Chapter III: Guadalcanal--The First Three Days

Task Group X-RAY, containing the Guadalcanal assault Forces, separated from Task Group YOKE northwest of Cape Esperance and approached the transport area off Beach Red in double column at 12 knots.1 No sign of enemy activity was encountered and the preliminary naval bombardment of the coastal area, which began at 0613, elicited no response. At 0645 the transport area was reached. Five minutes later the signal "Stop" and "Back" came from the Hunter Liggett, followed immediately by the traditional Navy signal to commit the troops--"Land the landing force." The double column dropped anchor and stopped in perfect formation, and landing craft began moving over the sides.

The command post of the division had opened on board the McCawley at 0519.2 Eight minutes later, General Vandegrift was asked by Admiral Turner to set the time for H-hour. He did so, selecting 0910, and then, as news of the successful landings on Florida and Tulagi began to come in, the first waves of assault troops moved toward Beach Red.3

There was still no sign whatever of incipient or potential opposition. At 0859, 11 minutes before the first troops were scheduled to go ashore from the landing craft, reassuring words came from an observation plane from Astoria operating in the vicinity of the beach. No enemy could be seen in that area.4 Fifteen minutes later, however, word came from the same source that trucks were moving on Lunga Field, several thousand yards west of the landing beach. This news brought forth a request for an air strike against that area. While the request was being complied with, by VS-3s from Saratoga5 a message came from the beach that the troops were going ashore at 0919.6 There was no opposition at the beach.

Intelligence Situation

In order to understand the plans for landing and for the subsequent maneuvering of the troops, it is necessary at this time to review the intelligence situation as it existed from the beginning.

To put the matter succinctly, the division planners and ultimately all echelons of command above that level had only the haziest idea of the terrain of Guadalcanal. Sailing charts which contained data assembled over a period of years from official hydrographic surveys and from information given by navigators contained little that could be used by a staff planning a land campaign on Guadalcanal. No maps worthy of the name were in existence. It had been necessary, as we have seen, for the Intelligence Officer of the division to enlist the services of former residents of the territory and get from them, by word of mouth, some idea of the terrain upon which the division was to operate.

The information so gathered was extremely


spotty in quality. Whereas it was possible for the planners to get a good idea of the character of that portion of the island immediately adjacent to the coast, and particularly that portion which was threaded by the small coastal roads and trails, it was impossible to obtain any detailed information whatever regarding that part of the island which lay inland, away from the cultivated coastal area. It soon became apparent, indeed, that while the British or Australian residents of the district had a fair knowledge of the territory immediately surrounding their area, they were nearly as ignorant as the newcomers about the terrain features of the jungle-covered interior of Guadalcanal.

Two outstanding examples of the faulty nature of this advance information can be pointed out here. The first example is more amusing than important--that which has to do with the scrambling of the names Ilu and Tenaru. Three rivers flow into the coastal plains in the vicinity of Beach Red. The largest and by far the most important is the Lunga, and about that there was never any doubt. The second most important from the point of view of the landing is a long and relatively swift stream arising in the foothills behind the plains and descending to the sea in a series of broad, shallow rapids and deep, almost stagnant pools. In general character it resembles the Lunga, which also is long and varied in appearance. Between these two rivers lies another completely unlike them in appearance and characteristics, a deep, sluggish, lagoon-like body of no great length.

According to the earliest chart available (No. GC 146, 1937), the river which constituted the western boundary of Beach Red is the Tenaru. Flowing into this river almost at the mouth and from the right bank, there is a small stream which, in the days before the war, was known to the natives as the Ilu. The sluggish stream between the Lunga and the Tenaru seems to have been unnamed by the natives since its relatively short length and its similarity to countless other lagoons on the coast reduced it to comparative unimportance from their point of view.

Basic to the misunderstanding in the subsequent Gordian tangle of the names is the fact that near the mouth of this lagoon there stood a native village called Tenaru. Pilot Officer Charles V. Widdy, RAAF, a former resident of the island and a volunteer for duty with the division, is said to have pointed out the village and said that it was Tenaru. He was thought to be referring to the river and the name was attached to the undeserving creek. Since it was also known that the Ilu was a stream lying to the east of the main body of the Tenaru, logic demanded that the name be attached to the long swift river at the extremity of Beach Red.

Much alter, when Army forces on the island were busy with cartography, they attempted to rectify the situation, and with something approaching logic they contended that if the Ilu really was the Tenaru, then the Tenaru must, of necessity, be the Ilu. This reasoning was followed on maps that were made subsequently, and the rivers were so marked on signs erected upon the bridges that crossed them. So involved did the whole matter become, indeed, that it had unforeseen repercussions, for by 1945, the natives, hopelessly confused by the controversy were themselves no longer sure which river was which.

So the matter stands today, but the first and perhaps the decisive land action of the campaign was fought at the mouth of the unnamed lagoon known temporarily and erroneously as the Tenaru. later in the campaign, the lagoon became known as the Alligator Creek--it had no native name--and the Tenaru once more bore its correct name. For the purposes of this monograph, the name Alligator Creek will be used, and the battle fought at its mouth will continue to be known as the Battle of the Tenaru.

