In essence, the business of logistics consisted of four parts: to identify in advance the needs of the theatre and provide for them; to develop a system that could and would respond in good time to the demands of the fighting formations; to ship what was required to the Continent, whether by ocean-going ship or coaster; and to deliver from ports or beaches what was needed where and when it was needed. Each of these steps posed its own problems, and the failure to solve each one before the autumn would have a negative effect on the campaign.1
By and large, the Allies did a good job of predicting the needs of the armies and air forces in northwest Europe. However, there was a general failure to anticipate the nature and thus logistic requirements of bocage fighting, a problem of particular concern to the Americans. Consequently, shortages developed as early as June. The infantry wanted more mortars, grenade launchers, light and submachine guns, and bazookas in particular. It also became apparent that the army had failed its armor branch by not pushing the earlier introduction of the M-26 Pershing or upgrading at least some Shermans with the 90mm gun (as the British had with their 17-pounders), thus condemning tankers to combat against often superior tanks. Many ammunition natures, particularly 105mm and 155mm howitzer and 81mm mortar rounds, were also expended at higher than predicted rates (often because of unjustifiably profligate usage, as even First Army’s chief of staff admitted). Occasional periods of poorly enforced rationing and appeals for self-restraint failed to change behavior significantly, so strict rationing was imposed in mid-July to build up stocks for Operation Cobra.2 The success of the breakthrough and transition to pursuit eased most weapon and ammunition shortages, but only for a time. The problem then became one of fuel supply, although this was mostly an issue of moving abundant fuel stocks. The ammunition problem then resurfaced in early autumn. Moreover, it became clear that the War Department, having cut back procurement a year before in response to congressional perceptions of waste, was not providing adequate quantities of some types of ammunition—a problem compounded by COMZ’s deficiencies. The sudden transition to a war of movement revealed another shortage—motor transport—which would eventually bring operations almost to a halt. Inadequate provision for a switch from semistatic to fast-moving operations was due partly to the War Department’s failure to procure adequate numbers of heavier trucks. But it could also be blamed on the field commanders overriding their logistic planners and according too low a priority to the buildup of truck fleets in the first two months.
The British and Canadians were generally conditioned to expect a less generous provision of materiel and supplies (and, perhaps in consequence, were better at stock control and accounting and the avoidance of waste). They apparently made fewer importunate demands of logisticians to meet unexpected needs. They too, however, suffered from periodic ammunition shortages and restrictions on usage until well into August. And both armies suffered again from a dearth of munitions in September and October. To a large extent, this was caused by a lack of logistic lift, a problem resulting from conscious trade-offs in the planning process and greatly exacerbated by the discovery that 1,400 newly procured British three-ton trucks suffered from engine defects (which afflicted replacement engines as well).
The major failing of each of the Allies was the inadequate provision of infantry replacements. Each army faced a manpower crisis sooner or later. For the British, it was sooner. As a result of drawing faulty lessons from the North African campaign, the War Office seriously underestimated the infantry’s likely casualty rate (by almost half) while simultaneously overestimating that of the other arms. Even without this miscalculation, the British had recognized a year before D-Day that they would likely face a shortage of 35,000 infantrymen by the end of September. This estimate proved to be only slightly too pessimistic. As forecast, to keep other formations up to strength, 21 Army Group had to disband two divisions (one in August and one in November) and the equivalent of two armored brigades. The Canadian army likewise suffered from shortages from August until mid-November (when a static role on a passive sector led to falling loss rates). The US War Department (and the ETO) were also far off in the prediction of infantry loss rates, especially riflemen. By the end of July, the replacement pool was almost dry of infantrymen. The transition from attritional to maneuver warfare eased the problem until November, when a full-scale manpower crisis blew up with the return of attritional warfare. This was not entirely the War Department’s fault; as it pointed out, the theatre was 70,000 men overstrength, of which 32,000 were overhead, and it could and should have been retraining some of the surplus to provide the needed riflemen.3 In the event, all three Allies mitigated the shortage by combing HQ and administrative units for potential infantry, retraining surplus air defense gunners and personnel from other overmanned combat units, and speeding up the flow of replacements. The results were not entirely satisfactory, as they produced large numbers of poorly trained and resentful riflemen. The British deficiencies also exerted a baneful effect on operational planning and execution that went beyond the disbandment of formations; commanders were so concerned with minimizing casualties that they became overcautious and sometimes missed opportunities.
While the macro-needs of the land forces were, for the most part, adequately anticipated and met, the same cannot be said for more detailed theatre plans. There was something of a logical disconnect between the structure of the Allied armies and their operational planning prior to the invasion. Although there was great emphasis on armor and mobility in their organizational development, suggesting a preparedness and a desire for maneuver warfare, the armies assumed that expansion of the lodgment to the Seine and the Loire and through Brittany would progress in a steady, predictable fashion as the enemy conducted an orderly withdrawal from one obstacle to the next. Once the Seine was reached, there would be a month-long operational pause while ports were opened, rail and pipelines extended, and more supplies and motor transport landed to establish the forward dumps necessary to support a further advance. The subsequent march on Germany was expected to take place in a similarly predictable fashion. In essence, the offensive would resemble that of the Allies in 1918 rather than the German blitzkrieg. The logistic system was configured for this essentially linear, even-paced development. To take the American model as an example, COMZ was supposed to receive supplies into the theatre (Normandy); transfer them to the appropriate dumps, duly sorted and inventoried; and forward them, on demand, through its various sections to the army rear boundary. Army would then distribute them to the divisions. The plan anticipated the creation by D+41 of an adequate reserve of fourteen days of supply for all items save ammunition, of which there would be five units of fire.4 By D+90, reserves would be built up to twenty-one days of supply for most items and five units of fire. Army levels would be maintained throughout at seven days of supply and seven units of fire. (Of course, the number of divisions in France would go up in the interim, from fifteen to twenty-one.) To achieve such levels of reserves, supplies had to be landed at a rate exceeding current demand by at least 50 percent. That is, given that a division slice initially required as much as 800 tons per day, upward of 1,200 tons would have to be landed daily—a figure excluding the air forces’ needs, coal, civil affairs supplies, preshipped equipment, materials for railway and bridge repair and pipeline construction, and other overheads. For the period up to D+41, this meant that 26,500 tons per day had to be landed. The figure would rise to approximately 45,000 tons a day by D+90.5
Operations did not develop as predicted. The D+17 phase line was only partially reached by the end of July. The Breton ports had not been taken by D+50. However, the sweeping post-Cobra advances carried the Allies eastward at such a speed that they went from the hoped-for D+20 line (south of the Sélune) to the D+90 line (the Seine and the Loire) in less than a month. While operational commanders were seizing the opportunities vouchsafed to them, the logisticians were increasingly pressed to adjust their plans and actions to reflect the new realities. When the offensive continued beyond the Seine without any pause, they progressively failed to cope with demand.
