CHAPTER IV
The Japanese

Just as the Allies were determined to advance against Rabaul, the Japanese were determined to hold it, and, indeed, to continue the advance that had been checked at Guadalcanal and Buna. The importance imparted to Rabaul by its airfield sites and harbor, as well as by its strategic location, had long been recognized by the Japanese. Imperial General Headquarters' instructions of November 1941 directed the capture of Rabaul at the earliest opportunity after the fall of Guam.1 Rabaul supported the offensives against the Allied lines of communication, and defensively was a bastion which would help defend the Caroline Islands, the Netherlands Indies, and the Philippines against attack from the south. It was one of the most important bases in the semicircular string of island fortresses that stretched from Burma through the Indies and the Bismarck Archipelago to the Marshall Islands, thence north and northwest to the Kuriles.2

Japanese Command and Strategy

By late 1942 Rabaul had been developed into the major air and naval base in the Japanese Southeast Area, and was the site of the highest headquarters in that area. Although smaller than most Allied areas in the Pacific, the Southeast Area was huge. Its western boundary, as set on 2 April 1943, was longitude 140° east.3 The northern boundary ran from 140° east just north of the Equator to a line drawn between Kapingamarangi in the Greenwich Islands to Nauru, thence southeast between the Fijis and Samoa. It thus embraced parts of both the South and Southwest Pacific Areas.

Unlike the Allied areas, the Southeast Area did not possess a unified command. The highest Army and Navy headquarters

--32--


Organization of Japanese Forces, Southeast Area, June 1943

--33--

co-operated closely with one another, but were responsible to different higher authorities. (Chart 3) In charge of Army operations in eastern New Guinea, the Bismarck Archipelago, and the Solomons was General Hitoshi Imamura, commanding the 8th Area Army with headquarters at Rabaul. Imamura was responsible to the Army Section of Imperial General Headquarters. The naval command was the Southeast Area Fleet or the Southeastern Fleet led by Vice Adm. Jinichi Kusaka. His immediate superior was the Commander in Chief of the Combined Fleet but on several occasions he seems to have dealt directly with Tokyo.

By the time the Guadalcanal and Papuan campaigns ended, the 8th Area Army included two field armies and one air division. The 17th Army operated in the Solomons; the 18th Army was to be responsible for the campaigns in eastern New Guinea. The 6th Air Division, with headquarters at Rabaul, generally operated in New Guinea under the tactical direction of the 18th Army.

Under the Southeastern Fleet were the land-based 11th Air Fleet, which operated principally in the Solomons, and the 8th Fleet with bases at Rabaul and in the Shortlands-Buin area. The 8th Fleet, whose strength and composition varied considerably, usually included cruisers, destroyers, submarines, transports, and naval base forces. An administrative rather than a battle fleet, its primary duties were patrol and escort. Large-scale combat operations were the mission of either the 3d or the Combined Fleet, both then at Truk.4

Both the 8th Area Army and the Southeastern Fleet had been set up in late 1942 when the Japanese, making their major offensive effort in the Solomons and still planning to drive the Americans from Guadalcanal, realized that they had to commit large forces to attain success. But Imperial General Headquarters then revised its strategy and decided to abandon Guadalcanal, evacuate the survivors, and withdraw to strong positions in front of Rabaul.

Under the revised strategy, Imperial Headquarters decided to shift its emphasis from the Solomons to New Guinea. A policy of "active defense" would be pursued in the Solomons in order to reinforce New Guinea and pursue an "aggressive offensive" there.5 Lae, Salamaua, Wewak, and Madang on New Guinea's north coast were specifically mentioned as bases to be held. Imamura therefore ordered Lt. Gen. Hatazo Adachi, commander of the 18th Army, to strengthen Lae, Salamaua, Wewak, and Madang. These points were valuable as harbors, airfield sites, or both. Lae and Salamaua were of great importance as their possessor could dominate Dampier and Vitiaz Straits and thus block any attempt to advance along the New Guinea coast to the Philippines or any other place in the Greater East Asia Co-Prosperity Sphere. The Japanese were determined not to yield "an operational route for the proclaimed enemy Philippines invasion."6 These bases would also be necessary to the Japanese if they were

--34--


VICE ADM. JINICHI KUSAKA


GENERAL HITOSHI IMAMURA

to realize their hopes of capturing Port Moresby.

Thus the Japanese survivors of Buna were ordered to Salamaua, and on Imamura's orders Adachi directed more elements of his army to move from Rabaul to the New Guinea bases. The 20th Division began moving to Wewak; the 41st sent elements to Madang, and part of the 51st Division was sent to Lae and Salamaua. The fixing of the west boundary of the Southeast Area on 7 January at the Dutch border apparently gladdened Adachi's heart. After being limited to the Buna region, "having suddenly obtained freedom of the operational area, it gave them [the 18th Army] bright and desirous hopes. . . ."7

At the same time, detachments of Field Marshal Count Hisaichi Terauchi's Netherlands Indies-based Southern Army were occupying areas along New Guinea's north coast from the Vogelkop Peninsula to Hollandia.

Imperial Headquarters' orders for the Solomons required the 8th Area Army, in co-operation with the Southeastern Fleet, to hold the central and northern Solomons. Army and Navy authorities at Rabaul disagreed over exactly where the forward defense lines should be located. The Army favored the Bougainville area, holding that it would be too difficult to supply the islands farther south. The Navy insisted on New Georgia and Santa Isabel as outposts for Bougainville. Each service went its own way. The Army assumed responsibility for the defense of the northern Solomons. The Navy took over land defense of the central Solomons. Imamura gave to Lt. Gen.

