Prepared by:

[Note: This manuscript was prepared during World War II by the historians assigned to the Supreme Headquarters, Allied Expeditionary Force (SHAPE), and subsequently forwarded to the Office of the Chief of Military History (now US Army Center of Military History) for use in the writing of the official history of World War II. The original is on file in the Historical Manuscripts Collection (HMC) under file number 8-3.6A CA, which should be cited in footnotes, along with the title. It is reproduced here with only those limited modifications required to adapt to the World Wide Web; spelling, punctuation, and slang usage have not been altered from the original, especially given that the original was prepared using British rather than American dialect. Where modern explanatory notes were required, they have been inserted as italicized text in square brackets.]

I. THE ORIGINS OF COSSAC

The great design for the return of Allied Forces to the Continent of EUROPE had its beginnings at DUNKIRK. From that time onwards, such a return became the ultimate goal of all whose responsibility it was to plan the military operations against GERMANY, but at first the dearth of men, of arms and of all the resources necessary for the prosecution of war made the study little more than an academic exercise. With the entry of the UNITED STATES OF AMERICA into the conflict, however, the potentialities of the situation charged rapidly, and the planning of a detailed campaign came within the realm of practical policy.

The successful conclusion of the campaign in NORTH AFRICA was necessary before the attention of the Allies could be devoted to a full scale attack upon EUROPE, but at the CASABLANCA Conference, in January 1943, it was felt that the time had come to evolve schemes for cross-Channel operations. It was deemed advisable that plans should be produced for a return to the Continent in the event of GERMAN weakening or collapse and for deceptive operations to be conducted during 1943. The suggestion was also made that in August 1943 a bridgehead might be established in the COTENTIN PENINSULA from which further offensive operations could subsequently be undertaken, but hopes of this possibility were speedily abandoned. It had been calculated that at least four UNITED STATES divisions would be available for the task by mid-August, seven by mid-September and fifteen by the end of the year, totaling 938,000 men. These figures, however, were found to be greatly exaggerated, and it was recognized that the full scale cross-Channel attack could not take place before 1944.

It was agreed at CASABLANCA, however, that the work of preparing for the grand assault on the fortress of EUROPE must go forward, and that it must be shared by the two nations which were eventually to cooperate in its execution. For the present it was decided to appoint a Chief of Staff to the Supreme Commander, under whom would be established a UNITED STATES–BRITISH staff, with the duty of driving forward the plans for cross-Channel operations–a task which had hitherto been entrusted to a body known as the 'Combined Commanders'. It was expected that the Supreme Commander ultimately to be appointed would be a BRITISH general and that he would have an AMERICAN deputy, so the nomination of the Chief of Staff was decided on parallel lines. Lieut-General F. E. MORGAN was appointed to this post, with Brig-General R. W. BARKER, of the U.S. Army, (who had previously been associated with the Combined Commanders) as his deputy. To these men accordingly fell the task of building up the organization which was to plan the allied invasion of North-West EUROPE.

Taking the initial letters of his appointment–Chief of Staff to Supreme Allied Commander–General MORGAN christened his organisation "COSSAC".

II. THE AIMS OF COSSAC

General MORGAN's first introduction to his task came when Lieut-General ISMAY, the Deputy Secretary to the War Cabinet, presented him with a file of papers "some inches thick'', containing the records of the Combined Commanders' work in connection with the proposed operations against the GERMAN forces in EUROPE. He was required to "elaborate the means by which the expedition was to be organised and undertaken" within twenty-four hours. This he did in the form of a memorandum on cross-Channel operations, which served as a basis for discussion when he met the Chiefs of Staff on 24 March 1943.

In this memorandum he dealt with both long- and short-term policies in respect of such operations. He stressed the acute difficulties which had in the past arisen in consequence of the divorce of planning from execution, and urged that the two functions be combined from the outset. He further postulated that there must in all services in the machinery of higher command be complete amalgamation between the BRITISH and the AMERICANS, since although they were not yet in the UNITED KINGDOM in great numbers, the ultimate issue lay with the "United Nations Strategic Reserve of land forces", the AMERICAN army. General MORGAN's view was that the Chief of Staff should be invested "with plenary powers, temporarily to impersonate the commander-to-be"; by that means it would be possible satisfactorily to "launch the organization and maintain impetus until the arrival of the Supreme Commander". In view of the short time available for planning, no delay or interference must be tolerated. It was of prime importance that the highest possible degree of autonomy, at least in the operational sphere, be granted at the earliest possible stage to all concerned; on the administrative side, such autonomy must be attained gradually, but it should be achieved in sufficient tine for the Supreme Commander to be able to test his administrative machinery before taking the field,

The principles thus enunciated were recapitulated by General MORGAN when he addressed his staff at the first meeting of COSSAC held on 17 April 1943. "I want to make it clear", he said, "that, although the primary object of COSSAC is to make plans, I am certain that it is wrong to refer to it in any way as a 'planning staff'. The term 'Planning Staff t has come to have a most sinister meaning–it implies the production of nothing but paper. What we must contrive to do somehow is to produces not only paper, but ACTION".

"In spite of the fact that it is quite clear that neither I nor you have by definition any executive authority, my idea is that we shall regard ourselves in the first instance as primarily a co-ordinating body. We plan mainly by the co-ordination of effort already being exerted in a hundred and one directions. We differ from the ordinary planning staff in that we are, as you perceive, in effect the embryo of the future Supreme Headquarters Staff. I do not think I can put the matter any better to you than by quoting to you the last words of the CIGS, who said: 'Well, there it is; it won't work, but you must bloody well make it.'"

General MORGAN then read to his staff the Directive to COSSAC which had been drawn up by the BRITISH Chiefs of Staff. The terms of this Directive had been evolved during March, and had been discussed with General Andrews, Commanding General of the UNITED STATES Forces in the European Theatre of Operations, on 2 April 1943. The amended draft was then submitted for the approval of the AMERICAN Chiefs of Staff in WASHINGTON, and finally issued on 26 April. When General MORGAN met his staff on 17 April, however, he had been authorized to proceed on the terms of the Directives assuming AMERICAN approval.

This Directive, which provided the basis for all the subsequent activities of COSSAC, ran as follows:

1. The Combined Chiefs of Staff have decided to appoint, in due course, a Supreme Commander over all United Nations forces for the invasion of the Continent of EUROPE from the UNITED KINGDOM.

The Supreme Commander will be responsible to the Combined Chiefs of Staff for planning and executing such operations, and for the coordination of the training policy of forces to be employed in amphibious operations against the Continent in accordance with this Directive,

2. Pending the appointment of the Supreme Commander or his deputy, you will be responsible for carrying out the above planning duties of the Supreme Commander. You will report direct to the BRITISH Chiefs of Staff with whom will be associated the UNITED STATES Commander of the European Theatre of Operations acting as the direct representative of the UNITED STATES Chiefs of Staff in the UNITED KINGDOM.

OBJECT

3. Our object is to defeat the GERMAN fighting forces in North-West EUROPE.

4. To this end the Combined Chiefs of Staff will endeavour to assemble the strongest possible forces (subject to prior commitments in other theatres) in constant readiness to re-enter the Continent if GERMAN resistance is weakened to the required extent in 1943. In the meantime the Combined Chiefs of Staff must be prepared to order such limited operations as may be practicable with the forces and material available,

PREPARATION OF PLANS

5. You will accordingly prepare plans for:

STAFF AND METHOD OF PLANNING

6. You will be provided with a small permanent Combined Staff drawn from the BRITISH and UNITED STATES Navies, Armies and Air Forces. For administrative planning you will make appropriate use of the planning agencies of the U.S. and BRITISH Services.

FURTHER GUIDANCE

7. You should maintain close contact with the BRITISH Chiefs of Staff and their organisation, through whom you will be given such further guidance as you may require. You should also maintain close contact with Headquarters European Theatre of Operations of the UNITED STATES Army.

ALLIED STAFFS

8. The Allied Military Staffs (other than BRITISH and AMERICAN) will not be brought into the planning at present. The BRITISH Chiefs of Staff will inform you at what stage these other Allied Staff should be consulted.

CANCELLATION OF PREVIOUS DIRECTIVES

9. This Directive cancels all previous directives issued to the Combined Commanders for amphibious operations launched from the UNITED KINGDOM against the Continent.

*********

Commenting on the terms of his Directive, General MORGAN stressed the need for a comprehensive view of the task which lay ahead of COSSAC: that task was nothing less than the reconquest of EUROPE. To achieve that, they would have to deal with an army of the order of 100 divisions—of which approximately 15 would be BRITISH and 85 AMERICAN—together with air forces of proportionate strengths and a large naval support. At the moment, only the vanguard of these mighty forces were in BRITAIN; the bulk still lay in the UNITED STATES, and, when the time came, they would enter the Continent direct from across the ATLANTIC. With so vast a task, in which such huge forces would be employed, COSSAC must work, as General MORGAN put it, with "a map which starts at one end with SAN FRANCISCO and ends at the other end in BERLIN".

Furthermore, COSSAC's work must have depth as well as breadth: the task was to reconquer EUROPE. To date, most of the planning had dealt only with the initial attack upon the shores of the Continent. "In our Grand Campaign", said General MORGAN, "our ultimate object is to wage successful war on land in the heart of EUROPE against the main body of the GERMAN strategic reserve. It is true that we have to cross the enemy's beaches, but that to us must be merely an episode. True, it is a vital episode and, if it is not successful, the whole expedition will fail. We must plan for the crossing of the beaches, but let us make sure that we get that part of the plan in its right perspective—as a passing phase".

Apart from the comparatively long-term planning for the assault on the Continent, COSSAC had to make preparations to deal with the situation should the AXIS collapse. It was difficult to say when this might occur; in 1918, GERMANY passed from her zenith to her defeat in eight months, and history might repeat itself in 1943. If such a break came, then COSSAC must be prepared to take full advantage of it with whatever forces might at the time be available.

Then there was the task of planning operations for 1943 on as comprehensive a scale as resources would permit. In some vital respects, such as land forces and landing craft, these resources were woefully slender; but, General MORGAN pointed out, 'operations' did not necessarily imply a land campaign. The object was to test the degree of enemy resistance, and there were more subtle ways of achieving that than by direct frontal assault in maximum force.

Keeping these aims in view, the first task of COSSAC must be to acquire the essential information upon which to base the plans for future operations. The unfortunate misapprehensions which arose through inaccurate information at CASABLANCA provided an object lesson in this respect. "Before we can begin to make any plan of any kind, we must have accurate knowledge of the resources of all kinds that are, or will be, at our disposal for the execution of that plan". It was essential that the vast intricacies of existing production programmes should first be mastered; then would come the time to consider what operations were feasible, and if necessary perhaps to suggest adjustments in matters of production.

What must ever be in the minds of all the staff of COSSAC was this: that it was they who were to produce the "Master Plan".

III. THE DEVELOPMENT OF COSSAC

The history of COSSAC, from its inception to the time when it was transformed into the Headquarters of the Supreme Allied Commander, is the story of the development of the three main operational plans which were evolved to meet the requirements of the original Directive, and of the internal expansion of organization and powers necessary to produce and eventually to implement these plans. This expansion was to transform COSSAC from a so all planning staff into a complex executive headquarters, under a Chief of Staff exercising practically all the authority of the Supreme Commander pending the latter's assumption of duty in January 1944.

The nature and development of the operational plans produced by COSSAC will be considered in detail in due course, but it is necessary here briefly to outline them in order that the internal affairs of the headquarters may be fully understood. The ''elaborate camouflage and deception scheme extending over the whole summer with a view to pinning the enemy in the WEST" took the form of a series of operations known under the general name of 'COCKADE'. This consisted of three component operations: the mounting of a threat, by forces based in SCOTLAND, against NORWAY, known as operation 'TINDALL'; a threat by UNITED STATES forces based in Southwest ENGLAND against the Western ports of FRANCE, known. as operation 'WADHAM'; and an operation involving an amphibious feint from South-East ENGLAND against the PAS DE CALAIS, known as operation 'STARKEY'. The plan for "a return to the Continent in the event of GERMAN disintegration" became known as operation 'RANKIN'; this also was sub-divided into component operations, Cases A B and C, according to whether the return was to be effected under conditions of a weakened GERMAN resistance, a withdrawal from the occupied countries or a complete collapse of GERMAN power within the Fatherland itself. Finally, the chief scheme of all, the plan for "a full-scale assault against the Continent in 1944", was accorded the name of operation 'OVERLORD'.

Of these three main schemes, operation 'COCKADE' was carried out in its entirety under the aegis of COSSAC, but the hoped-for GERMAN collapse or withdrawal did not materialize, and operations 'RANKIN' and 'OVERLORD' were accordingly handed over to General EISENHOWER for implementation when he became Supreme Allied Commander.

The central feature of the development of COSSAC in the evolution of these plans lay in its transformation from a planning staff into an operational headquarters. As has been noted, General MORGAN had pointed out the difficulties inherent in an organization where execution was divorced from planning, in his memorandum on the subject of the organisation of COSSAC before his headquarters actually began to function; and he was careful to stress, at the first meeting of his staff, the dangers which would result if they were to regard themselves simply as producers of paper plans to whom no responsibility attached for the execution of their schemes. On 20 May 1943, he suggested to his Principal

Staff Officers that they should always try to envisage themselves as working, not simply for a Chief of Staff, but for the Supreme Commander himself.

This theme was repeatedly stressed on subsequent occasions, and as the operational schemes were developed and the time for their implementation drew near, the necessity for the grant to COSSAC of executive powers, pending the appointment of the Supreme Commander, became increasingly evident. When the basic plan of operation 'OVERLORD' had been submitted to the BRITISH Chiefs of Staff, General MORGAN told his staff at their weekly meeting on 9 July 1943 that they must now "consider the reorganization of this Headquarters, with the object of transforming the present planning staff into an operational staff", as a necessary pre-requisite of proceeding to detailed planning. "The time is drawing near", he said, "when it will be necessary to define the relationship to exist between my staff and the outside world. Signs have recently become apparent that, although, by definition, I possess none, I am expected to exert some measure of executive authority."

He accordingly gave instructions that, pending further official ruling, the COSSAC staff should consider itself an executive one, and orders were issued that preparations should be made for the necessary internal reorganization. These steps were confirmed in August, when the QUADRANT Conference gave its official blessing to the basic 'OVERLORD' plan. Now, said General MORGAN, they must proceed with all speed to transform themselves into an operational headquarters, which being done, "we are off and nothing shall stop us."

