Volume Five "i"

THE PACIFIC: MATTERHORN TO NAGASAKI JUNE 1944 TO AUGUST 1945

THE ARMY AIR FORCES In World War II

j PREPARED UNDER THE EDITORSHIP OF

WESLEY FRANK GRAVEN

Princeton University

JAMES LEA GATE

University of Chicago

New Imprint by the Office of Air Force History Washington, D,G., 1983

For sale by tbe Superintendent of Documents, U.S. GoTemment Printing Office Washinjrton, D.O. 20402

The University of Chicago Press, Chicago 37

Cambridge University Press, London, N.W. 1, England

Copyright 1953 by. The University of Chicago. All rights reserved. Published 1953. Composed and printed by Kingsport Press^ Inc.^ Kingsport^ Tennessee^ U.S.A.

Copyright registration renewed 1981

This work, first published by the University of Chicago Press, is reprinted in its entirety by the Office of Air Force History. With the exception of editing, the work k the product of the United States government. )

Library of Congress Cataloging in Publication Data Main entry under title :

The Army Air Forces in World War IL

Vol. 1 originally prepared by the OfEce of Air Force History; v. 2, by the Air Historical Group; and v. 3-7, by the USAF Historical Division.

Reprint. Originally published : Chicago: University of Chicago Press, 1948-1958.

Includes bibliographical references and indexes.

Contents: v. 1. Plans and early operations, January 1939 to August 1942— V. 2. Europe, torch to point- blank, August 1942 to December 1943— [etc.]— v. 7. Services around the world.

1. World War, 1939-1945— Aerial operations, American. 2. United States. Army Air Forces Histdry-^World War, 1939-1945. I. Craven, Wesley Frank, 1905- . II. Cate, James Lea, 1899- III. United States. Air Force. OfBce of Air Force History. IV, United States. Air Force. Air Historical Group. V. United States. USAF Historical Division. D790.A89 1983 940.54^4973 83-17288 ISBNO-912799-03-X (v. 1)

ii

FOREWORD to the New Imprint

IN March 1942, President Franklin D. Roosevelt wrote to the Director of the Bureau of the Budget ordering each war agency to prepare "an accurate and objective account" of that agency's war experience. Soon after, the Army Air Forces began hiring professional historians so that its history could, in the words of Brigadier General Laurence Kuter, "be recorded while it is hot and that personnel be selected and an agency set up for a clear historian's job without axe to grind or defense to prepare." An Historical Division was established in Headquarters Army Air Forces under Air Intelligence, in September 1942, and the modern Air Force historical program began.

With the end of the war, Headquarters approved a plan for writing and publishing a seven- volume history. In December 1945, Lieutenant General Ira C. Eaker, Deputy Commander of Army Air Forces, asked the Chancellor of the University of Chicago to "assume the responsibility for the publication" of the history, stressing that it must "meet the highest academic standards." Lieutenant Colonel Wesley Frank Craven of New York University and Major James Lea Cate of the University of Chicago, both of whom had been assigned to the historical program, were selected to be editors of the volumes. Between 1948 and 1958 seven were published. With publication of the last, the editors wrote that the Air Force had "fulfilled in letter and spirit" the promise of access to documents and complete freedom of historical interpre- tation. Like all history, The Army Air Forces in World War II reflects the era when it was conceived, researched, and written. The strategic bombing campaigns received the primary emphasis, not only because of a widely-shared belief in bombardment's con-

tribution to victory, but also because of its importance in establish- ing the United States Air Force as a military service independent of the Army. The huge investment of men and machines and the eflFectiveness of the combined Anglo-American bomber offensive against Germany had not been subjected to the critical scrutiny they have since received. Nor, given the personalities involved and the immediacy of the events, did the authors question some of the command arrangements. In the tactical area, to give another example, the authors did not doubt the effect of aerial interdiction on both the German withdrawal from Sicily and the allied land- ings at Anzio.

Editors Craven and Gate insisted that the volumes present the war through the eyes of the major commanders, and be based on information available to them as important decisions were made. At the time, secrecy still shrouded the Allied code-breaking effort. While the link between decoded message traffic and combat action occasionally emerges from these pages, the authors lacked the knowledge to portray adequately the intelligence aspects of many operations, such as the interdiction in 1943 of Axis supply lines to Tunisia and the systematic bombardment, beginning in 1944, of the German oil industry.

