The aim of this study was to determine the effect of pulsing chlorhexidine gluconate, at concentrations commonly used in mouthwashes, on Streptococcus sanguis biofilms and microcosm dental plaques in vitro. Biofilms were grown on bovine enamel and nutrients were supplied in the form of artificial saliva. Pulsing experiments were carried out on steady-state biofilms using 0.05 or 0.2% chlorhexidine solutions delivered twice daily for 1 min. In a separate study, the enamel discs on which the biofilms were formed were pre-treated with chlorhexidine and pulsed directly after inoculation and then at regular intervals. With both concentrations of chlorhexidine used, a c.2 log10 reduction in the viable counts of S. sanguis was achieved with the initial pulse, but as pulsing continued, the bacterial population recovered, albeit not to the previous level. A c.1 log10 reduction in the total viable counts of the microcosm plaques was seen after the first pulse with 0.2% chlorhexidine. The total count then recovered rapidly and, after the fifth pulse, the total viable counts were not significantly different from those before pulsing. The total counts then remained at a similar level throughout the course of the experimental runs. Pre-treatment of the enamel discs with 0.2% chlorhexidine before inoculation produced viable counts of c.10(5) cfu/mm2, a 1 log10 reduction compared with untreated discs. After pulsing with 0.2% chlorhexidine at 8 h, a 3 log10 reduction was seen in the total aerobic and anaerobic counts, but again the viable counts subsequently increased despite twice-daily chlorhexidine pulsing. Regardless of the nature of the biofilm, pulsing initially achieved substantial kills, but the viability of the biofilms subsequently increased despite continued pulsing. Chlorhexidine was effective at reducing the viability of microcosm plaques when it was applied to the substratum before exposure to bacteria and subsequently pulsed on to the biofilms.