To the Editor—We read with interest the recent report by Gumbo et al., in which a hollow-fiber model was used to study the kinetics of killing of Mycobacterium tuberculosis H37Ra by isoniazid [1]. The main finding was that killing ceased after 3–4 days because of the emergence of resistant mutants. From this, the authors concluded that the “age-old” hypothesis that killing by isoniazid in vivo ceases because of depletion of log-phase bacilli is incorrect.