In 1948 Wooler reported the results of treatment in 47 patients with cardiospasm; he concluded that the treatment of choice was with the Negus hydrostatic dilator and, if this failed, Heller's operation. In 1950 Allison brought the total to 76 patients and recommended using the Negus dilator in all but three groups of patients, namely, (1) small children, (2) those not cured by the Negus dilator, and (3) those in whom the oesophagus had become so dilated and tortuous that the cardia could not be reached through the oesophagus. It is our intention to extend these reports and describe the results in 137 patients with cardiospasm who have been treated at Leeds. Many different procedures have from time to time been advocated in the treatment of cardiospasm. Most of these were recently reviewed by Acheson and Hadley (1958), who found that the only procedures to have withstood the test of time were dilatation of the cardia through the oesophagoscope, usually with a Negus dilator, and cardiomyotomy, first described by Heller (1914). However, no definite conclusions appear to have been reached as to which of these two procedures is the more desirable, and, in fact, there has been a reluctance to use Heller's operation. Olsen, Harrington, Moersch, and Andersen (1951) reported 601 cases treated by the Negus dilator, but in 1958 Ellis, Olsen, Holman, and Code rather cautiously reported their results in 55 patients treated by Heller's operation; a similar cautious attitude can be detected in the report of 22 cases by Hawthorne and Nemir (1953) and that of Acheson and Hadley (1958) in 35 patients. By reviewing both the results of treatment and the trend of treatment in this series, it is intended to demonstrate a choice between the two procedures and dispel some of the doubts about Heller's operation. MATERIAL.-One hundred and fifty-nine patients are listed in the records with a diagnosis of cardiospasm, and 22 of these were rejected because of doubt about the diagnosis or incomplete documentation, leaving a total of 137. These patients were seen between 1939 and August, 1956; 74 were women and 63 were men. The age distribution (as recorded when the patient was first seen) is shown in Fig. 1. The high incidence in the fifth and sixth decades was noted by Allison and is an expression of the time that many patients are prepared to tolerate symptoms.