconazole was monitored in replicated experimental populations founded from a single, drug-sensitive cell of the yeast Candida albicans and reared over 330 generations. The concentration of fluconazole was maintained at twice the MIC in six populations; no fluconazole was added to another six populations. All six replicate populations grown with fluconazole adapted to the presence of drug as indicated by an increase in MIC; none of the six populations grown without fluconazole showed any change in MIC. In all populations evolved with drug, increased fluconazole resistance was accompanied by increased resistance to ketoconazole and itraconazole; these populations contained ergosterol in their cell membranes and were amphotericin sensitive. The increase in fluconazole MIC in the six populations evolved with drug followed different trajectories, and these populations achieved different levels of resistance, with distinct overexpression patterns of four genes involved in azole resistance: the ATP-binding cassette transporter genes, CDR1 and CDR2; the gene encoding the target enzyme of the azoles in the ergosterol biosynthetic pathway, ERG11; and the major facilitator gene, MDR1. Selective sweeps in these populations were accompanied by additional genomic changes with no known relationship to drug resistance: loss of heterozygosity in two of the five marker genes assayed and alterations in DNA fingerprints and electrophoretic karyotypes. These results show that chance, in the form of mutations that confer an adaptive advantage, is a determinant in the evolution of azole drug resistance in experimental populations of C. albicans.
I realise that this example is for a yeast rather than a bacteria but the principle is the same.
The authors use exactly the sort of clonal expansion Crashfrog has already detailed. They get around the problem you put forward of killing off all the non-resistant cells by using a dose equivalent to the yeast strains
Minimium Inhibitory Concentration (MIC). An MIC dose will prevent the growth of a culture but does not neccessarily kill that culture off. The minimum lethal dose to kill off the culture, or Minimum Bacteriocidal Concentration (MBC), may coincide with the MIC for some antibiotics (Bacteriocidals) but not for all (Bacteriostatics).
If a dosage less than the MBC is used then you can tell that resistance is present in certain populations as they will have developed elevated MICs.
TTFN,
WK