Most shortages, including those of ammunition, arose not because of inadequate foresight by the War Department and War Office or even excessive demand but because of the difficulties of delivering adequate tonnages to the Continent and then distributing supplies to the armies.6 This was partly a function of limited port capacity (exacerbated during the period of the “great storm” in June, when the beaches became unusable for a few days). In larger part, it was due to the unexpected and therefore unplanned rapidity with which the offensive developed in August–September. The pace of operations would have led to logistic overstretch sooner or later. The fact that it happened sooner and affected the US forces more acutely was to a considerable extent due to flaws in the setup of logistic management. For eighteen months before the invasion, there had been a struggle of byzantine complexity between competing jurisdictions over who and which organizations should command, control, or coordinate what aspects of the American logistic effort. Whatever Lieutenant General J. C. H. Lee’s shortcomings as a logistician (he had a peacetime quartermaster’s mentality, lacking a can-do approach, initiative, and a willingness to improvise), he was a formidable bureaucratic infighter. Even Eisenhower’s directive issued on D-Day itself did not definitively settle any issues. The Americans went into the campaign with a command and supply system that could charitably be described as confused and more unkindly described as in disarray. A variety of commands existed (SHAEF, ETO, First Army and then 12 Army Group, COMZ and its ill-defined subordinates the Forward, Advance, and Base Sections), and they often expended much time and effort working against one another.7 Institutional responsiveness to the requirements of the field forces was not among the system’s defining characteristics. COMZ was divorced from the operational chain of command. Lee was responsible only to Eisenhower. Neither Smith, the SHAEF chief of staff, nor Bradley, let alone the army commanders, could issue orders to COMZ or its parts; they could only make requests. This had a negative impact on both their authority and their ability to plan with confidence. Given Lee’s powerful mentors in Washington (including Marshall himself), Eisenhower never felt confident enough to sack him, even when he failed to deliver, or even to impose a more rational system on him. In stark contrast was the simple, straightforward British system, in which the army group’s chief logistician was directly subordinate to the operational commander, an arrangement replicated down the chain of command.
COMZ, with its squabbling subcommands, was a complex piece of machinery supposedly designed to provide supply without waste or delay. In practice, it was overly bureaucratic and inefficient, quite unsuited to cope with the friction generated by battle and unable always to get what was needed to the fighting man in good time. The flexibility, adaptability, and responsiveness to rapidly changing circumstances required by maneuver warfare were lacking. Also largely absent was foresight, a prerequisite for effective logistic support; for instance, winter arrived long before appropriate clothing was provided because COMZ was purely reactive.8
The supply system was undesirably rigid in its organization of deliveries. Thus, the prescheduling of supply shipments into France for the whole of June, July, and August resulted simultaneously in shortages and unbalanced stocks. To mitigate the consequences, selective unloading was instituted to meet current priorities. This resulted in shipping lying idle, increasing turnaround times, and it meant that, at any given moment, a high proportion of theatre supplies was stuck in the pipeline. For example, the priority accorded to ammunition in July to alleviate shortages created deficiencies in other areas, such as replacement tanks and quartermaster and engineer stores. The inflexibility inherent in the system was alleviated to some extent by special express shipping services, sometimes by air, to meet unpredicted requirements.9 The disposition of goods once they were landed was another problem. With every sort of supply competing for scant space on and near restricted docks and beaches, those responsible for handling the supplies were sometimes overwhelmed. Stocks were piled up wherever space could be found, and record keeping was often scant. Lacking a proper system of inventory and control, supply officers lost track of stores and found it easier to put in a fresh demand than to hunt for the items required. Uncertainty about depot balances, or even contents, led to massive waste; supplies that are un-sorted, unrecorded, and unlocatable might as well not have been shipped in the first place. Nor did COMZ maintain close liaison with operations staffs to ensure the intelligent anticipation of needs; it was purely reactive. Moreover, the indenting system was insensitive to urgent demands and became chaotic as requisitions were resubmitted (often several times) when not honored; the problem was frequently compounded by the practice of substituting “filler” tonnage, things not requested, for the required items (both suppliers and movers were often satisfied with achieving a quantitative target, regardless of expressed needs). Waste, often incurred as a consequence of haste, was endemic. For instance, by mid-October, more than 3.5 million jerry cans supplied to the Americans in theatre and essential for the delivery of fuel from bulk storage and pipe heads to the front had been discarded, lost, or stolen, contributing significantly to the armies’ fuel supply crisis. Unquantifiable but substantial amounts of POL and other attractive items disappeared into the chaos of the resurrected French railway system (both individual cars and whole trainloads) or into the insatiable black market.
US field forces prided themselves on their casual, informal approach to supply. Administration was generally seen as pedantry, producing delays and interfering with the essential business of fighting. In their disdain for records or formal authorizations, they contributed significantly to the disorder and waste that increasingly characterized the system. When the armies advanced and turned over their rear supply dumps, COMZ usually inherited a mess; supplies were scattered around with little or no supporting paperwork. The armies were also guilty of ultimately self-defeating short-termism in their resort to “moonlight requisitioning,” hoarding, overrequisitioning, cannibalization, and subversion of the system through bribery, barter, and outright theft. In particular, their commanders’ habit of subordinating logistic considerations to operational aspirations was bound to overstretch COMZ sooner or later. Partly because of the extraordinary demands placed on it, and partly because of its structural and command deficiencies, the crunch point came earlier than it might have. Though systemically flawed, the logistic machine managed to function for some time with remarkable if steadily declining effectiveness, working imperfectly and staggering from crisis to crisis. This was made possible by the goodwill, ingenuity, and initiative of some senior and many mid- and junior-level officers who were determined to provide for the fighting formations in difficult circumstances.
US Army logisticians calculated that, to be fully effective in the offensive, fighting formations should be able to count on a daily supply of roughly 650 tons per divisional slice, with another 4,000 to 5,000 tons going to Ninth Air Force.10 By the beginning of September, the planners were counting on having, in addition to the Normandy beaches, Mulberry A (which was actually destroyed in a storm after only three days of operation), Cherbourg, and the Breton ports, principally Brest, Lorient, and Quiberon Bay. This would provide a daily discharge capacity of about 46,000 tons, including an expected 25,000 tons from the beaches and Cherbourg. This tonnage was calculated to be enough, after a portion was set aside for the British, for the maintenance of twenty-one divisions, the buildup of the reserves required for a sustained offensive, and the support of the air arm. It was further calculated that discharge capacity would be sufficient to support the planned buildup as far as the beginning of September (D+90), but only by a very narrow margin, and only if the ports were taken more or less as forecast (including St. Malo on 1 July, Quiberon Bay by 16 July, and Lorient and Brest by 26 July). However, by the end of September, capacity was expected to start to fall short of requirements, with a serious deficiency continuing for several months. Brittany’s perceived centrality to the success of Overlord is clear from the fact that it was expected to provide the bulk of supplies by the time the autumn weather circumscribed beach activity, with Brest being the point of entry for reinforcing formations and Quiberon Bay alone discharging 10,000 tons per day.11 No wonder Eisenhower wrote to Montgomery in late July: “We must get the Brittany peninsula. From an administrative point of view that is essential. We must not only have the Brittany peninsula—we must have it quickly. So we must hit with everything.”12
These expectations were not realized. In June and July, thanks to the late opening of Cherbourg, only 62 percent of planned discharges had actually been achieved, limiting the buildup of stocks; however, the much slower than anticipated rate of advance mitigated the effects initially. Thereafter, the plan to open the Breton ports as the main source of supply was progressively jettisoned. St. Malo was taken on 18 August and Brest on 19 September, but each was so comprehensively wrecked and blocked that neither was ever used. Neither Quiberon Bay nor Lorient was ever developed (the latter was never captured, and although the former had been taken on 16 July, no effort was made to seize Belle Isle, whose batteries controlled the entrance). Thus, the over 20,000 tons expected to flow from Brittany by the end of September (and 30,000 by early November) did not materialize; the 4,640 tons coming into minor, inefficient ports was an inadequate replacement, bringing the total discharges at the beginning of October to only 26,000 tons per day from all sources. To the logisticians, the upset to their plans was greater because St. Malo, Brest, Lorient, and Quiberon Bay were all deep-water ports and anchorages, and opening additional shallow-draft capabilities was pointless, as all available coasters were already in use. The port situation also contributed directly to the ammunition crisis that dominated operations in October.13 The Canadian army took Rouen on 30 August and Le Havre (a deep-water port) on 12 September, and SHAEF almost immediately allocated these ports to 12 Army Group. Like Cherbourg, St. Malo, and Brest, however, they had been thoroughly demolished. They were not open for business until October, when they discharged close to 90,000 tons between them; they did not take on major importance until November, when they handled almost 293,000 tons. Thus, at the beginning of October, with adverse weather disrupting or preventing unloading over the beaches, the outlook for the landing of supplies was bleak. Discharges in October averaged only around 25,000 tons per day, compared with an estimated requirement of at least 38,500 tons and a planned 49,750.14 This quantity was grossly inadequate to meet the army’s needs, and the rate of discharge was too slow to clear shipping as it arrived. By early September, a backlog of over 100 deep-draught Liberty ships (each with a cargo capacity of almost 11,000 tons) was awaiting an opportunity to unload. Until then, they acted as floating storage, a waste of shipping space the system could ill afford; the only alternative to lying idle was being forced into British ports for time-wasting transloading into coasters as these became available.