All historical works a generation old suffer such limitations. New information and altered perspective inevitably change the emphasis of an historical account. Some accounts in these volumes have been superseded by subsequent research and other portions will be superseded in the future. However, these books met the highest of contemporary professional standards of quality and comprehensiveness. They contain information and experience that are of great value to the Air Force today and to the public. Together they are the only comprehensive discussion of Army Air Forces activity in the largest air war this nation has ever waged. Until we summon the resources to take a fresh, comprehensive look at the Army Air Forces' experience in World War II, these seven volumes will continue to serve us as well for the next quarter century as they have for the last.

RICHARD H. KOHN Chief, Office of Air Force History

iv

FOREWORD

* * 5|C * * * * * * *

WITH the publication of this fifth volume of The Army Air Forces in World War II the narrative of AAF com- bat operations is completed. The plan of the series will be familiar to those readers w^ho have foUow^ed the story in earlier volumes; for others it may be helpful to place the present study in the / context of the whole series. Volume I carried the story of the AAF, both at home and abroad, through the first critical months of the war to the latter part of 1942, when it could be said that the Allied forces had seized the initiative in accordance with agreed-upon strategy. That strategy rested upon the assumption that there were in fact two wars, at least to the extent of permitting the war against the European Axis to be assigned a priority over that with Japan, and this assump- tion has been taken by the editors as warrant enough for a separate treatment of AAF operations in Europe and against Japan after the summer of 1942. In Volumes II and III the narrative of combat op- 42;ratiqps against the European Axis was carried forward from the beginning of Eighth Air Force bombing operations in August 1942 to the final collapse of Germany. In Volumes I and IV the fortunes of the AAF in the Pacific and CBI were followed from the initial at- tack on Pearl Harbor to the summer of 1944. Taking up the story at that point, the present study provides a narrative of combat opera- tions against Japan to the final victory in August 1945. The two re- maining volumes in the series will be devoted to the home front and to services, like that of the Air Transport Command, which do not readily fit into a discussion bound by theater limits.

At the close of Volume IV, MacArthur's forces had advanced along the northern coast of New Guinea to Sansapor and Admiral Nimitz' central Pacific forces had recently seized the Marianas, where engineers promptly undertook the development of airfields for use by the B-29's. A large part of the present volume, as would be expected, is devoted to the strategic air offensive against Japan, an

V

THE ARMY AIR FORCES IN WORLD WAR II

offensive opened by XX Bomber Command from Chinese bases on 15 June 1944 and continued with mounting fury after November by XXI Bomber Command from bases in the Marianas. But that offensive, like the Combined Bomber Offensive against Germany, was con- sidered officially as no more than an adjunct to other plans for the defeat of Japan, and it may be well to consider first the development of those other plans.

At the time of the launching of the B-29 offensive no final plan for the defeat of Japan had taken shape. Proponents of a strategy that would advance MacArthur's forces (mainly Army) northward from New Guinea by way of the Philippines toward Japan continued to press vigorously for a decision that would concentrate U.S. resources upon this line of attack; no less vigorous were the advocates of a strategy that would concentrate on a drive, under the leadership of Admiral Nimitz, for the establishment of air and sea bases on the China coast as a preliminary to the final assault on the home islands. By the summer of 1944 the debate was an old one and had been re- solved only to the extent of an agreement by the Joint Chiefs that for the time being there was some advantage in keeping Japanese forces under the pressure of a double attack. In a directive of 12 March 1944 Mac Arthur had been instructed to continue with operations necessary to support of an invasion by Nimitz of the Palaus on 1 5 September and to land with his own forces on Mindanao in the southern Philippines on 1 5 November. Depending upon subse- quent decisions, Nimitz would occupy Formosa on 15 February 1945 or MacArthur would land on Luzon in a move preliminary to a de- layed attack on Formosa. The Joint Chiefs again postponed a final decision when on 8 September 1944 they approved plans for the seizure of Leyte in the following December.

Meantime, plans had been laid by MacArthur for the capture of Morotai in the Moluccas as a stepping stone on the way to Mindanao and Leyte, the timing of the operation to coincide with Nimitz' in- vasion of the Palaus in order that a double advantage might be taken of available naval cover. Kenney's Far East Air Forces, which since 15 June had combined the Fifth and Thirteenth Air Forces, recipro- cated by collaborating with the Seventh Air Force, which until the summer of 1945 would continue to operate as a subordinate unit of Nimitz' central Pacific command, in pre-invasion bombardment pre- paratory to both landing operations. The landings were accomplished

vi

FOREWORD

Oil schedule at Morotai and Peleliu, and engineers followed hard upon the assault forces to make ready the airfields which gave to the islands their strategic significance.

