Chapter 8: The War Against the Sub Pens
SUBMARINES became the primary concern of the Eighth Air Force after 20 October 1942 and continued to preoccupy that organization until June 1943. In the fall of 1942, however, it was not at all clear whether striking the submarine operating bases on the coast of France, as the directive of 20 October stipulated, was an efficient method of reducing the submarine menace; nor was it clear that the day bombers could do that job effectively. The entire antisubmarine campaign constituted, in fact, a highly controversial problem, and one in which the essential data became too often obscured by the mysterious activities of that most mysterious of the enemy services. On the basis of information no doubt unavoidably insufficient, the Eighth Air Force became committed to a protracted campaign against the submarine operating bases on the French coast, which campaign, though unquestionably inconvenient and harassing to the enemy, proved on final analysis to have had no appreciable effect on the rate of operations.*
To those who had to cope with the steadily increasing submarine threat, several alternative courses of action suggested themselves, no one of which promised by itself to be entirely satisfactory. It would have been very natural for strategic bombing forces to have concentrated their efforts on the sources of the submarine fleet, as they planned to concentrate on the sources of the entire German war machine. The submarine construction yards and the component parts manufacturing plants provided tempting objectives, the complete destruction of which would eventually solve the problem. The RAF had
* See below, pp. 251-54.
already expended a not inconsiderable and sustained effort in this direction. Although few and light in the fall of 1942, British Bomber Command attacks during the fifteen months from April 1941 to June 1942 had damaged the ports of Rostock, Lubeck, and Emden and had dealt heavy blows to facilities at Bremen, Hamburg, Wilhelmshaven, Kiel, and Bremerhaven. In addition, the submarine Diesel factory at Augsburg and the component parts factories in Cologne had suffered in the attacks on those cities.1
The British effort had, however, been directed primarily against the towns themselves rather than against the port facilities and factories, in accordance with the RAF policy of area bombing. It was the opinion of the Ministry of Economic Warfare in July of 1942 that, apart from damage to the plant at Augsburg which was supposed to be producing up to 50 per cent of the total submarine Diesel engine requirements, little severe damage had been inflicted on component factories. In that instance, it estimated, probably one month's output had been lost, amounting to the Diesel requirements for ten submarines. As for the construction yards, repeated attacks on Wilhelmshaven, Kiel, Hamburg, and Emden had resulted in no detectable decrease in production, although the estimated schedule appeared to have been delayed by a few weeks as a result of a variety of factors, not all of which could be identified with the bombing offensive. This same agency further contended that these objectives were not well suited to aerial bombardment. Component parts plants were numerous, widely scattered, often inaccessible from the United Kingdom, hard to identify, and of a type difficult to destroy except by attacks of "exceptional weight and concentration." Moreover, it was reported that a surplus of suitable productive capacity existed. The shipyards presented targets too small, too isolated from other suitable objectives, and of a type not easily enough put permanently out of action to warrant a major share of the bombing effort. On the other hand, their proximity to the British air bases made them always useful secondary objectives.2
The increased accuracy possible with precision bombing by day promised greater effectiveness in attacks on targets of this nature. But even so, there was little hope of an immediate effect on submarine operations. It was estimated in August 1942 that the submarine fleet consisted of some 240 operational craft, with 120 training in the Baltic. Production at that date was believed to be in the neighborhood of twenty per month, ten to fifteen a month becoming operational; and
sinkings by Allied agencies were currently at the rate of from five to seven a month. It appeared, therefore, that no amount of damage done to the submarine construction yards and factories could reduce the operating fleet during the ensuing nine months; indeed, it was anticipated that accessions from the force in training would add to the fleet eight to ten s each month during that period.3 Bombing attacks on production facilities could have only a long-term effect on operations, and in the fall of 1942 the Allies were in no position to wait until the fleet perished from attrition.
With plans for the opening of an African campaign in November, the element of time became of the most urgent importance. If the Allies were effectively to supply the United Kingdom, the Middle East, and North Africa, it was clear that something drastic would have to be done for the restriction of enemy submarine operations,4 and one of the more favorable opportunities seemed to be offered by the enemy's operating bases on the western coast of France. The Germans had begun, immediately after the defeat of France, to develop facilities at Brest, Lorient, St. Nazaire, La Pallice, and Bordeaux in order to place the submarines as close as possible to Atlantic supply lines and as far as possible from British airfields. They had constructed elaborate pens to house and protect these craft during their stay in port and had built extensive repair and servicing facilities. Elaborate also was the schedule of turnaround by means of which a limited number of pens could be made to accommodate a large and growing fleet of submarines.5
Since the middle of 1942 the RAF Coastal Command had concentrated a considerable part of its antisubmarine forces in patrols over the Bay of Biscay. Practically all units of the Atlantic submarine fleet had to pass through the Bay of Biscay on the way to and from their French bases, and there thus existed a constantly high concentration of submarines in the bay and its approaches. By covering this transit area with long-range aerial patrols, Coastal Command hoped either to destroy a significant number of submarines by direct attack or, by forcing them to remain submerged for long periods, to reduce substantially their effective time in the open sea. As yet, the effort suffered from lack of enough long-range aircraft, lack of a "balanced" antisubmarine force capable of attacking both by day and night, and lack of adequate radar equipment and special weapons. Actual "kills" had been relatively few,6 but great hopes had been placed in an increased and improved Bay of Biscay offensive.
A logical development of that offensive was the employment of Eighth Air Force bombers against the enemy's operating bases. It was considered practically impossible to penetrate with any bombs then available the dozen or more feet of reinforced concrete that formed the roof of the U-boat pens.7 But it was believed that the facilities at these bases were so integrated, and the time schedule for repair and refitting so carefully adjusted, that any damage to the installations surrounding the pens would cause serious delay in turn-around and so, in effect, reduce the number of submarines in operation. Locks, floating docks, storage depots, railway yards, powerhouses, foundries, barracks, and submarines not actually in the pens all appeared to present vulnerable targets for bombing aircraft--especially for bombers equipped for precision operations.8 It was considered very probable that much of the servicing had been put under concrete along with the submarines themselves and that alternative power installations existed which could be used to relieve most emergencies affecting the power system. It was certain that the bases would be given powerful antiaircraft protection.9 Yet the prospect of disorganizing the U-boat campaign by harassing attacks on vital points, and eventually of neutralizing them, seemed reasonably bright.10 General Eaker in October expressed confidence that, given ten heavy bombardment groups, the VIII Bomber Command could effectively deny to the enemy the use of five Biscay bases.11 The Air Ministry in August had declared itself in favor of operations against the U-boats at sea and against their operating bases, in preference to the long-term policy of attacks against building yards and factories.12
Opinion in Washington was divided on the use of long-range, landbased aircraft in antisubmarine operations. The U. S. Navy favored extended convoy cover as the most effective use of this type of plane, but it also had advocated air attack on the operating bases as a helpful auxiliary measure.* Those actively identified with the AAF Antisubmarine Command argued for employing as many B-24's as possible on such projects as that already being conducted by the RAF Coastal Command in the Bay of Biscay. But Brig. Gen. C. W. Russell, AAF coordinator for antisubmarine activity, on 3 November placed primary emphasis on attacks against the operating bases and construction yards by heavy bombers of the Eighth Air Force, a policy which AC/AS, Plans indorsed.13
* See below, chap. 12.
