Wang (Can. J. Botany, 38, 635–642 (1960)) concluded that the accumulation of radioactivity observed on radioautographs at infection sites on rusted leaves fed with C14-labelled substances was 'apparent' rather than real. The ‘accumulation ratio’ is defined as the ratio of the specific activities (c.p.m./mg dry weight of intact tissue) of rust-infected to uninfected areas of infected leaves. Theoretical considerations relating to the radioautography of leaves labelled with C14 and to the measurement of ‘accumulation ratios’ by extraction of C14-labelled substances from rusted and uninfected segments of infected leaves, as well as experimental data, show that Wang's conclusion is not generally applicable.Experimentally, it was shown using polymethacrylate C14 sources that differences in distance between sources and X-ray film of the order of 100 μ had no effect on the intensity of autoradiographs. Rust-infected leaves, fed with radioactive glucose, were radiographed between X-ray plates. Localization of radioactivity at infection sites was observed on both ‘dorsal’ and ‘ventral’ radiographs, indicating a real accumulation per unit area. Ventral were more radioactive than dorsal surfaces. The main development of the fungus occurred on the former. Radioautography revealed that C14 from glucose-1-C14, glucose-6-C14, and uniformly labelled glucose fed to excised wheat leaves became localized at 10-day-old rust infections in 2 hours. ‘Accumulation ratios’ calculated from the specific activity of leaf segments remained close to 1.0 for at least 6 hours after introduction of the tracer, but increased to more than 2 after 24 hours. When ‘accumulation ratios’ were calculated from the specific activities of individual pustules (excised with a punch 1 mm in diameter) and interpustular disks, values greater than 1 were observed in 2 hours, thus confirming the results of autoradiography. Differences between the ‘accumulation ratios’ observed with glucose-6-C14 and glucose-1-C14 were consistent with an increased role of the pentose phosphate pathway at infection sites. Incorporation of C14 from uniformly labelled glucose into the alcohol-insoluble fraction of rusted leaf segments was 2.5-fold that in uninfected segments in 6 hours and 3.65-fold in 24 hours. The humin formed during hydrochloric acid hydrolysis accounted for approximately 50% of the activity of the alcohol-insoluble material. The ‘accumulation ratio’ for the alcohol-soluble material was only 1.56 after 24 hours.All the results support the view (Shaw and Samborski, Can. J. Botany, 34, 389–405 (1956)) that there is a quantitative, metabolically dependent accumulation of C14 from radioactive glucose at vigorous rust infections. The relative roles of fungus and host in this process are discussed briefly.