The second example of erroneous information--given and received in good faith--was that dealing with the terrain feature known as Mount Austen. In General Vandegrift's operation order, the 1st Marines were assigned an objective described as a grassy knoll four miles south of Lunga. This knoll had been described several times to planners


by former residents as the commanding terrain feature of the vicinity, and the implication was that it was close to the Lunga plains and easy of access from them.

The advance information about Mount Austen was correct in only one respect--it was a commanding terrain feature. The idyllic name of "Grassy Knoll" actually described a rugged truncated hill mass some 1500 feet in height, covered with alternating patches of rain forest and kunae grass, surrounded on its lower reaches by a chaotic system of cliffs, ravines, and gorges--all overgrown with lush and almost impenetrable vine growth. It lay, furthermore, perhaps six miles west and eight miles south of Lunga Point, which would place it some ten miles southwest of Beach Red, and was separated from that beach by rivers, jungles, precipitous ravines, and small plains covered with six-foot kunae grass. Parenthetically, it may be said here that Colonel Cates, commanding the 1st Marines, realized almost at once that the possibility of his attaining that objective was so remote as to be fantastic, and General Vandegrift, upon receipt of his advice concurred immediately. As a matter of act, the hill mass presented a constant threat to the airfield and to the troops in the vicinity of it in spite of constant patrolling begun immediately after the perimeter was set up.

Lest is seem that this matter of faulty intelligence is being labored, it must be emphasized, that in this, the first land offensive of American forces during the second World War, the plans which were formulated for the initial assault were based on nebulous, often contradictory, and seldom accurate information. The fact that the character of such information was recognized quickly, and that plans for continuing the campaign were changed radically in accordance with correct data, speaks eloquently for the ability of General Vandegrift and his staff to improvise quickly and effectively.

Information regarding the size of the enemy force on Guadalcanal was meager in the extreme. Coastwatchers on the island had kept as close a watch as was humanly possibly and had forwarded the information so obtained through their regularly established communications channels.7 They had likewise made use of information obtained from loyal natives whose good faith was unquestioned, but whose mathematical powers were not too well developed. On the basis of such data, it was at first estimated that approximately 5,000 enemy were present in the Lunga district.8 It was thought that this total included about 2,100 labor troops, a reinforced infantry regiment, and an antiaircraft battalion.9

The plans for the attack of necessity had to be based upon such information as was at hand, and they called for a landing which would take advantage of, first, the probably localization of enemy forces around Lunga Point and, second, the characteristics of the terrain itself. Since it was understood that the enemy would be concentrated in the vicinity of the airfield then under construction, and since a direct frontal attack was to be avoided if possible, it was decided to select a beach far enough from the area to allow time for landing and deployment of troops, and for maneuver prior to actual assault. It was desirable also to select a beach the configuration of which was such that some protection could be obtained from the terrain itself.10

The beach finally chosen lay about 9,000


yards southeast of Lunga Point. A strip 1,600 yards in length was selected at the deepest point of a gentle re-entrant in the coast-line. The right flank (northwestern extremity) of the beach was marked by the Tenaru River, which, it was thought, would offer a natural line of defense against any possible interference with the landing from the west or the northwest. The eastern extremity had no such natural protection.

LOYAL NATIVES such as these, together with their leader, Captain Martin Clemens, Australian government representative on Guadalcanal (even while in Japanese hands) rendered invaluable services to the Marines. These natives were all members of the Solomon Islands police force.

The First Day

Operation Order 7-42 called for a landing on Beach Red by the 5th Marines (reinforced), less 2d Battalion (employed on Tulagi). This group would land with its remaining two battalions abreast, 1st Battalion going into position on the right. The 3d Battalion was to go into position immediately on the left (southeast) flank to act as beachhead defense against enemy activity from that direction. The 1st Marines (reinforced) would land at H-hour plus 50 minutes in column of battalions, 2d Battalion leading and followed by 3d and 1st Battalions in that order. Immediately upon landing and assembling, the regiment would pass through the right section of the beach and attack almost due west on an azimuth of 260 degrees, battalions echeloned to the left rear in order of landing. Its mission was to seize Grassy Knoll and to be prepared for further advances. The

Map 4: Initial Dispositions -- Guadalcanal, 7 August 1942


1st Battalion, 5th Marines, meanwhile would advance west and secure the line of the "Tenaru" River (actually Alligator Creek.) The Support Group, consisting of Headquarters, 11th Marines, and all other division troops not employed on Tulagi, was to land on order over Beach Red.

Six floatplanes were supplied the assault forces by the heavy cruisers to Task Force 62. Three of these, from the USS Astoria, were used as liaison planes whose first duty was to mark the extremities of Beach Red with smoke so that the target area would be clearly visible to the transports and the landing craft. Three additional planes were supplied by the Quincy for use as spotting planes by the 11th Marines, the division artillery.11

The actual landing was accomplished as planned, with dispatch and without opposition. By 0938, regimental headquarters of the 5th Marines was ashore, and its first command post had opened about 100 yards inland. The 1st Marines had accomplished their part of the landing operation by about 1100, for by that time the regiment was beginning to pass through right elements of the 5th Marines.12

By 1400 the 1st Marines were well on the way to carrying out the assigned part of their mission. Message #75 (D-3 Journal, 7 August) contains the following message:

"1410 Troops getting along OK on beach river holding them back one-half mile inland."

