Geochemical patterns for elements, such as Sn, W and Au, present in drainage sediments as resistate heavy minerals are often erratic and difficult to interpret. To investigate the source of these problems and develop methods of eliminating them we have compared the behavior of Sn, present as cassiterite, and associated pathfinder elements downstream from a small primary Sn deposit in Perak, Peninsular Malaysia. Dispersion trains for the pathfinder elements are characterized by smooth decay patterns and differences in concentrations between high- and low-energy environments, characterized by coarse-and medium-grained sands respectively, are not significant. In contrast, Sn (and magnetite) concentrations are extremely erratic with significantly higher concentrations in high- compared to low-energy environments. As a result the dispersion train for Sn exhibits no regular decay pattern away from its source. These findings suggest that the action of the stream is analogous to that of sluice box, with light minerals being winnowed away and cassiterite, together with magnetite, accumulating. For all but the finest sizes this process, which is most efficient in high-energy environments, causes considerable local variability in Sn content of the sediments. However, because the hydraulic behavior of cassiterite and magnetite is similar, but magnetite is not associated with the primary mineralization, the Sn/magnetite ratio can be used to eliminate Sn anomalies resulting from local variations in hydraulic conditions. The concept of hydraulic equivalence of cassiterite and magnetite was extended to examining the relationship between Sn and different size fractions of the light minerals that constitute the bulk of most sediments. Greatest contrast is obtained when the Sn content of the −270 mesh (−53 μm) fraction is re-expressed as its hydraulic equivalent concentration in −65 + 100 mesh (−212 + 150 μm) material. For exploration purposes it is concluded that: (1) providing cassiterite is present in the fine size fractions, sampling of this material will reduce hydraulic effects, thereby reducing data variability, and can also increase the length of the anomalous dispersion train; and (2) hydraulic effects can also be reduced by re-expressing Sn concentrations as ratios to magnetite (provided this is not associated with the primary Sn mineralization) or a hydraulically equivalent size fraction of the light minerals that constitute the bulk of the sediment. Similar principles probably apply to the interpretation of geochemical data for other elements dispersed in drainage sediments as heavy minerals; this warrants further investigation.