Annexes

Annex A. Photography, 1914-18.

The RFC went to war primarily to gather intelligence. It followed that, since a photograph provided an accurate and permanent record of what had been seen during a sortie, cameras should have been the most useful items of role equipment in the inventory. This appears not to have been quite as blindingly obvious at the time as it is today, however, and in August 1914 the RFC had just five Ross ‘Panros’ folding bellows cameras available. By early 1915 three of these had been lost or damaged but the pictures that they had taken had been valuable enough to persuade the most sceptical of generals that aerial photography was essential. In January Lt J T C Moore-Brabazon was put in charge of a small experimental photographic section which had been set up within 1st Wg to review the situation and determine the way ahead. Within a few weeks, with the cooperation of the Thornton-Pickard Manufacturing Company, a new camera, the Type A, had been designed, built, tested and put into production.

The Type A was first used operationally in March 1915 and the lessons learned from this experience led to a series of progressively improved designs. The salient characteristics of the most common types of wartime camera are listed at Figure A1. It will be seen that each model generally tended to be heavier and more complex than its predecessor but, at the same time, it also became simpler to operate. Reliability was improved by the introduction of metal construction. Utility was enhanced by the provision of a range of lenses, so that by 1918 it was possible to take photographs from 20,000 feet with sufficient discrimination to be able to distinguish barbed wire entanglements. While the major customers, the planning and intelligence staffs, the map-makers, the artillerymen and, to an extent, balloon observers, all required vertical photographs, there was still a frequent need for oblique pictures to amplify specific details or to record fleeting events. These were usually taken by the observer, using a hand-held camera.

Fig A1. Salient characteristics of the most common types of camera used by the RFC/RAF.
* Including magazine.

Thornton-Pickard Type C. (B L Gray)

So far as organisation was concerned, on Moore-Brabazon’s recommendation, each wing was established to have its own photographic section in the spring of 1915.1 This arrangement sufficed for a while but within a year it was unable to cope with the volume of work. The problem was not so much the processing of plates, which were being exposed in steadily increasing numbers, but in supplying prints for which the demand was insatiable and which were almost always required urgently. To speed up the service it was decided both to expand and to decentralise the system. From April 1916, therefore, every corps reconnaissance squadron was established to have its own photographic section staffed by an NCO and three men.

By the summer of 1918 the establishment of photographic personnel on a corps reconnaissance squadron equipped with twenty-four RE8s or FK8s had grown to sixteen: a sergeant; a corporal and fourteen air mechanics. There was no corresponding provision on bomber or fighter reconnaissance units, which suggests that, although DH 9s, Bristol Fighters and other types could certainly be fitted with cameras, relatively little use was being made of them in other roles.

Some idea of the scale of the work carried out is conveyed by some statistics compiled shortly after the war which indicated that the numbers of prints produced annually in France were:²

Year	No of Prints
1915	(Est) 80,000
1916	552,453
1917	3,925,169
1918	5,946,096
Total	10,503,718

Compared to the work done by the Army, the Navy devoted relatively little effort to the development of cameras. This was because vertical photography had little practical application in the maritime field and for overland work the RNAS would have had access to the excellent range of cameras which was being developed by the RFC. Mounting cameras externally on seaplanes had proved to be problematic in any case due to salt deposits from sea spray. The Navy did have a need for hand-held cameras for very long range oblique work and suitable devices were developed to meet this requirement, but they were produced only in very small batches. As a result there was little standardisation but most naval cameras were of simple box design with a double dark slide for manual plate changing. A variety of lenses gradually became available, these giving focal lengths of between 4” and 15”. The proliferation of shore-based coastal reconnaissance and anti-submarine flights in 1918 created a demand for more cameras of this general type. The RAF eventually introduced standard models for this purpose, notably the P14 and P18.3 These were produced in substantial quantities, seeing extensive service with land-based squadrons in France and elsewhere as well as at sea.

Above – A Williamson Type L (designed by F C Laws – hence the ‘L’) camera fitted to an Avro 504J and protected by an airstream deflector. Operated remotely via a Bowden cable, the plates were changed automatically by the windmill-powered flexible drive. Below – another Type L, this time mounted on the lower wing of an AWFK 3, where it could be operated manually, while the man in the rear cockpit wields one of the handy P-series. (Both B L Gray)

A Bristol Fighter with its fabric stripped to reveal the observer groping towards a floor-mounted Type L camera in order to change the magazine. (J M Bruce/G S Leslie collection)

Annex B. Communications, 1914-18.

While crews had been using W/T to transmit to the batteries whose fire they were directing ever since 1914, this one-way system did not cater for long-range work, ie when the target was too far away for the ground signals of the ‘Gunners’ to be read from the air. To solve this problem it was necessary to provide two-way radio communications. Associated trials with airborne W/T receivers were well under way by 1916 and in the spring of 1918 a start was made on deploying an operational system. To begin with small numbers of suitably equipped Bristol Fighters were distributed among the existing corps reconnaissance squadrons.⁴

This piecemeal approach proved to be difficult to manage, a problem that was overcome by an organisational change. In July the modified aeroplanes available to I Bde were pooled to form an autonomous unit manned by crews who were to be dedicated to long-range artillery work. Once this unit, L Flight, had worked out the most appropriate operating procedures, it became clear that this was a much more efficient solution and it was decided that each RAF brigade should have its own Bristol Fighter-equipped long-range artillery flight. The next one, III Bde’s N Flight, was formed in September and three more had come into being before the Armistice. The equipment used comprised a Type 52A transmitter, powered by a wind-driven alternator, and a Mk III receiver. Although the long-range system had clearly served to demonstrate the advantages of two-way radio (W/T) communications, it had been introduced too late to have a major impact on operations. ⁵

While the provision of airborne Morse receivers had provided the RFC with a worthwhile increase in operational flexibility, it did not represent much of a technical advance, since the use of two-way W/T was already well-established elsewhere, notably within the RNAS. A far more significant innovation was the introduction of wireless telephony, ie voice communication, later to become better known as radio telephony or R/T. Experimental work on speech transmission was under way in the UK by May 1915 (possibly earlier) and the RFC first employed R/T in France during July 1917 when Nos 11 and 18 Sqns carried out a series of field trials. This activity was carried out behind the lines, using apparatus made up in the workshops of the Signals Experimental Establishment at Woolwich. The success of this programme led to large-scale orders being placed, the aim being to have all two-seater squadrons fully equipped by 1919.