The choice between operating bases and construction yards as targets for Eighth Air Force operations was for the time being simple enough. A campaign against the French bases was especially well suited to the capabilities and limitations of the American bomber force. Not only were the targets much better adapted to daylight precision methods than to those of the RAF night bombers, they were also within the area of occupied France to which Eighth Air Force operations had been temporarily restricted. Accordingly, General Spaatz pledged the maximum use of his force against the U-boat bases. At the same time, he made available to Coastal Command twelve B-24's to help cover the movement of shipping to Africa by expanding the system of long range air patrols over the sea lanes,14
The German Submarine Bases
The VIII Bomber Command flew its first mission against the submarine bases on 21 October, when it dispatched ninety bombers (sixty-six B-17's and twenty-four B-24's) to attack the enemy base at Lorient-Keroman. The objective was a small fishing port, situated about one and one-half miles southwest of Lorient on the Brest peninsula, which the Germans had developed as a major submarine base. Principal targets were the U-boat shelters: twelve completed ones and a block of seven pens then under construction. Typical of their kind, these shelters had been built on dry land, then connected with the harbor by channels, and provided with heavily reinforced concrete roofs. Immediately adjacent to the pens stood lighter and smaller buildings believed at that time to contain workshops, transformers, oil storage, offices, and other installations directly connected with the servicing of U-boats. Lorient had not been attacked by the RAF during 1942, nor had the British ever attacked the area of the submarine pens. In 1941 they had made thirty-three night raids, dropping 396.1 tons of bombs, mainly on the town itself.15
Though bad weather on 21 October forced all but fifteen B-17's of the experienced 97th Group to turn back before reaching the target, the bombing was unusually good. From a 17,500-foot altitude--a considerable departure from the 22,000-to 27,000-foot level usually reached--the bombers dropped thirty high-explosive bombs, each weighing one ton. With the exception of a few which fell some 1,100 yards from the pens, most of the bombs fell in the immediate target area. Of the thirty dropped, twenty-one fell within a radius of 1,000
feet from the aiming point.16 Five bombs were reported by ground observers to have hit the central block of shelters; but, according to a French underground informant, they did not penetrate more than five feet despite their weight. Among the surrounding buildings, the results were somewhat better. Three general workshops and a pair of floating docks were pretty thoroughly destroyed and two submarines were damaged by blast. Of the 150 reported killed, more than half were German workmen and about 40 were French.17
Although little major damage was done to the base itself, the bombing made a great impression on both French and German opinion. For once, the French people appear to have compared an attack by U. S. forces favorably with those made by the British. They seem to have been greatly pleased with the whole affair, standing in the streets, watching and smiling and applauding the accuracy with which the Americans dropped bombs on the German installations. It was, they felt, too bad that Frenchmen had also to be killed, but the victims had asked for their fate in accepting employment at the base for the sake of the high wages paid there. As for the Germans, they appear to have been taken completely by surprise. The alarm was not sounded, and the bombs had fallen before the antiaircraft guns went into action. The Germans were said to have been convinced that a formation of such size--fifteen aircraft--could only have been their own planes. The mission temporarily discredited the Quislings, who had insisted that Allied attacks were being made deliberately against the civilian French population and that the base was too well defended to be attacked. The controlled press remained silent.18
Despite the fact that the defenses at Lorient were caught napping and although the attacking force encountered no effective flak, they did run into stiff resistance from enemy fighters. As the formation crossed the enemy coast en route to the target, it met thirty-six Fw-190's which gave it continuous battle to a point not far from the objective. The bombers acquitted themselves well but lost three of their number.19
With this mission and these losses in mind, General Spaatz wrote in pessimistic vein to General Arnold on 31 October: "Whether or not these operations will prove too costly for the results obtained remains to be seen. The concrete submarine pens are hard, maybe impossible nuts to crack." "However," he added, "the bombing of the surrounding installations should seriously handicap the effective use of the
bases."20 General Spaatz had, in fact, undertaken this task with more determination than relish or optimism. It was not only a regrettable, if necessary, diversion of effort from the main mission of his force; it was also a job that would probably require the use of tactics very different from those for which his units had been trained. As early as 15 September 1942 he had expressed concern over this problem. Assuming that the pens themselves would be virtually impervious to normal high-altitude bombing and that they constituted the vital spot in the base installations, he predicted that if the bases were to be put out of operation some tactics in addition to ordinary high-level bombing would have to be used.21
By the end of October, General Spaatz was ready to operate against the submarine bases from lower altitudes. Evidently convinced that bombing from above the 20,000-foot level, as practiced heretofore, was not likely to yield accurate enough results to neutralize small targets, he planned to operate at altitudes possibly as low as 4,000 feet. In that event, he warned, much higher casualties than any so far sustained would have to be faced, for the objectives would certainly be heavily defended by antiaircraft. Other factors, he believed, would also lead toward a higher casualty rate. Low altitudes would favor enemy fighters. Since the French bases were beyond the range of available fighter escort (no P-38's or P-47's were available), the bombers would be without fighter support over the objective. Finally, the crews left after the assignment of the 97th and 301st Groups to the Twelfth Air Force were by no means seasoned, especially the gunners.22
On 9 November the VIII Bomber Command flew a mission at very much reduced altitude against the submarine installations at St. Nazaire. If it had been seriously expected that attacks at lower altitudes would increase effectiveness without at the same time producing prohibitive losses, those hopes were effectively dampened by the experiment.23 Thanks to a well-planned course and a large diversionary mission flown by the RAF, the fighter threat, heretofore the more serious, was circumvented, but the same could not be said of the antiaircraft batteries concentrated in the neighborhood of St. Nazaire.24 The twelve attacking B-24's, flying at 17,500 to 18,300 feet, suffered little, but the thirty-one B-17's, flying at 7,500 to 10,000 feet, fared badly. In the neighborhood of St. Nazaire they ran into very intense flak, extremely accurate both in altitude and deflection; at 10,000 feet both light and heavy fire was reported, of considerable intensity and accuracy. As a result of this
Battle Casualties, Eighth Air Force
Interrogations, 381st Group, Summer, 1943
barrage, three aircraft were lost and twenty-two others were damaged in some degree.25
It was a costly experiment. Flak hitherto encountered at higher altitudes had been relatively ineffective, and it was evident that the cost of low-altitude bombing could be justified only by appreciably improved accuracy. Only some 75 of the 344 bombs dropped could be plotted from strike and reconnaissance photographs, but of these, no more than 8 burst within 600 feet of either of the two aiming points--the shops of Chan tiers et Ateliers de Penhouet and the lock at the entrance to the Bassin de St. Nazaire. Considerable incidental damage was done, of course, especially to rail facilities.26
This mission apparently convinced the Eighth Air Force command that attacks at low altitude would not yield results commensurate with the losses likely to result from such undertakings. Subsequent attacks on submarine bases were made at altitudes ranging from 17,500 to 22,000 feet which, at least until the mission against St. Nazaire on 3 January 1943, effectively foiled antiaircraft fire.27
Prior to 3 January the VIII Bomber Command conducted six more missions against the submarine bases, concentrating on St. Nazaire and Lorient, with one relatively light and ineffective attack devoted to La Pallice. A total of 199 heavy bombers, in missions varying in strength from 11 to 53 aircraft, attacked according to a fairly consistent pattern. They approached the target area overland across the Brest peninsula and, in order to elude enemy fighters, returned over water, skirting wide around the French coast. RAF fighter forces provided support in the form of short-range escort and diversionary sweeps over enemy territory. In no instance did the bombers enjoy fighter cover over the target area. Flak accounted for only one of their number, although in many instances it caused minor damage. On four occasions, however, the bombers encountered stiff opposition from enemy aircraft, which resulted directly in the loss of five more planes. In addition, two bombers crashed and two were lost to unknown causes.28
This over-all loss rate of less than 5 per cent of the attacking forces justified, from a defensive point of view, the decision to abandon attacks at lower altitudes. And over against these losses could be placed the damage done to the U-boat installations. By the end of December, St. Nazaire and Lorient were both showing the cumulative effect of repeated bombardment. Although the accuracy achieved still left much to be desired, enough bombs had fallen within the target areas to cause
at least serious inconvenience to the enemy. St. Nazaire suffered especially heavy damage. In the course of the five missions from 9 to 23 November, 158 aircraft dropped a total of 771,000 pounds of high-explosive bombs on or in the vicinity of the port facilities.29 This damage no doubt made it more difficult to service and repair s. According to an account obtained from a German naval prisoner of war, work continued after the AAF raids only in the submarine shelters which, though hit at least six times, apparently suffered no lasting damage. This same informant spoke of large-scale evacuation of the working population, which left barely enough hands to continue the restricted scale of work required in the shelters. In one shop, he said, 200 apprentices had been killed and, owing to the lack of labor to remove them, the bodies had been left in the rubble.30
The repeated attacks made by the U. S. forces at St. Nazaire in November had apparently demonstrated the virtue of concentrated effort in this type of bombing. Undoubtedly St. Nazaire, the most important of Germany's Biscay bases, had suffered heavily. But the rapid recovery of that port after 23 November appeared also to demonstrate that, if such crippling effects were to last, attacks of similar weight would have to continue at a similar rate. No mission was conducted against St. Nazaire between 23 November 1942 and 3 January 1943. During that breathing period the servicing facilities were apparently put once more into running order. British observers estimated that by 6 December the port was again in full commission. In order to retrieve the earlier successes, the VIII Bomber Command struck St. Nazaire on 3 January in the largest attack made against the submarine bases to date. Some sixty-six aircraft bombed the port, dropping 342 x 1,000-pound high-explosive bombs.3l
Accuracy on this mission was better than on most of those since the first attack on Lorient. The points of burst of 107 bombs could later be identified, and of this number, 26 were located within 1,000 feet from the aiming point, in this instance a small torpedo warehouse which was hit and demolished. Considerable damage was done in the dock area.32 A ground report claimed that, for the time being at any rate, the works of Penhouet had been put completely out of action. Several bombs fell on and around the submarine base itself, but none penetrated the reinforced concrete roof and, except for some windows, doors, and electrical apparatus damaged by blast inside the shelter, the base escaped serious injury and work proceeded without let or hindrance.33
Significant as the results of the bombing appear to have been, the nature of the opposition encountered during the mission gave Eighth Air Force observers even more to think about. Heavy resistance from fighters over St. Nazaire itself accounted for three of the bombers lost. In return for these losses, bomber crews were finally credited with twelve of the enemy destroyed, eighteen probably destroyed, and four damaged. But the greatest surprise came from the intensity and accuracy of the flak which, unlike that previously experienced, was thrown up in a "predicted barrage" rather than in an attempt to follow the attacking force continuously. This unprecedented fire destroyed three more of the attacking planes and hit an additional thirty-nine. In personnel, the mission cost seventy men missing, five killed, nine seriously wounded, and twenty-one slightly injured. In terms of aircraft, the cost was seven destroyed and forty-seven damaged. Although the most successful mission to date against the submarine bases from the standpoint of destruction to enemy installations, it was fully as costly as the ill-fated low-altitude attack of 9 November against the same objective.34 Quite clearly, the submarine bases presented problems which U. S. bombardment experts had yet to solve.