--35--


LT. GEN. HATAZO ADACHI

Haruyoshi Hyakutake's 17th Army, then consisting chiefly of the 6th Division, responsibility for Bougainville and adjacent islands.8

Having insisted on the necessity for holding New Georgia and Santa Isabel, naval authorities then complained that this responsibility placed an excessive demand on naval strength, and asked Imamura for some Army ground troops for New Georgia in addition to the few who were already there. The general, still invoking the difficulty of supply, was at first reluctant. In March the Southeastern Fleet sent the 8th Combined Special Naval Landing Force to New Georgia, and another, the 7th, to Santa Isabel.9 After a good deal of negotiation, and perhaps on orders from Imperial Headquarters, Imamura acceded to Kusaka's requests and sent more Army troops to New Georgia under their own headquarters, the Southeastern Detachment, and some additional units to Santa Isabel. Both the 8th Combined Special Naval Landing Force and the Southeastern Detachment, as well as the Santa Isabel force, were under the tactical control of the 8th Fleet.

Thus in early 1943 the Japanese were holding a network of mutually supporting air and naval bases arranged in depth, running in two converging arcs through New Guinea and the Solomons to Rabaul. From the defensive point of view, these positions would serve to protect Rabaul, the Netherlands Indies, and the Philippines. Offensively, these bases could support advances southward, and although the Japanese had decided on delaying action in the Solomons, they were determined to take the offensive in New Guinea.

Japanese Offensives, January-June 1943

The Attack Against Wau

The first offensive effort under the revised strategy was directed against Wau in the Bulolo Valley goldfields southeast of New Guinea's Huon Peninsula. Wau, the site of a prewar airfield, lies 145 air miles north by west of Port Moresby, and 25 air miles southwest of Salamaua. (Map 4) Since May 1942 Wau had been held by a small body of Australians, known as the KANGA Force, who operated

--36--

under control of the New Guinea Force. As the Bulolo Valley could be reached overland from other Allied bases only over mountainous, jungled, and swampy routes, the KANGA Force was supplied largely by air. It had been ordered to keep watch over Lae and Salamaua and to hold the Bulolo Valley as a base for harrying the enemy until he could be driven out of the area.10 If the Japanese had been able to establish themselves at Wau, they could have reaped great gains. They could have staged aircraft from Madang and Wewak through Wau, thus bringing Port Moresby within effective range of their fighters.11 The 18th Army entertained ambitious plans for capturing Wau and crossing the Owen Stanley Range to seize Port Moresby. It is not clear, however, whether Adachi intended to proceed from Wau over the rough trail that led from Wau to Bulldog on the Lakekamu River, or to move against Port Moresby via Kokoda. Either route would have outflanked the Allied Gona-Sanananda-Buna-Dobodura-Oro Bay positions that had been won in the arduous Papuan campaign.

When 18th Army troops moved to New Guinea in early 1943, some went to Lae and Salamaua to strengthen naval forces already there.12 The reinforced 102d Infantry Regiment was sent in a convoy from Rabaul to Lae during the first week in January. But the Allies,


Map 4
The Wau Area

warned by the fact that the Japanese had given up their efforts to send troops to Buna, had anticipated that the Japanese might try to strengthen Lae and Salamaua and were therefore attempting to isolate that area by air action. Allied planes found the convoy, bombed it, and sank two transports. About three fourths of the 102d went ashore at Lae, but half its supplies were lost.

Once at Lae, the 102d was ordered by Adachi to seize Wau. This Allied enclave was connected to the north coast by several trails that could be traversed on foot. The Japanese commander at Lae, Maj. Gen. Toru Okabe, decided to begin his drive against Wau from Salamaua.

--37--


LT. GEN. HARUYOSHI HYAKUTAKE

By 16 January he had gathered his attacking force there.

The Allies, determined to prevent the Japanese from capturing Wau and threatening Port Moresby, had meanwhile acted promptly. Headquarters, New Guinea Force, decided to reinforce Wau, and in mid-January advance elements of the 17th Australian Infantry Brigade were flown from Milne Bay to Wau.13

After assembling at Salamaua, Okabe and the 102d Infantry made their way laboriously upward to the Bulolo Valley. They struck at Wau in a dusk attack on 28 January and pushed through to the edge of the airfield. But there they were stopped. For the next three days Australian soldiers of the 17th Brigade, plus ammunition, supplies, and two 25 pounder guns, were flown in by air. In three days troop carriers of the Allied Air Forces flew in 194 planeloads, or one million pounds. So critical was the situation on the 29th that the first load of troops practically leaped from the planes firing their small arms. The Japanese pressed hard, but by 30 January acknowledged failure and began to withdraw.14

Having broken the enemy's attack, the Australians kept pressing him back toward Salamaua. In April the 3d Australian Division took over direction of operations and the KANGA Force was dissolved. The Australians then halted short of Salamaua to wait until other Allied troops could be made ready for a large-scale attack against the entire Finschhafen-Lae-Salamaua complex.15

The Australians' gallant defense of Wau thus frustrated the last Japanese attempt to attack Port Moresby overland, and kept for the Allies an advantageous

--38--

position which would help support later offensives against the Huon Peninsula.