Training of Bristol Fighter and DH 4/9 crews began at Biggin Hill in January 1918, the Wireless Experimental Establishment accepting intakes of up to thirty students per week, drawn both from squadrons working-up in the UK and from those already in France. On 2 April, by which time sufficient aircrew had been trained to man ten squadrons, the task was taken over by the newly established Wireless Telephony School which moved to Chattis Hill a fortnight later. At much the same time No 1(T) Wireless School began turning out large numbers of appropriately trained technicians from Farnborough. The deployment of hardware started in March when the Bristol Fighters of Nos 11 and 22 Sqns began to be fitted with R/T equipment but the entire programme was thrown into disarray by Ludendorff s spring offensive. Once stability had been restored the process was resumed with No 22 Sqn being fully equipped by the end of April, Nos 11 and 88 Sqns during May and No 18 Sqn, along with some of the corps squadrons working with Second Army, in August.

Above – An observer demonstrating the way in which the Strange mount might permit him to use the gun to provide a modicum of defensive fire to the rear, albeit at some inconvenience (not to say risk!) to his pilot. Below – A sergeant demonstrating the use of a gun mounted on the No 4, Mk IV swivelling pillar mounting of an FE2b. Both guns are Lewis Mk IIs and both are fitted with bags to collect spent cartridge cases, a very necessary precaution to avoid their striking the pilot and, in the case of the ‘Fee’, the propeller. (Both J M Bruce/G S Leslie collection)

Steady progress was maintained thereafter but the programme still had a long way to go when the war ended, although all nine fighter squadrons based in the London Air Defence Area had been equipped before the end of the summer. Nevertheless, it is indicative of the RAF’s intention to provide telephony universally that the establishments of all squadrons were amended in July 1918 to reflect the provision of the necessary additional tradesmen, including a Technical Officer (Wireless), or two in the case of a corps reconnaissance squadron.6

Many applications for R/T were found and tried with varying results and in many cases it was the observer who handled the radio. The practical experience gained included the transmission of reconnaissance reports direct to Army Headquarters (by No 18 Sqn), the direction of tanks by aircraft (by No 8 Sqn) and two-way communication between forward troops and aircraft during contact patrols. R/T had also been used to pass landing information and/or diversion instructions from operating bases to returning night bombers of the Independent Force, and fighter reconnaissance squadrons were beginning to use air-to-air communications to control formations. The latter was found to be highly effective in training new crews behind the lines and before the war ended the use of R/T was being actively investigated for the tactical control of combat formations (by No 22 Sqn) and for the co-ordination of bombers and escorts (by Nos 88 and 103 Sqns).

There were many teething troubles due to equipment failures and limited power output, the practical range for an airborne transmitter being not much more than 10 miles. Problems were also experienced with discomfort from ill-fitting headphones and the concentration necessary to operate the equipment tended to detract from other vital tasks, not least the need to maintain a sharp look out for enemy aircraft. The greatest practical limitation, however, was the need to use a trailing aerial up to 150 feet long. This was a potential hazard when flying in formation and an unacceptable hindrance in air combat. If a fight developed the aerial had to be reeled in or jettisoned; in either case it put an end to the use of R/T when it might have been most valuable. Work was in hand to provide fixed or ‘cartridge’ aerials to solve this problem when the war ended. Despite the many difficulties being experienced, solutions to all of the problems which had been encountered were being devised and R/T showed great promise. Had the war continued into 1919 there can be little doubt that it would have provided the RAF with a significant tactical advantage.

Annex C. Gun Armament, 1914-18.

Although some pre-war experimental work had been done on the carriage of guns, when the RFC flew to France in 1914 none of its aeroplanes was armed. Indeed the most appropriate weapon(s) had yet to be identified and there were no practical gun mountings. To begin with, therefore, the more aggressive crews began to arm themselves with Lee-Enfield rifles, supplemented by the occasional shotgun or even a Service revolver, none of which was very effective.

What was needed was a machine gun, of which the British Army had two standard types. Both were of rifle (·303 inch) calibre, the recoil-operated, belt-fed, water-cooled Vickers, and the gas-operated, magazine-fed, air-cooled Lewis. Because it was a little lighter, generally ‘handier’ and lacked the complication of cartridge belts, the latter was considered to be best suited for use in the air and more than 1,000 had been ordered for the RFC by mid-1915. Although only a quarter of these had been delivered by that time, these guns sufficed to arm the relatively small force which was in the field.

Yoking two Lewis guns together, as on this RE8, C2731 of No 5 Sqn, created the so-called ‘Huntley and Palmer’. Note the pilot’s Vickers gun mounted alongside his cockpit, the mounting frame for a camera and the stencilled notice under the sill of the observer’s cockpit warning that 150 lb of ballast was necessary if the aeroplane was to be flown solo. (J M Bruce/G S Leslie collection)

While it was obviously important to have an effective weapon, if it was to be of any real use, it was also necessary to devise an efficient means of bringing it to bear. Many different types of mounting were designed, some of them specific to particular types of aircraft, and most of these were serviceable enough, at least in pusher aeroplanes. As discussed in the main narrative, however, none of them were of much real use to an observer occupying the front cockpit of a tractor biplane. The biggest constraint was the need to avoid damaging the aircraft’s propeller, which effectively created a large cone in the forward hemisphere into which it was impossible (or extremely inadvisable) to fire a gun (see photographs on page 21).

The same constraint applied to pilots, of course, although several methods of mounting a fixed forward-firing Lewis were devised. The best of these, designed by Sgt R G Foster in 1916, permitted the gun to be fitted, parallel to the aircraft centre-line, on top of the upper wing where it was high enough to shoot above the propeller disc. Foster-type mounts were used on single-seaters for the remainder of the war. While the Foster mount was a satisfactory means of outflanking the problem created by the propeller, what was really needed was a solution – a means of firing through the disc swept by the airscrew.

