The records were reviewed of 334 patients with a diagnosis of pectus excavatum at Presbyterian Hospital in New York, New York, from 1948 through 1977. No operation was done on 168; 174 operations were done on 166. In those not operated upon 18% of infants died of associated anomalies. The deformity improved or disappeared in half of the surviving infants and in some children up to the age of 6 years. Thereafter, it remained the same or worsened. Mild deformities were compatible with long life without symptoms; severe deformities were associated with chronic disability. In those operated upon, there was no surgical mortality. Results were satisfactory 5 plus years after operation in 68 of 102 patients. Late results deteriorated through adolescence and were more clearly related to method of correction than to sex, severity, age at operation, or time after operation. It is concluded that (1) operation is justified in appropriate cases; (2) it is seldom indicated before the age of 3 years; (3) results are best following operations between ages 3 and 6 years, although they may be satisfactory at any age; (4) results of radical operations are superior to results of simpler ones; and (5) better objective methods of evaluation over many years are needed to judge the value of any procedure.