An hour before dawn on 7 August, Dog Day, Fletcher’s three TF-61 carriers (with Noyes, CTG-61.1, in tactical command) closed Point Victor, thirty miles west of Guadalcanal. The main Watchtower objectives, Lunga on the big island’s central coast and Tulagi eighteen miles north, were sixty to seventy-five miles northeast of the carriers. The waning moon shone feebly in the clear skies, but a dark wall of clouds barred the way to the target. In fact the door was wide open, and the key components of the invasion force had already come in. After rounding the west tip of Guadalcanal, the two parts of Turner’s invasion force (TG-61.2) had passed on opposite sides of the mountainous mass of Savo. Scott’s Yoke Group aimed for Tulagi nestled within the south coast of Florida Island, while Turner’s own X-Ray Group cut directly across the large coastal bight bound for Lunga Point.1
Fletcher’s own TU-61.1.1 (Saratoga) constituted the point of the arrowhead, with Kinkaid’s 61.1.2 (Enterprise) five miles off its port quarter and Noyes’s 61.1.3 (Wasp) in the corresponding slot to starboard. “Blue flashes from engine exhausts flickered and burned—first only here and there, but all at once as if at the flip of a massive switch they covered two-thirds of each flight deck.” More than one hundred aircraft noisily warmed up on deck. Promptly at 0530 the carriers swung southeast into the eighteen-knot wind. The air operations plan called for simultaneous dawn attacks by ninety-three aircraft (forty-four fighters, forty-eight dive bombers, and one torpedo plane). The Enterprise’s initial deck load included eight fighters for screen combat air patrol over the distant cruisers and transports. Inevitably in the darkness given inexperienced aviators and three carriers in close proximity, the rendezvous miscarried. Flight leaders gathered whom they could and headed out to their assigned targets. Once the first deck load got away, the carriers spotted planes for follow-up attacks, combat air patrol, ground support, searches, and inner air patrol on lookout for subs. Growing daylight revealed the “inspiring sight” of “three carriers steaming at high speed in line abreast at intervals only great enough to prevent mutual interference of their destroyer screens.”2
Purple Base on board the transport Neville controlled air support flights in the Yoke sector. To southward, Orange Base in Turner’s flagship McCawley directed flights against Lunga. Fighters assumed two combat air patrol stations: the carrier combat air patrol above the carriers and the screen combat air patrol between Tulagi and Guadalcanal. Controlled by Lt. Henry A. Rowe’s Red Base in the Enterprise, the carrier combat air patrol comprised sixteen to twenty-four F4Fs in two-hour patrols. The fighters checked out suspicious radar contacts but spotted no enemy planes. The screen combat air patrol comprised eight to sixteen F4Fs directed by Lt. Robert M. Bruning, Black Base FDO in the Chicago in Crutchley’s TG-62.6. In between flight operations Noyes repeatedly reversed course in a rough fifteen-mile square to maintain relative position in relation to the targets while keeping twenty-five to thirty miles from Guadalcanal’s west coast. Strike planes and fighters shuttled back and forth from the invasion areas and combat air patrol stations as the carriers tried to hold to Noyes’s intricate timetable.
Turner, seeing what he took to be seaplanes rising off Lunga, announced at 0611, “Enemy planes now taking off.” No further word reached the carriers until 0647, when they were told incorrectly that U.S. planes landed at Lunga after destroying enemy fighters. No enemy aircraft were present. Kinkaid later explained the difficulty in piecing together the situation based only on radio transmissions, returning pilots, message drops, and visual signals. Boscoe Wright, Fletcher’s cruiser leader, called communications “terrible.” The carriers “knew nothing of what went on at Guadalcanal, though occasionally they picked up scraps of TBS.” Even worse for Fletcher personally, “The Saratoga couldn’t get anything on her radio or TBS” and “had to depend on the Minneapolis.” Turner’s communications were erratic at best and not all that informative.3
The invasion seemed to be proceeding well. A report at 0759 told of an unopposed landing at Tulagi, where Wasp planes claimed (and got) thirteen planes torched on the water. Saratoga pilots newly back from Lunga spoke of little return fire. At 0824 Turner confirmed the Tulagi landings took place on time with “no beach opposition.” The marines landed first on Florida, then Tulagi, where initial opposition grew heavy, and late that morning on tiny Gavutu, also against fierce resistance. A causeway linked Gavutu to Tanambogo, another stronghold. By 0913 when the First Marine Division splashed ashore at Red Beach on Guadalcanal, bombardment and air strikes scattered the defenders almost beyond redemption. Unfortunately Vandegrift did not understand he could seize the airfield immediately. At 0953 Turner reported that the landing on Red Beach was unopposed and that ships’ gunfire silenced Gavutu island. The dawn search northwest beyond Santa Isabel turned up just a small tanker east of New Georgia. Fresh from coordinating strikes over Lunga, Felt landed on board the Sara at 1058 and told Fletcher he had trouble finding any resistance at Guadalcanal at all.4
