From 1970 to 1983, 1,646 lung cancer patients were referred for treatment to the Hunter Radiation Therapy Center, Yale-New Haven Hospital. Forty-three patients had clinical Stage I non-small cell lung cancer felt to be surgically resectable but were treated with radical radiation therapy either for medical reasons (37 patients) or because the patient refused surgery (six patients). This group of clinical Stage I lung cancer patients is understaged by modern criteria since the majority of patients did not have thoracic CT scans and staging was based on fairly limited clinical and radiographic studies. The histological diagnosis was squamous cell carcinoma in 53% of the Stage I patients, adenocarcinoma in 25%, and other non-small cell histologies in 22%. All patients were treated with megavoltage irradiation and the mediastinum was treated in 88% of the patients. Eleven patients were treated with a continuous course (CC) and 32 patients received split course (SC) therapy based on physician preference. The CC consisted of a median fraction size of 200 cGy to a total median dose of 5900 cGy in 6-7 weeks. The SC used a median fraction site of 275 cGy to a total median dose of 5400 cGy over a 6-week period with a 2-week rest in the middle of treatment. The actuarial survival rate of the 43 clinical Stage I patients was 36% at 3 years and 21% at 5 years. Intrathoracic failures occurred in 39% of the patients. Despite the fact that the CC group was similar to the SC group in terms of age, histology, and tumor extent, the CC patients had a lower thoracic failure rate (2/11) versus 15/32), a longer median survival (51.6 months versus 27 months), and a better actuarial 5-year survival rate (45% versus 12%) when compared to the SC patients. Using Cox regression analysis to compare survival curves, the CC group had a significantly better survival compared to the SC group (p = .04).