Such a sequence long since had become a familiar feature of island- hopping operations in the Pacific, but the engineers on this occasion approached their tasks with an unusual sense of urgency. Admiral Halsey, commanding the U.S. Third Fleet in pre-invasion strikes, had picked up intelligence indicating that Leyte contained no Japanese forces. Moreover, the reaction to his attacks argued a general weak- ness of the enemy throughout the Philippines. On Halsey's initiative, therefore, it had been decided to cancel a projected occupation by Nimitz of Yap and to jump MacArthur's front forward in one leap from Morotai to Leyte, with a target date of 20 October. The de- cision was one of the major gambles of the war. Even with the rhost rapid develbpment of air facilities oh Morotai, Leyte would remain beyond the range of effective cover by Kenney's air forces, still based on New Guinea. The plan of the Leyte operation thus violated one of the cardinal principles of SWPA strategy: to keep each forward move within the reach of land-based air forces. But Halsey's estimate of the enemy's weakness in the Philippines was not out of line with SWPA assumptions that the Japanese air forces were in a state of near- collapse, and powerful uilits of the Navy's carrier forces promised protection during the interval before Itenney could get his air gar- risons forward. The gamble seemed to be one worth taking.

And so it was, as events proved. Yet, the risk was also proved to have been greater even than that anticipated. The report that there were no Japanese forces on Leyte was wrong; actually the veteran i6th Division was stationed there. Other intelligence regarding Leyte, intelligence affecting plans for airfield development and the build-up of an air garrison, turned out to be misleading. The enemy, correctly anticipating the general plan of U.S. leaders, was engaged in strength- ening his position throughout the Philippines. It was true enough that Japanese air strength was on the point of collapse, as the desperate tactics of kamikaze attacks soon made abundantly clear, but remain- ing resources could be and were concentrated on the Philippines to an extent that dangerously belied Allied estimates of the situation. A plan to concentrate Japanese naval forces for all-out resistance to an Allied invasion of the Philippines rested upon the hope that U.S. carriers might be decoyed away from the beachhead to permit its

vii

THE ARMY AIR FORCES IN WORLD WAR II

destruction by the main force. And the American naval forces which carried the responsibility for protecting the beachhead also carried orders, thoroughly consistent with naval doctrine, that an *'oppor- tunity for destruction of major portions of the enemy fleet" would become "the primary task."

The landings on Leyte were easily made. A now extended experi- ence with pre-invasion bombardment by Allied naval and air forces had persuaded the enemy to adopt the tactic of withdrawing from the beaches for concentration in the interior, and Allied air operations for isolation of the battle area had been effective enough to limit inter- ference by enemy air to sporadic though vicious attacks. During the weeks preceding the invasion, FEAF planes ranged widely over the area south of Leyte and, beginning ten days in advance of the land- ing, Halsey's Task Force 38 once more gave an impressive demonstra- tion of the carrier's power in destructive sweeps of the Ryukyus, Formosa, and Luzon. Despite the sweeps of Task Force 38, assisted by 302 B-29 sorties against a few selected air installations on Formosa, the enemy was able to begin moving air reinforcements into Formosa and Luzon almost as the carriers withdrew. And when the naval en- gagements with the Japanese fleet on 24 and 25 October drew off the protecting forces at Leyte, enemy air units were in position to punish the beachhead severely on the afternoon of the 24th and to follow through the next morning with no less than sixteen attacks upon the airfield seized by U.S. assault troops on the day of their first landing. The courage and daring of U.S. fleet units, coupled with blunders by the enemy, saved the beachhead from the intended assault by the main body of the Japanese fleet, but escort carriers in Leyte waters had spent themselves in desperate fleet actions, and Halsey's fast carriers, which had been decoyed far to the north, now had to be withdrawn for replenishment. The last of the fast carrier groups departed on the 29th, almost a week before FEAF planes were scheduled to take over responsibility for air defense of the beachhead.

Kenney reacted promptly to emergency demands for help. Though recently captured Morotai, nearest of his bases, as yet possessed facili- ties hardly equal to the requirements of a single bombardment squad- ron, he crowded substantial reinforcements onto the island. Attempts to attack enemy fleet units completely miscarried, but on Leyte ground crews which had been sent ahead of their planes labored night and day (and under repeated air attacks) with the engineers to lay the