The Battle of the Bismarck Sea

The Australian defense of Wau had a third consequence that was more far-reaching than even the most ebullient Bulolo Valley veteran (if anyone was ebullient after fighting in the mud, mountains, and heat) realized at the time. It helped lead to the destruction of an entire Japanese convoy and the subsequent weakening of Lae.16

Okabe's attacks against Wau had so depleted his meager force that the Japanese at Rabaul, who were determined to hold Lae and Salamaua at all costs, became worried. The 20th and 41st Divisions could not be spared from Wewak and Madang. Thus Imamura, Adachi, and the naval commanders decided to send the rest of the 51st Division in convoy to Lae. They planned very carefully.

They were well aware of the havoc that airplanes could wreak on troop transports. Guadalcanal had demonstrated that point, and if final proof was needed, Adachi had had it in the destruction of part of Okabe's shipment in January. Had it been possible for the convoy to sail from Rabaul to Madang and land the troops there to march to Lae, the ships could have stayed out of effective range of Allied fighters and medium bombers; heavy bombers, thus far relatively ineffective against ships, were not greatly feared. But there was no overland or coastal route capable of getting large bodies of troops from Madang to Lae. It was therefore necessary to sail directly to Lae and thus come within range of fighters and medium bombers. The Japanese, employing almost two hundred planes based at Rabaul, Madang. Wewak, Cape Gloucester, Gasmata, and Kavieng, hoped to beat Allied planes down out of the air and to provide direct cover to the ships.

But the Allies had deduced Japanese intentions. Ship movements around New Britain in late February, though not part of the effort to reinforce Lae, were noted by Allied reconnaissance planes. As a result air search was intensified and air striking forces were alerted. On 25 February General Kenney and his subordinates came to the conclusion that the Japanese would probably try to put more troops ashore at Lae or Madang.17

Not only were the Allies warned; they were also ready. By the end of February airfields in Papua, with those at Dobodura near Buna carrying the biggest load, based 207 bombers and 129 fighters. The Southwest Pacific had no aircraft carriers and few if any carrier-type planes that were specifically designed for attacks against ships. But Kenney and his subordinates had redesigned the nose of the B-25 medium bomber and installed forward-firing .50-caliber machine guns so that the bomber could strafe the deck of a ship and thus neutralize all her exposed antiaircraft guns. Further, they had practiced the skip-bombing technique that proved particularly effective in sinking

--39--


JAPANESE TROOP TRANSPORT UNDER ATTACK by B-25's, Battle of the Bismarck Sea.

ships.18 Once warned, the Allied airmen prepared detailed plans for striking the convoy and executed a full-scale rehearsal off Port Moresby.

At Rabaul, 6,912 Japanese soldiers boarded eight ships. The ships weighed anchor about midnight of 28 February 1943 and, with eight destroyers as escort, sailed out of Rabaul and westward through the Bismarck Sea at seven knots. At first bad weather--winds, mist, and rain--hid them from the air, but soon the weather began to break and Allied patrol planes sighted the convoy first on 1 March and again the next morning off Cape Gloucester. As it was still beyond the reach of medium bombers, heavy bombers from Port Moresby attacked it in the Bismarck Sea. They sank one transport and damaged two others, a good score for heavy bombers. Survivors of the sunken ship, about 950 in number, were picked up by two of the destroyers which made a quick run to Lae to land the men after dark. The destroyers returned to the convoy on the morning of 3 March. During the night the convoy had sailed through Vitiaz Strait and into the Solomon Sea, tracked all the while by an Australian Catalina.

But now the ships entered the Huon Gulf in clear daylight, and were within range of medium bombers. The Allied planes that had organized and rehearsed

--40--

for the attack assembled over Cape Ward Hunt at 0930 and set forth for the kill. The Japanese had failed to destroy Allied air power in advance, and the convoy's air cover was ineffective. Starting about 1000 and continuing until nightfall, American and Australian airmen in P-38's, P-39's, P-40's, Beaufighters, A-20's, A-29's, Beaufort bombers, B-17's, B-24's and B-25's pounded the luckless Japanese from medium, low, and wavetop altitudes with resounding success. All remaining transports, along with four destroyers, sank on 3 and 4 March. After night fell motor torpedo boats from Buna and Tufi swept in to finish off crippled ships and shoot up survivors in the water.

Of the 6,912 troops on board, 3,664 were lost. Including those taken by destroyer to Lae, 3,248 were rescued by the Japanese. The sinking of eight transports and four destroyers in "the most devastating air attack on ships" since Pearl Harbor was a tremendous victory, and it was won at a cost of thirteen killed, twelve wounded, and four Allied planes shot down.19

The Japanese quickly changed their plans for future shipments. They decided to send no more convoys to Lae. Large slow ships would be sent only to Hansa Bay and Wewak; high-speed ships and small craft would run to Finschhafen and Tuluvu on the north coast of New Britain. Small coastal craft would take men and supplies to Lae from Finschhafen and Cape Gloucester, and some men and supplies would be sent overland from Finschhafen to Lae. In emergencies supplies that were absolutely required at Lae would be sent in by high-speed ships or submarines. The main body of ground forces eventually intended for Lae would be sent overland after completion of a road, already under construction, from Wewak through Madang to Lae.