It remained only to issue the necessary amendment, conferring executive powers, to the COSSAC Directive. This was under consideration by the BRITISH Chiefs of Staff Committee at the end of August, and the Secretary, Brigadier HOLLIS, wrote to General MORGAN to ask for immediate detailed suggestions; "Never", commented the latter, "were so few asked to do so much in so short a time." He consulted General DEVERS, Commanding General, European Sheathe of Operations, UNITED STATES Army, concerning the position—"the present anomalous arrangement whereby I function without a Commander . . . indefinitely". "While I hate the sight of the whole of this business", he wrote, "I am completely at a loss to suggest anything better, short of course of appointing the great man himself, which appears to be utterly impossible". The only solution seemed to be to put a simple amendment to the Directive, and then "to rely on the good will of all concerned to make the impossible state of affairs thus created work in practice". This was the tenor of his reply to the Chiefs of Staff Committee: ''Essentially what we are here trying to do is to make an impossible situation reasonably possible for practical purposes". The best plan was to "reduce verbiage to its smallest limit", the only alternative being to go into immense detail which would entail a vast amount of work and lead nowhere. "So, as I say, for my part I incline to suggest to you use of the minimum words and the smallest possible piece of paper, and then to hope for the best".

This was accordingly done, and on 7 September 1943 the necessary variation of the original Directive was issued, paragraph 2 being amended to read: "Pending the appointment of the Supreme Commander or his deputy, you will be responsible for carrying out the above planning duties of the Supreme Commander and for taking the necessary executive action to implement those plans approved by the Combined Chiefs of Staff".

Fortified with this grant of executive authority, COSSAC was now able to proceed in the work of developing the basic plans evolved, to carry through the execution of operation 'COCKADE'; and to take the necessary steps for establishing the forces required for the accomplishment of operation 'RANKIN' or operation 'OVERLORD' when the time for them should come.

The implementation of COSSAC's Directive naturally involved a considerable amount of travailing about the world by the principal members of the staff. Thus in May 1943, Major-General BARKER went to AFRICA to study the preparations for operation 'HUSKY ' for the lessons they would afford in COSSAC's own tasks; while Colonel J. T. HARRIS went to WASHINGTON to explain COSSAC to General MARSHALL. On 28 July, a group of the COSSAC staff, consisting of Major-General BARKER, Captain HUTCHINGS (U.S.N.), Colonel ALBRECHT and Lieut-Colonel KUTZ, left for WASHINGTON to present the 'OVERLORD' plan to the U.S. Chiefs of Staff, to confer with the War Department as to the troop basis for the operation, and to discuss various questions concerning civil affairs and supply. Of this mission, Major-General BARKER, Captain HUTCHINGS and Colonel ALBRECHT subsequently proceeded to QUEBEC, where they were joined by Captain MANSERGH, Brigadier McLEAN and Air Commodore GROOM. It was the task of these six representatives of COSSAC to explain the basic operational plans evolved to the President, the Prime Minister and the Combined Chiefs of Staff assembled at the QUADRANT Conference. In October and November, General MORGAN himself went to WASHINGTON to discuss operation 'OVERLORD' with the Combined Chiefs of Staff. Missions were also exchanged at this period with Allied Force Headquarters, ALGIERS, to correlate the plans of offensive action in the MEDITERRANEAN and North-West EUROPE in 1944.

Under the terms of the original Directive to COSSAC, the Allied Military Staffs, other than BRITISH and AMERICAN, were not to be brought into the planning at present, and it was pointed out at the inaugural meeting that the very existence of the staff should be concealed as long as possible. Other nations had an inkling of what was contemplated, and it was important to keep the facts from them, for the security of the European exiled allied governments was notoriously bad. Contact with the CANADIAN authorities had, however, to be made from the outset, since the bulk of the forces ultimately to be employed in the invasion of North-West EUROPE was to come from the NORTH AMERICAN Continent.

By August 1943, however, despite the original ban on consultation with the other Allied governments, "signs and portents were not wanting that the time had now come when it was no longer possible to dispense with such contacts". Planning for operations 'RANKIN' and 'OVERLORD' had reached the stage when it was necessary to obtain information on such matters as port facilities, internal conditions and civil affairs of the occupied countries—information which could only be obtained from the allied nationals concerned. Moreover, the proposal to use FRENCH patriot forces in operations in Southern FRANCE made it inevitable that some disclosure of intended plans must be made to the Committee of National Liberation. Great caution would be necessary in making these approaches, for a salutary warning was available in the painful experience suffered by Combined Operations Headquarters as a result of the indifferent security of European allied staffs.

It was proposed that the heads of the Military Missions of the DUTCH, BELGIAN, NORWEGIAN, POLISH and CZECHOSLOVAKIAN governments should be accredited to the headquarters of the Supreme Allied Commander, so that he could consult with them as required, but that they should normally remain with their respective missions at 21 Army Group Headquarters, where their opportunities for interference and investigation in matters of higher policy would be limited. In the case of the FRENCH, however, it was necessary for a mission to be specially accredited to Supreme Allied Headquarters in view of the active share expected of the patriot forces in forthcoming operations. It was suggested that General MORGAN should be empowered to discuss the objects only of operations 'RANKIN' and 'OVERLORD' with the heads of these missions, without disclosing any details as to military or civil affairs plans, since the aim was ''to obtain information from them without disclosing any in returns". Military information to these missions was not to extend beyond a bald

statement that invasion, or the occupation of the Continent in the face of weakened GERMAN resistance, was intended, and it was to be made clear that the Supreme Commander would not be concerned with political problems.

As the planning by COSSAC developed, and its powers widened, there was naturally a steady increase in the size of the staff. General MORGAN, at the inaugural meeting, had insisted on strict limitation of numbers in the interests of efficiency: all demands for extra staff by the various branches were to be submitted for his personal approval. He desired the ultimate Supreme Commander's staff to be modelled on that employed by Marshal FOCH at the end of the war of 1914-1918: "a really small body of selected officers who dealt with the major decisions on broad lines, the day-to-day work of the war being delegated completely to commanders of army groups".

As originally organized, the staff of COSSAC was grouped in three main branches: Operations, Administrative and Intelligence. The Operations branch consisted of Navy, Army and Air Sections, each of which had AMERICAN and BRITISH components under their respective Principal Staff Officers. The Administrative branch also had an AMERICAN and a BRITISH P.S.O., but the Intelligence branch, although containing AMERICAN representation, was under the control of a single BRITISH P.S.O. In addition, there was established a Central Secretariat, serving the Chief of Staff himself and all the branches.

The three BRITISH and three U.S. Operational P.S.Os. were responsible to the Chief of Staff for the production of the main appreciations and outline plans. Collectively and individually, they were to direct their staffs in accordance with the policy of COSSAC and the requirements of their respective Services, with due regard to the available resources, administrative limitations and available intelligence. The Intelligence and Administrative branches had each their respective naval, army and air components for both the BRITISH and U.S. forces. It was their responsibility to supply necessary data to assist the Operational staffs in preparing their appreciations and outline schemes, and they also worked under the direction of the Major-Generals Intelligence and Administration for the production of intelligence and administrative plans respectively. The heads of these two branches attended the P.S.Os.' meetings to ensure that their own planning conformed to operational requirements, and to advise the Operational P.S.Os. where necessary.

The original Principal Staff Officers working under COSSAC were:

There were, however, a number of changes in the staff during the following months. On 1 May 1943, Commodore J. HUGHES-HALLETT succeeded Rear Admiral VIAN as the head of the Naval Branch (Br), while on 16 June Brig-General W. E. CHAMBERS and Captain G. HUTCHINGS became the respective heads of the Army Operations Branch (U.S.) and the Naval Branch (U.S.). The formation of the Naval staff was a major difficulty in these early days, the delays in effecting the necessary appointments being the cause of repeated complaint by COSSAC.

By July 1943, it was becoming clear that the prospective executive functions of COSSAC were going to necessitate considerable reorganization of the staff, More cohesion was needed, and a closer integration was

especially needed on the operational side. The Naval element had been of an executive nature from the start, and the Air element was becoming so. The Intelligence branch was satisfactory apart from a shortage of AMERICAN officers, while in the Administrative branch there was already integration in the sense that Maj-General BROWNJOHN was the Chief Administrative Officer, though it had been found necessary to retain the separation of the AMERICAN and BRITISH sections by reason of the difference between the two administrative systems. General MORGAN instructed that, to ensure closer co-operation in the operational planning, a single U.S.-BRITISH Operational Branch should be formed. The effect of this would be that in future there would be only three Heads of Branches on the military side—in charge respectively of Operations, Intelligence and Administration. Thus, there eventually came about in each branch the fusion of the BRITISH and AMERICAN staffs with a single officer—BRITISH or AMERICAN—at their head.

In early October proposals on these lines were drafted and forwarded to WASHINGTON and the War Office, ''in contemplation of the transformation of this staff from a planning to an operational Headquarters". The guiding principle henceforth, as Maj-General BARKER, Deputy Chief of Staff, expressed it, was to be that "any division along national lines should be abolished, and that we should reorganize functionally in accordance with the newly-prepared table of organization".

The process of integration was the natural outcome of the increasing weight of AMERICAN representation on the COSSAC staff, in respect of both numbers and ranks, by the autumn of 1943. At first, when the forces of the UNITED STATES in BRITAIN had been comparatively small, COSSAC had been predominantly BRITISH in character; but as the flow of troops across the ATLANTIC continued, so the inter-allied balance on the staff was adjusted. This growth of AMERICAN representation is reflected in many developments, such, for example, as the renaming of the former branches as "G Divisions". The Army Operations Branch became G-3 Division in October 1943, and subsequently the Intelligence Branch became G-2 and the Administrative Branch G-4. It was not until February 1944, however, that the Civil Affairs and Publicity and Psychological Warfare Branches became G-5 and G-6 Divisions respectively.

Meanwhile, in November 1943, the Allied Naval Commander-in-Chief, Expeditionary Force (Admiral SIR BERTRAND RAMSAY) and the Air Commander-in-Chief, Allied Expeditionary Air Force (Air Marshal SIR TRAFFORD LEIGH-MALLORY) were appointed, and their staffs were set up at their respective headquarters. In consequence, whereas the naval and air staffs (drawn from those of the Commander-in-Chief, PORTSMOUTH, and Air Officer Commanding-in-Chief, Fighter Command, respectively) had hitherto been regarded in some measure as branches of H.Q. COSSAC, they now became exclusively part of the staffs of their commanders. It was emphasized, however, that "an intimacy of thought and liaison must be maintained at all levels".

These changes, coupled with the steady expansion of COSSAC, necessitated the establishment of an "Inner Cabinet", consisting of the Chief of Staff, his Deputy, the Allied Naval Commander-in-Chief and the Air Commander-in-Chief. The function of this body was to ''make Command decisions on matters of major importance referred to it requiring inter-service co-ordination on matters of joint interest". The weekly meeting of the COSSAC staff would in future be attended only by persons detailed by the respective Chiefs of Staff and not by the Allied Naval Commander-in-Chief or Air Commander-in-Chief.

From the establishment of COSSAC until the time came for the Supreme Commander to take personal control, the processes of expansion and development proceeded in all departments of the staff along the

lines of General MORGAN's instruction of 5 June 1943: that always "the organization of H.Q. COSSAC should be considered in relation to the ultimate command set-up; the one should eventually merge into the other". One of the first requisites for the conversion of COSSAC into an operational headquarters was the establishment of an effective Signals organization, and by the end of October a Signals planning staff had been formed with Maj-General C. H. H. VULLIAMY at its head. At the same time, COSSAC found it necessary to extend its attention to spheres of activity which had not, in the first instance, been envisaged as coming under its control. Such matters, for instance, as the supervision of all reconnaissance activities and the work of the S.O.E. became inevitably the subjects of COSSAC's control when it was given the authority to proceed with the execution of the projects it had planned.

As early as 19 May 1943, the desirability was foreseen of a Civil Affairs organization within the COSSAC staff to deal with the problems of that nature which would arise in planning operations on the Continent; and General MORGAN had discussed the question with LORD RENNELL OF RODD, head of Civil Affairs in the MIDDLE EAST. On 21 June, application was made to the Chiefs of Staff for officers suited to work of this nature to be assigned to COSSAC.

At first, Civil Affairs matters were conducted in COSSAC by one U.S. and one BRITISH Lieut-Colonel, with certain technical advisers, working under the Maj-General Administration; but by August 1943 it was clear that something more was needed. A GERMAN collapse during the coming winter was thought possible, and the thesis of operation 'RANKIN' was such that combat effort might well be of secondary importance to the Supreme Commander, in comparison with ability to control and direct civil affairs satisfactorily. The existing independent bodies working under the BRITISH and U.S. Governments could not, by reason of their unilateral nature, cope with the tasks ahead; Civil Affairs would be a military responsibility when the ANGLO-AMERICAN forces entered the Continent, and as such should be controlled by an allied organization directly responsible to the Supreme Commander.

The chief difficulty in realising this aim lay in the chronic shortage of trained Civil Affairs staff officers, the recruiting drive for whom had not proved an outstanding success. In September, however, a Civil Affairs branch of COSSAC was at length formed, consisting of four units, to each of which was allotted the task of planning for a particular country—FRANCE, HOLLAND, BELGIUM and NORWAY. Maj-General Sir ROGER LUMLEY became the BRITISH head of this organization, with Colonel C. E. RYAN (shortly succeeded by Colonel KARL R. BENDETSEN) as the senior AMERICAN representative.

Little progress was made at first—partly because the country units worked on the assumption of full military government and direct control as had been the case in ITALY, whereas it was subsequently realised that the countries of North-West EUROPE would be 'liberated' territory, not 'occupied' enemy lands, and as such would require a different type of treatment. With the clarification of the operational plans, however, a further reorganization of the Civil Affairs branch was made. Sections were set up to deal with particular functions—legal, fiscal, supply, economic affairs and information—while the country units continued to devote their attentions to the individual problems affecting their respective lands. The new organization now proceeded to draw up a basis for its work in "Standard Policy and Procedure for Civil Affairs Operations in North-West EUROPE".

Another branch of activity that was to figure largely in the subsequent development of COSSAC was that of Publicity and Psychological

Warfare. Publicity and propaganda were regarded as essential elements in the success of the operational plans, and the establishment of a PPW branch of COSSAC was suggested at a War Office meeting held on 29 April 1943. With the approval of the Chiefs of Staff on 30 June, Lieut-Colonel THOR M. SMITH (U.S.) and Lieut-Colonel AYLMER VALIANCE (Br), with an officer lent by the Air Ministry, were appointed to a Directorate of Press and Propaganda under the Intelligence Branch. In October the DPP was transformed into the Publicity and Psychological Warfare Branch, and in November Brig-General ROBERT A McCLURE came from A.F.H.Q., ALGIERS, to take charge of this branch, which then became independent of the Intelligence Division. This expansion was the direct outcome of the lessons learnt during operation 'STARKEY' in September, when considerable difficulties were encountered in connection with press control and guidance and with the day-to-day co-ordination of propaganda and operations. It was evident that much had to be done to ensure greater success in these respects when the time came to implement operations 'RANKIN' or 'OVERLORD'. A report to this effect had been submitted to the BRITISH Chiefs of Staff on 10 September, and a proposed organization of the PPW Branch was sent to the Combined Chiefs of Staff on 27 December. On 7 January 1944 the PPW 'Charter' was issued, explaining the nature and purpose of the new branch.

