In the meantime, after Admiral Mitscher told TF 58 that it appeared the Japanese were coming their way, Gus Widhelm had been having a hard time getting takers for his $1,000 wager that the Japanese would be heading out for a carrier duel.
Far to the west of Saipan, Ozawa’s forces were plowing deeper into the Philippine Sea. At 1000 on the 16th, Admiral Ugaki’s battleship force rendezvoused with the 1st Supply Force at 11°00′N, 130°00′E. Fueling began immediately while the two units headed north toward the rest of the Mobile Fleet. At 1650 Ugaki joined Ozawa and the entire Mobile Fleet was finally assembled.
After the rendezvous, 1st Supply Force began fueling the rest of the Mobile Fleet in preparation for the coming action. Fueling was leisurely, not being completed until 1300 the next day, at which time the Mobile Fleet was at 12°15′N, 132°45′E. The oilers then broke off to join with the 2nd Supply Force which had left Guimaras on the 15th and had since been bringing up the rear of the Mobile Fleet. When the two provisioning units met, they turned northeast and headed for a position at 14°40′N, 134°20′E, where they were to stand by for further use. Ozawa was now a little over 750 nautical miles from Saipan.
Ozawa was biding his time. He and most of his other top commanders had great faith—misplaced, as it turned out—in the operations of their land-based planes in the battle. These planes would severely damage the U.S. forces, thus facilitating the Mobile Fleet’s later attacks. Ozawa also knew that he could stay out of range of TF 58’s planes because his planes had a much greater radius of action than the Americans. (Generally, they had an advantage of 350—560 miles in the search role, and 200—300 miles in the attack role.) One other advantage fell Ozawa’s way. The prevailing wind was from the east, which meant that he could launch and recover planes while heading toward the enemy. Mitscher, on the other hand, would have to keep turning east while air operations were in progress and would not be able to make much headway to the west.
A canny sailor, Ozawa was also pretty sure he knew the psychological makeup of his opponent, Spruance. The Japanese admiral figured that Spruance was a conservative and deliberate commander—one not inclined to take risks. He fully expected Spruance to sit close to Saipan and take no offensive action unless he had to.
Even though he was outnumbered fifteen carriers to nine, and two to one in planes, because of the “advantages” mentioned above Ozawa felt he had a fighting chance to destroy the Americans. The one factor that limited Ozawa was fuel. He had only enough to come straight at the enemy (a fact of which the Americans were, naturally, unaware), and the elaborate and complicated plans the Japanese loved to use could not be employed this time.
It appeared to the Japanese that part of their plans was already working, for on the 16th a Betty from the 755th Naval Air Group and four Jills of the 551st Group (all based at Truk) reported attacking U.S. vessels off Saipan. One cruiser was claimed sunk and two others damaged. Fifth Fleet kept operating, though, not even knowing it had been attacked.
Another of the ubiquitous U.S. submarines came across the Japanese on the 16th. The Cavalla (a new sub commanded by Commander Herman J. Kossler) was patrolling 360 miles east of San Bernardino Strait in company with the Pipefish. Although intelligence reports had presumably put them right on the track of the Mobile Fleet, a fruitless day of searching had turned up nothing. That evening Kossler headed for San Bernardino Strait to relieve the Flying Fish.
Shortly after 2300, while proceeding on the surface, Kossler got a radar contact. It was a small force, only four ships. Kossler brought the Cavalla in for a closer look. It was two oilers escorted by a pair of destroyers. He had stumbled on the 2nd Supply Force! Kossler ran ahead of the enemy ships and dived about 0340. Submerged, the Cavalla sneaked in for an attack. Just as Kossler was about to fire at an oiler, one of the destroyers charged. Kossler went deep to evade the attack. When he brought the Cavalla back up at about 0500, the enemy was nowhere in sight.
Kossler decided not to chase the oilers. His orders were to relieve the Flying Fish, and he had already wasted a day and a lot of fuel in the fruitless attack on the supply group. However, when he radioed his decision to Vice Admiral Charles A. Lockwood, ComSubPac and also commander TF 17, a quick change of orders came flashing back: “Destruction these tankers of great importance. Trail, attack, report.”