Road Building

Construction of the road had been started in January. This most ambitious project involved building a truck highway from Madang to Bogadjim, thence over the Finisterre Range and through the Ramu and Markham River Valleys to Lae. The 20th Division was given this work.

In early February the Allies, having received reports from natives, were aware of enemy activity in the Ramu Valley. Allied intelligence deduced that the Japanese were interested in an inland route to Lae. Intelligence also minimized the danger of a serious threat, for it seemed unlikely that the road could

--41--

be completed in time to be of much use.20

Allied intelligence was correct. The road-building projects were next to impossible for the Japanese to accomplish. Their maps were poor. The routes they selected, especially the inland route for the Madang-Lae road, led them through disease-ridden jungles and swamps, over towering mountains, and up and across canyons and gorges. They never had enough machinery and what they had was ineffective. Their trucks, for example, were not sufficiently powerful to climb steep slopes. Their horses fared poorly on jungle grasses. Bridges kept washing away on the Madang-Hansa Bay road. Combat troops were unhappy as laborers. Dense forests hid the road builders from air observation, but in the open stretches of the Finisterre Range they were constantly subject to air attack. By the end of June the Madang-Lae road had been pushed only through the Finisterre Range.

Lae therefore never did receive substantial reinforcements or supplies, despite the Japanese determination to hold it and dominate Dampier and Vitiaz Straits.

The I Operation

While Japanese Army troops were busy building roads in New Guinea, the Japanese Navy had also taken a hand in an effort to beat the Allies. Galled by the admittedly crushing defeat in the Bismarck Sea, fully aware of the threat that Allied air activity in the South Pacific presented to their shipments of troops and supplies to New Georgia and Santa Isabel, and concerned over their declining air strength, the Japanese decided to gather more planes, smash Allied air power, and attack Allied shipping in the Southeast Area.

Japanese air strength was somewhat less than substantial at this time. In March 1943 there were only about three hundred planes--one hundred Army and two hundred Navy--in the Southeast Area. Rabaul frequently complained that Tokyo never sent enough replacements to replace losses. Toward the end of March General Imamura asked Imperial Headquarters for more. Headquarters did send more, but not enough to satisfy Imamura, and some planes that were dispatched never arrived. For example, the 68th Air Regiment navigated so badly while flying from Truk to Rabaul that many of its planes failed to find Rabaul and were lost at sea.

Admiral Isoroku Yamamoto, Commander in Chief of the Combined Fleet, decided to take a hand in the attempt to beat the Allies out of the air. For this effort, given the code name I Operation, he sent the planes from the 3d Fleet carriers at Truk to join with 11th Air Fleet planes at Rabaul, Kavieng, Buin, Buka, and Ballale. He took headquarters of both the Combined and 3d Fleets from Truk to Rabaul to direct the I Operation, which involved more than three hundred aircraft.

Japanese aircraft had concentrated against the Allied New Guinea bases in March, and the month had been a quiet one on Guadalcanal. But that the Japanese had renewed their interest in the

--42--

Solomons was demonstrated to the Allies on 1 April when bombers and fighters struck at the Russells. Air combats raged for three hours as Allied fighters beat off the attackers, losing six of their number in the process.21

Six days later, 7 April, came the main phase of I Operation in the Solomons. It was a splendid opportunity for the Japanese, for there were many targets around Guadalcanal. A naval task force, having fueled at Tulagi, was steaming northwest en route to shell Vila and Munda that night. Including cargo ships, transports, and the task force, there were present about forty ships of corvette size or larger, and a larger number of smaller vessels. In addition much ammunition, fuel, and equipment were being stored on Guadalcanal in preparation for the invasion of New Georgia.

To attack these lucrative targets, Yamamoto dispatched 117 fighters and 71 bombers. Coastwatchers on New Georgia, counting more than 160 planes overhead, flashed warnings southward. Halsey canceled the scheduled bombardment; the task force rounded Florida and sped down Indispensable Strait. Other ships and craft started getting under way and most had reached open water when the Japanese arrived about 1500.

While Allied bombers flew to the southeast to avoid the Japanese, all available Allied fighters, seventy-six in number, took the air to intercept. P-38's (Lightnings) flew on top, and beneath them, at various altitudes, were F4U's (Corsairs), F6F's (Hellcats), and P-39's (Airacobras). As the Japanese planes broke up into separate flights, a general melee ensued. The skies above the Russells, Tulagi, and the waters between Guadalcanal and Florida saw violent combat. According to the Japanese, "resistance offered by the ten or so enemy Grummans [F6F's] and P-38's was beaten down and the attack on shipping was carried out." They reported seriously damaging most of the Allied ships, a claim that is as inaccurate as their statement that only ten Allied fighters tried to intercept.22 They sank the New Zealand corvette Moa, the U.S. oiler Kanawha, and the U.S. destroyer Aaron Ward, and damaged one other oiler. They apparently never sighted the task force. Seven Allied fighters and one pilot were lost, but the Japanese lost many more.23

Yamamoto, apparently satisfied with the performance over Guadalcanal, then turned against the Allies in New Guinea. On 11 April 22 bombers and 72 fighters struck at Oro Bay. They sank one merchant ship, damaged another so badly that it had to be beached, and hit an Australian minesweeper. Next day 131 fighters and 43 bombers flew over the Owen Stanleys to hit Port Moresby. There were few Allied fighters on hand to oppose them. As he himself points out, General Kenney had expected the attack to hit Milne Bay and had sent most