Movement of troops from the beachhead began at this time. At 1400 the 1st Battalion, 5th Marines, had been ordered to move off toward the west to secure and occupy the line of Alligator Creek, upon which it was to take up positions for the night. The battalion, with regimental headquarters and the Weapons Company, began to move at once, while the 3d Battalion remained in position covering the southeast flank of the beachhead.13

The 1st Marines, however, realized almost at once that it would be impossible to get on the Mount Austen objective.14 The hill mass which had been so designated, and which had been visible from the ships, could not be seen from the beach. Furthermore, the regiment was passing through unexpectedly rough terrain and was learning that the banks of the Tenaru, which would like a snake back and forth across the regiment's line of advance, were steep, thickly overgrown, and difficult of passage, and that the river itself was not fordable.15

At 1430, Colonel Cates reported that he was still passing through the right front of the 5th Marines and that no contact had been made as yet with the enemy.16

At about this time the Support Group was coming ashore from the ships of the Transport Group, which had moved in to within 2,000 yards of the beach. At 1515, General Vandegrift notified Admiral Turner that his command post was opened on shore, as 45 minutes later the General himself was on the beach.17


By nightfall, the units ashore were in position as shown on the accompanying map.

Several important lessons had been learned during the first day's operation. The ship-to-shore movement had been a gratifying success. Two reinforced regiments, and all division troops, had landed on Guadalcanal; the regiments had carried out their assigned missions where it was geographically possible to do so.18 The landing was carried out with something of the ease with which a similar maneuver might have been accomplished in a peacetime drill. The fact that no enemy opposition was encountered was conducive to smoothness of operation, but even taking this into consideration, the entire D-day operation from the point of view of the Commanding General, who was bringing his troops ashore for combat, was an outstanding success.

The movement of supplies, however, from the landing craft to the beaches and ultimately to the supply dumps, was an entirely different matter. Certain phases of this operation were characterized by the same smoothness and efficiency as the troop movement. Specifically, unloading from holds to landing craft and movement to the beach proceeded according to schedule. Once the landing craft touched the beach, however, they encountered a situation which, almost as the first craft landed, began to deteriorate. Captain L.F. Reifsnider, Commander of Task Group 62.1, reported as follows:

A serious situation developed early during landing operations when the labor section of the shore party was unable to cope with the rapidity and quantity of supplies and equipment delivered at the beach. The situation is ascribed to a total lack of conception of the number of labor troops required to unload boats and move material off the beach, failure to extend the beach limits earlier in the operation, and, to some extent, lack of control of troops on and in the immediate vicinity of the beach--it was definitely understood and agreed that the unloading of the boats and the removal of material from the beach would be done by the labor section of the Pioneer Battalion. .  .  .19

By the terms of Landing Operations Doctrine, United States Navy, (FTP 167) Chapter 2, Section II, paragraph 212 (a), the unloading of the material of the Landing Force form small boats was a task to be performed by the military component of the Shore Party. Reifsnider's reference to the Pioneer Battalion is interesting in that only 310 men of that unit were present as a body and even these had been assigned an active defense role as part of Colonel del Valle's Support Group. The rest of the battalion had been parceled out to various regiments as reinforcing elements. A feeling comment on the subject of the relationships between the Shore Party Commander and the Naval [Task Force] Commander is contained in a 12 February 1949 letter to the Commandant of the Marine Corps from Colonel Frederick L. Wieseman, who commented on the matter under discussion as follows:

The initiation of general unloading by the Naval (Task Force) Commander is a matter that should be prevented by custom, doctrine, high command, Joint Chiefs of Staff Directive, and force of arms if necessary. .  .  . Nothing should be landed on a beach unless the Shore Party Commander has OK'ed its dispatch.

Reifsnider's complaint, however, justified in terms of the undeniable fact that supplies were piling up on every beach, is nevertheless an example of the somewhat remote viewpoint of the observer afloat who may be conversant with only his own part of the picture. What had happened in fact was that General Vandegrift, on the basis of what intelligence he had, was anticipating momentarily commencement of a meeting engagement of major proportions, at some point short of his objective, the airfield. As a result, although the Division operation order had provided that the Shore Party might call upon unit


commanders for additional labor, no units as such were being held on the beaches, but were in or enroute to assembly areas preparatory to the anticipated battle. Consequently they could not be made available for labor purposes, however important this might seem. Captain Reifsnider's "hundreds of Marines" (as reported by seamen and observed from on board the Hunter Liggett) were largely the rear elements of units already engaged in the slow advance through the jungle toward the field.

