Islands of Destiny: The Solomons Campaign and the Eclipse of the Rising Sun

by John Prados

  Cynics nicknamed the twin-engine Betty, known to the JNAF as the Type 1 bomber, the “Type 1 Lighter,” referring to a cigarette lighter. The lack of self-sealing tanks added to aircraft fire hazards (JNAF would introduce armored tanks and automatic fire extinguishers rather than self-sealing tanks, but later), while the absence of fuselage armor and bulletproof glass increased the danger to crews. Just as losses forced Tokyo to spur aircraft production, shipping allocations required Japan to cut raw material imports, resulting in the need to reduce the weight of metal in airplanes, further limiting survivability improvements. At the same time, climactic factors—like Rabaul’s volcanic ash—increased wear just as battle damage boosted the proportion of aircraft not flyable upon landing.

  During August the JNAF would lose 214 planes in combat and 138 more in accidents, malfunctions, weather, crashes on landing or takeoff, and the like. The figures for September were 113 combat against 123 operational losses. The vast majority of this attrition took place in the Solomons, and the September figures, when incidental losses first exceeded combat ones, indicate the difficulty of flying there. They also suggest the declining experience levels of JNAF crews, and the growing shortage of mechanics. That December operational losses again exceeded the combat toll—and began a trend that continued throughout the war.

  By August 26, Admiral Tsukahara’s serviceable aircraft had declined to twenty-nine bombers and nineteen Zeroes. That was the reason for the temporary detachment of fighters from Nagumo’s carriers to Buka, as well as plans for a more comprehensive solution. By the end of the month strength had increased to thirty-eight Bettys, forty-one fighters, six dive-bombers, and three patrol bombers, plus more than a dozen floatplanes. With the addition of the 26th Air Flotilla, on September 20, Tsukahara would have 131 fighters, eighty-one bombers, four reconnaissance planes, fourteen patrol bombers, and the floatplanes. He also ordered construction of a new airfield near Buin. Tsukahara soon found it preferable to bring replacement aircraft to the Solomons rather than maintain existing planes, so serviceability rates often stood at half or less.

  At Henderson Field the stalwarts of the Cactus Air Force had problems too, including many of the same ones. The tropics often turned Henderson into a mudflat or a dustbowl, sometimes simultaneously. Rabaul had volcanic ash, Cactus black dust. The planes’ hard rubber tail wheels, designed for carrier landings, rutted the field. Planes of MAG-23, brand-new when they arrived, needed little more than gas and oil—a good thing, since most sailors of CUB-1 could not do much more than pump gas. And that was by hand, just like rearming the aircraft, which took hours. Of course, there were enemy bombs and naval bombardments to contend with, along with the Zeroes. Within a few weeks the planes were flying wrecks. On the day of the Battle of the Eastern Solomons, the thirty-one aircraft of MAG-23 had been reduced to twenty. Possibly the low point came on October 12, when Cactus had just five flyable fighters.

  Its fluctuating strength betokened Henderson Field’s importance, the pace of operations from Cactus, and strenuous Japanese neutralization efforts. Henderson Field posed a lethal threat to any Japanese force within about 300 nautical miles. That remained the standard engagement range for several reasons: Radio beacons and air navigation restricted long-distance flight; the longest-legged Allied aircraft were based back in the New Hebrides or on New Guinea and could not operate from Guadalcanal; aircraft actually at Cactus had limited ranges. So long as Henderson functioned, it inevitably dominated Japanese calculations. Conversely, the struggle to maintain its effectiveness became a major focus of all Allied activities. Airpower was the hammer that would smite the enemy.

  The SOPAC effort went to a new level beginning in September. The missing fighter and dive-bomber squadrons of MAG-23 arrived with thirty-one fresh planes and its commander. Transport William Ward Burrows brought MAG-23’s maintenance men to take over from the sailors. Japanese aircraft sank the ship, but only after the mechanics and critical equipment were unloaded. The liaison flight from Efate on the evening of September 4 carried Brigadier General Roy S. Geiger, who took charge of the air effort. A couple of transports also delivered the Seabees of the 6th Naval Construction Battalion. Geiger put the Seabees to work on a new airstrip, Fighter 1. One thing Seabees did excellently was build and repair, and that became central to the Cactus Air Force. Marston matting—perforated aluminum plates (named for a town in North Carolina where they were first manufactured)—helped solve the mud problem. The Seabees pre-loaded dump trucks with earth to fill a typical crater, and precut Marston matting to replace runway panels, discovering they could fix the damage from a Japanese 550-pound bomb in about forty minutes. Repairing the runway was vital.

