read this error will be of the order of 1.0 percent; if 30 frames are read from a given film the resulting reading error will be about 0.2 percent. The determination of fireball diameter is thus subject to four probable errors, Each of these is about 0.2 percent for the AEC data, resulttng in 0.4 percent in the diameter, and 2.0 percent inthe yield. For the U. N. data three of the measurements are uncertain to 1,0 percent, and the fourth 0.2 percent; the result is 1.8 percent in the diameter and 9.0 percent in yield. The timing error is in each case 0.2 percent, resulting in 0.3 per‘cent in yield, and the air-density uncertainty causes 0.2 percent in the yield. Uncertainty in the scaling constant may amount to as much as 3.0. ‘percent, owing primarily to the variable effects of the mass of the device, Shielding, and barge. Combining all of these possible errors we find the sresultant uncertainty in total yield to be about 4 percent for the AEC meas- ‘urements and 10 percent for the Observer check, both within the prescribed 5-percent and 15-percent error limits, | It must be recalled that the AEC figures will be given to the Observers. The purpose of the Observer check is to verify the USAEC figures to a reasonable precision and not to furnish the accepted yield for the shot. 9. Cameras and Lenses The growth of the nuclear fireball is recorded, for later analysis, by means of high-speed ciné cameras. Although many suitable cameras are available commercially, we find the Eastman High-Speed Type III to be most suitable for our purposes, These cameras employ continuous, instead of intermittent, motion of the film. A glass prism, placed between lens and film, rotates at such a speed that the image is translated in the direction of the film motion; synchronism between the motions of the image and of the film is imperfect, but is more than adequate for most purposes, The housing around the prism serves as the shutter, blocking off the light when the image deflection-angle becomes large, and permitting light to pass for the next frame as it rotates further. | The Eastman camera, Figure 4, utilizes a 100-ft spool of 16-mm film. The prism is "two-sided", so that a complete prism revolution is required for every two frames. As a result, the duration of the exposure