The flow of carbon from plant to fungus in ectomycorrhizal associations has not been well quantified. The objective of this study was to use 14C to quantify the increase in fixed carbon translocated below ground in ectomycorrhizal relative to non-mycorrhizal willow (Salix viminalis L, Bowles hybrid). Rooted cuttings were inoculated with Thelephora terrestris (Ehrh). Fr. or left non-mycorrhizal. Non-mycorrhizal plants were grown at the same (4 mg kg−l bicarbonate-extractable P) (NM-P) or at a higher (21 mg kg−1) soil P concentration (NM +P), one at which the non-mycorrhizal plants were similar in size to the mycorrhizal (M-P) plants. At 41, 51, 76 and 89 days after planting, the shoots were exposed to a pulse of 14CO2. Plants were harvested after a 202 h chase period. The 14Cactivity was quantified in live fractions: shoot tissue, shoot respiration, ‘root’ tissue (= roots plus fungi), ‘root’ respiration (= CO2 released below ground) and soil. Of the total 14C detected in these five fractions, M-P plants allocated from 3.9% (harvest 1) to 11.5% (harvest 4) more to the below-ground fractions (‘root’ tissue, soil and ‘root’ respiration), than did the NM-P plants. Differences between NM+P and M-P plants were only half of those above (3.1 % and 4.4% at harvests 2 and 3, respectively, compared to 6.4 % and 7.4%, respectively for the difference between NM-P and M-P). Correction for differences in root/shoot ratio between M-P and NM-P plants eliminated the observed differences in carbon distribution only at the first three harvests. There was no evidence for increased ‘root’ respiration fates or rhizodeposition being responsible for the increased carbon diverted below ground by M-P plants.