  The Cactus Air Force developed its own routines. Air raids typically arrived within a couple of hours either side of noon, due to the required flying time from Tsukahara’s bases, plus the need for the JNAF planes to roost before dark. Coastwatchers usually warned of the raids, giving Geiger’s pilots an hour or more notice. The Burrows delivered radar, enabling Henderson to pick up incoming aircraft and update the interceptors. Unlike carrier combat air controllers, who vectored fighters to their prey, the Marines developed “vertical interception.” This was also necessary because base radio, which worked poorly amid tropical humidity, could hardly be heard more than about twenty miles away. Scrambling fighters clawed for altitude, listened to the chatter until they could see the Japanese beneath them, dived in pairs to loose a burst of fire, then climbed to repeat the exercise. Machine guns had to be wiped clean of lubricants before fighters took off or the guns froze at altitude and jammed when fired. Pilots made no effort to maneuver against agile Zeroes, relying instead on their diving speed.

  Offensive operations developed their particular rhythm too. The natural counter to nightly Japanese naval activity was to put up experienced pilots to try to get in some licks. At dawn enemy task groups would be retreating, and at dusk approaching, so scout bombing became axiomatic for morning and afternoon searches. And the days would be punctuated by the noontime Japanese bombing and the nighttime shelling. Some of the most dangerous episodes occurred when the JNAF broke pattern, such as in late-afternoon or early-morning strikes. Pilots got whatever sleep they could. When the JNAF bombed—despite the difficulties noted earlier—there was always the possibility they would succeed in laying a tight pattern, cratering the airfield and wrecking parked planes. Thus the bombers and the Army Air Force P-400s, useful mainly for close air support, would scramble just to be safe. On August 28, for example, the Cactus Air Force might have been smashed save that its bombers took to the air. Harold L. Buell was one of the pilots of VB-10, the Enterprise Dauntless unit temporarily at Henderson. “Like a broken record,” he recalled, “the routine of noon bombings, morning and afternoon searches, night shellings, and harassment missions went on day after day with little variance.”

  The Allied analog to the unavoidable dribbling of reinforcements that hampered Japanese efforts was undoubtedly the flow of aircraft and crews to the Cactus Air Force. Of the twenty-one pilots who arrived with the original Marine Fighter Squadron 223, only nine left with it in October. Of the others four men were wounded or invalided off the island, seven were killed, and the fate of the last pilot is not clear. Leaders considered that pilots and crews could last a month at Cactus. General Geiger asked the high command for eighteen F-4Fs and another eighteen SBDs every ten days. The pressure was tremendous. Pilot Buell, injured in a nighttime crash landing, was sent up again after one day’s rest.

  In early September, Cactus had eighty-six pilots and sixty-four planes. By the tenth only eleven of the thirty-eight F-4Fs that had been delivered were operational; one day alone eight fighters crashed on takeoff. “At this rate we can whip ourselves without any assistance from the Japs,” a pilot groused. On October 14, Henderson had seven flyable Dauntlesses plus a number of Wildcats, with more than seventy aircraft in need of repair after a major Imperial Navy bombardment. During that interval Cactus had been reinforced by rep
lacement aircraft, flights of Army warplanes, a Marine dive-bomber squadron and one of fighters, plus Navy fighter, torpedo, and dive-bomber squadrons from the carriers Wasp and Saratoga. The redoubtable Marine Fighter Squadron 212, many of whose pilots had already fought at Cactus under VMF-223, entered the fray with its own banner. Eventually eight fighter, twelve dive-bomber, and two torpedo squadrons would fight from Henderson. Guadalcanal was a meat grinder for aircraft.

  It was a constant struggle to keep the planes flying—when they had gasoline and ordnance. At one point in October fuel was in such short supply that liaison planes were airlifting gas to Cactus. An index of the significance of this is that a C-47-type plane could carry a dozen barrels of aviation gas, each sufficient to fuel one Wildcat for one hour. Oxygen bottles for the Wildcats ran out within days of arrival and, along with critical spare parts, became a staple on the transport planes from Efate. One time the Japanese plastered the “Pagoda,” Henderson’s control tower, another its main ammunition dump. Shelling destroyed the food dump in October, scattering cans of SPAM everywhere. They would be found at odd moments for a long time afterwards.

