The U.S. Sixth Army, which would invade and occupy Kyushu, estimated 124,935 U.S. battle casualties, including 25,000 dead, plus 269,000 non-battle casualties (disease, accident, etc.) for Kyushu alone. The JCS came up with an estimate that a 90-day campaign on Kyushu would cost 156-175,000 battle casualties, with 38,000 killed in action. By late July, the JCS was forecasting 500,000 casualties at the high end and 100,000 at the low end. In late July 1945, the War Department provided an estimate that the entire Downfall operations would cause between 1.7 to 4 million U.S. casualties, including 400-800,000 U.S. dead, and 5 to 10 million Japanese dead. (Given that the initial Downfall plan called for 1,792,700 troops to go ashore in Japan, this estimate is indeed most sobering, and suggests many more troops than planned would need to be fed into a meat grinder). Other estimates in the U.S. government indicated U.S. deaths at 500,000 to 1 million. Which of these and other estimates would be the most accurate has been hotly debated over the years (and are caught up in the debate about whether the atomic bomb should have been used), and I’m not going to solve it. But it is clear that the cost of invading Japan would have been staggering for both the U.S. and the Japanese.