A neutron moderation meter was used to measure soil moisture 0–4 feet deep in plots of sugar beet carrying two plant populations (8800 and 54000 plants/acre), each with and without irrigation. Recordings began in April or May in each of three years (1967–9) after sowing the crop and continued at 1 or 2-;week intervals until harvest in October.

The measured soil moisture deficits were very similar to potential deficits calculated from meteorological measurements. This indicates that the crop could extract the water needed for transpiration from the soil even when the deficits were quite large (more than 5 in in 1967), which probably explains the small response to irrigation by sugar beet in England.

When the soil moisture deficit increased rapidly early during the season (1967), the crop extracted water from the soil by exhausting the available water from progressively deeper horizons, whereas when the deficit increased rapidly late during the season (1969) water was still being extracted from all horizons until harvest. Both decreasing the plant population and irrigating decreased the amount of water used from depth in the profile every year.

The total amount of water used (evaporation plus transpiration), on average, from soil reserves and rainfall, was 12·2 in by the small population and 13·4 in by the large population. When irrigated, the consumption increased to 14·2 and 15·4 in. respectively. The difference in usage between populations was almost entirely from the difference in leaf cover early during the season. The water consumption in 1968, when the summer was wet, was only two-thirds of that in 1967 and 1969 when the summers were drier.