Twilight on Sunday morning, 23 August, the seventeenth day after the invasion, faintly registered in the rainy skies fifty miles east of Malaita.1 Lacking the raw strength to meet the Japanese fleet head-on, Fletcher’s TF-61 (three carriers, 216 planes) resumed the role, familiar from Coral Sea and Midway, of defending an exposed position against a strongly supported amphibious attack. Overnight he deployed to 130 miles east of Lunga, within supporting distance, yet beyond the reach of float planes in the central Solomons. Starting 670 miles southeast of Rabaul and just within its powerful search and strike network, Fletcher intended to work southeast into the wind and remain alert, while Ndeni PBYs combed the crucial sectors north and northwest of Guadalcanal.2
Despite repeated forecasts from Ghormley (if not Nimitz), the enemy had not returned in strength after the devastating raid against Savo. Should this day prove as quiet as the last few, Fletcher must start detaching his task forces one by one, as Ghormley directed, to refuel from the Platte and Cimarron northwest of Espíritu Santo. Fifteen Enterprise dive bombers and nine torpedo planes left to examine the northern semicircle to 180 miles, as far west as Santa Isabel and north of Guadalcanal and Malaita. Two Saratoga TBFs departed at 0615 for Cactus carrying Schindler with the previously noted 22 August Comsopac message, radio matériel, and ten pounds of frozen strawberries from the flag mess. The Saratoga spotted a strike group on deck, and Schindler alerted Vandegrift that it might drop in at Henderson Field, possibly even that day. Sunrise brought stronger winds, thirty-two knots and gusts, accompanied by fierce squalls. At breakfast Fletcher read a special Cincpac Ultra message advising that the “Orange striking force” of two Shōkaku-class carriers, two fast battleships, and four heavy cruisers was now “indicated” to be “in or near Truk area,” and thus not nearer to Cactus than one thousand miles. “In Truk-Rabaul area” was “Cinc Second Fleet” with “possibly” two fast battleships and “definitely” four heavy cruisers. This valuable intelligence, however, failed to answer the prime question of when the assault on Guadalcanal might come. With the Japanese carriers so distant, such a move now seemed unlikely for several days.3
The Enterprise search encountered only two distant subs running south. One slithered beneath the waves one hundred miles north of the task force; the other, 150 miles northeast, first submerged, only to surface and be forced to dive again. The first boat was I-11 en route to the B deployment line southwest of Guadalcanal; the second, I-17 bound for the A deployment line east of Malaita. Neither suffered much damage. The Enterprise planes slotted in the westerly sectors, including southern Santa Isabel Island and northwest of Malaita, returned early because of foul weather. Fletcher could not rule out the enemy’s presence in that crucial quadrant, but he knew shore-based planes should cover at least part of the affected area.
First word from a Ndeni flying boat reached the Saratoga at 0942. Lt. (jg) Francis C. Riley of VP-23 placed eight ships roughly 250 miles north of Cactus and 350 miles northwest of TF-61.4 He gave no description, course, or speed. At 1013 the PBY in the adjacent sector to the east ran across two-float seaplanes seventy-five miles southeast of Riley’s contact and 275 miles northwest of TF-61. Either they were Rekata float planes patrolling ahead of the approaching force, or they came from cruisers. At 1135 when nothing more came of the supposed eight ships, Fletcher signaled Noyes and Kinkaid his intent to continue southeast at fifteen knots until sunset. That would cut by half the run to the fueling rendezvous, then 175 miles southeast. He already gave Noyes tentative orders to refuel at dawn on the twenty-fourth. The skies cleared, but rough seas persisted along with the brisk wind. At 1154 the Saratoga monitored Riley’s amplifying report identifying two cruisers, three destroyers, and four transports closing Guadalcanal from the northwest at seventeen knots. Finally with definite word, Fletcher swung TF-61 swung around to 310 degrees at fifteen knots. Felt summoned all dive bomber and torpedo pilots to the Saratoga’s cavernous wardroom to await attack orders when the carriers moved within range. That, too, became a lengthy process. TF-61 had to turn sharply away from the base course to launch or recover planes. The delay did allow shore-based air more time to seek carriers that must come before any other objective.5
At 1250 Ghormley granted Fletcher’s request, radioed that morning from Cactus, for three oilers. On the mornings of 24 and 25 August, the Platte and Cimarron were to be within fifty miles of the designated rendezvous northwest of Espíritu Santo. The Sabine would join them on the twenty-fifth. According to the noon fuel report, the TF-61 destroyers averaged 57 percent, although the heavy ships had plenty. Oil was not critical, but TF-61 required timely refueling to retain the freedom to maneuver at high speed should a fight break out. Battle was not the time to rely on heavy ships for fuel. Fletcher must soon get on with his refueling.6
Additional search contacts secured that afternoon from Ndeni-based PBYs solely concerned three subs located 180 miles and 150 miles northwest and 240 miles northeast of TF-61. Two were possibly the same boats the SBDs roughed up that morning. If heavy ships, including carriers, supported Riley’s convoy (as it seemed some must), they proved remarkably elusive. Clearly frustrated Fletcher did not storm into attack range as quickly as possible, Kinkaid suggested TF-16 dash northwest at twenty-five knots, then at 1500, when the convoy should be 250 miles away, unleash an Enterprise strike. After hitting the target about 1645, her planes would return after sundown. Kinkaid echoed his call on 2 August for TF-16 to operate well north of the other two carriers, securing the dispersion the pioneer aviators desired when confronting enemy carriers. Fletcher simply directed the Enterprise, the duty carrier, to execute a vital precautionary search of the northern semicircle to two hundred miles. Kinkaid resignedly complied, launching twenty Enterprise SBDs at 1425.7
Task force operations, 23 August 1942
Fletcher resolved to attack that afternoon if possible but resisted the impulse to hit the convoy at maximum range. He hoped instead to locate enemy carriers. His preferred weapon was the Saratoga’s strike group, with all thirty-one SBDs and six of thirteen TBFs ready to go. If the convoy held the reported course and speed, it should, by 1430, be 275 miles northwest (and 170 miles north of Lunga). No advocate of carrier dispersion, Fletcher did not relish splitting TF-16 off to attack on its own. Even Kinkaid’s plan to launch from a slightly lesser distance required the planes to return after dark, with the worry of night carrier landings by inexperienced pilots and danger from subs caused by illuminating the carriers. Fletcher’s hole card was Henderson Field. Should nothing better than the convoy turn up by 1415, he intended to play it. After an “interminable” two hours, Felt set down the telephone with a “satisfied grin” and joyfully announced: “Let’s go, boys! No carriers have been found. We will attack the cruisers.” Fletcher ordered him to strike the convoy, then stage south to Henderson Field, landing by sunset if possible, and return to the Sara at 0730 the next morning east of Malaita. No escort fighters would go; the flight was too long. At 1440 the thirty-seven planes departed northwest. Watching from the North Carolina, Michael Laing wondered if Fletcher acted from additional information of which he was unaware. Otherwise, “This seems like a long shot.” Truly, Fletcher took a grave risk similar to the Japanese search-strike mission on 7 May that caused Admiral Takagi and Admiral Hara so much grief in the Coral Sea. Much rode on Felt’s shoulders. Fletcher could hardly afford to lose his force. He judged the strike capable of defeating the surface force thought approaching Guadalcanal and relied on the marines to protect his planes while on the ground.8
Anxiously watching from Nouméa, Ghormley made certain Fletcher knew of the convoy. In turn McCain demanded the Mackinac (returning to Ndeni) maintain “contact with enemy force reported by [Riley].” At 1409 Leary placed two destroyers and two cargo ships near Shortland as of 1140. “Request you hit them now,” Ghormley shot back. Nearly five hours later, MacArthur explained his flyers needed “a minimum of seven hours for execution [of ] missions against targets given by you as it is impossible due to dangers involved to use Moresby or Milne Bay except as topping off fields.” He avowed the B-17 strike on Rabaul the previous day had “telling effect.” Ghormley cautioned Vandegrift, “Sightings indicate two groups enemy ships converging Guadalcanal.” They could arrive “tonight.” Vandegrift fully understood the threat. At 1615 he dispatched nine marine SBDs, thirteen F4Fs, and the Saratoga TBF against the convoy now believed well within two hundred miles of Lunga. McCain hurriedly organized a night strike by five PBYs from Espíritu Santo, with B-17s to go in the next dawn. He again demanded the Mackinac update her sighting reports. For his own part Turner ordered the Fomalhaut out that evening with the disabled Blue. The cargo ship cleared out as directed, but the destroyer had to be scuttled. In the past three days two cargo ships, six APDs, and one seaplane tender had deposited either all or a big portion of their cargoes at Lunga and Tulagi, relieving in part the severe shortage of supplies.9
