World War Two: New Georgia (10) After Munda

The hardest slugging was over, at least on New Georgia. But several tasks faced the troops. The airfield had to be put into shape at once and the remaining Japanese had to be cleaned out of New Georgia and several of the offshore islands.

The Airfield

Repair and enlargement of the battered airstrip began immediately after its capture.”…Munda airfield looked like a slash of white coral in a Doré drawing of hell. It lay like a dead thing, between the torn, coffee-colored hills of Bibilo and Kokengolo.” Seabees of the 73rd and 24th Naval Construction Battalions began the work of widening, resurfacing, and re-grading the field. On 9 August additional naval construction battalions added their tools and men to the task. Power shovels dug coral out of Kokengolo Hill, and bulldozers, earth-movers, graders, and rollers spread and flattened it. Good coral was plentiful, as were men and tools, and the work moved rapidly forward. By 7 August the field, although rough, was suitable for emergency wheels-up landings.

Advance parties from General Mulcahy’s air headquarters moved from Rendova to Munda during the second week of August. On the 14th, the day after the first Allied plane landed, General Mulcahy flew from Rendova to Munda in his amphibian plane and opened Headquarters, Air Command, New Georgia, in a Japanese-dug tunnel in Kokengolo Hill.

Two Marine fighter squadrons (VMF 123 and VMF 124), with twelve Corsairs (F4U’s) each, arrived on the 14th and began operations at once. Together with the fighters based at Segi Point, which were also under Mulcahy, they and other Allied squadrons covered the Allied landing at Vella Lavella on 15 August. There were some difficulties at first. Maintenance crews were inexperienced, there were not enough spare parts, the field was not complete, and taxiways and dispersal areas were small and in poor condition. Japanese naval guns, promptly nicknamed “Pistol Pete,” shelled the airfield from the nearby islet of Baanga intermittently from 16 through 19 August. But conditions quickly improved, and Pistol Pete, which had not done much damage, was captured by elements of the 43rd Division on 19 August.

As the field was enlarged, more planes and units continued to arrive. Operations intensified, and as the Japanese were cleared from the central Solomons Mulcahy’s planes began to strike targets in the northern Solomons. For this reason his command was removed from the New Georgia Occupation Force on 23 September and assigned as part of the Air Command, Solomons. Mulcahy’s fighters escorted bombers to the Bougainville bases, and Munda-based bombers soon began dropping loads there too.

Munda airfield, which by mid-October had a 6,000-foot coral-covered runway and thus was suitable for bombers, became the best and most-used airfield in the Solomons. The rotation of Navy, Marine Corps, and Army Air Forces commanders that was standard in the South Pacific had brought about the relief of Admiral Mitscher as Commander, Aircraft, Solomons, by General Twining, the Thirteenth Air Force commander. General Twining moved his headquarters on 20 October from Guadalcanal to Munda and made the most intensive possible use of the new base.


Airfield development, though of primary importance, could be of only minor interest to the ground troops who had the dreary task of slogging northward from Bibilo Hill in an attempt to trap the retreating Japanese. The job had been assigned to the 27th and 161st Infantry Regiments, both operating under their parent command, the 25th Division.

Addition of the 27th Infantry to the New Georgia Occupation Force had come about because of General Griswold’s urgent requests for more men. During July the Western Force of Task Force 31 had carried fully 26,748 men to Rendova, but by the month’s end not that many men were available for combat. Many of the arrivals were service troops. Further, casualties and disease had weakened the infantry regiments.

The three infantry regiments in the 37th Division (less the two battalions with Liversedge) had an authorized total strength of about 7,000 men. But the 161st Infantry, which entered the campaign below strength, was short 1,350 men. The two-battalion regiments were short too, so that the 37th Division’s rifle regiments had only 5,200 men. And the 43rd Division was in worse shape. With an authorized strength of 8,000 men, its three rifle regiments had but 4,536 men. Griswold had asked Liversedge if he could release two infantry battalions for the Munda drive but Liversedge replied that that would be possible only if Enogai and Rice Anchorage were abandoned. His raider battalions were then only 60 percent effective.

[NOTE 10-5A: See XIV Corps G-3 Jnl File, 28-29, 31 Jul 43. By 14 August sickness and casualties had rendered the 4th Marine Raider Battalion practically unfit for fighting. The battalion surgeon, Lieutenant J. C. Lockhart, USNR, reported that out of 453 men present only 137 were fit for duty. Memo, 4th Mar Raider Bn Surgeon for CO 4th Mar Raider Bn, 14 Aug 43, sub: Health of Personnel of 4th Raider Bn, in XIV Corps G-3 Jnl File, 23 Aug 43.]

On 28 July Griswold asked Harmon for replacements or for a regimental combat team less artillery. This request posed a grave problem for Harmon and Admiral Halsey. The injunction against committing major forces to New Georgia was still in effect; at least it was theoretically in effect, for in small island warfare, especially in 1943, 26,000 men constituted a major force. The only immediately available division was the 25th, and one of its regiments, the 161st, had been sent in fairly early. Further, Halsey and Harmon had planned to use the 25th for the invasion of the Buin-Faisi area of Bougainville.

Yet as long as the high command retained confidence in Griswold, there could be but one answer. As Harmon wrote to his chief of staff, Brigadier General Allison J. Barnett, “…we have to make this Munda-Bairoko business go—and as quickly as possible. It is the job ‘in hand’ and whatever we use we have to get it done before we go on to the next step.” One of the major difficulties, according to Harmon, was the fact that the Americans had underestimated the job in hand. “Munda is a tough nut—much tougher in terrain, organization of the ground and determination of the Jap than we had thought. . . . In both Guadalcanal and New Georgia we greatly underestimated the force-require to do the job.” Thus Harmon alerted the 27th Infantry of the 25th Division for transfer to New Georgia and recommended to Halsey that the 25th Division be taken off the list for Bougainville. [He suggested substituting the 2nd Marine Division or the 3rd New Zealand Division.] As soon as he received Halsey’s approval Harmon ordered up the 27th Infantry. On 29 July Colonel Thomas D. Roberts of Harmon’s staff arrived at Griswold’s headquarters to announce the imminent arrival of the 27th Infantry and some replacements.

At this time the Japanese were still holding the Ilangana-Shimizu Hill-Horseshoe Hill-Bartley Ridge defense line, and no one was anticipating the rapid advances that characterized the first days of August. Thus on 30 July with Colonel Roberts’ concurrence Griswold asked for more 25th Division troops and Harmon promptly promised the 35th Infantry.

Advance elements of the 27th Infantry, and Headquarters, 25th Division, landed on the barrier island of Sasavele on 1 August, and in the next few days the regiment was moved to the right (north) flank of the Munda front to protect the XIV Corps’ right flank and rear.

The Cleanup; North to Bairoko

The Japanese withdrawal from Munda released a sizable body of American troops to attempt the cleanup of the Japanese between Munda and Dragons Peninsula. After the rapid advances began on 1 August the 27th Infantry, temporarily commanded by Lieutenant Colonel George E. Bush, sent out patrols to the north before advancing to clear out the Japanese and make contact with Liversedge.

Meanwhile General Griswold decided that mopping-up operations would have to include a drive from Bibilo Hill northwest to Zieta, a village on the west coast about four crow’s-flight miles northwest of Bibilo, to cut off the retreating Japanese. On 2 August the 37th Division had reported that Fijian patrols had cut the Munda-Bairoko trail but found no evidence of any Japanese traffic. Lieutenant Colonel Demas L. Sears, Assistant Chief of Staff, G-2, 37th Division, offered the opinion that if the Japanese were evacuating New Georgia they were moving along the coast to Zieta rather than to Bairoko. This opinion was buttressed by reports from Colonel Griffith of the Raiders who radioed on 2 August that there had been no traffic in or out of Bairoko.

Next day, on orders from Griswold, the 3rd Battalion, 148th Infantry, left Enogai on a cross-country trek toward Zieta, a trek that was halted short of there on 5 August by additional orders Colonel Douglas Sugg had commanded the regiment until a few days before the move to New Georgia. He fell ill and was hospitalized, and his place was taken by Colonel Bush, the executive. Sugg resumed command of his regiment on 12 August from Griswold. He had decided to use the two 25th Division regiments under General Collins, the commander who had led the division on Guadalcanal, to drive to Zieta and Bairoko.

From then until 25 August, the 25th Division units slogged painfully along the swampy jungle trails in pursuit of the elusive enemy. The Japanese occasionally established roadblocks, ambushes, and defenses in depth to delay the Americans, but the worst enemy was the jungle. The terrain was, if anything, worse than that encountered on the Munda front. The maps were incorrect, inexact, or both. For example, Mount Tirokiamba, a 1,500-foot eminence reported to lie about 9,000 yards northwest of Bibilo Hill, was found to be 4,500 yards south by west of its reported position. Mount Bao, thought to be 6,000 yards east-northeast of Bibilo Hill, was actually 2,500 yards farther on.

As the regiments advanced, bulldozers of the 65th Engineer Battalion attempted to build jeep trails behind them. But the rain fell regularly and the trails became morasses so deep that even the bulldozers foundered. General Collins ordered the trail building stopped in mid-August. Now supplies were carried by hand and on men’s backs to the front, and when these methods failed to provide enough food and ammunition the regiments were supplied from the air.

The 1st Battalion, 27th Infantry, trekked north on the Munda-Bairoko trail and made contact on 9 August with Liversedge. The 2nd and 3rd Battalions, 27th Infantry, after some sharp fighting took Zieta on 15 August, then pushed northwest to Piru Plantation. The plantation lay about seven and one-half airline miles northwest of Bibilo Hill, but the regiment’s advance on the ground required a 22-mile march. The 161st Infantry, following the 1st Battalion, 27th, moved up the trail and after mid-August began patrolling to the west short of Bairoko Harbour. On 25 August, after Griffith had reported two nights of busy enemy barge activity in and out of Bairoko Harbour, the Americans bloodlessly occupied its shores.

The Japanese Evacuation

But the main body of Japanese survivors had slipped out of Zieta and Bairoko. Traveling light, they had evaded the slower-moving, more heavily encumbered Americans. On 5 August General Sasaki had decided that he could defend New Georgia no longer. He therefore sent the 13th Infantry and most of the Bairoko-based special naval landing force units to Kolombangara, the 229th Infantry, the 3rd Battalion, 230th Infantry, and the 3rd Battalion, 23rd Infantry, to Baanga, a long narrow island which lay across Lulu Channel from Zieta. These units, plus two 120-mm. naval guns, were ordered to defend Baanga, and the naval guns were to shell Munda airfield. Sasaki’s headquarters, having moved out of Munda, was established on Baanga until 7 August, and the next day he moved to Kolombangara.


The islet of Baanga, 6,500 yards long and some 4,000 yards west of Kindu Point on New Georgia, was captured to extend Allied control over Diamond Narrows and to stop the shelling of Munda by the two 120-mm. guns nicknamed “Pistol Pete” by the American troops.

Seizure of Baanga was entrusted to the 43rd Division, briefly commanded by General Barker after 10 August, when General Hodge returned to the Americal Division. [Barker was replaced several days later, on orders from the War Department, by General Wing, who was senior to him.] Patrolling started on 11 August, but the Japanese on Baanga fought back hard, and the 169th Infantry, which Barker initially assigned to Baanga, gave a “shaky performance.” The 172nd Infantry (less one battalion) joined in, and the southern part of the island was secured by 21 August. The 43rd Division lost 52 men killed, no wounded and 486 non-battle casualties in this operation.

The Japanese, meanwhile, had decided to get off Baanga. General Sasaki had evolved a plan to use the 13th Infantry, then on Kolombangara, to attack New Georgia, and dispatched his naval liaison officer to 8th Fleet headquarters to arrange for air and fleet support. But he was to get none. Moreover, no more ground reinforcements were to be sent to New Georgia. The last attempted shipment consisted of two mixed battalions from the 6th and 38th Divisions, to be carried to Kolombangara on destroyers. [Each battalion consisted of four rifle companies, a machine gun platoon, and a small artillery unit.]

But Commander Frederick Moosbrugger with six destroyers surprised the Japanese force in Vella Gulf on the night of 6-7 August and quickly sank three Japanese destroyers. The fourth enemy ship, which carried no troops, escaped. Moosbrugger’s force got off virtually scot free, while the Japanese lost over fifteen hundred soldiers and sailors as well as the ships. About three hundred survivors reached Vella Lavella. When Sasaki’s request reached the 8th Fleet, Vice Adm. Gunichi Mikawa, basing his decision on instructions from Imperial General Headquarters, ordered Sasaki to cancel his plan for attacking New Georgia and to concentrate the Baanga troops on Arundel to forestall further Allied advances. So the Japanese left on barges for Arundel, completing the movement by 22 August.

Vella Lavella: The Bypass

Meanwhile an Allied force had made a landing at Barakoma on Vella Lavella, which lay about thirty-five nautical miles northwest of Munda. This landing represented a major and completely successful departure from the original TOENAILS plan. The plan had called for the attack against Munda to be followed by the seizure of Vila airfield on Kolombangara, but the Japanese were now correctly believed to be established on Kolombangara in considerable strength.

Some estimates placed the enemy garrison at ten thousand, a little under the actual total. And Admiral Halsey did not want a repetition of the Munda campaign. As he later put it, “The undue length of the Munda operation and our heavy casualties made me wary of another slugging match, but I didn’t know how to avoid it.”

There was a way to avoid a slugging match, and that way was to bypass Kolombangara completely and land instead on Vella Lavella. The advantages were obvious: the airfield at Vila was poorly drained and thus no good while Vella Lavella looked more promising. Also, Vella Lavella was correctly reported to contain few Japanese. Vella Lavella, northwesternmost island in the New Georgia group, lay less than a hundred miles from the Japanese bases in the Shortlands and southern Bougainville, but a landing there could be protected by American fighter planes based at Munda and Segi Point.

The technique of bypassing, which General MacArthur has characterized as “as old as warfare itself,” was well understood in the U.S. Army and Navy long before Vella Lavella, but successful bypassing requires a preponderance of strength that Allied forces had not hitherto possessed. [N610-14] The first instance of an amphibious bypass in the Pacific occurred in May 1943 when the Allied capture of Attu caused the Japanese to evacuate Kiska.

[N610-14: Ltr, MacArthur to Smith, Chief of Mil Hist, 5 Mar 53, no sub, OCMH. In ground operations field orders usually specify that isolated pockets of resistance are to be bypassed, contained, and reduced later, so that the advance will not be held up.]

When members of Halsey’s staff proposed that South Pacific forces bypass Kolombangara and jump to Vella Lavella of the more euphonious name, the admiral was enthusiastic. On 11 July he radioed the proposal to Admirals Turner and Fitch. “Our natural route of approach from Munda to Bougainville,” he asserted, “lies south of Gizo and Vella Lavella Islands.” He asked them to consider whether it would be practicable to emplace artillery in the Munda-Enogai and Arundel areas to interdict Vila; cut the supply lines to Vila by artillery and surface craft, particularly PT boats; “by-pass Vila area and allow it to die on the vine”; and seize Vella Lavella and build a fighter field there. The decision on this plan would depend on the possibility of building a fighter field on Vella Lavella to give close fighter support for the invasion of Bougainville. Both Turner and Fitch liked the idea.


Reconnaissance was necessary first. Allied knowledge of Vella Lavella was limited. Coastwatchers, plantation managers, and such members of the clergy as the Rev. A. W. E. Silvester, the New Zealand Methodist missionary bishop whose see included New Georgia and Vella Lavella, provided some information but not enough to form the basis for the selection of an airfield site or an invasion beach. Colonel Frank L. Beadle, Harmon’s engineer, therefore took command of a reconnaissance party consisting of six Army, Navy, and Marine officers. Beadle was ordered to concentrate his reconnaissance in the coastal plain region of Barakoma and Biloa Mission on the southeast tip of Vella Lavella because it was closest to Munda, the natives were friendly, coastwatcher Lieutenant Henry Josselyn of the Australian Navy and Bishop Silvester were there, and the terrain seemed favorable. The Japanese had already surveyed the Barakoma area for a fighter strip.

Beadle’s party boarded a torpedo boat at Rendova on the night of 21-22 July and slipped through the darkness to land at Barakoma. Silvester, Josselyn, and two natives were on hand to meet the American officers. For six days Beadle’s party, the bishop, the coast-watcher, and several natives explored the southeast part of the island, and ventured up the west coast to Nyanga Plantation, about twelve crow’s-flight miles northwest of Barakoma. Returning to Rendova on 28 July, Beadle reported that the vicinity of Barakoma met all requirements, and that there were no Japanese on the southeast coast of the island. Beadle recommended that the landing be made on beaches extending some 750 yards south from the mouth of the Barakoma River, and suggested that an advance detachment be sent to Barakoma to mark beaches. These recommendations were accepted.

Admiral Wilkinson, Turner’s successor, chose an advance party of fourteen officers and enlisted men from the various units in the Vella Lavella invasion force and placed it under Captain G. C. Kriner, USN, who was to command the Vella Lavella naval base. This group proceeded from Guadalcanal to Rendova, then prepared to change to PT boats for the run through the night of 12-13 August toward Barakoma. The work of the advance party was of a highly secret nature. If the Japanese became aware of its presence, they could kill or capture the men and certainly would deduce that an Allied invasion was imminent.

On 11 August coastwatcher Josselyn radioed Guadalcanal to report the presence of forty Japanese. (Japanese survivors of the Battle of Vella Gulf, 6-7 August, had landed on Vella Lavella.) His message indicated that pro-Allied natives had taken them prisoner. From Koli Point General Harmon radioed General Griswold to ask for more men to accompany the advance party and take the prisoners into custody. Accordingly one officer and twenty-five enlisted men from E and G Companies, 103rd Infantry, were detailed to go along.

