changes in the media as interfaces and cavities. Here again each shot must be studied individ- ually. This factor has more effect on crater depth and profile than radius according to data from a high-explosive test, Reference 8. The above analysis has been for the determination of crater radius. Crater depth, it is felt, is a function of the same variables though numerical values of F, through F, maybe different for crater radius and depth. The problem now becomes oneof reconsidering the individual shots at the EPG and by assuming adjustment factors, using the above discussion as a guide, making a better correlation of the crater data. Ivy Mike was a 10.5 Mt device fired at a height of burst of 35 feet. Crater measurements show a radius of 2,810 feet and a depth of 120 feet. Crater pictures show an absence of anylip; a condition which indicates full washing. In using a factoring system to reduce the craterto NTS conditions, the radius should be reduced by a factor of two to account for saturation and washing but increased by a factor of 1/0.94 to compensate for height of burst. The adjusted radius would be approximately 1,500 feet. Castle Shot 1 was a 14.5 Mt device fired at a height of burst of 15.5 feet. Crater measurements showed a radius of 3,000 feet and a depth of 240 feet. This shot was also considered to be fully washed. A factor of two is, therefore, used for saturation and washing effects. The HOBfactor in this case would be 1/0.98 and the adjusted radius would be approximately 1,530 feet. Castle Shot 3 was a 110 kt device with a 13.6 foot height of burst. The portion of the island in which the crater was formed had a steep slope into the lagoon. The crater radius on the island side ranged from 380 to 410 feet. The radius on the lagoon side was in excess of 600 feet and indicated a possible venting of cratering energy in this direction. An average radius of . 460 feet does not seem improbable. The crater had a broken and irregular lip with an average height of 10 feet on the island side indicating that the crater was not completely washed from the wave action generated by the explosion. Since it is assumed that complete washing did not take place, a washing and saturation factor of only 1.8 instead of 2 should be used. The HOB factor in this case would be 1/0.84, The adjusted radius would then be approximately 300 feet. Shot Lacrosse was a 39.5 kt device with a height of burst of 17 feet. The radius was 202 feet. This crater was completely land locked and was not considered washed. The soil in this area consisted of some cemented sand and coral. Since it was more cohesive than NTS soil, it should have exhibited greater compressive strength. Therefore, a soil strength factor of 1/0.9 is used. Since the soil was fully saturated, at high tide the water covered ground zero, a saturation factor of 1.5 is used. The HOB factor is 1/0.7 and the adjusted radius would be approximately 215 feet. If a saturation factor of 1.4 instead of 1.5 is used as suggested in WI —1307 the adjusted radius is 230 feet. Both values are plotted in Figure 3. The Shot Seminole device is discussed later with reference to the Shot Koa crater. The Seminole radius due to its protected position and lip condition should be increased by about 20 percent in order to be fully washed. A saturation-washing factor of two can then be used. Due to the water tamping and near-surface placement this shot was considered to be a surface burst and an HOB factor was not needed. An energy containmentfactor of 30 percent for the water tank emplacement was found by comparing the adjusted radius of 198 feet to that from a surface burst, or 152 feet. The adjusted radius for the 13.5 kt Seminole device would be 152 feet. The straight line in Figure 3 was drawn using the intercept given in TM 23-200 of 64 feet and one third slope. The plotted points represent EPG data adjusted to surface detonations in dry soil as given above. Table 1 gives adjusted and unadjusted crater radii for EPG shots including Shots Cactus and Koa. SHOT PARTICIPATION Shots Koa, Cactus, and Fig, the only land-surface shots of Operation Hardtack, constituted the primary participation of this project. Limited participation was carried out on barge shots Linden, Oak, Yellowwood, Butternut, and Holly. 13