Ninety minutes after Felt’s strike group left, Ghormley warned Fletcher of the imminent danger to Cactus and alerted him to “be prepared send surface detachment to destroy enemy forces Cactus area tonight.” At that time TF-61 was 170 miles distant by sea from Lunga—seven hours at twenty-five knots. Fletcher could have a surface force there before midnight or by 0200 at a more economical twenty knots. To comply, though, it must leave almost immediately, but that was up to Ghormley. Fletcher certainly would not take his whole task force there. He already dealt strongly with the convoy to the north. Cargo ships starting out from Shortland could not possibly reach Guadalcanal before late the next morning. A sighting report that Leary relayed at 1701 confirmed that they had not even left Shortland.10
Fletcher experienced the familiar gut-wrenching wait for strike results that was, as one Japanese carrier captain described, a “thousand seconds” passing “like as many years.”11 Felt should have reached the target around 1630. Nothing was forthcoming from other commands, notably McCain’s Airsopac. Fletcher pondered the latest hot intelligence from Pearl. The Cincpac daily bulletin unequivocally placed the Shōkaku and Zuikaku “enroute Japan to Truk,” while a heavy cruiser and “possibly” two fast battleships were in the Truk “area” or “vicinity.” Thus Cincpac amended his earlier announcement that morning of two Shōkaku-class carriers “in or near Truk.” The term “in or near” is not the same as “enroute to,” although Nimitz seemed not to have noticed the difference. Even so, Fletcher now had good reason to believe the nearest carriers were still north of Truk, itself more than eleven hundred miles distant from his present position. In fact, Nagumo cruised only three hundred miles northwest of TF-61, and Kondō’s Advance Force drew even closer. The reason for this colossal failure by the same radio intelligence network that forecast the Coral Sea and Midway battles will be examined later. Its immediate effect was to delude Fletcher into believing it was safe to detach his task forces one at a time to fuel.12
After 1700 Kinkaid heard from the Enterprise pilots back from the afternoon search of a “slightly damaged” sub one hundred miles northwest of TF-61, and possibly a cargo ship and light cruiser or destroyer anchored off northern Malaita. The search again encountered bad weather to the northwest, this time about 145 miles out. Oddly Kinkaid did not think these contacts sufficiently important to forward to Fletcher until the next morning. At 1800, eleven minutes prior to sundown, Fletcher reversed course to the southeast. No planes to westward appeared on the Sara’s radar. Whether or not Felt’s raiders found the convoy, they must have gone to Guadalcanal as instructed. Twenty minutes later, “There being no enemy carriers in immediate vicinity,” Fletcher directed Noyes to proceed to the fueling rendezvous 210 miles south. The departure of the Wasp (sixty-three planes) left Fletcher 150 operational aircraft in the Saratoga and Enterprise, including the thirty-eight at Cactus.13
With Noyes’s TF-18, Richard Frank wrote, disappeared “the best chance in 1942 for the United States Navy to have potentially decisive quantitative superiority in a carrier action.” Morison considered detaching the Wasp at that juncture yet another example of Fletcher’s lamentable predilection for unnecessary fueling at critical times. “Once more, the destroyers’ logs prove that this anxiety was unwarranted.” Of course Morison, aware unlike Fletcher of enemy carriers in the neighborhood, applied his usual omniscient hindsight. Fletcher’s evaluation of the situation was a “bad guess.” He should somehow have divined the presence of carriers nearby. Frank faulted Fletcher for not finding “as much significance in the presence of a convoy and in the submarine sightings as did some of his subordinates.” Yet the only subordinate on record is Davis in the Enterprise, who wrote afterward that the discovery of three subs rushing south “pointed strongly to the possibility of a considerable enemy movement to the southward along the track of the advancing submarine screen.” Again, Davis showed wisdom after the fact. His boss Kinkaid did not even think it vital to advise Fletcher immediately of the last sub sighting on the twenty-third.14
The critics discounted the importance of Cincpac’s radio intelligence. That night’s summary by McCormick’s War Plans Section bespoke the favorable interpretation at Pearl, based largely on the same intelligence and sightings, that completely justified Fletcher’s action. Despite the “comparatively large number of enemy cruisers, destroyers, transports, and freighters” that turned up “recently in the Rabaul area,” McCormick judged they would offer “a fine target for Fletcher if they ever come within his reach.” In turn, “Enemy striking forces previously mentioned heading toward the Southwest Pacific” were “now indicated” to be “in the Truk area.” This “being the case, they cannot arrive in the Tulagi area before the 27th or 28th, local date.” Fletcher could also read a chart. He took advantage of that supposed window of opportunity to refuel his three task forces in succession. To wait risked all three carrier groups being caught short of oil when the Japanese carriers finally did show up.15
That evening Ghormley declined to order TF-61 cruisers and destroyers to proceed to Cactus, but time and distance had already negated that option. MacArthur relayed another coast-watcher report that five ships, “Including one very large,” were still at Shortland as of 1330. That again reduced any chance of the second pincer showing up at Guadalcanal. Later Leary detailed a concentration of ships off southern Bougainville that again argued any coordinated amphibious assault against Cactus had to be more than a day away.16
Fletcher maneuvered to approach his intended dawn position from the east. He would start ninety miles east of Malaita and 170 miles northeast of Lunga, and thus was forty miles farther east than on the twenty-third. Late that evening Vandegrift radioed that his own strike returned without finding the convoy in poor visibility, but he did not mention the Saratoga planes. Only well after midnight on 24 August did two messages clarify the situation for Fletcher. The first, received at 0124, elicited much anger in the Saratoga. The Mackinac, back at Ndeni since 1430, finally advised that the convoy of three transports, one light cruiser, and two destroyers had “radically” changed course at noon. Riley did not in fact shadow it but searched as much of the balance of his sector as the weather allowed. He reacquired the contact; when last seen at 1230 the convoy was retiring northwest. Thus the Japanese scooted out of range well before the launch of the Saratoga strike. Why it took thirteen hours to advise Fletcher of that crucial event no one ever explained. Nor did anyone bother to tell Fletcher that thick clouds forced the PBYs to curtail their search of certain vital sectors at five hundred miles instead of 650 miles. Even that reduced area received only 70 percent coverage. Schindler, waiting at Lunga, provided the best news. Although Felt never found the convoy, all the planes reached Cactus and were prepared to “hit enemy on arrival.” Due to “delay in gassing,” Schindler requested they return to Saratoga at 1030 instead of at 0730. Fletcher felt tremendous relief the Sara’s flyers were safe. Now he needed to retrieve them as quickly as possible and see what transpired on 24 August.17
The dawn of 23 August, the penultimate day of the KI Operation, saw Tanaka’s Second Echelon convoy plodding straight in from three hundred miles north of Guadalcanal. Angling southeast, Nagumo’s Kidō Butai and Kondō’s Advance Force crossed Tanaka’s track about two hundred miles north. Nagumo planned to continue southeast, but keep north beyond the reach of flying boats he thought were based no closer than Efate. By dawn on the twenty-fourth he would be roughly halfway between Malaita and Ndeni and 250 miles northeast of Lunga. From there he could destroy the U.S. carriers that patrolled south of San Cristóbal. Kondō, reinforced by three destroyers, was to shift from southwest of the Kidō Butai to sixty miles northeast and guard Nagumo’s left flank with seaplane searches three hundred miles to eastward. Only if Tsukahara’s Base Air Force failed to pound the Lunga airfield on 23 August would plans change. In that event Nagumo was to spin off a carrier detachment to raid Guadalcanal on the twenty-fourth and also give Tanaka close air cover. Frozen out of carrier command, Rear Admiral Hara now led the 8th Cruiser Division in the Kidō Butai. Should a carrier strike against Lunga be necessary, he would form a mini support force with flagship Tone, Ryūjō, and two destroyers. With only nine Type 97 carrier attack planes the Ryūjō offered little real offensive firepower, but her twenty-four Zero fighters could protect the convoy.18
Fully aware of his vulnerability to any form of attack, Tanaka noted the baleful appearance of Riley’s flying boat 275 miles north of Guadalcanal. To make matters worse, the I-11, one of two subs encountered that morning by Enterprise SBDs, radioed the sighting of carrier-type planes only 180 miles southeast of the convoy. Worried American carriers might have moved up more quickly than anyone realized, Mikawa once again told Tanaka to pull away until the situation cleared. Wearily complying at 1040, Tanaka apprehensively awaited results of searches by Base Air Force and Kondō’s whole force. Tanaka’s nearest source of support remained Mikawa’s Outer South Seas Force shuttling back and forth to refuel at Shortland.
