A coarse sandy soil and a sandy loam were compacted with a heavily loaded three-axle vehicle. A single pass and four repeated passes, the latter on topsoil as well as on the plough bottom, were applied. The compaction treatments were done only once and at a water content of about field capacity. The soil density was measured by means of a gamma-ray technique to a depth of 80 cm and the penetration resistance to a depth of 64 cm. The yield of a small grain cereal crop was followed during a 9 year period. A single pass with the high axle load vehicle gave a moderate compaction in both soil types, but no significant effects upon crop yield. Four passes increased the compaction effect, which could be detected to a depth of at least 60 cm in both soil types, and decreased the crop yield, but significantly for some years only. The compaction effect was highest when the load was applied to the firm plough bottom. In the loamy soil, this caused an additional reduction in yield, while the crop response in the coarse sandy soil was not dependent on the depth at which the repeated loading had been applied. In the coarse sandy soil subsoiling significantly decreased crop yield in non-compacted soil and significantly increased the yield in soil which had been compacted on the plough bottom. No effects of subsoiling could be detected in the loamy soil.