The purpose of this study was to test the mutant selection window (MSW) hypothesis in vitro and in vivo with Staphylococcus aureus exposed to fosfomycin. With the in vitro time-kill studies, S. aureus ATCC 29213 [with a minimal concentration that inhibits colony formation by 99% (MIC99) of 2.2 μg/mL and a mutant prevention concentration (MPC) of 57.6 μg/mL] lost fosfomycin susceptibility at antibiotic concentrations (2×, 4×, and 8× MIC) that are between the lower and upper boundaries of the MSW. In the tissue-cage model, S. aureus was exposed to fosfomycin pharmacokinetics at concentrations below the MIC99, between the MIC99 and the MPC, and above the MPC, respectively. Changes in susceptibility and counts of total and resistant viable bacteria were monitored in tissue-cage fluid obtained daily. However, the selection of resistant mutants was not observed during antibacterial treatment and 48 h after the termination of fosfomycin treatment, regardless of the fosfomycin dosage. Besides, we found no differences between the in vitro-isolated mutant and its sensitive parental strain, which indicates the absence of fitness cost of fosfomycin resistance in S. aureus ATCC 29213. These findings demonstrate that agar plate determinations do not fit the MSW for fosfomycin treatment of rabbits infected with S. aureus ATCC 29213; therefore, the existence of the window must be demonstrated not only in vitro but also in vivo. Further research is needed on the exact mechanism of resistance.