Also alerted by the I-11 to an enemy carrier much farther north than previously thought, Kondō and Nagumo supplemented the shore-based search effort. It seemed so many aircraft crisscrossing the eastern Solomons could not fail to ferret out the opposition, but that happy result did not occur. Tsukahara learned nothing from his own search, except the land-attack plane on the 115-degree line did not return. Perhaps it succumbed to poor weather; no one on the Allied side claimed it. At 1400 Nagumo executed his own precautionary scout 250 miles ahead, but at that time TF-61 was 375 miles southeast. Like Fletcher, Nagumo played hide-and-seek until he could locate the opposing carriers. He strongly suspected the Allies had not sighted his ships and sought to preserve that advantage by not closing prematurely within reach of the short-range planes at Lunga and the flying boats from distant Efate. At 1740 he radioed that unless otherwise instructed, he would retire north all night (150 miles), then at 0700 turn back southeast. Kondō’s afternoon search likewise yielded nothing. The Advance Force kept pace with the Kidō Butai until late afternoon, then forged ahead southeast before also turning north at 2115. Kondō never realized TF-61 was only 275 miles beyond. He planned to start the next morning as scheduled sixty miles east of Nagumo. Kondō cautioned Yamamoto the whole Support Force must soon refuel and set a convenient rendezvous for 26 August north of the Solomons.19
Overseeing the offensive as the Yamato fueled 180 miles southeast of Truk, Yamamoto learned that storms again prevented Tsukahara’s land-attack planes from taking out the Lunga airfield. He therefore ordered Nagumo to strike Guadalcanal on the afternoon of 24 August “with appropriate force.” He stipulated the attack take place only if in the meantime no U.S. carriers turned up. Consequently Nagumo directed Hara’s Support Force to depart south at 0400 on 24 August and hit Lunga that afternoon, while the rest of the Kidō Butai again sallied southeast. Tsukahara reluctantly permitted Mikawa and Tanaka to postpone the landing to the twenty-fifth and canceled the bombardment of Lunga scheduled for that night. The Kagerō did not get the word and briefly shelled the perimeter after midnight on the twenty-fourth. Evidently a flash report of the light bombardment of Lunga reached Fletcher, but he judged correctly it was not a serious attack.20
Had the clouds cooperated on 23 August, all six carriers would have fought. The Airsopac flying boats, of course, flew out of Ndeni rather than Efate 450 miles farther south. All things being equal they should have pinpointed Kondō and Nagumo as well as Tanaka. One of Kondō’s cruisers even glimpsed a PBY fly past in the overcast. Fletcher greatly benefited from the chaotic weather that disrupted searches by the Rabaul land-attack planes and Shortland-based flying boats that should have seen TF-61. The stage was set for the naval action that would come to be known to the Allies first as the Battle of the Stewart Islands (islets near TF-61) and ultimately the Battle of the Eastern Solomons.
Early on the morning of 24 August Fletcher’s reduced TF-61 continued toward the planned dawn position ninety miles east of Malaita. The Saratoga and Enterprise counted fifty-five fighters, thirty-three dive bombers, twenty-two torpedo bombers, and two photo reconnaissance planes, plus the Sara’s thirty-one dive bombers and seven torpedo bombers expected back at 1030 from Cactus. The Enterprise again scheduled routine forenoon and afternoon searches, while the Saratoga handled combat air patrol. Fletcher intended to work closer to Malaita into the same area as the previous afternoon. The Point Option course for the morning search was to be 280 degrees. Given the need to run into the brisk twenty-four-knot southeast wind for periodic flight operations, the actual westerly rate of advance might only be six knots. Fletcher did not anticipate significant action that day, but as usual he was wary.21
Twenty minutes before sunrise, the Saratoga lofted eight combat air patrol F4Fs, and the Enterprise twenty SBDs, to search the northern semicircle to two hundred miles. Three additional SBDs from the Enterprise were assigned to inner air patrol. The overcast skies offered hints of clearing. The Enterprise finally blinkered to Fletcher the results of her previous afternoon search, namely the sub attack and enemy ships possibly anchored off Malaita. Fletcher queried why Kinkaid had not reported sooner and if the Enterprise sent bombers to investigate. Kinkaid responded he did not do so the previous evening “because of doubt due difference of opinion [of ] observers,” and his disinclination to “use [blinking] lights after dark this area to make report of doubtful accuracy.” Nevertheless he arranged for the dawn search to check that site. The morning proceeded uneventfully. Fletcher directed the Enterprise, once she recovered her forenoon search, to take flight duty for the rest of the day, while the Sara rearmed and refueled the planes from Cactus and held them in reserve to strike.22
Like the previous day, the hardworking PBYs tended by the Mackinac at Ndeni sought Japanese forces north of the Solomons. Six VP-23 Catalinas covered an arc bearing 306–348 degrees to 650 miles, taking them east and northeast of Bougainville. Also like the previous day, that vast area encompassed all the task forces threatening Guadalcanal, but now visibility was better. All six PBYs contacted the enemy.