Looking back over this first phase of the effort against the bases, leaders chiefly concerned with its prosecution could come to few conclusions regarding its effectiveness; it was easy enough to compile and quote certain operational data; ground reports and aerial reconnaissance pointed to certain specific effects which have already been summarized. But it was much more difficult to determine whether any significant number of months of operations had been denied the enemy through these operations or to what extent, if any, the American bombing attacks had affected the number of s operating in the Atlantic. Information gained since the cessation of hostilities indicates that the s active in the Atlantic were steadily increasing in number during the period in question.35
Opinion varied as to the extent and relative importance of the damage inflicted by the bombing. Admiralty agencies seemed to have been warmly appreciative of the U. S. attacks, if necessarily vague in specifying their reasons.36 Late in November, Admiral of the Fleet Sir Dudley Pound, First Sea Lord and Chief of Naval Staff, wrote General Eaker praising the "fine achievement of the U. S. A/C employed in the precision bombing of the U/B bases in the French Biscay ports."37 Generally speaking, the Admiralty recommended intensifying the day offensive
against the submarine bases, with concentration on the installations in the neighborhood of the pens rather than on the pens themselves.38 Air Ministry and RAF Bomber Command opinion was comparatively lukewarm. A contemporary Air Ministry analysis, while granting that the U. S. attacks had undoubtedly embarrassed the enemy, placed greater confidence in direct sinkings of submarines by surface and air attack and in long-range antisubmarine air patrol in the areas where the U-boats operated. RAF Bomber Command continued to advocate the bombing of building yards.39
AAF Headquarters had other misgivings about the bombing of submarine bases. While generally elated over the fact that positive action was at last being taken against enemy installations by American heavy bombers, and although especially pleased with the fine series of attacks executed during November,40 headquarters agencies felt that the weight and nature of the attacks remained unequal to the task of doing "something drastic" about the menace that still threatened Allied supply lines.41 Also taken into account were the relatively high losses sustained during the last two missions--10 aircraft out of a total of 106 attacking.
Probably in an effort to allay doubts in AAF Headquarters, General Eaker had maintained a consistently optimistic tone with reference to the campaign against the submarine bases. The losses, though unfortunate, were to be expected in operations conducted repeatedly over the same objectives and in such a way that the enemy could tell by the hours of daylight and by the flight time to and from the target just when the bombers would arrive, even if their RDF had not already given fighter and flak defenses sufficient warning. Over against these losses, which were not actually prohibitive, should be placed the heavy toll taken of the enemy fighters by the American bombers. "We are still able," Eaker wrote on2 January 1943, "and shall continue to knock down better than 6-1 enemy fighters for our bomber losses. This is, we feel, an excellent exchange." Furthermore, improved tactics might in the future be expected to better the situation materially.42 The operations of November had confirmed him in the belief that with ten heavy bomber groups he could eliminate a large part--possibly 60 per cent--of the submarine menace in the Atlantic. In January he expressed the hope that as soon as it became possible for him to put 100 to 120 bombers in the air he would be able to hit submarine-building installations
in Germany proper whenever weather over the Brest peninsula was unfavorable for operations against the bases.43
The U. S. Navy contributed a more conservative estimate, both of results and of prospects, and one somewhat nearer the truth as ultimately determined. A report from the naval attache in London, for example, compared the bombing of the Biscay pens and base facilities unfavorably with other antisubmarine air operations, especially the escorting of threatened convoys.44
By January 1943, two things about the antisubmarine bombing program had become clear. In the first place, earlier assumptions regarding the imperforability of the pens were now borne out by experience. Even with the use of heavier armor-piercing ammunition it was considered doubtful whether significant damage could be done to the pen blocks. Consequently, all that could be hoped for from bombing of bases would be disorganization of the turn-around and servicing schedule.45 Secondly, in order to paralyze the operating bases and so to deny them to the Germans, it would be necessary to employ much larger forces and to bomb much more frequently than had hitherto been feasible. In answer to a direct question from Washington, the headquarters of Eighth Air Force replied that to neutralize these five bases completely 250 sorties against each base per week for eight weeks would be required.46 Both Air Ministry and Admiralty agreed in the necessity for increased frequency of attack by increased forces, for it was not an easy matter to inflict permanent damage on ports, as the RAF had found out at Bengasi and the Germans at Malta.47
The rest of the problem remained in the realm of opinion. Did results justify the effort expended against the submarine bases and the diversion from true strategic bombing which it involved? Was bombing of submarine bases the best use of the heavy bombers, or even a reasonably profitable way of reducing the submarine menace? These vital questions could not as yet be answered with any degree of finality. Involving, as they did, comparisons between divergent and even opposed schools of thought regarding the employment of heavy bombers, any tentative answers were unavoidably colored by the interests of the evaluating agencies. It was, however, generally recognized that no one method was likely to provide by itself the solution to the submarine problem, and opinion still gave the efforts of the Eighth Air Force a prominent, if somewhat indefinite, place in the antisubmarine campaign. The bombers may not as yet have affected the submarine situation
in any major way, but they had done their job well enough with inadequate forces to make most observers believe that, properly equipped, they could do it decisively.
It was not until the end of 1943 that USAAF surveys of strategic bombing results tended to confirm doubts hitherto hesitantly expressed regarding the value of bombing submarine bases. By that time the submarine had been defeated in the first round of the battle of supply, and it had become apparent that attack from the air against the at sea had been the most effective single factor in reducing the German submarine fleet, and that bombing of bases had contributed relatively little in that direction. Grand Adm. Karl Doenitz, who, as one-time commander of the fleet, was in a unique position to know whereof he spoke, later confirmed this opinion in an interview with Allied intelligence officers after his capture in 1945. Not only were the pens themselves impervious to anything but the heaviest type of bomb, he asserted, but they housed virtually all necessary repair and maintenance facilities. Bombing of surrounding installations did not therefore seriously affect the rate of turn-around. What slowed turn-around most effectively, he claimed, was the necessity for repairing the damage done to hull structure by aerial-bomb and depth-charge attacks delivered at sea.48 Undoubtedly the AAF raids caused temporary dislocations during the early months of the campaign, especially at St. Nazaire. Clearly, also, they harassed the enemy by destroying auxiliary construction plants and neighboring railway facilities and in a variety of minor ways, but these were not their primary purpose.
Enemy Aircraft and Transportation
Not all Eighth Air Force effort expended during November, December, and early January was directed against the submarine bases. Those installations enjoyed, or rather suffered, first priority; and, in fact, ten of the fifteen operations undertaken by the Eighth Air Force during that period involved attacks on the five Biscay ports. But the U. S. bombers had also been instructed to strike at the German Air Force and enemy-operated transportation facilities in occupied countries as matters of second and third priority respectively. Of the 407 bombers dispatched against targets other than submarine bases, 231 were detailed to attack airdromes and 176 to bomb targets of importance to German transportation. Owing to the vagaries of the weather, which on 12 December turned a major effort against the air installations
at Romilly-sur-Seine into a minor attack on the Rouen-Sotteville yards, only 89 of the 236 planes that completed their mission dropped bombs on aircraft installations, leaving by far the heavier weight of attack for transportation. As it happened, only one target in each category sustained any considerable pounding. Three missions against Lille accounted for almost all the damage inflicted on transportation, only one other attack, the slight and ineffective one against Rouen-Sotteville on 12 December, having been executed. In the aircraft category, although planes were sent three times to the Abbeville/Drucat airdrome and once to Cherbourg/Maupertuis, only the single raid on Romilly-sur-Seine on 20 December can be classified as effective.49
At Lille the locomotive and rolling-stock repair and construction works of the Ateliers d'Hellemmes and of Fives-Lille had been damaged in the USAAF attack of 9 October 1942, but had since been extensively repaired.50 They still constituted a composite objective of the utmost significance to Axis transportation chiefly because they served as the principal railway repair depot in France. RAF attacks on locomotives had created a serious repair situation. Consequently, the Lille shops were being taxed to the limit of their capacity.51 It was in the hope of still further constricting this bottleneck that, on 8 November, the daylight bombers were sent again to Lille52 where 30 of them dropped 293 x 500-pound high-explosive bombs intended primarily for the Hellemmes shops, which had hitherto escaped major damage. The repair shop and the machine shop both were damaged. Another attack, on 6 December, by thirty-six planes against the same target added materially to the damage already inflicted, but it is impossible to say to what extent it further retarded repair activities.53 The heaviest attack against Lille came on 13 January 1943, when sixty-four heavy bombers dropped approximately 125 tons of bombs on or in the neighborhood of the objectives. As a result of this attack, repair work and locomotive construction were seriously interrupted. At Hellemmes, for example, where locomotives awaiting overhaul had been piling up since the American raid of 8 November, work appears to have come to a complete standstill for some time.54
By mid-January 1943, USAAF bombing at Lille, at St. Nazaire where the locomotive sheds had been destroyed, and at Rouen-Sotteville, combined with RAF attacks on locomotive objectives, had created a situation in which Germany no longer could regard French railway facilities--developed before the war in excess of the demands
made on them by normal traffic--as a source of reinforcements for her own overtaxed lines.55
On 20 December the Eighth Air Force made its one effective attack of the period on the German Air Force in a relatively large-scale mission against the aircraft park and repair depot at Romilly-sur-Seine. This aircraft depot and airdrome, situated near the river Seinesome sixty-five miles southeast of Paris, held the reserve aircraft of all types for the German Air Force in France and the Low Countries and did much repair and re-equipment.56 Of the 101 bombers dispatched on this mission, 72 bombed the target area, releasing 306,000 pounds of high explosives and 25,000 pounds of incendiaries. Results were reasonably good. Damage was inflicted on hangars, barrack huts, and aircraft, and 138 craters were made on the landing ground, 10 of them on the perimeter or taxi-tracks.57 Of considerably greater historical significance, however, was the fact that, in the course of this deepest penetration yet made by USAAF planes into German-occupied territory, the bombers made contact with almost the entire force of enemy fighters located in northeast France. The ensuing air battle developed epic proportions and provided an important test of the American heavy bombers' ability to carry out unescorted missions deep into enemy territory.
