562 BOND, CRONKITE, SONDHAUS, IMIRIE, ROBERTSON, AND BORG than one-half the total time for bilateral, and the depth-dose curve is thus above that for bilateral. Thus, the difference noted is seen to result from the inverse square effect. It is important, however, to note that, although the crossfire technique has taken into account to a degree the inverse square effect, it has not, of course, in any sense eliminated the effect. It has averaged the entrance and exit exposure doses, and thus has raised the depth-dose curve, somewhat as might result if inverse square were negligible. An identical superimposed curve is obtained if, with bilateral, the average of the entrance and exit doses is used as the ‘air dose,” instead of the entrance air dose with each half-exposure. If the midline air dose is used with bilateral exposure, the curve is essentially identical in shape to the crossfire curve, but is placed a short distance above it. Of importance later in considering the curve for fallout radiation, if the half-exposure curves for bilateral radiation are corrected for inverse square falloff before addition, the resulting curve, although placed at approximately the level of the crossfire curve, is considerablyflatter than the crossfire curve (70.5 % at the edges, 69.0% at the midline). Ring and 4x exposures. With ring geometry, the phantomis at the center of a concentric ring of fixed sources (any of the “bands” shownin [ig. 2). The phantom pnt Pd cs placed in the geometric center of the y-ray generator shown in Fig. 2 is exposed under conditions closely approaching a 4r geometry. The depth-dose pattern for both exposures is shown as curve b, Fig. 4B. Theyare essentially identical and are negligibly different from those obtained with the crossfire technique. These types of exposure can be considered to bear a similar relationship to crossfire exposure. as does multilateral or rotational exposure to the bilateral technique. Inverse square is taken into account to a degree, but is not corrected for or eliminated. Bomb, inttial y-radiation. The measured depth-dose curve in phantom material exposed to the initial y-radiation from a nuclear device is shownas curve c, Fig. 3, and as curve b, Fig. 4C. The phantom employed was a cylinder measuring 25 cm in diameter, and measurements were taken approximately 3 feet above the ground. It is apparent that, although the rate of falloff of dose in tissue is still appreciable in a thickness of tissue approximating man, the exit dose of approximately 55% is well above the value of approximately 20% for Co® y-radiation in the laboratory. This is consistent with theory (1/2, 73) and other observations (14). The linear absorption coefficient for bomb immediate y-radiation observed at distances of biological interest (quoted on page 97 of reference 14) can be converted to the mass absorption coefficient that will apply to the phantom material by correcting tor the small difference in electron density and for inverse square (no detectable falloff through the 26--m phantom). Application of the absorption coefficient thus derived yields a depth-dose curve essentially identical to that observed. .\ similar, more approximate result is obtained by using the good geometry coefficient for y-rays of several Mev and applying the appropriate build-up