Their reports, at least those actually received, strongly influenced the course of the battle. The correlation of call signs (based on plane numbers) and the arrangement of search sectors (clockwise from west to east) were purely coincidental. Fletcher could not determine from its call sign what sector an aircraft actually searched, a fault Airsopac did not remedy until November.23
Table 28.1 The Ndeni Search
Yamamoto’s plan for 24 August set forth the extensive air search of the Solomons and the waters east and south to find the U.S. carriers that posed the greatest peril to Tanaka’s Second Echelon convoy slowly closing Guadalcanal. Edging southeast during the day, Kondō’s Advance Force (five heavy cruisers, one seaplane carrier, one light cruiser, five destroyers) and Nagumo’s Kidō Butai (two carriers with 142 planes, two battleships, three heavy cruisers, one light cruiser, eight destroyers) would do all in their power to destroy them. At the same time, Tsukahara’s Base Air Force was to blast the Lunga airfield and finally open the way for the landing the next day. Operating separately from Nagumo, Hara’s Support Force (one light carrier with thirty-three planes, one heavy cruiser, two destroyers) would dash past Tanaka’s convoy to draw into strike range of Guadalcanal. Should no U.S. carriers be detected by noon, Ryūjō aircraft were to surprise the Lunga-based planes on the ground. Yamamoto gave Hara the option of waiting to recover his strike planes or routing them northwest to Buka, while he headed north to refuel.24
At 0600 after recycling northward all night, Japanese forces assumed their former relative positions. Farthest east, Kondō’s Advance Force cruised southeast at sixteen knots while watching the Kidō Butai’s left flank. According to plan he was to keep station sixty miles east of the carriers. Nagumo, however, had diverged northwest instead of going north the previous night (and not told anyone), so Kondō was actually 120 miles southeast of him. The Advance Force itself was five hundred miles northwest of Ndeni and 360 miles north of TF-61. Out on the west flank, the Second Echelon convoy (one light cruiser, two destroyers, three transports, and four patrol boats) again inched southward from 275 miles northwest of Guadalcanal. Separating from Nagumo at 0400, Hara sped south at twenty-six knots on a course roughly parallel to Tanaka’s. Two hours later he was seventy miles northeast of the convoy. Two pairs of Mikawa’s heavy cruisers with a few destroyers converged on Tanaka from the west. The centerpiece of the array, Nagumo’s Kidō Butai (Shōkaku and Zuikaku), started southeast before dawn at fifteen knots from 140 miles northeast of the convoy, 415 miles north of Lunga, 450 miles northwest of TF-61, and six hundred miles northwest of Ndeni. In case U.S. carriers had worked around to the east, Nagumo dispatched nineteen carrier attack planes to cover the eastern semicircle to 250 miles.25
Task force operations, 24 August 1942
The first actual sighting by the Ndeni PBYs occurred at 0900 in one of the two central sectors. Ens. James Spraggins’s 7V37 happened on a light cruiser headed south. Neither Fletcher nor Kinkaid monitored his 0905 contact report, which was 310 miles north of TF-61. Spraggins glimpsed one of Hara’s two destroyers but placed it sixty miles east of its actual location. Continuing northwest, he unwittingly slipped between Hara’s Support Force and its parent, Nagumo’s Kidō Butai, off to the northeast. Spraggins made no other contacts during the mission. Thus the Ndeni search missed its first vital opportunity to pinpoint Nagumo promptly that morning. At 0935 the Enterprise copied a message from Ens. Gale Burkey’s 5V37, in the other center sector, of one carrier, two light cruisers, and one destroyer, headed south. Burkey nailed Hara’s Support Force 275 miles northwest of TF-61. Kinkaid thought Fletcher also knew of the contact, especially after he relayed that vital information by TBS at 0945 to the Saratoga and received an acknowledgment. The Saratoga’s air plot recorded that transmission (or at least part of it) but missed the call sign “5V37,” noting the “source of Enterprise information was unknown.” Even so, air plot did not bother to check with flag plot to see if Fletcher also had the message, which he did not. Other searchers working northwest of Guadalcanal likewise located and promptly reported nearly all the ships in their respective sectors. Sweeping west of Burkey, Lt. Joseph Kellam’s 3V37 found Tanaka’s convoy at 0953 situated 260 miles north of Guadalcanal and three hundred miles northwest of TF-61. Not surprisingly, Fletcher and Kinkaid never copied his contact report. Kellam continued searching his sector. About the same time, southwest of Santa Isabel, a B-17 from Espíritu Santo found a cruiser 125 miles west of Lunga and 250 miles west of TF-61. At 1003 Ens. Theodore Thueson’s PBY 1V37, covering the westernmost sector, discovered two light cruisers and a destroyer near Ontong Java atoll, 220 miles northwest of Lunga and more than three hundred miles from TF-61. Fletcher heard both of those reports immediately. The B-17 came across the destroyer Kagerō retiring up the Slot, while Thueson encountered Mikawa’s main body southwest of Tanaka’s convoy.26
Thus McCain’s searchers successfully charted enemy deployment north and northwest of Guadalcanal and quickly acquired the Ryūjō, the most important target, although Fletcher did not know it. He deduced the ships found by the B-17 and Thueson’s 1V37 had pulled out after shelling Cactus the previous night. He saw no evidence of the feared two-pronged attack on Cactus. Nevertheless at 1012 he alerted Kinkaid to ready a strike group “as soon as possible.” At that moment after the launch of eight SBDs and six TBFs to counter the serious sub threat, only two SBDs and nine TBFs remained on board the Enterprise. The morning search and first inner air patrol, however, waited to come on board, though they, too, must be serviced before a strike.27
At 1016 Fletcher monitored a report from PBY 9V37 of two cruisers and two destroyers “distance 450 bearing 337.” Lacking further information, he assumed the PBY’s point of origin was Ndeni, which, if correct, positioned those enemy ships 330 miles north of TF-61 and in excess of 250 miles northeast of the cruiser group discovered near Ontong Java. Now the Japanese appeared to be operating across a broad front. Very quickly Fletcher received evidence that a carrier might be involved, an alarming situation indeed. At 1030 9V37 radioed it was being attacked by three aircraft, type not given, while 5V37, a different PBY, reported: “Attacking [sic] by aircraft fighting plane type Zero. I am returning to base.” Due to the specific mention of Zero fighters, as opposed to seaplanes, Fletcher later described the 5V37 message as the “first indication that enemy carriers might be in the vicinity.” However, he had no idea where 5V37 was (having not received Burkey’s earlier report of a carrier) and, for that matter, only supposition as to the location of 9V37. It did not appear out of the realm of possibility that 5V37 ran afoul of Zeros based out of Buka. More ominous, at 1050 yet another PBY, 8V37, claimed hostile interceptors, but this time supplied a definite position that translated to 330 miles north of TF-61 and sixty miles southwest of where 9V37 might be. Now to Fletcher it seemed the enemy main force might be more concentrated. The reports of 9V37 and 8V37 provided even “stronger indications of an enemy carrier.”28
Principal sighting reports received by CTF-61, 24 August 1942
Key:
| 1. | 1003 | PBY 1V37 Thueson; 2 light cruisers, 1 destroyer |
| 2. | 1016 | PBY 9V37; 2 cruisers, 2 destroyers |
| 3. | 1016 | B-17 22V40; 1 cruiser |
| 4. | 1050 | PBY 8V37; attacked by aircraft |
| 5. | 1110 | PBY 5V37 Burkey; 1 aircraft carrier, 2 cruisers, 1 destroyer, originally sent 0935 |
| 6. | 1117 | PBY 8V37; 1 light cruiser |
| 7. | 1242 | PBY 1V37 Thueson; 2 cruisers, 1 destroyer |
| 8. | 1320 | Saratoga radar contact, bearing 350 degrees, 112 miles, course 220 degrees |
| 9. | 1405 | PBY 3V37 Kellam; 1 small aircraft carrier, 2 cruisers, 1 destroyer |
| 10. | 1440 | Air raid reported by Cactus |
| 11. | 1523 | TBF 1V395 Jett; 1 aircraft carrier, 1 heavy cruiser, 1 light cruiser, 3 destroyers |
| 12. | 1550 | Saratoga’s first strike attacked 1 aircraft carrier |
| 13. | 1607 | Saratoga radar contact, many bogeys bearing 320 degrees, 103 miles |
| 14. | 1700 | Returning Saratoga strike planes sighted Japanese strike group |
| 15. | 1735 | Saratoga’s second strike attacked 1 battleship, 4 heavy cruisers, 6 light cruisers, 6–8 destroyers |
| 16. | 1815 | B-17s Manierre attacked 1 aircraft carrier |
| 17. | 1815 | B-17s Sewart attacked 1 small aircraft carrier, 3 heavy cruisers, 2 destroyers |
| 18. | 2000 | SBD Davis sighted 2 aircraft carriers, 4 heavy cruisers, 6 light cruisers, |
| 8 destroyers, originally sent 1500 | ||
| 19. | 2130 | PBY 14V37 Corbett; 5 vessels |
| 20. | 2133 | PBY 13V37 Brady; unidentified vessels |