Although the morning went quite well despite heavy fighting at Tulagi and Gavutu, the question remained when Japanese aircraft would counterattack. MacArthur offered no clue how his crucial interdiction fared against the Rabaul airfields. Perhaps not well, for at 1025 Cincpac cautioned Ghormley and Fletcher that Rabaul unleashed eighteen bombers and seventeen fighters against the invasion. Nine minutes later Pearl advised, “Enemy subs are on move to attack Blue Occupation Force at Tulagi.” Sophisticated radio intelligence informed Cincpac of these distant events, but simple eyesight proved vital, too. About the same time as the Cincpac warnings, a plain-language message transmitted on a special frequency advised: “24 bombers headed yours.” The call sign identified the sender as PO Paul E. Mason, an Australian coast watcher hiding near Buin in southern Bougainville, three hundred miles northwest of Lunga.5
Rabaul and Tokyo awoke to grave peril in the Southeast Area at 0630, when the Tulagi garrison radioed: “Under intense enemy bombing.” In the short run only Admiral Mikawa’s Eighth Fleet and Rear Adm. Yamada Sadayoshi’s 5th Air Attack Force could respond. Within two hours they knew the worst. A huge amphibious force with at least one carrier and a battleship assaulted Tulagi and Lunga. Mikawa gathered his heavy ships to sortie that afternoon, directed subs to Tulagi, and arranged for naval troops to leave for Guadalcanal in the next few days. Most of Yamada’s twin-engine land-attack planes were preparing to strike the worrisome airfield at Milne Bay on the tip of New Guinea, but the Solomons must come first. His aviators favored aerial torpedoes against ships, but high explosive “land” bombs already crammed the bombers perched at Rabaul. Yamada feared to delay to rearm his planes with torpedoes lest they be caught on the ground, just as the U.S. Navy planners hoped MacArthur would do. Three land-attack planes left to examine the waters around Tulagi and Guadalcanal. Yamada needed to know where the U.S. carrier was in relation to the target area. He missed a great opportunity. For three hours a lookout station on the southwest coast of Guadalcanal watched two carriers and a battleship cavort to westward, but it could not establish radio communication with headquarters. At 0950 twenty-seven Type 1 land-attack planes under Lt. Egawa Renpei and eighteen Zero fighters started taking off from Rabaul to hunt U.S. carriers. Nine Type 99 carrier bombers of the newly arrived 2nd Air Group were to comprise a second wave to bomb transports and U.S. troops already ashore. Sending them was a quixotic gesture. Armed with just two small 60-kilogram bombs each, the carrier bombers lacked the range to fly to Guadalcanal and return. They were to ditch at Shortland south of Bougainville. Thus Yamada would expend all nine dive bombers but hopefully not their crews.6
Foreknowledge of the incoming aerial attack did not necessarily mean Noyes’s TG-61.1 could mount an adequate defense. The two threatened locales, carriers and invasion force, were more than sixty miles apart. Fletcher rightfully accorded the three precious carriers top priority, but even they spread across twenty miles while individually handling flights. At 1130 the carrier combat air patrol numbered twenty-four F4Fs, with twenty-four being readied to relieve them. The situation over the transports was not as good. The screen combat air patrol stood at fourteen F4Fs, but by 1300 all must depart for lack of fuel. Only eight F4Fs were to replace them. Given the tricky mechanics of carrier operations, the flattops simply could not launch or recover aircraft on demand, as Nagumo learned to his vast dismay at Midway. Bombers flying ground support missions could tie up flight decks and delay the launch of combat air patrol reinforcements. The screen combat air patrol needed thirty minutes to reach station, and dwindling fuel often forced it to leave early. TG-61.1 started advancing southeast roughly parallel to the Guadalcanal coast and twenty to thirty miles distant to keep sixty to seventy-five miles from Lunga and Tulagi. The wind unfortunately abated, forcing sprints above twenty-eight knots to conduct flight operations. That was hard on the Wasp’s uncertain turbines and enormously costly in fuel. The Saratoga’s next fighter launch, set for noon, comprised eight F4Fs for carrier combat air patrol and four Wildcats on screen combat air patrol. Noyes authorized four more F4Fs to depart seventy-five minutes early to reinforce the screen combat air patrol. The sixteen F4Fs took off at 1203. Four fighters headed out from the Enterprise on screen combat air patrol, along with a pair to relieve the section escorting the strike coordinator over Lunga. Four F4Fs flying ground support received no targets (neither did the previous batch) and circled as a sort of low-altitude combat air patrol. Noyes suspended fighter ground support flights and added them to the carrier combat air patrol.7
High mountains blocked the Chicago’s radar. The parade of planes only partly equipped with IFF radio transmitters likewise hampered Bruning’s efforts. Nevertheless he directed the four Enterprise F4Fs northwest after a large bogey. It is doubtful the Enterprise radar ever picked up the same contact, but Rowe, the carrier FDO, overruled Bruning. He diverted the same four fighters south still well short of Savo. Rowe also hastened northward four carrier combat air patrol F4Fs under Lt. (jg) Theodore S. Gay. The trap miscarried. The four F4Fs previously on screen combat air patrol ended up back at the carriers, their place to be taken by Gay. At 1311 the Enterprise, following the flight schedule, lofted Lt. (jg) Gordon E. Firebaugh’s four VF-6 F4Fs for screen combat air patrol, and another two for the Lunga strike coordinator. Fourteen VT-3 TBFs followed to patrol ahead for lurking subs. Their departure also freed up the flight deck for rapid combat air patrol recovery.