Kossler turned around and took his submarine back down the supply force’s estimated track. It took some time, but at 2000 on 17 June Kossler hit the jackpot. While the Cavalla was proceeding on the surface, her radar picked up seven large “blips” about twenty thousand yards away. Kossler dove and closed with the target.
The Mobile Fleet steamed by the Cavalla as Kossler took in the parade with interest. Although he could have attacked, Kossler knew he had to get the information out first. After surfacing at 2245, he reported fifteen or more ships doing 19 knots and heading due east. Their position was reported as 12°23’N, 132°26’E. (Actually, the Mobile Fleet was about sixty miles northeast of that position.) Because it was dark, Kossler missed many of the ships as they ran past, but the number of ships he did report—fifteen—would worry Spruance.
Although Lockwood appreciated the information Kossler had sent, he thought it was now time for action. He told Kossler and all the other submarine skippers in the area to shoot first and talk later. For his skippers’ further edification another message said, “The above list of enemy ships does not frighten our varsity. We have all that and plenty more ready and waiting and they are all rough, tough and nasty.” Lockwood further ordered Kossler to, “Hang on and trail as long as possible regardless of fuel expenditure. . . . You may have a chance to get in an attack.” Kossler had lost the Mobile Fleet by the time he got Lockwood’s message, but he determinedly took his boat toward where he thought the Japanese would be. He would be rewarded for his chase and the Cavalla’s crew would yet see some action.
Both sides were busy making final plans on the 17th. Operating under radio silence, Ozawa sent a Judy to Peleliu just before noon with a request for land-based air operations. Ozawa told Combined Fleet Headquarters and Fifth Base Air Force:
The First Mobile Force, being at location ‘E’ on the evening of the 17th and having finished supplying operations, will advance to a general location west of Saipan by dawn of the 19th, going via point ‘O’ [possibly a translation error for ‘C’] at 15.0 N latitude, 136.0 E longitude. In the meantime, this fleet shall guard against westerly advances of the enemy and their movements from the north. The objective is first to shell regular aircraft carrier groups and then, by employing all fighting power, to annihilate the enemy task forces and their invading forces. The following are the requests made of land-based air units:
1. It is requested that, from the evening preceding the decisive battle, you shall maintain a constant reconnaissance of the regular aircraft carriers of the enemy in the vicinity of the Mariana Islands. If this is impossible, notify us immediately of the condition and deployment of regular aircraft carriers as of noon.
2. We request intensified patrolling of the area west of the Marianas by each base on the day previous to the decisive battle. Special attention shall be paid to carry on reconnaissance in the sector from 160 degrees to 210 degrees from Iwo Jima. [Ozawa was figuring on the possibility of just the sort of end-run Jocko Clark had in mind.]
3. If the forces of the Yawata unit are not deployed on time, it is believed we shall be forced to delay the decisive battle by one day. Please notify us of such a probability.
Ozawa was as yet unaware that the land-based phase of A-GO had gone seriously awry. Kakuta certainly was not telling him.
Off Saipan, Reeves’s and Montgomery’s task groups finished fueling shortly after midnight. The two groups wound up much farther east than planned, and Mitscher ordered them to make 23 knots to the west. Task Force 58 had to get as far west as possible because launching and recovering its planes meant the carriers would have to turn back into the easterly wind and, consequently, would not make good much distance toward the enemy.
Because of their distance to the east, TGs 58.2 and 58.3 sent night searches out 270—350 miles at 0200, half an hour later than planned. The two groups kept heading west until 0430. At 0700 the searchers were recovered. They had not seen anything. As the first searchers returned, another group of Helldivers and Avengers were launched to search to the west and southwest, to a distance of 325—350 miles. A third search, launched at 1330 by the Bunker Hill and Lexington, was as unsuccessful as the earlier attempts. Clark and Harrill, meanwhile, were ordered to search as far west as possible and to keep the area east of 138 degrees and south of 12 degrees covered.