The whole party left Rendova at 1730 on 12 August. En route Japanese planes bombed and strafed the four torpedo boats for two hours. One was hit and four men were wounded but the other three made it safely. During the hours of darkness they hove to off Barakoma. Rubber boats had been provided to get the party from the torpedo boats to shore, but no one was able to inflate the rubber boats from the carbon dioxide containers that were provided. So natives paddled out in canoes and took the Americans ashore.

Meanwhile Josselyn had radioed Wilkinson again to the effect that there were 140 Japanese in the area; 40 at Biloa and 100 about five miles north of Barakoma. They were, he declared, under surveillance but were not prisoners. Once ashore Captain Kriner discovered there were many starving, ragged, poorly armed stragglers but no prisoners. He requested reinforcements, and in the early morning hours of 14 August seventy-two officers and enlisted men of F Company,103rd Infantry, sailed for Barakoma in four torpedo boats. This time the rubber boats inflated properly and the men paddled ashore from three hundred yards off the beach.

The advance party with the secret mission of marking beaches and the combat party with the prisoner-catching mission set about their respective jobs. Beach marking proceeded in a satisfactory manner although the infantrymen in that party were not completely happy about the presence of the 103rd troops. They felt that the two missions were mutually exclusive and that the prisoner-catching mission destroyed all hope of secrecy. Only seven Japanese were captured. F Company, 103rd, held the beachhead at Barakoma against the arrival of the main invasion force.

[NOTE: SOPACBACOM’s History of the New Georgia Campaign, Vol. I, Ch. VIII, pp. 20-21, OCMH, contains some fanciful data concerning the sinking of one PT boat and the killing of about fifty Japanese.]


Once Colonel Beadle had made his recommendations the various South Pacific headquarters began laying their plans. This task was fairly simple, for Admiral Halsey and his subordinates were now old hands at planning invasions. Actual launching of the invasion would have to await the capture and development of Munda airfield.

It was on 11 August that Halsey issued his orders. He organized his forces much as he had for the invasion of New Georgia. The Northern Force (Task Force 31) under Admiral Wilkinson was to capture Vella Lavella, build an airfield, and establish a small naval base. Griswold’s New Georgia Occupation Force would meanwhile move into position on Arundel and shell Vila airfield on bypassed Kolombangara. New Georgia-based planes would cover and support the invasion. South Pacific Aircraft (Task Force 33) was to provide air support by striking at the Shortlands-Bougainville fields. As strikes against these areas were being carried out regularly, the intensified air operations would not necessarily alert the enemy. Three naval task forces of aircraft carriers, battleships, cruisers, and destroyers, and the submarines of Task Force 72, would be in position to protect and support Wilkinson. On Wilkinson’s recommendation, Halsey set 15 August as D Day.

Admiral Wilkinson also issued his orders on 11 August. The Northern Force was organized into three invasion echelons (the main body and the second and third echelons) and the motor torpedo boat flotillas. Under Wilkinson’s direct command, the main body consisted of three transport groups, the destroyer screen, and the northern landing force. Each transport group, screened by destroyers, was to move independently from Guadalcanal to Vella Lavella; departure from Guadalcanal would be so timed that each group would arrive off Barakoma just before it was scheduled to begin unloading. Three slow LST’s, each towing an LCM, would leave at 0300, 14 August, six LCI’s at 0800, and seven fast APD’s at 1600. The motor torpedo boat flotilla would cover the movement of the main body on D minus 1 by patrolling the waters east and west of Rendova, but would retire to Rendova early on D Day to be out of the way. Preliminary naval bombardment would in all probability not be necessary, but Wilkinson told off two destroyers to be prepared to support the landing if need be. Two fighter-director groups were put aboard two destroyers. Once unloaded, each transport group would steam for Guadalcanal.

The second echelon, composed of three LST’s and three of the destroyers that would escort the main body, was to arrive at Barakoma on D plus 2, beach overnight, and return to Guadalcanal. The third echelon consisted of three destroyers and three LST’s from the main body. Wilkinson ordered it to arrive on D plus 5, beach throughout the night, and depart for Guadalcanal the next morning.

The northern landing force, 5,888 men in all, consisted of the 35th Regimental Combat Team of the 25th Division; the 4th Marine Defense Battalion; the 25th Cavalry Reconnaissance Troop; the 58th Naval Construction Battalion; and a naval base group.[N100-20] Command of the landing force was entrusted to Brigadier General Robert B. McClure, assistant commander of the 25th Division, who as a colonel had commanded the 35th Infantry during the Guadalcanal Campaign. General McClure would be under Wilkinson’s control until he was well established ashore. He would then come under General Griswold.

[N100-20: The 35th Regimental Combat Team consisted of the 35th Infantry; the 64th Field Artillery Battalion (105-mm. howitzer); C Company, 65th Engineer Battalion; Collecting Company B, 25th Medical Battalion, and detachments from other divisional services. Harmon, who had promised the 35th Infantry for New Georgia on 1 August, later considered using the 145th Infantry, but concluded that it could not be pulled out of New Georgia and brought back to Guadalcanal in time. Rads, Harmon to Griswold, 1 Aug and 6 Aug 43, in XIV Corps G-3 Jnl.]

The Japanese on Vella Lavella (no garrison at all but only a group of stragglers) were estimated to total about 250, with 100 more on nearby Ganongga and 250 at Gizo. Wilkinson warned that enemy air strength in southern Bougainville, less than a hundred miles away, and at Rabaul was considerable, and that naval surface forces were based at both places.

To carry off such a stroke almost literally under the enemy’s aircraft would require, besides fighter cover, considerable speed in unloading. Wilkinson planned to unload the main body in twelve hours. Troops debarking from APD’s were to go ashore in LCVP’s, forty to a boat. At the beach ten of each forty would unload the boat while the thirty pushed inland. Once emptied, LCVP’s were to return to their mother ships for the rest of the men and supplies. Sixty minutes were allotted for unloading the APD’s and clearing the beach. The LCI’s would then come in to the beach and drop their ramps. Passenger troops would debark via both ramps, ground their equipment, then re-board by the starboard ramps, pick up gear, and go ashore down the port ramps. One hour was allotted for the LCI’s. Then the LST’s, bearing artillery, trucks, and bulldozers, would ground. Trucks were to be loaded in advance to help insure the prompt unloading of the LST’s.

The 35th Infantry, commanded by Colonel Everett E. Brown, had been making ready for several days. It had been alerted for movement to Munda on 1 August, and on 9 August had received orders from Harmon’s headquarters to prepare for an invasion. The 1st and 2nd Battalions on Guadalcanal and the 3rd Battalion and the 64th Field Artillery Battalion in the Russells then began rehearsing landings. In the week preceding the invasion South Pacific Aircraft struck regularly at Kolombangara, Buin, Kahili, and Rekata Bay.

By 14 August the landing force and its supplies were stowed aboard ship, and all transport groups of the main body shoved off for Barakoma on schedule. Once on board, the men were informed of their destination. Japanese planes were reported over Guadalcanal, the Russells, and New Georgia, but Wilkinson’s ships had an uneventful voyage up the Slot and through Blanche Channel and Gizo Strait. The sea was calm, and a bright moon shone in the clear night sky. Northwest of Rendova the LCI’s overhauled the LST’s while the APD’s passed both slower groups.

Seizure of Barakoma

As first light gave way to daylight in the morning of 15 August, the APD’s carrying the 1st and 2nd Battalions of the 35th Infantry arrived off Barakoma and hove to. General Mulcahy’s combat air patrol from Munda and Segi Point turned up on schedule at 0605. With part of the 103rd Infantry and the secret advance party already on shore and in possession of the landing beach, there was no need for support bombardment. The APD’s swung landing craft into the water, troops of the 2nd Battalion climbed down the cargo nets and into the boats, and the first wave, with rifle companies abreast, departed. The 2nd Battalion hit the beach at 0624 and at once pushed south toward the coconut plantation around Biloa Mission, about four thousand yards from the beach. The 1st Battalion, having left the APD’s at 0615, landed with companies abreast at 0630 and pushed northward across the waist-deep Barakoma River. Thus quickly unloaded, the APD’s cleared the area and with four escorting destroyers proceeded toward Guadalcanal.

The twelve LCI’s arrived on schedule and sailed in to the beach, but quickly found there was room for but eight at one time. Coral reefs a few yards from shore rendered the northern portion of the beach unusable. The remaining four LCI’s had to stand by offshore awaiting their turn. The 3rd Battalion started unloading but had barely gotten started when enemy aircraft pounced at the invasion force.

This time the Japanese were not caught so completely asleep as they had been on 30 June. In early August Japanese radio intelligence reported a good deal of Allied radio traffic, and the commanders at Rabaul were aware that ships were again concentrating around Guadalcanal. They concluded that a new invasion was impending but failed to guess the target. At 0300 of the 15th a land-based bomber spotted part of Wilkinson’s force off Gatukai. Six dive bombers and forty-eight fighters were sent out on armed reconnaissance, and these found the Americans shortly before 0800. Mulcahy’s planes and ships’ antiaircraft guns promptly engaged them. The Japanese planes that broke through went for the destroyers, which steamed on evasive courses and escaped harm. The Japanese caused some casualties ashore by strafing, but did not attack the LST’s and LCI’s. They ebulliently reported repulsing fifty Allied planes.

This attack, together with the limitations of the beach, delayed the unloading of the LCI’s, which did not pull out until 0915.21 The three LST’s then beached and began discharging men and heavy cargo. Unloading continued all day.

The Japanese struck again at 1227; eleven bombers and forty-eight fighters came down from the north. Some attacked the LST’s but these “Large, Slow Targets” had mounted extra antiaircraft guns and brought down several Japanese planes.

At 1724, some thirty-six minutes before the LST’s departed, the enemy came again. Forty-five fighters and eight bombers attacked without success. The Japanese pilots who flew against the Northern Force on that August day showed a talent for making unwarranted claims. A postwar account soberly admits the loss of seventeen planes, but claims the sinking of four large transports, one cruiser, and one destroyer. It states that twenty-nine Allied planes were shot down and that four large transports were damaged. The ships retiring from Vella Lavella were harried from the air almost all the way, fortunately without suffering damage.

D Day, a resounding success, had proceeded with the efficiency that characterized all Admiral Wilkinson’s operations. Landed were 4,600 troops (700 of whom were naval personnel) and 2,300 tons of gear including eight 90-mm. antiaircraft guns, fifteen days’ supplies, and three units of fire for all weapons except antiaircraft guns, for which one unit was landed. The 35th Regimental Combat Team established a perimeter defense.

Field artillery was in position by 1700, and by 1530 the 4th Marine Defense Battalion had sixteen .50-caliber, eight 20mm., and eight 40-mm. antiaircraft guns and two searchlights in place. The guns engaged the last flight of enemy planes. There were some problems, of course. The LST’s had been unloaded slowly, but supplies came ashore faster than the shore party could clear them off the beach. Boxes of equipment, ammunition, and rations were scattered about. The troops had brought their barracks bags and these lay rain-soaked in the mud. Unused field stoves stood in the way for several days.

The bypass to Vella Lavella was easier and cheaper than an assault on Kolombangara. Twelve men were killed and forty wounded by air bombing and strafing, but D Day saw no fighting on the ground. There was never any real ground combat on Vella Lavella, because Japanese stragglers were mainly interested in escape rather than fighting. When it became known that Wilkinson was landing on Vella Lavella officers of the 8th Fleet and the 17th Army went into conference. They estimated, with accuracy unusual for Japanese Intelligence, that the landing force was about a brigade in strength. With blithe sanguinity someone proposed sending a battalion to effect a counter-landing. General Imamura’s headquarters took a calmer view and pointed out that sending one battalion against a brigade would be “pouring water on a hot stone.” The 8th Area Army stated that two brigades would be needed to achieve success, but that not enough transports were available.

In view of the general Japanese strategy of slow retreat in the central Solomons in order to build up the defenses of Bougainville and hold Rabaul, it was decided to send two rifle companies and one naval platoon to Horaniu at Kokolope Bay on the northeast corner of Vella Lavella to establish a barge staging base between the Shortlands and Kolombangara.

The real struggle for Vella Lavella took place in the air and on the sea. Japanese naval aircraft made a resolute effort to destroy the American ships bearing supplies and equipment to Vella.

Fighters and bombers delivered daylight attacks and seaplanes delivered a series of nocturnal harassing attacks that were all too familiar to Allied troops who served in the Solomons in 1942, 1943, and early 1944.

The combat air patrol from Mulcahy’s command made valiant efforts to keep the Japanese away during daylight, but as radar-equipped night-fighters did not reach the New Georgia area until late September shore- and ship-based antiaircraft provided the defense at night. For daylight cover Mulcahy had planned to maintain a 32-plane umbrella over Barakoma, but the limited operational facilities at Munda made this impossible at first. On 17 August only eight fighters could be sent up at once to guard Barakoma. To add to the difficulties, the weather over New Georgia was bad for a week, the fighter-director teams on Vella Lavella were new to their task, and one of the 4th Defense Battalion’s radars was hit by a bomb on 17 August.

The Northern Force’s second echelon, under Captain William R. Cooke, having departed Guadalcanal on 16 August, beached at Barakoma at 1626 the next day. The fighter cover soon left for Munda and at 1850, and again at 1910, Japanese planes came over to bomb and strafe. General McClure ordered Cooke not to stay beached overnight but to put to sea. Escorted by three destroyers, the LST’s went down Gizo Strait where they received an air attack of two hours and seventeen minutes’ duration. The convoy sailed toward Rendova until 0143, then reversed and headed for Barakoma.

One LST caught fire, probably as a result of a gasoline vapor explosion, and was abandoned without loss of life. The other two reached Barakoma, suffered another air attack, unloaded, and returned safely to Guadalcanal.

Captain Grayson B. Carter led the third echelon to Vella Lavella. It was attacked by enemy planes in Gizo Strait at 0540 on 21 August; one destroyer was slightly damaged and two men were killed. Planes struck off and on all day at the beached LST’s, but men of the 58th Naval Construction Battalion showed such zeal in unloading that the LST’s were emptied by 1600. The next convoys, on 26 and 31 August, had less exciting trips. The weather had cleared and the air cover was more effective. During the first month they were there, the Americans on Vella Lavella received 108 enemy air attacks, but none caused much damage. In the period between 15 August and 3 September, the day on which Wilkinson relinquished control of the forces ashore, Task Force 31 carried 6,505 men and 8,626 tons of supplies and vehicles to Vella Lavella.

During that period General McClure’s troops had strengthened the defenses of Barakoma, established outposts and radar stations, and patrolled northward on both coasts. On 28 August a thirty-man patrol from the 25th Cavalry Reconnaissance Troop that had accompanied radar specialists of the 4th Marine Defense Battalion in search of a new radar site reported considerable enemy activity at Kokolope Bay.

Capture of Horaniu

Having decided to establish the barge base at Horaniu, the Japanese sent the two Army companies and the naval platoon, 390 men in all, out of Erventa on 17 August. Four torpedo boats, 13 troop-carrying daihatsu barges,[N101-25] 2 armored barges, 2 submarine chasers, 1 armored boat, 4 destroyers, and 1 naval air group from the Shortlands were involved. The destroyers were intercepted north of Vella Lavella in the early morning hours of 18 August. The daihatsus dispersed.

[N101-25: Daihatsu is an abbreviation for Ogata Hatsudokitei which means a large landing barge. The daihatsu was 41-44 feet long; it could carry 100-120 men for short distances, 40-50 on long trips. The sides were usually armored, and it carried machine guns.]

The Americans sank the 2 submarine chasers, 2 motor torpedo boats, and 1 barge. The Japanese destroyers, two of which received light damage, broke off the action and headed for Rabaul.26 Harried by Allied planes, the daihatsus hunted for and found Horaniu, and the troops were ashore by nightfall on 19 August. About the same time, General Sasaki took alarm at the seizure of Barakoma and sent the 2nd Battalion, 45th Infantry, and one battery of the 6th Field Artillery Regiment from Kolombangara to defend Gizo.

When General McClure received the report from the reconnaissance troop patrol on 28 August, he ordered Major Delbert Munson’s 1st Battalion to advance up the east coast and take the shore of Kokolope Bay for a radar site. To take the 1st Battalion’s place in the perimeter defense, McClure asked Griswold for a battalion from New Georgia. The 1st Battalion, 145th Infantry, was selected.

On the morning of 30 August Major Munson dispatched A Company up the east coast ahead of his battalion, and next day, after the arrival of the 1st Battalion, 145th Infantry, the main body of the 1st Battalion, 35th, started north. Josselyn and Bishop Silvester had provided native guides and the bishop gave Munson a letter instructing the natives to help the American soldiers haul supplies. C Company, 65th Engineer Battalion, was to build a supply road behind Munson.

By afternoon of 1 September A Company had reached the vicinity of Orete Cove, about fourteen miles northeast of Barakoma. The main body of the battalion was at Narowai, a village about seven thousand yards southwest of Orete Cove. The coastal track, which had been fairly good at first, narrowed to a trail that required the battalion to march in single file. Inland were the jungled mountains of the interior. Supply by hand-carry was impossible, and McClure and Colonel Brown, who had been informed that higher headquarters expected the Japanese to evacuate, decided to use landing craft to take supplies to Munson. On 2 September supplies arrived at Orete Cove along with seventeen Fiji scouts under Tripp, who had recently been promoted to major.

From that day until 14 September Munson’s battalion, supported by the 3rd Battalion, 35th, and C Battery, 64th Field Artillery Battalion, moved forward in a series of patrol actions and skirmishes.

