The clinical characteristics and the kinetics of the disposition of the hepatomegaly associated with acute, uncomplicated Plasmodium falciparum malaria were investigated in 162 children in an endemic area of Nigeria. Hepatomegaly was significantly more common in the younger than in the older children. Complete resolution occurred in 48% following antimalarial chemotherapy. In the children in whom hepatomegaly did not resolve, a reduction in liver size of < 17% by the time parasitaemia was cleared (usually on day 3) was associated with non-resolution of hepatomegaly by days 7 or 14 of follow-up. An increase in liver size to at least 125% of the baseline value by day 4 or 5 was associated with a lack of therapeutic response, providing the child involved was aged < 5 years. In the children who had complete clearance of parasitaemia and resolution of hepatomegaly, there was no significant relationship between the parasitaemia-derived conventional indices of therapeutic response [i.e. time to clearance of 50% (PC50) or 90% (PC90) of the parasitaemia, and the parasite-clearance time (PCT)] and the corresponding parameters derived from measurement of liver size [i.e. time for resolution of 50% (HR50) or 90% (HR90) of the hepatomegaly and the hepatomegaly-resolution time (HRT)] in the same patients. However, as the HR50:PC50, HR90:PC90 and HRT:PCT ratios were similar (range = 1.6-2.1), the liver parameters may have therapeutic application. In the children with drug-sensitive P. falciparum infections and in whom hepatomegaly completely resolved, the area produced by plotting liver size against time (i.e. the area under the curve of hepatomegaly v. time, or AUChp) increased in proportion to the liver size below the costal margin (P = 0.02, from analysis of variance), but there was no significant difference in the half-lives of hepatomegaly (t1/2hp) or in the ratios of liver size to AUChp, indicating that the kinetics of the resolution of hepatomegaly were linear in the range examined. Comparison of the kinetic indices of hepatomegaly and parasitaemia showed that, although the half-lives of parasitaemia and hepatomegaly and the corresponding clearance values were similar, there was no correlation between these parameters among those in whom hepatomegaly completely resolved and parasitaemia completely cleared. These results indicate that routine clinical measurement of the liver size in children with hepatomegaly during acute, uncomplicated, P. falciparum malaria may have some use in evaluating and monitoring the therapeutic responses of infections. The resolution of hepatomegaly, a reflection of pathological changes, lags behind clearance of parasitaemia in children with P. falciparum malaria, and supports the use of the liver 'rate' as a malariometric index for assessing the intensity of transmission in endemic areas.