At 1321, Admiral Turner had directed Captain Reifsnider to expedite unloading.20 By 1449 the advance message center was advising General Vandegrift that the shore party commander was badly in need of at least 500 men to help unload boats.21 It said further that no men were available on the beach for that function. Less than an hour later, Reifsnider informed General Vandegrift that 15 men from each cargo ship were being detailed to help unload boats until they could be relieved by pioneers.22

This measure was in the nature of makeshift, and the constant stream of material arriving at the beach could not be handled. In spite of the fact that every effort was made, at one time 100 loaded boats had been beached while 50 more were standing by just outside the breakers waiting an opportunity to land. So chaotic had conditions become that, by 2330, the commander of the Shore Party reported to General Vandegrift, that unloading was entirely out of hand;23 that supplies were arriving much faster than they could be handled; and that it was imperative that the ships cease unloading until the beach could be cleared. Three hours later Admiral Turner advised General Vandegrift that his24 request was being carried out.

Enemy activity during the day had been confined to air strikes launched from Rabaul. At about 1100, a coastwatcher stationed in the Upper Solomons reported that a force of 18 bombers had just passed his station, headed for Guadalcanal. At 1320 this force appeared, striking ineffectually at ships in the transport area off Beach Red.25 The destroyer Mugford was hit by a 250-pound bomb and suffered 20 casualties, but no other vessel in the group was touched. The attacking planes were reported to be twin engine Type 97s. Two were shot down by antiaircraft fire from the ships.

A second attack came at about 1500, when the same area was struck by ten 99 Aichi dive bombers.26 No damage resulted from this strike, and again two enemy planes were brought down by antiaircraft fire. In neither attack was any attempt made by the Japanese to damage material on the beach.

During the first night the only disturbance which occurred resulted from the nervousness of some troops who lacked combat experience and who were in unfamiliar surroundings without definite knowledge of the size or whereabouts of the enemy forces to their front. No Japanese activity was noted, but there was considerable firing in the area to the west and northwest of the beachhead, in the zones occupied by the 1st and the 5th Marines. Both regiments suffered casualties as a result.27

General Vandegrift's Change of Plan

At 2200 on 7 August the two regiments were informed of a change in plans.28 As we have seen, the objective of the 1st Marines was too far removed to be possible of attainment. With less than a division of troops--for the reserve regiment, which would have brought the infantry strength up to normal,


THE TRANSPORT, GEORGE F. ELLIOTT, was the first to be sunk in the Guadalcanal campaign, on 8 August 1942. Before U.S. torpedoes administered the coup de grace, the burning Elliott silhouetted other U.S. men of war for the Savo Island disaster.

was committed on Gavutu-Tanambogo, and one battalion of the 5th Marines was busy on Tulagi--the great hill mass of Mount Austen could not be secured. Its assault would have required the separation of a major unit from the main body, and in such a case liaison (where possible) would have been difficult in the extreme. The logistics of the movement would have been appallingly difficult.

The plans for 8 August, therefore, called for occupation of the airfield and establishment of a defensive line along the Lunga River. Those elements of the division which had been assigned the mission of establishing and maintaining defensive positions along the eastern and southeastern boundaries of Beach Red were ordered to continue that mission.

At 0930, 8 August, the 1st Battalion, 5th Marines, supported by Company A, 1st Tank Battalion, crossed Alligator Creek at its mouth and advanced toward the Lunga with its right flank on the beach. The advance was cautious and deliberate. In the meanwhile, the 1st Marines, who had spent the night along the Tenaru River south of the positions occupied by the 5th, began their advance to the west. The 1st Battalion, 1st Marines, was somewhat hindered in its advance by the fact that it had to cross the creek some 500 yards inland from the mouth. The 2d and 3d Battalions, however, skirted the headwaters and plunged immediately into the thick jungle growth which lies to the south of the airfield. Contact between units was faulty due to their failure to patrol to the front and flanks.29

By nightfall the regiment had attained only part of its objective. The 1st Battalion had overrun the airfield and had succeeded in reaching the Lunga; the 2d and 3d Battalions, however, slowed down by the difficult terrain to a rate of approximately 500 yards per hour, went into bivouac south of the airfield.

The advance of the 5th Marines had been without incident of a serious nature. In reaching their objective they had traversed the area which had been occupied by the main Japanese force. First enemy resistance, from scattered individuals had been met and several prisoners had been taken. From them, as well as from captured documents, it was learned that no serious resistance was to be expected for the time being. The negative nature of all information from front line units verified this. At 1430, therefore, the regiment was ordered to contract its front, to cross the Lunga by the main bridge immediately north of the airfield, where the


main road cuts inland to avoid the several branches of the river near the beach, and to advance in the direction of Kukum. This movement was accomplished with Weapons Company in advance, and at 1500 Marines reached the main Japanese encampment.30

It was immediately obvious that the enemy force was smaller than had been thought and the fact was confirmed that the Japanese had retreated in a state bordering on complete demoralization. Large quantities of food, ammunition and equipment were found in perfect condition, and much engineering materiel, electrical apparatus, and radio equipment was captured intact. There had been no attempt at demolition by the enemy. There was some destruction by Marines initially because of lack of proper indoctrination, but this condition was corrected almost at once. The only damage apparently had been caused by the bombardment preceding the landing, and the completeness of the enemy installations are well described in the final action report as follows:31

The extent to which the enemy had been able to develop their Lunga Point positions was remarkable in view of the short time of occupation. Since July 4th they had succeeded in constructing large semi-permanent camps, finger wharves, bridges, machine shops, two large radio stations, ice plants, two large and permanent power plants, and elaborate air compressor plant for torpedoes, and a nearly complete airdrome with hangars, blast pens, and a 3600 foot runway.