Methods of doing this had been available since 1914 but it was not until 1916 that the British authorities actually adopted a system. The first to find official approval was the Vickers-Challenger synchronisation gear, which was followed by Ross gears and Kauper gears. Meanwhile, the potential of the Vickers gun had been re-examined. It had been found that, if it could be carried on a fixed, rather than a flexible, mounting, the belt feed need not be an encumbrance. It had also been shown that, in an airborne environment, it was possible to dispense with the water jacket. By 1917, therefore, the Vickers gun, in combination with the definitive synchronising system, the CC (Constantinesco-Colley) gear, became the standard forward-firing armament in practically all single-seat fighters. It was also adopted as the pilot’s (fixed) weapon in two-seaters.⁷

In the Sopwith 1½ Strutter of 1916, and in all subsequent two-seaters, the observer occupied the rear cockpit, giving him a relatively unrestricted field of fire. Since it needed to be flexibly mounted, the weapon of choice for back-seaters remained the Lewis gun, which was progressively refined with the aim of reducing its weight while increasing both its ammunition capacity and its rate of fire. Fitted with a 47-round magazine, the Mk 1 Lewis of 1915 weighed a little over 25lb and could fire 550 rounds per minute. Despite having a 97-round drum, the Mk III of 1918 weighed only 17lb and fired 700 rounds per minute.

Early methods of mounting the observer’s gun varied in detail but most designs involved the weapon’s being free to move in elevation, usually with a degree of counter-balance, while it was trained in azimuth by swinging it around a ‘rail’ fitted to the rim of the cockpit. By the end of 1916 the ‘Scarff ring’ (named for its designer, WO Frederick William Scarff, RNAS) had been adopted as standard, the most common pattern being officially identified as the No 3, Mk II Barbette.

By 1918 it was quite common to mount Lewis guns in pairs, this combination being known colloquially as the ‘Huntley and Palmer’. While it was quite a handful to manage in the better than 70 mph slipstream, since they could be fired together or independently, the two-gun arrangement gave the observer the very useful options of being able to double either the firing time or the weight of fire. That having been said, however, it should perhaps be pointed out that, for two reasons, it was unwise to fire the Lewis in spectacular Hollywood-style sustained bursts. First, because the barrel was susceptible to damage (‘blueing’) from overheating and, secondly, because of the limited capacity of its magazine. If the gun were fired continuously, the early 47-round magazine would be emptied in five seconds and, because of the increased rate of fire of the Mk III, even the later 97-round magazine would last for only eight. The idea was to fire the gun in well-aimed three-to-four round bursts rather than to use it as a hose.

With only minor modifications, the fixed Vickers gun and the Lewis gun/Scarff ring combination remained the standard armament of British aircraft until the adoption of the Browning machine gun and power-operated turrets in the late 1930s.

For more detail on the guns employed (by both sides) during WW I, the reader is referred to the late Harry Woodman’s excellent Early Aircraft Armament (1989) from which much of the above has been condensed.

Annex D. Navigation, 1914-18.

Although back-seaters and pilots had shared an interest in navigation from the outset, by 1918 it was becoming increasingly common for observers to assume responsibility for this task. The extent to which this was occurring varied from role to role but it was most marked where navigation was most demanding, in the squadrons of the long-range heavy night bomber force. Solutions to the problems of air navigation had been incremental, improvements in tools and techniques being introduced piecemeal from 1915 onwards, so that by 1918 substantial progress had been made. Despite these advances, practically all of which had been made by RNAS personnel, for most practical purposes air pilotage, ie navigation primarily by reference to a map, remained the basic means of navigation over land.

It was one thing to keep track of one’s position while flying in daylight over a relatively confined area of what soon became very familiar terrain, which is how the RFC’s tactical aeroplanes operated, but it was quite another to fly deep penetration raids into enemy airspace at night. Recognising the need to raise navigational standards, Lt Cdr Lord Tiverton,8 arguably the RN’s’s leading theoretician in the field of applied air power, nevertheless recommended that map-reading should remain the cornerstone of bomber navigation, rejecting other methods as being insufficiently precise to justify their increased complexity. This did not mean that crews were restricted to flying from one readily identifiable feature to another. To assist them with route navigation they were encouraged to apply simple air pilotage techniques. Drift, for instance, could be assessed by eye by reference to lines drawn on the map in advance and displaced at (say) 10° intervals from the desired track. Similarly, groundspeed could be measured by using a stopwatch to record the time taken to fly between two identifiable points, ideally cross-track line features. In fact, to assist the Anglo-French night bomber force based near Nancy, the French positioned lights at predetermined intervals so that an outbound crew could obtain an accurate groundspeed check by overflying them before crossing the lines.

Tiverton also advised that bomber crews should be required to spend many hours poring over appropriate maps so that they would become thoroughly familiar with the entire region over which they might be required to operate. They would also need to have a particularly detailed knowledge of their specific objectives and, to provide the material to meet this requirement, he recommended the use of dedicated photographic reconnaissance units. Many years later Tiverton’s ideas would be put into practice and ‘Target Study’ was to become a regular feature of the training routine for bomber crews of the Cold War era, but it never became a well established procedure in 1917-18.

While map-reading also sufficed for inshore coastal operations, it provided no solution to the navigational problems of crews operating well out to sea on antisubmarine patrols. This presented particular difficulties for the crews of the naval airships which often flew over the North Sea or the Western Approaches for many hours, even days, out of sight of land and in winds of unknown velocity. The speed of these winds could easily exceed the speed capability of the ‘ship and under these circumstances it was impossible to stay on station and extremely difficult to maintain a sensible plot of the craft’s position. Thus, while an airship might sight something of interest, this information was of limited value if the crew could not establish their own position with reasonable accuracy.