There was an additional complication to calculations of port discharge rates. Supplies had to be not only landed but also cleared. Cherbourg, the Americans’ only major port until Le Havre opened in the autumn, enjoyed only limited access by road and rail (and the same limitations applied to most smaller ports). Creating mountains of supplies at the dockside was arguably worse than nondelivery, as it led to further constriction, not to mention holdings of inadequately documented and therefore essentially unusable stocks. Moreover, as the demands for motor transport to move supplies to the armies escalated, trucks were taken from clearance duty, further limiting capacity.
British and Canadian requirements were proportionate to that of the Americans; fourteen divisions required 9,100 tons per day. Initially, they lacked a significant port like Cherbourg, but their mulberrry harbor was operating smoothly. By August, the average daily discharge over their beaches and through the artificial harbor was around 16,000 tons; this was sufficient to create fourteen days of reserves plus fourteen days of working margins or equivalent stocks of all commodities held in the rear maintenance area (RMA). However, the projected thrust northward would take the armies far from the RMA in Normandy, and the major obstacle of the unbridged Seine would complicate and slow the flow of supplies. On 30 August the weighty decision was made to rely on the early capture and opening of at least one Channel port and meanwhile reduce the amount of supplies and vehicles imported into Normandy to 7,000 tons per day. The first post-Normandy task of First Canadian Army was to capture the Channel ports, defended by substantial elements of Fifteenth Army, to support 21 Army Group’s advance. But the big prize would be Antwerp, the answer to all future capacity problems.
As the Americans had found at Cherbourg, the quick capture of a port did not necessarily mean a quick opening. This was, however, achieved at Dieppe, which began receiving vessels on 7 September and quickly reached a daily capacity of 6,000 to 7,000 tons. With Dieppe in operation and Boulogne about to fall, the decision to rely on ports north of the Seine to bring in sufficient supplies to meet 21 Army Group’s needs seemed justified. Both ports were much nearer the point of main effort and, being north of the Seine, enjoyed rail links northward that were only lightly damaged by the air interdiction campaign; they also joined with the Belgian system, which was untouched. However, Boulogne was not surrendered until 22 September and could not be opened until three weeks after its capture; even then, it could accept only coasters landing about 2,200 tons. It could not take substantial cargoes until two larger, deep-water quays were opened, and these were not discharging a respectable 11,000 tons daily until November. Ostend was receiving a mere 1,000 tons per day by the end of September (although it was also an entry point for bulk POL), and Calais could receive only landing ships tank (LSTs) and personnel.15 During the Market Garden operation, the army group still depended, but to a decreasing extent, on supplies trucked north from the RMA in Normandy; it would not enter the autumn campaign with enough supplies to meet all the needs of both its armies, never mind the stores and other items required to cope with a backlog of repairs and the replacement of defective equipment.
The US port situation has been dwelled on at greater length because the Americans faced the more challenging problem. Their component of the Allied force was also larger and set to grow larger still; six more divisions would land during September and October (all, with Brest unavailable, using discharge facilities at the expense of supplies). The American beaches were more exposed to autumnal gales and their mulberry was already wrecked, so the need for compensatory port capacity was growing more urgent. And by the beginning of September, the supply line from their only major port, Cherbourg, was twice the length of the British line; another port nearer to the front would save scarce motor transport (one argument against developing the south Breton ports in favor of Le Havre was that the latter would save the equivalent of seventy truck companies for every 5,000 tons landed).16 In addition, Marseille was expected to become another entry port for Third Army supply, although there would be no surplus capacity from the maintenance of 6 Army Group until late November. But what the Americans needed every bit as much as the British was Antwerp, captured essentially intact on 4 September and capable of handling an initial 40,000 tons daily (coincidentally, the logisticians’ estimated daily requirement by early September), to be shared on a 1.3:1 American-British ratio. The port also had massive storage capacity for bulk POL; excellent clearance facilities; road, rail, and canal links with the interior; and good inland communications. In addition, it was a much shorter haul from Antwerp to the front than from Cherbourg, let alone Brest.17 That is why Admiral Ramsay, the naval C-in-C, signaled to Eisenhower and Montgomery on the day the port was captured: “It is essential if Antwerp and Rotterdam are to be opened quickly enemy must be prevented from: carrying out demolitions and blocking ports; mining and blocking the Scheldt and the new waterway between Rotterdam and the Hook; both Antwerp and Rotterdam are highly vulnerable to mining and blocking . . .; it will be necessary for coastal batteries to be captured before approach channels to the river routes can be established.”18
Before Operation Cobra, the Allied lodgment was so small and the lines of communication were so short that delivery of POL and timely resupply were not problems for the transportation at hand (provided the items were available, which in July was increasingly not the case with ammunition).19 Nor, given its forecasted pace of operations, did SHAEF anticipate a sudden and dramatic change. It was expected that the progress of port opening in Brittany, the repair of railways and bridges damaged by air interdiction, the construction of pipelines, and the buildup of road transport (including much more heavy lift for lines of communication work) would ensure uninterrupted supply to the armies. It was not to be like that. From the beginning of August to the beginning of September, the armies advanced from the D+20 line to that anticipated for around D+240, distances of 550 to 650 km (340 to 400 miles). While the commanders were delighted, the logisticians were appalled.
It was always intended that railways would take care of most long-distance hauling; a single train could shift more than 1,000 tons—400 2.5-ton truckloads, or the lift of eight to nine light truck companies. The aim was to provide at least one double-track line from the rear base and RMA to the rear boundary of each army. This would mean that army transport, collecting from the theatre’s forward dumps, would have to haul loads only over the limited distances for which their transport was scaled. Of course, for the first two months of the campaign, the bridgehead was so constricted that the only rail rehabilitation that took place was of the yards at Cherbourg (port clearance was the first priority) and then of a line to Carentan, 50 km (30 miles) southward, hardly an economic distance for rail operations. With Third Army’s spectacular advance following Cobra, though, pressure was on COMZ’s engineers to abandon carefully prepared plans for lines into Brittany and to make rapid extension to the east their main effort. The five general service engineer regiments originally allocated were reinforced by another four; by the end of August, more than 18,000 men (including prisoners of war) were working on reconstructing the rail lines. They made impressive progress, restoring the marshaling yards at St. Lô (which had been so obliterated by air attacks that the engineers had to obtain the plans from SNCF, the French national rail company, to figure out what went where) and then, by 17 August, to Le Mans. By 27 August, a single-track line had been pushed forward to Chartres; by 30 August, despite the considerable damage inflicted by previous Allied interdiction attacks, it reached just west of Paris. But by that time, the American armies were already about 400 km (250 miles) further on, and the absence of bridges over the Seine and the Paris bottleneck slowed progress. Work was commenced on the only lightly damaged lines east of the city and reached Liège and Verdun by mid-September. However, rail capacity was limited by several problems. Until reconstruction of the devastated Parisian marshaling yards and the Seine bridges was completed, supplies had to be trucked from railheads at Chartres and Dreux to the other side of Paris before being reloaded onto trains. Parts of the system remained single-track for some time. Reliability was limited by several factors: the temporary nature of some urgent repairs, the meager quantities and worn condition of locomotives and rolling stock after four years of occupation and air attack (eventually, these were augmented through Cherbourg), the lack of fully functioning infrastructure (e.g., communications, signaling), and the time required to create an efficient movement staff.
