The stability of praziquantel (PZQ)-insusceptible S. mansoni isolates and the possible selection of PZQ-insusceptible parasites upon applying therapeutic pressure were examined over several life cycle passages (snails to mice). To test isolate stability, 3 PZQ-susceptible and 7 PZQ-insusceptible isolates were used to establish infection in mice, and they were passaged each for 2–5 life cycles. After each passage, 6 groups of mice were used to assess the PZQ dose at which the worm burden was decreased by 50% (ED₅₀). Five of them were treated with doses of PZQ (12.5, 25, 50, 100, and 200 mg/kg for 5 days) 7 wk after infection; the last group represented infected, but untreated, controls. Possible selection of PZQ-insusceptible parasites under therapeutic pressure was examined by subjecting 1 PZQ-susceptible and 1 PZQ-insusceptible S. mansoni isolate to therapeutic pressure by PZQ for 8 passages. After the final passage, PZQ ED₅₀ was estimated. All PZQ-susceptible S. mansoni isolates showed stable susceptibility to PZQ (mean PZQ ED₅₀ = 85 mg/kg) over all passages. Two of the 7 PZQ-insusceptible S. mansoni isolates (S47 and ER5) showed normal sensitivity to PZQ in 1–2 passages (although not the last passage, and without a declining ED₅₀ profile), whereas the remaining passages kept a sustained insusceptibility to the drug (mean PZQ ED₅₀ = 217 mg/kg). Worm maturity and sex were irrelevant to variability in drug ED₅₀ within an individual isolate over different passages, revealing the heterogeneous nature of the parasite. Therapeutic pressure for limited life cycle passages did not result in a significant increase in drug ED₅₀. The fact that reversion of some of the PZQ-insusceptible S. mansoni isolates to normal drug-sensitive state is not long lasting and that the therapeutic pressure by PZQ in the field is not comparable with that in the laboratory (unlimited), make monitoring the response of patients to the drug in the field an integral part of schistosomiasis control measures.