Horaniu fell on 14 September. The Japanese did not seriously contest the advance. Instead they withdrew steadily, then moved overland to the northwest corner of the island. Up to now troops of the United States had borne the brunt of ground combat in the Solomons, but Admiral Halsey had decided to give the 3rd New Zealand Division a chance to show its mettle. He had earlier moved the division from New Caledonia to Guadalcanal. On 18 September Major General H. E. Barrowclough, general officer commanding the division, took over command of Vella Lavella from General McClure. On the same day the 14th New Zealand Brigade Group under Brigadier Leslie Potter landed and began the task of pursuing the retreating enemy.[N103-27] Battalion combat teams advanced up the east and west coasts, moving by land and by water in an attempt to pocket the enemy. But the Japanese eluded them and got safely off the island.

[N103-27: A brigade group was similar in strength and composition to a U.S. regimental combat team. For details of Potter’s operations see Oliver A. Gillespie, The Pacific, “The Official History of New Zealand in the Second World War, 1939-1945” (Wellington, New Zealand, 1952), pp. 125-42.]

The Seabees had gone to work on the airfield at once. As at Munda, good coral was abundant. By the end of August they had surveyed and cleared a strip four thousand feet long by two hundred feet wide. They then began work on a control tower, operations shack, and fuel tanks. The first plane to use the field landed on 24 September, and within two months after the invasion the field could accommodate almost a hundred aircraft.

The decision to bypass Kolombangara yielded this airfield in return for a low casualty rate. Of the Americans in the northern landing force, 19 men were killed by bombs, 7 died from enemy gunfire, and 108 were wounded. Thirty-two New Zealanders died, and 32 were wounded.

Final Operations Arundel

About the time that Vella Lavella was being secured, General Griswold’s forces on New Georgia were carrying out their part of Admiral Halsey’s plan by seizing Arundel and by shelling Kolombangara to seal it off. The attack on Arundel, which is separated from the west coast of New Georgia by Hathorn Sound and Diamond Narrows, proved again that it was all too easy to underestimate the Japanese capacity for resolute defense.

The 172nd Infantry invaded it on 27 August, but the Japanese fought so fiercely that the 27th Infantry, two battalions of the 169th Infantry, one company of the 103rd Infantry, B Company of the 82nd Chemical Battalion (4.2-inch mortars, in their South Pacific debut), the 43rd Reconnaissance Troop, and six Marine tanks had to be committed to keep the offensive going.

Resistance proved more intense than expected in part because the indefatigable Sasaki had not yet abandoned his hope of launching an offensive that would recapture Munda. On 8 September he sent the 3rd Battalion, 13th Infantry, from Kolombangara to strengthen his forces on Arundel, and five days later, when Allied air and naval forces had practically cut the supply lines between Bougainville and Kolombangara and his troops faced starvation, he decided to attack Munda or Bairoko via Arundel and seize the Americans’ food. He therefore dispatched Colonel Tomonari (who was slain in the ensuing fight) and the rest of the 13th Infantry to Arundel on 14 September.

Thus the battle for Arundel lasted until 21 September, and ended then, with the Americans in control, only because Sasaki ordered all his Arundel troops to withdraw to Kolombangara.

The Japanese Evacuation

Sasaki had ordered the evacuation of Arundel because Imperial General Headquarters had decided to abandon the New Georgia Islands completely. While the Americans were seizing Munda airfield, the Japanese naval authorities in the Southeast Area realized that their hold on the central Solomons was tenuous. But they resolved to maintain the line of communications to Kolombangara, so that Sasaki’s troops could hold out as long as possible. If Sasaki could not hold out, the next best thing would be a slow, fighting withdrawal to buy time to build up defenses for a final stand on Bougainville.

Such events in early August as the fall of Munda and the Japanese defeat in Vella Gulf on 6-7 August precipitated another argument between Japanese Army and Navy officers over the relative strategic merits of New Guinea and the Solomons. This argument was resolved in Tokyo by the Imperial General Headquarters which decided to give equal priority to both areas. Tokyo sent orders to Rabaul on 13 August directing that the central Solomons hold out while Bougainville was strengthened, and that the central Solomons were to be abandoned in late September and early October. The decision to abandon New Georgia was not made known at once to General Sasaki.

Sasaki, with about twelve thousand men concentrated on Kolombangara, prepared elaborate defenses along the southern beaches and, as shown above, prepared plans for counterattacks. Finally on 15 September, after Sasaki had sent the 13thInfantry to Arundel, an 8th Fleet staff officer passed the word to get his troops out.

Southeastern Fleet, 8th Fleet, and Sasaki’s headquarters prepared the plans for the evacuation. A total of 12,435 men were to be moved. Eighteen torpedo boats, thirty-eight large landing craft, and seventy or eighty Army barges (daihatsus) were to be used.[N104-28] Destroyers were to screen the movement, aircraft would cover it, and cruisers at Rabaul would stand by in support.

[N104-28: 17th Army Operations, II, Japanese Monogr No. 40 (OCMH), 54, says 138 “large motor boats” were to be used; Southeast Area Naval Operations, II, Japanese Monogr No. 49 (OCMH), 52, lists 18 torpedo boats, 38 large landing barges, and about 70 Army craft]

The decision to use the daihatsus was logical, considering the destroyer losses in Vella Gulf and the success the nocturnal daihatsus had enjoyed. American PT boat squadrons, four in all, had been operating nightly against the enemy barges in New Georgian waters since late July, and had sunk several, but only a small percentage of the total. Destroyers and planes had also operated against them without complete success. The Japanese put heavier armor and weapons on their barges for defense against torpedo boats, which in turn replaced their torpedoes—useless against the shallow-draft barges—with 37-mm. antitank and 40-mm. antiaircraft guns. The barges were too evasive to be suitable targets for the destroyers’ 5-inch guns. Planes of all types, even heavy bombers, hunted them at sea, but the barges hid out in the daytime in carefully selected staging points. Those traveling by day covered themselves with palm trees and foliage so that from the air they resembled islets.

Sasaki ordered his troops off Gizo and Arundel; those on Arundel completed movement to Kolombangara by 21 September. The seaplane base at Rekata Bay on Santa Isabel and the outpost on Ganongga Island were also abandoned at this time. The evacuation of Kolombangara was carried out on the nights of 28-29 September, 1-2 October, and 2-3 October.

Admiral Wilkinson had anticipated that the enemy might try to escape during this period, the dark of the moon. Starting 22 September American cruisers and destroyers made nightly reconnaissance of the Slot north of Vella Lavella, but when Japanese submarines became active the cruisers were withdrawn. The destroyers attempted to break up the evacuation but failed because enemy planes and destroyers interfered. The Japanese managed to get some 9,400 men, or some 3,600 less than they had evacuated from Guadalcanal in February, safely off the island. Most of them were sent to southern Bougainville. Twenty-nine landing craft and torpedo boats were sunk, one destroyer was damaged, and sixty-six men were killed.

The final action in the New Georgia area was the Battle of Vella Lavella on the night of 6-7 October, when ten Japanese destroyers and twelve destroyer-transports and smaller craft came down to Vella Lavella to rescue the six hundred stranded men there. Facing odds of three to one, American destroyers engaged the Japanese warships northwest of Vella Lavella. One Japanese destroyer was sunk; one American destroyer was badly damaged and sank, and two more suffered damage. During the engagement the transports slipped in to Marquana Bay on northwest Vella Lavella and got the troops out safely.30 The last organized bodies of Japanese had left the New Georgia area.

When the 1st Battalion, 27th Infantry, landed at Ringi Cove on southern Kolombangara on the morning of 6 October, it found only forty-nine abandoned artillery pieces and some scattered Japanese who had been left behind. The long campaign—more than four months had elapsed since the Marines landed at Segi Point—was over.


New Georgia had been lengthy and costly. Planned as a one-division affair, it had used up elements of four divisions. It would be months before the 25th and 43rd Divisions were ready to fight again. American casualties totaled 1,094 dead, 3,873 wounded. These figures do not tell the complete story, for they count only men killed or wounded by enemy fire. They do not include casualties resulting from disease or from combat fatigue or war neuroses. For example the 172nd Infantry reported 1,550 men wounded or sick; the 169th Infantry, up to 5 August, suffered 958 nonbattle casualties. The 103rd Infantry had 364 “shelled-shocked” and 83 non-battle casualties.Japanese casualties are not known, but XIV Corps headquarters reported a count of enemy dead, exclusive of Vella Lavella, of 2,483.

The Allied soldiers, airmen, marines, and sailors who suffered death, wounds, or illness, and those who fought in the campaign without injury, had served their cause well. New Georgia was a success. The bypassing of Kolombangara, though overshadowed by later bypasses and clouded by the fact that the bypassed troops escaped, was a satisfactory demonstration of the technique; the seizure of Vella Lavella provided Halsey’s forces with a good airfield for a much lower price in blood than an assault on Kolombangara. The Allies swiftly built another airfield at Ondonga Peninsula on New Georgia. This gave them four—Munda, Barakoma, Ondonga, and Segi. The first three, the most used, brought all Bougainville within range of Allied fighters. When South Pacific forces invaded the island, they could pick an undefended place and frustrate the Japanese efforts to build up Bougainville’s defenses and delay the Allies in New Georgia.

The New Georgia operation is also significant as a truly joint operation, and it clearly illustrates the interdependence of air, sea, and ground forces in oceanic warfare. Victory was made possible only by the close co-ordination of air, sea, and ground operations. Air and sea forces fought hard and finally successfully to cut the enemy lines of communication while the ground troops clawed their way forward to seize objectives intended for use by the air and sea forces in the next advance.

Unity of command, established from the very start, was continued stood off nearly four Allied divisions in throughout with obvious wholehearted-the course of the action, and then success by all responsible commanders. Successfully evacuated 9,400 men to fight No account of the operation should again. The obstinate General Sasaki, who be brought to a close without praising disappears from these pages at this point, the skill, tenacity, and valor of the deserved his country’s gratitude for his heavily outnumbered Japanese who gallant and able conduct of the defense.

SOURCE: Cartwheel: Reduction of Rabaul; By John Miller Jr. (United States Army Center of Military History)

World War Two: Munda Trail (9); XIV Corps Offensive


World War Two: North Africa (6-28); Gafsa, Maknassy, and El Guettar (17-25 March)

18 Army Group’s Plan for II Corps The operations by II Corps in March were intended to accomplish a threefold purpose. Headquarters, First Army, issued a formal directive prescribing the corps mission shortly before II Corps passed to 18 Army Group’s direct control, The II Corps was to draw off reserves from the enemy forces facing the Eighth Army; to regain firm control of forward airfields from which to furnish assistance to Eighth Army; and to establish a forward maintenance center from which mobile forces of Eighth Army could draw supplies in order to maintain the momentum of their advance. This prospective supply point was to be established at Gafsa, which the II Corps was to retrieve from an Italian garrison by an attack to start not later than 15 March.

Troops not required for the defense of Gafsa could then demonstrate toward Maknassy as a menace to the enemy’s line of communications along the coast. In the meantime, the passes in the Western Dorsal from Sbiba southwestward to EI Ma el Abiod were to be firmly held, while the airfields at Thelepte, in front of the Allied defensive line, were to be regained for the use of Allied fighter units. Of the enemy’s combat troops, AFHQ estimated that 45,100 Germans and 28,000 Italians were in the Mareth-Matmata defenses, and 11,400 German and 12,800 Italians in the Gabes-Gafsa-Chott Position area. The garrison at Gafsa, with security forces to the west of it, amounted, AFHQ thought, to 7,100 Italians of the Centauro Division; eastward, from Sened to Maknassy were only 800 German and 750 Italian combat troops. If this appraisal of the opposing forces was correct, the enemy would be forced to send reserves to try to stop the II Corps. The American forces had chiefly to avoid being caught in a weak situation during an enemy spoiling attack or by a counterattack provoked by an initial American success. They were not expected to advance southeast of Gafsa.

The plans of 18 Army Group for the II Corps prescribed a subsidiary role which naturally disappointed its more confident American officers. The directive in effect prohibited an American advance to the sea and envisaged a hesitating movement subject at all times to 18 Army Group’s approval, a program which indicated lack of confidence in the capacity of II Corps to execute a full-scale operation on its own responsibility. The higher echelons of command apparently considered the capabilities of the American units to be only partly developed. The February setback had revealed deficiencies and had presumably shaken the morale of participants. The forthcoming operations were therefore designed to permit small successes and the application of training lessons taught in battle schools instituted by 18 Army Group during the preceding fortnight. A few such victories, it was hoped, even though minor, would bring the performance of American units up to the required level by developing their capacities and fortifying their self-respect.

But the Americans, particularly the more aggressive ones like the new corps commander, General Patton, tugged against the restraining leash from the start. Both 18 Army Group and Army Group Africa thought of the area in which the II Corps would undertake its March offensive operations as the deep northwestern flank of the Axis forces in the coastal region south of Gabes. A potential Allied thrust from the Tebessa area had available two major avenues of approach. The main route ran through the mountain-walled valley, north of the chotts. This valley could be entered at Gafsa and followed to either a southeastern or northeastern exit. The second route lay just to the north of that valley, and was separated from it by the long mountain ridge between Gafsa and Mezzouna. It too was protected throughout its eastern half by a mountain screen, for north of the ridge between Gafsa and Mezzouna a loop of somewhat lower hills extended between Djebel Goussa (625), near Station de Sened, and Djebel Zebbeus (451 ), north of Maknassy. The Allied forces intending to proceed along either of these avenues would necessarily begin by taking Gafsa.

That oasis was the key. It lay in an exposed position, from a military standpoint, and had changed hands several times. When Field Marshal Rommel broke off his February offensive, he left there elements of Division Centauro which were linked with the main body of Fifth Panzer Army by small forward defense units in the Faid-Maknassy area, and with the Army Group Africa reserves nearer Gabes. Whether in Allied or Axis possession, Gafsa was near the outer end of a line of communications, and correspondingly vulnerable. Tebessa was eighty-four miles to the northwest, while Sbeitla, either via Feriana and Kasserine, or by way of Bir el Hafey, was almost as far to the north. Faid pass, on the other hand, was seventy-two miles northeast by the most direct highway, and Maknassy was fifty miles east-northeast. Although American troops were relatively familiar with part of this area, they knew the Gafsa-Gabes valley chiefly from the maps, on which important topographical features were inexactly represented. Initially, Operation WOP, as the II Corps designated the undertaking, did not provide for sending American troops, except defensive and reconnaissance forces, beyond Gafsa into this valley. The 1st Infantry Division was to make the attack on Gafsa, with the 1st Armored Division initially protecting the northeastern flank of the advance, while troops of the French Southeast Algerian Command were to operate on the other flank south of the line Metlaoui-Djebel Berda (926).

If II Corps should continue toward Maknassy, an advance contemplated as the second phase of the attack, its elements would be on both sides of the mountain ridge extending between Gafsa and Mezzouna. At its western end the bare and rugged slopes of Djebel Orbata (1165) rose abruptly to a crest of about 3,500 feet. The contours of this somewhat twisting ridge softened, and the crests were lower, along its eastern half. Trails through its deeply eroded gulches and defiles were narrow and few. Contact between the two forces on either side would be restricted to the barest minimum from Gafsa to Sened village, that is, about halfway to Maknassy, and from that point to the tip of Djebel Bou Douaou (753), five miles east of Maknassy, would be severely limited. Simultaneous attacks along both sides of the ridge would have to be relatively independent of each other.

The Corps Plan: Operation WOP

General Patton took over command from General Fredendall at Djebel Kouif on the morning of 6 March, after conferences with Generals Eisenhower and Smith in Algiers and with his new superior, General Alexander, and others in Constantine. Time before the operation was to begin was already short, and plans and preparations had to be expedited and co-ordinated. The tentative plans for II Corps, 1st Infantry Division, and 1st Armored Division, were considered in a commanders’ conference on 8 March 1943, and with some minor modifications, were then approved by the new commanding general. General Alexander and the Chief of Staff, 18 Army Group, General McCreery, spent the next two days at II Corps headquarters while inspecting the principal elements of the corps. D Day was set back from 15 March to 17 March, closer to the Eighth Army’s scheduled attack, by army group orders of 13 March, despite Patton’s apprehensions that the enemy might strike first. The II Corps’ role in the forthcoming army group operations was, Patton decided, to be like that of Stonewall Jackson in the Second Battle of Manassas. The corps was to conduct a flank battle which would assist Eighth Army to break through enemy lines, as Jackson had aided Longstreet’s corps.

[NOTE 6-29NA2: (1) Patton Diary, 13 Mar 43. (2) The II Corps staff then included the following: chief of staff, Brigadier General Hugh J. Gaffey; G-l, Colonel Samuel L. Myers; G-2, Colonel B. A. Dickson; G-3, Colonel Kent C. Lambert; G-4, Colonel Robert W. Wilson; artillery, Colonel Charles E. Hart; antiaircraft, Colonel Robert H. Krueger. The staff of the 1st Armored Division included: chief of staff, Lieutenant Colonel Grant A. Williams; G-l, Lieutenant Colonel Loris R. Cochran; G-2, Lieutenant Colonel M. M. Brown; G-3, Lieutenant Colonel Hamilton H. Howze; G-4, Lieutenant Colonel Percy H. Brown, Jr.; artillery, Colonel Robert V. Maraist. The staff of the 1st Infantry Division included: assistant division commander, Brigadier General Theodore Roosevelt, Jr.; chief of staff, Colonel Stanhope B. Mason; G-l, Lieutenant Colonel Charles Ware; G-2, Lieutenant Colonel Robert W. Porter, Jr.; G-3, Lieutenant Colonel Frederick W. Gibb; G-4, Lieutenant Colonel C. M. Eymer; artillery, Brigadier General Clift Andrus. • Patton Diary, 14 Mar 43.]

