Rooting intensity and root distribution of Brussels sprouts and leeks were assessed in field experiments and in the Wageningen Rhizolab, a rhizotron facility built in 1990, in relation to their contrasting N uptake behaviour (Brussels sprouts usually take up N rapidly after planting, resulting in soil depletion, while leeks generally use only half of the available N). In field experiments rooting was quantified in the various layers of the profile as the volumetric root length density (RLD; cm/cmsuperscript 3) and in the Rhizolab with horizontal glass minirhizotrons as the number of roots per cmsuperscript 2 (NR). Regression of RLD and NR on thermal time (accumulated average daily temperature above ground) after planting revealed that the rooting depth of Brussels sprouts increased faster with thermal time than that of leeks (0.13 vs. 0.08 cm/degree-day, respectively in one of the Rhizolab experiments). Furthermore, leeks showed an unusual distribution of roots in the profile with maximal rooting intensity at depths of 10-20 cm. In all experiments the proliferation of roots (calculated as the increase in NR or RLD with thermal time) was slower in leeks than in Brussels sprouts, especially in the deeper layers of the soil profile. Therefore leeks can be considered as a shallow rooting crop. The field experiments generally showed the same difference between the crops, Brussels sprouts having greater root density, deeper rooting and faster root proliferation than leeks. For both crops, however, root proliferation at the deeper layers was much slower in the field than in the Rhizolab, probably due to unfavourable conditions in the subsoil caused by higher bulk densities or temporary high water tables. The consequences of different rooting patterns for N utilization are discussed.