The results of membranectomy and deep myectomy in the left ventricular outflow tract were compared to those of membranectomy and myotomy in 42 patients who underwent surgical repair of discrete and tunnel subaortic stenosis. Fifteen consecutive patients (Group A) underwent membranectomy and myotomy, and 27 consecutive patients (Group B) underwent membranectomy and myectomy. Two patients of Group A and nine of Group B had tunnel subaortic stenosis. The preoperative mean (+/- standard deviation) peak systolic gradients across the left ventricular outflow tract in patients with discrete subaortic stenosis types I and II were 64 +/- 29 mm Hg in Group A and 52 +/- 3 mm Hg in Group B (p = not significant). In the patients with tunnel subaortic stenosis the preoperative mean gradients were 97 +/- 74 mm Hg in Group A and 73 +/- 26 mm Hg in Group B (p = not significant). In patients with discrete subaortic stenosis types I and II, postoperative catheterization at a mean follow-up of 21 months revealed residual mean gradients of 29 +/- 24 mm Hg in Group A and 10 +/- 13 mm Hg in Group B (p less than 0.01). In the patients with tunnel subaortic stenosis, the postoperative mean gradients were 25 +/- 7 and 30 +/- 30 mm Hg in Groups A and B, respectively (p = not significant). We conclude that in the surgical management of discrete subaortic stenosis types I and II, deep myectomy (in addition to membranectomy) produces better relief of the left ventricular outflow obstruction than do membranectomy and myotomy. In patients with tunnel subaortic stenosis myectomy is less effective than in the non-tunnel type but still produces acceptable results and may delay radical procedures to a later age.