Introduction: Subjects with a weak cold-induced vasodilatation response (CIVD) to experimental cold-water immersion of the fingers in a laboratory setting have been shown to have a higher risk for local cold injuries when exposed to cold in real life. Most of the cold injuries in real life, however, occur in the foot in cold air rather than in the hand in cold water. Therefore, an experiment was conducted to investigate the within-subject relation between CIVD in the fingers and toes exposed to cold water and cold air. Methods: In 4 experimental sessions, 11 healthy male subjects immersed their toes and fingers in 5°C water and exposed the fingers and toes to -18°C cold air for 30 min. The pad temperature of the middle three digits was measured. Results: CIVD in water was more pronounced in the fingers (onset time 5.1 ± 1.8 min; amplitude 5.0 ± 2.1°C) than in the toes (onset time 10.6 ± 6.0 min; amplitude 3.0 ± 1.0°C). Out of 22 skin temperature responses to cold air, 13 were not identifiable as CIVD. The mean skin temperatures for fingers and toes during the last 20 min of cold exposure were 25.6 ± 7.1°C and 20.9 ± 6.8°C, respectively, for air and 9.3 ± 1.9°C and 7.1 ± 1.3°C for water immersion. There was a strong relation between the mean temperature of the fingers during cold-water immersion and toes during cold air exposure (r = 0.83, P < 0.01), showing that a weak CIVD response in the hand is related to a weak response in the foot. Discussion: We conclude that the cold-water finger immersion test is related to the temperature response in the toes and may thus continue to serve as a valid indicator for the risk of local cold injuries. Copyright © by the Aerospace Medical Association.