A notable RNAS officer, Flt Cdr A W Bigsworth, gave his name to this combined chartboard, straightedge and protractor. The ‘Bigsworth Board’ was produced in substantial quantities and it remained in service well into WW II when it was still providing a conveniently portable and self-contained navigation station in aeroplanes (like the remaining open-cockpit biplanes, and even Blenheim Mk Is) in which adequate facilities for the observer were still lacking.

Navigating by reference to the Sun, Moon and stars was a familiar procedure to sailors and (when cloud cover permitted) the crews of RNAS airships and flying boats made some attempt to employ these techniques.9 The use of astronomical sightings was only moderately successful, however, as established nautical practice did not transfer easily to the airborne environment.¹⁰ The problems that were encountered were compounded in the case of airships by the vast bulk of the vessels themselves which would have obscured the upward view from the underslung crew car like a permanent overcast. A real enthusiast for celestial navigation, lucky enough to be assigned to airships of the North Sea or Coastal classes, could overcome this limitation, however, so long as he was prepared to abandon the relative security of the crew car and clamber up a fabric tube through the ship to a gun position perched precariously on top of the envelope.

While operating within sight of the coast it was, of course, far more practical to take compass bearings of coastal features to establish an aircraft’s position. Furthermore, when escorting a convoy, such missions representing a substantial proportion of airship activity, navigation could generally be left to the ships which were being accompanied.

Calculators and tools.

During 1917, a gradual increase in the intensity of relatively long-range overland operations, particularly bombing sorties, began to focus attention on the potential of dead reckoning.¹¹ Here the RFC/RAF was able to take advantage of the fact that RNAS activities over the sea had already led to the development of a range of useful tools. Most of these had been devised by naval officers, extrapolating their experience of nautical practice, who gave their names to these devices. Thus in the summer of 1916, for instance, Lt Cdr G R C Campbell and Lt Cdr G B Harrison were able to adapt the principle of the naval ‘Battenberg’ to produce the Campbell-Harrison Course and Distance Calculator. From July 1917 this instrument was embellished by Cdr Rollo Appleyard RNVR who added his eponymous Time and Distance dials. A combination of the two produced a handy calculator which could solve all of the common problems involved in aerial navigation and by September 1918 more than 16,000 examples were on order. Other aids to navigation included Flt Cdr A W Bigsworth’s plotting board, Cdr H P Douglas’ protractor (still standard issue to all RAF navigators until, at least, the 1990s) and the Aero Bearing Plate, the latter being additionally available in a luminous model for night use.

The bearing plate was an adaptation of another long-established nautical device, the Pelorus. The version developed for use in the air had a fixed, depressed sighting angle which made it particularly useful. Having first aligned the plate with the passage of objects on the ground, thus establishing drift, one could time the interval from an object’s entering the field of view until it was directly beneath the aircraft. This, combined with a knowledge of the aircraft’s height, the depression angle of the sight and some elementary trigonometry, provided groundspeed, which, since heading, airspeed and drift were already known, permitted the wind velocity to be determined. Once a means had been found to assess the local wind in flight it became theoretically possible to deal with the triangle of velocities, the key to aerial navigation.

The availability of these tools and the increasing awareness of the need for all aircrew to be able to handle them constructively led to navigation displacing artillery cooperation within the curriculum of the SoMAs in January 1918. It also began to feature increasingly in the syllabuses of a number of role-related courses, particularly those concerned with day and night bombing.

Compasses.

The RFC was not alone in increasing the amount of attention being paid to air navigation during training and as late as March 1918 the RNAS was considering setting up a course to train specialist instructors in map-reading, compasses and aerial navigation. Once qualified they were to preach the gospel at naval flying training units where classroom exercises were to include the application of the wind velocity to calculate the headings and times to fly to negotiate a triangular course. Since the RAF assumed responsibility for all aircrew training only a few weeks later, it is doubtful whether this RNAS-sponsored scheme was ever implemented. Nevertheless, it is informative to consider the content of one of its proposed practical exercises. In the air, from a known start point, a pilot would have had to fly a precalculated heading for ten minutes ‘under the hood’ after which he would have had five minutes to locate himself by reference to his map. The interesting thing is that the average permitted error in the heading steered was to have been as much as 20°, it being envisaged that the instructor would monitor this by logging the compass reading every fifteen seconds.

This level of tolerance may seem remarkable today, but, as had been demonstrated by a recent series of trials, it was a reasonable reflection of the performance to be expected using a contemporary air compass. In November 1917 the RFC Experimental Station at Orfordness had been engaged in an investigation of the feasibility of instrument flying. As part of this work, a number of pilots had each been required to demonstrate their ability to fly due north, south, east and west. During each exercise the heading was to be maintained for five minutes without using any external reference. A test observer recorded the compass reading at fifteen second intervals. The trial results were eventually presented as graphs indicating error to the left or right of the required heading. At Figure D1 the outcome of one of these test flights has been replotted to show the headings actually steered by the aeroplane on each of its four runs.

It is significant that the logged readings were to an accuracy of only 5°. This will have been partly because it was difficult to read the instrument with any greater precision, but this problem would have been considerably aggravated by the instability caused by the meandering course being steered as the pilot ‘chased the needle’. Of the eighty readings which are reflected in Figure D1 there were only eight instances of this particular pilot holding the same heading for 30 seconds. Six (unidentified) pilots took part in the trial. The example illustrated here was the worst case but none of the others was particularly impressive. The best that anyone achieved was to wander 10° either side of a mean course, this mean course often being displaced by as much as 30° from that which was intended. The results of this trial were not so much a reflection on the flying skills of the pilots involved, however, so much as an indication of the limitations imposed by the state of the art of instrumentation.

Fig D1. The headings actually flown by an unidentified pilot on 4 and 5 November 1917. He was attempting to maintain courses of due north, east, south and west using an RAF Mk II Air Compass but with no external reference. Weather conditions were noted as being good with smooth air. The aircraft was a BE2e, 4560.