viii

FOREWORD

Steel matting that permitted a force of thirty-four P-38's to move in as the initial air garrison on 27 October. The Navy having indicated its inability to fulfill its original mission of air defense, the job was promptly given to Kenney. Anxious days remained. Jammed together on a single strip with no provision for dispersal yet possible, the P-38's constituted an inviting target for enemy attack. Between zy October and 3 1 December the enemy sent more than a thousand sorties against Leyte. The American defense force, which by December included Marine air units, proved itself superior to the enemy, and losses in combat were relatively small. But most planes continued to be based on Tacloban, the original field, where damaged aircraft were pushed into the sea to make room for reinforcements. All Philippine targets had been cleared for FEAF attack on 27 October, with instructions to the Navy to coordinate with FEAF before attacking. With both heavy bombardment groups of the Thirteenth Air Force brought forward to Morotai by mid-November, FEAF attacks on Philippine airfields began to count. Halsey's carriers were back by 5 November for heavy blows, and from its base in the Palaus the Seventh Air Force's 494th Group added weight to the attack. But the enemy had developed new skills in dispersal, and only with mid-November could it be said that U.S. forces asserted a telling superiority in the air. Meanwhile, the enemy had reinforced his ground troops on Leyte by 22,000 men within the first two weeks after the U.S. landing, and other thousands would follow, though at times without getting their equipment ashore. The evidence indicates that some 19,000 enemy troops were on Leyte at the time of our landing. At the close of land operations on Leyte in May 1945, totals showed some 56,000 enemy troops killed or captured.

The entire Leyte operation is extremely complex and at many points debatable. For so long as men study military history, the opera- tion will retain a special fascination of its own. The editors of this volume have gone into some detail here, not so much because of a desire to enter into a debate as because of the belief that the experi- ence at Leyte, in reverse so to speak, lends a special emphasis to the principles on which air operations had been successfully coordinated with the advance of ground and naval forces in the southwest Pacific. Those principles were grounded upon the assumption that air forces must first be in a position to assert and maintain superiority in the area of battle. It had been repeatedly recognized, as at Hollandia in

ix

THE ARMY AIR FORCES IN WORLD WAR II

New Guinea, that carrier-based air power could extend the reach of amphibious operations and safely so, provided land-based air power was in a position to take over promptly the primary responsibility. The advantage belonging to land-based air power obviously was its staying power: the capacity to stay there and fight it out for what- ever term might be necessary to maintain air superiority and to do this without reference to any other competing obligation.

Fortunately, the U.S. command, given time, had more than enough resources to make good the gamble at Leyte. Fortunately, top, leaders showed a continued willingness to gamble on the declining power of the enemy by adhering to a stepped-up timetable of operations. The Joint Chiefs in October finally had resolved the question of an inter- mediate strategic objective by agreeing that MacArthur should ad- vance by way of Mindoro to Luzon on 20 December and stand ready to support Nimitz in a later occupation of Okinawa, which at Nimitz' suggestion had been chosen in lieu of Formosa. Mindoro, which was to serve as an advanced air base for cover of the landings at Luzon, was scheduled for 5 December. Disappointing delays in the develop- ment of airfields on Leyte threatened the plan, for without a greatly increased capacity there Kenney would be unable to cover Luzon for the Mindoro operation. Happily, a rescheduling of Mindoro for 15 December and postponement of Luzon to 9 January 1945 made it possible for Halsey's carriers to cover Luzon while FEAF concen- trated on the southern Philippines and protected the convoy en route to Mindoro. The convoys had a rough time of it, even though Kenney had stripped Leyte of air defense to provide a cover; but the schedule was kept and, with the protection of Mindoro-based planes and the assistance once more of the carriers, MacArthur reached the Lingayen beaches on time.

In the rapid development of the Philippine campaign, during which U.S. forces not only overran Luzon but in a series of brilliantly exe- cuted operations retook the whole of the Philippine archipelago by the summer of 1945, AAF forces demonstrated an extraordinary ver- satility both in the fulfillment of primary responsibilities and in the support of other services. As expanding facilities on Morotai and Mindoro and the capture of airfields in the Philippines made possible the forward staging of FEAF strength, Kenney 's ''boys" gave re- peated demonstfation of the full meaning of air supremacy. If the relative ease with which they asserted and maintained that supremacy

X

FOREWORD

bespoke the advantage gained from an earlier victory over the enemy air forces in the battles of New Guinea and the Solomons, the fact takes nothing from the evidence of skills which had been well de- veloped. Only in the direct support of ground troops in a land cam- paign of the magnitude developed on Luzon did AAF crews face a task for which they had limited experience, and even here their sup- port more than met the test of battle.