The British experience, thanks partly to geography, was somewhat easier. By 1 September, a single line had been opened as far as Argentan, but operating it was complicated by limited siding and loading point capacity and a shortage of engines. A day or two later, a through rail route was discovered nearly intact between Beauvais and Brussels, although the 280 km (175 miles) included a 65 km (40 mile) stretch of single line working north of Amiens. A link between the RMA in Normandy and Beauvais had to wait for the completion of a single-line Seine bridge at Le Manoir on 22 September, but by the end of the month, there was a continuous connection between the RMA and Brussels. Even before that, on 6 September—the day before the first coaster discharged in the port—Dieppe was linked to the main route, and upward of 2,800 tons per day were being moved right up to the Brussels area. By 6 October, the railhead had been pushed forward to Eindhoven. The shortage of locomotives and rolling stock was not resolved until the opening of the Dieppe rail ferry on 29 September; by 7 October, SHAEF had met 80 percent of British requests for both. Until then, Cherbourg had been their only point of entry, and the Seine rail break prevented their import north of the river. Inefficiency in operating the system, similar to that in the US area, was also a limiting factor, and the service was unreliable up to the end of September.
Despite the endeavors of the engineers and railway operators, rail transport was beset by too many problems to contribute significantly to the supply of the armies in the first crucial weeks of September. Even in the middle of the month, it was only beginning to carry the weight the logistic theoreticians had intended—6,000 to 7,000 tons per day as far as Paris, and 5,000 to 6,000 from the east side of the city to the army rear. Moreover, the Americans were obliged to use scarce capacity to provide 1,500 tons of food and coal each day to meet the needs of the people of Paris and to move engineer stores for bridge and rail rehabilitation and pipeline construction. It was not until November that rail managed to carry half the tonnage required by the armies.
From July, most fuel was delivered to the Continent by tanker, although from mid-August, the submarine pipeline PLUTO was also pumping it ashore at Cherbourg. Having decanted POL into jerry cans for delivery to the troops, supplying units by vehicle was not a problem as long as the bridgehead was constricted. Indeed, reserves grew to very comfortable levels. The exploitation of the Cobra breakthrough changed all this. By mid-August, the increased distance from depots to American spearheads meant that twice as much time was needed by a greater number of trucks—at a time when demand showed every sign of growing exponentially. It was time to press forward with a more economical solution: delivery as far forward as possible in bulk, by pipeline. Three general service engineer regiments and other elements (almost 9,000 men, including prisoners) were put to work to construct a triple line (two for POL and one for aviation gasoline). By the end of the month, the pipe head had reached Alençon, and over the following fortnight, a single POL line was pushed forward beyond Chartres to within about 30 km (18 miles) of the Seine. At that stage, further progress was dramatically slowed as COMZ ended the brief, ten-day-long priority given to moving pipeline materials by rail and took truck units away to meet the immediate needs of the armies at the expense of long-term savings. The decision was made to concentrate on building shorter lines from Le Havre and Antwerp (in the event, purely local pipelines were built from Le Havre and Rouen, and although a line was constructed from Antwerp to Maastricht, it was not completed until December). The pipe head of the line from Cherbourg did not reach the southeastern outskirts of Paris until 6 October.
British construction, too, lagged far behind, reaching 30 km (18 miles) northeast of Rouen during September. Mid-October saw the opening of a PLUTO terminal near Boulogne, but bad weather delayed the start of pumping until 27 October. That month, too, a line was constructed between Ostend and Ghent, but not until December was another put into operation from Antwerp to Eindhoven. Ample bulk POL installations had been created in Boulogne and Ostend and were inherited in Antwerp to support operations north of the Seine, but the fighting formations needed packaged fuel. Similar to the Americans’ situation, a growing shortage of jerry cans and longer vehicle turnaround times were limiting factors.
Although the pipelines cut the demand for motor transport (except, of course, the trucks supporting construction), their performance was not an unalloyed success. Much of the pipe laying was done by troops who were undertrained for the task, leading to poor workmanship. In addition, most of the pipeline was laid along the sides of hard-surface roads; leaks and breakage from traffic accidents were frequent until the engineers learned to reduce the latter by laying pipes on the other side of roadside hedgerows. There was also pilferage via deliberate punctures to feed the black market. Communications between pumping stations, let alone tank farms, to control use and report failures were also inadequate. Aspirations to move the pipe heads close to army dumps fell far short of achievement until winter had arrived. All these problems resulted in interrupted supply, increased the turnaround time for transport units, and aggravated the fuel shortage at the most critical period of the pursuit.
Before the invasion, SHAEF’s refusal to provide COMZ with the 240 truck companies (averaging forty-five to fifty trucks) it had requested and allocating shipping space for only 160 seemed to entail no undue risk. In the event, the immediate needs of the restricted bridgehead and the priority accorded to the delivery of other items resulted in only 94 of those companies being in France by the time of the breakout. Moreover, the Transportation Corps sought a ratio of 2:1 between heavy and light trucks (i.e., between 6-, 10-, and 12-ton semitrailers and 2.5-tonners), but the War Department failed to provide the former in adequate quantities, and the actual ratio was 1:2.5. If operations had unfolded more or less according to forecasts, this would not have mattered. Railways and pipelines would have shifted the vast bulk of supplies from the Normandy base area to Advanced Section railheads, and motor transport would not be required to haul supplies more than 250 km (155 miles) to the army rear. The unanticipated successes of August made nonsense of these calculations. Transport provision had not been scaled for far-reaching mobile operations with longer journeys and longer turnaround times. With the aid of 300 to 360 British vehicles borrowed for a month, the Americans coped for the first three weeks. But the decision to move in strength beyond the Seine without a pause to establish intermediate supply dumps between the Normandy base area and the armies and to acquire sufficient additional transport required hasty improvisation to meet rapidly growing demand.
COMZ set up the Red Ball Express, a one-way loop highway for operation twenty-four hours a day to be used exclusively by its own fast through traffic. Starting on 25 August, the original highway ran from St. Lô through Alençon to Chartres and returned via Mayenne. Initially, 118 truck companies were employed, but that number, augmented by reinforcements, quickly grew to a peak of 132 by 29 August. They moved 89,000 tons to the dumps in the Dreux–Chartres–La Loupe service area by 5 September (the date the original mission was considered completed). Averaging this out over the twelve days, that amounted to around 7,400 tons per day. In the same period, 70 trains carried slightly more than 30,000 tons as far as Chartres, averaging 5,000 tons a day by the beginning of September. These quantities were theoretically enough to provide for the immediate needs of the sixteen divisions involved in the pursuit, cover air force requirements, and deliver forward materials for railway repair and pipe laying, but not enough to create any intermediate dumps or reserves. In practice, the unpredictability of deliveries and the unfortunate fact that they often failed to include the most needed items often left formations critically short of current requirements. And of course, by 5 September, army-level transport had a very long haul to fetch supplies from the dumps and get them to the divisions; for instance, First Army was soon trucking supplies 480 km (300 miles) from west of Chartres all the way to Liège. For this reason, armies formed their own provisional truck companies with vehicles from grounded air defense, medium and heavy artillery, engineer, and other units. They also tried to require COMZ convoys to deliver east of the Seine. Even divisions felt compelled to send their own trucks in search of supplies; for instance, trucks from 1 Infantry Division made two round-trips of 1,120 km (695 miles) each.