The most serious enemy activity during the day had consisted of a repetition of the retaliatory measures of 7 August. Soon after 1100, warning was received from Pilot Officer Cecil John Mason, RAAF, a coastwatcher stationed on Bougainville, that a large number of planes had been sighted over his post, flying southeast.32 At noon they appeared over the area to find that all ships of the task force, alerted by the warning, were at maximum speed.

Approximately 40 twin engined torpedo planes participated in the attack. Two ships were lost. The destroyer Jarvis, hit forward by a torpedo, managed to leave the area under her own power but disappeared with all hands while making her way to the southeast. The transport Elliott, set afire when an enemy plane crashed on board, had to be beached and destroyed by her sister ships. Survivors were taken aboard the Hunter Liggett.

Twelve of the attacking planes were shot down in the vicinity of the ships by antiaircraft fire from the ships and by fighter planes from Admiral Noyes' carriers operating to the south. Two more were accounted for by shore-based antiaircraft weapons. Still others were shot down by carrier-based fighters to the west of the transport area. A total of seven American planes were lost.33

The Naval Withdrawal

The burning of the Elliott had two adverse consequences, entirely apart from the loss of the ships herself. Included in the supplies aboard her had been a good shore of the material of the 2d Battalion, 1st Marines, and that was lost.34 The second, and more serious, consequence was the fact that the glare caused by her burning allowed enemy observers in the neighborhood of Tassafaronga to see the cruisers and destroyers which were shortly to be attacked on that night of 8-9 August, and to report their presence to the advancing enemy task force.

In the evening of 8 August, General Vandegrift


was called to a conference aboard the USS McCawley, flagship of Task Force 62. While there he was told that Admiral Turner had decided to remove all transports and cargo vessels from the area at 0600 next morning, 9 August. The reason given for this decision was the fact that advice had come from Admiral Fletcher, Commander, Task Force 61, telling of a shortage of fuel and of the loss of 21 of his 99 planes, and of his consequent decision to withdraw.35

This posed a new and most alarming problem for General Vandegrift and his staff. Plans made by the division had been formulated on the assumption that the ships would remain for four days in the target area so that all supplies could be put ashore. however, even with the removal of all supplies to the beach, the division would have been in a somewhat precarious position, for the shortage of shipping and the unforeseen demand for haste had made necessary a cut below the basic allowances ordinarily prescribed. The unloading process, as we have seen, had been complicated by a condition approaching chaos on the beach, and the movement from ship to shore had been stopped as a result. The withdrawal of the supply ships, therefore, was, from a troop standpoint, little short of a catastrophe, but Admiral Turner's decision was not changed.

Shortly after midnight of 8-9 August, moreover, friendly surface forces operating in the Solomons area suffered a sudden and overwhelming defeat. The events leading up to the disastrous Battle of Savo Island are interesting.

There can be no doubt at this time that the American attack on Tulagi and on Guadalcanal came as a surprise to the enemy at Rabaul as well as to the smaller forces in the target area.

The American convoy had been sighted as it approached the area by an enemy lookout in the vicinity of Cape Esperance. There appears to have been a breakdown in communications between his post and Tulagi, for his warning did not alert the people on the latter island.36 The attack, moreover, cut the area off from communication with the enemy rear areas (the radio installations on Tulagi, Gavutu, and Tanambogo had been destroyed by the prelanding bombardment by the San Juan and the two destroyers which accompanied her). Captain Miyazaki, of the Imperial Japanese Navy, who was on duty at Rabaul at the time, was questioned on November, 1945, as a prisoner of war. He said, in speaking of the events of 7 August 1942, "Early in the day we lost communication with Gavutu, so did not know what happened."37


VICTIM OF SAVO ISLAND, the heavy cruiser Astoria burned throughout the morning of 9 August 1942, until, engulfed by uncontrollable flames, she sank.

Communication must have been reestablished quickly, however, or else the enemy must have been able to deduce, from the silence that had fallen over its forces in the Tulagi area, that an attack had been mounted. By afternoon of 7 August a naval task force was being assembled from units in Kavieng and Rabaul.38 It was formed from elements of the 8th Fleet, and consisted of five heavy cruisers--Chokai (flying the flag of Rear Admiral Gunichi Mikawa, CinC, 8th Fleet), Kako, Furutaka, Aoba (Rear Admiral Goto), Kinugasa, the light cruiser Tenryu (Rear Admiral Matsuyama), and Yubari--with one old destroyer Yunagi from the 4th Destroyer Division. Rendezvous was effected northwest of Bougainville, and the force came down the stretch of water which lies between the parallel chains of islands of the group and which was later to become known as the "Slot".