Eight RAF and three U. S. fighter squadrons, all flying Spitfires, conducted diversions over areas where the German aircraft were known to be based. Enemy reaction to these efforts amounted to probably eighty-nine aircraft, but no encounters took place. In addition, 35 Spitfire IX's of the RAF escorted the bombers as far as Rouen, and 107 provided cover for them on their return trip. These operations also proved uneventful for the escorting fighters.58 It was against the bomber force that the Germans concentrated the full weight of their attack. It may have been that they were prepared for just such a mission as this, for on 12 December, the date of the preceding American raid, the bombers had flown toward Romilly, intending to attack that objective, but on finding it closed in by weather they had fallen back on a target of lower priority. At any rate, the escort this time had barely turned back (at 1150 hours) when an estimated sixty German fighters, mostly Fw-190's from the Pas de Calais area, attacked the formation.59 They came in well above, peeled off, and closed in from the front, either slightly above, dead level, or slightly below. One B-17 of the 91st Group was observed to hit the ground at Vascœuil, and a few minutes later another B-17 from the same group began to lose altitude rapidly
with a number of enemy fighters following it down. At about 1205 hours, the enemy planes were relieved by fifty to sixty fresh fighters from Caen/Bougie, Paris, and possibly Evreux. These planes continued the fight almost to the target, which was reached between 1240 and 1245. During this phase of the battle a number of Me-109's joined in the attack, some approaching from above at 10 or 11 o'clock, flying through the formation and diving out at 3 o'clock. One B-17 of the 306th Group was hit about ten minutes before the target, but it was not until a few minutes later that it started down.60
On the return trip the bomber formation suffered almost continuous attack from fighters, most of which had apparently taken part in the earlier stages of the engagement and were now making second sorties. Two B-17's of the 306th Group went down in the vicinity of Paris. Over the Channel another bomber, the sixth to be lost on the mission, went down and was last seen smoking badly and approaching the English coast at very low altitude. In addition to the six bombers lost, two more were so badly shot up that they crash-landed in England. Twenty-nine others sustained damage in some degree.61 These losses probably all resulted from enemy fighter action, for the flak encountered proved consistently inaccurate and ineffective. The losses were heavy, and they reflected the success of the German fighter pilots in adjusting their method of attack to the peculiarities of the American bombers. It was on 23 November, during the attack on St. Nazaire, that the bomber crews had first reported a change in the direction from which the fighter passes were launched.62 Hitherto, attacks had come mainly from the rear, but as the enemy discovered that the B-17's and B-24's were weakest in forward firepower, he changed abruptly to head-on attacks, which during December and January seriously embarrassed the U. S. force.63 Interrogation of crews returning from Romilly indicated that seven enemy planes had been seen to crash, that eighteen broke up in mid-air, and that twenty-seven more went down in flames. Total claims originally registered included fifty-three destroyed, thirteen probably destroyed, and eight damaged.64 These claims seemed to be excessive in view of the number of aircraft--estimated at not over 120--which could have intercepted. An Air Ministry analysis set probable figures for the Romilly action at a much lower level. Keeping in mind the heavy fire power of the U. S. force, the fact that this force had been under attack by fighters for nearly two hours, and that the visual evidence of planes
destroyed--even allowing for duplication in claims--pointed to heavy enemy losses, this report suggested thirty enemy fighters destroyed and fifteen to twenty damaged as a not unreasonable estimate.65 Subsequently, VIII Bomber Command itself lowered even this figure; the revision of 5 January listed claims of 21/31/7.66 German records suggest instead that the Americans shot down only two planes and damaged a third. Three other enemy fighters lost that day could be indirectly credited to the AAF mission, as also ten fighters listed as damaged for reasons not directly attributable to "enemy" action.*
Logistical and Tactical Problems
It was not enough to evaluate Eighth Air Force operations in terms simply of the results obtained. From the very beginning it had been apparent that its achievements would have to be interpreted in relation to the factors that limited both the scope of its operations and the degree to which those operations were effective. These limiting factors loom especially large during the early months dealt with in these chapters, for it was then that long-term plans were being laid for the air war against Germany--and, be it noted, always with an eye to what the Eighth Air Force was doing in these essentially experimental operations. The problems themselves fall into two large categories: (1) tactical problems and (2) those of supply, maintenance, and operations. Most of them had made their appearance and had received initial consideration in the weeks prior to November 1942. During the period covered by this chapter they developed rapidly and much thought and effort were expended in an attempt to solve them.
Basic, of course, among these factors was the size of the operating force itself, essentially a problem of logistics. The departure of the Twelfth Air Force in early November had left the parent organization with combat units amounting to seven heavy bombardment groups less one squadron† (two of which groups were scheduled for TORCH at some later date), one single-engine fighter group minus its ground echelon, and one observation group scheduled eventually for North Africa. Of the heavy bomber units, four (the 9st, 303d, 305th, and
* This category, for example, might include planes suffering accident as a result of an exhausted fuel supply after combat. German records credit the mission with destruction on the ground by bombing of five GAF bombers and one fighter, and with four additional bombers damaged. (Information supplied through courtesy of British Air Ministry.)
† See above, p. 235.
44th) became operational only on 7 November; two (the 306th and 93d) had at that date been on an operational status for only one month. The 92d was used, from November 1942 until May 1943, for training purposes only. On 5 December the 93d Group was ordered to move its air echelon, minus one squadron, to North Africa for a temporary tour of duty which lasted until the end of February 1943. Meanwhile, one squadron of the 93d Group had been on antisubmarine duty with RAF Coastal Command from 25 October to 25 November, and another from the same outfit had been detailed as an experimental unit to work on the blind-bombing project. During November, December, and January, therefore, General Eaker could count on a combat force of at most six heavy bombardment groups.67 The Twelfth Air Force had also left the Eighth so low in air force service elements that General Spaatz in November expressed doubts whether sustained operations could be maintained by the remaining combat units.68
Moreover, the prior demands of TORCH made it impossible to keep up to full strength those units that were regularly available. The problem of replacements received a great deal of attention during the fall and winter of 1942 both in Washington and in Headquarters, Eighth Air Force. Early in November, General Spaatz urged that the rate of replacement for units in the United Kingdom be stepped up to the level proposed by the War Department in July 1942. The plan then presented had provided for 20 per cent replacement in heavy bombers per month, additional aircraft for reserve and for the augmentation of units through December 1942 at the rate of two per month per group, and combat crews for 75 per cent of the aircraft thus provided.69 On 2 December 1942 he cabled from Algiers urging that replacements for the African theater be expedited in order that no further drain would be necessary on the already strained units of the Eighth Air Force. Further withdrawals, he warned, would seriously affect operations from the United Kingdom which were of vital importance not only in themselves but because they prevented the enemy from diverting air strength to the Mediterranean.70
AAF Headquarters, while sympathizing fully with the plight of the Eighth, was apparently unwilling to jeopardize more critical projects in order to build up the force in the United Kingdom, especially in view of the fact that shipping space was no less at a premium than were men and materiel.71 Moreover, the estimate of Eighth Air Force requirements in Washington seems not to have coincided exactly with
that made by Spaatz and Eaker, for records in AAF Headquarters did not indicate so serious a situation as that reported from the theater.72 Be that as it may, by the end of January 1943 the Eighth Air Force was not receiving replacement planes and crews as fast as it was expending them.73
The result was that under existing operational conditions the force employed in the day bombing program was not large enough to accomplish any major item of the task it had undertaken, a fact which had become apparent during the campaign against the submarine bases. The size of the operating force also limited the choice of targets, for it was felt that only a force large enough to protect itself readily should be dispatched over the Reich. Yet, on the other hand, the necessity of restricting activity to a single, relatively narrow area in occupied France made it impossible to disperse the enemy fighter defenses and so tended to increase combat losses. In any event, it was obvious that a lower rate of loss might be expected when a force could be employed which was large enough to saturate any given system of defense.74
Regardless of the number of aircraft and crews on hand, the number that could be sent out on any particular mission depended on the ability of maintenance crews and depots to keep the aircraft in operational order, to repair battle damage, and to make such modifications as combat experience demonstrated to be necessary. That ability, in turn, depended on an adequate supply of parts and a force of trained personnel large enough and in a position to devote enough time to this work to keep up with the requirements of the operational units. In the fall and winter of 1942 neither of these conditions existed, and so it was not possible to realize fully the potential strength of the bomber force available. On the 15 missions studied in this chapter the VIII Bomber Command was able to dispatch an average force of 70 bombers with a maximum of 101, yet these figures represented a discouragingly low percentage of the total aircraft on hand in the theater. Through November, for example, only 51 per cent of this total was in combat condition.75
The Twelfth Air Force had left the Eighth depleted in service units, and those left in the United Kingdom were still required to give high priority to equipment destined for North Africa.76 For similar reasons, the Eighth continued also to suffer from an insufficient flow of parts and tools.77 And there appeared little likelihood that the situation would improve for some time to come, for, when shipping became available
to carry the required personnel and equipment, it would probably have to be used for transporting combat units. Although General Spaatz warned that "a marked reduction in the rate and efficiency of air operations must be expected until the required service elements have caught up with the combat elements," he advocated allocating available shipping to combat replacements as a matter of first priority, since the latter required more time to become acclimatized and might, if necessary, be supported on an emergency basis until normal service units arrived.78
At the same time, the Eighth Air Force was facing a rising rate of battle damage which placed an increasing load on the already inadequate repair facilities, with the inevitable result that a large proportion of heavy bomber strength remained inoperational. In September, 13.3 per cent of the attacking planes suffered reparable damage; in October, 37.7 per cent. By December the percentage in this category had risen to 42.1, with January promising an even higher proportion of damaged planes.79 Still further to complicate matters, it had been found necessary to modify the heavy bombers to meet unforeseen tactical and operational conditions and, moreover, to do the work to a large extent in the theater.* Until a standard model could be turned out in the United States, fully equipped for combat in the European theater--and even then special projects would require special modifications--changes had to be made at almost all echelons by the cut-and-fit method, which again increased the load on available maintenance facilities.80
Maintenance difficulties were reflected in the relatively high rate of abortive sorties resulting from mechanical failure. Since October that rate had increased considerably, amounting in November to 23 per cent of all abortives. Crews were instructed to return without entering enemy territory if turrets became inoperative, if guns jammed, or any other important items of equipment failed.81 It is possible, of course, that the anxiety of group commanders to get as many of their planes in the air as possible had the effect of starting some that, under less hectic operating conditions, might have been left on the ground for more thorough overhaul. In any case this was a serious matter, for the total abortive rate was itself high. Of the 1,053 bombers dispatched from 21 October 1942 to 13 January 1943, 421 had failed to attack.82 In January, General Eaker admitted that, next to the large number of aircraft
* See below, chap. 18.
