Screening of the newborn for cystic fibrosis by measurement of immunoreactive trypsin has been undertaken on alternate weeks in Wales and the West Midlands for five years since 1985 to evaluate the possible clinical benefits of early diagnosis. Patients detected by screening and those diagnosed by clinical symptoms alone were assessed annually for differences in clinical, anthropometric, and biochemical variables. Fifty eight infants not considered to be at risk of cystic fibrosis (they did not present with meconium ileus and do not have a sibling with cystic fibrosis) have been detected by screening and they have been compared with 44 children who were diagnosed clinically. This latter group includes nine children whose screening was negative but who were recognised subsequently to have cystic fibrosis. The mean age at diagnosis of the screened group was significantly lower than that of the group diagnosed clinically. Excluding admissions for diagnostic tests for cystic fibrosis, the screened group spent a significantly shorter time in hospital during the first year of life. The results of all other comparisons made between the screened group and those diagnosed clinically were similar up to the age of 4 years.