The general effect of all the reorganisation of COSSAC was to produce closer integration of effort between the AMERICAN and BRITISH component staffs. The elimination, as far as possible, of national divisions in the work of his staff was a fundamental feature of General MORGAN's policy; and despite the difficulties encountered in the attainment of this ideal—such as the contrasting logistical systems in the Administrative Branch—there was achieved by January 1944, when COSSAC was transformed into the Supreme Headquarters, Allied Expeditionary Force, a very real unity of aims and methods, on the lines which General EISENHOWER had himself laid down for his NORTH AFRICAN headquarters a year earlier. A similar felicity of inter-allied relationship was created in the dealings of COSSAC with outside authorities—with the War Department at WASHINGTON and the War Office in LONDON, with ETOUSA, and with the Chiefs of Staff of both nations.

A necessary concomitant of the expansion of COSSAC's functions was the development of adequate machinery to deal with the paper work involved. In May 1943, the Central Secretariat was set up "to co-ordinate the work of the Headquarters". This was to develop in due course into the Office of the Secretary, General Staff, when COSSAC became the headquarters of the Supreme Commander. Its original duties were defined as the general secretarial work, including the preparation of minutes, in connection with the meetings presided over by the Chief of Staff and the Principal Staff Officers' meetings; the distribution to branches of internal papers prepared by the Chief of Staff; the registration and despatch of the Chief of Staff's external correspondence; and the distribution of correspondence received for the various branches. These duties inevitably became more and more extensive as the work of planning proceeded.

The office procedure of COSSAC as a whole was further crystallized in October 1943. The maintenance of duplicate files (in order that a complete set should be available to each government for historical purposes) was considered but was deemed impracticable, and instead a policy of microfilming was adopted. It was ruled that divisional registries were to keep their own papers while the Central Secretariat coped with general policy ones, that the BRITISH system of filing should be used, and that the U.S. system of correspondence should be adopted. Here was an interesting illustration of inter-allied co-operation and the readiness to take the best from both sides which had become characteristic of the spirit of COSSAC,

Inevitably, as COSSAC grew, difficulties arose in connection with the accommodation available for it. The question was raised in August 1943, when it was seen that the detailed planning for operation 'OVERLORD' would involve staff decentralisation, and a month later Maj-General BROWNJOHN warned COSSAC that some accommodation would have to be found further than NORFOLK HOUSE, where the headquarters had until then been situated. By 29 October, pressure in NORFOLK HOUSE was reported to be increasing weekly, and "a spirit of compromise and forbearance" was required on the part of all concerned.

It was not merely pressure of space that demanded the finding of further accommodation for COSSAC. If it was to be an operational headquarters, conducting operations 'RANKIN' or 'OVERLORD', accommodation was required which would afford adequate protection both to staff personnel and to communications. In all, accommodation was needed for a staff of 320 officers and 600 other ranks, of whom approximately 300 officers were considered the minimum operational staff who would have to function alongside the Supreme Commander himself. At present, the only available premises were the SOUTH ROTUNDA, to which part of the staff was moved in November and December, 1943, NORFOLK HOUSE and 80 PALL MALL. The two latter had neither protected staff accommodation nor protected signal communications. The SOUTH ROTUNDA, which had originally been fitted up as an anti-invasion base, was well protected in both these respects, and was connected to the various ministries by the WHITEHALL TUNNEL. It was, however, too small for COSSAC's needs, and there was considerable haggling with the Air Ministry, which, while not actually wanting to occupy it, desired to maintain a lien on it for use in an emergency, such as a possible intensive rocket bombardment of LONDON.

Since the Supreme Commander must be able to maintain close contacts, it was desired to provide him with suitable protected accommodation outside the central LONDON area but within 30-45 minutes' journey by car of WHITEHALL (for the War Cabinet, Ministries and BRITISH Chiefs of Staff), St. Paul's School, HAMMERSMITH (where 21 Army Group Headquarters, together with the Allied Naval Commander-in-Chief, Expeditionary Force, was located), and STANMORE (where the Headquarters of the Air Commander-in-Chief, Allied Expeditionary Air Force, was established). It was at first thought that no such suitable accommodation existed outside LONDON, and that it would be necessary to instal the key personnel in the ROTUNDA, with the remainder as far afield as ALDERSHOT.

At the New Year, however, it was decided that adequate accommodation for the whole of the staff could be arranged at BUSHY PARK, and, this according with the desires of General EISENHOWER, now to become Supreme Commander, work on the preparation of the buildings was at once commenced; it was expected that they would be ready for occupation by 15 February,

So everything was prepared for General EISENHOWER to take over his command. COSSAC had grown from a small planning staff into what was in all but name the headquarters staff of the Supreme Allied Commander; and the plans had been produced for the operations which it was to be the responsibility of that Commander to execute. The last weekly staff meeting under General MORGAN was held on 14 January 1944, when it was announced that General EISENHOWER had signified that his designation was to be "Supreme Commander, Allied Expeditionary Force", and that he was expected to be at work by 17 January. COSSAC was then wound up as such, to continue work henceforward as "Supreme Headquarters, Allied Expeditionary Force".

IV. OPERATION 'COCKADE'

As has been seen, it was early realized that the CASABLANCA Conference hopes for an invasion of the Continent during 1943 were doomed to disappointment. Not only were the AMERICAN forces not available for European operations in the anticipated numbers, but also the impending moves in the MEDITERRANEAN absorbed nearly all the resources of landing craft and other essentials of amphibious warfare. Nevertheless, as the Prime Minister pointed out, it was of the utmost importance that our true weakness in respect to cross-Channel operations should be concealed from the enemy. A broad strategic deception policy was needed, which aimed at threatening the GERMANS and ITALIANS on every front where a threat could plausibly be sustained. By that means, the enemy could not only be discouraged from transferring forces to the RUSSIAN front, where they were sorely needed, but he would also be induced to undertake defensive activities such as to cause a wastage of effort and material. Even though we were not strong enough to undertake actual operations in North-West EUROPE, preparations for invasion at different points could be simulated which would pin down enemy forces and might induce the GERMAN Air Force to give battle to our superior air formations.

Thus it came about that the Directive to COSSAC required General MORGAN to prepare plans for "an elaborate camouflage and deception scheme extending over the whole summer with a view to pinning the enemy in the WEST and keeping alive the expectation of large scale cross-Channel operations in 1943".

It was considered essential, if the enemy forces were to be fully contained, that more than one threat should be levelled—not merely directly across the Channel, but against NORWAY and Western FRANCE as well. This would entail the dispersion of such resources as were available to maintain the threats, but it was important that they should not be frittered away in such a manner that in no one place were there any concentrations of a strength likely to make the enemy believe in the possibility of invasion. GERMAN refusals in the past to 'bite' at unconvincing deception threats mounted against NORWAY had to be borne in mind. However, the Allies had an advantage in that they would be working on interior lines from the UNITED KINGDOM, and it was expected that what was lacking in actual resources, such as vehicles, aircraft and boats, could be concealed by the use of dummies and camouflage.

It was evident from the experience of the DIEPPE adventure in August 1942 that something more than a large scale raid was necessary if the GERMAN reserve divisions in FRANCE were to be pinned down as well as their coastal defence forces, since the latter had been able to cope with the raid unaided. Considerable reductions of these reserves had been made when divisions were withdrawn to deal with the RUSSIAN offensive (for the GERMAN army had lost a million men in the past two years), but it was felt that the nature of the enemy communications in FRANCE was such that still further reductions could be made without any essential weakening of the backing to the coastal 'crust'. In the air, the GERMAN fighter force had been strengthened, but only at the expense of other fronts and of reserves. Consequently, despite its apparent strength, it was not thought likely to be committed to intensive fighting unless the GERMAN command was convinced that a serious invasion attempt alas contemplated. The Allied air forces had to provoke the enemy to battle in order to use their superior strength to shatter his power.

In NORWAY the situation was of a different nature. The country was divided into more or less isolated areas, between which it was impossible to move fluid reserves. Moreover, the fact that no GERMAN troops had been removed from NORWAY for use on the RUSSIAN front seemed to indicate that the existing forces there were deemed the minimum necessary to hold the country. Thus, whereas in FRANCE and the LOW COUNTRIES it was essential that the threat of invasion should be levelled against more than one point, in NORWAY an operation of sufficient size seriously to threaten any one area (other than the northernmost tip) would be enough to contain the enemy forces throughout the country.

In considering where the threatened blows were to fall, it was felt that in FRANCE the area in which the GERMAN air force could be brought to battle with the greatest advantage to our own fighters was the PAS DE CALAIS. The smallness of port capacities there, however, was such that if the enemy was to believe in an intended invasion, a feint elsewhere was also necessary, and the most obvious place for a secondary move would be BRITTANY. An intended attack here would have to be simulated on a long-voyage basis, since there wore not enough landing craft available to mount an operation in the WEST on the same scale as that against the PAS DE CAIAIS.

In NORWAY, since no operation could be covered by home-based fighter aircraft, the threat had to be aimed at an area which contained landing fields to which planes could be flown from BRITAIN and from which further operations could then be covered. In this respect, STAVANGER (which had been the hinge on which the whole GERMAN plan of capturing NORWAY in 1940 had turned) was considered the best target.

By 3 June 1943, the appreciation and outline plans, based upon these considerations, were ready. The scheme as a whole was given the code name 'COCKADE', and it consisted of three component operations, 'STARKEY', 'TINDALL' and 'WADHAM'.

Operation 'STARKEY' was the threat against the PAS DE CALAIS area of Northern FRANCE. The object was defined as being to convince the enemy that a large scale landing in the PAS DE CLARIS area was imminent, and thereby to compel the GERMAN fighter force to engage in air battles of attrition in circumstances advantageous to us.

It was planned that the core of the land forces needed to carry out this threat should be provided by the CANADIAN Army, with its deficiencies in respect of combined operations elements made good by the attachment of the Marine Division and four Commandos. In addition, the presence in KENT and EAST ANGLIA of parts of the BRITISH 2nd Army was to be used to convey to the GERMANS that a force of some nine divisions was to take part in the operation. 2nd Army could not be employed in a more active role since it still lacked a commander, one of its corps was engaged in essential training in SCOTLAND, and the organization of the Army was not complete.

By sea, though no figures could be given at this stage, it was decided to assemble on the SOUTH coast a larger number of landing craft than had ever before been seen there. Every endeavour was to be made to provide sufficient lift for one division at assault scales plus four commandos. The naval supporting forces were to include two R-class battleships which were to supplement the air bombardment in neutralising the GERMAN long-range shore batteries.

In the air it was anticipated that the Allies would have a numerical superiority in single-engine fighters of about four to one, though enemy reinforcement might reduce this proportion to two to one. These forces had to be concentrated in South-East ENGLAND, and reserves made available to enable them to engage in intensive fighting over a limited period.

In addition, large day and night bomber forces would be needed and it would be required to divert a considerable proportion of the heavy bomber effort from strategic targets to objectives related immediately to cross-Channel operations for fourteen days prior to the feint attack. It was estimated that 3,000 U.S. heavy and medium day bomber sorties and 3,000 BRITISH heavy and medium night bomber sorties would be necessary, with 15,000 fighter sorties.

The target date for the operation was to be "as late as possible consistent with seasonal suitability of weather conditions", and, all factors being duly weighed, the culmination of the feint was arranged to take place between 8 and 14 September 1943. The operation was to be planned in detail and controlled by the Air Officer Commanding-in-Chief, Fighter Command, in collaboration with the Commanding General, U S. Eighth Air Force, and Naval and Military Commanders nominated by the Commander-in-Chief, PORTSMOUTH and Commander-in-Chief, HOME FORCES, and subject to the latters' concurrence in the Naval and Military operation orders.

During the summer months, deception measures were to be put in hand. Additional landing craft were to be simulated (it was anticipated that 200 of those 'WETBOB' and 'BIGBOB' dummies would be available by August), gliders were to be erected on EAST ANGLIAN airfields, and tented camps set up in assembly areas. Security measures were to be instituted between GRAVESEND and SOUTHAMPTON, and the Ministry of Home Security was to be asked to investigate such matters as evacuation and emergency water supplies in the embarkation zone. Raids by sea, for reconnaissance purposes, were to be made along the FRENCH and BELGIAN coastline, and increased air reconnaissance was to concentrate on the PAS DE CALAIS area.

For a fortnight prior to the date of the feint assault, an air onslaught eras to ho made against enemy fighter airfields and the fighter defence organization in the PAS DE CALAIS, to disrupt enemy defences and bring on air battles whenever possible. The normal sequence of convoys was to be interrupted at the same time, and the troops were to be marshalled to their final concentration points. Between D-9 and D-5, minesweeping was to take place from DUNGENESS and the SOUTH FORELAND towards BOULOGNE. At D-9 the air and sea bombardment of the GERMAN coastal batteries was to commence (in the hope of damaging enemy administration and morale rather than the guns, which were heavily protected); and this was to be followed by day and night attacks on supply dumps, military headquarters and focal points of the communications system. Finally, during the 24 hours preceding the launching of the feint, 'softening-up' attacks were to be concentrated on the beach defences from HARDELOT to the RIVER BRONE, and from AMBLETEUSE to AUDRESSELLES.

In the feint itself, vehicles and drivers, with A.A. personnel, only were actually to embark; but to give realism—and valuable practical experience—the follow-up and build-up formations were to move forward as if to embark. The LSI were to sail from SOUTHAMTON, the balance of the assault craft from DOVER, FOLKESTONE and adjacent ports. The craft were to put to sea, under strong fighter protection and covered by bombing attacks on the enemy coast, and then were to turn about and return to ENGLAND.

Both operation 'WADHAM' and operation 'TINDALL', while similarly deceptive in aim, were concerned only with the building up of sham assault preparations, without proceeding to actual amphibious moves as was the case with 'STARKEY'.

The object of operation 'WADHAM' was defined as being to convince the enemy that a large scale operation was being prepared for the capture of the BREST Peninsula by AMERICAN forces in the early autumn. The culminating date of the operation was fixed as 30 September, in order to convey to the GERMANS that the execution of the operation was dependent upon the prior success of 'STARKEY'. 'WADHAM' was to be planned in detail and controlled by the Commanding General, ETOUSA, in close consultation with the Air Officer, Commanding-in-Chief, Fighter Command, and the Commander-in-Chief, PLYMOUTH.