out of commission, the large number of abortive sorties had been his chief worry during the preceding weeks.83
Even more important than mechanical failure as a cause of abortives was the weather. Too often the bombers had to take off in mud and water or fly in rain which caused their guns to freeze or their windows and sights to become blurred at high altitude. And there was always a very good chance that, regardless of expert predictions, the bombardiers would find their targets partially or totally obscured by clouds. As high as 50 per cent of the abortive sorties could be traced directly or indirectly to the weather. Things were improving slightly by January: crews were, for example, learning how, when runways were covered with water, to prevent icing of guns and turrets by the use of oil; and in some instances malfunction of bomb-bay doors owing to the same conditions was prevented by removing the doors completely.84 But, in the final analysis, only fine weather could entirely eliminate these operational hazards.
Weather, indeed, continued to act as the greatest of all the factors limiting Eighth Air Force bombing. Only once during November and December was it reported that a mission had actually been canceled because of maintenance and repair difficulties, and then the trouble arose only after three successive days of operations. On the other hand, weather conditions held available aircraft on the ground on numerous occasions.85
High hopes continued to be placed on blind-bombing techniques, and much study was being devoted to the use made of special navigational devices by British Pathfinder units. It was hoped specifically that development in that direction, together with the improved weather which might be expected during the coming spring and summer months, would permit an average of six missions per month per operational bomber instead of the three missions averaged during the fall of 1942.86 But the initial experiments in using single B-24's equipped with Gee for "moling" * or "intruder" missions gave little ground for optimism. On 2 January 1943, four B-24's of the special experimental squadron made the first of these blind operations over manufacturing cities north of the Ruhr with the object of alerting air-raid crews and otherwise harassing the enemy. It had been specified that, in view of the
* Missions flown by single radar-equipped bombers sent out in overcast weather for the purpose of alerting the enemy's anti-aircraft defenses and generally contributing to a feeling of insecurity on his part.
valuable equipment carried and the small intrinsic value of any bombing done, the airplanes should return to base if cloud cover proved insufficient to give protection. Perversely, the weather cleared and all four returned without bombing. Twice more in January B-24's went out on expeditions of this sort, only to be foiled again by fine weather. Short of resorting to night bombing (the RAF had conducted eighteen missions during January) the Eighth Air Force had little choice but to wait for favorable weather and a wider selection of targets.87
The Eighth Air Force also faced certain major tactical problems. Success depended primarily on the ability of the day bombers to hit and destroy their objective and on their ability to defend themselves against flak and fighter attack. Questions on both these accounts had dogged Eighth Air Force operations from the beginning. During the fall and winter of 1942 they became rapidly more pressing. In order to hit such relatively small, isolated, and invulnerable targets as submarine base installations, better offensive tactics--particularly improved accuracy--would have to be developed. At the same time, the vigorous growth of German countermeasures called attention even more urgently to the problems of defense. Prior to 21 October neither flak nor fighters had seriously threatened the American bombers. Clearly, the Germans had been caught unprepared for a weapon such as the day bomber, which not only could do real damage from extreme altitudes but could also shoot it out with the best fighters in the Luftwaffe. However, as many observers, including General Spaatz, had foreseen, they lost no time in adjusting defensive tactics to cope with this unprecedented attack. If they adjusted neither so rapidly nor so radically as some had feared, they nevertheless gave the Eighth Air Force grounds for serious concern and taxed the ingenuity of its tactical experts.
Except for the few seconds of the bombing run, when the purpose of the heavy bomber is realized, all phases of a bombing mission are dominated by considerations of defense. But considerations of defense had to be carefully balanced against those of offense, for they were not always reconcilable; and they had also to be weighed in relation to each other, for what would offer protection against flak might increase vulnerability to fighters. For example, high-altitude bombing reduced risk from flak, but it also reduced bombing accuracy. Bombings by a single aircraft might, under ideal conditions, be best for both accuracy and protection from flak but would not provide sufficient defense against fighter attacks. Large bombing units flying in formation would
give adequate protection against fighter attacks but would increase flak hazards and at the same time reduce accuracy by enlarging the resulting bomb pattern. As experience was gained, constant adjustment was made in the multilateral compromise necessitated by this problem of integrating defensive and offensive tactics.88 By early 1943 many of the basic lessons had been learned, much of the pioneer work having been done by the 1st Bombardment Wing under the successive command of Generals Longfellow, Kuter, and Hansell.89
German flak defenses at first had proved ineffective in opposition to aircraft flying at altitudes above 20,000 feet, and German fighter pilots had been unwilling to come very close, preferring to stand off just outside the range of the bombers' guns and wait for a favorable opportunity to duck quickly in and out of the formation. During October, enemy fighter tactics reflected a feverish determination to find a way to stop the day bombers. Though many types of attack were tried, tail attacks predominated. This had been the accepted angle of attack against bombers, however, and it was the type against which the USAAF had undertaken to protect its heavy bombers by the addition of especially heavy armament and armor plate.90 The climax in this phase of the German attack came on 21 October when the Fw-190's (bearing the yellow nose paint characteristic of Goering's elite fighter wing) made a series of desperate attacks from the rear in an apparent effort to find a blind spot safe from both dorsal and ball turrets. They came in, sometimes in formations of three, at flight level, opening fire at 800 yards. Three bombers were lost as a result of this action, and six others damaged.91
Beginning with the St. Nazaire mission of 23 November, the Germans changed their tactics abruptly. Oberleutnant Egon Mayer, who commanded the attacking fighters that day, is credited with developing the head-on attack. Having studied the largely unsuccessful efforts made so far to stop the heavy bombers, he ordered a frontal attack, leading one element personally. The tactic worked well, for it caught the American bombers in their most vulnerable spot.92 At that time some B-17's had one.30-cal. hand-held gun, firing through one of four eyelets just off center, and some mounted two.50-cal. side nose guns. In either case, a blind spot was left in front which neither the upper turret nor the ball turret could reach. The B-24's were equipped with.50-cal. side nose guns, and a single.50-cal. center nose gun mounted to fire below horizontal only. This armament also left a blind spot which the
upper turret could not cover.93 The only disadvantage to the head-on attack from the enemy point of view was that it made necessary a high degree of skill and training on the part of the fighter pilots in order to make effective use of the short time allowed by the very rapid rate of closure, even when the approach was executed at low speed; and it was for that reason that it was not universally adopted and, indeed, was officially frowned upon in August 1943 when the number of inexperienced pilots had increased so rapidly that such attacks could only prove disastrous. They continued to be made, however, throughout the air war in Europe.94
Through January 1943 nose attacks continued to predominate and accounted for most of the losses suffered by the VIII Bomber Command as a result of encounters with enemy fighters.95 Losses from enemy fighter fire, in turn, constituted by far the larger proportion of total losses, which had risen from an average of 3.7 per cent of the attacking force in November to 8.8 and 8.7 per cent in December and January respectively.96 Bomber crews had to face the enemy's frontal attacks very frequently just over the target, when the confusion inevitably resulting would be most likely to spoil the bombardier's aim. In fact, it was believed that to break up the bombing run had now become a primary objective of the German fighters.97 The frontal attacks, therefore, came during these months to be the chief defensive problem of the Eighth Air Force.