The aircraft used (a BE2e) would have covered about five miles in five minutes, yet it is plain from the diagram that in three of the four cases shown the (presumably relatively experienced) trials pilot was already at least a mile away from where he would have expected to have been. The student pilots on the projected RNAS course would have been even less capable of steering a heading accurately. Since they would have been required to remain ‘heads down’ for twice as long, it was only to be expected that they would be displaced by several miles by the time that they were permitted to look out. As they were also quite likely to have been further disorientated by finishing up pointing in the ‘wrong’ direction, it is perhaps understandable why the Navy had thought it necessary to allow them as long as five minutes to sort themselves out.

If nothing else, the Orfordness experiment had served to highlight the inadequacy of the most common compass then in use (the RAF Mk II).12 This will hardly have been a surprise, as all early aircraft compasses were acknowledged to be notoriously inaccurate. They were all far too easily disturbed by a variety of factors, including the magnetic field created by the whirling metallic mass of a rotary engine, the electromagnetic effects generated by magnetos, mechanical vibration, the accelerations caused by aircraft manoeuvres and the magnetic implications of the carriage and release of bombs. Vibration alone, for example, was known to be capable of inducing a permanent deflection in excess of 40° in the Type 259 Compass.

Left, a Type 259 air compass and, right, a Type 5/17 (Both images courtesy of COMPASSIPEDIA, The Online Compass Museum)

Apart from these problems, early aircraft compasses were deeply mistrusted, indeed often ignored, as they were all prone to the mysterious, so-called, ‘northerly turning error’. For no apparent reason, beyond the fact that his aeroplane just happened to be pointing north(ish) at the time, if the pilot applied bank (and nothing else) he would find that the compass immediately indicated that his aircraft was actually turning. This was the most straightforward example of a false reading but, when manoeuvring, as the aircraft’s orientation and attitude changed, the compass could be induced to present a variety of contradictory indications. Confusing at the best of times, northerly turning error could sometimes be quite alarming, especially in conditions of poor visibility.

This problem was investigated by Capt K Lucas at F arnborough in 1915 and his explanation of the phenomenon resulted in the RAF Mk II Compass.¹³ This minimised the problem, but at the expense of the instrument’s being somewhat insensitive and sluggish. Once the Mk II had been deflected it would tend to swing lazily back and forth before settling back on north. If the pilot attempted to follow its indications too closely, while it was still swinging, this could set up a cycle in which he ‘chased’ the instrument. The error could sometimes build up to the extent that he could be seduced into flying in a circle, following the compass as it eventually took the ‘short way’ back to north (Figure D1 almost certainly shows instances of this having occurred).

Thus, while Lucas’ rationale had served to dispel the mystery surrounding the previously unpredictable behaviour of air compasses, the RAF Mk II was insufficiently responsive to cope with violent or prolonged manoeuvres and thus it too tended to inspire little confidence among many of the crews using it. On the other hand, those obliged to fly in the ‘bumpy’ conditions which were frequently encountered in the Middle East are said to have come to appreciate the Mk II’s relative lack of sensitivity.¹⁴

Although Farnborough continued to make its contribution, the acknowledged repository of compass lore lay within the Admiralty Compass Branch and in February 1917 it was given responsibility for all aircraft compasses by the Cowdray Air Board. This led to the Branch being elevated to Department status, still under the overall control of its original Director, Capt Frank Creagh-Osborne (RN Retd). The Department was subdivided into a Magnetic Compass Branch under Lt Cdr Colin Campbell, a Gyro Compass Branch under Lt Cdr Geoffrey Harrison and an Air Compass Branch. Since the need for an efficient airborne compass was common to both the RNAS and the RFC, in an early example of ‘jointry’, the latter was to be staffed by the Army, Capt Max Cooper-King RFC being installed as Superintendent.¹⁵ Shortly afterwards the whole empire moved from its central London address to the Admiralty Compass Observatory at Slough. A period of intensive effort soon produced the Type 5/17 Air Compass, the designation indicating the month in which it became available for general service use, ie May 1917. While still subject to, the now understood and therefore no longer alarming, northerly turning error, the Type 5/17 was more sensitive, yet more stable, than the RAF Mk II and thus had a significantly improved performance.

A Bristol Fighter, F4405, having its compass swung. (J M Bruce/G S Leslie collection)

Not satisfied with this, however, the team at Slough went on to develop the ultimate wartime compass, the Type 6/18, Lt Cdr Campbell and Dr G T Bennett generally being given the credit for having led this project. The Type 6/18 was the first practical aperiodic compass, which is to say that, while being very responsive, it was also ‘deadbeat’, ie if the needle was deflected it would return rapidly to north but without any significant overswing. Its relatively late introduction meant that insufficient Type 6/18s could be produced in time to equip more than a tiny fraction of the 3,300 aeroplanes that were on charge to operational units when the war ended, let alone the other 19,000 which were flying with training units and/or being held in reserve. Nevertheless, where it was available it had the potential to enhance the precision of dead reckoning and its reliability, its accuracy and its predictable behaviour encouraged pilots to place more faith in all of their instruments.

Along with progressively upgraded versions of both the RAF Mk II and the Type 5/17, the Type 6/18 remained in use as a standard air compass well into the post-war era. In fact, so successful was the basic design of the latter that the post-war P and O Series compasses which began to appear in the 1920s (and which could still be found in the cockpits of many RAF aeroplanes forty years later) were really progressive refinements of the Type 6/18 rather than new concepts.

As a footnote to the above, it is interesting to note that in 1919, when the post-war RAF required all of its officers to become pilots, one of the tests that they were required to undertake involved an exercise not unlike that which had been proposed by the RNAS in March 1918.16 The subject was required to fly on a nominated compass heading, his instructor checking the reading every 30 seconds. The average error was not to exceed 15°, the maximum permitted deviation being 30°. After fifteen minutes the student was handed a map on which he was expected to mark the position of the aircraft which, in view of the anticipated performance of the compass, was quite a challenge.

Dead Reckoning.

