182 RADIATION BIOLOGY 



unless the object is very close, in which case 0.2 mm of celluloid mil 

 suffice; if Fi is of copper, about 0.2 mm of aluminum plus 0.2 mm of 

 celluloid will be required for Fo, and if Fi is made of tin, F2 should be made 

 of 0.2 mm of copper plus 0.4 mm of celluloid (Thoraeus, 1934; MarinelH, 

 1937; Wrenshall and Nichols, 1942; Morrow, 1946). If the object and 

 the irradiated portion of the supports are very light, the only question 

 remaining is whether the scattering from E' is hkely to introduce errors in 

 the measurement. Qualitatively the situation is analogous to that just 

 considered for the filter, except for the fact that E' is usually lead and the 

 whole residual energy of the beam is converted therein. Since the 

 Compton component of the scatter and the K radiations from lead are 

 likely to be registered correctly by the thimble chamber because of their 

 hardness, the only precaution necessary is to absorb the L radiation (15 

 kev maximum for lead), and, by so doing, eliminate the electronic com- 

 ponent as well. This can be done in practice either by covering the 

 irradiated portion of E' by a few tenths of a millimeter of aluminum or by 

 maintaining the distance PE' at about 20 cm. It should be remarked at 

 this point that keeping the distance between the object and the scatterer 

 (filter or shield) to about 20 times the diameter of the chamber is always 

 the preferable procedure, because, in addition to diffusion of poorly 

 measurable scatter away from the object, smoother gradients of dosage 

 rates at the point of measurement are always attained. This condition, 

 as already mentioned in the section on thimble-chamber calibration, is to 

 be implemented as much as possible in the interest of accurate dosimetry. 



In the event that biological experiments — because of high dose-rate 

 demand — require the placement of small biological specimens close to the 

 window of the tube or the filter, measurement must be made with very 

 shallow chambers. In this respect the extrapolation chamber presents 

 attractive features, since spacings as close as 0.25 mm can be easily 

 attained, the thickness of the upper electrode can be varied according to 

 the needs of the experiment (see p. 176), and fair corrections can be made 

 for the "wall effect" by the extrapolating procedure (Quimby and Focht, 

 1943). It should not be forgotten, however, that, as for the case of 

 particle radiation, the physical support of the specimen should be identical 

 to the lower electrode of the chamber. 



When the biological object is thick, so that t is of the order of the 

 relaxation length ^^ of the radiation, the dose will vary within the beam in 

 a complex manner, on account of photon scatter, and some energy will be 

 expended beyond the limits of the geometrical beam. The description 

 of the variation of the dose within large-scattering media has been 

 treated extensively (under the generic term of depth dosage) in the radio- 

 logical literature (Glasser, Quimby, et at., 1944; Quimby, MarinelH, and 

 Farrow, 1938; Loevinger, Wolf, and Minowitz, 1950; Quimby and 



" See section entitled Dose from Radioelements Deposited in Biological Systems. 



