EXPERIMENTAL studies of Murphy, Maisin, and Sturm (1), Russ and Scott (2, 3) and Mottram (4) have clearly demonstrated that the lethal dose of roentgen rays or gamma rays from radium for tumors growing in animals is much smaller than the dose required to kill the same tumors in vitro. The results are confirmed and extended by our recent studies (5, 6, 7). Figure 1 shows the survival curves of Mouse Sarcoma 180 exposed in vivo1 and in vitro2 to 200 kv. roentgen rays, filtered with 0.5 mm. Cu and 1.6 mm. A1, at a distance of 50 cm. It shows that when Mouse Sarcoma 180 is irradiated in vitro the dose of filtered roentgen rays necessary to kill all the fragments of tumor is between 2,800 and 3,000 roentgens (measured in air). When tumor implants growing in animals are treated with similar radiation, the dose necessary to cause complete regression is about 1,800 roentgens. The nature of this reaction is complex, and little light has been thrown on its mechanism in the last ten years (8). The present study was undertaken to gain further information regarding the mechanism of the radiosensitivity of tumors. Irradiation of Tumors in vivo In a previous study (7) we found that when tumors, 0.3 to 1.5 cm. in diameter, were irradiated in vivo through a hole in a lead shield slightly larger than the cross-section of the tumor, no tumor regressed with a dose of less than 500 r, and very few with less than 750 r. With a dose of 1,000 r, about 50 per cent of cases showed regressions. With a dose of 1,500 r, about 85 per cent of cases showed complete regression, but with a dose greater than 2,000 r, tumor regression occurred in 100 per cent of cases. These successfully treated tumors disappeared completely in about 24 days. However, we do not know the length of time which elapses from the first disturbance until the tumor cell dies. There are, however, a limited number of reports upon the latent effect of radiation in vivo. Thus Mottram (9) irradiated tumors (Mouse Sarcoma 180) in mice with 20 mg. of radium element for 20 minutes. After varying lengths of time the tumors were removed and fragments were implanted into normal animals. The results showed that if tumors are left for four or five days in the animal after irradiation, and then removed and transplanted into other animals, they are less apt to grow than when they are irradiated and transplanted immediately. He attributes this difference to the effect of radiation on the blood supply. Kawakami (10) reported that if Kato rabbit sarcomas are irradiated with x-rays (3 to 20 S.E.D.) while in the hosts, and immediately extirpated and implanted in other non-irradiated rabbits, they will grow progressively. But, if they are allowed to remain in the original hosts for 17 hours after irradiation with a dose as small as 5 S.E.D., many of them fail to grow.