In the Philippines, as earlier in New Guinea, AAF planes struck ahead of land and amphibious forces to clear the way, protected con- voys and other troop movements, delivered by air emergency sup- plies and paratroopers, kept enemy air beaten down on fields far and near, joined with naval forces to deny the enemy opportunity to re- inforce his positions, maintained daily patterns of search covering thousands of miles for the intelligence of all services, and withal kept the flexibility necessary to meet emergency demands. In addition to commitments to the fighting in the Philippines, FEAF shared in the increasingly successful effort by U.S. submarines to cut the sea com- munications joining Japan to the southern parts of its Empire, found the reserve strength to assist the Australians in the reconquest of Borneo, and assumed responsibility for the neutralization of Formosa, a key enemy base that acquired special significance with the U.S. landing on Okinawa in April 1945. When kamikaze attacks seriously endangered U.S. naval forces supporting the Okinawa operation, some disagreement developed between naval and air leaders as to the source of these attacks. Having reason to believe that its pre-invasion bombardment of Formosa had reduced enemy air there to a state of impotency, FEAF argued that the attacks came from Kyushu, as postwar evidence indicates most of them did; the Navy suspected that most of them came from Formosa, as indeed perhaps 20 per cent of the attacks did. Though loath to waste any effort needed elsewhere, FEAF repeatedly stepped up its continuing operations against For- mosa air installations in response to urgent appeals from the Navy. It was difficult, however, to cope with a well conceived program of dispersal that was implemented on a much larger scale and with far more determination than was at any time suspected by FEAF in- telligence. And even had the intelligence been more accurate, it is doubtful that any of the conventional forms of air attack could have accomplished more than some reduction of the enemy effort. In retrospect, perhaps the kamikaze form of attack will serve chiefly to

xi

THE ARMY AIR FORCES IN WORLD WAR II

remind us that air supremacy can never be conceived of as an abso- lute.

When the war ended, AAF units flying from the hard-won bases of Okinawa had already brought Kyushu, southernmost of the ene- my's home islands, under an attack preparatory to a scheduled am- phibious landing in the following November. Earlier assumptions that the establishment of some lodgment on the Chinese mainland would be a necessary preliminary to the final assault on Japan had been abandoned. Difficulties arising from the question of command, which in the Pacific often had complicated the problem of agreement on strategy, had been resolved by a decision that MacArthur would command all Army, and Nimitz all naval, forces, with dependence upon the principle of cooperation in joint actions. FEAF, enlarged by the addition of the Seventh Air Force redeployed to Okinawa, continued to serve as MacArthur's air command. A new air com- mand, the United States Army Strategic Air Forces in the Pacific (USASTAF) , would control the Twentieth Air Force and the Eighth Air Force when redeployed from ETO to Okinawa.

The decision to mount the invasion of Japan from island bases with- out benefit of a lodgment on the east China coast meant that the war would end, as it had been waged throughout, with no real connection between the Pacific theaters and China-Burma-India. In the latter theater problems of strategy and command had been even more diffi- cult of solution than in the Pacific, being rooted in the divergent in- terests of the three Allied nations and made bitter by the personal animosities of some leaders. China-Burma-India, lying at the extreme end of the supply line from America, was accorded a very low pri- ority, and geographical factors within the theater made it difficult to use the bulk of the resources in combat: most of the tonnage available was spent merely in getting munitions to the various fronts. There were few U.S. ground forces in CBI, most of the troops being air or service forces whose mission was to see that a line of communication was preserved whereby China could be kept in the war.

The Tenth Air Force, having earlier protected the southern end of the Assam-Kunming air route that was long the only connection between China and U.S. supply bases in India, was committed in mid- 1944 to a campaign in northern Burma whose dual objective was to open a trace for the Ledo Road into China and to secure bases for a

xii

FOREWORD

more economical air route over the Hump. By that time Allied air forces, combined in the Eastern Air Command, had control of the skies over Burma; they helped isolate the strategic tov^^n of Myitkyina, supplied by airlift the ground forces conducting the siege, and ren- dered close support in the protracted battle that dragged on from May to August. After the fall of Myitkyina, the Tenth Air Force participated in the drive southward to Rangoon, a campaign that would seem to have borne little relation to the primary American mission. In both air support and air supply the Tenth showed skill and flexibility, but these operations absorbed resources that might have accomplished more significant results in China. After the Burma campaign EAC was dissolved in belated recognition of differing in- terests of the Americans and British, and at the end of the war the Tenth was moving into China to unite with the Fourteenth Air Force.

