We set out to study the relationship between the timing of hospital admission and the progress of labour. In all there were 591 healthy primiparous women with normal pregnancies and who anticipated normal deliveries; 436 of them had come because of contractions. When the intrinsic speed of labour (estimated from the status of the cervix at the time of admission in relation to the duration of regular contractions) was allowed for, women coming early (regular contractions for 4 h or less before admission) as compared to those coming late stayed longer in the hospital before delivery but had a shorter total length of labour. They had more interventions during labour, more caesarean sections, longer postpartum hospital stay, and their children had more diagnoses of difficult delivery. Women who came because of ruptured membranes without contractions had shorter labours, more interventions during labour, longer postpartum stay and their children had more discharge diagnoses and longer hospital stay than women coming late. This study suggests that too early admission to the hospital may negatively affect the progress of labour, and controlled trials are needed to confirm or to disprove this suggestion.