  Japanese airmen may have been able to relax with geishas at Rabaul, but there was no such luxury for the Cactus Air Force. The men were on half rations until mid-September, and that was made possible by captured provisions. Even later they existed on dehydrated potatoes and SPAM or cold hash. When carrier Wasp embarked a flight of Marine planes for delivery, the pilots were enchanted by its quality food and accommodations, only to be subjected to the reality of Guadalcanal. The tentage, bedding, and blankets were all captured from the Japanese. On Cactus there was no way even to wash laundry except take it to a stream and do it yourself. Ensign Buell was washing his one day in the knee-deep Lunga River when the bombers came. He and everyone else streaked for cover buck naked.

  The pilots’ experience would be of a piece with that of ordinary Marines. Herb Merillat is a good example. Lieutenant Merillat held an anomalous position in the 1st Marines’ hierarchy. A lawyer by trade, he had been recruited to a Marine information office, then assigned to Vandegrift’s division, which seconded him to its intelligence staff, but as a historian. Merillat was in a good position to observe the inner workings of Vandegrift’s staff, kept careful notes and a diary, and used them to produce a wartime account of Guadalcanal he titled The Island. That book takes the campaign as a simple matter of units and battles and says little of the trials and tribulations the Marines faced. But Merillat’s later reflections, which include many entries from his diary, tell another story. Soon after the landing Colonel Randolph Pate, Vandegrift’s supply officer, informed the general they had food for five days. A more careful review on August 15, after the Marines had already gone on half rations, found field rations for seventeen days, C rations for three, and captured food sufficient to extend that another ten.

  That day Marines grabbed some wicker baskets that had been airdropped to the Japanese in the bush. They contained mimeographed leaflets—unwelcome to the Americans—designed to boost morale with news of the Imperial Navy’s victory at Savo Island. But what the men really wanted from the baskets was their canned goulash. Some units were subsisting on rice and fish heads. Captain Nikolai Stevenson, commanding the 1st Company, 1st Marine Regiment, recalled the meager diet. Almost everyone got dysentery. On August 12, Corporal James R. Garrett of I Battery, 11th Marines, enjoyed his first hot food in days. A few days after that, Glenn D. Maxon, a lieutenant in the 1st Marines, noted a big food shortage. The men exploded grenades in the Lunga River to catch fish. Like many others, Maxon later contracted malaria. It was common to lose dozens of pounds of body weight on Guadalcanal. When Marines launched a raid into the Japanese rear area early in September, every man loaded himself down with tins of crab and beef taken from enemy stocks before destroying the rest. The raiders also brought back twenty-one cases of beer and seventeen half-gallon bottles of sake. The admirals had code names, but Marines called their invasion “Operation Shoestring.”

  War correspondent Richard Tragaskis went out with a Marine patrol. When they halted, instead of resting the men scavenged for good coconuts. When they found a dead Japanese, one needled their cook, attributing the enemy’s demise to Marine chow. Dinner was a candy bar. When Tregaskis left Cactus late in September, one of the reasons, he records, was that he had worn out his last pair of shoes. On September 18 the food problem was finally solved with a large shipment of supplies—so big it included post exchange materials. Suddenly Marines had tins of white shoe polish nobody wanted or could use. Typically, the cargo ships were able to unload less than two-thirds of their freight before they pulled out, but even so, arriving Marines ate beef stew that night.

  Marines suffered a big setback soon after the Watchtower landing. Vandegrift’s intelligence chief, Lieutenant Colonel Frank Goettge, learning from a prisoner that a group of the enemy wanted to surrender, insisted on leading the twenty-six-man patrol sent to find them. Lieutenant Jack Clark of the naval operating base provided the Higgins boat that carried the patrol to its doom. He never forgot joking with the Marines before they left, and since this was a routine mission, the landing craft returned after dropping them off. The Marines were ambushed. Goettge sent a runner for help, but the man had to swim more than four miles through shark-infested waters, and by then it was too late. Scuttlebutt had it that one Marine’s tongue was cut out and another’s hands cut off. Only three men returned, including the runner. Frank Goettge was not among them. Goettge’s disappearance was a potential disaster far greater than the simple loss of an intelligence officer, bad as that was, because he knew about Ultra. It is fortunate the bloodthirsty Japanese killed these men rather than interrogating them.