Lt. Leo Riester’s 9V37, flying the easternmost sector from Ndeni, had in fact reported the two cruisers and two destroyers at 1016. He snooped Kondō’s Advance Force, and his navigation was off by only a few miles to the east. The seaplane carrier Chitose scrambled three nimble Type 0 observation seaplanes (code-named Pete) that hounded the PBY for more than an hour. The distant Mackinac mistakenly assumed Riester’s assailants to be Zero fighters and broadcast word to that effect. At 1144 when Riester turned his damaged plane back toward base, he correctly identified his foes as three “cruiser seaplanes,” but Fletcher did not copy that message. The 5V37 warning, of course, originated from Burkey, whose lumbering PBY evaded two Ryūjō Zeros for more than an hour before escaping. PBY 8V37 was Lt. (jg) Robert Slater, handling the sector between Spraggins (7V37) and Riester (9V37) to the east. Slater sent no messages at that time, so any reference to 8V37 was a garble. Passing near enough to Nagumo’s Kidō Butai to be seen by its lookouts, Slater never noticed the carrier force in the haze. That was the final missed opportunity for the Ndeni search to have found Fletcher’s most dangerous opponent.29
While contemplating the prospects for battle, Fletcher was gratified to see Felt’s raiders appear overhead fresh from the muck of Guadalcanal. At 1035 the Saratoga launched a dozen combat air patrol F4Fs and landed eight combat air patrol F4Fs, twenty-nine SBDs, and seven TBFs. The additional TBF was Schindler’s. The past evening after the harrowing long-range search/strike through intermittent squalls (“Some just terrible pile of weather,” Felt recalled), Felt got the strike down safely at Henderson Field, greatly helped by flares set alongside the runway. The ground crews refueled the planes, a laborious process given the primitive means available. Later that night while marine SBDs patrolled adjacent waters on the lookout for approaching enemy ships, the Saratoga crews waited alongside their planes ready to hasten aloft and attack. The only opposition was a brief shelling by what was thought a sub, actually the Kagerō. After dawn with both skies and coast clear, Felt had the SBDs toggle their defused 1,000-pound bombs into the mud for their grateful marine hosts, then took off for the rendezvous east of Malaita. Two SBDs turned back due to engine trouble, but the other Saratoga flyers uneventfully reached TF-61 after spotting the ships from a long way off. Correspondent Clark Lee, who flew the mission, described the vicinity of home plate: “Long swells were running but the air was clear and unclouded, with an after-the-storm freshness.” Of course the brilliant skies offered no hindrance to others seeking TF-61.30
Lee noticed “an atmosphere of quiet excitement” on board the Saratoga. “This was the day.” At 1103 as the last of Felt’s planes touched down, Fletcher alerted all ships that he required full boiler power in thirty minutes. At the same time Kinkaid advised that “Art’s boys” were “ready now,” meaning the Enterprise strike group (twenty-five SBDs and nine TBFs, plus fighter escort) could take off at any time. “Quiet excitement” pervaded the Saratoga, but the Enterprise seethed. Kinkaid knew of Burkey’s 5V37 carrier contact for ninety minutes and waited for Fletcher to charge north so his “boys” could attack. Inaction seemed inexplicable. At 1116, in exasperation, Kinkaid queried Fletcher by TBS: “Are you guarding for contact reports?” Fletcher replied, “Many reports are garbled. Do you have any definite reports of enemy?” Kinkaid certainly did. All proved familiar to Fletcher except the startling “5V37: 1 CV 2 CL [light cruiser] 1 DD Lat 04–40 Long 161–15 course 180.” He responded that he had all except that one and asked Kinkaid to repeat the longitude. That of course was the original contact of the Ryūjō force that Burkey radioed at 0935. Fletcher’s failure to copy the 0935 message and the repeat Kinkaid relayed by TBS at 0945 exemplified the terrible communications that plagued the U.S. Navy in the South Pacific. Naturally assuming Burkey’s sighting had just taken place, Fletcher charted the enemy carrier force bearing 343 degrees, distance 281 miles. Kinkaid, who knew the contact was nearly two hours old, reckoned its current position at 245 miles northwest, bearing 336 degrees.31
Fletcher now understood the enemy “had commenced his full scale attempt to retake Guadalcanal.” The contact reports that he actually received revealed a concentration of ships 275 to three hundred miles northwest of TF-61 and advancing across a front of forty to sixty miles. That impression was reinforced after 1117, when someone rebroadcast the original 7V37 (Spraggins) 0905 contact report of a light cruiser but wrongly attributed it to PBY 8V37, which previously warned of being attacked by enemy planes. That light cruiser (actually one of Hara’s destroyers) was forty miles east of Burkey’s carrier contact. Fletcher now must fight his battle without the Wasp, a situation resembling that of Confederate Lt. Gen. James Longstreet on the second day at Gettysburg. Lacking one of his three divisions, Longstreet complained to Maj. Gen. John B. Hood that he did not care “to go into battle with one boot off.” Fletcher would have agreed. Cincpac intelligence obviously erred badly by placing all the Japanese carriers north of Truk. Now the discovery of one carrier almost certainly meant two or three others were stashed nearby. “We did not believe [Japan] would attempt this attack without at least 3 carriers,” recalled Pederson. The “invasion force which had turned back the previous day was somewhere to the northward of the striking force, ready to move in when the way was clear.” Again Nagumo held the weather gauge. The southeasterly winds allowed him to conduct flight operations while continuing to close his enemy.32
Word of the carrier sighting filtered down to Felt, newly back from Cactus, along with orders to go again right away. Given the long distance to the target, he commented, “That’s beyond our range.” His tired pilots needed some rest. “Let’s just take it easy,” he added. “You keep getting intelligence and when those things are within our range, we’ll go.” Fletcher concurred. At 1127 he turned north and ordered the refueling and rearming of the Saratoga planes be expedited “for an attack mission pending the closing of the enemy and further information from reconnaissance planes.” The situation again resembled 7 May at Coral Sea, where Fletcher loosed ninety-three planes against what he thought were two big carriers, only to net the small carrier Shōhō while the Shōkaku and Zuikaku remained hidden. The following day the dangerous pair inflicted fatal damage to the Lexington. Now this day they, plus perhaps some friends, lurked somewhere north of TF-61, possibly within strike range. Above all Fletcher must know whether that were so. “In order to verify these [search] reports and guard against a repetition of the attack made under similar circumstances on the 23rd which never found its objective, I reluctantly ordered Commander Task Force Sixteen to immediately launch a search.”33
Kinkaid advised Fletcher the Enterprise had twenty F4Fs, twenty-five SBDs (with 1,000-pound bombs), eight TBFs (two 500-pound bombs each), and the group commander’s TBF ready to go. From this pool of planes he proposed to employ twelve SBDs and eight TBFs for the afternoon search. Fletcher told him at 1137 to dispatch that planned search “as soon as possible” along the suggested arc of 290–090 degrees to 250 miles. TF-61 would steer a Point Option course due north and average ten knots. Fletcher’s “reluctance” stemmed from having to waste so many dive and torpedo bombers for a follow-up search. Had the Airsopac searchers discovered all the carriers, those planes could have joined the attack. He held back the fifty Saratoga strike planes (twelve F4Fs, thirty SBDs, and eight TBFs) for the time being to see if the Enterprise or the shore-based search could find the other carriers. With the Sara’s flight deck spotted for takeoff, he directed the Enterprise to maintain the combat air patrol and inner air patrol and also recover and service the Saratoga’s fighters. Rooted in Pederson’s preinvasion planning, that unprecedented example of enhanced coordination between carriers greatly improved combat air patrol flexibility. Once the Saratoga’s strike departed, the role of the two carriers would switch. The Enterprise was to form a second strike group from her remaining SBDs and TBFs; the Sara would refuel fighters from both carriers and arm her own few TBFs (those that could not accompany the first strike) with torpedoes so they could join the second wave.34
Fletcher constantly worried about having to absorb a carrier attack without knowing its point of origin. The likelihood of that calamity increased after 1150, when radar detected a bogey. Four F4Fs encountered a sleek, fast, huge flying boat thirty miles southwest and shot it down in flames at 1213 after a long chase away from the task force. Given the distant intercept, Fletcher doubted this snooper ever sighted TF-61. Nor did Slonim’s radio intelligence team hear any radio transmissions. Fletcher knew he could not count on such good luck all afternoon. The victim was a Type 2 flying boat. Shortland indeed copied no messages from the missing plane. That proved terribly bad luck for Hara’s Ryūjō force.35
Given the wide area to be covered, Kinkaid increased his search to sixteen SBDs and seven TBFs and delayed departure until the SBDs, to augment their range, switched to single 500-pound bombs. The Big E also prepared sixteen F4Fs for combat air patrol and six SBDs to relieve the antisubmarine patrol. While waiting for the launch, Kinkaid informed Fletcher that the Enterprise’s earlier search had only eyeballed another sub to the east. Poor visibility prevented the check for possible ships off northern Malaita, but the afternoon search would not proceed far enough south to do so. He also advised that the intermediate air patrol had bombed yet another sub twenty-five miles southeast of the task force.