The Axis top command had recognized the Allied threat to the garrison at Gafsa but discounted it a few days before the attack, after Kesselring’s assurances to Mussolini that the place was in no danger. The force there was strong, he declared, and the approaches were very heavily mined. Kesselring, in the same spirit of optimism, also advised the Duce to expect a defensive success even at the Mareth Position, where the British attack was expected as soon as the moon turned full. General von Arnim thought well of a project to make a swift spoiling attack through Gafsa as far, perhaps, as Tozeur. At his prompting, General Hildebrandt went to El Guettar and on beyond Gafsa toward Feriana on the morning of 15 March, arranging tentatively with General Calvi de Bergolo for an attack by his 21st Panzer Division and part of Division Centauro to take place about 19 March, and to extend to Feriana and MetIaoul.

The II Corps on 15 March consisted of the 1st, 9th, and 34th Infantry Divisions, the 1st Armored Division, the 13th Field Artillery Brigade (Brigadier General John A. Crane) and the seven battalions of the 1st Tank Destroyer Group (Colonel Burrowes G. Stevens), which had been parceled out to the four divisions and to corps reserves, plus corps troops-in all, 88,287 men. [NOTE 6-11NA] Army group retained control of the 9th Division (minus Combat Team 60), the 34th Division, and the 13th Field Artillery Brigade.

For training, 75 officers and 175 enlisted men of the U.S. 2nd Armored and U.S. 3rd Infantry Divisions were attached to II Corps. British raiding parties from Eighth Army were active in front of II Corps behind the enemy’s lines, while in the general area of the road through Bir el Hafey to Sidi Bou Zid and in the mountains to the south, two squadrons of the Derbyshire Yeomanry under II Corps control engaged in energetic reconnaissance.

[NOTE 6-11NA: (1) 1st TD Gp Outline Plan WOP, as given in M5g 1355, 12 Mar 43, Entry 64, in II Corps G-3 jnl. This lists the following tank destroyer battalions: 6015t attached to the lst Infantry Division, 701st and i76th attached to the 1st Armored Division, 811th attached to the 14th Infantry Division, 894th attached to the 9th Infantry Division, 805th and 899th in II Corps [(‘serves. (2) II Corps AAR,2 May 43, App. A,]

In the seizure of Gafsa, General Allen’s reunited 1st Infantry Division was to be reinforced by the 1st Ranger Battalion and by several battalions of field artillery and tank destroyers. Elements of General Ward’s 1st Armored Division, with Combat Team 60 attached, had the initial mission of providing protection against Axis attacks from the directions of Sidi Bou Zid or Maknassy.

Meanwhile the two reinforced divisions completed preparations behind the main line of defense at the Western Dorsal, which General Ryder’s 34th Infantry Division held in the Sbiba sector and General Eddy’s 9th Infantry Division (less the 60th Combat Team), from Kasserine to EI Mael Abiod. General Allen’s command was scheduled to approach Gafsa via Feriana on the night preceding its 17 March attack. General Wards forces were to emerge through Kasserine pass and via Thelepte move to an assembly area east of Djebel Souinia (679), near the Gafsa·Sidi Bou Zid road.

The spring rains began falling heavily a few days before 17 March, filling the gullies, soaking the ground, and confining heavy vehicles to the roads. The II Corps sent a provisional detached armored flank force to the Sbeitla area under command of Colonel Clarence C. Benson, Commanding Officer, 13th Armored Regiment, on the night of 14-15 March, while Combat Command B of the 1st Armored Division started by daylight on 16 March to make certain of reaching its positions in time for the next day’s attack. The 19th Combat Engineers had in only three days graded a new dirt road from the vicinity of Thelepte directly to the Gafsa-Sidi Bou Zid road. This new route, named “the Welvert road,” was used by General Ward’s units. The 1st Engineer Battalion lifted some 2,000 mines along the routes of approach beyond Feriana used by General Allen’s attacking elements.

The Occupation of Gafsa

General Allen’s plan for taking Gafsa, in this first operation by his whole division against Axis troops, was thorough. Division intelligence estimates of the defending forces likely to be found there ran to only two battalions of infantry, one or two of field artillery, up to two companies of tanks, and a military police battalion-all Italian. The enemy troops at Gafsa could summon reinforcements from El Guettar (one infantry battalion), and from the stations along the railroad to Maknassy, or from the Faid area.

Within one day, troops from as far away as Gabes could be brought to Gafsa, provided they were not needed for the impending Mareth battle. But the enemy garrison at Gafsa was expected only to delay the American advance and then to fall back to prepared positions in the mountains either toward Gabcs or toward Maknassy. No enemy counterattack was deemed likely unless the Mareth Position was abandoned. Indeed, road traffic observed just before the day of the attack indicated that most of the Gafsa force had already withdrawn.

Following a half hour’s air bombardment, regimental combat teams of the 16th (Colonel d’ Alary Fechet) and 18th Infantry (Colonel Frank U. Greer) and one reinforced battalion of the 26th Infantry (Colonel George A. Taylor) were to make the assault. Five complete battalions of field artillery and one battalion of antiaircraft artillery were to be employed. The 1st Ranger Battalion would be ready, after first supporting the cast flank, to reconnoiter to El Guettar and to seize an area from which, subsequently, Combat Team 26 might attack beyond that village. Following the assault, the 601st Tank Destroyer Battalion was to be ready to intervene south of Gafsa against any enemy mechanized forces that might arrive there next day. The 1st Armored Division would hold one combat command of three battalions ready to furnish support from the Station de Zannouch area. Participation by the 13th Field Artillery Brigade and by the 1st Battalion, 213th Coast Artillery (AA), was to be under corps control.

In preparation for the attack the 1st Infantry Division made an approach march of about forty-five miles during the night of 16-17 March, one that was executed speedily and on a close schedule. The attacking elements detrucked north of the Gafsa area before daylight. The artillery was ready to support an assault at 0800, but the infantry was held back until 1000, despite the enemy’s visible retreat, to await a scheduled preparatory air bombardment. By noon, 17 March, the 18th Infantry had reached the eastern edge of Gafsa; shortly thereafter Company I, 16th Infantry, was leading the 3rd Battalion, 16th Infantry, into the village from the northwest; and the 3rd Battalion, 26th Infantry (reinforced), was near the western limit. They had overrun enemy security detachments and although mines and booby traps were plentiful, they found Gafsa to be free of defenders. While 1st Armored Division reconnaissance entered Station de Zannouch northeast of Gafsa on the route to Maknassy, reconnaissance and outpost forces of the 1st Infantry Division continued toward EI Guettar and Djebel Mdilla. At Mdilla they made contact with some French troops, after initially mistaking them for an enemy group. The seizure of Gafsa was an encouraging exercise rather than a hard battle like those in which the 1st Infantry Division was later to earn fame.

During 18-19 March, a period marked by drenching rainstorms, the entire division occupied the place and organized for defense against counterattack or air bombardment. At the same time the 1st Ranger Battalion occupied El Guettar on 18 March and sent patrols to establish contact with the enemy who was holding prepared defense positions behind the Keddab wadi.

The Seizure of Station de Sened

The situation at Gafsa justified General Patton on 18 March in concluding that the second phase of the II Corps’ attack could be undertaken next day. While the 1st Infantry Division organized Gafsa strongly for defense, the 1st Armored Division (reinforced) could be committed to the seizure of Station de Sened. The bulk of General Ward’s division was already in areas selected with a view both to defense against incursion from the northeast and to the concentration needed to attack Station de Sened.

Although some elements, and in particular Benson’s armored task force directly under II Corps control, were in the vicinity of Sbeitla, and Combat Command B (Robinett) was near Bir el Hafey, Combat Command A (McQuillin) was in the Zannouch area and Combat Command C (Stack) with the 60th Combat Team (de Rohan) were southeast of Djebel Souinia. It was General Ward’s intention to bring Combat Command A northeastward along the railroad line from Gafsa toward Station de Sened in conjunction with an approach by Combat Command C and Combat Team 60 over hills north of the objective. But if the military situation near Gafsa permitted an immediate start of the second undertaking, the weather made postponement unavoidable. Much against his wishes, General Patton was forced to accept the fact that the mud had made an armored attack on 19 March out of the question.

[NOTE 679NA2: (1) II Corps AAR, 10 Apr 43. (2) Patton Diary, 19 Mar 43. (3) Msg, Entry 181, in II Corps G-3 J nl. This message reports the presence at Sbeitla at 1930, 14 March, of the following units of the 1st Armored Division: the 3rd Battalion, 13th Armored Regiment (mediums); the 68th Field Artillery Battalion; Company A, 701st Tank Destroyer Battalion; Company C, 16th Armored Engineers (C). By an agreement between Generals Ward and Eddy, the 84th Field Artillery Battalion was to remain at Sbeitla. The 3rd Battalion, 47th Infantry (9th Division), was already there. The 81st Reconnaissance Battalion moved to the area east and southeast of Sbeitla.]

Streams were full to overflowing. The earth was soggy and in many places there were extensive shallow pools. Bivouac areas were flooded. The soft roads were quickly cut into deep ruts by heavy trucks or churned into a viscous blanket by tank tracks. Travel cross-country became impossible for wheeled vehicles. Indeed, to assist them in reaching the roads from their parks required extraordinary effort and much extra time. The weather’s one compensation was the fact that it restrained enemy air activity.

While most of the 1st Armored Division remained immobilized on 19 March, Patton drove through the downpour to review the situation with Ward and Robinett. He was enthusiastic and confident, concerned only that the enemy should not be given opportunity for a spoiling attack while the Americans waited for conditions to be wholly satisfactory. He would have preferred to have the attack on Station de Sened made by as much infantry and artillery as could be moved on tracked vehicles rather than to wait for full co-ordination between the elements approaching Station de Sened from the north and McQuillin’s armored force from Zannouch. Ward’s orders for the attack to be made early on 20 March were issued to McQuillin, de Rohan, and Stack, while Robinett shifted his forces southwestward to the divisional assembly area, and Benson, under II Corps’ control, took up the position north and east of Djebel el Hafey (682) thus vacated by Combat Command B.

The plan of attack involved a march extremely taxing for de Rohan’s combat team. It was to climb the western slopes of Djebel Goussa to reach the dominating terraces along its southern face, which rises abruptly some 600 feet above the floor of the valley, in order to take the enemy’s hill positions from the rear. From the heights, the attacking force could command the entrenched positions which the enemy had constructed on the flats below them near Station de Sened. At the same time, Combat Command C would be crossing a difficult series of ridges and shoulders at the southwestern extremity of the Djebel Madjoura (874) across a valley from Djebel Goussa, protecting the northeastern flank and giving fire support. Its objective was the exit (three miles north of Station de Sened) from this valley to the broad Maknassy Valley, Working together, the 60th Combat Team and Stack’s force would be able either to cut off the enemy force defending Station de Sened or to compel it to hasten its retreat in order to avoid encirclement.

McQuillin’s armored force, which would at the same time attack the enemy with infantry and artillery from the west, might be the beneficiary of the outflanking movement by Stack’s and de Rohan’s commands, Combat Team 60 (9th Infantry Division) then consisted of: the 60th Infantry Regiment; the 60th Field Artillery Battalion; Company C, 15th Engineers: Company C, 9th Medical Battalion; three platoons, 443rd Coast Artillery (AA) Battalion (SP): detachment, 9th Signal Company; Provisional Truck Company, Headquarters, 60th Infantry.

[NOTE 6-100NA: (1) Combat Command C, 1st Armored Division, then consisted of the following: the 6th Armored Infantry (less the 2nd and 3rd Battalions): the 1st Battalion, 13th Armored; the 68th Field Artillery Battalion; Company B, 16th Engineers. It was reinforced for later operations east of Maknassy } the 3rd Battalion, 60th Infantry; the 58th Field Artillery Battalion: and Company B, 776th Tank Destroyer Battalion. (2) CCC 1st Armd Div AAR, 18 Apr 43. (3) 60th Inf Hist, 1943.]

General Ward’s plan for the seizure of Station de Sened worked in general as he had foreseen. The 60th Combat Team approached Djebel Goussa during the night of 19-20 March and throughout the next day pushed to the heights while Combat Command C, 1st Armored Division, was gaining its objective somewhat earlier and holding positions from which to support Combat Team 60, if necessary. By evening of the same day Combat Command A had struggled from Zannouch to a mine field west of Station de Sened. The enemy was driven out by artillery fire, some fleeing northeastward at dawn of 21 March into the path of Combat Command C, which took seventy-nine prisoners and two 47 -mm. guns after a brief engagement. Another part of the garrison took refuge at Sened village, where it finally surrendered on 23 March to Company G, 60th Infantry, supported by elements of the 91st Field Artillery Battalion. A few escaped from Sened to Sakket on the other side of the mountains only to be captured there by troops of the 1st Infantry Division. Station de Sened had thus been taken on 21 March by maneuver rather than by storm, and without the losses normally to be expected in a frontal attack. But the exertion left the infantry, particularly of the 60th Combat Team, too tired to begin another attack effectively.

The Advance Beyond Maknassy

The capture of Gafsa and Station de Sened left only a demonstration to be made toward Maknassy, twenty miles farther east in order to complete execution of 18 Army Group’s original instructions to II Corps. But those instructions had already been changed. On 19 March, Patton returned to his headquarters in Feriana after his rain drenched visit to the headquarters of the 1st Armored Division to find General McCreery, Chief of Staff, 18 Army Group, with General Alexander’s new plans and orders for II Corps. The Corps was now to seize the high ground east of Maknassy and to send a light armored raiding party to the Mezzouna airfields to destroy enemy installations there. No large forces, however, were to pass beyond a line extending from Gafsa through Maknassy heights, Faid, and Fondouk el Aouareb. Later, after the British Eighth Army had passed up the coast beyond Maknassy, the II Corps was to be reduced by the transfer of its U.S. 9th Infantry Division to relieve the British 46th Division on the far northern flank and to operate under British 5 Corps within British First Army. The 34th Infantry Division would at about the same time sideslip to the north in order to attack Fondouk el Aouareb gap along the axis Maktar-Pichon.

It thus appeared that after the enemy had moved north of Fondouk el Aouareb, the II Corps would be faced with the ignominious prospect of being pinched out of the Allied line. These instructions would not only prohibit any American advance to the sea but would confine the role of II Corps to merely threatening the enemy’s western flank without ever actually attempting to cut him off; they would also prevent the corps, except for 9th Infantry Division, from participating in the last stage of the campaign. In accordance with orders, the II Corps sent attacking forces not only to Maknassy but to a defile east of El Guettar, on the southern side of Djebel Orbata, the purpose being- to draw off enemy troops which might otherwise strengthen the defense of the Mareth Position. There the main attack was to begin, it will be remembered, on the night of 20-21 March.

On 21 March General Patton drove to General McQuillin’s command post in order to hurry Combat Command A to a hill mass five miles northeast of Station de Sened which appeared to the corps commander a possible place of advantage which must be denied to Maknassy’s defenders. At the same time, Combat Command C moved northeastward, along a camel trail, and then swung south to reach the main route from Station de Sened to Maknassy at a point about halfway between the two places. For a stretch, Combat Command B followed, but instead of turning south, continued eastward in the valley to an area from which to guard the northern flank of the attack on Maknassy, and assist in preparatory artillery fire on the village. The exhausted troops of the 60th Combat Team, meanwhile, assembled just north of Sened station.

Advance elements of Combat Command C, 1st Armored Division, approached Maknassy before midnight and subjected the place to an interdictory shelling, hoping to discourage the enemy from laying mines and booby traps before withdrawing. Colonel Stack received reinforcements during the night and disposed these troops for the approaching action. Not until 0715 next morning did reconnaissance discover that Maknassy was free of the enemy, whom some of the inhabitants declared to have withdrawn onto the hills near the road to Mezzouna, east of the village.

General Ward was then faced with a critical choice. He could attempt to occupy the hills five miles east of Maknassy in a daylight attack without waiting to reorganize or to prepare for stiff resistance. Enemy aviation could strike from airfields only a few minutes away, as it had during the battles along the Medjerda river in November and December. If General Ward waited until fully ready, the enemy might be able to strengthen his hold on the hill position controlling the exit from the Maknassy to the Mezzouna side of the mountains, so that the effort to dislodge him would be difficult and costly. The incentive to take great risks was slight, because of the orders the Americans were then operating under to threaten the Axis line of communications but not to commit any large force beyond the hills. Ward decided to make a deliberate, soundly organized attack. Although there is a certain amount of inconclusive evidence that this choice was approved at the time by General Patton, the stronger evidence is to the contrary, and Patton was later to conclude that the choice had been founded upon considerations which were unduly defensive in character, and to condemn himself for not having gone forward that day to urge on the advance.

18 Army Group Revises II Corps Mission, 22 March At this point, the role of the II Corps was again modified. On 21 March General Montgomery, when he recognized that the Eighth Army would be engaged for several days in trying to breach the main Mareth Position near the coast, and before he decided to shift the principal effort to the El Hamma gap, suggested to General Alexander that the U.S. II Corps could be of substantial assistance by a strong armored thrust through Maknassy to cut the Sfax-Gabes road. At 18 Army Group, such a project was considered to be too ambitious, particularly in view of the likelihood that the 10th Panzer Division would intervene during its execution. But General Alexander issued instructions to II Corps on 22 March to prepare for a possible effort to disrupt the enemy’s line of communications and destroy his supply dumps southwest of Mahares. General Patton was to make ready a strong mobile column for such a mission.

Meanwhile, the limited missions of the 1st Armored Division east of Maknassy and of the 1st Infantry Division east of EI Guettar remained unchanged. Only the tempo was accelerated somewhat. That afternoon General Patton gave oral instructions to General Ward to seize the heights from Meheri Zebbeus, north of Maknassy, to Djebel Bou Douaou, southeast of it; to organize and occupy that ground; to send a light mobile armored force to raid the Mezzouna airfields; and, in addition, to prepare a second armored force and keep it in readiness to harry the enemy’s lines of communications in the vicinity of Mahares, more than fifty miles east of Maknassy on the coast. Patton directed that the attack be made that night.