It was immediately clear that the only effective countermeasures would be the addition of increased forward firepower in the bombers and an improved defensive formation which would give all planes the benefit of mutual protection. Of these remedies, the addition of nose guns was the more critical item, because it would involve a great deal of time-consuming modification both in the United Kingdom and in the United States. Meanwhile makeshift tactics were devised. One method of countering the front-quarter, level attack--the method reported in December as the one officially approved--consisted of a diving turn into the attack, which uncovered the top turret and, incidentally, tended to spoil the enemy pilot's aim. It was hoped that in this way any such attack would encounter not only the front, side-firing guns but the top turrets of at least some bombers in the formation.98
Modification for nose guns began promptly. Pending the installation of a standard power-driven turret in the B-17, flexible, hand-held. 50-cal. nose guns were provided in most of the Fortresses destined for
the European theater; and the standard B-24 center nose gun was modified in such a way that it could fire above the horizontal99 In the theater, similar modifications were undertaken on as many aircraft as could be accommodated in the depots. The need for such modifications was so great that improvised field installations were authorized as long as they conformed to basic requirements. By mid-January, most heavy bombers in the United Kingdom were equipped with effective forward fire, if only from single, improvised,.30-cal and.50-cal. hand-held guns.100 Complete satisfaction could only result from the installation of a turret in the nose, but it was not until August and September of 1943 that the improved B-17's and B-24's arrived in the theater complete with this power-driven equipment.101
Although it was a standard defense against all fighter attack, the large formation of bombers so stacked as to provide mutual fire support proved especially helpful in countering the frontal attacks. Indeed, it was during the fall and winter of 1942, and primarily in answer to this particular problem, that the 1st Bombardment Wing evolved a system of formations which became the prototype for operations in the theater. When General Kuter took over the wing on 6 December 1942, he found the four groups each operating according to its own tactical doctrine. No wing organization existed for tactical purposes, and consequently the groups collaborated only in the sense that they all attacked the same target roughly at the same time. No effort was made to secure additional fire support by coordinating group tactics. Squadrons and groups had developed into cohesive teams, but the wing as a whole had not become a combat unit. Acting on the assumption that the larger the formation, consistent with requirements of maneuverability, accuracy, and control at high altitudes, the more mutual fire support would be obtained, General Kuter set about to weld the squadrons and groups into the largest practicable combat units.102
At first the groups had bombed in elements of three aircraft, but fighter attacks demonstrated that bombing by elements, however satisfactory from the point of view of accuracy, did not provide sufficient defensive power. Bombing by squadrons composed of two elements of three aircraft each was then tried. The intensity of enemy attacks soon made it necessary to resort to bombing by groups of three squadrons. Thus a formation composed of eighteen to twenty-one bombers, known as a combat box, became the standard minimum combat unit, and it was stacked in such a way as to uncover as many of the top and
bottom turrets as possible in order to bring the maximum fire to bear on the critical forward hemisphere. It was considered the smallest unit feasible for defensive purposes and the largest that could be handled readily on the bombing run.103
But it appeared, especially on the trip toward the target and again on withdrawal, that mutual fire support could be greatly increased by combining two or more combat boxes into a single defensive formation. It was not, however, considered practicable to fly the entire bombardment wing in one formation. Anything larger than a formation of two or three combat boxes would have required deployment in such depth that the differences in wind velocity and aircraft performance at different altitudes would have aggravated the tendency of any formation to telescope and lose effective position. Moreover, two groups were about all that could be readily briefed and controlled by a single combat commander. Accordingly, the 1st Bombardment Wing formed two combat wings of two groups each. In each of these combat wings the senior group commander assumed command and was given full authority in planning and executing the mission. This organization existed for tactical purposes only and in no way affected the administrative organization of the bombardment wing.l04
The combat wing, consisting of two or three combat boxes, thus became the maximum defensive formation. It was generally deployed in echelon up, in a vertical wedge similar in principle to that of the combat box, although in the period under review many variations occurred. In early 1943 it was apparently also planned to use the combat wing as a unit in formation bombing whenever the fighter opposition seemed likely to be strong enough immediately over the target to warrant its use; this, despite the fact that it would be a clumsy formation to maneuver around the initial point onto the bombing run and that the resulting bomb pattern would tend to be too large for the desired accuracy.105
Fighter cover, and lots of it, had originally been held a prerequisite to day bombing, and the early missions had been flown under a huge umbrella of friendly fighters. But after October most priority targets lay beyond the range of the Spitfires, which for the time being were the only fighters available for such operations. They usually accompanied the bombers part way in toward the target area and provided withdrawal support on the way out. During the missions, moreover, large fighter forces, still for the most part RAF, conducted diversionary sweeps to confuse the enemy RDF; but the bombers were generally
left to the as yet uncertain protection of their own gunners during the critical time over the objective.l06
As losses mounted during these partially or entirely unescorted missions, and especially as the time drew near when operations would have to be conducted over the Reich itself with its presumably denser fighter defenses, it began to look as if the long-range fighter would after all be a necessary part of a successful day bombing offensive. By 15 January the 78th Fighter Group was due to have its quota of P-38's, and General Eaker hoped that with these he could reduce bomber losses over the submarine bases by one-half. Unfortunately this group soon followed its predecessors to Africa, and the need for long-range fighters remained.l07
As an alternative to the long-range fighter, the escort-bomber, known provisionally as the XB-40 or YB-40, was in the process of being developed. A B-17 especially equipped to combat enemy fighters, it carried extra armament, armor, and ammunition in place of the usual bomb load. Conceived as a possibility for the European theater in 1941 and actually planned in the summer of 1942, it was scheduled to appear in the European theater by March 1943.* Although contemplated without enthusiasm by General Eaker, it was favored by many commanders who hoped that, by mixing it with the bombers in ratio of 1 YB-40 to 2 or 3 bombers, they might free the latter from the limitations of fighter range and send them over Germany as far as their fuel would take them.108
The Eighth Air Force had less reason to fear antiaircraft fire than fighter attacks during the period under review. Barely one-fourth of the bombers lost in action could be credited to flak alone, and only a few more bombers suffered flak damage than were hit by enemy aircraft. But, whereas the percentage of damaged aircraft that were hit by fighter action showed little tendency to increase, the percentage of damaged bombers that had been hit by flak appeared definitely to be rising. And on two occasions, at St. Nazaire on 9 November and on 3 January, substantial losses had been sustained as a result of antiaircraft fire.l09
The increase in flak damage reflected a marked improvement in German antiaircraft technique. Flak batteries were now concentrated in such a way as to fit the pattern of USAAF targets, with special attention
* See Volume I, p. 604.
given to the submarine bases on the Bay of Biscay. Originally the only type of fire encountered was that termed a "continuous following," which required the gunners to make a continuous prediction of the position of the target aircraft. Often this type of fire was thrown up behind the formation and gradually worked forward. It was essentially a trial-and-error method in which altitude could be estimated more easily than deflection, since the gunner had to predict some twenty seconds in advance the point at which the target aircraft would be and since his 88-mm. shells had a lethal radius of only thirty feet. Although the gunners seem to have improved the accuracy of their fire, this method proved relatively ineffective at high altitudes, provided that positive--though naturally not regular--evasive action were taken by the bombers.110
A much more effective technique, if the target could be determined in advance, was that called a "predicted barrage," in which flak was thrown up throughout a limited area through which it had been calculated the attacking aircraft would have to fly. It was a method uniquely adapted for use over the submarine bases, which were well-known objectives not easily confused with neighboring targets. In fact, it was at St. Nazaire, on 3 January 1943, that a predicted barrage was first encountered--with serious results to the attacking bombers. The technique was not, however, one likely to succeed in such areas as the Ruhr, where targets of high priority abounded.111
The tactics best suited for penetrating heavy flak defenses were simple enough, but they almost all necessitated some degree of compromise with the requirements for accuracy or for defense against enemy fighters. Positive evasive action might be taken for as long as possible, the length of the level bombing run being reduced to the shortest commensurate with careful aiming. The bombers might converge on the target nearly simultaneously on at least two axes; they might maintain a substantial differential in altitude between units; and they might take maximum evasive action immediately after release of the bombs. But care had to be taken not to disperse them to such an extent that elements would fall prey to fighter attack. Finally, to escape flak most effectively, the bombers had to fly at the highest altitude compatible with accurate bombing.ll2 Here, in a sense, was the most difficult compromise to make. It might fairly be said that, in these early months at any rate, flak handicapped effective bombing operations less by destroying or damaging bombers than by forcing the attacking planes to bomb
at altitudes too high for their more or less inexperienced crews to achieve consistent accuracy.
