As the 1st Marines battled to secure the left flank, and as the 7th Marines fought to isolate and then reduce the Japanese defenses in the southern end of Peleliu, the 5th Marines, Colonel Harold D. Harris commanding, was charged to drive across the airfield, cut the island in two, and then re-orient north and drive to secure the eastern half of the island. Shortly after the scheduled H plus one schedule, the 2d Battalion, 5th Marines, Major Gordon D. Gayle commanding, landed over Beach Orange 2, in trace behind 3/5. It moved directly east, through the dunes and scrub jungle, into and out of the antitank barrier, and to the west edge of the clearing surrounding the airfield. Passing through the lines of 3/5, Gayle’s battalion attacked By this time, the antitank ditch along the center and right of Orange Beaches 1 and 2 was notable for the number of command posts located along its length. Shofner’s 3/5 was there, as was Harris’ 5th Marines command post. Then an advance element of the division command post under Brigadier General Oliver P. Smith, the assistant division commander, landed and moved into the antitank ditch within sight of the airfield clearing area. Simultaneously, important support weapons were moving ashore. The 1st Tank Battalion’s M-48A1 Sherman medium tanks, one-third of which had been left behind at the last moment because of inadequate shipping, were landed as early as possible, using a novel technique to cross the reef. This tank landing scheme was developed in anticipation of early Japanese use of their armor capability. Movement of this fire and logistical support material onto a beach still close to, and under direct observation from, the commanding Umurbrogol heights was an inescapable risk mandated by the Peleliu terrain. So long as the enemy held observation from Umurbrogol over the airfield and over the beach activity, there was no alternative to driving ahead rapidly, using such fire support as could be mustered and coordinated. Continuing casualties at the beaches had to be accepted to support the rapid advance. The commanding general’s concern for early momentum appeared to be eminently correct. Units on the left had to assault toward the foot of Umurbrogol ridges, and quickly get to the commanding crests. In the center, the 5th Marines had to make a fast advance to secure other possible routes to outflank Umurbrogol. In the south, the 7th Marines had to destroy immediately those now cut-off forces before becoming freed to join the struggle against central Peleliu. The movement of the 5th Marines across the airfield and to the western edge of the lagoon separating the airfield area from the eastern peninsula (Beach Purple), created a line of attacking Marines completely across that part of the island oriented both east and north, toward what was believed to be the major center of Japanese strength. The 7th Marines, pushing east and south, completed splitting the enemy forces. Colonel Hanneken’s troops, fully engaged, were generally concealed against observation from the enemy still north of the airfield and from the heights of Umurbrogol. There was a gap between Nevertheless, it was by now apparent that the D-Day phase-line objectives were not going to be met in either the south or the north. Alarmed at the loss of the desired momentum, General Rupertus began committing his reserve. First, he ordered the division reconnaissance company ashore, then, pressing commanders already on the island, he ordered his one remaining uncommitted infantry battalion, Lieutenant Colonel Spencer S. Berger’s 2/7, to land. No commander ashore felt a need for 2/7, but Colonel Hanneken said he could find an assembly area where it would not be in the way. General Rupertus ordered it to land, remarking to his staff that he had now “shot his bolt!” Ashore, it was apparent that what was needed on this hectic beachhead was not more troops, but more room in which to maneuver and more artillery. General Rupertus began to make plans to land himself and the main elements of his command group. Advice from the ADC ashore, and his chief of staff, Colonel John T. Selden, convinced Rupertus to stay on the flagship. He compromised that decision by ordering Colonel Selden ashore. By now, the shortage of LVTs was frustrating the timely landing of following waves. In consequence, neither Selden’s small CP group, nor Berger’s 2/7, could get past the transfer line in their landing craft, and had to return to their ships despite their orders to land. Into this division posture, at about 1650, Colonel Nakagawa launched his planned tank-infantry counterattack. All Marine commanders had been alerted to the Japanese capability to make an armored attack on D-Day, and were well prepared. The attack emerged from the area north of the airfield and headed south, generally across the front of the 1st Marines’ lines on the eastern edge of the airfield clearing. The attack moved directly into the 5th Marines’ sector where Boyd’s 1/5 was set in, and stretched across the southern area of the airfield. Marines in 2/1 and 1/5 took the attackers under fire, infantry and tanks alike. A bazooka gunner in 2/1’s front hit two of the tanks. The commanding officer of 1/5 had his tanks in defilade, just behind his front lines. They opened up on the Japanese armor, which ran through the front lines and virtually into his forward command group. Boyd’s lines held fast, taking the attackers, infantry and tanks alike, under fire with all available weapons. Major John H. Gustafson, in 2/5’s forward command post mid-way across the airfield, had his tank platoon close at hand. Although the enemy had not yet come into his zone of action, he launched the platoon of tanks into the melee. Accounts vary as to just who shot what, but in a very few minutes it was all over. The attacking tanks were all destroyed, and the Japanese infantry literally blown away. Colonel Nakagawa’s attack was courageous, but proved to be a total failure. Even where the tanks broke through the Marine lines, they induced no Marine retreat. Instead, the Japanese armor became the focus of antitank fire of every sort and caliber. The light Japanese tanks were literally blown apart. More than 100 With the counterattack over and the Japanese in apparent disarray, 2/5 immediately resumed its attack, moving north along the eastern half of the airfield. The battalion advanced halfway up the length of the airfield clearing before it stopped to organize for the night. It was the maximum advance of the day, over the most favorable terrain in the division front. It provided needed space for artillery and logistic deployment to support the continuation of the attack the next day. However, that relatively advanced position had an open right, south, flank which corresponded to a hole in the regimental command structure. At that stage, 3/5 was supposed to maintain the contact between north-facing 2/5 and south-oriented 3/7. But 3/5’s battalion command and control had been completely knocked out by 1700. The battalion executive officer, Major