The distribution coefficient, Kd, is often used to quantify heavy metal mobility in soils. Batch sorption or column infiltration tests may be used to measure Kd. The latter are closer to natural soil conditions, but are difficult to conduct in clays. This difficulty can be overcome by using a laboratory centrifuge. An acceleration of 2600 gravities was applied to columns of London Clay, an Eocene clay sub-stratum, and Cu, Ni, and Zn mobility was measured in centrifuge infiltration tests, both as single elements and in dual competition. Single-element Kd values were also obtained from batch sorption tests, and the results from the two techniques were compared. It was found that Kd values obtained by batch tests vary considerably depending on the metal concentration, while infiltration tests provided a single Kd value for each metal. This was typically in the lower end of the range of the batch test Kd values. For both tests, the order of mobility was Ni>Zn>Cu. Metals became more mobile in competition than when in single-element systems: Ni Kd decreased 3.3 times and Zn Kd 3.4 times when they competed with Cu, while Cu decreased only 1.2 times when in competition with either Ni or Zn. Our study showed that competitive sorption between metals increases the mobility of those metals less strongly bound more than it increases the mobility of more strongly bound metals.