The general conception Was that a combined seaborne and airborne assault was to be made against the coast of BRITTANY with a view to the early capture of BREST and the subsequent clearance of the whole of BRITTANY. The initial assault was to be carried out by the AMERICAN V Corps, and was to be so timed after the capture of the PAS DE CALAIS area that maximum possible support could be given by the Metropolitan RAF and U.S. Eighth Air Force. Immediate follow-up formations were to sail from WEST coast ports, while the ultimate build-up of an army group was to be effected direct from the UNITED STATES.

Headquarters of U.S. V Corps, with one division, were already in South-West ENGLAND in June, and by the autumn it was expected that a second infantry division, an armoured division and leading elements of an airborne division would have arrived. The fact that there was to be a rapid increase in the arrivals in BRITAIN of U.S. troops from September onwards was to be used to convey the impression of intended build-up of troops sailing direct to FRANCE after the capture of BRITTANY.

No special naval resources were available for the operation, but full use was to be made of the development of the APPELDORE Combined Operations training area and of Exercise 'JANTZEN' on the SOUTH WALES coast in July. Until August, moreover, the enemy could be left in doubt as to whether the main concentration of landing craft in the PORTSMOUTH— SOUTHAMPTON zone (for 'STARKEY') was to sail EAST or WEST.

Air resources were also limited, but a certain number of Thunderbolt (P-47) fighters were available, some gliders could be provided, and bombing attacks could be made by formations of Eighth Air Force prior to their use in 'STARKEY'.

With such slender resources, the element of bluff had to be considerably larger in 'WADHAM' than in 'STARKEY'. Dummy planes and gliders had to be used to give the impression of extensive air preparations. Exercise 'JANTZEN' in July had to be portrayed as a rehearsal for a major attack, and combined operations training activities at APPLEDORE had to be intensified. The concentration of AMERICAN troops in Southwest ENGLAND had to be made as conspicuous as possible, and naval and air reconnaissance visibly increased.

Most important of all, a belief had to be propagated in GERMANY that both existing and forthcoming numbers of U.S. troops for the attack on the Continent were greater than was actually the case. It was believed that the GERMAN Intelligence service was sufficiently uncertain about such questions as to make it reasonable to suppose that exaggerations of the order of 30% might be swallowed. To achieve this, controlled 'leakage' of information was necessary in the UNITED STATES as well as in BRITAIN, and accordingly Maj-General ROSCOE B. WOODRUFF, Commanding VII Corps, came over in July for indoctrination into the part to be played in the STATES. The readiness of certain divisions to sail for EUROPE was to be given out before they were actually prepared, and the numbers who did sail were to be exaggerated. A statement was also required that the headquarters of expeditionary forces had been set up on both sides of the ATLANTIC; and the names of commanders and formations affected were to be 'revealed'. Not only

would the establishment of an AMERICAN Expeditionary Force Headquarters in BRITAIN have the effect of inducing the enemy to believe that there were many more U.S. forces in the country than was actually the case, but it would also be of positive value in that it would enable the staff nucleus to get 'run in' in good time for later operations when the AMERICAN forces would in truth be in the UNITED KINGDOM in strength for the 'OVERLORD' plan of invasion.

The object of operation 'TINDALL' was defined as being to mount a major deceptive threat aimed at STAVANGER with a view to pinning down GERMAN forces in NORWAY during the summer and early autumn. The culminating date of the operation, though later postponed, was originally envisaged as between 12 and l8 September 1943. The operation was to be planned in detail, and controlled by the General Officer Commanding-in-Chief, Scottish Command, in close collaboration with the Air Officer Commanding to be deputed by the Air Ministry, and in consultation with the Commander-in-Chief, ROSYTH.

The general conception was that preparations were to be made for a heavy scale airborne attack, followed by the landing of seaborne forces, with a view to the capture of the port and airfields of STAVANGER. The force was to consist of a corps of three divisions (one mountain), three parachute brigades and two airlanding brigades, with considerable heavy and medium bomber support, and under cover of the Home Fleet. The initial assault would have to be airborne, with commando assistance, on a scale large enough to allow of the neutralisation of the coast defences prior to the arrival of the main seaborne expedition; the defences were too strong and the beaches too difficult to allow of a frontal assault from the sea in the fires place.

To support this conception, Headquarters I Corps (British), with at least three field force divisions—including 52nd (Mountain) Division—would be in training in SCOTLAND during the summer. Sufficient reconnaissance and operational training resources existed in SCOTLAND to cover the air aspect of the deception scheme. No special naval resources were available, but in view of the constant arrival and departure of convoys in the CLYDE, and the physical presence of the Home fleet with auxiliary aircraft carriers in SCOTTISH waters, it cannot considered that anything further was required.

As with operation 'WADHAM', an essential feature in 'TINDALL' was the bringing to the enemy's notice, by special means, of the supposed invasion plan and the preparations being made with parachutists' training for the task. In the sphere of visual misdirection, gliders were to be assembled on airfields in the KINLOSS-EASTHAVEN group, dummy Blenheim and Boston aircraft displayed, and training flights made in the direction of STAVANGER. In addition, minor reconnaissance activities as soon as the hours of darkness permitted, were to be arranged by the Admiral Commanding ORKNEYS and SHETLANDS in collaboration with the Chief of Combined Operations.

It is to be noted.that, whereas WADHAM was essentially a diversionary operation intended to reinforce 'STARKEY', the 'TINDALL' scheme was of an independent nature, designed to contain GERMAN forces in a comparatively isolated portion of Occupied EUROPE.

Operation 'COCKADE' as a whole was intimately connected with the 'RANKIN' plans for a return of Allied Forces to the Continent in the event of GERMAN weakening or collapse. No one could tell with certainty how soon the GERMAN disintegration might occur, but hopes ran higher in the early summer of 1943 than was subsequently to prove justified. It might be found that GERMANY would retire before the threat of invasion offered by 'COCKADE'—a possibility which was devoutly to be desired: ''If we

can win the war without a battle", said General MORGAN, "so much the better". He pointed out to his staff the necessity of remembering that the pretence of invasion might be transformed into the actuality of occupation: "The deception staffs should frame plans that will derive reality from the physical preparations which must in any case be made for our return". If GERMAN forces were withdrawn from the occupied countries, then the resources assembled for the sham operations of 'COCKADE' would have actually to be employed. The landing craft on the SOUTH coast would not put to sea only to turn back again, but could be used to transport all available forces to FRANCE; the AMERICAN forces in the U.S.A. of which use was being made in the 'WADHAM' bluff, would have to cross the ATLANTIC to Western FRENCH ports; and the airborne attack on STAVANGER would become an actual seizure instead of a feint. It was expected that by mid-August it would be evident whether such a GERMAN withdrawal was going to take place in 1943 or not.

'COCKADE' might also be regarded as an essential prelude to 'OVERLORD'. The manner of the enemy's reaction to the former would give some indication of the reception likely to be accorded to a full scale invasion attempt, and the air attacks on Northern FRANCE were part of the necessary softening process which must precede a major assault.

With the approval by the Chiefs of Staff of the outline plans for 'COCKADE', the detailed planning of a component schemes went ahead rapidly, and by l July 1943 it has reported that all three parts had been issued to the Commanders concerned to prepare for executive action.

Inevitably, difficulties were not lacking. In the air sphere, it was found that the gliders which were required for operation 'TINDALL' were needed elsewhere for training purposes, the interruption of which would entail delays in the preparations for the 1944 invasion of the Continent. The gap had to be filled as far as possible with dummies. At the same time difficulties were caused by naval assault forces which were needed to support the 'TINDALL' deception being utilized in raids against the NORWEGIAN coast without correlation to the 'TINDALL' plans,

Operation 'WADHAM' suffered delays in July and early August owing to the dearth of aircraft, while the dummy glider programme was held up by the shortage of mechanics to assemble the machines. In mid-August, however, matters improved and preparations were accelerated, despite interference by bad weather which put a stop to both reconnaissance and bombing flights. Meanwhile the 'leakage' of information in the UNITED STATES, concerning the training of forces and their sailings to BRITAIN, was under way.

Even for 'STARKEY', air power was not so readily available as had been hoped. The Vice-Chief of the Air Staff pointed out on 26 August that the operation plans would interfere with the strategic bombing offensive against GERMANY which had been accorded the highest priority. Consequently the 'STARKEY' requirements had to be filled from O.T.U. and Wellington squadrons, but extra weight was promised for the final night of the operation to give at least 200 sorties. The assault against the FRENCH coastal defences was also weakened by the non-availability of the two battleships which it had been hoped to employ for naval bombardment purposes, these being taken instead for duties in the MEDITERRANEAN. COSSAC had considered that the presence of these vessels in the Channel would act as "cheese in the mousetrap" to the GERMAN Air Force. On the analogy of the run through the Channel by the SCHARNHORST and GNEISENAU, it was hoped that the bait would be sufficient to lure the GERMAN planes within range of the allied fighters; but the Admiralty, remembering perhaps the fate of the RENOWN and the PRINCE OF WALES, were unwilling to take the risks involved.

The chief difficulty in respect of operation 'STARKEY', however lay in the shortage of landing craft sufficient to give an air of realism to the 'invasion' preparations. As early as 10 May this shortage was accepted as a possible limiting factor, and COSSAC urged the necessity of obtaining further craft from reserves which were understood to be in AMERICA. In June the shortage was aggravated by the allocation of a number of craft to operation 'LIFEBELT', and it was feared that the amphibious feint element in 'STARKEY' might have to be abandoned. So drastic a revision of plans was not resorted to, however, and the shortage of craft was to some extent made up by the use of dummies. It was expected that at least 75 more major and 100 more minor landing craft would be available than had been used at DIEPPE, and it was intended to display a further 175 dummies. This shortage of craft, on the scale necessary for invasion purposes was, however, to remain a stumbling block in the way of planning operations 'RANKIN' and 'OVERLORD' for many months t o come.

The lack of essential resources indeed caused some doubts during the summer as to the feasibility of attempting to carry through operation 'STARKEY' at all. COSSAC reported in July that the scale of preparation required for the attack on SICILY made it an "open question'' whether 'STARKEY' would achieve its object; but it was hoped that the increasing pressure in Eastern and Southern EUROPE would produce the required reactions to the feint of invasion in the North-West.

The strain on the resources necessary for the accomplishment of 'COCKADE' was to some extent eased by the decision reached in August to postpone operation 'TINDALL' until November. The suggestion for this postponement was agreed by the Chiefs of Staff Committee on the grounds that on our side greater forces would be available at a season when 'STARKEY' was concluded and cross-Channel operations no longer feasible, while on the GERMAN side it would probably be necessary then to make withdrawals of troops from NORWAY to reinforce the RUSSIAN front during the winter. The scheme was accordingly adopted to proceed with the preparations for 'TINDALL' as originally arranged until 25 August, when it would be postponed, and apparently dismounted, until after the conclusion of 'STARKEY' . When then resumed, the GERMANS, having learnt that 'STARKEY' and 'WADHAM' were only feints or rehearsals, might conclude that the invasion of NORWAY was after an going to be the only invasion operation to be carried out in 1943.

Meanwhile the process of putting 'STARKEY' and "WADHAM', into operation went forward. By 9 August it was reported that 'STARKEY' was "gathering momentum", and on 16 August that the enemy was showing signs of beginning to react with reconnaissance and bombing activities over the SOUTH coast. By 23 August it was considered that these preliminary reactions were as great as could be expected since the enemy was faced with so much greater commitments in ITALY and RUSSIA than had been envisaged in the spring. On 2 September, however, General MORGAN reported that the break-up of the weather had upset the plans for air operations and minesweeping, and that when the minesweepers eventually got out their activities produced nothing more than desultory gunfire from the enemy side of the Channel. The operation was nevertheless persevered with, and on 9 September both 'STARKEY' and 'WADHAM' were concluded and the forces assembled for them began to be dispersed.

During the remainder of September and October, attention was concentrated on the implementation of operation 'TINDALL'. The display of aircraft and gliders—real and dummy—was again set up, and preparations for an. airborne invasion simulated until on 5 November 'TINDALL' was concluded and the preparations dismantled. This brought the whole

of 'COCKADE' to an end.

A major problem which had presented itself during the planning of 'COCKADE'—and especially of 'STARKEY'—lay in the question of the publicity to be given to the enterprise. The fundamental difficulty, as the COSSAC Intelligence staff pointed out as early as May 1943, lay in the reconciliation of the dual necessities of persuading the GERMANS that operation 'STARKEY' was to be an invasion and the patriot forces that it was to be only a feint. Even though no actual landings were to take place, the effect of the preparations upon the patriots might well be to heighten to flash-point expectations of relief and then, at the very onset of winter, to disappoint them. If a premature uprising did take place, the effect of drastic GERMAN punitive measures might well be to render any future patriot co-operation with the Allies out of the question. To counter this risk, it was decided to employ every means—radio, leaflets and secret agents—to impress upon the subject peoples the necessity of accepting discipline from LONDON and of taking no violent action until they received definite orders from that source.

The troops taking part in Exercise 'HARLEQUIN' (as the amphibious operation implementing 'STARKEY' was called) were also told that they were going on an exercise, not that they were about to invade EUROPE. This drew forth protests from those responsible for propaganda to the enemy that all their attempts to persuade the GERMANS of the imminence of invasion were being undone by the disclosures of the true nature of the operations. General MORGAN however stood firm by his policy of telling the truth to the troops, but, as he said, it was quite another matter to make them believe it.

A further complication was introduced by the manner in which the AMERICAN and BRITISH press reacted to the 'STARKEY' preparations—"exactly as the Chiefs of Staff would not have wished". The episode nevertheless provided a useful lesson for those whose task it would be to exercise control over the press when the time came for the much greater undertaking of operation 'OVERLORD'.

Prior to the culmination of operation 'STAPKEY', a draft communique was framed for issue to the press on the conclusion of the enterprise. This was to have told the public that a full scale amphibious exercise had taken place, with concentrations of land and air forces in South-East ENGLAND and of landing craft on the SOUTH coast. The operation had been carried out in "realistic detail down to embarkation points". The assault troops had not embarked, but a proportion of the landing craft had been exercised in the Channel, with naval and air protection. Valuable lessons had been learned by the Service and Civil authorities. The enemy had reacted with mine-laying and coastal artillery fires and had suffered considerable casualties in air and E-boat attacks.

In view of the actual lack of enemy reactions, this communique was in fact never issued. On 8 September the Chiefs of Staff decided on a much briefer account: "A full scale amphibious exercise has recently taken place in the ENGLISH CHANNEL. It has been most successful, and valuable lessons have been learned by the Service and Civil Authorities concerned".

