The effect of tampon usage on the vaginal microflora of 35 healthy women was determined following their random allocation to either tampon or napkin use for three consecutive menstrual cycles. Sequential and semiquantitative vaginal cultures were obtained on days 3 +/- 2, 15 +/- 2, and 25 +/- 2 of the menstrual cycle (day 1, first day of menses) before and after randomization. Before randomization, the rate of isolation and median counts of facultative lactobacilli were significantly higher (P less than .05) and that of eubacteria was significantly lower (P = .026) among regular tampon users than among exclusive napkin users. After randomization, only median counts of coagulase-negative staphylococci were significantly increased (P = .025) during tampon use compared with the rates for the same women during napkin use. These shifts in vaginal microflora occurred only in samples obtained during menstruation and not in those obtained at other sampling times. The data presented here support the notion that the use of tampons may result in alterations in the autochthonous vaginal microflora. It remains to be determined if these ecologic shifts in the vaginal microflora may adversely affect resistance to colonization by potential pathogens in the lower female genital tract.