Fletcher laid out his plans to Kinkaid at 1216. The Saratoga had ready a strike group to go once searchers established fresh contact with one or more carriers. The Enterprise was to service the dozen Saratoga combat air patrol fighters sent aloft at 1040. Fletcher queried Kinkaid how many planes he would have after the Enterprise search departed and what would be left for a second strike. TF-61 turned southeast at 1229 for the Enterprise to launch the forty-five planes on deck. After the next launch and recovery cycle, around 1330, Kinkaid advised he would have an attack group of twenty-three SBDs (sic, actually thirteen) armed with 1,000-pound bombs, six TBFs with torpedoes, the command TBF, and twenty escort F4Fs. Hoping Fletcher would relent and release him to attack immediately, he again emphasized, “Enemy sighted by plane consists of 1 CV 2 cruisers 1 DD bearing 340 distance 250 miles.” Fletcher, though, waited a little longer to see if the new search turned up something more substantial before committing the Saratoga group. “Will hold your attack group in reserve for possible second carrier,” he replied at 1242. “Do not launch attack group until I direct you.” Nothing more regarding the enemy’s carriers was immediately forthcoming. At 1242 Thueson’s PBY 1V37 radioed that the force previously reported near Ontong Java included heavy cruisers and was retiring north at twenty knots. At 325 miles from TF-61, this group was not only far out of strike range but also lacked carriers.36
After the search took off, the Enterprise recovered eight combat air patrol F4Fs, six TBFs, and eight SBDs from antisubmarine patrols and the twelve Saratoga combat air patrol F4Fs, who had just ambushed a low-flying twin-engine bomber. The Japanese plane splashed only seven miles short of TF-11, after having veered directly toward the Sara. It was thought to have carried a torpedo. The crew never raised headquarters by radio, and Slonim’s team heard no suspicious messages. Yet it seemed certain to Fletcher: “The presence of these planes clearly indicated that our position must be known to the enemy.” Ironically he actually knew much more about Japanese dispositions than they did of his, plus he enjoyed the fleeting advantage of understanding a battle was in the offing. Yamamoto’s admirals strongly suspected, but could not prove, the presence of U.S. carriers within strike range. Neither plane downed by the TF-61 combat air patrol in fact got the word out before being lost. The odds, though, would shift decisively in favor of whoever found his opponent’s main body first.37
Startled by U.S. flying boats farther north than expected, Nagumo fretted they already spotted him. Thus he belatedly implemented a tactical plan, the rationale behind the concept of the Third Fleet, devised following the Midway debacle. A vanguard of battleships, cruisers, and destroyers was to form a scouting line one hundred to 150 miles ahead of the carriers, conduct its own search, warn the main body of enemy scouts, and finish off the enemy task force once carrier air crippled it. The vanguard would also absorb air attacks that might otherwise hit the carriers. The rapid deployment south afforded the Kidō Butai no opportunity to test the new tactic. This day Rear Adm. Abe Hiroaki took the vanguard of two battleships, four cruisers, and two destroyers only ten miles ahead of the main body and spread them along a twenty-mile front—a belated separation, although far less than doctrine, that would still pay great dividends. So did a last-minute decision that modified the previously scheduled search southeast by battleship and cruiser float planes. Concerned at the last minute that the waters nearest to Guadalcanal lacked adequate coverage, the Third Fleet staff fortuitously added a plane to fly the 165-degree line to three hundred miles.38
At noon, hearing no word of U.S. carriers, Hara proceeded with the Lunga airfield strike. Six Ryūjō carrier attack planes (each with six 60-kg land bombs) and nine strafing Zeros hoped to surprise the enemy planes on the ground. Six other Zeros served as escort. By 1250 the twenty-one planes departed south for Lunga two hundred miles distant. Hara could have had the strike stage to Buka while he withdrew his ships safely north to refuel. Thinking he was safe, he lingered to await the return of his planes in four hours. Hara’s overconfidence doomed the Ryūjō.39
At 1320 while Fletcher awaited possible fresh search contacts before committing the Saratoga strike, her new SG radar registered “many” bogies bearing 350 degrees, distance 112 miles. Instead of closing, the contact headed away southwest (220 degrees) on a tangent directly toward Lunga 235 miles distant. Only a carrier could have furnished such a strike, if indeed that was what it was. Given it was thought Japanese planes did not attack beyond 250 miles, the supposed flattop might lurk only 120 miles north of TF-61. In that case Enterprise searchers should have readily sighted her, but they remained silent. Fletcher ordered the Sara’s strike to launch immediately. The Japanese could attack at any time and possibly catch her planes on deck, a catastrophe to be avoided at all cost. Moreover, considerations of time and distance dictated his strike must depart very soon or risk returning after sundown.40
Lacking additional intelligence, Fletcher’s only viable target remained Burkey’s lone carrier actually reported at 0935, but which he thought was only found about 1100. Still reckoning a wrong position that was too far north, Pederson, allowing for the enemy’s maximum southward advance, specified the outbound strike leg as 320 degrees to 216 miles. If the target did not turn up there, Felt could count on it being off to the north. Such a long mission again precluded fighter escort, so Fletcher released the dozen Sara F4Fs to augment the combat air patrol. At 1340 TF-11 bent on twenty-five knots southeast into the brisk wind and lofted the fifty planes. Fletcher launched the attack “on the most meager information which was three hours old and had never been confirmed.” Such were the charming quirks of carrier warfare. For the time being he ignored the source of the supposed Guadalcanal strike, the possible carrier not far to the north. That was wise, for that flattop never existed. The radar contact was a phantom due to atmospheric interference. Hara’s Support Force was in excess of 150 miles northwest of the putative point of origin of that “strike.” No other ships were closer than one hundred miles.41
At 1338 while Fletcher unleashed the Sara’s strike group against the cold carrier contact, the Enterprise mustered eight F4Fs against a new bogey. The wily twin-float seaplane finally succumbed at 1401 near the Stewart Islands, thirty miles northwest of TF-61. Fletcher failed to realize the significance of one more snooper, but it meant the world to Nagumo. In this game of high stakes he, unlike Fletcher, did not juggle confusing and conflicting sighting reports. Nagumo had only one. Special Duty Ens. Fukuyama Kazutoshi’s Chikuma Number Two aircraft flew the special 165-degree line that was the last-minute addition to the late morning search. His all-important farewell was: “Spotted large enemy force. Being pursued by enemy fighters, 1200 [Z-9, local time 1400].” Nagumo did not receive that report until 1425 after the Chikuma blinkered it up the line. The message gave no position, but from the flight plan and elapsed time, Nagumo’s staff estimated with uncanny accuracy the target bore 153 degrees, distance 260 miles. The bearing was almost dead on and the actual distance to TF-61 only ten miles farther. The force catapulted float shadowers to amplify the contact.42
Nagumo now could attack. The Shōkaku and Zuikaku counted fifty-one Zero fighters, fifty-four Type 99 carrier bombers, thirty-six Type 97 carrier attack planes, and one Type 2 carrier reconnaissance plane. Nine Zeros flew combat air patrol. Swearing never again to be caught with armed planes on board, the Third Fleet staff devised three strike waves, each a small, handy deck load in size. Reckoning carrier bombers could more easily breach strong fighter defenses than torpedo planes, the first two waves would comprise all fifty-four carrier bombers, escorted by twenty-four Zeros. Their mixed loads of high explosive and semi–armor-piercing bombs would blast flight decks and antiaircraft guns and penetrate vitals to knock out propulsion. Only then would the third wave of thirty-six carrier attack planes and twelve fighters finish off the target with torpedoes. That tactic arose from the heavy loss of torpedo planes at Coral Sea and Midway. The sequence also offered the advantage of not having armed planes stacked in the hangars until the enemy’s offensive threat was neutralized.43
Unaware the die was cast, the Saratoga recovered three SBDs—the two orphans from Cactus and one strike abort. At the same time the Enterprise sent up eight combat air patrol F4Fs and prepared a strike “ready for possible second enemy carrier in area.” The remaining few Saratoga TBFs would join them “if ordered.” Fletcher stood northwest at fifteen knots to close the returning Enterprise afternoon search. News of a second enemy carrier reached him, even as the daily Cincpac intelligence bulletin falsely soothed the Shōkaku and Zuikaku had only come as far south as the “Truk area,” with “all other carriers [except the auxiliary Kasuga Maru] believed home waters.”44
At 1357 Kellam’s 3V37 warned of “suspicious vessels” and eight minutes later described them as one small carrier, two cruisers, and one destroyer headed southeast. The position attributed to Kellam’s contact converted to 345 degrees, two hundred miles from TF-61. That was “far from the position in which the first carrier was expected to be.” The Saratoga strike group, if it sighted nothing on its outbound leg (a distinct possibility because the original search contact was hours old), could end up nearly one hundred miles southwest of this new second carrier. Therefore at 1435 Fletcher radioed Felt to tell him that at “course 350 T[rue,] distance 150 from your present position 1 CV 2 CL 1 DD.” Fletcher had no idea whether Felt ever heard the message. The reported longitude of the 3V37 contact was actually 2 degrees of longitude farther west, for Kellam indeed spotted the already discovered Ryūjō force. This misstated position only added to the confusion. While Fletcher considered the grim implications should his main blow go astray, Kinkaid queried whether he, too, knew of the second carrier. Impatient to launch the Enterprise’s strike of eleven SBDs, eight TBFs, and eight F4Fs, Kinkaid also worried TF-61 would not advance far enough by 1630 to be where his returning afternoon searchers could find it. “We must make good course 335 speed 20 to intercept Point Option.” Fletcher at once complied and asked Kinkaid whether he had “heard anything of your [search] boys that went out about two hours ago?” Kinkaid replied, “Negative.”45
Fletcher still had only bits of the puzzle. At 1440 Cactus radioed it was under attack by twin-engine bombers, fighters, and carrier-type bombers, seemingly confirming the Sara’s radar contact of planes launched from a carrier not far north of TF-61. Yet the Enterprise searchers, who fully covered that area, never reported a carrier. Actually it was only the twenty-one Ryūjō planes that struck Lunga. Bad weather forestalled an earlier attack by the Rabaul-based land-attack planes. The Mackinac relayed the opinion of an unidentified PBY pilot (evidently Burkey) that the “carrier previously reported” was the distinctive, island-less Ryūjō. Fletcher could only hope Felt’s Saratoga flyers would see that for themselves.46 At 1500 the Saratoga landed twenty F4Fs (including sixteen Enterprise) and took over combat air patrol support. Kinkaid asked Fletcher whether he also copied a message from Enterprise scout 1V395, inevitably to be told no. He informed Fletcher that 1V395 sighted one carrier, one heavy cruiser, one light cruiser, and three destroyers bearing 320 degrees and 210 miles from TF-11. That was just about where Fletcher could expect the original 5V37 carrier contact to be by that time. It was, of course, once again Hara’s Ryūjō force that Burkey found nearly six hours before.
