The effects of asthma per se and of various oral steroid-treatment schedules on linear growth in 231 children with chronic severe asthma (6.2-16.2 years of age) were evaluated. Asthmatic children who had never received steroids and those who received occasional steroids, had comparable growth retardation, averaging approximately 1 standard deviation (s.d.) below their age- and sex-specific predicted means. Children who had been treated for 2 years or longer with alternate-day or daily steroids also did not differ from each other, with both groups averaging approximately 2 s.d. below their predicted means. Growth stunting, intermediate in severity, was found in children treated with frequent intermittent steroids. For groups with steroid medication-histories of never, occasional, intermittent, alternate-day and daily, the percentage of children with 2 or more s.d. below their predicted heights were 25, 17.4, 28, 50 and 45.5 respectively. The duration and dosage of daily steroid-treatment correlated positively with severity of growth retardation. In addition to norms, the children were also compared with a healthy, same-sex sibling and with predictions from their parents' heights. Regardless of the comparison used, the results were similar. Since genetic and socio-economic factors are controlled by the latter two comparisons, the data indicate that these factors played no role in growth retardation in these children. Growth retardation in asthma appears to be due to the disease itself, and is accelerated by steroid therapy.