Using precision calorimetric techniques for measuring extremity heat loss, possible contraindications or advantages of alcoholic or hot drinks or smoking just prior to cold exposure were assessed. Ten subjects were given 60‐min exposures with one hand and one foot immersed in 15° C water, following the drinking of 1. 200 ml of 23° C water (control), 2. 200 ml of 23° C water containing 22 ml of absolute alcohol, 3. 200 ml of 55° C bouillon, 4. 200 ml of 23° C water and smoking 2 cigarettes. No significant effects were noted in the foot, but significant differences were observed in hand heat loss to the calorimeter; smoking depressed hand heat loss, alcohol elevated it. Smoking had a persistent influence, but the effect of alcohol had disappeared after 30 min. Bouillon had an effect similar to alcohol, but of shorter duration, and also resulted in a significantly reduced heat loss after 30 min. Alcohol produced a significantly higher amount of cold induced vasodilatation (CIVD) with greater total temperature elevation in 3 fingers than did smoking, although smoking did not prevent CIVD. With intake of bouillon, CIVD appeared earlier than with alcohol, and during the 1st 15 min, total temperature elevation in the 3 fingers was markedly higher than in the control run, but not as large as with alcohol. All differences were relatively small and would seem to be of little practical importance.