This force was sighted at 1130 on the morning of 8 August by a U.S. observation plane which maintained contact with it for about an hour.39 The results of the observation were reported at once, but through some mixup in the communication chain which as never been satisfactorily explained, the screening force of United States and Australian ships apparently was not apprised of the potential danger which the enemy task force presented.40

The screening force, divided in two groups, was patrolling the approaches to the transport area on each side of Savo Island when, at about 0130 of the morning of 9 August, it was attacked and overwhelmingly defeated by an enemy force which immediately retired from the area. No attempt was made by the Japanese to pursue the advantage which had been gained, and the transport area was left unmolested. The attack had been preceded by the dropping of flares from Japanese cruiser-based planes, and information subsequently got from prisoners indicates that the attacking force was aided by observation from Cape Esperance made possible by the illumination from the flares and from the burning transport, Elliott.41


The results of the attack were little short of catastrophic for the Allied forces. Of the five cruisers on station at the time, four were sunk and the other badly damaged. Chicago sustained damage, while Astoria, Vincennes, Canberra, and Quincy sank during the night and the early morning.42

Post-war interrogation of Japanese prisoners answered a question which arose immediately after the Japanese withdrawal--why had the attacking force refrained from annihilating the then defenseless transports? It appears that one 8-inch round fired from the second group to be attacked--the Northern Group--penetrated the operations room of Chokai, destroying all equipment and charts.43 This together with the fact that there was some delay in resuming proper formation, impelled Vice Admiral Mikawa to withdraw rather than run the risk of being overtaken by planes during a later withdrawal.

A belated vengeance overtook another ship of the force when Kako, about to enter the harbor of Kavieng the next morning, was sunk off Simberi Island by an American submarine, the old S-44.44

Withdrawal of Task Force 62

There is no evidence to show that the proximity of the enemy task force had any bearing upon the decision of the Commander of Task Force 62 to retire, although the position, speed, and direction of the enemy group were known to him prior to his conference with General Vandegrift. The exact composition of the force was not known, and the enemy's intentions were not accurately estimated.45

According to the War Diary of the Commander, Task Force 62, the following troops were left in the Guadalcanal-Tulagi area when the transports and supply ships withdrew:

The 2d Marines, under Colonel John M. Arthur, had formed the Division Reserve, and had, as a part of its ultimate mission, the occupation of Ndeni, in the Santa Cruz Islands.

As we have seen all three battalions of the regiment were committed. This was done by orders from Rupertus direct to the battalion


commander in each case and not through the normal chain of command.46 The regimental headquarters remained afloat, however, as did all administrative personnel and all records of the regiment and its battalions. In addition to these, working parties from all companies, as well as most of the Headquarters and Service Company, Regimental Weapons Company, and G and Headquarters and Service Batteries, 3d Battalion, 10th Marines, remained aboard ships.

When the transports left the area on 9 August, this group (which totaled 1390 officers and men) were taken, in Transport Divisions 2 and 12, to Espiritu Santo, where, according to the War Diary of Commander, Task Force 62, they were used to "reinforce the garrison there." On 14 August, Colonel Arthur was ordered by Admiral Turner to report to the Commanding General, Espiritu Santo, for duty. He did so, but within a few days he was ordered to return to Tulagi with a small number of officers and men of his command.47

A good-sized complement of Marines was left on Espiritu Santo, however, for elements of the 2d Marines were subsequently ordered by Admiral Turner to form a "2d Provisional Raider Battalion", to consist of a small headquarters and six rifle companies. This was done, but the unit was disbanded on 22 September, two days after arrival of the authentic 2d Raider Battalion, under Lieutenant Colonel Evans F. Carlson. Eight hundred men of the 2d Marines and Battery G, 10th Marines, were then ordered to return to Tulagi. They arrived 9 October with 358 men of the 6th Naval Construction Battalion.

The forming of the "2d Provisional Raider Battalion" gave rise to a considerable correspondence by Admirals Turner, Ghormley, and Nimitz, and the Commandant of the Marine Corps, General Holcomb.

It will be recalled that General Vandegrift was operating under immediate command of Admiral Turner, Commander Amphibious Force, South Pacific. In a letter written to ComSoPac on 29 August, Turner said in part:

6. In many circumstances in the future amphibious warfare in the South Pacific, it is believed that a Marine Regiment, or a part of a Marine Regiment, or two Marine Regiments, will be the size of a force appropriate for offensive and defensive amphibious operations. The employment of a division seems less likely. .  .  . The problem of mopping up outlying detachments will exist throughout the campaign. For this reason the Commander, Amphibious Force, South Pacific, has reached the conclusion the Marine regiment (sic) will not be an entirely suitable combat unit for operations in the South Pacific unless it has, as an integral part of its organization, either a Raider or a Parachute Battalion. The previous concept that Raider and Parachute Battalions are always division or corps troops is no longer agreed to.