Lacking carrier contacts and the freedom to seek them on his own, Egawa simply kept his bombers pointed southeast toward the ships known to be off Lunga. In fact he caught another bad break when Rowe’s quarry, the land-attack plane searching south of Guadalcanal, ran into thick clouds after 1230. It unwittingly discontinued its mission only a few miles short of the carriers, but scouting eastward, it discovered twenty-seven escorted transports off Lunga. For the time being only the eight VF-5 F4Fs patrolling at twelve thousand feet defended the Amphibious Force. At 1315, near Savo, Lt. James J. Southerland suddenly encountered Egawa’s bombers descending through the clouds. His four F4Fs and the four led by Lt. Herbert S. Brown tore into the intruders in a fierce air battle that cost five Wildcats.8
At 1315 the carrier combat air patrol stood at twenty-four F4Fs and the screen combat air patrol eighteen, including ten VF-6 fighters under Gay and Firebaugh en route. Fletcher ordered TF-61 to prepare to repel air attack. Soon after, he heard a report that twenty-five enemy aircraft bombed the transports. Sixteen Wasp F4Fs raised the carrier combat air patrol to forty fighters. The ratio changed in the next half hour to forty-four Wildcats on carrier combat air patrol and twenty-nine on screen combat air patrol. Finally realizing no bogeys threatened the carriers, Rowe rushed eight F4Fs north to Tulagi, but they arrived too late to fight. Gay and Firebaugh bravely harassed the withdrawing Japanese planes far up the Solomons, but lost four of their number. For the time being Fletcher and Noyes only knew of a ruckus over the invasion area and continued the air support flights on schedule. The three battered survivors of the VF-5 combat air patrol, including one badly wounded pilot, returned to the Saratoga. After taking down their statements, the VF-5 yeoman hastened to flag plot to read them to Fletcher, where he enjoyed “a cup of coffee with the Old Boy & everything.” The VF-5 pilots estimated thirty-six to fifty twin-engine medium bombers, strongly escorted, sliced through defending F4Fs and bombed the transport force anchored off Lunga. Perhaps three bombers and two Zeros went down. One ship was on fire. About twenty-five bombers wheeled eastward over Florida Island and disappeared north. Nothing was known of the Gay and Firebaugh flights until later, when six surviving F4Fs landed.9
As many as fifty medium bombers assailing Guadalcanal and Tulagi meant, as Maas wrote that night, “MacArthur (Dug-Out Doug) did not neutralize Rabaul.” Japanese air responded to the invasion in strength, if not as yet effectively. The land-attack planes claimed one destroyer sunk and a hit on a transport. Actually they bombed Crutchley’s screen off Lunga without damage. Perceiving land-based Zeros were possibly superior in performance even to those on the carriers, the U.S. Navy long wondered how its carrier planes would fare against them. The truth proved demoralizing. Seventeen Zero escorts engaged thirty-four U.S. aircraft (eighteen F4F-4s, sixteen SBD-3s). U.S. Navy credits were seven twin-engine bombers destroyed (plus five probable) and two fighters. Nine F4Fs and one dive bomber went down. Despite fierce opposition, Japanese losses proved surprisingly light. Four land-attack planes fell to fighters, one ditched off Buka, and another crashed back at Rabaul. VF-6 destroyed two escort Zeros. The ten Zeros that returned to Rabaul (five more stopped at Buka) completed an extraordinary eight-hour mission hitherto unprecedented for single-engine fighters. The land-attack crews claimed fifteen Grumman fighters, and the Zero pilots no less than thirty-six (including seven unconfirmed), plus seven U.S. carrier bombers. The few U.S. aircraft actually brought into battle greatly impressed their opponents, who reckoned upward of ninety intercepting Grummans. Even the Japanese official history declared its strike battled sixty-two U.S. Navy fighters.10
At 1400 Turner warned Fletcher of enemy dive bombers as well. Another partially garbled message reported about twenty-five planes withdrawing southeast at three thousand feet. As yet Fletcher and Noyes could not confirm the presence of the dive bombers. Jumpy ship lookouts might have mistaken SBDs as hostile planes. Inasmuch as dive bombers could not normally fly the nearly twelve-hundred-mile round trip between Rabaul and Guadalcanal, perhaps they came from a carrier. That would have meant that radio intelligence, God forbid, completely missed a sortie of first-line flattops from home waters. Unwilling to go to that extreme, Fletcher wondered whether the culprit might be the auxiliary carrier Sowespac located the previous day north of Rabaul. At 1455 he suggested to Noyes and Kinkaid there be a search the next morning “toward Rabaul in view enemy CV reported there.”11
Radio transmissions copied at 1455 described a dive bomb attack on Guadalcanal, this time battled by the twenty-four F4Fs on screen combat air patrol. The nine carrier bombers flew south along the northern boundary of the island chain, and, shielded by mountains from the Screening Group’s land-blocked radars, they achieved complete surprise. A 60-kilogram bomb killed twenty-one men in the destroyer Mugford. Fifteen VF-5 and VF-6 F4Fs claimed fourteen dive bombers, five more than were there, and did bag five. The four survivors reported a light cruiser on fire. Yamada’s one-way mission actually damaged one destroyer, but did pay off another way. The appearance of carrier-type aircraft so far from home multiplied the confusion on the U.S. side. Kinkaid commented to Noyes that the “presence [of ] dive bombers indicates possibility enemy CV this vicinity” and suggested a search. Concerned only three Enterprise F4Fs could scramble, he asked the Saratoga and Wasp to reinforce the combat air patrol. Only one example of the many appalling lapses in communications, Fletcher received these 1514 and 1527 TBS messages at 1555 and 1558 respectively, possibly via the Minneapolis. Fletcher certainly shared Kinkaid’s fears of possible carrier opposition. That evening Maas mused: “If [enemy] carrier got to rear (CV Dive-Bombers) then search and interdiction by MacArthur very poor (failure) and same by McCain not very good.” Fletcher noted the dive bombers were last seen retiring west instead of northwest directly to Rabaul. That only heightened suspicion of a carrier. He anticipated an all-out air attack the next day. By moving farther away from Rabaul, he might reduce the vulnerability of his own carriers to a double punch from land and carrier-based planes. At 1600 he told Noyes to shift the predawn launch point one hundred miles southeast from the one that day. Noyes chose Point Roger, thirty-two miles southwest of Cape Henslow (seventy miles from Lunga, eighty-five miles from Tulagi), and devised the maneuvers to bring the carriers there at 0600 on 8 August.12