That force, ably commanded by Maj. Gen. Claire E. Chennault, had developed tactics so effective that its planes had been able to sup- port Chinese ground forces and strike at shipping through advanced bases in east China while giving protection to the northern end of the Hump route. Chennault believed that if his force and the airlift upon which it depended could be built up, air power could play a decisive role in ejecting the Japanese from China. The promised build-up came too slowly. In the spring of 1944 the Japanese started a series of drives which gave them a land line of communication from north China to French Indo-China, a real need in view of the insecurity of their sea routes, and the drives in time isolated, then overran, the east- ern airfields which had been the key to much of Chennault's success. In the emergency, a larger share of Hump tonnage was allocated to the Fourteenth and totals delivered at Kunming by ATC grew each month, dwarfing the tiny trickle of supplies that came over the Ledo Road. Chennault received too some additional combat units, but the time lag between allocation of resources and availability at the front was fatal. Different views of strategy and personal disagreements be- tween Chennault and Chiang Kai-shek on the one side and Lt. Gen. Joseph W. Stilwell, the theater commander, on the other resulted in the relief of Stilwell and the division of CBI into two theaters, India- Burma and China, with Lt. Gen. Albert C, Wedemeyer commanding the latter. Heroic efforts by air, including mass movements of Chinese ground forces by plane, prevented the Japanese from overrunning all

xiii

THE ARMY AIR FORCES IN WORLD WAR II

China. In the last months of the war the combined Fourteenth and Tenth Air Forces were preparing for a final offensive, but the sur- render came before this could be developed.

The command system in CBI and logistical problems as well were made more complicated by the deployment in that theater of XX Bomber Command, an organization equipped with B-29 bombers and dedicated to a doctrine of strategic bombardment. The plane, an un- tried weapon rated as a very heavy bomber, had been developed dur- ing the expansion of the Air Corps that began in 1939, and its combat readiness in the spring of 1944 had been made possible only by the Air Staff's willingness to gamble on short-cuts in testing and procurement. The bomber command, which resembled in many respects an air force rather than a command, had also been put together in a hurry, and the mission in CBI was conceived both as a shakedown for plane and organization and as an attack on Japanese industry. Early plans had contemplated using the B-29 against Germany, but by the sum- mer of 1943 thoughts had turned to its employment against Japan. The prospect that some time would elapse before appropriate bases in the Pacific could be seized plus the desire to bolster the flagging Chinese resistance to Japan, a need in which President Roosevelt had an active interest, led to a decision to base the first B-29 units in CBI. The plan looked forward also to VHB operations from the Marianas, where U.S. Marines landed on the same day that XX Bomber Com- mand flew its first mission against Japan.

To insure flexible employment of a plane whose range might carry it beyond existing theater limits, the JCS established the Twentieth Air Force under their own control with Arnold as **executive agent." Theater commanders in whose areas B-29 ^^^^^ operated would be charged with logistical and administrative responsibilities, but opera- tional control would remain in the Washington headquarters. This system of divided responsibilities found its severest test in CBI where the command system was already confused and where the dependence on air transport led to fierce competition for all supplies laid down in China.

Operational plans (MATTERHORN) for XX Bomber Command involved the use of permanent bases at Kharagpur near Calcutta and of staging fields near Chengtu in China, within B-29 radius of Kyushu and Manchuria but not of Honshu. Supplies for the missions were to be carried forward to Chengtu by the B-29's and by transport planes

xiv

FOREWORD

assigned to the command. Delays in the overseas movement of the B-29's and in airfield construction held up combat operations, the first regular mission being sent against Yawata on 15 June.

The earliest target directives gave precedence to the steel industry, to be attacked through bombing coke ovens. This target system was basic to the whole Japanese war effort and it had the tactical advantage of lying within range of the Chengtu bases. Little damage was done in Japan proper, but a few missions against Manchurian objectives were more effective than was then realized. From the beginning, operations were strictly limited by the difficulty of hauling supplies, especially fuel and bombs, to the forward bases. It was impossible for XX Bomber Command to support a sustained bombardment program by its own transport efforts, and the Japanese offensive in east China which began just before the B-29 missions prohibited any levy on normal theater resources, When the B-29's were assigned a secondary mission of indirect support of Pacific operations, logistical aid was furnished in the form of additional transport planes which were first operated by the command, then turned over, to ATC in return for a flat guarantee of tonnage hauled to China.

Because support of Pacific operations was designed to prevent the enemy from reinforcing his air garrison during the invasion of the Philippines, XX Bomber Command shifted its attention to aircraft factories, repair shops, and staging bases in Formosa, and factories in Kyushu and Manchuria. This shift from steel, considered a long-term objective, to aircraft installations reflected recent decisions to speed up the war against Japan. Attacks against the newly designated tar- gets, begun in October, were moderately successful, but a new Japa- nese drive lent urgency to the need for additional logistical support for ground and tactical air forces in China; consequently, at the re- quest of General Wedemeyer, the command abandoned its Chengtu bases in mid- January 1945.

