Mycelial cord systems, up to 50-cm diameter, of the basidiomycete Phanerochaete velutina (DC.: Pers.) Parmasto, a common woodland saprotroph, grown on non-sterile soil in model laboratory microcosms were baited, after 27 d, with pairs of fresh beech wood blocks (baits), placed at 10 d intervals behind the foraging colony margin. System development was quantified by image analysis. Mean radial extent and hyphal cover increased linearly with time until day 21, but declined before the mycelial systems reached the edges of the laboratory microcosms. The mass (DBM) and border (DBS) fractal dimensions of the mycelial systems changed with time but the ratio DBM∶DBS became constant after 14 d. A separate central compartment containing the inoculum was supplied with 32P orthophosphate and its translocation to wood baits monitored non-destructively for 73 d. Whilst total 32P acquisition by wood baits increased linearly with time, the proportion of total allocated to baits varied significantly both temporally and according to the length of time that baits had been in contact with the mycelium. Most recently supplied wood baits were not the main sink for supplied phosphorus; rather, the rate of 32P acquisition was initially greatest in baits from which egress of the fungus had already occurred. The rate of 32P acquisition by the most recently added baits increased with time, supported by efflux from other wood baits, which had initially been the main sinks for translocated phosphorus. The results raise important questions about the ecological and functional significance of nutrient partitioning in cord systems and imply that ‘observed’ translocation, rather than being an absolute measure, indicates the degree to which phosphorus is loaded from a translocation stream in regions where it is being actively utilized and/or stored.