Fletcher’s 1600 message also directed Noyes to “make early search,” by which he meant prior to sunset at 1818. At the same time he had the Saratoga reallocate a dozen SBDs from a scheduled ground mission to a brief sweep for a “suspected carrier” fifty to ninety miles northwest. Calls for air support waned, and some planes already returned without being allocated targets. In the meantime Noyes suggested to Fletcher at 1627 that the dive bombers were “probably land based from Rabaul via Buka or Kieta.” That was correct in the sense Rabaul was actually the point of origin, but no one on the U.S. side appreciated that Yamada simply expended his carrier bombers. However by informing Fletcher that Kinkaid “has already been told to search,” Noyes demonstrated he, too, hedged his bets. Kinkaid switched to northward eight of fourteen Enterprise TBFs patrolling ahead for subs. None of the dusk searches found enemy ships or planes because none were there.13
Task Group 61.1 track chart, 7–9 August 1942
Still concerned over possible carrier opposition, Noyes also modified the normal rotation of carrier duties on 8 August. The Wasp was to search northwest in the direction of Rabaul “primarily for reported Cast Victor” and handle standby combat air patrol, while the Enterprise took primary responsibility for the carrier and screen combat air patrols, and additional ground support missions “if requested.” In contrast, Ramsey was to keep the Saratoga in strict reserve, spotting fighters and an attack group for immediate launch should a carrier be discovered or radar detect an inbound bombing attack. At noon on 8 August the carriers would exchange roles, the Enterprise handling the afternoon search, Sara assuming the main combat air patrol and ground support load and the Wasp going into reserve. Thus Noyes anticipated the Saratoga’s flight deck being locked up tight from dawn to noon on 8 August, but there would be unforeseen complications.14
By sunset on 7 August TG-61.1, cruising twenty miles south of Guadalcanal, had come nearly as far east as Cape Henslow, near where flight operations were to resume the next dawn. The afternoon passed without further enemy contact, but three fighters could not find their way back from the invasion area and ditched. That day the three flattops conducted an amazing 703 takeoffs and 687 landings, the most ever in one day up to that point. Nine F4Fs and one SBD fell to enemy action, and operational accidents cost five F4Fs and one SBD. Another Wildcat had to be jettisoned because of crash damage, and five were badly shot up. The exhausted carrier aviators had the satisfaction of knowing “Terrible Turner” praised “all air squadrons for their excellent work.” Deeply surprised the carriers were not attacked, Fletcher, Noyes, and Kinkaid all sensed they got away with something so far. Maas wrote in his diary, “Expected attack on our carriers did not materialize tho we were fully expecting one. Tense all day.” Later that evening Fletcher learned that as of 1600, radio direction finders at shore bases placed Japanese subs just to eastward between Malaita and San Cristóbal. Thus TG-61.1 was wise to undertake a seventy-mile dogleg southwest during the night to clear the area.15
Kinkaid’s TU-61.1.2 went its own way southward. That afternoon he had been “very much annoyed” to discover one of his destroyers, the Gwin, was “dangerously low on fuel.” He later complained she was never topped off while on detached duty from 30 July and 3 August with Fletcher’s TU-61.1.1. On 3 August before being returned, the Gwin had in fact fueled along with all the TG-61.1 destroyers and was at 82 percent capacity. Her log for 6 August showed 52 percent. The Gwin was not the only one of Kinkaid’s destroyers low on fuel. The Grayson was down to 55 percent fuel on 6 August (42 percent on 7 August); the others had little more. Kinkaid cautioned Fletcher on the seventh that most of his destroyers had oil for two days at fifteen knots, but the Gwin only enough for somewhat over a day at that speed. It appears in this instance Kinkaid greatly overstated the deficiency in that he erred in using fifteen knots instead of twenty-five knots. He later explained that if he had been “well out to sea,” one of his heavy ships would have immediately fueled the Gwin, but “that was not advisable” because of possible air and sub attacks. Therefore he secured permission from Fletcher to go off alone, fuel the Gwin, and rejoin the other carriers at Point Roger just prior to dawn. The North Carolina commenced fueling the destroyer in the “pitch black night” without showing any lights. After “efficiently” executing the “slow and delicate operation,” the Gwin by midnight had drawn 56,642 gallons of fuel oil, which restored her to 79 percent of capacity.16