The enemy had already recognized the operations by Eighth Army and U.S. II Corps as concentric attacks upon General Messe’s First Italian Army. Kesselring decided to commit the reserves of the Fifth Panzer Army in holding the heights east of Maknassy and to release the 10th Panzer Division from Army Group Africa reserve, for a counterattack toward Gafsa under the control of General Cramer’s Headquarters, German Africa Corps. Colonel Dickson, G-2, II Corps, correctly concluded on 22 March that an attack by the 10th Panzer Division was imminent, either at Maknassy or El Guettar, and General Patton acted accordingly.

The day’s reconnaissance on 22 March confirmed the reports by inhabitants of Maknassy village that the enemy had retreated into the hills, or beyond. The main ridges of the Eastern Dorsal at this southeastern extremity are in the pattern of a large figure 5, with Maknassy directly south of the vertical stem and about five miles from the heights which form the curving section.

Djebel Zebbeus (451) and Djebel Djebs No.1 (369) are north of Maknassy. Djebel Dribica (209), Djebel Naemia (322), Djebel Bou Douaou (1753), and Djebel Bou Hedma (790) form the semicircle, with the latter’s long ridges extending far to the southwest of Maknassy. Between Djebel Djebs No.1 and Djebel Dribica, the Leben wadi, crossed by a railroad and by a highway bridge north of the village, drains from the Maknassy plain to the coastal flats. With its tributaries, the Leben wadi forms a lengthy tank obstacle north and northwest of Maknassy. Djebel Dribica leads southeasterly to Hill 322 on Djebel Naemia, with which it is connected by an L-shaped ridge.

The road and the railroad from Maknassy to Mezzouna and Mahares run side by side over the southern shoulder of Djebel Naemia, except for a distance of nearly two miles through an opening between Hill 322 and a second Djebel Djebs, where the railroad loops less than a mile south of the road. This Djebel Djebs No.2 (312) rises east of a broader gap between Hill 322 and Djebel Bou Douaou, like a stopper barely removed from a bottle, and is admirably situated to control movement through the defiles northwest or southwest of it. Djebel Bou Douaou and Djebel Bou Hedma extend south and west with crests somewhat higher than Maknassy’s other neighboring hills and with good observation of the “Gumtree road” running along their southern bases from El Guettar to Mahares.

Combat Command B’s reconnaissance observed a few groups of enemy vehicles pulling back through the gaps northeast of B, 1st Armored Division, on reconnaissance, Maknassy valley, MAknassy, and drew artillery fire from the yicinity of Meheri Zebbeus; but along the route of the road and railroad, the enemy kept out of sight.

Enemy Defense of Maknassy Pass,23-25 March

At 1415 on 22 March General Ward issued written orders for an assault at 2330 that night. a2 He specified that Colonel Stack’s forces should attack Djebel Dribica and Hill 322 north of the pass and Djebel Djebs (2), beyond and southeast of it, while Combat Team 60 (-) simultaneously gained control of Djebel Bou Douaou. He directed Combat Command B (General Robinett) to protect the northern flank in the vicinity of Djebel Zebbeus and the upper reaches of the Leben wadi. Of the two bare and rocky hills north of the pass, Djebel Dribica seemed the more imposing obstacle and was assigned to the more experienced 1st Battalion, 6th Armored Infantry (Colonel Kern), while Hill 322 on Djebel Naemia was to be taken by the 3rd Battalion, 60th Infantry (Lieutenant Colonel John J. Toffey, Jr.), attached to Combat Command C. This battalion was still weary from climbing Djebel Goussa near Station de Sened, and from trying for most of the day to work through traffic-clogged roads to the assembly area after a night move toward Maknassy.

The attack beyond Maknassy was too late and too weak. The two battalions under Colonel de Rohan, to be sure, met no opposition in securing Djebel Bou Douaou, and next morning were ordered to occupy the area north of it as far as the road and railroad, and to assist by fire the southern wing of Combat Command C. On the northern flank after a three-battalion artillery preparation of thirty minutes duration, the 1st Battalion, 6th Armored Infantry, took Djebrl Dribica in competent fashion against only light opposition. But the 3rd Battalion, 60th Infantry, found that Diebel Naemia, with Hill 322 in the center, was by far the most strongly defended area. A mine field barred the approach to the crest, and was covered by concentrated fire from well placed machine guns both on the hill ahead and on adjacent slopes to the right and left. One company commander was killed. The battalion commander went forward to reconnoiter and was pinned down. The attack stopped. The men dug in to wait for daylight.

Next morning, 23 March, the 3rd Battalion, 60th Infantry, pressed forward again. This time the attack was supported by the 1st Battalion, 6th Armored Infantry, and by fire from the remainder of Combat Team 60, across the road. ] just as the enemy was beginning to fall back in the belief that the Americans were breaking through, Colonel Lang reached the battlefield and was able to make the small German force hold on until the first reinforcements arrived. At about the same time, on the Allied side, Colonel Toffey was wounded. The German force that held the vital Hill 322 was Rommel’s former personal guard. At the time of the imminent break-through it had been reduced to only eighty infantrymen. At another critical juncture the covering fire of a few tanks aided the enemy to occupy positions along the eastern edge of Djebel Dribica. Hill 322, dominating the pass, remained the objective of a succession of American attacks that evening, including one supported by four artillery battalions. Success seemed so likely that routes were reconnoitered for a light armored force preparing to raid the Mezzouna airfields during the night of 23-24 March.

After the first attacks were stalled on Hill 322, a stronger force concentrated for an assault at 0700, 24 March. Three battalions of infantry, supported by two companies of tanks and four battalions of artillery, plus some 75-mm. tank destroyers, attacked the defenders from the north, west, and south. The 3rd Battalion, 60th Infantry, on the west, and the 3rd Battalion, 6th Armored Infantry (less Company I), on the south, received direct supporting fire from a company of medium tanks, while the 1st Battalion plus companies G and I, 6th Armored Infantry, was protected by others on its eastern flank as it attacked along the ridges from the north.

[NOTE: The reinforcements hastily thrown into the threatened sector were elements of Kampfgruppe Lang (Colonel Lang, commanding officer of the 69th Panzer Grenadier Regimen(. 10th Panzer Division).These German units were nominally under General Imperiali’s 50th Special Brigade. See Fifth Panzer Army, KTB, 23 Mar 43, and is # D–166 (Lang), Part II.]

Throughout the night the enemy had been digging in while small groups of reinforcements built up his total number to an estimated 350. The Americans, their difficulties greatly increased because of the enemy’s excellent air-ground co-operation, were unable to dislodge these troops after hard ground fighting. Continued failure to gain Hill 322 threatened to frustrate that part of the mission assigned to II Corps which appealed most to General Patton. He had protested to General Eisenhower the prohibition On American advance beyond Fondouk el Aouareb. He probably counted on the raiding party from Maknassy to Mezzouna to demonstrate that if such an advance were authorized, it could lead to even bigger successes, and for that reason had instructed General Ward to prepare for an as yet unauthorized aggressive action in the area near Mahares. 

[NOTE: (1) Kampfgruppe Lang, sent to this area by Fifth Panzer Army on 22 March 1943, began arriving next day, and by 26 March consisted of: Regimental Staff and 1st Battalion, 69th Panzer Grenadier Regiment; 1st Battalion, 86th Panzer Grenadier Regiment; 26th Africa Battalion; nine Mark VI plus fifteen Mark III and IV tanks of 501 st Panzer Battalion; 580th Reconnaissance Battalion; about thirty men from Kasla O. B. (Rommel) and one company of Kesselring’s headquarters guards, and of artillery, 9th Battery, 90th Artillery Regiment; one battery of Italian 105’s; two bat· teries of 210-mm. mortars supported by one 170-mm. gun; two batteries of 88-mm. dual-purpose Flak guns; and one platoon of 20·mm. antiaircraft guns. Hq, Fifth Panzer Army, KTB, 26 Mar 43. MS # D-166 (Lang), Part II. (2) Lang was correctly identified and his units ascertained in the 18 Army Group Appreciation, 2000, 25 March 1943. (3) Kampfgruppe Lang passed to the command of Headquarters, DAK (Cramer), on 28 March 1943. “Rpt, 1st Armd Div to II Corps, 1420, 24 Mar 43, Entry 211, in II Corps G-3 Jnl. ‘” (1) Patton Diary, 22 March 1943, says that he sent General Bradley to Algiers on this mission. (2) Bradley, A Soldier’s Slor}’, pp. 56-58, says that Bradley obtained Patton’s permission to go.]

He returned from the greatly extended 1st Infantry Division, where he had spent much of 24 March, to discover that Ward’s attack on the narrower front east of Maknassy was still unsuccessful. The tanks had thrown their tracks on the rough and rocky ground and had been of minor service. The infantry had lacked the impetus to carry the attack through to success. One estimate of the time still needed to get the pass was twenty-four hours. The enemy was described as “obviously moving to concentrate” on the 1st Armored Division’s front and as stepping up the pace of his frequent air attacks. General Ward made a request for “all available air cover for our troops moving up now, and for active reconnaissance, and strafing of the enemy moving up on our front this p. m. and tomorrow.

Calling General Ward to the telephone, General Patton ordered him personally to lead an attack next morning which had to succeed. The orders were partially carried out, for the division commander did lead an attack by three battalions of the 6th Armored Infantry, which began without artillery preparation, and won a brief success.

But they could not hold the hill under heavy fire from German mortars, machine guns, and artillery. By noon, 25 March, General Ward decided that he must suspend the attack while the troops recovered from near exhaustion and reorganized. In view of the enemy’s decision to concentrate mobile armored elements against the weakening American forces of the 1st Infantry Division east of El Guettar, II Corps ordered a provisional armored protecting force under Colonel Benson to be shifted to the vicinity of Gafsa. H The initiative at Maknassy was allowed to pass to the enemy while the situation east of El Guettar was being corrected.

The 1st Infantry Division Holds the Enemy Near El Guettar

The 1st Infantry Division’s part in the operations specified by 18 Army Group’s instructions of 19 March was an attack east of El Guettar along the Gumtree road south of the Gafsa-Melzouna mountains, to be launched after the 1st Armored Division had neutralized Station de Sened. This drive easterly from El Guettar was to be made down the great valley on a front which furnished a severe test of tactical efficiency and which broadened beyond the capacities of a single infantry division to defend or control.

From Gafsa to El Guettar, the road curves around the southwestern portion of Djebel Orbata (1165) and runs along its southern base past the Chott el Guettar to a road fork some twelve miles distant from Gafsa. At this fork, the road to Gabes branches off to the southeast, passing between the Djebel Berda (926) complex on the south and jumbled hills at the north which rise to the horseshoe-shaped Djebel cl Mchehat (482). The other branch, Gumtree road, continues along the southern base of Djebel Orbata and the ridge of which it is a part, and strikes across the coastal plain north of the Sebkret en Noual to Mahares on the sea coast. Three miles east of the El Guettar road fork, the Gumtree road enters a long narrow defile between a spur of the Djebel Orbata and the Djebel el Ank (621). Near the eastern end of the defile are the village of Bou Hamran and the junction with a track which leads through the mountain hamlet of Sakket and over the ridge to the village of Sened.

[NOTE: H ( 1) MS)2;, Col Akers to Col Benson, 2000, 21 Mar 43, Entry 56, in II Corps G~3 Jnl. (2) II Corps AAR, 24-25 Mar 43, 10 Apr 43. The provisional force, consisted of a battalion of medium tanks, a battalion of artillery, and a battalion of infantry.]

The 1st Infantry Division was expected to capture the defile at Djebel el Ank and to press eastward along the Gumtree road. It could not advance very far without becoming vulnerable to an attack up the road from Gabes on the southern side of Djebel cl Meheltat, an attack which could strike the division on the flank and either cut it off from Gafsa or force a withdrawl The road to Gabes was connected with the Gumtree road by a dirt road through El Hafay at the northeastern end of Djebel Chemsi (790) some twenty-five miles east of EI Guettar. A force from either Gabes or from a point much farther north could therefore approach El Guettar from the southeast, via the Gumtree road and its El Hafay connection with the Gabes-Gafsa road. The 10th Panzer Division, once it had assembled, could strike either at Maknassy or toward El Guettar and, if the latter, then by either the Gabes road or the Djebel el Ank defile.

After a day of reconnaissance, observation, and preparations, the 1st Infantry Division opened its attack on the night of 20-21 March. To capture the defile on the road to Mahares, the 1st Ranger Battalion was sent on a circuitous and difficult ten-mile detour over the shoulders of Djebel Orbata which brought it back to the road at a point east of the Italian defenders for a surprise attack from the north same time, the 26th Infantry, aided by artillery preparation and the cover afforded by foothills and great rocks, pushed straight along the road, closing in to gain a complete victory. The Americans took over 700 prisoners and after converting captured positions for defense toward the east, continued the attack in order to gain control over all of Djebel el Ank before going on to Bou Hamran.


The 18th Infantry’s attack moved southeastward along the road from El Guettar to Gabes and into the adjacent hills against dements of Division Centauro. At night, and under a hazy moon, the infantry crossed a plain where mere six-inch grass was the main cover, and where the enemy had already demonstrated that his observation in daylight was alert and accurate, and his artillery fire swift and precise. Getting through a mine field, the troops infiltrated past Hill 336 north of the road to take it at daybreak with a rush from the rear. General Patton and some of his staff visited the hill soon afterward. The first stage of the attack “according to plan” yielded 415 prisoners. During the remainder of 21 March, the 18th Infantry, limited by the enemy’s rapid laying of artillery fire on moving men, and by very active air strafing and dive bombing, particularly of command post and artillery positions, strove to press through foothills onto the Djebel el Mcheltat. Battery A of the 33rd Field Artillery Battalion lost two 105’s and two vehicles in a direct hit. At noon, 21 March, the divisional objective was established as a line from a crest three miles northeast of Bou Hamran to a point about fourteen miles to the southeast of El Guettar, a road junction east of Djebel el Kheroua (369). Casualties by nightfall were fourteen killed and forty-one wounded.


The Americans during the next day consolidated their gains east of El Guettar, for the Italians made no resolute counterattack and were driven off by artillery fire and Allied air action whenever their guns opened up or small groups of their tanks assembled. American advance was not as rapid as had been expected. The troops improved their positions along the Gumtree road by having the 26th Infantry probe east of Bou Hamran. The 16th Infantry occupied the western foothills of the Djebel eI Mchehat, and two battalions of the 18th Infantry occupied heights south of the Gabes road, at the northeastern tip of Djebel Berda. Contact with the enemy in this area late in the day (22 March) promised action of about the same tempo for the following morning.”

The operations by II Corps had been so well correlated with attacks by British Eighth Army as to make 22 March a critical day for the enemy. The seizure of Gafsa on 17 March had coincided with initial British exploratory attacks in the Mareth coastal sector. The capture of Station de Sened on 20-21 March had occurred while British 30 Corps began trying to punch through the M1areth Line and when the New Zealand Corps was about to seize Hill 201 south of El Hamma. By the time the 1 st Armored Division reached Maknassy on 22 March and the 1st Infantry Division pushed beyond El Guettar, the enemy was committing some of his reserves near Mareth and sending others to the El Hamma gap. He had only one mobile division available with which to try to check the American approach to the Gabes area, the 10th Panzer Division (- ) (General Fritz von Broich).

On 21 March, as noted, Army Group Africa released the 10th Panzer Division to the German Africa Corps for employment in a counterattack toward Gafsa. It was not apparent to the Allies where this attack would be made, although they were aware that it was building up. Actually, the German force assembled during the night of 22- 23 March near Djebe1 Ben Kheir (587, east of El Hafay ) . At 0300 the 10th Panzer Division attacked along the axis of the Gabes- Gafsa road to push northwestward against the southern flank and perhaps the rear of the 1st Infantry Division before daylight.

Reinforced American Infantry Versus German Armor, 23 March General Allen that night had the 26th Combat Team advancing along the Gumtree road, while the 1st and 2nd Battalions, 18th Infantry, were engaged, as they had been since the preceding afternoon, in driving out Italian forces northeast of Hill 772 on Djebel Berda, thus widening the front considerably. The trains of the 2nd Battalion, 18th Infantry, barely reached the cover of the foothills, after crossing in darkness from the northern side of the valley, when sounds of motors to the northeast were followed by an eruption of tracer fire and the echoing rumble of guns. At 0500, while darkness was still complete, an enemy motorized force was reconnoitering by fire the southern slopes of Djebel el Meheltat.

Daylight revealed the presence of the 10th Panzer Division, which was methodically sweeping the foothills and the lower ground north of the road before undertaking bolder measures. The 3rd Battalions of both the 18th and 16th Infantry were under direct attack. A spearhead moving up the valley was engaged by the 60 1st Tank Destroyer Battalion until about 0700. The main body of the enemy force was in full view of American observers on heights above the valley on either side, and from the German Africa Corps’ command post on Hill 369.

At first the battle ran entirely in favor of the attacking Germans despite determined and courageous opposition. Their tanks and self-propelled guns, interspersed with infantry in carriers, rolled westward in a hollow square formation and at a slow but steady pace. Behind them, a column of trucks drove to a predetermined point at the western end of Djebel el Meheltat and unloaded more infantry, which followed closely the armored rectangle ahead of them. Then the mass of the enemy separated into three prongs. One group turned northwest among the foothills east of Hill 336 overrunning the 32nd Field Artillery and part of the 5th Field Artillery Battalions; another continued along the road; and the third, and much the largest, force tried to sweep the hills and northward along the edge of the Chott el Guettar.