--43--

of his fighter strength there.24 Fortunately the damage was very light. Two days later the Japanese fulfilled Kenney's expectations by attacking Milne Bay, but they did little damage. One Dutch merchant ship was a total loss, and a British motorship and another Dutch ship were damaged. Yamamoto then concluded the I Operation, which he regarded as highly successful, and returned the carrier planes to their parent units at Truk. The Japanese, apparently misled by optimistic pilots' reports, boast of destroying 1 cruiser, 2 destroyers, 25 transports, and 134 planes, while losing 42 planes themselves. But actual Allied losses in the Solomons and New Guinea were 1 destroyer, 1 tanker, 1 corvette, 2 Dutch merchant ships, and about 25 planes.

Ambush Over Kahili

Yamamoto then decided to pay a morale-building visit to the Buin area. He, his chief of staff, and other officers left Rabaul on 18 April in two twin-engine bombers escorted by fighters. When the party reached a point thirty-five miles northwest of Kahili, the airdrome near Buin, they were jumped by eighteen P-38's from the South Pacific's Thirteenth Air Force, which had been sent there for that very purpose.

When Admiral Halsey returned to Noumea after conferring with MacArthur in Brisbane, he learned that American intelligence officers had discovered the exact time on 18 April Yamamoto was due to reach the Buin area from Rabaul. Admiral Nimitz and his staff agreed that disposing of Yamamoto would advance the Allied cause, so the Commander, Aircraft, Solomons, was told to shoot him down. The eighteen P-38's, manned by picked pilots and led by Maj. John W. Mitchell, were sent on the mission. Taking off from Henderson Field on Guadalcanal, they flew low over the waves for 435 miles by a circuitous route to the interception point northwest of Kahili. Yamamoto's flight hove in sight just as its fighter escort was leaving. Mitchell's attack section, led by Capt. Thomas G. Lanphier, Jr., bored in and Lanphier made the kill. Yamamoto's plane crashed in the Bougainville jungle. He died. The other plane fell in the sea, but the chief of staff, whom it was carrying, survived. One American pilot was lost. This Lucifer-like descent of the aggressive, skillful Yamamoto, perhaps the brightest star in the Japanese military firmament, was a severe blow to the morale of the Japanese armed forces.25

The Big Raid

By early June, the Allies in the Solomons realized that the Japanese were again determined to accomplish what the I Operation had failed to do--cut the lines of communication to Guadalcanal by air action. Yamamoto fell from the skies believing that I had succeeded, but by June the enemy leaders at Rabaul knew that the Allies were freely building up supplies on Guadalcanal. On 7 June the Japanese inaugurated another series of fighter-escorted bombing attacks against Guadalcanal. Planes from the

--44--

Russells made the first interception that day. According to Allied accounts, the Japanese lost twenty-three fighters, four of them to P-40's of the No. 15 Royal New Zealand Air Force Fighter Squadron in its Solomons debut. Nine Allied planes were shot down but all pilots were recovered. In a second attack five days later, the Japanese are reported to have lost thirty-one planes, the Allies, six.

By mid-June Allied reconnaissance planes were reporting 245 planes at Rabaul, with the forward fields in the northern Solomons filled to capacity. What some Allied veterans of this period call "the big raid" on Guadalcanal came on 16 June when a large force of enemy bombers and fighters, numbering over 100 planes, flew down to attack Guadalcanal.26 The coastwatchers again had sent their timely warnings, and 104 Allied fighters were ready. As in April, they intercepted promptly, the Japanese formations broke up into smaller flights, and air combats raged. Whenever possible ship- and shore-based antiaircraft took the enemy under fire. The Japanese hit three Allied ships, two of which had to be beached, and did some damage to shore installations before they were driven off. Six Allied fighters were shot down. The number of enemy planes destroyed was large, although the exact total cannot be determined. The Allies claimed 98. One Japanese account admits the loss of about 30 planes.27

Neither the I Operation nor "the big raid" achieved substantial results. The Japanese failed, partly because their efforts were brief and sporadic rather than long and sustained, and partly because Allied resistance had been vigorous and generally skillful.

Japanese Strength and Dispositions, 30 June 1943

In June Japanese strategy was still substantially what it had been in January.28 Late in March Lt. Gen. Rimpei Kato, the 8th Area Army's chief of staff, and other officers had gone to Imperial Headquarters, apparently to explain things after the Bismarck Sea debacle. The result of the visit was an Army-Navy "Central Agreement" which was really a reaffirmation of the policies laid down earlier. The Japanese still planned to defend the Solomons while strengthening the bases in New Guinea and the Bismarck Archipelago in preparation for future offensives, especially against Port Moresby. Ambitious plans for air supremacy were prepared, including one for maintaining 641 planes (284 Army and 357 Navy) in the Southeast Area, but, as has been shown above, these were destined to fail.

In April Imamura summoned his army commanders to Rabaul and gave them orders based on the Central Agreement. Instructions to Adachi emphasized holding Lae and Salamaua, building the Madang-Lae highway, and establishing coastal barge lines from western New Britain to Lae and Salamaua. In fulfillment of the policy of using naval air in

--45--

the Solomons and Army air in New Guinea, the entire 6th Air Division was told to move to New Guinea.