COMZ had to extend the operation of the Red Ball Express for an indefinite period, and the round-trip became much longer as well. By 20 September, the outward run was from St. Lô, through Versailles, and around Paris to Soissons (still 300 km [185 miles] short of Liège) in support of First Army. The route branched off from Versailles through Melun and on to Sommesous (200 km [125 miles] west of Metz and 160 km [100 miles] from Nancy) to supply Third Army. Each round-trip was approximately 1,600 km (990 miles), meaning a five-day turnaround time.20 Because transport resources were overstretched, COMZ had to inform the armies that they would each receive only 3,500 tons per day for the foreseeable future. But even that amount—enough for five divisions, or enough for seven if the opposition was very slight—could not be achieved. By midmonth, approximately 45,000 tons in total had been delivered, not even two-thirds of the already inadequate allocation.21 Coincidentally, Lee chose the first fortnight in September, when the transport shortage was most acute, to move his ultimately 29,000-strong HQ from its hutted accommodation in Normandy about 450 km (280 miles) to more luxurious quarters in 167 Parisian hotels.
The situation began to improve in the second half of September with the rehabilitation of the railways. A driblet of supplies started to run on lines from east of Paris on 7 September, growing to 5,000 to 6,000 tons per day as the month wore on. When there was sufficient capacity to move 2,000 tons per day to each army, it was decided that most Red Ball convoys should terminate east of Paris, where cargoes would be transferred to railcars. This would save on motor transport without penalizing the fighting formations. In all, the Red Ball Express lasted another seventy-two days, until 16 November. It transported in excess of 315,200 tons to forward depots or to Paris for transfer to railway trains. And from 6 October it was supplemented by the White Ball Express, running the somewhat shorter route from newly opened Le Havre to Reims. However, the period of greatest need, when resistance started to harden and pursuit ceased to result in easy gains, was the period when supply was at its most disorganized and inadequate.
To find the number of trucks required by the Red Ball Express (and, later, the Red Lion), COMZ reduced base area and port and beach clearance activity by 50 percent, thus adding to the backlog of coasters waiting to be unloaded and the proportion of supplies stuck in the pipeline. This reduction in activity also contributed to growing confusion as to what could be found in which dump as an increasingly harassed Normandy Base Section attempted to make space at quays by depositing supplies wherever it could find room. COMZ and the armies also formed provisional transport companies by stripping out vehicles from many service and combat units (e.g., maintenance, engineer bridging, corps artillery) and from three immobilized infantry divisions. When the pace of the pursuit started to slacken as resistance stiffened in mid-September, many of these formations and units would be missed. Despite all these efforts to find more trucks, their numbers actually started to decline as accidents, wear and tear, and lack of maintenance took their toll; by the first week in September, only 115 companies were running, almost 15 percent fewer than just a week before. Nevertheless, the improvised system was fairly effective in keeping the armies advancing with its “only just in time” deliveries, but it was not efficient. Deliveries were unpredictable, making even short-term planning somewhat speculative, and they often fell short of demand. Both armies, the Third in particular, resorted to the shortsighted expedient of hijacking convoys meant for others and even depriving COMZ trucks of fuel for their return journey; COMZ became understandably reluctant to send its vehicles into Third Army’s rear. Moreover, the Express was becoming increasingly uneconomical, consuming enormous quantities of fuel—300,000 gallons (815 tons) per day, equivalent to two-thirds of Third Army’s daily demand at the end of August. It was also becoming increasingly creaky and unsustainable in its demands on both troops and their vehicles.
Success came at a heavy cost to COMZ and, ultimately, to the field forces. Lack of maintenance took its toll as grossly overloaded vehicles were run into the ground and spare parts, tires, and tools ran short; driver exhaustion led to a significant rise in the rate of accidents (and minor sabotage by drivers seeking a rest was not unknown, as was the selling of desirable loads on the black market). The number of major repairs required rose from 2,500 in mid- September to 5,750 by the end of the month. This was more than the repair facilities could cope with and denuded them of spares, with debilitating effects that lingered for months. For instance, when Ninth Army received five truck companies relieved from Red Ball duty, it found that only 60 percent of the vehicles were in operable condition. The operation of the system, improvised as it was, left much to be desired, especially in its early stages. Coordination among the various sections of COMZ and with the armies was poor, and the latter’s unilateral ordering of convoys further forward than planned did not help matters. Loading and unloading of convoys were unduly time-consuming (11.5 to 39 hours), thanks to the scattered location of dumps and the difficulty of finding items within them. Traffic control suffered from a shortage of military police and efficient communications, which led to time- and fuel-wasting encroachments by other units onto Red Ball routes. Sometimes, as journeys lengthened and the operational situation changed, convoys could not find army dumps and had to travel an extra 80 to 150 km (50 to 90 miles) to hawk their loads around, or they were diverted to divisional supply points over unfamiliar routes. All this substantially lengthened turnaround times.
The 21 Army Group experience was in many ways similar to that of 12 Army Group, though somewhat less fraught. The American forces were much more widely spread. The British established rail links sooner than their ally as well. There was the same heavy initial reliance on road transport, and the plan to establish road heads 80 to 160 km (50 to 100 miles) apart and stock them with up to five days’ holdings of all stores immediately fell victim to the pace of the advance. Conversely, at 510 km (315 miles), the distance from the RMA to Brussels was much shorter than from the Normandy base to Liège or Metz; with the opening of Dieppe, 340 km (210 miles) distant, the problem was further eased. Though this was still double the distance used to calculate the allocation of transport, considerable reinforcements were provided to mitigate the problem. During August, Second Army increased its number of general transport companies (each with 90 to 120 trucks) from six to fourteen, and a tank transporter company was converted for load carrying as well. Like the Americans, the British needed even more; accordingly, imports over the beaches were limited, two divisions of VIII Corps and artillery and engineer units were grounded to free up transport for Second Army, and more tank transporters were pressed into service to carry supplies. After the fall of Le Havre, I Corps was also stripped of its trucks to reinforce supply efforts for the Canadians. Up to the first week in September, neither the British nor Canadian armies were seriously discommoded by a lack of transport—albeit at the cost of immobilizing substantial combat forces that would soon be needed. However, the decision to mount Operation Market Garden still threatened overstretch. From the reserves, 1,820 trucks and 154 extra POL tankers were added, and an additional seventeen general transport companies were provided on permanent loan. Unfortunately, 1,400 new trucks were found to be inoperable because of faulty engines, resulting in a transport deficiency that necessitated reliance on American help for Market Garden. On Eisenhower’s command, the Red Lion Express was set up to haul 500 tons per day, mostly POL, from Bayeux to Brussels. The Red Lion ran from 16 September to 12 October; about half the amount delivered went toward the maintenance of the two US airborne divisions involved, at least partly refuting American claims that American needs were being subordinated to those of the British.
