Dan Ford's books
For print editions of Dan's books, go here      For the e-books, go here

HOME > JAPAN > INVASION

Waiting for the invasion

[During the U.S. occupation of Japan, demobilized officers were hired to write a series of campaign histories, which were translated and mimeographed as the Japanese Monograph Series. Following are my notes from Japanese Monograph No. 23, dealing with the Imperial Army air force in the defense of the home islands. The mixture of metric and "English" measurement appears in the original. A complete set of Japanese Monographs is on microfilm at the Library of Congress. -- Dan Ford]

An American invasion was expected "late summer or early fall" of 1945. Resources were to be husbanded for that. "In general, a counterattack will not be made against enemy fighter units attacking independently except when the situation is specially advantageous or when absolutely necessary."

On 1 Aug 1945 the Air General Army had 800 front-line fighters and bombers, plus 2,000 "special suicide attack planes." Another 1,000 planes were scheduled for conversion for suicide missions by end of October. Their disposition is detailed in the monograph. [Navy air forces in Japan were very much larger.]

Imports of gasoline ended May 1945. The army had 13 million gallons of avgas on 1 Aug, with 10.5 million gal reserved for the Ketsu-go suicide attack on the U.S. fleet. Half of trainers and 10 percent of transports converted to run on alcohol. Shokon oil from pine-tree roots was in production but no deliveries before war's end.

Used Ki-67 heavy bombers with radar for patrol. Ground radar not particularly effective: range 300 km, couldn't spot low-altitude a/c or those above 8,000 feet[!]. Communication by telephone land lines which had been destroyed by American bombing. As a result, radar was useful only to check information otherwise received. Most reliable way to track U.S. formations was by listening to American radio frequencies, one of "the few weaknesses which were exploited to good advantage." Could predict raids with 80-90 percent accuracy, 4-5 hours in advance.

Question? Comment? Newsletter? Send me an email. Blue skies! -- Dan Ford