Generally speaking, operation 'COCKADE' did not achieve the results expected of it. Operation 'WADHAM' had never been regarded, even by its begettors, as likely to produce startling results; it was little more than a diversion from 'STARKEY'. The enemy took practically no notice of it, and indeed the absence of Naval lift other than a very few dummy landing craft made it obvious that no mayor danger was to be anticipated.

Concerning operation 'TINDALL', the General Officer Commanding-in-Chief, Scottish Command, reported that "no reports received covering the period of the operation give any indication that the GERMANS suspected that anything

was on foot". Although wireless and visual deception had been carried out, the time for adequate preparations had been too short. Glider flights could not be carried out owing to the absence of tug aircraft, and the airfield defences were unconvincing by reason of the lack of real LAA guns or flash simulators for the dummies. In any case, the GERMANS had ignored the whole business, and, judging by this lack of enemy air reconnaissance which the operation was designed to deceive, it would appear that 'TINDALL' was a failure.

A similar disappointment in respect of the main object of the operation was reported concerning 'STARKEY' by the Air Officer Commanding in Chief, Fighter Command. Neither the air bombardments of the enemy defenses nor the sailing of the assault and M.T. convoys produced any appreciable reaction by the GERMANS in the form of air or surface attacks or of coastal artillery shelling. Only the minesweeping provoked any reply in the form of counter minelaying and some gunfire. The desired air battles which were to lead to the attrition of the GERMAN Air Force did not materialise, although a considerable number of sorties were flown over the PAS DE CALAIS when the weather permitted. At the same time, the Intelligence Branch of COSSAC reported that there was no evidence of any military defensive moves in FRANCE; on the contrary, the movement of troops from FRANCE to ITALY had not been interrupted, and the GERMAN forces had not therefore been pinned down away from the active fronts. Apparently the enemy had realised that the Allies were in no position to undertake extensive operations in North-West EUROPE at the same time as they were engaged so fully in the SOUTH.

BERLIN radio might report that troops severe concentrated in South-East ENGLAND and that "as far as the second front in Western EUROPE is concerned it is beyond doubt that the BRITISH have made preparations on a considerable scale"; but it was clear that GERMANY would not believe in the invasion bogey until it became evident that the Allies had adequate resources in landing craft to enable them to carry out such an enterprise on a fuel scale.

On the other hand, 'WADHAM' had been a useful exercise to the staff of the UNITED STATES forces which were now being built up in BRITAIN; and Exercise 'HARLEQUIN' had provided an invaluable experience to the naval, army, air and civil authorities in the work of marshalling forward troop concentrations and of preparing for an invasion of the Continent such as it was planned to carry out in 1944. If only in these respects, operation 'COCKADE' was not entirely without its value.

V. OPERATION 'RANKIN'

In the original Directive to COSSAC, the second object for which General MORGAN's staff was instructed to prepare plans was "a return to the Continent in the event of GERMAN disintegration at any time from now onwards with whatever forces may be available at the time". The schemes evolved to meet such an eventuality were given the name of operation 'RANKIN'.

The basic principles upon which the planning was to be made were outlined by General MORGAN in a paper which he issued to his Principal Staff Officers on 22 May 1943. The first necessity, he pointed out, was for the scheme to be elastic, since action under it might have to be taken at any time: the choice of date lay not in the allies' hands but in those of the enemy. Thus the action, whenever the time came, would have to be taken on 'ad hoc' lines. The task before COSSAC was comparable to that of producing in peace-time a scheme of mobilization to take place on some unpredictable day, until which the normal peace-time avocations of the country must suffer dislocation to the minimum extent. Plans had to be drawn up to

facilitate speedy action at any moment, but they must not be such as to interfere with the other allied schemes for deception on the one hand or invasion in the face of full GERMAN resistance on the other.

One thing was certain: that any GERMAN collapse would be sudden. Yet it should be possible for the Intelligence staff to obtain information which would enable COSSAC to forecast the event with some margin of time for active preparations. The enemy 'disintegration', however, might not take the form of immediate complete collapse; GERMANY might withdraw gradually from the territories which she had occupied since 1940, or it might be necessary for the allies to break through "an enemy screen, still fighting but lacking supplies or reserves". Whichever was the case, COSSAC had to plan not only for the landing on the Continent but for the occupation of all the countries at present under GERMAN control.

The forces to be employed in the task would have to be whatever might at the given moment be available. At present, in May 1943, these were such as to make it impossible to consider anything more than a return under conditions of complete GERMAN collapse and the abandonment of all resistance; but the time would soon come when greater risks could afford to be taken and when the occupation could be carried out with the aid of AMERICAN forces sailing direct to the Continent from the UNITED STATES. In the meantime the bulk of the forces employed in a return to the Continent would have to be BRITISH, though at least token U.S. contingents would accompany them. Every available man and every available weapon of war would be needed, and they would be drawn from every possible source—including training schools, depots and defence formations, since those would no longer be a necessity; the occupation Could become "the one interest, paramount in the world". Speed in action would be all-important also, and the possibility of using the resources of Bomber and Training Commands to transport forces to the Continent had to be considered, for their normal functions also would be at an end.

It was of the utmost importance that official conventions should not be permitted to interfere with proceedings of so necessarily revolutionary a nature. Every regulation, every restriction must go by the board if necessary. The responsible authorities must be made to realize that the object in this case should be "to transport the army to EUROPE rather than to obey Board of Trade Regulations". The forces would have to be crammed into whatever floated or flew, and risks would have to be run for the sake of speed in attaining the main object.

On the other side of the picture, plans had to be made for the handling of the liberated peoples, for there was bound to be a "tremendous surge from beneath amongst people of all classes and all callings". Hence, an effective Civil Affairs organisation was essential, and contact with the European Allied governments in exile would have to be made.

A month later, on 23 June 1943, the Principal Staff Officers were instructed to consider the advisability of setting up the advanced guards of the forces of occupation forthwith, in order that when the time came the enemy-held territories could be entered at a moment's notice. That some delay should elapse before the main forces could follow those advanced guards was inevitable but it might not be a disadvantage. There was bound to be a certain amount of 'blood-letting' among the liberated peoples eager for revenge upon their quislings, and the delay in completing full occupation would both save the ANGLO-AMERICAN authorities from contamination and yet avoid the necessity of interfering and being compelled morally to save their enemies from very well-merited fates.

In the summer of 1943, it was felt that GERMAN weakening or collapse might become manifest very soon, and the COSSAC staff was told in July that planning for 'RANKIN' was to be considered its most urgent task.

GERMANY might retire the threats offered in operation 'COCKADE', or the pressure in RUSSIA or ITALY might bring about as sudden a collapse as in 1918. Even though the scales might appear to be weighted in favour of GERMANY until May 1944, COSSAC had to be prepared for a return to the Continent as from 1 August 1943, and that would, in fact, be the target date for operation 'RANKIN'. If GERMAN resistance ceased entirely, the Allies would have to go direct to the Fatherland itself; if the enemy retired from NORWAY or withdrew his FRENCH and LOW COUNTRIES garrisons behind the SIEGFRIED LINE, the Allies would have to aim first at the seizure of air bases in the erstwhile occupied countries from which support could then be afforded to further land operations against GERMANY.

Despite the apparent urgency of the task, the work of producing the necessary plans did not proceed as rapidly as had been hoped. By the end of July 1943, General MORGAN complained that, despite a vast amount of discussion, "no progress whatever" had been so far made in the evolution of any sort of scheme to cope with the state of affairs outlined in his Directive. Yet "the sum total of all the various factors now operating cannot be far from that of the factors which caused the collapse in 1918". The apparent difficulties in respect of lack of resources must not stand in the way of making adequate preparations: "I do not accept limitations imposed by the absence of craft and shipping. Should the enemy retire, we shall have to go after him, even if we have to swim".

It was necessary to evolve "an organisation for adequate and continuous reconnaissance, centrally controlled; an organisation for the throwing out of advanced guards in the appropriate directions and appropriately composed to suit the terrain of the moment; an organization to seize and secure and bring into use an adequate group of airfields on the Continent; and an organization for the 'ad hoc' co-ordination of all available resources in transportation and maintenance". Given these, the Allies would know when to strike, where to strike and how to strike

Despite the delays, by 13 August COSSAC was able to submit the basic plans for operation 'RANKIN' to the BRITISH Chiefs of Staff for their consideration. In the accompanying letter it was pointed out that there were still considerable difficulties to be overcome. It was, for instance, difficult to plan operations in detail before the clear definition of the political policy to be adopted as between BRITAIN and the UNITED STATES on the one hand and the occupied countries—FRANCE, BELGIUM, HOLLAND, DENMARK and NORWAY—on the other. Secondly, the institution of a Civil Affairs staff was necessary to carry out the orders of the Supreme Allied Commander for the control of the civil populations within the zones of his forces. Collaboration with RUSSIA would be necessary in determining spheres of occupation, and propaganda at home would be required to secure the support essential for the great effort which operation 'RANKIN' would entail.

The outline plans for operation 'RANKIN' as now evolved were designed to cope with three situations any one of which might possibly arise. 'RANKIN' Case A provided for a return to the Continent under conditions of such substantial weakening of the strength and morale of the GERMAN armed forces as would permit of successful assault with the ANGLO-AMERICAN forces available prior to the target date of operation 'OVERLORD'. 'RANKIN' Case B was concerned with the method of return should the GERMANS withdraw, totally or partially, from the occupied countries. 'RANKIN' Case C dealt with the action to be taken in the event of unconditional surrender by GERMANY and the cessation of all organised resistance in North-West EUROPE. Thus in Cases A and B the object was to effect a lodgement on the Continent from which the defeat of GERMANY could be completed; in Case C, on the other hand, the aim was to occupy as rapidly as possible appropriate areas from which to take steps to enforce the terms of unconditional surrender laid down by the allied governments.

It was assumed in making these plans that GERMANY's general situation was already ''verging on the desperate". She was faced with a serious situation on the RUSSIAN front and another threat was developing in ITALY. The U-boat campaign had suffered a severe set-back, and the strain of the Allied air offensive was beginning to tell on production and morale alike. No general reserves of forces remained untapped by her, and she could only reinforce one front at the expense of weakening another. Should the GERMANS have to withdraw anywhere, it would probably be in the WEST first; with the recollection of their own conduct in the RUSSIAN territories which they had overrun, they might well fear the retribution which would be their due were the RUSSIANS to continue their advance into the Fatherland.

In 'RANKIN' Case A—the return to the Continent in the event of a substantial weakening of GERMAN resistance in FRANCE and the LOW COUNTRIES—no assault against organized resistance was feasible before the end of 1943 unless there were clear indications that GERMANY was very near collapsing and unless a suitable naval assault force was available to the Allies. Subsequent to 1 January 1944, however, an assault could be undertaken against weak opposition to secure a strictly limited objective, and from March 1944 onwards conditions would become more favourable until an 'OVERLORD' situation was reached. It was therefore considered feasible from January onwards to undertake operations, against a weakened enemy, to capture the COTENTIN PENINSULA, provided that the port of CHERBOURG could be seized within the first 48 hours. Diversionary operations would probably be necessary in the PAS DE CALAIS area and in the SOUTH of FRANCE. All this, as will be seen, was in effect merely a modified form of the 'OVERLORD' plan of invasion.

In 'RANKIN' Case B—the course of action for a return to the Continent in the event of GERMAN withdrawal from the occupied countries—it was considered that the enemy would retire gradually from the extremities of the area at present under his control: Western FRANCE and NORWAY. In this case, it was necessary, for political as well as strategic reasons, to occupy the areas vacated, but it was important that the main forces of the Allies should not thereby be tied down far from the eventual centre of action.

The chief task in NORWAY would be the establishment of air bases, with RADAR installations, from which the northern supply route to RUSSIA could be guarded and the entrance to the BALTIC blockaded. One brigade group for Northern NORWAY and one division for Southern NORWAY would be enough to secure these bases and to support the NORWEGIAN contingent in the work of rehabilitation.

In FRANCE, BORDEAUX and the Western Ports would probably be the first places to be evacuated, and the PAS DE CALAIS (being an essential protection to the enemy lines of communication) last, until ultimately all the GERMAN forces were behind the SIEGFRIED LINE. It would be necessary to send one brigade group each to occupy BORDEAUX, BREST and NANTES (partly for rehabilitation purposes, and partly to prepare the ports for the reception of AMERICAN forces sailing direct from the UNITED STATES), but the first point of entry for the main forces would be CHERBOURG. Then, as the GERMAN withdrawal proceeded, HAVRE and ROUEN would also be utilised as principal bases. The seizure of CHERBOURG would be by U.S. forces, while HAVRE and ROUEN would be taken by the BRITISH. From these bases, the Allies would establish a line along the SOMME from which to press North-Eastwards through the PAS DE CALAIS to BELGIUM, opening up ports as they proceeded and establishing airfields from which the advance could be covered and the attack on GERMANY itself concentrated. At the same time, MEDITERRANEAN forces would be required to occupy MARSEILLES and TOULON, and to move Northwards thence in LYONS and VICHY.

The object of 'RANKIN' Case C—the return to the Continent following GERMAN unconditional surrender and the cessation of all organized armed resistance—was to occupy, as rapidly as possible, appropriate areas from which to take steps to enforce the terns of surrender and also to carry

out the rehabilitation of the liberated countries. For this purpose, it was considered that the areas first to be seized should be the JUTLAND PENINSULA, the ports of BREMEN, HAMBURG and KIEL, and the large towns in the valleys of the RUHR and the RHINE. From those bases, the allied forces would occupy the valley of the RHINE from the SWISS to the DUTCH frontiers, together with the RUHR area, and also DENMARK, SCHLESWIG and HOLSTEIN. They would establish air bases from which to overawe the whole of GERMANY and would disarm the GERMAN forces both stationed in these zones and returning through them from the formerly occupied countries. At the same time, selected ports in Western FRANCE and the LOW COUNTRIES would be opened, and steps taken to establish control in the capital cities, to institute measures of rehabilitation and to assist the process of GERMAN disarmament. Control would likewise be established for the same purposes in NORWAY, while MEDITERRANEAN forces would land in Southern FRANCE and occupy VICHY.

It was estimated that 24 divisions would be required for the primary areas of occupation: 7 for DENMARK and North-West GERMANY, 6 for the RUHR and 11 for the RHINE valley. Other troops—non-field force formations wherever possible—would be required in support of the various national contingents charged primarily with the rehabilitation of their respective liberated countries. These requirements would not, however, become available from the U.S. and BRITISH forces stationed in the UNITED KINGDOM until March 1944, and it would be necessary to make good the shortage with formations from the MEDITERRANEAN and with AMERICAN forces in the UNITED STATES who had been earmarked to sail direct to FRANCE under the 'OVERLORD' plan of invasion.