In June Imamura issued more orders, which restated the importance of Lae and Salamaua. The 18th Army was told to strengthen them as well as Wewak, Madang, and Finschhafen. Adachi was to regroup his forces at Lae and Salamaua and prepare to capture the Allied outposts and patrol bases at Wau, Bena Bena, and Mount Hagen, and to infiltrate up the Ramu and Sepik River Valleys.

In anticipation of the operations against Bena Bena and Hagen, Imperial Headquarters transferred the 7th Air Division from the Netherlands Indies to the 8th Area Army about July, and shortly afterward placed Headquarters, 4th Air Army under Imamura to coordinate operations of the two air divisions.

Imamura also developed an ambitious airfield construction program which involved building new fields or enlarging old ones. By June, too, all divisions of the 18th Army--the 20th, 41st, and 51st--were concentrated in New Guinea. The 17th Army, still consisting chiefly of the 6th Division, was in Bougainville. The Southeastern Detachment and the 8th Combined Special Naval Landing Force were dug in deeply in New Georgia, and the 7th Combined Special Naval Landing Force and Army elements still held Rekata Bay at Santa Isabel.

It is not possible, on the basis of existing information, to state positively just how many troops Imamura had under his command at this time. These figures, and those in the table of strength and dispositions, are guesses based on available enemy data: for the Solomons, 25,000; for the New Guinea coast east of the Dutch border, 55,000; for the Bismarck Archipelago, perhaps 43,000 ground troops, for a total of 123,000.29

In aircraft, the Japanese possessed a total of something over 500 planes in June, though some of them were usually

--46--

TABLE 1
COMPARISON OF ALLIED INTELLIGENCE ESTIMATES WITH JAPANESE STRENGTH AND DISPOSITIONS, SOUTHEAST AREA, 30 JUNE 1943
GROUND FORCES
Allied Estimates Japanese Units and
Approximate Strength
130,250/131,550

123,000
Bismarck Archipelago: 42,800 43,000
  (Rabaul, Talasea, Gasmata, Cape Gloucester, Tuluvu, Kavieng, Lorengau)   8th Area Army Hq
    8th Area Army Hq   6th Air Div Hq
  18th Army Hq   38th Div
  Elements 38th Div   65th Brig
  65th Brigade   Naval Units
  Naval Units  
New Guinea (east of Dutch border): 47,200/48,300 55,000
Lae-Salamaua-Finschhafen: 9,000/9,100 15,000
  51st Div   51st Div
  SNLF   SNLF
  Misc   Misc
Saidor: 500 unidentified   ?
Madang: 12,000/13,000 20,000
  Adv Hq, 18th Army   18th Army Hq
  20th Div   20th Div
Hansa Bay: 5,000 Misc   ?
Wewak: 20,000 20,000
  41st Div   41st Div
  Misc   Naval Units
Aitape: 200 unidentified   ?
Vanimo: 500 Misc   ?
Solomons: 40,250/40,450 25,000
Buka: 4,000 1,130
  6th Cav   6th Cav
  SNLF   SNLF
Bougainville-Shortlands:  
  26,700/26,900 8,480
  17th Army Hq   17th Army Hq
  6th Div   6th Div
  SNLF   SNLF
New Georgia: 10,550 10,500
  229th Inf   13th Inf
  SNLF   229th Inf
      SNLF
Santa Isabel: 2,000 3,100
  SNLF   SNLF
  Misc   1 Army Inf Bn
Choiseul; none listed 1,500/2,200
NAVAL FORCES
  1 light cruiser 1 cruiser
  9/10 destroyers 8 destroyers
  5 submarines 8 submarines
AIR FORCES
461 planes 540 planes
    (390 operational)

Source: Allied estimates are from GHQ SWPA G-2 Int Bulletin, Summary Enemy Dispositions, 30 Jun 43, in GHQ SWPA G-3 Jnl, 30 Jun 43; and CTF 31 Opn Plan A8-43, 4 Jun 43, Int Annex, Off of Naval Rcds and Library. Approximate strengths of Japanese units are from Japanese Monographs Nos. 37, 40-42, 48-60, 72, 110, and 107 (OCMH). The Australian Official War historians regard the postwar estimates as too low. See Ltr, Long to Greenfield, 7 Dec 66. They are, however, admittedly conservative, and based on the best Japanese data available in Washington.


out of action. For example, of the 300 planes assigned to the 11th Air Fleet on 30 June, only 25 were ready for combat operations. Of 240 belonging to the 6th Air Division, 50 needed minor attention and 25 required major repairs.

Planes were given a high degree of tactical mobility by the large number of conveniently spaced air bases in the area. Kavieng had one field. Rabaul boasted four--Lakunai, Vunakanau, Rapopo, and Keravat (which never amounted to much)--with one more under construction at nearby Tobera. In addition, the 8th Area Army was improving fields, or building new ones, at Wewak, Hansa Bay, Alexishafen, Madang, Lae-Salamaua, Tuluvu, and Talasea. The same situation prevailed in the Solomons. Besides the New Georgia fields and the seaplane bases at Rekata Bay and Shortland-Faisi, there were fields at

--47--

Kahili, and Ballale in the Buin-Shortlands area, and another, Kara, soon to be built. There was one at Buka, with another, Bonis, under construction just across Buka Passage. On the east coast of Bougainville the Tenekau and Kieta strips wre being built, apparently under orders of the 8th Area Army.30

The 8th Fleet, in June, had one cruiser, eight destroyers, and eight submarines. The potential of this fleet had been cut somewhat by Admiral Mineichi Koga, who had succeeded to command of the Combined Fleet. Because the recapture of Attu in May was regarded as a direct threat to the Japanse homeland, he diverted 20 percent of the forces (apparently including aircraft) "in the course of being assigned or available for assignment" in the Southeast Area to the Aleutians and to Saipan.31

Table 1 compares Allied estimates of enemy strength, and dispositions of that stength and dispositions based on available enemy records. It will be noted that allied estimates for Japanese strength and dispositions throughout the Southeast Area were quite accurate.