Since it was often necessary to maintain some kind of plot in the context of maritime operations, the RNAS had been well ahead of the RFC in the application of dead reckoning (DR), having adapted long-established nautical practices for airborne use. Regardless of how accurately it was maintained, however, a DR plot provided only an indication of where an aeroplane would be in still air. For this to be of any practical value in establishing where the aeroplane actually was, it was also necessary to know the speed and direction of the wind. Over the sea it was possible for an experienced observer to assess the local surface wind by inspection (by ‘drift and wind-lane’) and then, using a rule of thumb, to adjust this to provide a likely approx.-imation of the wind at his operating height. Alternatively, he could measure the actual wind velocity aloft by, for instance, tracking back and forth over a nominally static object, like a smoke float, and using a suitable device, eg a bearing plate or a bomb sight, to measure the drift on a variety of headings.

Wind finding was a time-consuming procedure, however, and in practice aeroplane crews would usually have accepted and applied the wind values provided at briefing. Since most sorties would have been of relatively short range, if not duration, this would normally have sufficed, although one imagines that, at least, some crews might have been able to request an update from their parent ship or shore station by W/T, should they ever have considered it necessary.

Airships carrying out extended patrols over the North Sea or the Western Approaches could also attempt to assess the wind themselves. One way of doing this would have been to head the ‘ship into wind until its drift had been neutralised and then to have reduced power until it ceased to make any headway. Once this ‘hovering’ condition had been achieved, the reciprocal of the craft’s heading and its true airspeed would have matched the direction and speed of the local wind. If this method was ever used, however, it would still have been a fairly laborious process and it seems far more likely that crews would have been notified by a broadcast from a shore station if a significant change in the wind had occurred or have requested one if they thought this to be the case. Alternatively, they could probably have obtained the current surface wind value from any shipping in the area.

Despite the increased attention being paid to the problems inherent in air navigation, it is questionable to what extent DR techniques were actually used in the air. Although improvements in instrumentation were being made, it was still not possible to measure airspeed, altitude or temperature to any great degree of accuracy and, as Figure D1 clearly shows, the performance of most compasses left much to be desired. Even the relatively advanced Type 6/18, would show an apparent change of heading whenever an adjacent Lewis gun was swung on its Scarff ring. All of this would, of course, have tended to confound any attempt at maintaining a precise plot. Furthermore, simply handling a chart, let alone plotting on it, at sub-zero temperatures with a gale of sleet howling through an open cockpit, presented considerable practical difficulties. The Bigsworth Plotting Board, with its chart clamps and integral compass rose, parallel rule and measuring scale, all mounted on an articulated arm, was an attempt to minimise this problem but it was hardly a complete solution.

The earliest lighthouses, like the one on the left were lit by acetylene. Later models (top) were electrically powered. Most (possibly all) were French-built.

Within the RFC/RAF the most useful practical application of DR techniques was at the planning stage of a long-range sortie. Although the science of meteorology was still relatively primitive, forecasts were regularly produced (see Annex H) and the observer could use these to lay his course using the best available wind information. Having made allowance for the anticipated wind effect, crews were then able to fly precalculated headings, refining these as necessary en route by observation of the actual drift being experienced and checking progress by reference to a map. Thus, despite the notional introduction of DR, it was essentially only an adjunct to map-reading and, so long as the weather did not change suddenly, such changes being very difficult to predict at the time, air pilotage provided an adequate practical solution to the problem of aerial navigation over land.

When the weather did change, however, it could cause considerable difficulty, especially at night. Contemporary instrumentation made prolonged flying in cloud an unrealistic proposition and if the ground became obscured the crew was deprived of its only fixing aid. If, as was very likely, the sudden development of clouds was associated with a marked change in the speed or direction of the wind (or both) a crew’s problems were considerably exacerbated. Bombing raids, even those carried out by the big Handley Pages of the Independent Force, only occasionally ranged as far as 150 miles from base and most targets were within 100 miles, but even so the return leg could become a painfully long drawn out affair if the prevailing westerly wind changed unexpectedly. In this event the headwind component at 5,000 feet could easily approach that of the aeroplane’s cruising speed of about 70 mph; conversely a strong wind from the beam could result in a disorienting 45° of drift.

Lighthouses.

Locally devised illuminated beacons had been in use on a makeshift basis before the end of 1914 but there was little call for night flying facilities in France prior to the advent of the Light Night Bomber Force in early 1917. The first lighthouses employed by the British were lit by acetylene but these were being progressively replaced by electric arc lights when the war ended. All of the lights were mobile and they were periodically moved to reflect squadron deployment patterns (and to confuse the enemy?). To support night operations from a total of eleven designated aerodromes, thirteen navigational lighthouses had been deployed by June 1917.¹⁷ By early November 1918 the RAF had twenty-four lighthouses available, of which seventeen were actually in use, the remainder being held in reserve.18 There was a further concentration of ten British lights in the vicinity of the Independent Force’s aerodromes in Lorraine.

‘You dunk it. I’ll listen’. A hydrophone being deployed from a Felixstowe F.3, N4230) during trials conducted off Shoeburyness in May 1918. (J M Bruce/G S Leslie collection)

It is worth noting that British crews were often able to exploit the enemy’s night navigation facilities (this undoubtedly, if inadvertently, having been a reciprocal arrangement), although the Germans tended to use pyrotechnics, rather than lighthouses. While these were probably less reassuring than the more or less permanent British beacons, they did have the significant advantage of being visible above low cloud and fog banks. Despite their apparent preference for fireworks, the Germans did deploy some lighthouses, a very powerful one near Lille being of particular value to British bomber crews, since it could be seen on a clear night virtually from take off.

Annex E. Hydrophones, 1918.

Trials with hydrophones designed to be used in aeroplanes began in 1917. While these early devices did permit the presence of a submarine to be detected, they could give no indication of its location which meant that they were of very limited tactical value. By early 1918 a much more useful directional system had been developed and, following successful experiments using flying boats, it entered service.

While the trials work had demonstrated that a suitably equipped seaplane could alight to listen for a contact and, if it found one, take off again to engage it, it had also shown that it was quite likely to lose track of its prey in the process. To solve this problem a second aeroplane needed to be on hand, airborne and ready to strike. Large twin-engined, manpower-intensive flying boats were not an economic proposition, however, so the preferred operational tactic was to employ pairs of floatplanes. Nevertheless, some big ‘boats were also equipped with hydrophones and by the end of the war a further variant of the system optimised for use by airships had been developed, although this never became operational.