In considering the respective spheres of occupation to be assigned to the U.S. and BRITISH forces, it was thought best that the former should be responsible for FRANCE, BELGIUM and the RHINE valley from the SWISS frontier to DUSSELDORF, while the BRITISH should take charge of HOLLAND, DENMARK, NORWAY and North-West GERMANY from the RUHR valley to LUBECK inclusive. This division of spheres of occupation would accord with the intended plan of entry of the Allies into EUROPE under 'OVERLORD'—the BRITISH on the left flank, the AMERICANS on the right. In BERLIN itself an international force could be set up.

These basic plans for operation 'RANKIN' Cases A, B and C wore submitted to the QUADRANT Conference in August 1943, and there approved by the Combined Chiefs of Staff. Planning in detail was accordingly not put in hand. There was, however, some difference of opinion as to which of the 'RANKIN' Cases should receive priority attention. The Chiefs of Staff considered that Case A was the most likely to be needed, whereas COSSAC wished to concentrate on case C—even before 'OVERLORD'—still believing in the possibility of a sudden GERMAN collapse under the strains to which she was likely to be subjected during the coming winter. In accordance with this belief, it was ruled in early October that Case C be given priority, and on 30 October the directive for the operation (excepting the parts concerning NORWAY and the CHANNEL ISLANDS) was issued to the Commanders of the U.S. First Army Group and the BRITISH 21st Army Group for detailed preparations to be made.

The special outline plans for 'RANKIN' Case C, NORWAY and CHANNEL ISLANDS, were drafted and issued to the Commanders concerned by mid-November. The detailed preparation of the NORWEGIAN scheme was made the responsibility of the General Officer Commanding-in-Chief, Scottish Command, in collaboration with the Commander-in-Chief, ROSYTH, and an officer deputed by the Air Officer Commanding, Air Defence of GREAT BRITAIN; while the CHANNEL ISLANDS scheme was handed over to the General Officer Commanding-in-Chief, Southern Command, in collaboration with the Commander-in-Chief, PLYMOUTH, and an officer deputed by the Air Officer Commanding, Air Defence of GREAT BRITAIN. In both instances, of course, the execution of the plans was to be under the general supervision and control of the Supreme Allied Commander, whenever ho should be appointed.

The planning of Operation 'RANKIN' was necessarily bound up with that of 'COCKADE' and 'OVERLORD', since the three were so intimately interconnected. On the one hand, the threats held out by operation 'COCKADE' might lead to GERMAN withdrawal from the occupied countries, thus creating the conditions envisaged in 'RANKIN' Case B. On the other hand, 'RANKIN' Case A was in effect merely a modified form of operation 'OVERLORD', and by the end of 1943 the planning for it became merged with the preparations for the larger undertaking. Again, there was a possibility that GERMANY would collapse after the initial success of the 'OVERLORD' invasion and thus 'RANKIN' Case C conditions would be created. All the 'RANKIN' plans, moreover, were essentially temporary in nature, for by the time 'OVERLORD' was mounted and prepared the Allies would be in a position to dictate to the enemy concerning the times and places of his withdrawal. Pending that date, 'RANKIN' had to be an elastic plan, designed to take advantage of any deterioration in the enemy situation.

In December 1943, consequent upon the discussions at the SEXTANT Conference, the UNITED STATES Chiefs of Staff suggested that the spheres of occupation allotted under 'RANKIN' Case C should be revised so that the BRITISH took over the Southern Sector and the AMERICANS the Northern Sector. Generals MORGAN and BARKER, however, pointed out that the existing scheme was bound up with the operational plans for the implementation of 'OVERLORD', and that both operations would have to undergo complete revision if the new U.S. suggestions were to be adopted. For the AMERICAN forces landed in the WEST to have to cross the BRITISH operational sphere to take up their new proposed occupational zone in North-West GERMANY would be a process both difficult and dangerous if GERMANY did not collapse promptly on the commencement of the invasion. Moreover, the delay which would be occasioned by the necessary re-planning might seriously jeopardize the chances of success in 'OVERLORD'. If the new spheres of occupation were insisted on, the only practicable solution would be to carry out the existing plan and then to reallocate the forces later when the general situation was sufficiently under control. It was unlikely that such a reallocation would be possible until three months after the armistice, and it would take at least two months to effect.

In the light of these considerations, the suggested changes were referred to the President and Prime Minister, and, as a decision was delayed, the matter ceased to be of primary operational significance. The work of detailed preparation proceeded during December on the lines already determined, and by 14 January 1944 it could be reported that "there did now exist the basis of a plan that could function in some manner" so far as 'RANKIN' Case C was concerned.

Most attention during these winter months had to be devoted to the problems arising in the sphere of Civil Affairs. As has been seen, considerable delays were experienced in the establishment of a full Civil Affairs branch of COSSAC, and this consequently held up the 'RANKIN' planning. The problems of dealing with the populations both of the occupied countries and of GERMANY herself were in the forefront of the questions arising concerning the development of the 'RANKIN' schemes, and especially was this the case in 'RANKIN' Case C. Not only had the feeding, policing, housing and general administration of the ordinary civil populations to be considered, but there was also the enormous problem of the 'displaced persons'—the millions who had been uprooted from their homes as evacuees from war-stricken areas, political refugees, escaped prisoners of war and all those who had been forced to labour in GERMAN factories or under the TODT organization throughout occupied EUROPE. All these would begin the trek homewards as soon as GERMANY collapsed, and it would be the responsibility of the Allies to check indiscriminate migration, to canalize the movements off military routes, to prevent the spread of disease, and to provide food and clothing. All this, in the event of a GERMAN collapse, would be a task more urgent perhaps than any purely operational duties of the Supreme Commander.

By the New Year, however, it was realized that the likelihood of 'RANKIN' conditions developing in EUROPE prior to the implementation of operation 'OVERLORD' was becoming more and more remote. In the autumn of 1943, the deception threats of operation 'COCKADE' had failed to shake GERMANY. Then it was hoped that the RUSSIAN winter offensive, coupled with the Allied pressure in ITALY, might compel the enemy to withdraw in the WEST in order to reinforce the Eastern front There the actual threat was greatest. In ITALY, however, the CASSINO Line held despite the ANZIO diversion, while in RUSSIA the GERMANS retreated before the Red Army onrush in a planned and orderly manner and succeeded in keeping their armies intact. It was clear that the conditions envisaged in 'RANKIN' Cases A, B and C were not to materialise yet. The conclusion was inevitable that not until an invasion in force was made was it likely that GERMANY would retire from the occupied countries or her military porter disintegrate.

The 'RANKIN' plans thus became subsidiaries to 'OVERLORD', their functions being to indicate the steps to be taken subsequent to the enemy's weakening, withdrawal or collapse as a result of the mass Allied invasion of the Continent in 1944.

VI. OPERATION 'OVERLORD'

The third, and greatest, task for which COSSAC was instructed by its original Directive to prepare plans was "a full scale assault against the Continent in 1944 as early as possible". This undertaking became known as operation 'OVERLORD'.

Although the problems involved in a return to the Continent had been discussed by the Combined Commanders prior to the creation of COSSAC, the task before General MORGAN's staff was in many respects a novel one. In the past the prospect of adequate forces being available for an invasion of EUROPE was so distant that most attention was necessarily focussed on a long term policy of attrition against GERMANY; but now AMERICAN forces would be flowing into BRITAIN at an ever-increasing rate, and it was vital therefore that the COSSAC staff should accept a certain reorientation of outlook. They were to prepare, not for a gradual wearing out of the enemy, but for his destruction by a knock-out blow at the earliest possible moment,

This knock-out blow was to be delivered from the UNITED KINGDOM through Northern FRANCE and the LOW COUNTRIES. The ANGLO-CANADIAN armies would be based on South-East ENGLAND, with the advanced guard of the AMERICAN armies to the WEST of them. These would together make the initial assault, and then the main bodies of the AMERICAN forces would sail direct from the UNITED STATES to the captured bridgehead until Allied armies of the order of 100 divisions mere concentrated on the Continent to complete the destruction of GERMAN power. It was upon this general scheme of action that COSSAC's subsequent planning of the operations was founded.

On 25 May the Combined Chiefs of Staff issued a supplementary directive to COSSAC on these lines. The aim was defined as being to secure a lodgement on the Continent, from which further offensive operations could be carried out; and to this end plans were to be drawn up for the seizure and development of Continental ports in order that the initial assault and build-up of forces might be augmented by shipments from the UNITED STATES or elsewhere of additional formations at the rate of three to five divisions per month. The target date for the operation was to be 1 May 1944, and the outline plans were to be in the hands of the Combined Chiefs of Staff by 1 August 1943. A total of 29 divisions in all could be available for the assault and immediate

build-up. Of these, 5 infantry divisions were to be simultaneously loaded in the landing craft, with 2 more to follow at once, as well as 2 airborne divisions, making a total of 9 assault divisions; the remaining 20 divisions would be available for movement to the lodgement area as quickly as the build-up could be achieved. One FRENCH division might also take part in these initial operations.

During June and July 1943 the development of an outline plan based on these estimated resources proceeded. The fact that simultaneous landing craft lift for only 5 divisions had been allotted made any dispersion of effort out of the question, but it was nevertheless decided that more than one line of action should be investigated. The alternatives considered were for assault in the CAEN or the PAS DE CALAIS areas. Both, naturally, presented difficulties, but it was hoped that these would be largely overcome by the Allies' Naval and Air supremacy. The BRITISH and U.S. Naval Sections of COSSAC were accordingly instructed to prepare a joint appreciation and plan to ensure the safety of the attacking forces from surface and submarine naval attack, while the BRITISH and U.S. Air Sections were similarly to consider how freedom of action by land and sea from air attack might be obtained between CHERBOURG and DIEPPE. The Army Operations Sections of both nations, with naval assistance, were to report on the feasibility of varying the organization of the assault forces. An appreciation and outline plan for an assault in the CAEN region was to be prepared by the U.S. Army members of the Operations Section, while a similar plan for an assault in the CALAIS region was to be produced by the BRITISH Army members of the Operations Section; in each case liaison with the other nation was to be maintained in drawing up the schemes. In the meantime the Administrative Section was allotted the task of producing an appreciation and outline plan for the preliminary concentration of troops in the UNITED KINGDOM.

By these means the necessary data was accumulated upon which it was in due course possible to base the 'OVERLORD' appreciation and outline plan. As the work proceeded, the General Officer Commanding-in-Chief, 21 Army Group, and the Air Officer Commanding Tactical Air Force, were brought into the discussion; and particulars were also known to the Commanding General, ETOUSA, the Air Officer Commanding-in-Chief, Allied Expeditionary Air Force (Designate), and the Allied Naval Commander-in-Chief (Designate). At the end of June, a very valuable conference, attended by General MORGAN and his Principal Staff Officers, was held by the Chief of Combined Operations at LARGS, when many of the major problems connected with the projected Continental assault were resolved.

The basic factor in determining where the initial assault was to be made lay in the requirement that the lodgement area should contain sufficient port facilities to maintain a force of some 26 to 30 divisions and enable that force to be augmented by follow-up shipments from the UNITED STATES or elsewhere of additional divisions and supporting units at the rate of 3 to 5 divisions per month. In the earlier phases of the assault, however, maintenance would necessarily have to take place over the beaches until the ports could be put into working order.

For such purposes, the PAS DE CALAIS region offered advantages in that its proximity to ENGLAND would facilitate air support and a quick turn round for shipping. On the other hand, its beaches, while favourable to the actual landing, lacked good exits to the hinterland, it was the most strongly defended area on the whole FRENCH coast, and it was a focal point of the enemy fighter air forces disposed for defense. Moreover, the area did not offer good opportunities for expansion of the lodgement zone, and it would ho necessary to develop the bridgehead to include either the BELGIAN ports as far as ANTWERP or the CHANNEL ports Westward to include HAVRE and ROUEN.

An attack on the COTENTIN PENINSULA would have a reasonable chance of success and could gain the invaluable port of CHERBOURG but the area lacked suitable airfields, and it would be difficult to expand the bridgehead beyond the neck of the Peninsula itself.

In the CAEN sector, defences were relatively light and the beaches were of high capacity and sheltered from the prevailing winds. The terrain, moreover, was suitable for airfield development and for the consolidation and subsequent expansion of the bridgehead. Air support to the initial assault at such a distance from the ENGLISH coast would present the chief difficulty, but the enemy was also unfavourably situated in this respect. From a bridgehead in the CAEN sector it was thought possible to seize the BRITTANY ports between CHERBOURG and NANTES, and through them to build up sufficient forces for the final advance Eastward.

In the light of these considerations, it was decided that the initial landing on the Continent should be effected in the CAEN area with a view to the eventual seizure of a lodgement area comprising the CHERBOURG-BRITTANY ports.

The preliminary requirements to ensure the success of this operation included first and foremost a reduction of the GERMAN fighter air force and its pre-occupation away from the assault area. This condition above all others would, it was considered, dictate the date on which the amphibious assault could be launched. It was also essential that not more than 12 reserve mobile field divisions should be available to the GERMANS in FRANCE, and that not more than 3 of these should be available in the CAEN area on D-Day, 5 by D plus 2 or 9 by D plus 8. The number of landing craft in the hands of the Allies was also dangerously low, and at least a 10 per cent increase was considered necessary to provide a safety margin, while lift for a further assault division striking at other beaches would be an advantage. Finally, some form of artificial harbour would be necessary to facilitate maintenance over the beaches for a period of some three months pending the full opening-up of the deepwater BRITTANY ports. These conditions having been fulfilled, it was thought that the proposed operations would have a reasonable prospect of success.

The preliminary phase of the operation was to start at once with action designed to reduce the GERMAN air forces and to ensure the progressive destruction of the GERMAN economic system and the undermining of GERMAN morale. To these ends, every form of pressure—air and sea activities, propaganda, political and economic moves, and sabotage—would be applied. At the same time, diversionary operations would be instituted to contain the GERMAN forces away from the CAEN area.

During the preparatory phase—the period immediately preceding the launching of the assault—air action against airfields and communications would be intensified, while three naval assault forces were assembled at ports along the SOUTH coast of ENGLAND and two more, carrying the immediate follow-up forces, in the THAMES ESTUARY and on the WEST coast respectively.

The assault would be launched with a short air bombardment of the beach defences, after which three assault divisions would be landed simultaneously on the CAEN beaches, followed by the equivalent of two tank brigades (U.S. 'regiments') and a brigade group (U.S. 'regimental combat team'). At the same time, airborne forces would capture the town of CAEN, while subsidiary operations would be conducted by commandos and airborne units to neutralize certain coast defences and to seize important river crossings. The object of' the assault forces would be to occupy the general line GRANDCAMP-BAYEUX-CAEN.

