The Medical Research Council (MRC) United Kingdom trial for childhood acute lymphoblastic leukaemia (UKALL X) randomised patients aged 0-14 years inclusive with an initial white blood count of less than 100 x 10(9)/l to receive an early intensification block, a late intensification block, both, or neither. The next trial, UKALL XI, for children aged 1-14 years, randomised between different central nervous system (CNS) directed therapies. At the beginning of the trial, all patients were also randomised between late intensification alone and both early plus late blocks. The effects of both the early and the late block in UKALL X alone have been reported previously. This paper examines the effect of the addition of the early intensification block to treatment which included late intensification, combining the data from UKALL X and the first part of UKALL XI. Early intensification was associated with fewer bone marrow relapses and a reduction in the odds of death of 0.63 (95% confidence interval: 0.46-0.87). Survival was significantly improved with an increase at 5 years of 8%, from 79 to 87%. Following this demonstration that early intensification improves survival, the effect of a third intensification block is under investigation.