We report our experience using the buttoned device to close defects within the oval fossa and probe-patent oval foramens, comparing the findings with those obtained with the Amplatzer septal occluder. From 1992 to 1997, we used the buttoned device to close defects in 73 consecutive patients, 64 with defects in the oval fossa and nine with patent foramens. We compared this experience with a further series of 62 patients seen from 1997 to 1999 in whom the Amplatzer septal occluder was used. Successful implantation was achieved in three-quarters of those with septal defects in whom the buttoned device was used, in all of those in whom the buttoned device was used for patent foramens, and in nine-tenths of those in whom closure was attempted using the Amplatzer occluder. Immediate surgery was needed in 3 patients in whom a buttoned device was used, one because of embolization and two with residual shunts and a straddling device. Similar immediate surgery was needed to retrieve one embolized Amplatzer occluder. During follow-up, surgery was needed in a further 7 patients, all having had insertion of a buttoned device, because of atrial perforation in one and a significant residual shunt in the remainder. At late follow-up, the rate of complete occlusion was 69% in the patients in whom the buttoned device was used to close a septal defect, 100% when the buttoned device was used for patent foramens, and 95% in those treated with the Amplatzer occluder. Our experience shows that the Amplatzer occluder produced a significantly higher rate of occlusion for larger defects, and with a shorter fluoroscopy time than the buttoned device. The Amplatzer septal occluder, therefore, is our preferred device for closure of defects within the oval fossa.