Soon after receiving word his planes bombed Lunga, Hara was snooped almost simultaneously by three separate Enterprise search sections: Lt. Cdr. Charles M. Jett’s two VT-3 TBFs (1V395 in the 320–330 sector), Lt. Stockton Birney Strong’s two VS-5 SBDs in the next sector north, and Ens. John H. Jorgenson’s mismatched pair of VS-5 SBD and VT-3 TBF closing from the southwest. Yet a fourth search section, two VT-3 TBFs led by Lt. John N. Myers, showed up a few minutes later. After transmitting its contact report, Jett’s section executed a horizontal bombing run from twelve thousand feet and missed the Ryūjō’s stern by only 150 meters. Zeros broke up Myers’s attempt at a similar attack and shot down his wingman, Mach. Harry Corl, a Midway TBD survivor. Unless Felt turned north against the second (Kellam’s 3V37) carrier contact, he aimed directly for Jett’s carrier, but Fletcher was unsure where Felt was going. He mulled whether to launch the second strike group (mostly Enterprise planes) against Jett’s target, but those planes must return in the dark. Making things much worse, Fletcher heard a different Enterprise searcher (9V395) advise at 1525 that he bombed, but missed, one of a pair of carriers. Maddeningly, the position of this vastly significant encounter did not come through, and Fletcher did not know the particular sector 9V395 handled. Kinkaid did but seems never to have provided Fletcher an estimate based on that knowledge. To bemused CTF-61, the enemy force now appeared to number four carriers: one pair, location regretfully unknown, and the two already reported, widely separated singletons.47
The adventures of the search planes covering the skies over Kondō and Nagumo exceeded even the swarm around the Ryūjō, but only snippets leaked to those who most needed to know. On its return leg Slater’s 8V37 PBY ran across the Advance Force, fought through the pugnacious Chitose seaplanes, and at 1346 transmitted a contact report of one heavy cruiser and four unidentified ships on course 140 degrees and 475 miles northwest of base. Not surprisingly, neither Fletcher nor Kinkaid received that particular message that would have reemphasized the peril north of TF-61. By 1430 the first representatives of the far-ranging Enterprise afternoon search also latched onto Kondō. Lt. John T. Lowe’s two VB-6 SBDs counted three heavy cruisers, several destroyers, and other ships on course 180 degrees, speed twenty knots. He tapped out a contact report and at 1447 shook up the Advance Force with near misses against the heavy cruiser Maya. The Enterprise immediately copied Lowe’s report of three heavy cruisers and three destroyers, course 180 degrees, speed twenty knots, and located them 240 miles NNW of TF-61 and just north of the false carrier contact attributed to Kellam. At 1449 Kinkaid forwarded the contact by TBS to Fletcher, who invariably did not receive it. Both admirals relied overmuch on TBS and should have used visual signals to confirm the receipt of key messages.48
A more spectacular discovery occurred at the same time Lowe attacked Kondō. Flying the adjacent sector to westward, VB-6 skipper Lt. Ray Davis (9V395) discerned Abe’s vanguard in the haze. While maneuvering to bomb one of the cruisers, he sighted farther north a very large carrier loaded with planes. She was Nagumo’s flagship Shōkaku in the process of launching her first wave of eighteen carrier bombers and nine Zeros. Several miles astern the Zuikaku followed suit with nine carrier bombers and six Zeros. Lt. Cdr. Seki Mamoru quickly formed up his attack group and started southeast. Eager to bag one flattop, Davis and wingman Ens. Robert C. Shaw swung upwind while climbing to fourteen thousand feet. They adroitly evaded combat air patrol Zeros but not the Shōkaku’s radar. Word of incoming enemy planes, the first to be detected by Japanese shipborne radar, failed to reach the Shōkaku’s bridge in timely fashion. No one warned the nine Zeros on combat air patrol of nearby intruders, although five of the nine Shōkaku escort Zeros broke away in a futile bid to catch the two SBDs. Only at the last instant did Shōkaku lookouts discern the danger. The carrier executed a sharp evasive turn just in time to evade one 500-pound bomb ten meters to starboard. The second bomb landed ten meters beyond. Then a terrible snafu bedeviled the American side. Davis’s carefully composed report (“2 CV’s with decks full,” with position and course) had, as noted, only reached the Enterprise and Saratoga in garbled form. Thus exceptionally poor radio reception prevented Fletcher and Kinkaid from hurling the second strike directly against Nagumo. The SBDs certainly shook up the Kidō Butai. Nagumo scrambled fifteen Zeros, raising the combat air patrol to twenty-nine, including the five escorts whose absence reduced Seki’s fighter protection to ten. The spotting and arming of Lt. Takahashi Sadamu’s second strike wave of twenty-seven carrier bombers and nine Zeros proceeded at a frenzied pace before more surprises dropped out of the sky.49
Ignorant of the position of Nagumo’s two carriers to the north, Kinkaid notified Fletcher at 1530 that Jett’s report of one carrier to the northwest was “good.” Fletcher replied: “If you consider contact good send your boys in.” Kinkaid confirmed at 1536 he would “launch when ready unless otherwise directed.” Four minutes later Fletcher advised he would turn southeast at 1800 to retire at twelve knots. Kinkaid must work in that course change when planning the second strike.50 Dwindling destroyer fuel was certainly at the back of Fletcher’s mind. The noon fuel reports revealed the TF-11 destroyers averaged 45 percent and the TF-16 destroyers 44 percent. Since then TF-61 steamed at high speed, with prospects for more sprints. The heavy ships still had plenty of oil, but they could not readily fuel destroyers in the midst of a battle.51
Kellam’s PBY 3V37 came across TF-16 while on its return leg to Ndeni. Unsure whether base ever copied his original contact reports, Kellam flashed by blinking light to the Enterprise: “small enemy carrier bearing 320 True distance 195 miles.” Along with the correct position (as opposed to the earlier error in transmission), Kellam gave its course and speed and mentioned he saw a separate group of three light cruisers, two destroyers, and three transports fifty miles beyond. Kinkaid forwarded Kellam’s carrier contact to the Saratoga at 1543 but oddly omitted the convoy. The position Kellam gave for the carrier proved almost identical to what Jett provided. Fletcher replied, “[It] appears to be same one as from your aircraft.” A few minutes later he overheard welcome radio traffic from Saratoga flyers pounding an enemy carrier. Which of the two solitary flattops was still unclear, but Fletcher was greatly relieved his strike found one of them. Kinkaid, though, balked at dispatching the second strike unless Fletcher assured him TF-61 would continue northwest until they could be recovered. Otherwise he judged the mission as too long. Fletcher responded at 1558 that he would not hold Kinkaid’s proposed northwesterly Point Option course, “Due to possibility of night torpedo attack.” Again like Spruance at Midway, he feared bumping into a surface force after dark, a unappealing prospect especially when the hidden main body included battleships and heavy cruisers in addition to a swarm of torpedo-wielding destroyers.52