7. In view of foregoing, and in order to prosecute promptly the operations required by prospective tactical situations, the Commander Amphibious Force, South Pacific, will, unless directed to the contrary, proceed with the organization of Provisional Raider Battalions in the Second, Seventh and Eight Marines, and give these already trained troops such additional specialized training as seems appropriate. Furthermore, he recommends that Marine Corps Headquarters issue directions for the permanent organization of Raider Battalions as integral units of all Marine regiments now attached to, or ultimately destined for, the Amphibious Force, South Pacific. It is not recommended that the total personnel strength of the regiments be increased.

Vice Admiral Ghormley forwarded the letter to Commandant Marine Corps on 6 September, via Commander in Chief, U.S. Pacific Fleet, with his own reactions to Turner's suggestions attached in the form of a first endorsement:

The organization of Raider Battalions from the organic troops of the 7th and 8th Marines should be withheld pending a declaration of policy by the Commandant, Marine Corps. .  .  .

A copy of this endorsement was sent to Turner.

Admiral Nimitz, in turn, added a second endorsement, also directed to General Holcomb, on 24 September:

The Commander in Chief, U.S. Pacific Fleet is of the opinion that Raider Battalions are specialized troops and should be reserved for appropriate tasks--and that extemporized organization of Marine Forces should be made only in case of dire necessity.


This correspondence elicited the following reply on 3 October from General Holcomb, in the form of a letter to Nimitz:

The Commandant noted with much concern the order from the Commander, Amphibious Force, South Pacific Force, to organize a raider battalion, on a temporary basis from the units of the 2d Marines, less its battalions. He made no comment at the time for reason that at this distance no one could form an opinion as to the necessity of carrying out this most unusual procedure. The objections thereto are clearly set forth by the Commander-in-Chief, U.S. Pacific Fleet, and appear to need no further comment. .  .  . It is noted with regret that Admiral Turner's letter of August 29th does not contain the views of the Commanding General, 1st Marine Division, in a matter in which he is particularly qualified and concerned.


Table of Contents
Previous Chapter (2) ** Next Chapter (4)


1. Action Report, Task Group 62.1, p. 3.

2. D-3 Journal, Initial entry, 7 August 1942.

3. Ibid., Messages 2 and 4.

4. Ibid., Message 13.

5. Action Report, USS Saratoga.

6. D-3 Journal, 7 August 1942. Messages 14, 15, and 16.

7. Reference is directed to The Coastwatchers, by Commander Eric Feldt, RANR.

8. An estimate of the situation, prepared by the division while in Wellington, on 11 July 1942, suggested a total of 5,000 including one infantry regiment 3,360), one engineer battalion (375), and miscellaneous combat and labor personnel (1,265). A revised estimate, dated 27 July, gives the same total but a slightly different breakdown. Both documents are contained in folder 3K, Guadalcanal Documents, Marine Corps archives. No basis can be found in documents for the estimate of 2,100 laborers which is contained in the Final Report, Phase II, p. 9.

9. It is interesting to note that even Japanese sources are ambiguous on the point of the actual number present on the island on 7 August 1942. Wording of translated Japanese documents is such as to justify a round figure of perhaps 2,000, of which the greater part were laborers.

10. Final Report, Phase II, p. 9.

11. Final Report, Phase I, Annex G, p. 2. This use of planes caused much adverse criticism both on the part of the pilots and the Marines. It was pointed out by them that such tactics, if employed against a defended beach, would result in the immediate loss of all planes, for the marking process made necessary a series of no less that eight runs at very low altitude. Ibid., Phase II, p. 15.

12. Final Report, Phase II, Annex L (Record of Events, 5th Marines).

13. Final Report, Phase II, Annex L, p. 1.

14. At about noon, the regimental executive officer and the operations officer requested Colonel Cates to have the objective re-examined in view of the unexpected roughness of the terrain and distance of Mt. Austen. Cates did so that night at a conference with Vandegrift. Interview, Colonel William W. Stickney, 15 January 1949.

15. Here also the amphibian tractor's versatility was demonstrated, on its first day of trial in action. Warned by Pilot Officer Widdy, during the rehearsals at Fiji, that the rivers and lagoons would be deep, members of the Engineer Battalion rigged a rough bridge of dunnage and timber (irregularly acquired on board ship) and sent it ashore on board two tractors. The two vehicles were driven into the Ilu backwash, a small stream flowing into the Tenaru at its mouth from the southeast, and formed pontoons across which the jury-rigged bridge was thrown. Final Report, Phase II, Engineer Annex.

16. Final Report, Phase II, p. 10. See also Annex M to that document and History of the First Marines, MCR.

17. D-3 Journal, messages 85 and 91. MCR.

18. Final Report, Phase II, p. 15. See also introductory remarks, same document, p. 1. MCR.

19. Action Report, Task Group 62.1, p. 9. Provision for handling of supplies was set up in Force General Order No. 7-42<, 25 February 1942. This in turn formed the basis for the supply section of 1st Marine Division Administrative Order 2a-42, which accompanied Operation Order 7-42. Paragraph 6 (a) of the Administrative Order directed Shore Party Commanders to call upon troop commanders in their immediate vicinity for assistance in the handling of supplies from landing beaches to dumps. It nowhere directed that a troop commander upon whom the request was made would necessarily comply.

