A cold provocation test (measurement of finger systolic pressure during combined body and local finger cooling) was performed on 111 male patients exposed to vibration and with a typical history of cold induced white finger. A new method of calculating the test result is described--namely, digital systolic blood pressure in the cooled test finger as a percentage of the systolic pressure in the arm (DP%). The conventional way of calculating the result, the systolic pressure in the cooled test finger as a percentage of the systolic pressure in the test finger when heated to 30 degrees C, corrected for changes in systemic pressure by the use of a reference finger (FSP%), requires the measurement of the systolic pressure in a reference finger. The two ways of calculating the test results give a similar sensitivity (74% for FSP%, 79% for DP% if all histories are regarded as true) but the new method does not require pressure measurements in a reference finger. This makes the test easier to perform and the result easier to understand.