Earlier, the B-29's had run a number of training missions in south- east Asia and one strike, from a staging field that had been built in Ceylon, against the great oil refinery at Palembang in Sumatra; now the giant bombers continued with attacks against Burma, Thailand, Indd-China, and Malaya. Strategic targets, as defined by the Twen- tieth Air Force, were lacking, and though some important damage was done to the docks at Singapore, operations had taken on an air of anticliinax long before the last mission was staged on 30 March, At

XV

THE ARMY AIR FORCES IN WORLD WAR II

that time the command was in the midst of the last and most sweeping of a series of reorganizations: the 58th Bombardment Wing (VH), its only combat unit, was sent to Tinian where it became part of XXI Bomber Command, while the headquarters organization went to Oki- nawa to be absorbed by the Eighth Air Force.

Measured by its effects on the enemy's ability to wage war, the MATTERHORN venture was not a success. For want of a better base area it had been committed to a theater where it faced a fantasti- cally difficult supply problem. Something of the difficulty had been realized in advance, but the AAF's fondness for the concept of a self- sufficient air task force had perhaps lent more optimism to the plan than it deserved. Certainly the desire to improve Chinese morale was a powerful argument, and here there may have been some success, though it would be difficult to prove. Powerful also was the desire of AAF Headquarters to use the B-29 for its intended purpose, very long-range attacks against the Japanese home islands, and in justice to that view it must be noted that the planners from the beginning expected to move the force to island bases when they were available, just as was done. As an experiment with a new and complex weapon, MATTERHORN served its purpose well: the plane was proved, not without many a trouble, under severest field conditions; tactics were modified and the organization of tactical units streamlined. The lessons learned were of great value to XXI Bomber Command, but the necessary shakedown might have been accomplished at less expense elsewhere, perhaps in the southwest Pacific. At any rate, the editors find no difficulty in agreeing with USSBS that logistical support af- forded to XX Bomber Command in China would have produced more immediate results if allocated to the Fourteenth Air Force, though it seems dubious that the alternate policy would have made for an earlier victory over Japan.

In his remarkable fictional account of a future American-Japanese war, published in 1925,* Hector Bywater referred to a news dispatch describing

American preparations to recover Guam by a sudden attack in overwhelming strength, this being but the first move in a great offensive campaign which was to be carried on with the utmost vigour until the Philippines were again in American hands. Further, it was hinted that the war would then be carried to the coasts of Japan proper, and allusions were made to the gigantic fleet of air-

* Hector Bywater, The Great Pacific War: a History of the American Japanese Campaign of i $31-1^33 (2d ed.; Boston, 1932), p. 244.

xvi

FOREWORD

craft which was building for the express purpose of laying waste to Tokyo and other great Japanese cities when the Americans had secured a base within striking distance.

Written two decades in advance, this proved to be an uncannily shrewd forecast of plans for the real war as they developed from the spring of 1944. First Saipan, then Tinian and Guam, were seized by Nimitz' forces for the primary purpose of serving as bases for VLR bombers, and while the Philippines were being secured, airfields were built in the Marianas and the bombardment of Japan was begun. Base development in the Marianas was delayed by the prolonged resistance of the Japanese garrisons, by competition for priorities with the Navy, and by fluctuations in deployment plans* However, minimum facili- ties were available to accommodate the 73d Bombardment Wing (VH) when its B-29's began to arrive at Isley Field on Saipan in October, and to receive each of the succeeding wings— the 313th (Tinian), 314th (Guam), 58th (Tinian), and 315th (Guam). The schedule was met only by the unprecedented device of basing each wing on a single field, all served by a depot field at Guam, which was also the site of the several headquarters connected with the B-29 proj- ect—XXI Bomber Command, AAFPOA, and after July 1945 the Twentieth Air Force and USASTAF.

Much of the credit for securing adequate priorities for B-29 build- ing programs that frequently ran counter to Navy demands in a Navy theater is due Lt. Gen. Millard F. Harmon, who became commander of AAFPOA upon its activation on i August 1944. That headquarters was established primarily to centralize, under Nimitz, logistical and administrative responsibility for all AAF forces in the central P^icific. The maintenance and repair system for B-29's in the Marianas de- veloped great efficiency, while supply problems never affected opera- tions as seriously as they had in the CBI: the one major crisis was caused by a threatened shortage of incendiary bombs that actually failed to materialize. Harmon, as commander of Task Force 93, had operational control of all land-based planes in the theater. Navy and Marine as well as Seventh Air Force units reinforced by VLR fighter groups. As deputy commander of the Twentieth Air Force he was responsible for coordinating B-29 operations with other theater ac- tivities, and he himself was inclined to interpret that duty to mean virtual control of all B-29 operations. This interpretation Arnold's office refused to accept, maintaining its direct control over the com-

xvii

THE ARMY AIR FORCES IN WORLD WAR II

manding general of XXI Bomber Command, to whom was accorded a considerable latitude in the fulfillment of directives. In July 1945, as a part of the general reorganization of U.S. forces in preparation for the invasion of Japan, a new headquarters, United States Army- Strategic Air Forces, was established at Guam under Gen. Carl Spaatz, its constituent air forces being the Twentieth (formerly XXI Bomber Command) and the Eighth, now converting to a VHB organization in the Ryukyus,

The B-29 assault began on 24 November 1944 with a strike against Nakajima's Musashino aircraft plant at Tokyo, a target chosen ac- cording to current directives which gave precedence to aircraft en- gine and assembly plants in that order. For the next three and a half months most of the missions were directed against such targets, with Musashino and the even more important Mitsubishi complex at Na- goya bearing the brunt of the attacks. High-level precision tactics were used, but with cloudy weather generally prevailing bombing ac- curacy was not up to expectations; damage was in most cases negligi- ble to moderate, but the threat of more effective attacks forced the Japanese into a badly planned dispersal program which materially re- duced the output of engines and planes. Although in this period, as throughout the rest of the war, weather constituted the most serious obstacle to successful operations, some of the difficulties were those commonly associated with the breaking-in of a new air force under arduous combat conditions; it was a tribute to the leadership of first Brig. Gen. Haywood S. Hansell, Jr., then Maj; Gen. Curtis E. LeMay, that the period of adjustment was so brief.

Losses were relatively heavy, both those inflicted by recently rein- forced defenses in Japan and the operational losses incident to the long overwater flight to Japan and return. The Japanese were also able to destroy some B-29's on the ground at Saipan by staging down through Iwo Jima in small raids that were annoying if not actually dangerous to the strategic campaign. Iwo Jima and its neighboring is- lands of the Nampo Shoto had been under attack since August by AAFPOA B-24's as a part of a general program of interdiction, but neither these attacks nor those occasionally delivered by B-29's and surface ships were sufficient to keep the air strips out of use. Iwo Jima, directly along the route to Honshu, was also a menace to B-29's in their missions northward, but in American hands the island could be developed into an emergency landing place, an advanced staging area,

xviii

FOREWORD

a base for VLR escort fighters, and an air-sea rescue station. These were the motives that led to the seizure of Iwo in a bitter struggle that began on 19 February and was finished only on 26 March. Pre- liminary bombardment by aircraft and surface ships failed to knock out the island's underground strongpoints, and the skilful and fanati- cal resistance of the enemy took heavy toll of the Marine invaders. The unexpectedly long struggle delayed the development of airfields, though one Japanese strip was rapidly overrun and rehabilitated for use of AAFPOA's fighters, which flew in to lend ground support to the Marines. Base development, still unfinished at the end of the war, turned the island into a complex of fighter and bomber strips. The fighters were used as escorts less frequently than had been expected, since waning enemy strength and a turn to night missions cut B-29 losses, but the fighters helped police the other Bonin Islands and made offensive sweeps over Japan. The use of Iwo as a staging base was less frequent than had been anticipated, also. As an emergency land- ing field, however, the island fully lived up to expectations; about 2,400 B-29's made unscheduled landings there and the crews saved thereby, and in the improved air-sea rescue service made possible by possession of Iwo, perhaps balanced the number of casualties suflFered during its capture.

On 9 March XXI Bomber Command began a series of incendiary attacks against urban areas that profoundly changed the nature of the strategic bombardment campaign. In spite of a general bias in favor of precision techniques, Twentieth Air Force headquarters had from the first been interested in the possibilities of incendiary attacks against the crowded and inflammable cities of Japan, and a few staflf members in Washington and in the field believed that such raids, con- ducted at night, would be far more destructive than conventional precision tactics. Early test raids were inconclusive (though a daylight incendiary raid on Hankow in China by Chengtu-based B-29's was highly successful), but under directives from Washington other at- tempts were made early in 1945 which aflforded more positive evi- dence. The tactics finally adopted by LeMay involved low-level night attacks with a heavy concentration of incendiaries of mixed types. The low approach and the partial stripping of defensive armament allowed a great increase in bomb load, but those measures were con- sidered by some as adding gravely to the danger from Japanese de- fenses. Fortunately the new tactics did not result in heavy losses, and

XIX

THE ARMY AIR FORCES IN WORLD WAR II

offensively they proved extraordinarily successful. The first attack, against Tokyo, burned out 15.8 square miles of the city, killed 83,793 people, and injured 40,918, being perhaps the most scathing air attack of the whole war. In rapid succession Nagoya, Osaka, Kobe, and