Admiral Spruance ordered a minimum of air operations for the day, primarily search missions, but Mitscher thought it necessary to send in more strikes over Guam. About seventy-five sorties were flown in the afternoon. The strike “temporarily” closed Agana, but was costly to the Americans; several planes and pilots were lost to the deadly flak. The fliers were somewhat bitter, feeling that TF 58’s battleships (at this time only preparing to form TG 58.7) should have been used to knock our the antiaircraft guns before the planes went in.
In the afternoon Mitscher radioed Spruance giving him the present and planned dispositions and movements of TF 58,
1.Present status:
(a)Task Group 58.2 is 12 miles south of Task Group 58.3
(b)Task Group 58.3 will be in Lat. 15°N, Long. 144°30’E at 1600 today.
(c)A search was launched at 1330 distance 325 miles, betweeen bearings 215—285. This search is to be recovered about 1830 in vicinity Lat. 15°N, Long. 144°30’E.
2.Recommended disposition upon the joining of forces from Task Force 51:
(a)Task Group 58.2 composed of carriers, CruDiv 13, DesRon 52, DesDiv 1; 12 miles south of Task Group 58.3.
(b)Composition of Task Group 58.3: carriers, CruDiv 12, DesRon 50 and DesDiv 90.
(c)Task Group 58.7 composed of battleships, CruDiv 6, DesDiv 12(16 torpedoes each), DesDiv 89, and DesDiv 106; stationed 15 miles west of Task Group 58.3.
3.(a)If battle is joined before Task Groups 58.1 and 58.4 join us, Task Group 58.2 will be designated battle line carriers.
(b)When Task Groups 58.1 and 58.4 join, propose to put Task Group 58.1 12 miles north of Task Group 58.3, and Task Group 58.3 and Task Group 58.4 12 miles south of Task Group 58.2.
(c)As soon as Task Groups 58.1 and 58.4 join, propose to have San Juan join Task Group 58.2 and Reno join Task Group 58.3 so that one CL(AA) will be with each carrier group.
(d)If battle is joined after Task Groups 58.1 and 58.4 join us, Task Group 58.4 will become battle line carrier group.
(e)After first air battles have been fought and we have control of the air, recommend CruDivs 10, 13 and 12 and DesDivs 11, 1 and 90 be released from carrier groups to join Task Group 58.7.
(f)After initial air battle, or before if it becomes feasible, recommend Task Group 58.1 take station about 50 miles to the northwest of Task Group 58.3 in order to hit Japs from northern flank and cut them off from escaping to the north.
4.Recommended movement tonight; at 1800 course 310° until reaching Lat. 16°N, then course 270° until after daylight launch. It is hoped this will permit us to flank the enemy, keep outside of 400 miles range of Yap and keep as far from other shore-based air flown in to Rota and Guam as practicable, and still be in position to hit enemy carrier groups (downwind from us).
5.As soon as things quiet down a bit, one Task Group at a time should be refueled in vicinity of Marianas, during which time it can assist Task Force 51 on Guam, Rota, or Saipan as directed.
Spruance initially approved Mitscher’s plan, but late the next day would change his mind and hold TF 58 close to Saipan. Mitscher, in the meantime, was going ahead with his preparations: “Proposed plan for strike on enemy surface forces,” he signaled his carriers. “Make deck load launch from CVs consisting of 16 VF, 12 VB and 9 VT. Second deck load prepared for launch as second wave unless situation indicates delay advisable. Augment VT from CVLs as practicable. Arming VT half torpedoes, VB half GP, half SAP. Later strikes include AP as targets indicate.”
Admiral Spruance issued his battle plan at 1415, saying, “Our air will first knock out enemy carriers as operating carriers, then will attack enemy battleships and cruisers to slow or disable them. Task Group 58.7 will destroy enemy fleet either by fleet action if enemy elects to fight or by sinking slowed or crippled ships if enemy retreats. Action against the retreating enemy must be pushed vigorously by all hands to insure complete destruction of his fleet. Destroyers running short of fuel may be returned to Saipan if necessary for refueling.”