German tank-infantry teams overran the American artillery and infantry positions east of Hill 336 in engagements which brought some hand-to-hand encounter, and heavy American losses. A curving belt of mines extended from Chott el Guettar across the road and along the Keddab wadi to the southeastern base of Hill 336. There the tide of battle changed. American artillery and the tank destroyers of the 601st and 899th Tank Destroyer Battalions knocked out nearly thirty enemy tanks, and the mine field stopped eight more. Eventually, the morning attack was contained. The 10th Panzer Division pulled back a few miles to the east and prepared for a second attack. During this withdrawal enemy artillery and aviation harassed the American defenders, and Allied air units struck back repeatedly. The Germans towed their disabled tanks to a prepared maintenance point not far from where their infantry had first detrucked. During this interlude, and running a gantlet of enemy shells and Stukas, nineteen American jeeps rushed back for ammunition, all but six returning safely in time to oppose the next assault. Elements of the 16th Infantry and the 1st Ranger Battalion were put into the line along the Keddab wadi. Headquarters, 1st Division, was ready for the second attack (1645 hours) and aware of the enemy forces to be committed.

[Note: The 601st Tank Destroyer Battalion lost twenty-four guns, of which two were recoverable, and had only nine others after this action. The 899th Tank Destroyer Battalion lost seven tank destroyers (M10’s). ]

Preceded by a German air strike, and on the signal of a siren, the ground troops of the enemy attacked once more toward El Guettar. At 1830, word came that they were still advancing, with thirty-eight tanks in one group, but barely fifteen minutes later, an exultant report arrived from the 18th Combat Team: Enemy attacked as scheduled, preceded by dive-bombers which did little damage. Troops started to appear from all directions, mostly from tanks. Hit Anti-Tank Company and 3rd Battalion. Our artillery crucified them with high explosive shells and they were falling like flies. Tanks seem to be moving to the rear; those that could move. 1st Ranger Battalion is moving to protect the flank of the 3rd Battalion, which was practically surrounded. The 3rd Battalion and the Rangers drove them off and the 1st Battalion crucified them.

 [NOTE: (I) The enemy units were: 2nd Battalion, 69th Panzer Grenadier Regiment; 2nd Battalion, 86th Panzer Grenadier Regiment; 10th Motorcycle Battalion; 1st and 2nd Battalions, 7th Panzer Regiment; and the 4th Battalion, 90th Artillery Regiment.]

Thus at the close of 23 March, a reinforced American infantry division, well supported by Allied aviation, mainly by their artillery had stopped the bulk of an enemy armored division. Kesselring’s report to the OKW acknowledged substantial tank losses in the morning attack at the Keddab wadi; it called the efforts of the 601st and 899th Tank Battalions a counterattack with tanks against the German north wing which had been repulsed, and it attributed the failure of the afternoon attack to superior Allied forces and a threatened penetration through the Italian-held positions on the northern edge of Djebel Berda. Although the armored counterattacks of 23 March were beaten off, the enemy by no means lost his determination to maintain an aggressive defense of the routes of Allied approach to the rear of Italian First Army. He might not succeed in plunging through the Americans with his armor, but he held strong defensive positions and could nibble incessantly with infantry and artillery, and with tanks used as artillery, at the American positions in the hills north and south of the Gabes and Gumtree roads.

On 24 March he made some progress in each sector, especially in the high ground on opposite sides of the Gabes road, and on 25 March, he succeeded in recapturing from the 1st and 2nd Battalions, 18th Infantry, their most exposed position on one of the northeastern buttresses of Djebel Berda. When a German patrol reached the bare summit of Hill 772 there, mortar fire brought it scampering down. But the two battalions of the 18th Infantry, even with the 1st Ranger Battalion, managed only to hold their position; they could not extend it without larger reinforcements, perhaps another entire regiment. With that much strength, they believed that they could take all of Djebel Berda and the hills east of it, and thus open the road to Gabes for American armor. Had such a regiment been sent, or had the two battalions simply remained in possession, the Germans might not have been able to withstand the 9th Infantry Division’s efforts later to drive them off. But during the night of 25-26 March the battalions were ordered to withdraw through the 1st Ranger Battalion. Colonel Darby’s Rangers, with a purely defensive role, held a south flank position in the foothills west of Djebel Berda for the next two days.

SOURCE: Northwest Africa: Seizing The Initiative In The West; by George F. Howe (United States Army Center of Military History)

World War Two: North Africa (6-27); From Mareth to Enfidaville – 8th Army

World War Two: Munda Trail (9); XIV Corps Offensive

The plan for the XIV Corps’ drive against Munda was completed shortly after Griswold took over. Colonel Eugene W. Ridings, Griswold’s assistant chief of staff, G-3, flew to Koli Point to confer with General Harmon and Admiral Wilkinson on naval gunfire and air support. Ridings also asked for, and obtained, a better radio (SCR 193) for Liversedge, to improve communications between him and Griswold. Harmon stressed the importance of submitting a precise plan for air support to Admiral Mitscher. Dive bombers would naturally be the best for close work, while mediums and heavies should be used for area bombing, he asserted. Harmon agreed to send in more tanks at Griswold’s request.
[4N2 It was at this conference that he approved the transfer of the 161st Regimental Combat Team to New Georgia.]

The American Plan

Naval support plans called for a seven-destroyer bombardment of Lambeti Plantation shortly before the infantry’s advance. Commander Arleigh A. Burke, the destroyer division commander, came to Rendova on 23 July to view Roviana Lagoon and select visual check points. Air support for the offensive would include, besides the normal fighter cover, pattern bombing by multiengine planes in front of the 43rd Division about halfway between Ilangana and Lambeti Plantation. Single-engine planes would strike at positions north and northeast of Munda field. Artillery spotting planes and liaison planes would be on station continuously.

Artillery support would be provided by Barker’s artillery from its island positions. Plans called for fairly standard employment of the field artillery, providing for direct and general support of the attack, massing of fires in each infantry’s zone of advance, counterbattery fire, and the defense of Rendova against seaborne and air attack. One 105-mm. howitzer battalion was assigned to direct support of each regiment, one 155-mm. howitzer battalion to general support of each division. Except for specific direct and general support missions, all artillery would operate as the corps artillery under Barker. The XIV Corps had neither organic artillery nor an artillery commander.

Griswold’s field order, issued on 22 July, directed his corps to attack vigorously to seize Munda airfield and Bibilo Hill from its present positions which ran from Ilangana northwest for about three thousand yards. The 37th Division was to make the corps’ main effort. Beightler’s division was to attack to its front, envelop the enemy’s left (north) flank, seize Bibilo Hill, and drive the enemy into the sea. At the same time it would protect the corps’ right flank and rear. The 43rd Division was ordered to make its main effort on the right. Its objectives were Lambeti Plantation and the airfield. Liversedge’s force, depleted by the abortive attack on Bairoko, was to continue patrolling and give timely information regarding any Japanese move to send overland reinforcements to Munda. The 9th Defense Battalion’s Tank Platoon would assemble at Laiana under corps control. The 1st and 2nd Battalions, 169th Infantry, at Rendova, constituted the corps reserve. The 37th Division was less the 129th Regimental Combat Team and the 3rd Battalions, 145th and 148th Infantry Regiments, and was reinforced by the 161st Regimental Combat Team (less its artillery) and a detachment of South Pacific Scouts. The 43rd Division was less nearly all its headquarters, whose officers were filling most of the posts in Occupation Force headquarters, and less two battalions of the 169th Infantry and the 1st Battalion,103rd Infantry.

All units were ordered to exert unceasing pressure on the enemy. Isolated points of resistance were not to be allowed to halt the advance, but were to be bypassed, contained, and reduced later. Griswold ordered maximum use of infantry heavy weapons to supplement artillery. Roads would be pushed forward with all possible speed.

D Day was set for 25 July. The thirty-minute naval bombardment was to start at 0610, the air bombing at 0635. The line of departure, running northwest from Ilangana, was practically identical with the American front lines except in the zone of the 161st Infantry where the existence of the Japanese strongpoint east of the line had been determined on 24 July.

The XIV Corps was thus attempting a frontal assault on a two-division front, with the hope of effecting an envelopment on the north. In the initial attack it would employ two three-battalion regiments (the 161st and the 172nd) and three two-battalion regiments (the 103rd, the 145th, and the 148th).

Enemy Positions and Plans

On 22 July the Japanese front line ran inland in a northwesterly direction for some 3,200 yards. This line was manned by the entire 229th Infantry, and at the month’s end the 2nd Battalion, 230th Infantry, was also assigned to it. In support were various mountain artillery, antitank, antiaircraft, and automatic weapons units. [4-n5: These included the Antitank Battalion, 38th Division; 2nd Independent Antitank Battalion; a detachment of the 2nd Battalion, 90th Independent Mountain Artillery Regiment; thirteen 7.7-mm. machine guns, and two 75-mm. antiaircraft units.] The positions were the same complex of camouflaged and mutually supporting pillboxes, trenches, and foxholes that had halted Hester in midmonth. The pillboxes started near the beach at Ilangana and ran over the hills in front of the 103rd, 172nd, 145th, and 161st Regiments.

A particularly strong series lay on a tangled set of jungled hills: Shimizu Hill in front of the 172nd Infantry and Horseshoe Hill (so named from its configuration) in front of the 145th and 161st Regiments. Horseshoe Hill lay northwest of Kelley Hill and west of Reincke Ridge. East of Horseshoe Hill lay the Japanese pocket discovered by the 161st. The pocket lay on a north-south ridge that was joined to Horseshoe Hill by a rough saddle. The pillbox line terminated at about the northern boundary of the 161st Infantry. When the 2nd Battalion, 230th Infantry, was committed it did not occupy carefully prepared positions. From the end of the pillboxes the line ran west to the beach, and this north flank does not seem to have been strongly held.

XIV Corps headquarters still estimated that four enemy battalions faced it; three at Munda and one at Bairoko. This was a fairly accurate estimate of strength on the enemy line, but Sasaki had an ace up his sleeve—the 13th Infantry. This regiment, which was not in full strength, was stationed on the American right flank about 4,900 yards west by north from Ilangana. Sasaki’s plans to use his ace were similar to his earlier plans. On the same day that Griswold issued his field order, Sasaki directed Colonel Tomonari to attack the American right flank in the vicinity of Horseshoe Hill on 23 July, then drive east along the Munda Trail. But the Americans struck before Tomonari made his move.

Ilangana and Shimizu Hill: The 43rd Division

In the 43rd Division’s zone, the offensive began as scheduled on the morning of 25 July. For once the weather was favorable. D Day dawned fair and clear, with visibility as good as could be expected in the jungle.

Naval gunfire, air, and artillery preparations went off as scheduled. Commander Burke’s seven destroyers had sailed up from Tulagi. At 0609 the two screening destroyers fired the first of four thousand 5-inch shells at Lambeti Plantation; these were followed by the main group at 0614. Visibility to seaward was good, but the morning haze still hung over Lambeti Plantation. Fifteen minutes later visibility had improved but now the target area was partly obscured by smoke and dust raised by the bombardment.Firing ceased at 0644.

[4-N6: When the infantrymen later reached Lambeti Plantation they found that although the bombardment had done extensive damage many positions, which could have been destroyed only by direct hits, remained intact. The theoretical density of this shelling was 70 rounds per 100 square yards. Admiral Wilkinson, who thought the target area was too far west of the front lines, later observed that 200 rounds per 100 square yards would be required to achieve complete destruction. Because of the difficulty in distinguishing targets in morning haze, Commander Burke recommended that shore bombardments should not start earlier than twenty minutes before sunrise. See CTF 31 (Com III AMPHFOR) Action Rpt for Morning 25 Jul 43, The Bombardment of Munda, 3 Sep 43, in Off of Naval Rcds and Library; and ONI USN, Operations in the New Georgia Area, pp. 53-54.]

From 0630 to 0700, 254 aircraft unloaded 500,800 pounds of fragmentation and high explosive bombs on their target area, a 1,500-by-250-yard strip beginning about 500 yards west of the 103rd Infantry’s front lines. No corps artillery concentrations were fired on 25 July, but the 43rd Division’s supporting artillery began before 0700 the first of more than 100 preparations that were fired that day. The 103rd and 152nd Field Artillery Battalions fired more than 2,150 105-mm. howitzer shells; the 155-mm. howitzers of the 136th Field Artillery Battalion threw 1,182 rounds at the enemy.

With the din subsiding as the artillery shifted its fire to positions farther west, the infantrymen of the 43rd Division moved to the attack at 0700. In the 172nd Infantry’s zone the 2nd and 3rd Battalions on the left and right attacked westward against Shimizu Hill. But by 1000 they had run into the enemy pillbox line and halted. Colonel Ross then requested tanks, got some from the corps reserve, and attacked again. By 1430 three tanks were disabled, and the attack stalled. A little ground had been gained on the regimental left.

The 103rd Infantry, now commanded by Colonel Brown, attacked alongside the 172nd with little more success. [4-N7: Colonel Brown, formerly commander of the 2nd Battalion, took over regimental command when Colonel Hundley replaced the 43rd Division chief of staff on 22 July.]The 3rd Battalion, on the left, pushed forward against machine gun and mortar fire, but immediately hit the Japanese line and stopped. The battalion attempted to move around the pillboxes but found that this maneuver took its men into other machine gun fire lanes.

The 2nd Battalion, 103rd, in the center of the 43rd’s zone, did better. It moved forward two or three hundred yards against light opposition. By 1040 E Company’s leading elements had advanced five hundred yards. The company kept moving until noon, when it had reached the beach near Terere. Here it set up a hasty defense position. But the companies on either flank had not been able to keep up, and the Japanese moved in behind E Company to cut the telephone line to battalion headquarters.

To exploit E Company’s breakthrough, General Hester took the 3rd Battalion, 169th Infantry, out of division reserve and ordered it to push through the same hole E Company had found. But the Japanese had obviously become aware of the gap, and as the 3rd Battalion marched to the line of departure it was enfiladed by fire from the south part of Shimizu Hill and from the pillboxes to the south. It halted. Five Marine tanks were then ordered to push over Shimizu Hill but could not get up the steep slopes. When three of them developed vapor lock all were pulled back to Laiana. In late afternoon the E Company commander decided to abandon his exposed, solitary position, and E Company came safely back through the Japanese line to the 2nd Battalion.

North of the 43rd Division the 37th Division had made scant progress.Thus the first day of General Griswold’s offensive found the XIV Corps held for little gain except in the center of the 43rd Division’s line.

The 43rd Division was weakened by almost a month’s combat, and its reduced strength was spread over a long, irregular, slanting front. It was obvious that combat efficiency would be increased by narrowing the front, and this could be done by advancing the left and straightening the line. Consequently Hester’s plan for 26 July called for the 172nd to stay in place while the 103rd Infantry attempted to advance the eight hundred yards from Ilangana to Terere.

Strong combat patrols went out in the morning of 26 July to fix the location of the Japanese pillboxes as accurately as possible. After their return, the artillery began firing at 1115, one hour before the infantry was to attack. At 1145 the 103rd’s front was covered with smoke and under its cover the front-line companies withdrew a hundred yards. At noon the artillery put its fire on the Japanese positions directly in front. As the tanks were not quite ready at H Hour, 1215, the artillery kept firing for ten more minutes. It lifted fire one hundred yards at 1225, and the 103rd started forward. The tanks led the advance in the center; behind them was the infantry. Attached to the 103rd for the attack were 2nd Lieutenant James F. Olds, Jr., the acting corps chemical officer, and six volunteers from the 118th Engineer Battalion. Each carried a flame thrower, a weapon which the 43rd Division had brought to New Georgia but had not used up to now. Griswold, whose headquarters had conducted flame thrower schools on Guadalcanal, was aware of the weapon’s possibilities. That morning the six engineers had received one hour of training in the use of the M1A1 flame thrower.

The flame throwers went forward with the infantry, which halted about twenty yards in front of the pillbox line and covered it with small arms fire. Under cover of this fire the flame thrower operators, their faces camouflaged with dirt, crawled forward. Operating in teams of two and three, they sprayed flame over three barely visible pillboxes in front of the center of the 103rd’s line. Vegetation was instantly burned off. In sixty seconds the three pillboxes were knocked out and their four occupants were dead. [4-N10: Captain James F. Olds, Jr., “Flamethrowers Front and Center,” Chemical Warfare Bulletin, Vol. 30, No. 3 (June-July 1944), pp. 5-9. This account, while valuable, seems to have telescoped two situations and actions into one, for Olds asserts that the 103rd was at Lambeti Plantation on 26 July. From the fact that three pillboxes had only four occupants, it would seem that this part of the Japanese line was lightly manned.]

Operations of the infantry, tanks, flame throwers, and supporting heavy weapons and artillery met with almost complete success. The 103rd Infantry encountered seventy-four pillboxes on a 600-yard front, but by midafternoon, spurred on by pressure from General Wing, it had reduced enemy resistance at Ilangana. From there it continued its advance through underbrush and vines and gained almost 800 yards. By 1700 the left flank rested on the coastal village of Kia. The 43rd Division’s line, formerly 1,700 yards long, was now much straighter by 300 yards.

From 28 through 31 July, the 43rd Division inched slowly forward, a few yards on the right flank and about five hundred yards along the coast. This was accomplished by “aggressive action and small unit maneuver, combined with constant artillery and mortar action [which] gradually forced the enemy back from his high ground defenses.” The 172nd ground its way over Shimizu Hill, the last real ridge between it and Munda airfield, and in doing so it helped unhinge the main Japanese defense system in its zone, just as the 103rd’s drive through Ilangana had broken the enemy line on the left. [4-N12: Unfortunately the records are too scanty to provide details showing just how the 172nd took this position. During the attack 1st Lieutenant Robert S. Scott almost singlehandedly halted a Japanese counterattack and for his gallantry was awarded the Medal of Honor. WD GO 81, 14 Oct 44.]

Major Zimmer’s 1st Battalion, 169th Infantry, was brought over from Rendova on 29 July; the 3rd Battalion, now commanded by Major Ignatius M. Ramsey, was taken out of division reserve and the 169th (less the 2nd Battalion) was assigned a zone between the 172nd and the 103rd. [Colonel Reincke was now regimental executive officer.] As the month ended the 169th (less its 2nd Battalion in corps reserve) was in the process of extending to the northwest to pinch out the 172nd.