The interpretation of hydrophone signals was made the business of the observer and to teach the necessary techniques, rather than continuing to use an RN training facility at Portland, the RAF established its own school at Aldeburgh. The system did become operational late in the war but the onset of autumnal weather, and seas too choppy to permit the frail floatplanes of the day to alight, prevented its achieving any success against U-boats.

Annex F. Wireless Direction Finding, 1914-18.

Ground-based Wireless Direction Finding.

The problem of fixing an aircraft’s position when operating out of sight of land had been very apparent from the outset and steps were taken to solve it in August 1915 when the Admiralty authorised the Marconi Company to build direction-finding (D/F) stations at a number of coastal sites. In effect these were an extension of a pre-existing system set up to intercept radio transmissions in order to detect and track enemy submarines – and the much feared Zeppelins. By December the Admiralty was able to issue standing instructions to aircrews operating over the North Sea. Thereafter it became routine for patrolling aircraft to make an hourly Morse transmission consisting of its callsign repeated ten times. This was sufficient to permit a receiving station to establish the bearing of the incoming signal and, by comparing this with bearings taken at the same time by other stations, it was possible to fix the source of the transmission by triangulation and relay this information to the crew.

The system was far from perfect, however, and in July 1917 there was an exchange of concerned correspondence in naval circles to establish why the system was not being exploited to the full. An investigation, based on the use of the D/F stations at Lowestoft and Flamborough Head by Howden-based airships, revealed that the best accuracy that could be expected would provide a fix within a 5 mile square but that 10-15 miles was more usual and greater than 20 miles far from unknown.

Naval aircrews were already well aware of this, of course, and tended not to trust radio fixes, putting their faith in them only when no better information was available. Since RNAS airships and seaplanes routinely carried wireless sets, it was probably easier for their crews to keep track of their whereabouts by simply asking one of the friendly ships in the relatively crowded North Sea. By September, with attention being focused on the D/F system, the incidence of gross errors had been significantly reduced, with the notable, and critical, exception of fixes provided for airships operating close inshore. Unfortunately, without the provision of many more D/F sites, the geometry of bearings taken by coastal stations on an airship which was also close to the coast was bound to result in a poor ‘cut’ and a fix of doubtful accuracy. Under foggy conditions, therefore, a crew carrying out an inshore patrol or one attempting to return to base had little alternative but to hold off a safe distance out to sea and wait for the visibility to improve.

While W/T was used extensively for air-to-ground communications over the Western Front, the RFC made only limited use of direction-finding. This was largely because the majority of air operations were conducted within a relatively confined area which meant that navigation presented few problems. A D/F service might have been of some value to crews operating at longer ranges but, since few of the RFC’s aeroplanes carried receivers, it would not have been possible to inform their crews of their position. Nevertheless, one practical application was developed, following the establishment of a network of D/F sites in France during 1917. These were able to pinpoint the location of German aeroplanes by triangulating their radio transmissions, this information then being relayed to patrolling fighters by visual signals. This primitive form of ‘GCI’ (Ground Controlled Interception) did achieve some success in frustrating enemy air activity.

In contrast to RFC practice, most of the RNAS patrol aircraft based on the coast were equipped with two-way radios. Early in 1917 three naval D/F stations were deployed to France with the primary aim of detecting and plotting the movements of Zeppelins and of enemy submarines operating from Belgian ports. After some initial instability these facilities were more or less permanently located at Coxyde (La Panne), Lampernisse and St Pol and they soon began to provide a similar oversea fixing service to that offered by the D/F stations in the UK.

Airborne Wireless Direction Finding.

While ground D/F stations had considerable potential, the success attending experiments with airborne direction finding probably had even greater significance for the long-term future of air navigation. This work was begun by the RNAS and continued under the auspices of the RAF. Initial trials were conducted at Cranwell in 1917 under the direction of Lt S Smith RNVR, using an RE7 with a single fixed coil installed athwartships to see whether it was possible to home on a transmission. While not entirely successful, this experiment had shown some promise. Once the problem of magneto interference had been overcome (by Capt J Robinson RFC) and a second, fore-and-aft, coil had been included, the system was demonstrated to work. By turning the aeroplane until the strength of the received signal indicated that it was pointing directly at the transmitter, bearings were successfully obtained from stations as close as Cleethorpes and as distant as Paris.19

The next improvement was to devise a set of moveable coils so that bearings could be measured without the aeroplane having to deviate from its course and by October 1917 a system suitable for use in Handley Pages had been developed. An early example was sent to Dunkirk for trials but concern over the possibility that the technology would be compromised if it fell into enemy hands caused these to be cancelled before the equipment had even been installed.²⁰

In the meantime trials work continued and by January 1918 Cranwell was able to demonstrate a maximum error of 2°20′, average accuracy being less than 1°, although the system was still confined to taking fore and aft readings.²¹ At the same time another report observed that ‘the accuracy of the D/F system is much greater than that with which the pilot can steer’; this may have been of only incidental interest to the sharply focused trials team but it had significant implications for the practical utility of the technique (see Note 29). Nevertheless, the concept was considered to show sufficient promise for plans to be drawn up for its introduction. In February two equipment schedules were published for O/400s, one of which reflected the requirements of the new W/T arrangement,²² and No 1(T) Wireless School was directed to start teaching direction finding technology to wireless operators earmarked to work on Handley Pages.²³

It was also necessary to lay down the composition of the crew of a Handley Page equipped with the prospective navigational system and this evidently proved to be easier said than done, because three successive directives were issued in just four weeks. The first had specified two pilots, an observer (status unspecified, but, at the time he would have been assumed to have been an NCO) and an assistant armourer. This was subsequently amended to provide two pilots, an NCO (specified this time) observer and a wireless operator. A fortnight later, this was superseded by a far more detailed instruction, directing the Training Division to make arrangements to provide the personnel noted at Figure F1.²⁴ In the event, this all proved to be somewhat academic as the RAF eventually found that the ideal crew for an O/400 was a pilot and two observers, all of them commissioned, and it never actually got around to deploying the D/F system, although, if it had done, it is probable that a wireless operator would also have been carried.