At 1602 with the second TF-61 strike still under deliberation, both carrier radars registered a large group of bogeys one hundred miles northwest. Seki’s first strike of thirty-seven planes was coming straight in. At that same moment, 225 miles NNW of TF-61, the Shōkaku and Zuikaku dispatched Takahashi’s nearly identical second wave. Abe’s vanguard force surged ahead to finish off the U.S. carrier force in night surface combat. Time indeed expired for Fletcher, for TF-61 was about to suffer the feared air attack from an unlocated force. Even so, the gremlins that interfered all day with the American search effort finally afflicted the Japanese. The Hiei seaplane in the sector adjacent to the dead Fukuyama reestablished contact with the U.S. striking force. However, its contact report placed Fletcher’s carriers fifty miles farther south than their actual position and also had them moving away southeast at twenty knots. Evidently it must have sighted TF-61 during one of its brief stints in the wind for air operations in between lunges to the northwest. Thus ill advised as to the enemy’s position and course, Takahashi started southeast expecting to find the target around 1800, close to sundown. Nagumo’s carriers turned east to await the results of the two strikes.53
When the incoming attack first appeared on radar, TF-61, heading northwest at twenty knots, was 140 miles east of Malaita and thirty miles southeast of the Stewart Islands. Fletcher’s own TF-11 (Saratoga) was in the lead, Kinkaid’s TF-16 (Enterprise) holding station ten miles off its port quarter. Fletcher brought TF-61 all the way around southeast into the wind to deploy additional combat air patrol fighters and unleash his hoarded second strike. Working up to twenty-five knots, the two task forces burrowed through “long heavy swells,” with destroyers “bouncing almost out of the water.” Fletcher told Kinkaid to “close me immediately” and ordered him to dispatch his strike. “To say the least we were in a bad predicament,” Pederson recalled. “All of our attack planes were committed on missions with the main enemy force still unlocated and his planes coming in to attack us. The best we could do was to get ready for an air attack and hope for the best.”54
TF-61 finally had an adequate number of defending fighters. Twenty-four F4Fs circled overhead, with another twenty-two Wildcats poised on the Saratoga. Rowe, the Enterprise FDO, requested the Sara immediately scramble all sixteen Enterprise Wildcats and hustle them northwest for distant intercept. The Saratoga then cleared her deck of all remaining planes: a half-dozen VF-5 Wildcats to reinforce the combat air patrol and the five torpedo-equipped VT-8 TBFs and two VB-3 SBDs armed with 1,000-pound bombs. At 1620 the Saratoga told VT-8 skipper Lt. Harold H. Larsen and Lt. (jg) Robert M. Elder, leading the VB-3 section, to join the Enterprise strike group about to be launched. The Saratoga then landed four fighters, whose pilots jumped out to help the plane handlers push the planes aft so they could refuel and get away as quickly as possible. Spurred by reports of enemy aircraft at twenty-seven miles and closing, Kinkaid released the Enterprise to launch seven F4Fs, eleven SBDs, and eight TBFs. Aware of a firm position only for Burkey’s original single-carrier contact (amplified by Jett and Kellam), he reluctantly directed Max Leslie, the group commander, to attack that target even though the first Saratoga group may already have done the job. The distance to the objective, 260 miles, precluded fighter escort. Kinkaid reassigned the seven F4Fs to the combat air patrol. He also authorized Leslie to take refuge at Cactus should he find it too hard to return to TF-61.55
Eleven combat air patrol F4Fs hunted forty miles northwest. Sixteen more hastened out in support. Twenty-six fighters, including the four about to take off from the Saratoga, remained near the ships in reserve. That raised the total to fifty-three Wildcats. At 1629 the fighter circuit radio receiver in the Sara’s flag plot sounded warning from Gunner Charles E. Brewer of VF-6: “At 1 o’clock above this is Red 2, tally ho! There are about 1 or 2 . . . 9 bogeys unidentified about 12,000 feet, 300. . . . Many ahead of those. . . . Bogey bears 350 [degrees], 20 [miles], Angels 12 [altitude twelve thousand feet]. They’re dive bombers.” The fighter circuit became so overloaded, no one could figure out exactly what was going on. Fletcher continued relaying the bearing and distance to the enemy planes and hoped “for the best.” Garbed in blue “anti-flash” suits to help prevent burns, he and his staff moved outside to the flag bridge for a ringside view. One level above, Clark Lee stood alongside Schindler and the lookouts ranged around sky forward. Beyond the port quarter, nearly ten miles off and in direct line of approaching enemy planes, TF-16 presented a brave sight. Eight screening ships in tight Victor formation deployed in a two-thousand-yard radius around the Enterprise, with the massive North Carolina five hundred yards farther astern. Kinkaid bent on twenty-seven knots (top speed for the battlewagon) and readied for more, while TF-11 made twenty-eight knots. The last plane, Leslie’s TBF, lumbered down the Enterprise’s flight deck at 1639. Greatly relieved to get away prior to the actual attack, he sought his flock of SBDs and TBFs forming up for the strike. Low on gas, three antisubmarine SBDs and at least half the twenty Enterprise search SBDs and TBFs circled hoping to keep out of trouble.56
At 1640, as Brewer radioed, “OK, let’s go give them hell,” watchers in the Saratoga noticed “a puff of smoke” materialize northwest beyond TF-16. A Zero downed by the combat air patrol fell with a “long plume of smoke trailing into the sea.” Instantly, “Another Jap plane plunged in flames, until the air was full of those long trailers of smoke.” The Enterprise and her screen “burst into flame.” The “sky blossomed with the black and yellow bursts of their exploding shells.” Overhead, planes “flashed through the smoke, the sunlight glinting on their wings.” The North Carolina looked like she was burning, “So concentrated and intense was her AA fire.” Kinkaid maneuvered evasively, while Fletcher turned TF-11 northeast out of the wind to force dive bombers to attack crosswind. At first the Enterprise weathered the assault with only splashes from near misses, but at 1644, after a bomb struck aft on the flight deck, a “long streamer of smoke trailed out to the rear of the still speeding ship.” Onlookers in TF-11 also noted at least one other hit, but the Enterprise did not appear badly damaged. They also perceived a “deliberate attack” against the North Carolina.57
Observers in the Sara also wondered when the numerous attackers would hit TF-11. Next to Lee a lookout “tensed suddenly” and screamed, “Dive bombers overhead coming down out of sun!” Then, “Torpedo planes coming in from port quarter!” The Enterprise likewise reported torpedo planes, but none flew the strike. The cause of the Sara’s alarm proved to be a hurt F4F seeking to land. At 1655 Fletcher turned into the wind to accommodate the cripple, which flipped over and ensnared itself in the wire barrier. TF-16 ceased fire at 1656. To Lee it seemed the Japanese came and went with “incredible speed.” Fletcher radioed Ghormley and McCain: “First attack completed; Enterprise hurt; another attack coming in.” Surprised neither the fifty-odd dive bombers nor deadly torpedo planes ever pounced on TF-11, he did not know only Seki’s twenty-seven dive bombers and ten Zeros executed the strike. Eighteen Shōkaku carrier bombers went after the Enterprise, and nine from the Zuikaku stalked the Saratoga. Intense antiaircraft caused three Shōkaku carrier bombers to switch to the North Carolina, while F4Fs turned away the Zuikaku planes short of TF-11. Three joined the assault on the Enterprise, and four took on the North Carolina. Now all depended on whether the second Japanese strike could follow up this attack.58