20. D-3 Journal, Message 70, 7 August 1942.

21. Ibid., Message 82.

22. Ibid., Message 89.

23. D-3 Journal, Message 104, 7 August 1942.

24. Ibid., Message 105. )This is entered in a transcribed version of the Journal as having gone out on 7 August, whereas it actually bears the time notation of 0242 8 August.)

25. Report of Action, Commander Task Group 62.1, p. 4.

26. Preliminary Report, CinCPac, pp. 2 and 3.

27. Interview, Captain John J. Jachym, November 1944.

28. Final Report, Phase II, p. 11. (The narrative says only that orders were issued during the night of 7-8 August. Annex M, p. 1, puts the time of the change of orders at 2000, 7 August.)

29. Final Report, Phase II, pp. 10 and 11. Annex M, p. 1. Annex L, p. 1.

30. Final Report, Phase II, Annex L, p. 11.

31. Final Report, Phase II, p. 12. In expansion of this data, LtCol Robert G. Ballance, USMC, wrote in 1949: Item "an elaborate air compressor plant" was actually a liquefaction plant for the manufacture of oxygen by compressing it to liquefaction and then "fractionally" distilling it. This plant would have provided compressed air for torpedoes as well as aviation, medicinal, and welding and cutting oxygen. The wanton destruction of this plant, then 70 to 75% complete, with remaining components on hand ready for installation was truly deplorable, for eventually just such a plant had to be hauled out from the United States and erected on the island. (Letter, Ballance to CMC, 12 February 1949.)

32. Preliminary Report Solomon Islands Operation. CinCPac, 23 August 1942. "The warning from this coastwatcher was of the greatest value."
[NOTE: According to The Coast Watchers, Cecil John Mason was killed during a mission to New Ireland in early July. Paul Edward Mason made this report from Bougainville. --HyperWar]

33. Ibid., p. 4.

34. Ibid., p. 13.

35. It has been widely believed that Admiral Fletcher's decision was reached on the spot late in the day of 8 August. This is not the case. A dispatch sent to Fletcher by Vice Admiral Ghormley on 2 August (after Callaghan had reported the gist of the Fiji conference) shows that even prior to that date Fletcher had planned to withdraw his task force, and therefore also the transports, before D-day plus three. Ghormley, not protesting, thereby gave tacit approval to Fletcher's plan to withdraw. War Diary, ComSoPac, 2 August. The division commander had no information that such a withdrawal was planned, the first indication of its coming to him as described above. Operation Plan 1-42 dated 16 July 1942, ComSoPac, had directed that all vessels fuel to capacity during the rehearsal at Fiji. Tankers used were the Platte and the Cimmaron each with 12,000 tons of fuel oil and 500,000 gallons of aviation gasoline, and the Kaskaskia, with 13,000 tons of oil. This order had been carried out at the conclusion of the exercises, immediately prior to departure fro the objective. War Diary, ComSoPac, June, 1942, contains the text of the plan.

When Fletcher, the man on the spot, informed me he had to withdraw for fuel, I approved. he knew his situation in detail; I did not." Ghormley ms. p. 93. According to data secured from ships' logs, every ship in Task Force 61 had on hand enough fuel when the log was closed on 8 August to last for at least four days. The carriers could have operated for 17 days, North Carolina for 18 days, the cruisers for about 11 days, and the destroyers, most voracious of the lot, for about seven days. These figures are based on the average daily consumption of each individual ship taken over the period of 1 to 8 August, 1942, and take into account the fact that two "short-legged" destroyers--Benham and Grayson--had fuel enough for three and two days respectively. Logs may be found in the National Archives.

36. Japanese Campaign in the Guadalcanal Area, Headquarters, USAFISPA. (Hereinafter referred to as the USAFISPA Report), p. 3.

37. Interrogations of Japanese Officials. USSBS (Pacific), Volume II, p. 419.

38. USAFISPA Report, p. 4.

39. Campaigns of the Pacific War, USSBS (Hereinafter referred to as Campaigns, USSBS), p. 106.

40. Preliminary Report Solomon Islands Operation, CinCPac, p. 11, paragraphs 45 and 46. "Why no enemy report from anyone?" From letter Rear Admiral Crutchley to Admiral Turner, 10 August 1942. Admiral Crutchley was in command of the screening force.

41. Campaigns, USSBS, p. 106.

42. Included among Marine casualties aboard the three American ships were two pilots of VMO-251, who went down with the Vincennes. Three others, aboard the Astoria, were rescued.

43. Interrogations, USSBS, Volume II, p. 362.

44. These details are taken, with the author's permission, from an exhaustive manuscript by Walton L. Robinson. Hereinafter referred to as Robinson ms.

45. Preliminary Report Solomon Islands Operation, CinCPac, p. 6, paragraph 23.

46. Interview, Colonel R.E. Hill.

47. War Diary CTF 62. September and October, 1942.

Compiled and formatted by Patrick Clancey, HyperWar Foundation