Lacking detailed summaries from Turner, Fletcher still knew very little of what actually occurred in the landings. The X-Ray Group at Guadalcanal seemed to fare well, with few calls for ground support. Fresh from duty as the Yoke strike coordinator, Felt described the fighting still raging on Tulagi and tiny Tanambogo. Confirming the serious nature of the situation there, the carriers monitored a message that evening from Scott to Turner advising that neither Tulagi nor Tanambogo were taken. “Unloading not yet commenced; estimated time complete unknown.” Felt told Ramsey that Turner himself requested a bombing mission against Tanambogo at dawn on 8 August, and that he promised to do it. However, Kinkaid’s overnight absence greatly complicated the situation. The Enterprise was to be duty carrier and should handle that attack, but Fletcher rightly refused to break radio silence (TF-16 was already too distant for TBS) to alert Kinkaid and transmit the lengthy details regarding times and targets. Instead at Ramsey’s urging, he authorized the Saratoga to furnish the flight, even though Noyes directed she remain in reserve. The Saratoga prepared a strike of eighteen SBDs and eight bomb-armed VT-8 TBFs to depart just after 0600.17
That evening Maas critiqued air operations based on what he learned from Felt and others. It was obvious that dividing shore-based air between MacArthur and McCain was not working. “Army air-operating over water must be under Navy (or Marine) command.” Likewise Maas faulted the arrangements for air support in the target area. “Air operations in support of Marine landing operations should be headed up by Marine aviation.” He deplored that Turner at Guadalcanal insisted also on running the air show over distant Tulagi, instead of letting Purple Base do its job. “Air missions broke down because Turner (Adm.) gave our planes no mission when they reported on station over Tulagi. When the Air Group Comdr. [Felt] in the air over area gave them missions (targets) in the absence of orders from Director Control on McCawley, he was reprimanded by Turner, who couldn’t even see what was going on.”18
MacArthur’s much-anticipated interdiction of Rabaul air power amounted to little. Fletcher learned that evening that eight B-17s reportedly “put all bombs” on the airfield that morning, but with no estimate of damage. Maj. Gen. George C. Kenney, the new commander of the SWPA Allied Air Force, had hoped to employ twenty Fortresses in the AAF’s greatest concentration yet in the southwest Pacific. In truth it was only on the afternoon of 7 August that thirteen B-17s bravely bombed Vunakanau Field just west of Rabaul and lost one bomber to an estimated twenty-six fighters. The great hope vested in MacArthur’s heavy bombers proved sadly misplaced. The raid cost Yamada no planes either aloft or on the ground, and the runways were repaired before the Guadalcanal strike returned. Late that afternoon nine land-attack planes arrived from Tinian, with eighteen more set to follow the next day. In a message later deciphered by Allied intelligence, Yamada listed thirty land-attack planes available for the eighth after starting with thirty-two the day before. When Kenney read the intercept he boasted to MacArthur that on 7 August his B-17s destroyed or damaged the other hundred or so bombers he believed had been there. His self-serving postwar memoir crowed that his planes blew up seventy-five of 150 bombers parked wingtip to wingtip on Vunakanau Field and forestalled further air attacks against Guadalcanal. In truth 7 August 1942 was no replay in reverse of the 8 December 1941 debacle at Clark Field, because no land-attack planes even remained at Vunakanau to be bombed. It was a ridiculous claim, a disservice to the gallant AAF flyers who flew the perilous long-range mission to Rabaul.19
The Sowespac update on the night of 7 August also revealed that the B-17s had seen six large and fourteen smaller ships in Rabaul harbor. In nearby St. George’s Channel, six more ships steamed southeast toward the Solomons, while northwest of Rabaul, one cruiser, three light cruisers, and a destroyer headed west at twenty knots. Other air searches that day proved negative. Thus as of midnight on 7–8 August there seemed to be no immediate surface threat to the invasion area. In fact Mikawa sortied that afternoon from Rabaul with five heavy cruisers, two light cruisers, and a destroyer. If unmolested during daylight on 8 August, he could deliver a night surface attack at Tulagi around midnight on 8–9 August.20
Before dawn on 8 August the Saratoga and Wasp task units approached Point Roger from the southwest. The Enterprise force, back from its fueling errand, closed from the southeast. Alerted to Kinkaid’s presence by radar, Fletcher maneuvered to let him pass ahead and take his place on the port side of the formation. Sliding into proper position, Kinkaid felt “very much pleased” with his navigation. By jockeying for position, though, the Saratoga was a dozen miles short of Point Roger at 0600, when Fletcher turned TG-61.1 northeast into a gentle breeze for flight operations. With the marine success, the need for ground support flights lessened, but the transport force, located seventy-five to ninety miles northwest of the carriers, still needed protection. Twelve Wasp SBDs fanned out to the west, north, and northeast (280–040 degrees) to 220 miles to check an area that included Santa Isabel and the whole New Georgia group. The Enterprise deployed four F4Fs as screen combat air patrol and four SBDs against subs. By far the busiest carrier in the first launch cycle was the Saratoga, which by rights was not supposed to fly any aircraft except in an emergency. Before daybreak she dispatched Felt with nineteen SBDs and eight bomb-armed TBFs to handle Turner’s urgent request the previous evening to strike Tanambogo at dawn. Fletcher and Ramsey would be ill served by this act of kindness. When the time came, Turner’s Orange Base, micromanaging the Tulagi sector from distant McCawley, failed to provide a target and wasted much of the Sara’s effort. The SBDs eventually left still lugging their bombs, but Orange Base finally released the VT-8 TBFs against Tanambogo.21