Command of the 43rd Division changed hands on 29 July when Major Gen. John R. Hodge, the tough, blunt commander of the Americal Division, came up from the Fijis to take over from Hester. This change was ordered by General Harmon who felt that Hester had exhausted himself. General Hodge had served as assistant commander of the 25th Division during the Guadalcanal Campaign, and thus had had more experience in jungle warfare than any other general then in New Georgia. Hodge, Harmon wrote, was the “best Div Comdr I have in area for this particular job.”

The 43rd Division, having cracked through the Shimizu Hill-Ilangana positions, was in a favorable position to drive against Munda under its new commander, while the 37th Division on its right fought its way through the enemy positions in its hilly, jungled zone.
The Attack Against the Ridges: The 37th Division

The dawn of D Day, 25 July, found the 43rd Division committed to a general attack, but the 37th Division was forced to postpone its advance. General Beightler had issued a field order on 23 July calling for a general attack by his three regiments, to start at 0700, 25 July. The 145th, 161st, and 148th Infantry Regiments were to attack due west, toward Bibilo Hill, on the division’s left, center, and right, with the 145th maintaining contact with the 172nd Infantry on its left and the 148th Infantry covering the corps’ right flank and rear. But the discovery of the strong Japanese position east of the 161st Infantry’s line of departure altered the plans.

On 24 July Beightler ordered Colonel Holland not to advance his 145th Infantry, but to stay in place. He told Colonel Baxter to move the 148th Infantry only up to the line of departure. The two regiments would hold while part of the 161st contained the Japanese position and the rest of the regiment bypassed it and came up on a line with the 145th and 148th.

After Baxter received the commanding general’s orders, he suggested that his regiment could perhaps help the 161st reduce the pocket by making a limited advance. Baxter hoped to establish an observation post on high ground from which both Munda airfield and the pocket in front of the 161st could be seen. Beightler assented to this request at 0910, 25 July. Patrols went out, and on their return the direct support artillery battalion laid a ten-minute preparation 400 yards in front of the line of departure while mortars covered the 400-yard gap. The 2nd Battalion, 148th, commanded by Lieutenant Colonel Herbert Radcliffe, started forward, met no Japanese, and gained 500-600 yards. The 1st Battalion, Lieutenant Colonel Vernor E. Hydaker commanding, moved up to the 2nd Battalion’s old positions.

Bartley Ridge and Horseshoe Hill

When I Company, 161st Infantry, had been unable to reduce the Japanese strongpoint on Bartley Ridge on 24 July, Colonel Dalton issued orders for its seizure on D Day.[Bartley Ridge was named in memory of 2nd Lieutenant Martin E. Bartley, a I Company platoon leader killed on 25 July.] I Company was to contain the Japanese pocket by attacking to its front while the 1st Battalion and the rest of the 3rd Battalion executed a double envelopment. The 1st Battalion was to move around the Japanese left (north) flank while the 3rd Battalion went around the right, after which the two battalions would drive southward and northward for two hundred yards. Fifteen minutes of mortar fire would precede these moves. Beightler arranged for the 145th and 148th Regiments to support the 161st with heavy weapons fire. He also asked corps headquarters for tanks to help the 161st, but the 43rd Division had been given the tanks for the D-Day attack.

From positions near the Laiana Trail eight 81-mm. mortars opened fire at 0745, 25 July, in support of Dalton’s attack. Heavy weapons of the adjoining regiments attempted to deliver their supporting fires, but the denseness of the jungle prevented forward observers’ controlling the fire. The unobserved fire began obstructing rather than helping the 161st, and that part of the plan was abandoned.

The 3rd Battalion, commanded by Lieutenant Colonel David H. Buchanan, was unable to gain. Shortly after 0800, when the attack began, I Company reported that its attack against the ridge strongpoint had stalled. A knob projecting east from Bartley Ridge and the heavy undergrowth provided enough cover and concealment to let the infantrymen reach the base of the ridge, but uphill from the knob, where the growth was thinner, all movement was halted by fire from the crest. The 161st Infantry had made plans to use flame throwers, and an operator carrying his sixty-five pounds of equipment made two laborious climbs and silenced an enemy machine gun, but many other Japanese positions remained in action. The main body of the 3rd Battalion, attempting to get around the south end of Bartley Ridge, was also halted.

The 1st Battalion was more successful. At 1035 its commander, Lieutenant Colonel Slaftcho Katsarsky, radioed Dalton that he had found the north flank of the Japanese position on Bartley Ridge, and that he was moving his battalion around it. Shortly afterward Beightler, Dalton, and staff officers conferred and decided that the 3rd Battalion should contain the strongpoint while the 1st Battalion pushed westward with orders to develop enemy positions but not to engage in full-scale combat. The 37th Division could not advance westward in force until Bartley Ridge had been cleared.

The 3rd Battalion established itself in containing positions north, east, and south of Bartley Ridge. E Company was released from reserve and sent into line on high ground just north of Bartley to secure the right flank in the 161st’s zone. The 1st Battalion advanced to a point about four hundred yards west of Bartley and halted on a small rise northeast of Horseshoe Hill. Tanks of the newly arrived 10th Marine Defense Battalion were to be committed to support the 37th Division the next day, and in the afternoon the tank commander made a personal reconnaissance of Colonel Buchanan’s zone in preparation for the attack.

Six light Marine tanks were to lead out in the attack at 0900, 26 July, after preparatory fire by machine guns and 81-mm. mortars. L and K Companies, in column, would move behind the tanks, which were supported by infantrymen armed with .30-caliber M1 rifles, .30caliber Browning Automatic Rifles (BAR’s), and two flame throwers. Tank-infantry communication was indirect. The tank radios formed a net within the Tank Platoon, and a 161st radio car maintained radio contact between the Tank Platoon commander and Colonel Buchanan.

When the tanks, with their hatches closed, got off the approach trail that had been bulldozed by members of the 65th Engineer Battalion, Colonel Buchanan directed the infantrymen to lead them forward. It was 0925 before the attack got started. In two lines of three vehicles each, the tanks lumbered over the littered undergrowth, steep slopes, and felled logs toward the southeast slope of Bartley Ridge. The Japanese quickly responded with fire from antitank and 70-mm. battalion guns, machine guns, and mortars.

The attack went well at first. About a dozen pillboxes were reported knocked out by 1110, and Buchanan ordered his men to occupy them to keep the Japanese from moving in again at night. Unfortunately, the tanks had encountered exactly the sort of difficulties that might be expected in tangled terrain with communications uncertain. In their lurches and frequent changes of direction they injured some of the accompanying foot troops. Poor visibility caused them to get into untenable positions from which they had to be extricated with consequent delays to the attack. During the morning a Japanese soldier stole out of the tangled brush and planted a magnetic mine that disabled one tank. A second tank was halted by a ruptured fuel line. The remaining four withdrew at 1110 to reorganize.

The flame thrower operators, carrying their bulky, heavy fuel tanks on their backs, were not properly protected by the riflemen and were soon killed.

In the course of the day’s fighting some fourteen pillboxes and a number of machine gun positions were knocked out, and the 3rd Battalion advanced about two hundred yards up Bartley Ridge.[4-N18] But it met such heavy fire from Bartley and Horseshoe Hill that its position clearly could not be held. Attempts to pull out the disabled tanks were unsuccessful. The battalion withdrew to its previous positions. The attack had disclosed the existence of so many more positions that Dalton received Beightler’s permission to make a thorough reconnaissance before attacking again.

[4-N18 In the course of the day’s action Captain Paul K. Mellichamp, battalion executive officer, picked up a radio from a wounded operator and directed mortar fire. He was wounded, but continued to direct the fire until he collapsed. He died shortly afterward, and was awarded the Distinguished Service Cross.]

While the 161st was attacking Bartley Ridge on 25 and 26 July, Colonel Holland’s 145th Infantry stayed in its forward positions on Reincke Ridge and Kelley Hill. During this period it sent out patrols to the north and west to try to find the source of the 90-mm. mortar fire that had been hitting the regiment since 22 July. It received no artillery support at this time because its front line was considered too close to enemy targets for the artillery to fire without hitting American infantry. By the end of 26 July the 1st Battalion, 161st, had fought its way forward to come up on line north of the 145th, but near the regimental boundary the line sagged eastward in the shape of a great U. Colonel Parker’s 2nd Battalion, 145th, was occupying positions in rear of the 1st Battalion, 145th, commanded by Lieutenant Colonel Richard D. Crooks.

General Beightler ordered Holland to commit his 2nd Battalion, in order to reduce the Japanese positions on Horseshoe Hill that had fired on the 3rd Battalion, 161st, during its 26 July attack against Bartley Ridge. Doubtless because troops of the 161st had not been able to get past the south end of Bartley, Colonel Parker’s battalion was to march northward right around the 3rd Battalion, 161st, push west around the north of the enemy positions on Bartley Ridge, then attack to the southwest. This maneuver would entail a march of about one and one-half miles to the assembly area.

Parker’s battalion moved out in the early morning of 27 July. It reached its assembly area on the north flank of Bartley Ridge without incident. After a preparation of one hundred rounds by the division artillery, which cleared some of the foliage, the battalion advanced to the attack in column of companies. “Having to fight every foot of the way,” it gained about three hundred yards before 1300, when E Company in the lead moved south off slopes of a ridge and started up a small knob projecting from Horseshoe Hill. As the company ascended the hill it was struck by fire from pillboxes. Among the first men killed was Captain Gardner B. Wing, E Company’s commander, in whose honor the 145th christened the knob.

On the same day, while American mortars fired intermittently at Bartley Ridge, patrols from the 3rd Battalion, 161st Infantry, examined the Japanese lines to procure data for a preparation by the corps artillery. In the course of the reconnaissance Colonels Dalton and Buchanan observed enemy pillboxes on the right flank of the 1st Battalion, 145th Infantry, and recommended that the attack be delayed until 29 July. General Beightler gave his assent. Dalton and Buchanan also decided to attack from the northwest instead of the southeast. Reconnaissance and pressure were to continue on the 28th.

On the morning of the 28th a ten-man patrol from I Company, led by Lieutenant Walter Tymniak, set out in a southerly direction toward the top of Bartley Ridge. To their surprise and satisfaction, the Americans met no fire, got safely to the top, and found several abandoned pillboxes. They occupied them, and I Company followed to the crest and began infiltrating the pillboxes. Not all were vacant, but the task of the attackers was eased as each pillbox was taken, for its fire could then no longer be used with that of its neighbors to make crossfire or interlocking bands of fire. Because the Japanese appeared to be evacuating and the American front was intermingled with the enemy front, the artillery preparation was called off. The 3rd Battalion continued its infiltration on 29 July. At the end of the day it was relieved by Major Francis P. Carberry’s 2nd Battalion and went into division reserve.

The 145th Infantry’s zone was shifted farther north on 30 July as part of a general shift in boundaries that General Griswold was making in order to widen the 43rd Division’s front. This move placed the southern half of Bartley Ridge within the 145th’s zone. Colonel Parker’s 2nd Battalion, 145th, had just completed its move around the 161st’s north flank, thence southwest against Horseshoe Hill. On 30 July it was attached to the 161st for the completion of the reduction of Bartley Ridge and Horseshoe Hill.

Carberry’s and Parker’s battalions pushed their attacks on 30 July. In contrast with Carberry’s battalion, which met little resistance, Parker’s men engaged in sharp fighting in the west. The Japanese who had evacuated the position facing Carberry had apparently moved into positions facing Parker. With grenade, rifle, machine gun, mortar, and flame thrower the two battalions fought all day and part of the next, until by midafternoon of 31 July the Japanese rear guards on Bartley Ridge were either dead or in flight, and the 2nd Battalion,161st, had advanced west and was on a line with the 1st. Bartley Ridge had contained forty-six log and coral pillboxes and thirty-two other lighter positions. First attacked by a company, it fell only after seven days’ fighting by two battalions.

On Horseshoe Hill the Japanese resisted from their pillboxes and foxholes with equal skill and enthusiasm. The Americans used small arms, grenades, automatic weapons, mortars, flame throwers, and field artillery as they systematically reduced the enemy positions, almost pillbox by pillbox.[4-N21] On 1 August Parker’s battalion received orders to attack in late afternoon, obeyed, and took Horseshoe Hill without firing a shot or losing a man. The Japanese had gone.

[4-N21: During these operations Private First Class Frank J. Petrarca, a medical aid man, so distinguished himself by gallant, selfless devotion to duty that he was posthumously awarded the Medal of Honor. WD GO 86, 23 Dec 43.]

Advance and Withdrawal of the 148th Infantry

On 1 August, the day on which the Americans completely occupied the ridge positions, the 148th Infantry returned eastward to the 37th Division’s lines after an advance which had taken it almost to Bibilo Hill. The 148th Infantry was the only regiment not confronted by prepared enemy positions, and it had made comparatively rapid progress from the first. When Colonel Baxter moved his regiment forward on 25 July, it went around the north flank of the Japanese defense line and met no resistance. However, none of the Americans then knew that the major part of the enemy 13th Infantry lay to the north of Baxter’s right flank. Patrols, accompanied by Fiji scouts, went out and reported the presence of a few Japanese to the west, none to the south. Generals Griswold and Beightler had emphasized the importance of maintaining lateral contact and Beightler had expressly directed that the 148th was to maintain contact with the 161st, and that all units were to inform their neighbors and the next higher unit of their locations. The 148th, however, was not able to make contact on its left with the 161st Infantry.

Baxter’s two-battalion regiment advanced regularly for the next three days. Colonel Radcliffe’s 2nd Battalion led on 26 and 27 July; on 28 July Colonel Hydaker’s 1st Battalion bypassed the 2nd and led the advance to a point somewhere east of Bibilo Hill.[4-N22] Patrols went out regularly and at no time reported the presence of a sizable body of the enemy. On 27 July Baxter reported that he had established “contact with Whiskers.” Colonel Dalton, the “guest artist” regimental commander of the attached 161st Infantry, sported a beard and was dubbed “Whiskers” and “Goatbeard” in the 37th Division’s telephone code.

[4-N22: The total of daily yardage reported in the journals, if correct, would have placed the 148th west of Bibilo Hill on 28 July, but the 148th soldiers, like almost everyone else in the jungle, overestimated the distances they had traveled.]

But the 148th’s front was almost a thousand yards west of Whiskers’ 1st Battalion, and the contact must have been tenuous. Next day G Company was ordered to move to the left to close a gap between the two regiments, but the gap stayed open.

During the move troops of the 117th Engineer Battalion labored to push a supply trail behind the advancing battalions. The rate of march was in part geared to the construction of the supply trail. As Baxter told Radcliffe over the telephone on 27 July, “I am advancing behind you as fast as bulldozer goes.”

Next day, however, there occurred a disturbing event. A platoon from A Company, 117th Engineer Battalion, was using a bulldozer to build the trail somewhere north of Horseshoe Hill when it was ambushed by the enemy. Three engineers were killed and two were wounded before elements of the Antitank Company and of the 1st Battalion rescued the platoon and extricated the bulldozer.

Japanese movements during this period are obscure, but this and subsequent attacks were made by the 13th Infantry coming south at last in accordance with Sasaki’s orders.

The situation became more serious on 28 July, the day on which Baxter’s aggressive movement took him almost to Bibilo Hill. At this time the regiment was spread thinly about fifteen hundred yards beyond the 161st; its front lay some twelve hundred yards west of the regimental ration dump and eighteen hundred yards from the point on the supply line “which could be said to be adequately secured by other division units.” There was still no contact with the 161st, and in the afternoon a group of the 13th Infantry fell upon the ration dump. From high ground commanding it the enemy fired with machine guns, rifles, and grenade discharges at men of the regimental Service Company. The Service Company soldiers took cover among ration and ammunition boxes and returned the fire. The dump, under command of Major Frank Hipp, 148th S-4, held out until relieved by two squads of the Antitank Company and one platoon from F Company. East of the dump, troops of the 13th Infantry also forced the 148th’s supply trucks to turn back. Baxter, stating “I now find my CP in the front line,” asked Beightler to use divisional units to guard the trail up to the dump.

All the 148th’s troubles with the Japanese were in the rear areas. The westward push, which took the leading battalion as far as one of the Munda-Bairoko trails, had been practically unopposed. But early on the morning of 29 July General Beightler, unaware of the 13th’s position, telephoned Baxter to say that as the Japanese seemed to be moving from the southwest through the gap between the 148th and 161st Regiments, and around the 148th’s right, Baxter was to close up his battalions and consolidate his positions. At 0710 Beightler told Baxter to withdraw his battalions to the east, to establish contact with the 161st, and to protect his supply route. Baxter, who had sent patrols out in all directions early in the morning, at 0800 ordered one company of the 2nd Battalion to clear out the supply trail to the east. At 0941, with Japanese machine guns still dominating the supply trail, Beightler sent Baxter more orders similar to those of 0710, and also ordered forward a detachment of the 37th Cavalry Reconnaissance Troop to help clear the east end of the supply trail. The telephone, so busy with conversations between Beightler and Baxter on 29 July, was then quiet for an hour.

Meanwhile Beightler had been conferring with Dalton, Holland, and members of the divisional general staff. As a result he had decided that the 161st should continue reducing Bartley Ridge, that the 145th should stay in place, and that the 148th would have to withdraw. So at 1055 Baxter ordered his regiment to turn around and pull back to the east. The 2nd Battalion, 148th, was to use at least one company to establish contact with the 161st while the rest of the battalion withdrew toward the ration dump. The 1st Battalion would move back to the 2nd Battalion’s positions. At 1150 Baxter reported the 2nd Battalion in contact with the 161st, and shortly afterward Beightler ordered Baxter to move the 1st Battalion farther east, putting it in position to deliver an attack the next morning against the rear of the Japanese holding up Dalton’s regiment.