  Goettge’s loss was offset to a degree by the appearance of Martin Clemens, the coastwatcher, who joined the Marines and worked closely with Vandegrift’s intelligence people, setting up a radio post that gave Cactus direct contact with Mason, Read, and other key observers. Clemens, joking with staff officers, admitted that he too had had just a few tins of food left when he appeared. The dangers were starkly demonstrated less than two weeks later. Clemens sent his trusted indigenous aide, Sergeant Major Jacob Vouza, to scout a suspected Japanese lookout post. The enemy caught Vouza with a small American flag he carried to identify himself. The Japanese tied Vouza to a tree and tortured him but got no information. Then they bayoneted Vouza and left him to die. The coastwatcher managed to free himself and regain U.S. lines. Vouza distinguished himself again later as a guide for the Marine Raider Battalion.

  Vandegrift soon got his own Ultra radio section. Lieutenant Sanford B. Hunt ran the (captured Japanese) radio that kept Vandegrift in touch with SOPAC. He delivered what intel the theater command sent. A radio intelligence element arrived in mid-September with the convoy that brought the 7th Marine Regiment. This was a tiny two-man direction-finding unit that helped generate target information for the Cactus Air Force. A couple of weeks later they were augmented by a full mobile radio detachment led by Lieutenant Commander Daniel J. McCallum. The “cryppies” knew their unit as Station AL. Marines like Herb Merillat knew them more informally as the “Cactus Crystal Ball.” McCallum flew up from Espíritu Santo with four sailors on one of the daily flights. The C-47 could not carry all their gear, so they took half and left the rest to arrive a month later by ship. McCallum brought two special-keyboard typewriters that could print Japanese characters, several Hallicrafters radio receivers, and a Copek—the encryption device that provided secure communications within the Allied radio intelligence net. As a Japanese-language officer, McCallum could translate decrypted enemy messages. Naturally there were teething troubles. Chief Radioman James J. Perkins, who arrived with the direction-finding team, lost a receiver in the confusion of unloading the convoy under Japanese bombardment. Months later he found its shell on the beach—Marines had been using the thing as an oven.

  With just one, later two, receivers or “posts,” the mobile radio unit had limited
capacity, but it put Cactus on the circuit for traffic analysis and decrypts obtained elsewhere. Some cryppies groused that McCallum was not a technician and wondered why he was there. His shortcomings put much of the weight on Lieutenant Charles “Homer” Kisner. Commander McCallum could use JN-25 codebooks, and, coached by Kisner, did his best. The commander passed his data to Marine Lieutenant Hunt, who continued to be Vandegrift’s Ultra liaison. Among their most important information was a daily list of radio fixes on Japanese ships, used to target air strikes. The Cactus Crystal Ball established itself in a tunnel dug into a hillock below Henderson Field.

  Though the JNAF concentrated on Henderson Field, their bombing affected everyone. On August 16, bombs wounded five Marines of Corporal Garrett’s I Battery. On the twenty-fifth, enemy ordnance missed the air base but struck near Vandegrift’s nearby headquarters. One bomb left a fifteen-by-twenty-five-foot crater, sprayed the command post with shrapnel, and sent shards into the general’s own tent. Lieutenant Merillat, who had taken up residence under a coral outcropping, traded that for a slit trench when bombs showered him with bits of broken coral. As another 1st Division Marine said of a later U.S. invasion, coral comes high.

  Beyond enemy action were the dangers of tropical disease, which were only magnified for men losing weight on half rations. In September, when the 1st Marine Division had incurred only about a thousand casualties, twice that many were laid up with malaria or dysentery. And that was before the rainy season brought tons more mosquitoes and widespread disease. Vandegrift’s whole command went on a preventive course of antimalarials (quinine, but mostly atabrine) that month. A rumor quickly arose that atabrine caused impotence, and Marines resisted taking it. Even having medics stand in chow lines to make Marines quaff the pills before receiving food proved none too successful, judging from medical casualties. More than 8,500 of Vandegrift’s Marines were felled by malaria, some more than once, nearly three-quarters of them in the period from mid-September to December. When the 7th Marines arrived, it had the only battalions anywhere near full strength.