L. H. GRAY 



Table 1. Dependence of aberration coefficients on temperature 



Temperature 



i C 18 C 36=C 



Minimum (anaerobic) value of a 1-68±011 1-70 ±0-11 2 -03 ±013 



Maximum (oxygen) value of a 4-54±0-28 5-25±0-33 5-87±0-37 



2-70 + 0-25 3-10±0-27 2-90±0-27 



aO 



IN VIVO IRRADIATIONS 



The in vivo measurements posed problems different from those which we 

 encountered in our attempt to measure the relation between radio-sensitivity 

 and oxygen tension in vitro, since the only control which we could exert over 

 the oxygen tension in the peritoneal cavity was through the composition 

 of the gas respired by the animal, and it was necessary to try to measure the 

 oxygen tension in the fluid in the peritoneal cavity of the living animal. The 

 bubble technique was applied again, but in the following way : 



Typically, a five-day old tumour in our animals consisted of 1 to 2 ml. of 

 fluid containing about 10^ tumour cells/ml, some inflammatory cells, and 

 a variable amount of blood. A small bubble of gas was introduced by 

 hypodermic needle into the peritoneal cavity 30 min before irradiation, and 

 equilibration was assisted by massage of the abdomen. The size of this 

 bubble, which consisted of 0- 1 ml. of either nitrogen or a 5 per cent oxygen 

 in nitrogen mixture, was chosen to be small enough to come into equilibrium 

 with the fluid during the 30 min before irradiation but large enough not to 

 follow the rapidly falling oxygen tension of the fluid which results from con- 

 tinued cellular respiration after the death of the animal. Sampling the 

 bubble was a delicate operation which was carried out as quickly as possible. 

 The spinal cord was severed immediately after irradiation, the muscular 

 wall of the abdomen was exposed, and the animal suspended ventral side 

 upwards by clamping the abdominal wall. The bubble rose and became 

 located near the clamp. A gas pipette having a sharp glass tip was made to 

 pierce the abdominal wall below the level of the fluid, and the tip was moved 

 up into the bubble. A small sample of gas was drawn into the pipette. 

 Before withdrawing the pipette the tip was once again brought into a position 

 below the fluid into the peritoneal cavity, and some fluid was taken in to seal 

 the gas sample. The volume of the sample was noted and its oxygen content 

 assayed by means of the Hersch analyser, as already described. A second 

 sample of gas was taken in another pipette about 1 • 5 min later. Comparing 

 the second sample with the first, it was evident that oxygen tension in the 

 bubble was falling at about 3 mm mercury per min. Extrapolation back to 

 zero time gave an estimate of the oxygen tension in the bubble at the moment 

 when the animal was killed, i.e. at the end of irradiation. The range of values 

 observed and their mean is given in Table 2 for animals breathing gas which 

 contained 8-5, 10, 21 (air), and 100 per cent oxygen. 



The table shows the corresponding mean values of the aberration coeffi- 

 cient a and of the ratio a/ocj^. 



Two additional measurements of radio-sensitivity were made shortly after 

 the introduction of 10^^ or lO^^ moles hydrogen peroxide into the peritoneal 

 cavity of tumour-bearing animals which were respiring air. When such 



81 