Air was routinely used to supplement the trucking effort from mid-August. But between 29 August and 3 September, First Allied Airborne Army successfully argued for the return of its aircraft (i.e., IX Troop Carrier Command and 38 and 46 Groups, RAF); after 3 September, Brereton, the army commander, had to relinquish half the force but wrested the aircraft back on the fourteenth for Operation Market. Between 20 August and 16 September, air transport (including 200 unsatisfactorily converted bombers) delivered 12,800 tons to 12 Army Group, just under 7,800 tons to 21 Army Group, and 2,650 tons of emergency supplies to Paris (the last an unanticipated drain on scarce resources that tied up land transport for some time as well). Toward the end of the month, some air transport operations in support of First and Third Armies were resumed, but only until the first week in October, when weather largely shut them down. Given the limited lift (3 tons) of the standard transport aircraft, the C-47, air transport could never be more than a palliative (save for the fast delivery of certain high-value items). As Bradley saw it, diverting the vast bulk of Troop Carrier Command to airborne operations in support of 21 Army Group also made availability unpredictable, as did the weather; these factors not only reduced the flow but also inhibited planning. Moreover, poor planning characterized air resupply. Requests were often duplicated, and motor transport was frequently delivered to the wrong airfields or was not waiting to collect at the other end. Forward airfields were in short supply, lacked adequate ground facilities, and were often summarily taken over by combat units.22
The logistics planners were not prepared for the sequel to breakthrough, even though that was the avowed and oft-repeated aim of offensive operations.23 They were consistently very conservative, indeed, overcautious. They had expressed doubts about the feasibility of the deep post-Cobra exploitation without prior capture of the Breton ports and had estimated that only twelve divisions could be maintained as far forward as the Seine by D+90. By then, they argued, a shortage of 127 truck companies would preclude any additional advance until railway reconstruction had progressed much further. Actually, by D+79, sixteen divisions were being maintained in action on the Seine and another five in Brittany; three weeks later, in mid-September, sixteen American divisions were being sustained, albeit increasingly inadequately, 320 km (200 miles) farther east in addition to those in Brittany. This was despite the fact that the preconditions related to ports, railways, and pipelines remained unmet. The logisticians’ achievements confounded their earlier pessimism. Their improvisations and exertions, especially the Red Ball Express, were an effective, if not efficient, response to an unforeseen challenge. But it was a response with limits. By 5 September, excluding Dragoon forces, there were twenty-four US divisions on the Continent; by 1 October, that number would rise to thirty-three (including two airborne), but only twenty of them could be maintained in combat as far forward as the Rhine, never mind farther east. Moreover, the answers to the supply crisis were short-term ones, applied at the expense of other long-term, long-lasting problems. In mid-September the system was unable to meet more than 80 percent of the minimum daily requirements of over 13,000 tons per day in the combat zone.24 In addition, there was a backlog of 150,000 to 180,000 tons of supplies needed in the forward areas to repair or replace materiel, replenish basic loads, and build up reserves. If the Germans did not give up (as they had in 1918) as the Allies approached their border, the armies’ combat effectiveness would become increasingly degraded. Indeed, if the campaign stretched into the autumn, this hand-to-mouth system would collapse. In the longer term, there could be no shortcutting of the requirement for adequate levels of all-weather, deep-water port discharge capabilities; developed railway and pipeline networks; and the intermediate, well-stocked depots and dumps between base areas and the armies that had been planned before the invasion. Albeit later than they had thought, the planners were proved correct in their assertion that logistic realities could be defied for only so long. Commanders remained willfully blind to this until too late in the day.
Concerns over ammunition stocks that had led to rationing before Operation Cobra disappeared in the wake of the breakout. They were replaced by worries about fuel supply. For instance, on no day between 26 August and 5 October did First Army have more than a one-day supply of POL on hand; on thirteen days there was half a day or less, and on sixteen days there was none at all, as it was issued as soon as it arrived. Both armies found their actions severely circumscribed by fuel shortages throughout September, on some days to the extent of almost total immobility. The crisis both coincided with the stiffening of enemy resistance and contributed to it by easing the pressure on the Germans. And once the pursuit gave way to increasingly heavy fighting, ammunition shortages and the lack of medium and heavy artillery and engineer bridging equipment caused by the diversion of vehicles to the Red Ball Express contributed to the Americans’ inability to bring their full combat power to bear. By the second week in September, Third Army’s emphasis in demand had switched from fuel to ammunition. First Army followed suit a week later. By the beginning of October, both had to reimpose rationing.25 And growing shortages of spare parts and replacements for worn-out equipment and all the other items neglected in the failing effort to meet demands for fuel and then ammunition contributed significantly to the degradation of fighting power. By early September, only about one-third of 3 Armored Division’s tanks were in operable condition. With the expansion and development of the railway system and, later, pipelines, the supply outlook improved. COMZ was able to raise each army’s allocation to 5,000 tons per day on 23 September (still enough for just seven or eight divisions) and then to 5,400 tons on 25 September and 6,500 tons on 4 October. Targets once again proved to be too optimistic and unattainable in the short term (First Army did better than Third), but by mid-October, the firepower situation was improving. Unfortunately for the Allies, this improvement came too late to restore the tempo of the advance. The offensive had culminated and stalemate had set in once again. Not until 7 November, much later than COMZ promised, were enough stocks accumulated for 12 Army Group to resume the offensive.
The fault was not solely that of the logisticians. Field commanders persisted in producing and following operational plans that ignored supply realities and COMZ’s very real problems. Hodges and Patton, far more than their more logistics-conscious British counterparts, were impatient of such considerations in their desire to concentrate on combat; the former was described by a troubleshooting War Department observer as “a man intolerant of supply shortcomings who has not studied supply and does not intend to.”26 In consequence, the combined effects of time, space, and friction slowed the advance to an intermittent crawl before the end of September. But the crisis was also caused in part and then exacerbated by an ill-thought-out administrative system poorly suited to the inconvenient realities of maneuver warfare. The armies, entirely overlooking their own contribution to the crisis, heaped blame onto COMZ and its commander for the growing difficulties. COMZ countered by blaming unforeseeable events, the failure to capture ports on schedule, the irresponsible behavior of combat formations, the shortage of service troops, and the weather—everyone and everything except itself. In truth, although these excuses had some foundation, COMZ was a bureaucratic and inefficient organization of overlapping jurisdictions and reliant on inadequate communications; as such, it was able to support only linear, attritional, relatively slowly developing warfare.