The question of bombing accuracy overshadowed all others pertaining to the offensive aspect of bombardment. Unfortunately it is not possible to say anything very precise about the degree of accuracy achieved in those days, for the information available is too incomplete, too inconsistently reported, and filled with too many variables to permit any worth-while conclusions. Despite the fact that AAF Headquarters exhibited an anxious interest in the subject, it was only on data accumulated after 1 January 1943 that any systematic analysis became feasible.113 This much, however, is incontestable: results in the fall and winter of 1942, though initially encouraging, especially for inexperienced crews, were disappointing to all those who, trained in the "pickle-barrel" school of bombing, knew how accurate the American bombers could be.1l4 An average of only about 50 per cent of the bombs dropped could be identified by photographic reconnaissance. Although many "duds" were reported by ground sources, it may be assumed that a large proportion of the unidentified bomb falls represented "gross" errors.115
It was this prevalence of so-called gross errors that concerned bombing analysts most acutely. Under practice conditions, accuracy might conceivably be improved indefinitely by training the bombardiers to set their sights more precisely and the pilots to hold a steadier course during the run on the target. There were thus in practice exercises few gross errors to contend with and few errors stemming from intrinsic faults in the equipment. Most errors were errors of adjustment alone. Things were very different in combat, where the confusion and excitement increased the incidence of gross errors to the point where they became the dominant factor governing bomb dispersion. Clearly, then, if the cause of these sizable errors was not discovered and removed, the Norden bombsight with its delicate adjustment would be valueless. It was, in fact, considered possible that in such an event an inferior sight requiring less careful adjustment might have to be adopted, a step which would seriously have compromised the ideal of precision which underlay the American bombardment theory.116
Undoubtedly many gross errors resulted from mechanical failure, the bombs either hanging up or salvoing prematurely. At high altitude the extreme cold, in addition to the strain on the airplane caused by the bomb load, sometimes impaired the functioning of the release mecha-
nism. Much more important was the frequent failure of pilots, bombardiers, and navigators to identify the target. Although an extreme case, it is instructive to notice that on the operation of 18 November 1942 one formation was able to bomb St. Nazaire under the impression that it was bombing La Pallice, 100 miles away. A more typical case occurred in the Lille raid of 8 November when some twenty to twenty-five bombs struck near a factory three miles short of the intended target, which was also a factory but situated in quite different surroundings.ll7 The development of perspective maps, then well under way, helped reduce the likelihood of mistakes of this sort by providing the bombardier with a picture of the target as he was likely to see it rather than as it appeared on the older type of vertically projected target map.118 Then, too, it was often difficult to follow a set course in the face of unexpectedly strong cross winds. And many errors arose from failure to set instruments properly, either because of combat excitement or because the severe cold and the encumbrances of oxygen apparatus, heavy clothing, and parachutes prevented dexterous manipulation.l19
Most unsettling of all factors making for inaccuracy was the necessity of conducting a steady bombing run in the face of enemy antiaircraft or fighter action. To one observer, bombing accuracy appeared to be inversely proportional to the resistance encountered at the target. In order to guard against flak, evasive action was normally taken for as long as possible on the approach to the target, leaving a maximum of fifty seconds for the level bombing run. During that time delicate adjustments had to be made with extreme dexterity and speed, and often under enemy attack.120 An additional difficulty arose from the fact that, in order to maintain an effective defense against fighters, the formation was likely to be too large to produce a satisfactory bombing pattern.121
Various solutions to these bombing problems were suggested. One obvious way to increase accuracy, though not of course to reduce the number of gross errors, was to bomb at lower altitudes. But the experiment of 9 November at St. Nazaire discouraged further planning in that direction, and a lower probability of error was exchanged for lower vulnerability to antiaircraft. Much naturally depended on a constantly improved state of training and experience, which alone would remove many of the causes of error.122 To insure a steady bomb run and so give the bombardier time to set his sights, pilots and bombardiers were urged to use their automatic flight-control equipment
(AFCE) which, when it functioned properly as at that time it did not always do, gave more precise results than manual flying.123
Some commanders believed that one way to get accurate aiming in formation bombing would be to have the leader in the formation set his sights accurately for deflection, even at the expense of accuracy in range, and leave the remaining crews to set theirs for range only, taking their direction simply by holding their place in the formation.124 In this way group bombing could be accomplished without the risks and confusion likely to ensue should each plane in the formation attempt to make its own adjustment for deflection. In a further effort to exploit the possibilities of group bombing, and incidentally to escape from the irregularities that seemed always to crop up when bombardiers of uneven ability bombed individually, some groups resorted in January 1943 to bombing entirely "on the leader," each bombardier taking his signal from the lead plane. Initial results of this method, though not at that time conclusive, proved very encouraging.125 Finally, one of the most urgent requirements for improved accuracy was some sort of improved firepower by means of which the frontal attacks, made so consistently by the German fighters in December and January, could be effectively countered and the morale of the bomber crews be correspondingly raised.126
The problem of accuracy, and indeed that of bombing in general, thus became inextricably entangled with that of defense. The method of bombing as worked out by the 1st Bombardment Wing during late 1942 tended to be dictated more by the nature of the opposition met than by the theoretical requirements of precision bombardment. The enemy practice of attacking during the bombing run, even in the presence of antiaircraft fire, made it advisable to preserve as large a formation as possible and one so arranged as to give all elements the maximum of mutual protection. A large formation (and it was tentatively suggested that bombing might be done in combat wing formation) increased vulnerability to flak and, if the bombing were done on the leader, it was likely to produce a larger bomb pattern than when the work was accomplished by smaller formations. If, on the other hand, flak defenses were known to be concentrated, it was necessary to accept higher vulnerability to fighters by splitting the formation so as to reduce risk from flak.127
In this chapter and the one immediately preceding it, a story has been told of things accomplished and problems encountered by the Eighth
Air Force prior to mid-January 1943. It was on the basis of these achievements and in the face of these half-solved problems that General Arnold took his stand on behalf of the daylight precision bombing of Germany at the Casablanca conference in January. The record was incomplete and the conclusions it warranted were necessarily tentative; but it enabled him to state the case for the daylight bombardment campaign strongly enough to insure for it a place, and an important one, in the plans forged at that time for the defeat of the European Axis.
Notes to Chapter 8:
1. A-5, 8th AF, An Appreciation of the Air Effort against Submarines, 16 Jan. 1943, included as Tab G in Intel. Service, Hq. AAF, An Evaluation of the Air Effort against Submarines (hereinafter cited as Evaluation). See also ibid., App. C.
2. Ibid., Tab N, Bombing Attacks on the German Industry, 21 July 1942.
3. Ibid., Tab G, App. C.
4. AC/AS, A-2, Hq. AAF, Air Estimate: The Submarine Situation in Europe (hereinafter Cited as Air Estimate, Subs.), Pt. I, Opns. Sec., 12 Feb. 1942, p. 1. Cf. ltr., Stratemeyer to Spaatz, 26 Nov. 1943; AAF Reference History No. 7, The Antisubmarine Command, chap. iii.
5. Air Estimate, Subs., Pt. I, pp. 4-7.
6. Ibid., pp. 94-5. See also memo for CG AAF from Brig. Gen. C. W. Russell, 3 Nov. 1942.
7. 2d ind. (no basic), CG 8th AF to CG AAF, 15 Sept. 1942; ltr., Spaatz to Arnold, 31 Oct. 1942.
8. Air Estimate, Subs., Pt. I, pp. 4-7.
9. Evaluation, Tab N.
10. Ibid., Tab G, App. C.
11. Ltr., Eaker to Spaatz, 14 Oct. 1942.
13. Memo for CG AAF from Russell, 3 Nov. 1942. Cf. Plans Div. Digest, 7 Nov., 21 Dec. 1942.
14. First 1100 Bombers Dispatched by Eighth Bomber Command (hereinafter cited as First 1100), 11, 383; CG 8th AF, extract of Daily Diary, 2 Nov. 1942; ltr., Arnold to Eaker, 2 Dec. 1942. The B-24's referred to were to be replaced from allocations to the AAFAC.
15. First 1100, Mission 15.
17. OSS Rpt., 26 Dec. 1942; OSS Rpt. P.1/3780, 27 Jan. 1943; First, 1100, Mission 15; Interp. Rpt. S. A. 154; ORS Day Raid Rpt. 14, 30 Dec. 1942.
18. OSS rpts. cited in n. 17. See also translation of memo from Col. Coustet, 30 Oct. 1942.
19. First 1100, Mission 15.
20. Ltr., Spaatz to Arnold, 31 Oct. 1942.
21. Ltr., Arnold to Spaatz, Oct. 1942; 2d ind. (no basic), CG 8th AF to CG AAF, 15 Sept. 1942.
22. Ltr., Spaatz to Arnold, 31 Oct. 1942.
23. First I100, Mission 18.
24. Ibid., 11, 401 ff.
25. Ibid., Mission 18.
26. ORS Day Raid Rpt. 17, 30 Dec. 1942; First 1100, Mission, 8. See especially Interp. Rpt. K.1433 and Statistical Analysis of Opns.
27. See Tac. Mission Folders, passim.
28. First 1100, Mission 21; reports on Missions 19 to 27, in Tac. Mission Folders.
29. First 1100, 11, 369.
30. Industry Dmg. Rpt. 62, 11-22 Nov. 1942, 23 Dec. 1942; OSS Rpt., 11 Jan. 1943; N. I. D. U. C. Rpt. E964C, Attacks on Bases by U. S. Aircraft, 3 Dec. 1942. See also interp. rpts. in respective mission rpts., especially rpt. on info. recd. from German naval POW taken in Dec. 1942.
31. Intel. Staff, Air Ministry, Analysis of Results of USAAF Bombing Attacks on Submarine Bases in France during November 1942 (hereinafter cited as Analysis of Results); VIII BC Narrative of Opns. in Tac. Mission Folder 28.
32.ORS Day Raid Rpt. 28, in Tac. Mission Folder 28; Ind. Dmg. Rpt. 66, 23 Feb. 1943.
33. OSS Rpt. P.1/4330, 2 Mar. 1943.
34. AAF Info. Intel. Rpt. 43-4, 26 Mar. 1943; ORS Day Raid Rpt. 28; Evaluation, Tab G, App. A. Figures on USAAF losses represent losses from all causes, including crashes. (Cf. incl. to rpt., CG 8th AF to CG AAF, 3 Mar. 1943.)
35.ONI, The Conduct of the War at Sea (translation of essay by Adm. Karl Doenitz), 15 Jan. 1946, p. 19. Cf AAFRH7, chap iii; Naval Attache Rpt., 12 Feb. 1943; ltr., Eaker to Arnold, 11 Jan. 1943.
36. First 1100, 11, 367; N. I. D. U. C. Rpt. 272, 21 Nov. 1942.
37. AFSHO Special File 77 (1trs.).
38. Ibid.; ltr., Naval Staff Officer, Hq. RAF Bomber Comd. to Eaker, 25 Nov. 1942, in First 1100, 11, 380; N. I. D. U. C. Rpt. E964C, 3 Dec. 1942.
39. Analysis of Results; ltr., Hq. VIII BC to Eaker, 2 Jan. 1943; ltr., Eaker to Arnold, 11 Jan. 1943.
40. Ltr., Stratemeyer to Spaatz, 26 Nov. 1942; ltr., Arnold to Eaker, 2 Dec. 1942.
41. Ltr., Stratemeyer to Spaatz, 26 Nov. 1942.
42. Ltr., Eaker to Stratemeyer, 2 Jan. 1943; ltr., Eaker to Arnold, 11 Jan. 1943.
43. Ltr., Eaker to Arnold, 26 Nov. 1942; ltr., Eaker to Arnold, 11 Jan. 1943.
44. Naval Attache Rpt., 12 Feb. 1943.
45. Evaluation, Tab G, App. F; ORS Special Rpt. I, Perforability of German Submarine Pens, 8 Dec. 1942.
46. ORS Special Rpt. p. 3. Cf. Evaluation, Tab G, p. 3.
47. Analysis of Results. Admiralty was reported to be in agreement with conclusions arrived at in this paper.
48. USSBS Interview 59, Grand Adm. Karl Doenitz, 28 June 1945. Cf. AC/AS, Intel., Strategic Aerial Bombardment of Europe, 10 Dec. 1943.
49. See ORS Day Raid Rpts., in Tac. Mission Folders.
50. Interp. Rpt. K.1449, 16 Dec. 1942, in Tac. Mission Folder 24 (Lille, 6 Dec. 1942).
51. Ind. Dmg. Rpt. 63, 5 Jan. 1943.
52. Ibid.; Ind. Dmg. Rpt. 66, 23 Feb. 1943.
53. Tac. Mission Folders 17 (Lille, 8 Nov. 1942),24 (Lille, 6 Dec. 1942).
54. ORS Day Raid Rpt. 29, 19 Feb. 1943, in Tac. Mission Folder 29 (Lille, 13 Jan. 1943); Ind. Dmg. Rpts. 66 (23 Feb. 1943), 74 (11 Nov. 1943); Interp. Rpt. F.S.85, 3 Feb. 1943, in Tac. Mission Folder 29 (Lille, 13 Jan. 1943).
55. Ind. Dmg. Rpt. 66.
56. Tac. Mission Folder 26 (Romilly, 20 Dec. 1942).
57. Intel. Narrative 26 and ORS Day Raid Rpt. 26, 2 Feb. 1943.
58. Rpt., USAAF Attack on Romilly, 20 Dec. 1942, n.d., in File 77 (Studies).
60. Ibid. See also Intel. Narrative 26.
61. Rpt., USAAF Attack on Romilly, 20 Dec. 1942, n.d.
62. CM-IN-10858, USAWW to AGWAR, 239, 25 Nov. 1942.
63. See appropriate mission rpts. in Tac. Mission Folders.
64. Intel. Narrative 26.
65. Rpt., USAAF Attack on Romilly, 20 Dec. 1942, n.d.
66. Incl. to rpt., CG 8th AF to CG AAF, 3 Mar. 1943.
67. Hq. AAF, Statistical Sum., 8th AF Opns., 17 Aug. 1942--8 May 1945. Cf. A/P Status Rpts., BOLERO Papers; History, 2d Bomb. Div. (and 2d Bomb. Wing), Activation to 31 Dec. 1943, pp. 37-39.
68. Ltr., Spaatz to CG ETOUSA, 9 Nov. 1942.
69. CM-IN-4417 (11-11-42), London to AGWAR, S, AFN 307A, 10 Nov. 1942; CM-OUT-2851 (7-11-42), WDOPD to USFOR, 10 July 1942. Cf. ltr., Spaatz to CG ETOUSA, 9 Nov. 1942; ltr., Spaatz to Stratemeyer, 7 Nov. 1942.
70. CM-IN-825 (12-2-42), London to AGWAR, 1026, 2 Dec. 1942.
71. Ltr., Arnold to Eaker, 2 Dec. 1942; CM-OUT-4254 (11-13-42), AFDAS to CG 8th AF, A-753, 13 Nov. 1942.
72. See sources in n. 71.
73. A/P Status Rpts., BOLERO Papers; ltr., Eaker to Stratemeyer, 30 Jan. 1943.
74. See ltr., Eaker to Stratemeyer, 2 Jan. 1943.
75. ORS Day Raid Rpts.; CM-OUT-4874 (1-14-43), AFDMC to CG 8th AF, A-1335, 14 Jan. 1943.
76. History, VIII AFSC, chap. v, p. 117.
77. First 1100, 11, 458; CM-IN-1368 (11-4-42), London to AGWAR, 498, 3 Dec. 1942.
78. Ltr., Spaatz to CG ETOUSA, 9 Nov. 1942.
79. History, VIII AFSC, chap. v, pp. 91 ff.
80. Ibid., pp. 151 ff., especially p. 152.
81. Report of Lt. Gen. Ira C. Eaker on USAAF Activities in the U. K. Covering Period from Feb. 20, 1942, to Dec. 31, 1943 (hereinafter cited as Eaker Rpt.), Tab E, Exh.3.
82. ORS Day Raid Rpts., passim. See also History, VIII AFSC, chap. v, p. 108.
83. Ltr., Eaker to Stratemeyer, 2 Jan. 1943.
84. Ibid.; ORS Day Raid Rpts., passim ; Eaker Rpt., Tab E, Exh. 3.
85. See cables, 8th AF; CM-IN-4417 (11-11-42), London to AGWAR, S, AFN 307A, 10 Nov. 1942; Eaker Rpt., Tab E, Exh.3.
86. Ltr., Eaker to Stratemeyer. 2 Jan. 1943; CM-IN-10920 (1-24-43), USAWW to AGWAR, 516, 23 Jan. 1943; ORS Memo 13, 11 Jan. 1943; Eaker Rpt., Tab E, Exh.3.
87. ORS Day Raid Rpt. 27A; History, 2d Bomb. Wing, pp. 35-36.
88. Opns. Anal. Sec., Hq. 8th AF, An Evaluation of Defensive Measures Taken to Protect Heavy Bombers from Loss and Damage, Nov. 1944, p. 3.
89. See History, 1st Bomb. Wing.
90. Evaluation of Defensive Measures, p. 36; First 1100, p. 395.
91. ORS Day Raid Rpt. 15; ltr., Eaker to CG 8th AF, 23 Oct. 1942, in Tac. Mission Folder 15; VIII BC Weekly Digest of Enemy Activity, 28 Oct. 1942.
92. Hq. USSAFE Air Staff Post Hostilities Intel. Requirements on German Air Force Tactical Employment, Sec. IV C, Fighter Operations of the GAF, Volume II, App. XLII; CM-IN-10858 (11-25-42), USAWW to AGWAR, 239, 25 Nov. 1942; interview with Brig. Gen. H. S. Hansell, 9 Aug. 1943.
93. Evaluation of Defensive Measures, P.36.
94. Fighter Operations of the GAF, I, p. 112.
95. Ltr., Eaker to Stratemeyer, 2 Jan. 1943; 8th AF Weekly Intel. Digest, 2 Jan. 1943.
96. Evaluation of Defensive Measures, p. 94. These figures include bombers having received class E damage, which involved virtual destruction.
97. Interview with Lt. Col. E. H. Snavely, 27 July 1943.
98. CM-IN-9057 (12-21-42), London to AGWAR, 183, 20 Dec. 1942.
99. CM-OUT-1338 (1-5-43), AFRDB to CG 8th AF, A-1260, 4 Jan. 1943.
100. Hansell interview; History, VIII AFSC, chap. v, pp. 118-19.
101. Evaluation of Defensive Measures, pp. 83, 89; interview with Col. L. C. Craigie, 12 Jan. 1943.
102. Hansell interview; History, 2d Bomb. Wing, p. 41; interview with Col. F. Armstrong, 24 Nov. 1942.
103. Armstrong interview. See also Evaluation of Defensive Measures, pp. 43-46; Tactical Doctrine, 1st Bomb. Wing, 1 Feb. 1943.
104. Tac. Doctrine, 1st Bomb. Wing, 1 Feb. 1943.
105. Evaluation of Defensive Measures, P.46.
106. Ltr., Eaker to Arnold, 11 Jan. 1943; Armstrong interview; interview with Capt. H. L. Stouse, 10 Jan. 1943.
107. Ltr., Eaker to Arnold, 11 Jan. 1943; Craigie interview.
108. Paper sgd. Brig. Gen. A. J. Lyon, 8 Aug. 1942; Craigie interview; interview with Col. P. W. Tibbets, 20 Feb. 1943; rpt. by Col. I. C. Eaker to C/AAF on trip to England, 30 Aug. to 1 Oct. 1941.
109. ORS Day Raid Rpts., passim; Evaluation of Defensive Measures, p. 94.
110. Tac. Doctrine, 1st Bomb. Wing, 1 Feb. 1943; Armstrong interview.
111. Armstrong interview.
113. Interview with Lt. Col. Leslie H. Arps, Dir., Opns. Anal. Div., 25 July 1945. Cf. CM-IN-6635 (12-16-42), London to AGWAR, 984, 15 Dec. 1942.
114. Interview with Lt. Col. Carl Norcross, 5 Jan. 1943; Craigie interview.
115. ORS Day Raid Rpts., passim; ORS, VIII BC, Preliminary Report on Bombing Accuracy, 4 Jan. 1943.
116. Prelim. Rpt. on Bomb. Accuracy.
118. Report on Perspective Maps, 31 May 1943.
119. Prelim. Rpt. on Bomb. Accuracy.
120. Craigie interview.
121. Eaker Rpt., p. 4.
122. First 1100, pp. 395-6.
123. Craigie interview; ltr., Stratemeyer to Spaatz, 24 Dec. 1942.
124. Prelim. Rpt. on Bomb. Accuracy.
125. ORS Report on Advantage of Dropping on the Leader over Sighting Individually for Range When Bombing in Formation, 20 Mar. 1943, in Opns. Anal. Div. Rpts., Volume I, No. 2.
126. Prelim. Rpt. on Bomb. Accuracy.
127. Tac. Doctrine, 1st Bomb. Wing, 1 Feb. 1943.