About the time the first wave left the Saratoga, Ramsey reminded Fletcher of the loss of five VF-5 pilots the previous day, one of whom SBDs sighted on a beach on northwest Guadalcanal. He requested Turner check out that area, while planes from the other carriers kept a sharp lookout. Fletcher relayed the message to Turner, who on his own detailed two high-speed minesweepers to that task. Ramsey sent four F4Fs north on an unscheduled screen combat air patrol to overfly the area where their fellow pilots went down. He also deployed four F4Fs on carrier combat air patrol, the first from any carrier that morning. As evidence of a more relaxed atmosphere the whole staff, less Fletcher and the duty officer, joined Maas at 0715 for breakfast in the captain’s mess. The Enterprise also took advantage of the lull for a special search of six SBDs between Guadalcanal and San Cristóbal looking for two fighter pilots lost the previous evening. Unfortunately all the mercy missions found no one.22
By 0800 sixteen F4Fs circled the carriers, and the same number manned the screen combat air patrol. Rowe ran short of fighters. Limited fuel capacity forced frequent combat air patrol rotations, and the inability of the Saratoga fighters to participate greatly hampered matters. Thus at 0758 Kinkaid requested permission from Noyes to add eight VF-5 F4Fs to the 0830 and 1030 carrier combat air patrol cycles. Noyes swiftly disapproved, reminding Kinkaid, “Saratoga must be ready to launch for or in case of attack.” By 0925 Noyes finally discovered Sara aircraft aloft. He bristled the Saratoga “does not appear to be complying with my orders for today’s operations which require her until noon to maintain fighters and attack group ready for launching at all times in case of bombing attack or locating of enemy CV.” In future, Noyes told Ramsey, “Please refer conflicting request to me.” This message crossed one from Ramsey at 0946 stating, “Red Base again requests 8 VF for Combat Patrol from me. Advise.” Obviously expecting Noyes’s approval, Ramsey lofted eight F4Fs for carrier combat air patrol at 0952, before the Wasp could reply. Wisely he told them to stay close and watch for new orders. Ramsey then respotted the Sara’s deck forward and started recovering eight F4Fs and the eighteen SBDs back from their fruitless mission over Yoke Group. Felt’s SBD and the eight VT-8 TBFs had yet to return from the Tanambogo strike. At 1015 an exasperated Noyes directly challenged Fletcher, who in this case acted as a subordinate task unit commander. “Your 072246 negative. Invite your attention to present situation if enemy CV should be located and I ordered your attack group launched. Your fighters should also be ready for launching for actual bombing attack until noon.” Certainly the Saratoga had not complied with his orders. Ramsey used visual signals to recall the eight F4Fs at 1023. The VT-8 landings, however, did not go well. One Avenger that was unable to attack Tanambogo could neither drop its tail hook nor bomb load. Finally at 1045 after some ten to fifteen minutes of fruitless attempts, the errant TBF set down, followed by Felt three minutes later.23
Thus by 1049 all Sara’s planes were back on board, but it was already too late. Watching from the flag bridge alongside Fletcher, Maas fumed at the failure to respot the huge deck to launch fighters. As it turned out, there was very good reason to be irritated. At 1038 the Saratoga monitored a coast-watcher message from northern Bougainville stating that forty large twin-engine planes passed overhead at 0942 headed southeast. FDO Bruning in the Chicago repeated that warning at 1044. He estimated they could reach Tulagi as early as 1115, an error in judgment as will be shown. Fletcher directed TF-61 to “prepare to repel air attack,” but the main fighter reserve of twenty-six F4Fs sat immobilized until plane handlers could reshuffle the Sara’s jammed flight deck. It is interesting that Dyer, Turner’s biographer, blamed Fletcher’s supposed anxiety over the possible presence of an enemy carrier for the delay in providing fighter defense for the invasion force. The exact opposite is true. Convinced no carriers prowled the Solomons, Ramsey declined to tie up his air strength so needlessly. His persuasiveness and reputation as a pioneer aviator swayed Fletcher, always conscious of his own lack of aviation experience, to permit the Saratoga to defy Noyes’s orders. He gambled that she could furnish Turner’s vital ground support mission and still respond to a threat. Ramsey’s estimate of the enemy carrier situation proved justified, when the Wasp’s morning search turned out negative. However, his affirmation that the Saratoga could quickly provide F4Fs to defend the task force was sadly in error.24
Conscious of lost opportunities, Yamada determined to strike hard and fast. From the many carrier aircraft his aviators encountered the previous day, he concluded two or possibly three carriers lurked east or northeast of Tulagi. Following a comprehensive search of the whole southern Solomons, all available land-attack planes, armed properly with torpedoes, would attack. Yamada concentrated his search of three bombers and two flying boats mostly northeast and east of Tulagi. Lt. Kotani Shigeru led twenty-six land-attack planes (three soon aborted) and fifteen Zero fighters southeast at low altitude along the northern fringe of the Solomons chain. He listened for search reports to learn where the enemy carriers were located. If found in range he would attack them, or otherwise hit the invasion force. TF-61 again enjoyed vital warning. The low-flying bombers rattled the jungle hilltop hideout of Lt. W. J. Read, guarding Buka Passage off northern Bougainville, four hundred miles northwest of Lunga. He radioed the sighting report that TF-61 received at 1038. Turner got TF-62 under way. Each transport squadron and its screen aimed for the center of the channel for freedom of maneuver to evade air attacks. So far, so good. However, Bruning’s estimate that these enemy planes could threaten Tulagi by 1115 actually required an average speed for the more than the four hundred miles of an impossible 270 knots. It appears he confused Read at Buka in north Bougainville with Mason, located one hundred miles closer in southern Bougainville. Unfortunately Rowe, the TF-61 FDO, was not thinking more clearly. At 1050 he estimated that bombers could reach Tulagi between 1100 and 1200, a time interval that was for the most part still far too sanguine.25