This “bare bones” plan sounded aggressive enough—but, like so many good plans, it never bore fruit. Also striking, given Spruance’s concern about such a tactic, is the fact that this plan makes no mention of a Japanese end run. A message to Spruance from Admiral Nimitz in the afternoon should, nevertheless, have given the Fifth Fleet commander some thoughts about sticking with his battle plan. “On the eve of a possible fleet action,” Nimitz radioed, “you and the officers and men under your command have the confidence of the naval services and the country. We count on you to make the victory decisive.”
At 1741 Spruance in the Indianapolis, plus CruDiv 12 (the Cleveland, Montpelier, Birmingham), joined TG 58.3. After joining, Spruance signaled Mitscher, “Desire you proceed at your discretion, selecting dispositions and movements best calculated to meet the enemy under the most advantageous condition. I shall issue general directives when necessary and leave details to you and Admiral Lee.”
This message left Mitscher in a quandary. It had been sent by Spruance in reply to a query by Mitscher regarding the tactical command of TF 58. The tenor of the message, however, suggested to Mitscher that he would not have complete control over TF 58. Spruance would be continually hovering nearby to approve or disapprove any orders. To preclude any conflict between Mitscher and Spruance, the TF 58 commander “preferred to submit his proposed courses of action to Admiral Spruance before they were put out of order, which meant that Admiral Spruance, although not taking OTC, was actually operating as OTC.”
During the afternoon Mitscher sent word to the two task groups at hand to prepare to dispatch their battleships and some escorting vessels to form the battle line, TG 58.7. Mitscher figured that with the usual confusion when ships move in and out of formation, it would be better to form TG 58.7 early, rather than wait until the Japanese were nearby to interfere with this movement. At 1730 the seven fast battleships (the Washington, North Carolina, Iowa, New Jersey, South Dakota, Alabama, and Indiana), along with four heavy cruisers and thirteen destroyers, left the carrier groups and formed the battle line. Able and aggressive Vice Admiral Willis A. “Ching” Lee commanded the task group. After all the ships had rendezvoused, they took up station fifteen miles west of TG 58.3. Mitscher had made his final dispositions and was now awaiting the return of his two other groups the next day.
The searches during the day had turned up no ships, but the afternoon search had run into several enemy planes which had soon been shot down. Surprisingly, it appears that the American commanders took little note of these planes. The positions of the air actions and the apparent direction the Japanese planes had come from did not lead anyone to guess the line of advance of the Mobile Fleet. Captain Burke said only, “those enemy planes probably did report our position.”
While the Americans were still vague about the position of the Mobile Fleet, Japanese land-based air knew full well where the action was. On the evening of the 17th, several heavy attacks were launched by the Japanese Base Air Force against U.S. forces off Saipan. At 1750 five Jills and an Irving night fighter from Truk attacked an “enemy transport convoy.” What they actually found was Captain G. B. Carter’s tractor group (TG 53.16) of sixteen LSTs, seven LCI(G)s, nine sub chasers, and the destroyer Stembel. The group was part of the Southern Attack Force and was killing time waiting for the expected Guam invasion. The Japanese fliers claimed that they sank “thirteen transports and left one destroyer listing heavily.”30 What they actually got in return for the loss of three planes was a torpedo hit on LCI(G)-468. The explosion killed fifteen men and wounded three. The gunboat was severely damaged and eventually had to be scuttled.
A second, larger attack hit the Americans unloading at Saipan around dusk. This attack was apparently picked up by TG 58.2 radars about 1735, heading toward Guam. Because the estimated twenty to thirty planes were over one hundred miles away, no interception was attempted. But there were more than thirty planes. Thirty-one Zekes, seventeen Judys, and two Frans from Yap ignored the fire-support group and went for the ships unloading off Charan Kanoa. An LST was hit and caught fire, but the blaze was put out and the vessel salvaged. Turning back from this attack, the enemy planes ran across the jeep carriers maneuvering offshore. Although radar picked up the incoming planes, the fighter direction was inaccurate, and forty-six Wildcats went off on a wild goose chase.
Now, with only the ships’ antiaircraft fire to confront them, the Japanese attacked. In the dim light many of the pilots thought they were attacking the fast carriers of TF 58. Bombs just missed Gambier Bay and Coral Sea. But the Fanshaw Bay was not as lucky. A 550-pound bomb sliced through her after elevator and exploded on her hangar deck. Eleven men were killed. The fires caused by the explosion were quickly put out by her crew, but the Fanshaw Bay had to retire to Eniwetok for repairs.