Subsequent action would take the form of a strong thrust Southward and South-Westward with a view to destroying enemy forces, acquiring sites for airfields, and gaining depth for a turning movement into the COTENTIN PENINSULA directed on CHERBOURG. When sufficient depth had been gained, a force would advance into the COTENTIN and seize CHERBOURG. At the same time a thrust would be made to deepen the bridgehead South-Eastward to cover the establishment of additional airfields in the area South-East of CAEN.

It was estimated that within fourteen days of the initial assault CHERBOURG should be taken and the bridgehead extended to include the general line TROUVILLE-ALENCON-MONT ST. MICHEL. By that time it should have boon possible to land some 18 divisions and to have in use about 14 airfields from which 28 to 33 fighter type squadrons could operate.

Later operations would necessarily be dictated to a large extent by the enemy's reactions. If he proved sufficiently weak, an immediate advance might be made to seize HAVRE and ROUEN; but more probably it would be necessary to take and bring into use the BRITTANY ports first in order to build up forces sufficient to breach the line of the SEINE upon which the enemy might be expected to make a stand with the bulk of his forces. For this purpose a thrust Southward would be made to seize NANTES and ST. NAZAIRE, followed by BREST and the smaller BRITTANY ports. The bridgehead would be consolidated on the left flank along the River EURE from DREUX to ROUEN and thence along the SEINE to the sea, while CHARTRES, ORLEANS and TOURS would also be occupied. As soon as the Lines of Communications organisation in this area had been developed and sufficient air forces established, operations would be instituted against PARIS and the SEINE ports, with subsidiary operations to clear the BISCAY ports for the reception of additional AMERICAN troops and of food supplies for the civil population of FRANCE.

The appreciation and outline plan for 'OVERLORD' was ready by August 1943, and was considered by the statesmen and Combined Chiefs of Staff at the QUADRANT Conference in that month. Of the statesmen, the Prime Minister thought more diversionary attacks desirable and the addition of at least 25 per cent to the strength of the initial assault; lack of resources, however, appeared to stand in the way of this. The Combined Chiefs of Staff thought that the planned rate of build-up and advance was over-optimistic. The outline plans were nevertheless approved, and COSSAC was authorized to proceed with the planning and preparations on the lines laid down. 'OVERLORD' was recognized as the chief task now facing the UNITED NATIONS, and by 29 November sufficient progress had been made to permit of Directives for the operation being issued to 21 Army Group and 1st U.S. Army Group.

The Supreme tallied Commander was to be responsible for the co-ordination of planning and would control the execution of the operation as a whole. The Commander-in-Chief, 21 Army Group, would be jointly responsible, with the Allied Naval Commander-in-Chief and the Air Commander-in-Chief, Allied Expeditionary Air Force, for the detailed planning of the operation and, when so ordered, for its execution until such time as the Supreme Allied Commander should allocate an area of responsibility to the Commanding General, First Army Group. This allocation of an area to First Army Group would not take place until sufficient U.S. troops had landed on the Continent. The general principle would be that the U.S. forces would operate on the right flank (being the nearest to the ports to which they would sail from AMERICA), while the BRITISH-CANADIAN forces could be on the left. Under the direction of the Commander-in-Chief, 21 Army Group, the initial assault would be carried out by a composite army consisting of approximately two BRITISH-CANADIAN corps and two U.S. corps, under the unified control of the Commanding Generals First (U.S.) Army. The latter would then remain in immediate charge of land operations, until such time as, in the opinion of the Commander-in-Chief, 21 Army Group, the forces landed warranted the introduction of a second army headquarters (BRITISH-CANADIAN) to take over a portion of the front. Except in an emergency, however, and then only for the shortest possible time, a BRITISH or U.S. formation lower than a corps would not be placed under command of another nationality.

The shipment of AMERICAN troops to the UNITED KINGDOM was scheduled to increase rapidly after July 1943. It was expected that 900,000 to 1,000,000 men would have arrived by the end of the year, and 1,250,000 by March 1944, although the bulk of the forces subsequently to be committed

to action in 'OVERLORD' would sail direct to FRANCE after the lodgement area had been secured. All these, however, as well as a proportion of the BRITISH forces, would be new to the realities of war, and it was felt that the initial assault was a task for which battle-experienced men were needed. Although special assault training areas were established, the fact remained that, as General MORGAN pointed out to his Principal Staff Officers on 19 May 1943, 'battle-inoculation' and intensive training could not achieve the same results a. actual fighting experience; "As those of us know who have taken part in battle, it is one thing to manoeuvre freely when secure in the knowledge that the man behind the gun is doing his best to miss us, but it is quite another thing when that same man is doing his utmost to liquidate you". Complete battle-tried formations from NORTH AFRICA were required, men who would inculcate the battle spirit into the novices—"high grade stock from which we must breed with the utmost rapidity". The same need applied equally to the staffs, to ancillary and administrative troops, and to air force ground crews as it did to the forward fighting elements of the army. These proposals were approved by the Combined Chiefs of Staff, and at the QUADRANT Conference in August it was agreed that three BRITISH and four U.S. Divisions should be returned to the UNITED KINGDOM commencing in November, for employment in operation 'OVERLORD'.

The greatest difficulty which exercised COSSAC during the months of planning for 'OVERLORD' lay in the provision of the requisite numbers of landing craft for the operation. It was recognized from the outset that a start had at once to be made in building up the numbers of craft necessary if the operation was to be launched by the target date of 1 May 1944, but very considerable obstacles lay in the way. It was not easy to obtain precise information as to the numbers of boats likely to be available at any one time, largely because it was anticipated that they would have partly to be made up from MEDITERRANEAN resources as and when these should become free.

The situation was summarized by General MORGAN at the end of September as still far from satisfactory. He interpreted the Prime Minister's suggestions at the QUADRANT Conference concerning the strengthening of the assault as advocating an increase not in the size of the force which would carry out the initial landings but in the numbers of troops which would compose the immediate follow-up forces. While it was expected that the available landing craft would be adequate for the primary assault divisions, it was on D plus 1 day, when the immediate follow-up divisions were due to land, that the shortage of craft would create difficulties. "In the popular phrase", wrote General MORGAN, "we already have far too high a proportion of our goods in the shop window. To consider any increase in this proportion without adequate stocking of the back premises would in my opinion be basically unsound". It would also be a great asset if sufficient craft were available to permit of a diversionary landing (as the Prime Minister himself desired) elsewhere than on the CAEN beaches soon after the initial main assault had been launched; as things were at present, it would be 7 to 14 days after this main attack before craft could be freed for such a task. What was needed was sufficient extra craft to provide lift for a fourth assault division which would be "a floating reserve formation in the full sense of that terms."

At this time, in September 1943, despite U.S. efforts to increase the rate of production of landing craft, it was estimated that the deficiency by l May 1944 would amount to 164 L.C.T. and 7 L.C.I.(L). The task of making up this deficiency was to remain in the forefront of COSSAC's work until the establishment of SHAEF in January 1944.

Not only the shortage of craft but also the provision of crews for them was a problem to which an answer had to be found. In June 1943 it was calculated that there was a deficiency of 9,000 men for this purpose. Suggestions that some AMERICAN crews should be employed

to man BRITISH craft were not favoured; there were obstacles in the way of their training, and it was believed that the difficulties which would result from mixed manning would only serve to increase the complications inherent in the operation. An alternative solution lay in the utilization of the personnel of the Royal Marine Division for this purpose, and this policy was ultimately adopted.

Closely connected with the problem of landing craft was that of artificial harbours. As has been seen, it was recognised in the outline plan of operation 'OVERLORD' that it would be necessary to introduce supplies over the beaches for some three months after the initial assault before sufficient ports to maintain all the forces in the lodgement area could be captured and brought into full working order after the inevitable GERMAN demolitions. During that interval the operation would be at the mercy of the weather unless some form of artificial harbour was available, under the shelter of which the reinforcements of troops, weapons and stores could be unloaded. No experience of such harbours was available in May 1943, however, although it was understood that experiments had been made by the AMERICANS in the PACIFIC. Plans had to be developed for some sort of breakwaters which could be moved quickly into position, and for piers or other type of shore connection at which vessels could unload.

The tentative conception of the use of artificial harbours was developed at the LARGS Conference, 29 June-4 July 1943, into what became known as the 'MULBERRY' scheme, and during the following months various suggested forms of floating breakwater devices reached the stage of practical experiment. The types considered included the 'LILO' (which presented to the sea a flexible working face of rubber air-filled chambers, thereby avoiding the heavy stresses normally set up by the waves on a rigid body); the "bubble" breakwater (in which air bubbles were used to create a surface current to counteract the waves); the steel and concrete cruciform 'BOMBARDON'; and the floating ship breakwater. Trials were also made with concrete caissons ('PHOENIX') which could be towed to a given point and there sunk, and the possibility of using sunken blockships was investigated.

Of these, the bubble breakwater, the floating ship breakwater and the 'LILO' were subsequently abandoned, and production was concentrated on the 'BOMBARDON' (after the structural defects revealed when three prototypes broke their backs in a half gale had been remedied) and the 'PHOENIX' concrete caisson. Tho general plan was for the huge 'BOMBARDON' units to form outer anchorages, inside which the 'PHOENIX' caissons would form harbours wherein the unloading of ships could proceed whatever the weather. To provide shelter for the landing craft during the earlier stages of the assault, sunken blockships ('GOOSEBERRIES') were to be employed; it was anticipated that these would be in a position on the first day after the initial landing.

Various types of ship-to-shore connections ('WHALES') were also examined, including the D TN Spud Pierhead and Flexible Pier, the 'Swiss Roll' (a floating canvas roadway) and the U.S. NL Pontoon. The latter was developed as the 'RHINO' ferry, composed of a number of joined pontoons propelled by out-board motors—particularly adapted for the landing of M.T. The 'Hughes Pier', which was non-floating and unsuitable for use where there was a considerable tidal range, was abandoned.

Difficulties abounded: time was short, labour was scarce, constructional facilities were lacking, the weather interfered with practical trials, and the spheres of responsibility were uncertain. It was not until December that proper co-ordination of effort between COSSAC, the Admiralty and the War Office in this work was achieved, and then the arrangement was necessarily complex. The Admiralty was deemed responsible for floating ship breakwaters (if any were to be used), blockships and the towing of

concrete units; the War Office was responsible for the design of concrete units and the erection of piers and pontoons; while a further complication was created by the share of the Ministry of Supply in making available the necessary constructional materials.

The task of driving this team fell to Maj-General Sir HAROLD WERNER, who, as Co-ordinator of Ministry and Service Facilities, was given charge of the 'MULBERRY' project in early August when the whole scheme was in danger of being overwhelmed by the difficulties encountered. Only by persistent effort were these difficulties overcome. To provide the urgently needed skilled labour, men wore released from the army, while the unskilled work was largely done by IRISH labour. Some of the 'PHOENIX' units, which were mainly constructed in the THAMES and SOUTHAMPTON WATER, had to be built in improvised basins excavated alongside the water and flooded by the demolition of the retaining bank when all was ready, because sufficient dry docks and slipways were not available. To achieve the construction of the piers and pontoons, frigate and aircraft carrier production had to be surrendered by the Admiralty.

As a result of these energetic measures, by January 1944 the 'MULBERRY' project was well under way, and it was possible to report that there was every hope that the harbour units would be available by the 'OVERLORD' target date.

Not only had the artificial harbours to be constructed, but tugs had to be provided to tow them into position as soon as possible after the assault forces had landed and set about establishing a bridgehead. The shortage of tugs available for this purpose was brought to notice in September 1943, COSSAC's original estimate of the number required was 130, but of those only 65 could then be provided by BRITAIN and 25 by the UNITED STATES, leaving a deficit of 40. It was considered impossible by the Admiralty to withdraw any more from their normal duties. COSSAC was asked if the number demanded could be reduced, but the reply eras that the 130 were considered essential if concrete caissons were to be used, although fewer might be needed for 'BOMBARDONS'. Caissons were then the only certain form of artificial harbour know, the other types being still in the experimental stage. When the 'BOMBARDON' was subsequently satisfactorily developed, however, the whole 'MULBERRY' scheme underwent revision and expansion, and eventually the number of tugs needed was determined as 158 to cope with the task of moving units totalling approximately one million tons. The provision of tugs consequently remained in the forefront of COSSAC's problems throughout its existence.

For the maintenance of the fuel supply to the armies in the lodgemont area, special arrangements were evolved by which this was to be effected by means of submarine pipe-lines from shore to shore. This was knows as operation 'PLUTO'. The idea did not originate with COSSAC—it was in active preparation when COSSAC was first set up, and experiments were carried out in the BRISTOL CHANNEL during 1943—but it became an integral part of the major 'OVERLORD' plan.

In order to make accurate preparations for 'OVERLORD' it was essential to carry out systematic reconnaissance by air, sea and land for months before the actual operation was due to take place—not only to investigate the enemy's strength and dispositions, but also to collect geographical and geological data and to evaluate the effectiveness of all aspects of the allied preliminary operations.

At the outset, however, COSSAC had no direct control over such reconnaissance operations, but in September 1943 General MORGAN pointed out to the Chiefs of Staff that the present lack of co-ordination led to wastage of effort and possibly even more undesirable consequences.

He asked them to grant him authority to co-ordinate the requirements and priorities of all reconnaissance agencies which would execute missions in the operational area allotted to the Supreme Allied Commander in North-West EUROPE, other than those required specifically in connection with the long-term bombing programme known as operation 'POINTBLANK'. On 13 October this policy was endorsed by the Vice-Chief of Naval Staff: COSSAC, he considered, must control all reconnaissance activity directly connected with the 'OVERLORD' plans, and must also be 'kept in the picture' by the Admiralty and Air Ministry concerning activities not directly affecting the invasion scheme. Lord LOUIS MOUNTBATTEN, Chief of Combined Operations, was likewise in agreement that the responsibility for planning and mounting small scale raids should rest with COSSAC, and, since General MORGAN was unwilling to undertake this without the assistance of the experienced staff at Combined Operations Headquarters, he recommended that the Planning and Intelligence Sections of that Headquarters be put under COSSAC to afford guidance.

The Chiefs of Staff agreed that COSSAC should be granted authority "to control and co-ordinate, within his sphere of operations, all raids and reconnaissance other than purely naval and air reconnaissance not connected with COSSAC operations". Reconnaissance for operations other than those which General MORGAN was preparing would be co-ordinated with his requirements by the Admiralty and Air Ministry in their respective spheres, and if any conflict of interests arose the matter was to be referred to the Chiefs of Staff for decision. A Reconnaissance Committee was accordingly set up, on which wore represented COSSAC, the Allied Naval Commander-in-Chief, the Air Commander-in-Chief, Allied Expeditionary Air Force, the Air Officer Commanding, British Tactical Air Force, the Commanding General, Ninth U.S. Air Force, the Commanders-in-Chief, 21 and U.S. Army Groups, the General Officer Commanding-in-Chief, Scottish Command, and Combined Operations Headquarters.

