On the basis of clinical and experimental observations, it has been recorded by many authors1 that the use of tobacco produces definite effects on the vascular system, which are particularly important in certain disease conditions, such as thrombo-angiitis obliterans and angina pectoris. In spite of this widely recognized association, the available evidence has tended to show that the smoking of tobacco must be looked on as an aggravating rather than an etiologic factor. Following the work of Maddock and Coller,2 experiments have been reported by Barker,3 Johnson4 and Wright,5 proving that the smoking of cigarets will produce in the great majority of normal individuals, under certain conditions, a marked change in the surface temperature of the extremities. This can be measured at the tips of the fingers and toes. We noted that in many instances slowing and even stoppage of the blood flow occurred in