The main constraint limiting COMZ’s ability to supply the fighting formations in September was the lack of in-theatre transport capacity, exacerbated by systemic failings (some of which were belatedly addressed in October). There were plenty of stocks of most items on the Continent (some ammunition natures being the most important exception) but as of 1 September, 90 percent of them were still in the Normandy base area. Even by the end of October, when the rail system was hitting its stride and shifting increasing tonnages forward, that figure had only dropped to 70 percent. In November there were still 600,000 tons in stockpiles at or near the invasion beaches and Cherbourg. Given that port capacity was not the principal bottleneck affecting supply in the late summer, the decision to abandon the project of exploiting Quiberon Bay’s natural harbor and the other Brittany ports cannot be blamed for the onset of a new stalemate in October. Imports were not a major problem in the late summer of 1944. They would, however, become one if deep-water ports were not opened rapidly to ensure adequate and continuous supply of the whole Anglo-American force in increasingly demanding conditions of combat and, at the same time, cope with the growing inflow of reinforcing formations (eight US divisions in September alone). In practice, this meant bringing at least Antwerp on line. Antwerp could handle more than the Normandy beaches and all the Channel ports combined, and it was five times nearer to Aachen than Cherbourg was and more than twice as close to Metz.27 The failure to open Antwerp until 29 November would have a profoundly deleterious effect on the campaign.28
The idea that only POL shortages stopped the pursuit is false. Other problems and deficiencies grew from neglect and sapped fighting power. Tanks and other vehicles became badly worn, immobilized, or total losses. In a November report, SHAEF provided the War Department with statistics on the loss of equipment: every month, 700 mortars, 375 medium and 125 light tanks, 900 2.5-ton trucks, 1,500 jeeps, and 100 cannon of various calibers had to be written off, not to mention 5,000 tires every day; theatre tank losses in August and September amounted, respectively, to 25.3 percent and 16.5 percent of establishment, and with reserves exhausted, even if the official replacement rate of 9 percent were met, it would not be enough to keep up unit strengths. Repairable items were being neglected owing to a lack of maintenance facilities and spare parts; thus, 15,000 vehicles lay deadlined in November. Advance airfields were not being built, and air units were not being moved up. Even when ammunition became available, needs could not be met in good time. Increasingly during September, the armies were forced to live a hand-to-mouth existence. This made forward planning a somewhat dubious, even nugatory exercise and led to an offensive that proceeded in fits and starts. All this allowed the enemy to rally and reconstitute some of his shattered units and formations and find fresh troops from the interior of the Reich and other fronts. It was not, by the standards of July, an impressive force in terms of size, quality, and training and equipping of the troops. But, given terrain favoring the defense, it was enough to stall weakened Allied forces, whose commanders were still infected with an optimism that was becoming harder to justify and increasingly mismanaged the campaign.
At the end of September, SHAEF and COMZ brought an overdue awareness of reality to operational planning. They calculated that the supply needs of 12 Army Group and air force essentials in the first half of October would amount to 18,800 tons per day, assuming the employment in combat of twenty-two divisions. This figure would go up to 22,700 tons by 1 November, when there would be twenty-eight divisions. The army group had also demanded another 100,000 tons over and above daily requirements to both repair deficiencies and create minimal reserves. However, deliveries in September were running at only 8,000 to 10,000 tons daily to the forward areas, and even in October it would be impossible to supply the minimum amounts necessary. It would take all of two months to improve the port and transport situation sufficiently to not only satisfy daily necessities but also create the required stocks in the forward dumps. Moreover, this mid-November target would be met only if railway rehabilitation proceeded on schedule and Antwerp started to ease the port discharge bottleneck. The possibility of restoring the momentum of August and delivering a fatal blow to the Germans before the winter had disappeared.
The autumn operational pause led to much ill feeling and recrimination between the American armies and even more between them and COMZ, which the army group did little to smooth over. The logistic crunch also contributed in no small measure to a souring of Anglo-American relations, with denigration of the Supreme Commander an unfortunate by-product. The US armies knew what they needed and, somewhat cavalierly, assumed that the required supplies were available somewhere in the system. When these supplies failed to materialize, they blamed each other for hogging supply, COMZ for its inefficiency, and the high command for diverting transport and POL to the British. Thus, one of Hodges’s staff colonels declared that in his army’s “race across France we were badly crippled for want of gasoline . . . [because] gasoline was diverted to the British to enable them to take the Channel ports, to capture the V-1 launch sites and to bring them faster into Antwerp.” Similarly, one of Patton’s officers wrote: “The situation now had all the indications of deliberate withholding of gas from Third Army. . . . Bradley told Patton that ‘Montgomery had won.’ That he had succeeded in inducing Eisenhower to make the main effort . . . in the strongly held and organized north instead of in the weak and disorganized south, as favored by Bradley.” Patton’s diary entry for 30 August echoed the sentiment: “The Brits have put it over again. We got no gas because, to suit Monty, the First Army must get most of it.”29
The actual unfolding of events demonstrated that Eisenhower’s decision in favor of an advance right across the front was mistaken. Each of the armies in his Northern and Central Groups culminated short of its assigned geographic objective, and the German army was vouchsafed enough breathing space to recover its strength and balance in the operational pause that followed. Operational and tactical failings undoubtedly contributed to this disappointing result, but the main cause of the Allied failure was the logistic system’s inability to cope, especially on the part of the Americans. Difficulties in supply prevented a significant and growing proportion of US combat power from being used to maximum effect; by 1 October, only twenty of 12 Army Group’s thirty-one divisions could be maintained as far forward as the Rhine, but weeks earlier, shortages and deficiencies had been acting as a brake on operations. Just as important, the resultant uncertainties complicated and distorted planning. Moreover, by trying to do too much with too little for too long, the Allies exacerbated their problems and ensured that recovering from them would be a slow process. It would take months to remedy the results of prolonged neglect of vehicles and equipment, acquire and ship forward the necessary spares and stores and replacements, and establish the intermediate depots, even minimally stocked, that were necessary for the prosecution of sustained operations. This raises the question: would two powerful but sequential thrusts, rather than a simultaneous advance, have been more logistically feasible for 12 Army Group?
Had Third Army been halted on the Meuse at the end of August (but still with its bridgeheads) and left on the defensive for a time, what lift would have been freed to support a stronger effort north of the Ardennes, the supposed main effort? Assume that XII and XX Corps, with six divisions between them, were left to defend the passive sector; assume that the Germans could not mount a serious threat against them, so POL and ammunition requirements were less than normally allocated in defense. In this case, the two corps would be able to get by on a meager average of 250 tons per division per day, a total of 1,500 tons. The theoretical allocation of supply to the two US armies was 3,500 tons apiece daily, so in theory, Hodges would be gaining 2,000 tons. In practice, however, during the first half of September, COMZ was averaging only 2,500 tons to Patton and 3,300 to Hodges, so the actual gain to First Army would be halved. Air resupply could reliably add about 650 tons per day, up to 1,000 in a surge. With 4,950 to 5,300 tons a day, First Army would be able to maintain sixteen to seventeen divisions in pursuit mode or seven to eight in attack mode, similar to its actions around Aachen in mid-September, but now properly resourced. Thus, Third Army’s transition to defense probably would have enabled First Army, nine to ten divisions strong, to penetrate as far as the Rhine in September in a continuous rather than a stop-and-go push. Had the assault on Brest been abandoned, there would have been enough resources to dispatch VIII Corps with perhaps three or four divisions to augment Hodges’s efforts or to reinforce the underresourced 21 Army Group.
It is likely that, having ensured that First Army had enough forces and staying power to reach the Cologne-Bonn area without giving the enemy any respite to reestablish a coherent defense, an operational pause would have been necessary to sort out accumulated problems in the logistic system. This would mean postponing Patton’s offensive until well into the autumn—probably November—with all the adverse consequences that would entail. However, the offensive would have been more powerful, and adequate levels of supply would have been assured. Patton’s army would not have been alone in the assault on the southern flank, as 6 Army Group, its own supply in order, would have been attacking from the south. Chapter 7 considers more fully the operational calculus involved in a phased effort to close on the Rhine.
