This randomised controlled trial of routine amniotomy was carried out in a developing country setting to investigate the effect of this common procedure on the duration of labour, intra-partum interventions and selected newborn and maternal outcomes. In a Jerusalem teaching hospital, 533 multiparous and 157 nulliparous low-risk women were randomised to either amniotomy or intent to conserve membranes. For multiparae, the median duration from randomisation to full dilatation was 95 and 160 min, respectively in the intervention and control arms (p < 0.001); for nulliparae it was 210 and 270 min, respectively (p < 0.001). In both groups, oxytocin was used less in the intervention arms (p < 0.001), and no difference in mode of delivery and immediate outcomes was detected. However, given the risks of this intervention and these study findings indicating an overall short duration of childbirth, amniotomy should be limited to cases of abnormal progress of labour.