In June 1943 the Japanese still cherished ambitions toward future offensives. It is clear in retrospect that their resources made them capable of defensive action only. But, as at Guadacanal and Buna, the Japanese were so skillfull in defensive operation that Allied troops were faced with a long series of hard battles.

--48--

Table of Contents ** Previous Chapter (III) * Next Chapter (V)


Footnotes

1. Unless otherwise indicated all data on the Japanese in this chapter are derived from the following monographs in the series Japanese Studies in World War II: 17th Army Operations, Vol II, Monogr No. 40 (OCMH); 18th Army Operations, Vols. I, II, Monogrs No. 41, 42 (OCMH); Southeast Area Naval Operations, Vols. I-III, Monogrs No. 48-50 (OCMH); History of the Army Section, Imperial General Headquarters, 1941-45, Monogr No. 72 (OCMH); Outline of Southeast Area Naval Air Operations, November 1942--June 1943, Pt. III, Monogr No. 107 (OCMH); and 8th Area Army Operations, Monogr No. 110 (OCMH). Monograph No. 110 is a greatly improved revision of History of the 8th Area Army, 1942-44, Monograph No. 37 (OCMH).

2. For early Japanese planning see Morton, The Fall of the Philippines, pp. 51-61; Milner, Victory in Papua; Miller, Guadalcanal: The First Offensive, pp. 4-7.

3. From 7 January to 2 April 1943, the western boundary was the border of Dutch and Australian New Guinea--longitude 141° east. In addition to Japanese sources cited see U.S. Strategic Bombihg Survey, The Allied Campaign against Rabaul (Washington, 1946), pp. 10, 83.

4. USSBS, The Allied Campaign Against Rabaul, pp. 11, 43-44.

5. Southeast Area Naval Operations, 1, Japanese Monogr No. 48 (OCMH), 55.

6. 18th Army Operations, 1, Japanese Monogr No. 41 (OCMH), 55-56.

7. Ibid., p. 110.

8. Of the 17th Army units which served on Guadalcanal, the 35th Brigade (Kawaguchi Force or Detachment) went to Burma; the 2d Division, to the Philippines; the 38th Division, to New Britain under direct control of the 8th Area Army.

9. Composed of the Kure 6th and the Yokosuka 7th Special Naval Landing Forces, the 8th Combined had been activated in Japan for service on Guadalcanal but did not get there before the evacuation.

10. Milner, Victory in Papua, Chs. I, III; ALF, Rpt on New Guinea Opns: 23 Sep 42-22 Jan 44.

11. AAF Int Summary 74, 3 Feb 43, in GHQ SWPA G-3 Jnl, 2 Feb 43.

12. Interrogation of Lt Gen Hatazo Adachi, Lt Gen Rimpei Kato (former CofS, 8th Area Army), Lt Col Shoji Ota (former stf off, 8th Area Army), and Capt Sadamu Sanagi (former Senior Stf Off, Southeastern Fleet), by members of the Mil Hist Sec, Australian Army Hq, at Rabaul, no date, OCMH.

13. ALF, Rpt on New Guinea Opns: 23 Sep 42-22 Jan 44; NGF OI 60, 13 Jan 43, in GHQ SWPA G-3 Jnl, 14 Jan 43.

14. Craven and Cate, The Pacific: Guadalcanal to Saipan, pp. 136-37; Kenney, General Kenney Reports, pp. 186-87; ALF, Rpt on New Guinea Opns: 23 Sep 42-22 Jan 44.

15. USSBS, Employment of Forces, p. 17; GHQ SWPA G-3 Opns Rpt 380, 21-22 Apr 43, in GHQ SWPA G-3 Jnl, 22 Apr 43.

16. Unless otherwise indicated, this brief account is based on the Japanese monographs and on definitive accounts in Craven and Cate, The Pacific: Guadalcanal to Saipan, pp. 141-51, and Samuel Eliot Morison, History of United States Naval Operations in World War II, Vol. VI, Breaking the Bismarcks Barrier: 22 July 1942-1 May 1944 (Boston: Little, Brown and Company, 1950), Ch. V.

17. See Kenney, General Kenney Reports, p. 197. See also GHQ SWPA G-2 Est of Enemy Sit 343, 28 Feb-1 Mar 43, in GHQ SWPA G-3 Jnl, 1 Mar 43.

18. For details see Craven and Cate, The Pacific: Guadalcanal to Saipan, pp. 140-41; Kenney, General Kenney Reports, pp. 21-22, 105, 144, 154-55, 162, 164. Maj. Gen. Henry H. Arnold had observed the RAF practicing skip-bombing in 1941 and introduced it to the AAF.

