From a prospective study concerning 576 patients with Raynaud's phenomenon, the authors studied the results of Allen's clinical test and their etiologic significance. These results were compared to those obtained by the instrumental technique using the Doppler probe to locate and compress radial and ulnar arteries and digital plethysmography to measure the effects of this compression. This instrumental technique makes it possible to investigate hand vascularization when Allen's clinical test is non interpretable or impossible which corresponds to 30% patients. A pathological Allen's test, whether it be clinical or instrumental, rather favors a Raynaud's syndrome (p < 10(-4)). The absence of ulnar vascularization was the abnormality most often noted and was rather found in severe Raynaud's phenomena, the sclerodermatous particularly (p < 10(-4)).