Experiments on specially constructed microplots were undertaken to overcome some of the difficulties associated with population studies using field plots. In these microplots Cruciferae caused greater increases in beet‐eelworm populations than Chenopodiaceae when the initial population was low and greater decreases when it was high. Non‐hosts and inefficient hosts caused reductions of the same order as host plants in the same families.When sugar beet was grown at varying initial populations, the final population tended to rise to a ‘ceiling’ which varied with soil and season. The ‘ceiling’ effect is also observed in the field and in pot tests, and may be used as a measure of the efficiency of different host plants or, if the same host plant is grown, as a measure of the effect of different soils and seasons. A linear relationship exists between the logarithm of the initial eelworm population and the yield of roots and tops.Eelworm attack upon sugar beet renders the plants incapable of profiting from favourable growing conditions and reduces the size of the plants without affecting the sugar content of the roots. Within certain limits, the final eelworm population is independent of yield or plant density. Cruciferae are far less susceptible to injury by root invasion than the cultivated varieties ofBeta vulgaris.It is suggested that host efficiency in raising and supporting eelworm populations and susceptibility to injury are distinct and independent attributes of host plants. Decay of populations in land under fallow or non‐host crops in non‐host families is approximately 20, 40 and 50% per annum for cysts, cysts with contents, and eggs respectively, and appears to be largely independent of initial population level. The average egg content of cysts with contents falls from 177 eggs/cyst in the first winter after the cultivation of a host crop to 131, 118 and 114 in the second, third and fourth respectively. Thein vitrohatch from beet‐eelworm cysts is affected by previous cropping. Low hatches are sometimes obtained in the winter and spring after the cultivation of host crops, especially where there has been considerable increase. A period of maturation appears to be necessary before eggs are fully sensitive to hatching stimuli.The implications of the microplot results are briefly discussed. Population increase appears to be limited by intraspecific competition rather than by specific enemies. The development and decline of infestations follows a course similar to that in other pests but, because of lack of mobility, the time factor is greatly extended. The differential effects of soil, season and host crop render set rotations ineffective as a measure of control once land has become generally infested. At this stage a system of advice based on soil samp