Patton and Bradley would have preferred to put the main US effort south of the Ardennes from the earliest days of exploitation to enable Third Army to execute a quick drive across the Rhine with ten to twelve divisions. They accepted that this could be done only if Third Army were given priority on all available supplies and transport. They envisaged the British, then on the Seine, being halted in place. First Army, too, would remain inactive, with the possible exception of a limited advance to protect the northern flank. Even with this concentration of logistic effort, they recognized that the limit of exploitation would be no further than Frankfurt; this, however, was considered sufficient, as such a rapid and deep advance, penetrating into Germany itself, would likely induce collapse and frighten the enemy into immediate capitulation. SHAEF studied the concept carefully in late August and rejected it. Logistically, it was a nonstarter. The force would require 6,500 to 8,000 tons per day, more than the entire amount COMZ delivered on an average day to the whole army group through most of September. In other words, assuming Hodges’s army still needed at least 1,500 tons, even seven to eight divisions in the Frankfurt area would be ambitious. The Red Ball Express would have to be extended by over 50 percent to Saarbrücken, and the increased turnaround time would require even more trucks and fuel to maintain the same rate of supply. Army transport would still have a 200 km (125 mile) haul from there to Frankfurt. The wear and tear on vehicles and men and the depletion of spare parts, tires, and so forth would increase. Long lines of communication within Germany, running through difficult territory, would be vulnerable to disruption by raids, guerrilla activity, and sabotage. Air resupply would help, to the tune of 500 to 1,000 tons per day, but only in good weather and in the absence of the Luftwaffe. The latter could not be guaranteed beyond fighter cover, and without forward airfields, that would become steadily scarcer. Above all, considering purely logistic factors for a moment, the Patton-Bradley scheme did not begin to address the issue of ports. The US forces would enter the autumn campaign still reliant on Cherbourg and the beaches, with the insignificant addition of some minor ports and the imminent closing, or at least drastic limiting, of the beaches as the weather made its impact.
Montgomery’s narrow-front concept was based on a keener appreciation of logistic factors, and it was aimed at a more important geographic objective. Montgomery was famous for his conservatism in the area of logistic risk; Americans habitually accused him of overinsurance. Nevertheless, the SHAEF study of early September agreed that Montgomery’s initial idea of a thrust as far as Berlin was theoretically feasible. Three British and two US corps could be employed initially, with three going all the way to Berlin (albeit on reduced maintenance), a British one being halted in the Bremen-Hamburg area, and an American one being halted in the Frankfurt-Magdeburg area. They believed, however, that several preconditions for success would have to be achieved. Both army groups would have to be on or over the Rhine by 15 September to provide an adequate LD. The Channel ports and Antwerp would have to be discharging 7,000 tons per day by that date. Railway rehabilitation would have to progress further and faster than was actually being achieved to free up road transport. Supplying the offensive would require the equivalent of 489 truck companies, but only 347 were available; air could compensate for 60 companies’ worth of the deficit (they assumed a highly optimistic lift of 2,000 tons per day), but the remaining deficiency would have to be found by the wholesale deprivation of divisional organic transport. Five American corps would have to be grounded or left “quiescent.”30 As none of the SHAEF stipulations were met, Eisenhower rejected this overly ambitious scheme but later agreed to Montgomery’s more modest Market Garden proposition.
With the resources available as of mid-September, was the scheme to encircle the Ruhr really as viable as Montgomery claimed? Second Army’s administrative position at the time was by no means good. Commodity stocks in the Brussels area road head were low, and the road head was too far behind the forward troops, with the Brussels bottleneck intervening. A month’s pause was considered necessary to reestablish administration on a sound footing, but none would be vouchsafed. That said, “at no time were there insufficient stocks in army depots to meet the requirements of Market Garden.”31 During the early period of the offensive through Holland, the line of communications’ supply of the army was still largely dependent on the 500 km (300 mile) haul from the RMA to Brussels. Although this was alleviated for more than half the journey by rail transport, army transport still faced a 230 km (145 mile) lift to get to Arnhem. Nevertheless, the system coped, helped to a considerable extent by the 500 tons per day trucked by the Red Lion Express, the use of tank transporters, and air resupply, which averaged 550 tons each day for four weeks into mid-October.
Of course, the Rhine crossing at Arnhem was merely an intermediate objective, not the final one. Montgomery aimed to move round the north face of the Ruhr as far as Münster and Osnabrück via Deventer, an additional 170 to 185 km (105 to 115 miles). Could Second Army, whose nine divisions required 5,850 tons daily, be supplied over this additional distance? Several factors were favorable. By the end of September, Dieppe and Ostend were discharging up to 8,000 tons daily; by 12 October, Boulogne was adding another 2,200 tons. This would go a long way toward meeting 21 Army Group’s total need for 9,100 tons. Dieppe enjoyed a through rail link with Brussels via Amiens from 6 September, and with the bridging of the Seine on 22 September, so did the RMA. The rail system would provide up to one-third the tonnage 21 Army Group required—more as time passed; moreover, Dempsey’s railhead would reach Eindhoven by 6 October.32 The road head was advanced to the Neerpelt area by 4 October, 80 km (50 miles) nearer to Arnhem than Brussels. These improvements would considerably ease, but not eliminate, the line of communications part of the problem; however, army transport would still face hauls of up to 300 km (185 miles). Fortunately, in mid-September the War Office agreed to loan an additional seventeen general transport companies (normally about 120 trucks each) to the army group, and these had all landed, preloaded with supplies, by 3 October. All this added up, in theory, to the conclusion that Montgomery’s concept was logistically feasible, particularly if he continued to receive American and air force help, which together amounted to 1,000 tons daily, and sometimes more.
Theory, however, does not take into account the problem of organizing, managing, administering, and controlling a substantially improvised system over vastly increased distances, an area of growing concern. Weather would increasingly impact air resupply, and the Luftwaffe might become a factor if Allied fighters could not quickly relocate forward. Once into Germany, attempts at sabotage could be expected. The line between logistic success and disastrous overstretch was probably quite thin and vulnerable to the ever-present friction of war. As it was, the failure of Market Garden necessarily resulted in a reappraisal of the logistic situation. Second Army calculated that a resumption of the offensive would require a buildup of the Antwerp-Brussels area advanced base until it contained 20,000 tons of supplies and 40,000 tons each of ammunition and fuel, not to mention large quantities of engineer, ordnance, and medical stores. By the end of October, there were only 23,000 tons of POL and a mere 4,000 tons of ammunition, even though supply targets were almost reached.33 Whether Montgomery’s intended outcome actually fit into a wider operational perspective and whether it was therefore desirable are considered in the next chapter.
Such calculations are highly hypothetical and owe much to hindsight. Armies at the time overestimated their own capabilities and underestimated the enemy’s, and neither they nor COMZ was entirely honest about the true status of supply—or even sure what that status actually was, given the somewhat chaotic nature of American logistic administration. Eisenhower lacked the necessary data to make a truly informed choice, and wishful thinking distorted the intelligence picture of the German situation. That said, one thing was crystal clear: until the port of Antwerp was discharging significant tonnages, which it could do very rapidly once the Scheldt estuary was cleared, the Americans would be unable to exploit fully more than two-thirds of their combat power. In addition, if the Germans did not collapse and sue for peace when the Rhine was reached or crossed, neither army would be in good enough administrative shape to pursue the campaign into the heart of Germany. Unless and until the Allies took advantage of Antwerp’s huge capacity, the excellent road and rail links to the interior, and their relative proximity to even the most southerly portion of 12 Army Group’s front, the campaign was likely to degenerate into a largely pointless, attritional semistalemate. If the Allies were not sure of their ability to bring the campaign to a successful conclusion before autumn was well advanced (as they could not be), they would have been well advised to make clearing the Scheldt estuary the first priority of post-Normandy operations. The issue could not be postponed and fudged, as had been done with the Brittany ports. The longer the supply problem was neglected, the worse it would get and the longer it would take to put it right. In the event, this was not actually achieved until well into December, by which time the Allies had lost the initiative, wrested so convincingly from the enemy back in the heady days of August.