19. Morison, Breaking the Bismarcks Barrier, p. 59. Allied casualty figures are from Kenney, General Kenney Reports, p. 206. Official communiques at the time, based on pilots' reports, claimed twenty-two ships and fifteen thousand men, and Kenney, in his book and in his comments on the draft manuscript of this volume, claimed six destroyers or light cruisers sunk, two destroyers or light cruisers damaged, and from eleven to fourteen merchant vessels in the convoy sunk; he also included, in his total for the Bismarck Sea, two small merchant ships that were sunk at Lae and Wide Bay. According to Kenney a second enemy convoy joined the first, which explains the disparity. However, Craven and Cate, in The Pacific: Guadalcanal to Saipan, pp. 147-50, and Morison, in Breaking the Bismarcks Barrier, after surveying all available enemy records, maintained that the convoy consisted of eight transports and an equal number of destroyers, that there was no second convoy, and that eight transports and four destroyers were sunk. 8th Area Army Operations, Japanese Monogr No. 110 (OCMH) supports them.

20. See GHQ G-2 Daily Summary of Enemy Int and G-2 Est of Enemy Sit 317, 2-3 Feb 43, in GHQ SWPA G-3 Jnl, 3 Feb 43. See also Australian Mil Forces Weekly Int Review 28, 5-12 Feb 43, in GHQ SWPA G-3 Jnl, 5 Feb 43.

21. Craven and Cate, The Pacific: Guadalcanal to Saipan, p. 212.

22. Southeast Area Naval Operations, II, Japanese Monogr No. 49 (OCMH), 13.

23. Craven and Cate, The Pacific: Guadalcanal to Saipan, pp. 212-13; Morison, Breaking the Bismarcks Barrier, pp. 120-24. Allied airmen claimed thirty-nine Japanese planes downed in air combat while surface ships claimed twenty-five. There were undoubtedly many duplications. The Japanese admit losing twenty-one.

24. Kenney, General Kenney Reports, pp. 225, 228-29.

25. Craven and Cate, The Pacific: Guadalcanal to Saipan, pp. 213-14; Halsey and Bryan, Admiral Halsey's Story, pp. 155-57; Morison, Breaking the Bismarcks Barrier, pp. 128-29. There are some differences in these accounts, chiefly regarding Yamamoto's destination and time of arrival. Halsey and Bryan, and Craven and Cate say he was to arrive at Ballale at 0945. Morison says he was due at Kahili at 1145.

26. USSBS, The Allied Campaign Against Rabaul, p. 46, says 150-60 planes attacked on 6 June, but 16 June is apparently intended.

27. Craven and Cate, The Pacific: Guadalcanal to Saipan, pp. 218-19; U.S. Strategic Bombing Survey, The Thirteenth Air Force in the War Against Japan (Washington, 1946); Southeast Area Naval Operations, III, Japanese Monogr No. 50 (OCMH), 53.

28. The basic research for this section was performed by Messrs. Stanley L. Falk and Burke C. Peterson.

29. Strength tables and reports do not seem to have survived the war, and available wartime and postwar documents and accounts are inexact, contradictory, or both. The most detailed figures on Japanese Army strength in the Southeast Area for June 1943 are contained in Southeast Area Naval Operations, Vol. II, Japanese Monograph No. 49 (OCMH), pp. 22-23, but this account gives no hint of the source of the figures that are not supported by the few strength figures in Army records. For example, the naval account states that there were 55,000 men holding Lae, Salamaua, Finschhafen, Madang, and Wewak--the 18th Army's area of responsibility--in June 1943, but an 18th Army statistical table indicates that the 18th Army then contained 80,000 men. The office of Official War History, Department of the Interior, of the Commonwealth of Australia, believes that the figure of 80,000 for the 18th Army is correct. Ltr, Mr. Gavin Long to Dr. Kent Roberts Greenfield (Chief Historian, OCMH), 7 Dec 55, OCMH. Japanese documents give no data on ground-troop strength in the Bismarck Archipelago for June 1943. The nearest date for which there are anything like valid figures is November 1943, and those figures are given in the U.S. Strategic Bombing Survey's The Allied Campaign Against Rabaul. Its text asserts that in November 1943 there were 97,870 (76,300 Army and the remainder Navy) in the Bismarck Archipelago, but this statement does not seem to be solidly supported by the interrogations upon which the text is based. General Imamura is quoted on pages 10, 11, and 82, as stating that of 90,000 troops in the Rabaul area in November, 55,000 belonged to his army. From whatever November figure is selected, the strength of the 17th Division, which arrived in New Britain in September and October of 1943, must be subtracted. Unfortunately, there are no figures on that division's strength. The figure 43,000, given in the text above, for the Bismarcks was obtained by the arbitrary method of subtracting 12,000, a reasonable estimate of the strength of the 17th Division, from Imamura's November figure of 55,000. There are fewer problems for the Solomons, which seem to have contained about 25,000 Japanese ground troops.

30. In addition to the Japanese monographs cited above, see USSBS, The Allied Campaign Against Rabaul, pp. 11-12, 46; USAFISPA Photo Int Unit Periodic Rpt, Airdromes and Seaplane Anchorages, Jul 43.

31. Southeast Area Naval Operations, II, Japanese Monogr No. 49 (OCMH), 18.



Transcribed and formatted for HTML by Jerry Holden for the HyperWar Foundation