Five of the Judys and several Zekes landed after dark on the “temporarily out of commission” field on Guam. The other planes retired to Yap. The Japanese fliers were jubilant, thinking they had sunk two or three fast carriers and had left another burning. To the Japanese, who were not yet aware of the miscarriage of the land-based phase of A-GO, it appeared that these planes were doing their job well.
The problems for the escort carriers were not over yet. As two White Plains Wildcats were returning to their carrier, they were fired on by “friendly” ships, then jumped by four other Wildcats of the CAP. Although neither of the two fighters was shot down, one was so badly damaged that on landing it crashed into five other aircraft and all six had to be written off.
Affairs around Saipan finally quieted down during the night, but tension was steadily building among the sailors of TF 58. Where were the Japanese? It was not too long before TF 58 found out where the Mobile Fleet had been. The Cavalla’s contact report reached Spruance at 0321 on the morning of the 18th, and Mitscher had it twenty-four minutes later. But now a divergence of viewpoints between the two commanders emerged.
Mitscher and his two “braintrusters,” Burke and Hedding, did some quick figuring. If the Japanese kept coming at 19 knots, they would be about 660 miles from Saipan at dawn and 500 miles from TF 58’s proposed 0530 position. That was still too far for any attack on the enemy, but by steaming directly for the enemy’s estimated 1500 position, TF 58 might be able to get in one strike in the late afternoon. But TF 58 was then widely separated, with TGs 58.1 and 58.4 far to the north of the other two groups. A noon rendezvous was planned and Clark and Harrill were ordered to link up with the rest of TF 58 as soon as possible. Mitscher could have headed west with the two groups he had on hand, letting the other groups catch up as best they could, but he preferred to be sure all his forces were concentrated for the coming action. When all his ships were together Mitscher would take TF 58 westward for the attack on the Mobile Fleet.
Admiral Spruance had other plans. By the evening of 17 June he was worrying about a flank attack. He later commented, “At dark on 17 June the situation appeared to be as follows: Enemy forces probably consisting of 5 BB, 9 CV, 8 CA and a number of destroyers were at sea east of the Philippines for the purpose of attacking our amphibious forces engaged in the capture of Saipan. The task of Task Force 58 was to cover our amphibious forces and to prevent such an attack. The enemy attack would probably involve a strike by carrier-based aircraft, supported and followed up by heavy fleet units. The possibility existed that the enemy fleet might be divided with a portion of it involving carriers coming in around one of our flanks. If Task Force 58 were moved too far from Saipan before the location of the enemy was definitely determined, such a flank attack could inflict heavy damage on our amphibious forces at Saipan. Routes of withdrawal to the northward and to the southwestward would remain open to such a flanking force. The use of enemy airfields on Guam and Rota were available to the enemy except as our carrier-based aircraft were able to keep these fields neutralized.”
The fifteen ships that the Cavalla had reported also worried him. “It appeared from the Cavalla reports, however, that the entire enemy force was not concentrated in one disposition; that if the force sighted by the Cavalla was the same as that sighted by the Flying Fish in San Bernardino Strait, a speed of less than 10 knots had been made good; and that the position of the Cavalla contact indicated a possible approach to the Marianas by this task group via the southern flank.”
Spruance apparently did not consider that in the darkness the Cavalla could have missed many of the ships (which she had). Forty ships spread over a wide expanse of ocean cannot be easily seen by a submarine at periscope depth. He also reasoned that if the force the Cavalla sighted was the same one the Flying Fish had reported, it had made only a very slow advance. Even with the sighting of oilers nearby, it apparently did not occur to Spruance’s staff that this slow advance could have been because the Japanese were fueling (which they were). And always there was Spruance’s extreme concern, almost obsession, with the possibility of a flank attack. Yet this concern did not cause him to modify his battle plan, which never mentioned that possibility. Despite his aggressive battle plan of the day before, Spruance was beginning to settle into a defensive posture.