Robert M. Ash, had been killed earlier in the day by a direct hit upon his landing LVT. About the time of the Japanese tank attack, a mortar barrage hit the 3/5 CP in the antitank ditch near the beach, killing several staff and prompting the evacuation of the battalion commander. As of 1700, the three companies of 3/5 were not in contact with each other, nor with the The 5th Marines commanding officer ordered his executive officer, Lieutenant Colonel Lewis W. Walt, to take command of 3/5 and to redeploy so as to close the gap between 5th and 7th Marines. Major Gayle moved 2/5’s reserve company to his right flank and to provide a tie-in position. Walt located and tied in his 3/5 companies to build a more continuous regimental line. By 2230, he had effected the tie-in, just in time. Beginning then, the salient which the 5th Marines had carved between Peleliu’s central and southern defenders came under a series of sharp counterattacks that continued throughout the night. The attacks came from both north and south. None of them enjoyed any notable success, but they were persistent enough to require resupply of ammunition to forward companies. Dawn revealed scores of Japanese bodies north of the Marine lines. Elsewhere across the 1st Division’s front there were more potentially threatening night counterattacks. None of them succeeded in driving Marines back or in penetrating the lines in significant strength. The most serious attack came against the Company K, 3/1, position on the Point, at the 1st Marines’ left. In the south, the 7th Marines experienced significant night attacks from the Japanese battalion opposing it. But the Marines there were in comfortable strength, had communications to bring in fire support, including naval gunfire illumination. They turned back all attacks without a crisis developing. At the end of the first 12 hours ashore, the 1st Marine Division held its beachhead across the intended front. Only in the center did the depth approximate that which had been planned. The position was strong everywhere except on the extreme left flank. General Smith, from his forward command post was in communication with all three regimental commanders. The report he received from Colonel Puller, on the left, did not afford an adequate perception of 1st Marines’ tenuous hold on the Point. That reflected Colonel Puller’s own limited information. The other two regimental reports reflected the situations adequately. In addition to the three infantry regiments, the 1st Division had almost three battalions of light artillery At least two colonels on Peleliu ended their work day with firm misconceptions of their situations, and with correspondingly inaccurate reports to their superiors. At day’s end, when General Smith finally got a telephone wire into the 1st Marines’ CP, he was told that the regiment had a firm hold on its beachhead, and was approximately on the O-1 objective line. He was not told about, and Colonel Puller was not fully aware of, the gaps in his lines, nor of the gravity of the Company K, 3/1, struggle on the Point, where only 38 Marines were battling to retain the position. Colonel Nakagawa, on the other hand, had reported that the landing attempt by the Marines had been “put to route.” Inconsistently, he had also reported that his brave counterattack force had thrown the enemy into the sea. A Horrible PlaceAmong the few civilian news correspondents who chose to share the fate of the Marines on shore on Peleliu was Robert “Pepper” Martin, of Time, who furnished the following description of what it was like there:
On the second day, the temperature reached 105 degrees in the shade and there was very little shade in most places where the fighting was going on, and arguably no breeze at all anywhere. It lingered around that level of heat as the days dragged by (temperatures as high as 115 were recorded). Water supply presented a serious problem from the outset. This had been anticipated and in actual fact the solution proved less difficult than expected; the engineers soon discovered that productive wells could be drilled almost anywhere on the comparatively low ground, and personnel semi-permanently stationed near the beach found that even shallow holes dug in the sand would yield an only mildly repulsive liquid which could be purified for drinking with halizone tablets. But it continued necessary to supply the assault troops by means of scoured-out oil drums and five-gallon field cans. Unfortunately, steaming out the oil drums did not remove all the oil, with the result that many or most of the troops drinking water from the drums were sickened. When the captains of the ships in the transport area learned of this and of the shortage of water, they rushed cases of fruit and fruit drinks to the beaches to ease the problem somewhat. The water situation presented a problem even in the case of troops operating on comparatively level and open ground. Once the fighting entered the ridges, terrain difficult merely to traverse without having to fight, the debility rate shot upward so alarmingly that an emergency call was sent to all the ships off-shore to requisition every available salt tablet for issue to the 1st Marines. The statement that heat prostrations equalled wound casualties is apt to be misleading. Most of those evacuated were returned to duty after a day or two of rest and rehabilitation; hence, their absence from the frontlines did not permanently impair the combat efficiency of their units. But such numerous cases did strain the already overburdened Medical Corps elements. Special Reef-crossing TechniquesInasmuch as Peleliu’s fringing reef would not permit landing craft within 700 yards of the beach, such craft deposited tanks at the reef’s edge. There the depths permitted tanks to operate in most areas, without being submerged, but not in all. A plan was devised to form tanks into small columns, each to be led by an LVT. So long as the LVT was grounded on the reef, the tanks could follow in trace. But when the LVT encountered a depth which floated it, tanks halted while the LVT literally “felt” out a suitable shallow path. Then the tanks followed, still in small columns, and so arrived at the shore at the earliest possible hour. The technique was one of the keys to timely employment of armor ashore before D-Day was over. Two other reef-crossing innovations were used on D-Day. A large number of amphibious trailers were incorporated into the logistic plan, to be towed behind landing craft, and later, at reef’s edge they would be taken in tow by amphibian tractors. Ashore, trucks took them into tow, enabling critical supplies to be moved well forward to supply points just in rear of the fighting. Newly available crawler cranes were emplaced on barges near the reef’s edge. They could lift nets full of ammunition and other vital supplies from boats to tractors at the transfer line. Other such crawler cranes were landed early and positioned by the shore party to lift net-loads from LVTs to trucks for expeditious delivery forward. |