At 1050 ten F4Fs flew screen combat air patrol, with eight other Wildcats en route. Eight more F4Fs circled TG-61.1, which was twenty-five miles south of Cape Henslow and thus seventy-five miles from Lunga, ninety miles from Tulagi. Fifty F4Fs (nine Wasp, fifteen Enterprise, and twenty-six Saratoga) perched on board the three flattops. Unaware of the Saratoga’s current unavailable condition, Rowe suggested she rush all her fighters to Tulagi, there to be joined by half the Wasp fighters. That would leave the Enterprise F4Fs and the rest of the Wasp fighters to defend the carriers. At 1101 the Enterprise scrambled thirteen F4Fs for carrier combat air patrol. Before Ramsey could follow suit, Noyes must free the Saratoga from the obligation to remain in reserve until noon. Fletcher also had his say in the matter. Sent by Ramsey to the flag bridge to explain the combat air patrol options, Felt outlined the threat to the Amphibious Force. He strongly recommended the Saratoga comply with Rowe’s call for most of the fighters to defend Turner’s ships, rather than the carriers. Fletcher retorted the Japanese “won’t attack the transports. They’re going after us and get us.” Pointing out that Turner’s ships were between the oncoming strike and TF-61, Felt counseled: “Let’s get our fighters off and protect those fellows.” Fletcher “gave in and said all right.” Noyes concurred. At 1105, finally convinced the morning search discovered nothing, he released all twenty-six Sara F4Fs for Tulagi. Due to incredibly poor fleet communications characteristic of the whole operation, the Saratoga did not log his approval until 1135.26
Even had Noyes’s authorization arrived at 1105, the Saratoga could launch no fighters at once and later only a few at a time—a crucial setback for Rowe’s plan to defend the transports. It was necessary to respot the SBDs and TBFs to free up the F4Fs for takeoff, and other problems arose as well. Maas deplored the “confusion and delay,” due to “continual and last-minute changes in number of planes to be launched and resultant changes in frequency set up because of different missions installed using different frequencies.” The radios in fighters originally assigned as strike escorts had to be reset from attack to combat air patrol frequencies but required three or four changes to get the proper settings. The pilots were even uncertain which planes to man and complained that deck crews removed belly tanks from some F4Fs. Those tanks could only be replaced on the hangar deck, and no one had time for that. Only at 1141 did just eight of twenty-six fighters get away, one of which lost a shoddily installed belly tank while taking to the air. The preceding hour had not offered an edifying demonstration of carrier skill to Fletcher, who certainly could contrast the Sara’s dismal performance with the crack Yorktown.27
Although Bruning did not have the incoming raid on radar, he expected it shortly. Therefore he deployed the eighteen screen combat air patrol F4Fs at seventeen thousand feet to ensure altitude superiority, but the enemy did not appear. At 1130 Noyes, conscious his own fighters did not have belly tanks, recalled the fifteen Wasp F4Fs on screen combat air patrol for fuel. Inexplicably he included the eight that just took off at 1043 and which still had plenty of gasoline. At 1140 the Wasp scrambled nine F4Fs to relieve them, as well as thirteen SBDs for anti-torpedo-plane patrol, and cleared her deck for rapid recovery. Rowe hurried the first eight VF-5 F4Fs northward along with the nine VF-71 F4Fs, but they would need nearly thirty minutes to reach Tulagi. Six Saratoga F4Fs followed at 1150. Until these reinforcements could arrive, only three VF-6 Wildcats under Mach. Donald E. Runyon protected the entire invasion force, a shocking lapse of fighter defense. At 1202 Rowe implored the Saratoga to rush all remaining fighters to Tulagi. True to form it took eight minutes to get an affirmative back from Noyes. By that time the air battle had ended.28
Turner’s two separate squadrons cruised in Sealark Channel. At 1155 lookouts suddenly spied to eastward a gaggle of medium bombers cutting around the east tip of Florida Island. Because the search never turned up the U.S. carriers, Kotani continued southeast to Tulagi. Fletcher received a huge break. About 1000 the Kawanishi Type 2 flying boat on the 130-degree line from Rabaul passed within a few miles of the carriers but never saw them. By keeping north of the Solomons and flying at low level (and hoping until the last moment to pinpoint enemy carriers northeast of Tulagi), Kotani completely avoided radar detection. The Yoke Squadron cut loose with heavy antiaircraft at twin-engine torpedo bombers skimming the waves. To the southwest, Turner’s flagship McCawley led the X-Ray Squadron into two 30-degree turns away from the oncoming Japanese. His four divisional columns steaming abreast resembled a school of minnows, but they possessed a strong bite. “The fire of all these ships was so extensive and of such volume that the Japanese pilots showed utter confusion and state of mind and reacted accordingly.” Kotani swung the whole strike south away from the Yoke ships to take on the X-Ray Squadron, which presented him only their sterns. Fierce gunfire dropped one bomber after another. Some foolhardy Japanese actually flew among and hopped over the ships or maintained highly vulnerable parallel courses. Few ever released torpedoes, and many were seen to crash in flames. A bomber plowed into the boat deck abaft the stack of the transport George F. Elliott and set her afire. Only the hapless destroyer Jarvis absorbed a fish. Fortunately Turner’s stout defense repulsed the strike with only a little help from the few U.S. Navy planes in the area. One SBD shot down a Zero south of Yoke Group. Runyon’s three F4Fs circled seventeen thousand feet over the X-Ray Squadron until the radio revealed torpedo planes at low level. Diving in, the Wildcats engaged bombers retiring westward and splashed four land-attack planes and one Zero without loss.29