COSSAC also found it necessary to seek control over Special Operations activities. It was considered vital that such work be correlated with a plan of such importance as 'OVERLORD' in which every branch of warfare was concerned. Converscly, COSSAC needed information which the SOE (Br.) or S.O. (U.S.) authorities alone could supply, to afford guidance in the general work of planning. The operational control to be exerted by COSSAC over SOE/SO activities was, however, to be essentially on broad lines, designed to secure conformity with the broad strategy of COSSAC's plans, and would not be concerned with matters of detailed methods or administration.

At home, COSSAC had to make the necessary arrangements for adequate security concerning 'OVERLORD'. This was discussed in the summer of 1943, but it was not until the end of the year that the proposals began to take shape. It was agreed by COSSAC that a visitors ban should be imposed in the coastal areas, to a depth of 10 miles, from the WASH to LANDS END, and from ARBROATH to DUNBAR, excluding EDINBURGH; but the decision to implement this recommendation rested with the War Cabinet, which only agreed in March 1944 after a personal appeal had been made by the Supreme Commander. It was deemed desirable for some guidance to be given to the Press to prevent reckless speculation as to the time, place and scale of the invasion, and it was suggested that the Prime Minister should have a confidential talk to the BRITISH editors on the subject while similar action would also be taken in the UNITED STATES. The cover plans which would be evolved to disguise the true nature of 'OVERLORD' would also have to be explained by COSSAC to the permanent heads of the BRITISH civil ministries affected, including those of Home Security, Health, Food, Information and War Transport. General MORGAN also suggested that the Prime Minister might give a broadcast address on the subject of invasion in order to assist in the guidance of public opinion.

In framing the operational plans for mounting the 'OVERLORD' scheme, the COSSAC staff had always to bear in mind the possible effects of an intensive 'CROSSBOW' bombardment of the invasion preparations. From July onwards, the enemy was known to be engaged in the feverish development of sites in Northern FRANCE whence attacks by rocket projectiles and pilotless planes might be launched against Southern ENGLAND and LONDON. So serious indeed was the threat that an appreciable diversion of effort from the 'POINTBLANK' strategic bombing Programme was necessary, and at the end of 1943 the possibility of a drastic revision of plans, involving the launching of the 'OVERLORD' operation from an area out of range of 'CROSSBOW', was considered. It was decided, however, that the risks must be run. Any attempt to move the assault forces further from their objective than the SOUTH coast would involve difficulties of such an extent as to more than counterbalance the 'CROSSBOW' threat, which, it was considered, would not preclude, though it might prejudice, the launching of the attack from the originally planned areas. The best that could be done was to ensure the maximum dispersion of craft within the assembly areas, and to use the allied air strength against the enemy's 'CROSSBOW' sites to reduce their efficiency to a minimum.

In September 1943, following the QUADRANT Conference, the difficulties surrounding the preparations for 'OVERLORD' led the Chiefs of Staff, at the instigation of the Prime Minister, to issue instructions to COSSAC to consider the possibility of undertaking operations against NORWAY in the early summer of 1944 "in case circumstances render the execution of 'OVERLORD' impossible". Plans were required for an occupation of NORWAY by the forces assembled for 'OVERLORD', and for an occupation of DENMARK, using bases in Southern NORWAY. COSSAC was asked to report on the effect this would have upon the 'OVERLORD' preparations, and what extra staff would be required. The reply was that the nature of 'JUPITER', as this proposed operation was titled, was so radically different from 'OVERLOAD' in respect of the distances to be covered and the problems of air support, that the preparations for the one could not well be applied to the other, and for COSSAC to undertake the planning of the 'JUPITER' enterprise would jeopardise the chances of success in 'OVERLORD'. Moreover, extra staff would not only be required for COSSAC, but in the lower echelons shore would have to be virtual duplication of the existing organization. On the UNITED STATES side it would be impossible to make this extra staff available in the time allowed, and 'JUPITER', if undertaken, would perforce have to be a predominantly BRITISH operation. As COSSAC mastered its chief difficulties, however, the 'JUPITER.' scheme fell into the background, and by the New Year the necessity for an alternative to 'OVERLORD' was no longer felt.

The QUADRANT Conference agreed in August 1943 that a diversion from the attack in North-West FRANCE should be created by simultaneous activity in the SOUTH, and in September a Directive was issued to General EISENHOWER, instructing him to prepare outline plans for such an operation by the MEDITERRANEAN forces, to he known as operation 'ANVIL'. General MORGAN sought authority from the Chiefs of Staff to deal direct with General EISENHOWER concerning the co-ordination of operations, and October missions were exchanged between COSSAC and Allied Force Headquarters, ALGIERS.

When General EISENHOWER's plan was submitted in November, however, a difficulty arose. Operation 'ANVIL' was originally conceived as a deceptive threat, designed to pin GERMAN forces in Southern FRANCE as far as possible from the area where the 'OVERLORD' attack was to be 1aunched in the North-West, but the proposals put forward by General EISENHOWER, under the terms of his directive, were for an actual assault by two or three divisions against TOULON and MARSEILLES. When a bridgehead had been established, the invading force, including FIGHTING FRENCH troops, was to be built up to a strength of ten divisions and to strike Northward towards VICHY, in co-operation with FRENCH guerrillas.

It was agreed by COSSAC that the abandonment of this plan of operations, once it had been put forward, would place the Allies in a difficult position via-a-via the FRENCH; moreover, the RUSSIANS had been told at TEHRAN that a blow would be struck in the SOUTH, simultaneously with 'OVERLORD'. At the same time, however, it was felt by both General MORGAN and General MONTGOMERY that it would involve "an unsound diversion of forces from the main 'OVERLORD' assault area to a subsidiary assault area where they were unlikely to pay the same dividend". As a mere threat, operation 'ANVIL' had been calculated to require assault craft sufficient to lift one assault division only, and the extra resources now demanded would have a serious offset in view of the bare margin allowed for 'OVERLORD'. 'ANVIL', it was considered, would not pin down more than two or three GERMAN divisions as Southern FRANCE was too far from the centre of GERMANY's chief interests to be worth defending in strength at a time when a much more serious menace was developing to the NORTH. It was suggested that the FRENCH troops could be compensated for their disappointment in the SOUTH by being sent in through the BISCAY ports at the earliest possible moment.

It was clear that either 'OVERLORD' or 'ANVIL' must undergo modification in view of the limited resources available, but a solution to the problem was not reached until after General EISENHOWER had assumed duty as Supreme Commander of the Allied Expeditionary Force in 1944.

In order to disguise the Allied intentions in operation 'OVERLORD', plans were drawn up by COSSAC for a deception scheme (suggested by Colonel J. H. COVET, Controlling Officer to the BRITISH Chiefs of Staff Committee), intended to cover the period from the conclusion of operation 'COCKADE' in November 1943 until the time came for the preparatory phase of 'OVERLORD' itself. This scheme, originally known as Plan 'JAEL', was cosigned to induce the enemy to make faulty strategic dispositions which would ultimately assist 'OVERLORD' or even render 'RANKIN' Case A feasible. Under this plan, Allied threats against Northern ITALY, the BALKANS, NORWAY and DENMARK were to be brought to the enemy's notice, while the real preparations going on in Southern ENGLAND were to be concealed as far as possible. Every indication was to be given that the Allied main effort in 1944 would be concentrated in the MEDITERRANEAN. It was hoped to persuade the enemy that the Allies considered the Northern FRANCE defences too strong to be tackled before the garrisons had been weakened to meet a MEDITERRANEAN offensive, and that much greater forces than was actually the case were massing in NORTH AFRICA for that purpose. Diplomatic activity in TURKEY and SPAIN was to be used to foster the impression that the aid of these countries was being sought in support of the supposed moves. Similarly the impression was to be given that in an attack on DENMARK and NORWAY, SWEDISH co-operation was contemplated.

This plan was evolved in October 1943, but subsequent events necessitated its modification, and in December, following the TEHRAN Conference, it was superceded by Plan 'BODYGUARD'. The basic aims were as before—to induce the enemy to make faulty strategic dispositions—but greater prominence was now given to the Eastern MEDITERRANEAN to distract attention from the 'ANVIL' scheme. It was to be made to appear that the Allies' main effort in the spring of 1944 would now be directed against the BALKANS, with RUSSIAN and TURKISH co-operation. In NORTH AFRICA, ANGLO-AMERICAN troops were to be alleged to be withdrawing, leaving only FRENCH forces there. In the UNITED STATES, the difficulties experienced in the provision of landing craft were to be portrayed as ruling out the possibility of a cross-Channel offensive until late summer at the earliest, by which time a further RUSSIAN offensive might have drawn off more of the defenders of the 'Western Wall'. The idea was also to be fostered in GERMANY that the UNITED NATIONS were hoping to win the war in EUROPE by the long range bomber offensive without resorting to land operations on the Continent until GERMANY was on the point of collapse.

This deception scheme was intended to be operative until about the end of February 1944. By that time, the preparations for operations based upon the UNITED KINGDOM would have reached a stage when they could no longer be disguised, and then it would be necessary to mislead the enemy as to the exact place and time of the intended blow. The basis of this Cover Plan for 'OVERLORD' was worked out by COSSAC in the autumn of 1943, and the provisional outline was completed in November.

The main feature of the Cover Plan was that, when the time came that the GERMANS must realise that something was brewing in ENGLAND, the invasion preparations in the South-East should be discreetly displayed, while those to the South-West should continue to be concealed. By this means it was hoped that the enemy would be led to anticipate that the main attack would be delivered against the PAS DE CALAIS region instead of against the CAEN area—and, even after the assault on the latter had commenced, to believe that a subsidiary operation was to be mounted against the PAS DE CALAIS. Furthermore, uncertainty as to the time of the intended invasion would be fostered by the holding of large-scale exercises and by wireless activity designed to simulate preparations for an attack,

To these ends, during the preliminary phase of operation OVERLORD, wireless deception would be directed to indicate the supposed presence of additional amphibious forces in the South-East while concealing the true nature of intended moves, while intermittent periods of wireless silence would lull the enemy into a false sense of security when silence was eventually imposed for the real D-Day. In addition, tented camps, deceptive lighting and exaggerated constructional activities would be displayed in the THAMES ESTUARY area, while installations to the WEST would be carefully camouflaged.

During the preparatory phase of 'OVERLORD', dummy aircraft would be displayed in the South-East, while dummy landing craft would be assembled from YARMOUTH to HASTINGS in order to make the NORE follow-up force appear to be an assault one. Large scale movements of troops not immediately required overseas would be carried out in the EAST and South-East, while civil activities in the same areas would include the reinforcement of fire services and other civil defence organizations and the encouragement of voluntary evacuation.

After the CAEN assault had begun, any spare craft would be concentrated with the dummies in DOVER and NORE Commands, dummy aircraft would continue to be displayed and deceptive flying carried out. A force of about six divisions, consisting of troops not required during the first fourteen days of the assault and others not forming part of the expeditionary force, would be stationed in the South-East, while deceptive wireless activity would continue.

Thus, from the early summer of 1943 until the end of its existence as such in January 1944, COSSAC hewed out the plans for the great return of the forces of liberation to the Continent of EUROPE. By the New Year, the Chiefs of Staff were satisfied that the work had reached a stage when the Supreme Allied Commander might be appointed to take charge and to put the plans evolved into final preparation for the day of action. The appointment had been long delayed: it had been repeatedly demanded by COSSAC for months past as an essential step to be taken before the task of planning could be carried to completion. After the SEXTANT Conference had ruled that nothing was to be undertaken in any part of the world which hazarded the success of 'OVERLORD' and 'ANVIL'—"the supreme operations for 1944''—there was no reason for further postponement of the appointment. The Allied Naval Commander, Expeditionary Force, the Commander-in-Chief, 21 Army Group, and the Air Commander-in-Chief, Allied Expeditionary Air Force, were already named; and on 14 January 1944, General MORGAN announced that General DWIGHT D. EISENHOWER would be arriving by 17 January to take up his duties as Supreme Allied Commander. COSSAC was now transformed into the Supreme Headquarters, Allied Expeditionary Force.

In the early autumn of 1943, Lieutenant-General F. H. MORGAN directed that a Headquarters sign be designed in anticipation of the appointment of the Supreme Commander.

A committee, composed of Mr. Charles PEASE, Captain H. O'D. MORRISON, U.S.A.A.C., and Subaltern E. SEROCOLD, A.T.S., commissioned the College of Heralds to produce a sign symbolic of the task which the Supreme Commander would undertake and of the inter-allied character of the forces under his command. While the resultant design was not considered entirely appropriate, the central feature "The Flaming Sword of Freedom" was accepted as the basic concept for further development.

Major-General R. W. BARKER, U.S.A., assumed direction of the detail of further designing. With the assistance of Colonel M. LACK, U.S.A., and the artistic skill of Corporal D. Q. GOODALL, A.T.S., the final sketch was produced. (It is to be noted that the "Flaming Sword of Freedom" in the accepted sketch was modelled on that of the Second U.S. Division Memorial (1917-18) in WASHINGTON).

Upon assuming command, General D. D. EISENHOWER approved the design. It was brought into general use in March 1944 as his Headquarters shoulder patch and vehicle mark.

The following are the chief sources used in the compilation of this outline History of COSSAC:

Note on the Custody and Indexing of Records.

A Central repository for COSSAC/SHAEF Records is maintained by the AD Branch of the Headquarters. The usual practice is, however, for Divisions to retain files in their own hands until the subject matter becomes 'dead'. This applies particularly to G-2 (Intelligence) and G-3 (Operations) Divisions.

The Office of Secretary, General Staff, Records Section, holds records of the highest levels, but these also will eventually be deposited with AG Records. The SGS Records Section is at present (May 1944) undergoing reorganization, involving the transformation of its filing system from an English to a Dewey-Decimal system. 'Dead' files, such as those dealing with Operation 'COCKADE' and its component schemes, will be transferred AG Records without prior re-numbering; 'live' files, such as those dealing with operations 'RANKIN' and 'OVERLORD', will be reallotted Dewey-Decimal reference numbers. A list of SGS file references is given in SGS.313.3, dated 28 May 1944.

A proportion of the COSSAC period records of Divisions has now been deposited with AG Records, where the files have been broken down and re-numbered according to details of subject matter. Cross references from old to new file numbers are listed in file AG.313.3-2.

Micro-filming is in progress, and all important documents will be dealt with by this process.

Verbal Information

Verbal information from the following has been utilized in compiling this History of COSSAC: