Despite the widespread use of ice packs in conjunction with ace bandages, padded ace bandages and compression dressings in the management of acute soft tissue trauma and sprains, the role of these barriers in the conduction of cold has not been adequately evaluated. Thermal probes were secured to both ankles of 62 healthy volunteers. The selected barrier was applied bilaterally and 500 g of chipped ice in a plastic bag was placed over the barrier on the right ankle for 30–45 minutes. A significant reduction in temperature of the right ankle was noted in all groups except in the padded ace group. The most rapid decrease in temperature was noted during the first 2 minutes of treatment. During the first minute, the surface temperature decreased an average of 22°C (ace), 3·8°C (dry washcloth), 5·2°C (no barrier) and 5·4°C (damp washcloth). After 10 minutes, the mean rate of cooling was 0·1–0·2°C and was approximately the same in all four groups. The Scheffe procedure indicated three homogeneous subgroups at 30 minutes: no barrier, damp washcloth; ace, dry washcloth; and padded ace. These findings question the clinical usefulness of the application of cold over padded aces and compression dressings and the use of a damp washcloth to ‘protect’ the sk