Objectives ‐ Patent foramen ovale (PFO) is a risk factor for stroke of undetermined (cryptogenic) origin. Low cost and non‐invasive bedside tests for detection of PFO are needed as alternatives to contrast transesophageal echocardiography. We investigated whether dye dilution curves and oximeter recordings are useful for detecting PFO and what is the prevalence of PFO in patients with cryptogenic stroke determined with these bedside methods. We also studied whether stroke risk factors, number of brain lesions, and stroke recurrence rates were different in patients with an unexplained stroke with and without PFO. Material and methods ‐ Dye dilution curves and oximeter recordings with non‐invasive earpiece apparatus were obtained in 59 patients aged under 50 years who had had a cryptogenic brain infarction. The number of ischemic lesions in the brain was counted by MRI. Results ‐ PFO was found in 24 (41%) of 59 patients. There was a 100% concordance in results obtained by dye dilution and by oximetry. Risk factors for stroke were similar in subjects with PFO and those without PFO. No significant association was found between PFO and Valsalva‐like activity at stroke onset. Those with PFO did not have more ischemic lesions detected by MRI nor did they have more recurrent ischemic episodes. Conclusion ‐ Dye dilution and oximetry are cheap and useful methods for detection of PFO and could be used for screening of the risk of paradoxical embolism. Because these 2 methods were not compared with the golden standard, transesophageal echocardiography, the specificity and sensitivity of the tests remain unsettled.