We present the methodology and data indicating the feasibility of a five-study collaborative stroke trial. The core study, the Warfarin-Aspirin Recurrent Stroke Study (WARSS), is a prospective, randomized, double-blind multicenter trial comparing warfarin and aspirin in preventing the recurrence of ischemic stroke. The WARSS primary endpoint is the earlier of death or symptomatic ischemic stroke recurrence. The primary null hypothesis is that, over 2 years, warfarin therapy will not differ in frequency of death or ischemic stroke recurrence compared with platelet antiaggregant therapy (aspirin). The target sample size is 1,920. Four related studies use the WARSS data. The Antiphospholipid Antibodies and Stroke Study (APASS) investigates the significance of antiphospholipid antibodies in ischemic stroke; the Patent Foramen Ovale in Cryptogenic Stroke Study (PICSS) assesses the risk of stroke recurrence or systemic embolization in medically treated cryptogenic stroke patients with and without patent foramen ovale; and the Hemostatic System Activation Study (HAS) studies the F1 + 2 fractions of prothrombin in a subgroup of patients to allow comparison between the international normalized ratio (INR) and the evidence of prothrombin split products activity. The Genes in Stroke Study (GENESIS) investigates whether the double deletion genotype of the angiotensin converting enzyme is a risk factor for stroke. Features of the collaboration are cost savings in research arising from the use of a common data base; standardization through use of one central laboratory for processing of INR values, and double blinding by fabrication of INR values for aspirin patients. WARSS randomization began June 22, 1993. Cumulative patient recruitment through June 1, 1996 exceeds 1,400 at 50 sites.