The entire attack lasted only ten minutes, and the scattered raiders fled west at high speed. Yamada was shocked how few land-attack planes turned up at Rabaul. Of the twenty-three, no fewer than seventeen succumbed, mostly to Turner’s remarkably effective antiaircraft fire. All surviving bombers were badly shot up, and one later crashed. Those who returned bore tales of incredible success: four large cruisers, three light cruisers, two destroyers, and three transports sunk, and severe damage to a large cruiser, a destroyer, and six transports. Yamada himself reduced this hysterical tally to one heavy cruiser, one destroyer, and nine transports sunk, plus three light cruisers and two unknown ships badly damaged—still incredibly wide of the mark. In truth the toll was heavy damage to one transport (which later sank) and one destroyer.30
At 1203 the carriers monitored a voice report of forty twin-engine bombers attacking the transports with bombs and torpedoes. Turner warned Fletcher of forty torpedo planes, and for once the message actually got through. Twenty-three F4Fs from the Saratoga and Wasp arrived too late to engage the enemy. At 1215 the Saratoga finally managed to loft her last eleven F4Fs and hasten them north. At the same time Noyes reinforced the carrier combat air patrol to thirty-four F4Fs. Exuberant VF-6 pilots could be heard assuring each other all enemy planes were shot down. Likewise at 1219 Fletcher was gratified to learn from Turner that the raid caused “very little damage.” It could have been much worse, given the confusion in calculating the time of the attack and the failure to muster fighters to meet it.31
The combat air patrol simply could not maintain its vast strength of nearly seventy fighters. By 1315 it looked to Fletcher and Noyes as if the carriers might not be attacked after all. It was also high time for the carriers to shift roles in accordance with Noyes’s operating schedule. The Saratoga was to replace the Enterprise as the main duty carrier, the Enterprise go on standby and search, and the Wasp take the Sara’s place in reserve. Thus Rowe suggested to Noyes the Wasp recall all her F4Fs. The other two carriers should deploy six F4Fs on carrier combat air patrol and six on screen combat air patrol and recover the rest for refueling. In the next hour the Wasp landed all twenty-four F4Fs and also eleven SBDs back from an air support mission to Tulagi. Only the thirteen SBDs on anti-torpedo-plane patrol remained aloft. After more arduous respotting of cumbersome SBDs and TBFs, the Sara took on board eleven fighters at 1328. At 1345 nine Enterprise SBDs sortied for Tulagi on one last air support mission, and fourteen TBFs departed on a comprehensive afternoon search. Its western sectors actually originated at the northwest tip of Guadalcanal (ninety miles northwest of the carriers) and extended two hundred miles to the midpoint of Choiseul and well into the Coral Sea. The other half of the search commenced from the east tip of Guadalcanal and covered Indispensable Strait and northeastern Malaita.32
Suddenly a coast watcher warned of inbound enemy planes. At 1356 Fletcher cautioned the task force, “Standby for air attack within next half hour.” The threat arose at a particularly bad moment when all three carriers were busy recovering Wildcats. The carrier combat air patrol still numbered forty fighters, but twenty-two were about to land. At the same time the screen combat air patrol comprised sixteen F4Fs, all of which must soon leave because of fuel. Turner likewise raised the alarm and got the transports moving again, to the great detriment of his plans to finish unloading that day. At 1410 Fletcher ordered TF-61, “Prepare to repel air attack.” He finally had his fill of the Saratoga’s chronic congestion. Thirty-one SBDs lumbered into the sky for no other reason than to free deck space to handle fighters. Five TBFs also departed to watch for subs. The scare was a false alarm, and by 1440 Rowe implemented his combat air patrol plan. All that time the carriers maintained a constant relative position with regard to Lunga (seventy-five miles) and Tulagi (eighty-five miles), while maneuvering in the strait between Guadalcanal and San Cristóbal. That refuted charges, to be examined below, that Fletcher already started withdrawing the carriers at noon on 8 August.33
Toward sundown the carrier portion, at least, of the Watchtower landings was winding down. Total takeoffs and landings on 8 August, although numerous, were half the previous day. Only one F4F ditched, pilot recovered. Just three F4Fs and four SBDs actually engaged enemy planes, with claims of four bombers and two fighters. The marines had, so far as Fletcher knew, captured all of the initial objectives. The last air support mission that afternoon received no targets from Orange Base. On both days the carriers provided remarkably effective ground support, considering the rudimentary state of the art, but combat air patrol protection was dismal. Turner’s invasion force weathered three fierce air assaults with relatively little help. The biggest surprise so far was that the carriers themselves had not been attacked, particularly as the enemy demonstrated he could strike hard at long distance. Kinkaid recalled, “We could not reasonably hope that Japanese search planes would fail to locate us during the period of our operations south of Guadalcanal—it seemed obvious we were there—but that is just what happened. Nor were we molested during this period by Japanese submarines.” Fletcher could not count upon such good fortune lasting very long, and other serious matters demanded his immediate attention. It was high time to reevaluate the situation, and that is precisely what he did.34