The division commander again emphasized the necessity for maintaining firm contact with the 161st. [As the Americans still did not know the13th Infantry’s location, they thought the attack had originated from the southwest rather than from the north.] At 1305, with the 148th moving east, Colonel Katsarsky reported that his 1st Battalion, 161st, had as yet no contact with the 148th. Beightler at once told Baxter that, as Japanese machine gunners were operating between the two regiments, the gap must be closed before dark. An hour later Baxter called Beightler to say that he was too far west to close with the 161st before dark. When Beightler ordered him to close up anyway, Baxter demurred. Asking his general to reconsider the order, he stated that he could almost, but not quite, close the gap. Beightler thereupon told Baxter to comply with his orders as far as was physically possible.

The 2nd Battalion had meanwhile been pushing east, except for F Company’s main body, which was advancing west toward the ration dump. Both bodies were encountering enemy resistance, and the day ended before the Japanese were cleared out. The Reconnaissance Troop cleared some Japanese from the eastern part of the supply trail, but at 1758 Baxter reported that the trail had been closed by raiding Japanese.

The 148th Infantry, in examining the personal effects of some of the dead Japanese, found that the men belonged to the 13th Infantry. Some of them had been carrying booty taken in the raids east of the Barike several days earlier. Colonel Baxter later estimated that the enemy harrying his regiment numbered no more than 250, operating “in multiple small light machine gun and mortar detachments and . . . [moving] from position to position utilizing the jungle to its maximum advantage. You can well imagine what we could do with our M-1’s, BAR’s and Machine Guns if all we had to do was dig in and wait for the Jap to come at us.

General Beightler, a National Guardsman most of his life, was an affable man, but he was far from satisfied with the outcome of the day’s action. At 1832 he radioed Baxter that General Griswold had ordered the 148th to establish contact with the 161st early the next morning and to protect the supply route. “Use an entire battalion to accomplish latter if necessary. At no time have you been in contact on your left although you have repeatedly assured me that this was accomplished…Confirmation of thorough understanding of this order desired.” Baxter thereupon telephoned division headquarters and put his case before a staff officer. General Beightler’s criticism, he felt, was not justified. “Please attempt to explain to the General that I have had patrols in contact with the 161 and have documentary evidence to substantiate this. I have not, however, been able to maintain contact and close the gap by actual physical contact due to the fact that the 161st had been echeloned 600 to 800 yards to my left rear. I have been trying and will continue tomorrow morning to establish this contact. It is a difficult problem as I have had Japs between my left flank and the 161st.”

Rain and mud added to Baxter’s troubles on 30 July. Still harried by enemy machine guns and mortars, the 2nd Battalion pushed east and south toward the 161st as the 1st Battalion covered the left (north) flank. Elements of the 37th Cavalry Reconnaissance Troop, C Company, 117th Engineer Battalion, and the 3rd Battalion, 161st Infantry, pushed north to give additional protection to the division’s right (north) flank and to protect the east end of Baxter’s supply route.

Baxter attempted to cut a new trail directly into the 161st’s lines, but Japanese rifle fire forced the bulldozer back. Some of the 148th’s advance elements side-slipped to the south and got through to the 161st that day, but the main body was still cut off. [F and H Companies, part of E Company, and the 2nd Battalion Headquarters Company were the elements that got through to the 161st] Some of the Japanese who were following the 148th attacked the 1st Battalion, 161st, but were halted. This action then settled down into a nocturnal fire fight. The plight of the rest of the regiment was still serious. Water was running low. Part of the Reconnaissance Troop tried to take water forward to the 148th on 30 July. It was stopped by Japanese fire. But rain fell throughout the night of 30-31 July and the thirsty men were able to catch it in helmets and fill their canteens.

On 31 July Beightler suggested that Baxter destroy heavy equipment and break his regiment into small groups to slip northward through the jungle around the enemy. The 148th blew up all the supplies it could not carry but it had to fight its way along the trail. It had over a hundred wounded men and could not infiltrate through the jungle without abandoning them.

Toward the end of the day B Company, which had been trying to clear the Japanese north of the supply trail, was ordered to disengage and withdraw slightly for the night. One of B Company’s platoons, however, had come under fire from a Japanese machine gun about seventy-five yards to its front and found that it could not safely move. Private Rodger Young, who had been wounded in the shoulder at the first attempt to withdraw, told his platoon leader that he could see the enemy gun and started forward. Although a burst from the gun wounded him again and damaged his rifle, he kept crawling forward until he was within a few yards of the enemy weapon. As a grenade left his hand he was killed by a burst that struck him in the head. But he had gotten his grenade away, and it killed the Japanese gun crew. His platoon was able to withdraw in safety. For his gallantry and self-sacrifice Young was posthumously awarded the Medal of Honor.

Colonel Baxter’s radio fairly crackled the next morning, 1 August, with orders from General Beightler: “Time is precious, you must move.” “Get going.” “Haste essential.” Thus urged on, Baxter ordered an assault by every man who could carry a rifle. He formed all his command—A, E, B, and G Companies—in a skirmish line with bayonets fixed, and assaulted by fire and movement at 0850. The attack succeeded. By 0930 the leading elements, ragged, weary, and muddy, reached Katsarsky’s area. The 148th was given fresh water and hot food, then passed into division reserve. As the men struggled in after their ordeal, all available ambulances, trucks, and jeeps were rushed up to transport the 128 wounded men to the 37th Division’s clearing station at Laiana.

Capture of the Airfield

The first day of August had broken bright and clear after a night of intermittent showers. It is likely that the spirits of the top commanders were also bright, for things were looking better. With Ilangana and Shimizu Hill reduced, the 43rd Division was in possession of the last piece of high ground between it and Munda airfield. Bartley Ridge had fallen; Horseshoe Hill was about to fall, and the 148th was completing its retirement.

General Griswold had issued no special orders for the day; the field order that had started the corps offensive was still in effect. In the 37th Division’s zone the most significant development was the return of Baxter’s men. The 145th Infantry was patrolling; the 161st was mopping up. In the 43rd Division’s area of responsibility, General Hodge had ordered an advance designed to bring his division up on line with the 145th Infantry.

The 103rd Infantry began its attack at 1100. E, G, and F Companies advanced in line behind patrols. Meeting practically no opposition, they gained ground rapidly and by 1500 were nearing Lambeti Plantation. The 2nd Battalion, 169th Infantry, then in process of pinching out the 172nd, attacked northwest across the front of the 172nd and established contact with the 145th Infantry. The 172nd completed a limited advance before going into division reserve. The 3rd Battalion, 169th, on the left of the 2nd, attacked in its zone and at 1500 was still advancing. For the first time since it had landed on New Georgia, the 43rd Division could announce that the going was easy.

The day before, Generals Hodge and Wing, accompanied by Colonel Ross, had visited the command and observation posts of the 1st Battalion, 145th Infantry, from where they could see part of Munda airfield. They detected evidence of a Japanese withdrawal, which seemed to be covered by fire from the enemy still on Horseshoe Hill.

Thus at 1500, 1 August, with the 43rd Division still moving forward, General Griswold ordered all units to send out patrols immediately to discover whether the Japanese were withdrawing. “Smack Japs wherever found, even if late.” If the patrols found little resistance, a general advance would be undertaken in late afternoon. Colonel Ridings telephoned the orders to 37th Division headquarters, and within minutes patrols went out. They found no enemy. At 1624 Ridings called Beightler’s headquarters again with orders to advance aggressively until solid resistance was met, in which case its location, strength, and composition were to be developed. The 148th Infantry was to have been placed in corps reserve with orders to protect the right flank, patrol vigorously to the north, northeast, and northwest, and cut the Munda-Bairoko trail if possible. Since the 27th Infantry of the 25th Division was arriving and moving into position on the 37th Division’s right flank, Beightler persuaded corps headquarters to let him use the 148th Infantry. Ridings required, however, that the 148th be given the mission of protecting the right flank because the 27th Infantry would not have enough strength for a day or two.

All went well for the rest of the day. The 103rd Infantry reached the outer taxiways of Munda airfield; the 169th pulled up just short of Bibilo Hill. The 37th Division’s regiments plunged forward past Horseshoe Hill, which was free of Japanese, and gained almost seven hundred yards.

The Japanese Withdrawal

The Japanese positions facing the XIV Corps had been formidable, and the Americans had been held in place for long periods. But the Americans had wrought more destruction than they knew. The cumulative effect of continuous air and artillery bombardment and constant infantry action had done tremendous damage to Japanese installations and caused large numbers of casualties. By late July most of the Japanese emplacements near Munda were in shambles. The front lines were crumbling. Rifle companies, 160-170 men strong at the outset, were starkly reduced. Some had only 20 men left at the end of July. The 229th Infantry numbered only 1,245 effectives. Major Hara, Captain Kojima, and many staff officers of the 229th had been killed by artillery fire. Hospitals were not adequate to care for the wounded and sick. The constant shelling and bombing prevented men from sleeping and caused many nervous disorders.

To compensate for the diminution of his regiment’s strength, Colonel Hirata ordered the soldiers of his 229th Infantry to kill ten Americans for each Japanese killed, and to fight until death.

Higher headquarters, however, took a less romantic view of the situation. On 29 July a staff officer from the 8th Fleet visited Sasaki’s headquarters and ordered him to withdraw to a line extending from Munda Point northeast about 3,800 yards inland. The positions facing the XIV Corps, and Munda airfield itself, were to be abandoned. Sasaki and his subordinates thought that it would be better to withdraw even farther, but the views of the 8th Fleet prevailed over those of the responsible men on the spot. The withdrawal, which was deduced by XIV Corps headquarters on 1 August, was accomplished promptly, and except for detachments at Munda and in the hills the main body of Sasaki’s troops was in its new position by the first day of August.
Jungle Techniques and Problems

The Americans did not yet know it, but the worst was over. All regiments began making steady progress each day against light, though determined and skillful, opposition.

By now all regiments, though depleted by battle casualties and disease, had become veterans. Pockets that once would have halted an entire battalion or even a regiment were now usually reduced with speed and skill. The flame thrower, receiving its most extensive use in the Pacific up to this time, was coming into its own as an offensive weapon. All regiments employed it against enemy positions, both in assault and in mopping up. The flame thrower did have several important disadvantages. The equipment was large and heavy, and required the operator to get very close to enemy positions, then expose his head and body in order to use his weapon. He needed to be protected by several riflemen. But even with its disadvantages, it was useful in destroying enemy positions.

Tanks, too, were of great value. General Griswold felt that, despite the difficulties inherent in operations over hilly jungle, the actions of the Tank Platoons of the 9th and 10th Marine Defense Battalions had been successful. On 29 July,

looking forward to fighting over easier terrain around Munda airfield, he asked General Harmon for more tanks. Corps headquarters, he also announced, was preparing to mount flame throwers on tanks.[4-N36] The operation ended before flame throwing tanks could be used, but the idea was successfully carried out in later campaigns.

[4-N36: Rad, Griswold to Harmon, 29 Jul 43, in XIV Corps G-3 Jnl, 30 Jul 43. Part of the 754th Army Tank Battalion was alerted at Noumea for transfer to Guadalcanal to be equipped with flame throwers for employment in New Georgia but Munda airfield had fallen before it was moved.]

The technique of reducing a pillbox, whether isolated or part of a defensive system, was now mastered. The official records unfortunately do not give much exact information on the reduction of specific pillboxes, but after the battle the 37th Division gave a valuable general description of the methods employed.

The first essential was a complete reconnaissance to develop the position, intention, and strength of the enemy. This was quite difficult in the jungle. “To one unskilled in jungle fighting, it is inconceivable that well trained reconnaissance patrols insufficient numbers cannot develop the situation in front of the advancing forces.”  Because they could not see far enough, because they could not always get close enough, and because Japanese fire discipline was sometimes so good that a given position would not fire until actually attacked, reconnaissance patrols could not always develop positions. The next step was a reconnaissance in force by a reinforced platoon.

This often uncovered a portion of the enemy position but not all of it. Usually the complete extent of a center of resistance was determined only by the attack. The attack itself consisted of three parts: artillery preparation, 81-mm. mortar fire, and assault, the artillery preparation had a threefold effect. It improved visibility by clearing away brush and foliage. It destroyed or damaged enemy positions. And it killed, wounded, and demoralized enemy soldiers.

The 81-mm. mortars, using heavy shell that had a delay fuze, fired on observed positions and usually covered the area between the American infantry and the artillery’s targets. They frequently drove the Japanese soldiers out of their pillboxes into the open where they became targets for rifle and machine gun fire. The 60-mm. mortars, though more mobile than the 81-mm.’s, threw too light a shell to be very effective in these attacks. Their shells usually burst in the trees, but the 81-mm. heavy shells penetrated the treetops and often the tops of the pillboxes themselves before exploding.

The assault consisted of a holding attack by a company or platoon delivering assault fire to cover a close-in single or double envelopment. BAR’s, M1’s, and grenades were used extensively, and flame throwers were employed whenever possible. Units of the 25th Division, which later drove northward from Munda to Zieta, encountered pillbox positions that were too shallow, and in country too dense, for artillery and mortars to be used without endangering the attacking infantry. Men of this division therefore advocated flame throwers, infantry cannon, and tanks for pillbox reduction.

These techniques, which simply represented the application of established tactical principles, were being applied well in early August, but several problems remained. Because the infantry units did not advance at the same rate, the front line became irregular and the supporting artillery was thus unable to capitalize on the advantages of firing at right angles to the axis of advance. All unit commanders were eager to employ artillery and mortar support to the utmost, but they frequently complained that neighboring units’ supporting artillery and mortar fires were falling in their areas and endangering their troops. They had a tendency to forget that the enemy also used artillery and mortars and, when receiving American artillery fire, frequently lobbed 90-mm. mortar shells into the American front lines to convince the American infantrymen that they were being fired on by their own artillery. In most cases the complaints were probably caused by Japanese rather than American fire.

Because maps were inaccurate and reconnaissance was inhibited by poor visibility, it was extremely difficult to determine the exact location of friendly units. In the 37th Division’s zone several artillery preparations were called off because of uncertainty about the position of the 148th Infantry. Flares and smoke pots, and sometimes flame throwers, were used to mark flanks, but usually could not be seen by anyone not in the immediate vicinity. Griswold had ordered the front line battalions to mark their flanks daily with white panels twenty-five feet long by six feet wide. These were to be photographed from the air. Reconnaissance planes made daily photographic flights, but there were no clearings in the New Georgia jungle large enough to permit the panels to be spread out, and this effort failed. By plotting close-in defensive artillery fires, forward observers were able to provide some reliable information on the location of front lines. When the 37th Division rolled forward after 1 August, it estimated positions and distances on the basis of speedometer readings from locations that had been plotted by air photography and interpolated on maps.

The difficulties of scouting and patrolling naturally affected nearly every aspect of the operation. Because enemy positions could not be fixed in advance, the troops often attacked terrain rather than the enemy. This procedure resulted in slow advances and in a high expenditure of mortar ammunition on areas actually free of the enemy. And mortar ammunition supply was laborious; shells had to be hand-carried from trail-end to the mortar positions. Poor scouting caused battalions to advance on narrow fronts and thus be halted by small enemy positions. One regimental operations officer asserted that inadequate reconnaissance was due in part to the fact that “higher commanders” did not issue orders until the late afternoon preceding an attack. Thus battalions did not have time for full reconnaissance: “Many times, units were committed in an area which had not been reconnoitered. This fact resulted in commanders having to make decisions concerning a zone of advance in which he knew little or nothing about the enemy positions. Enemy strong points encountered in this fashion often times resulted in hasty withdrawals which were costly both in men and weapons.

“Munda is yours”

The XIV Corps maintained the momentum of its advance against the enemy delaying forces. On 2, 3, 4, and 5 August the advance continued all across the corps’ front. The 103rd and 169th Infantry Regiments, which had gained the outer taxiways of the airfield on 1 August, kept going. The 3rd Battalion, 172nd, was committed on the 169th’s right on 4 August. In the more open terrain around the airstrip the troops were able to use 60-mm. mortars effectively, and their advance was consequently speeded. Kokengolo Hill, the rise in the center of the airfield where a Methodist mission had once stood, held up the advance temporarily. Bibilo Hill, whose fortifications included six 75-mm. antiaircraft guns that the Japanese had been using as dual-purpose weapons, was reduced in three days of action by elements of the 169th, 172nd, 145th, and 161st Regiments, supported by Marine tanks. The 148th Infantry, on the north flank, established blocks and ambushes on a north-south track action near the base of Bibilo Hill. that was presumed to be the Munda-Bairoko trail.

On 5 August, with Bibilo Hill cleared, the units of the 37th Division crossed the narrow strip of land between the hill and the water. This tactical success had one effect of great personal importance to the soldiers: many had their first bath in weeks.

In the 43rd Division’s zone on 5 August, the infantry, with tank and mortar support, killed or drove the last Japanese from the tunnels, bunkers, and pillboxes of Kokengolo Hill. Here were found caves stocked with rice, bales of clothing and blankets, and occupation currency. Crossing the western part of the runway, with its craters, grass, and wrecked Japanese planes, the infantrymen secured it in early afternoon. General Wing telephoned General Hodge from Bibilo Hill: “Munda is yours at 1410 today.” Griswold radioed the good news to Admiral Halsey:”…Our ground forces today wrested Munda from the Japs and present it to you…as the sole owner…” Halsey responded with “a custody receipt for Munda…Keep ’em dying.”

The major objective was in Allied hands. The hardest part of the long New Georgia battle was over.
SOURCE: Cartwheel: Reduction of Rabaul; By John Miller Jr. (United States Army Center of Military History)

World War Two: New Georgia (10) After Munda

World War Two: Munda Trail (8); Griswold Takes Over