Fig F1. Crew of a D/F-equipped 0/400 as specified in late February 1918.

To support the airborne D/F system a series of radio beacons would be required and the identification and/or construction of a suitable selection of transmitters began in February 1918, the chosen sites eventually including Poldhu (Helston), Stonehaven, Ipswich, Chelmsford and Horsea Island (Portsmouth). With each of these transmitting for three (different) minutes in every hour, it was theoretically possible to take a bearing from several stations in turn, permitting the position of an aircraft to be established with reasonable accuracy. Development work continued at Cranwell with Capt J D Greenwood RFC supervising the installation of a set of rotating coils in an O/400. After some local trials, the first true radio navigation exercise, a flight from Cranwell to Stonehenge, was undertaken on 23 March 1918. The navigator was Sqn Cdr H F Towler, a pilot who, following a lengthy stint as CO of the RNAS Observers School at Eastchurch, had become a prominent member of the development team.

As a result of the success of this flight, preparations were put in hand to introduce airborne D/F training at No 1 School of Navigation and Bomb Dropping. In April 1918 No 97 Sqn moved to Stonehenge where its crews were to be taught how to use the system. In the event, the Sunbeam engines of the squadron’s Handley Pages proved to be troublesome and little progress was made until they were replaced by Rolls-Royce powered aeroplanes. In the meantime, confidence was gained by driving the trainee observers and wireless operators around Salisbury Plain in a Crossley tender fitted with the coils so that they could take bearings at ground level.

Although trials had shown that the equipment worked well enough when it was maintained and operated by skilled personnel, there was some doubt as to whether this would be the case if it were in the hands of an ordinary squadron. Furthermore, large scale deployment would demand the provision of many more highly trained technicians, as the equipment required a certain degree of skilled adjustment and fine tuning. There were also reservations over the acceptability of the disruption in aircraft production which might occur if the equipment were to be installed. To cap all of this, doubts were being expressed as to the real extent of the tactical advantage that deployment would provide.

One of two types of moveable coil developed for Handley Pages. Both stood about five feet tall and occupied virtually the entire depth of the centre fuselage, immediately aft of the rear cabin. The upper ring was graduated in degrees. The whole device was rotated to find the position of maximum signal strength (this procedure was not quite as straightforward as it sounds) and the bearing, relative to the aeroplane’s heading, read off against a datum fixed to the aircraft’s centre line.

With enthusiasm for the new-fangled technology rapidly waning, it was decided to postpone deployment, pending the availability of the very long-range Handley Page V/1500. Early in July, therefore, No 97 Sqn’s trained personnel were transferred to the recently formed No 2 School of Navigation and Bomb Dropping at Andover where they were to continue directional wireless training. The squadron’s five O/400s (B9446, C9636, C9641, D4571 and D5417) plus a couple of FE2bs and a Curtiss JN4 were bequeathed to No 115 Sqn at Netheravon. No 97 Sqn was issued with unmodified aeroplanes and the following month it joined the Independent Force as a standard heavy bomber squadron.

In a memorandum dated 1 October 1918, Lt-Col G P Grenfell (O.4 at the Air Ministry), summarised the overall position regarding directional wireless.25 All associated aircrew training had been suspended, since the technology had ‘not yet proved its value as a means of navigation’. Qualified personnel had been earmarked for recall in the event of a resumption of the programme but were otherwise available to the Director of Training for other instructional duties. So far as specialist technicians were concerned, training, which had been under way at No 1(T) Wireless School at Farnborough since January under the direction of Capt C K Chandler, had been severely cut back, although it had not been stopped altogether.

This sketch shows the alternative wiring arrangement for the airborne D/F facility in an 0/400. Fixed coils, aligned with the lateral and longitudinal axes, could be wound within the wing structure and/or a rotating coil could be mounted in the rear fuselage. Since there was ample space, it was recommended that both systems should be installed in HPs, but in early post-war continuation trials conducted using Bristol Fighters from Biggin Hill only the fixed arrays were used.

In the meantime tests were to continue with an O/400 at Bircham Newton, where Lt-Col R H Mulock was then forming the 27th Group to control the operations of the projected four-aircraft squadrons of V/1500s. Further trials would also continue to be conducted by the Wireless Experimental Establishment at Biggin Hill. In addition, two sets of D/F equipment and a working party²⁶ were being despatched to eastern France to permit a pair of O/400s to be fitted out for field trials and familiarisation with the Independent Force, at least one of these evidently still being in commission at the Armistice (see below).

To support this residual experimental work the high-power transmitters at Chelmsford and Poldhu would be available, as were low-power stations at Andover, Borden and Milton. It was also hoped to persuade the French to cooperate, permitting use of their very powerful transmitters at Paris and Lyons. All three locations still working on the system, ie Bircham Newton, Biggin Hill and Courban, were to exchange reports and keep the Air Ministry informed of progress. At much the same time American requests to be given access to British D/F technology were rebuffed. It was argued that the system was not yet sufficiently developed for it to be deployed and that widening the circle of those ‘in the know’ might compromise the techniques involved. Furthermore, it was considered that, working in isolation, the Americans might well make some unforeseen breakthrough.²⁷

So far as feedback from France was concerned, Sir Hugh Trenchard showed no great enthusiasm for the system. His overriding priority was to sustain the strength of his existing squadrons and, if possible, to obtain more of them. The GOC was certainly not prepared to accept an interruption in the supply of replacement aeroplanes in return for the uncertain advantages which, its advocates claimed, might accrue from the deployment of an airborne D/F capability. He even declined the offer of a local low-powered beacon to support development work in France, since he was content that the existing lighthouse network provided adequate navigational facilities, particularly in assisting in recoveries to base.

Only a few weeks after Lt-Col Grenfell had circulated his memorandum, the situation was transformed by the flight, made in conditions of marginal visibility, of a wireless equipped O/400 (C9694) from Biggin Hill to Paris on 19 October; a return flight being completed three days later. In addition to a qualified W/T Operator, AM J Joyce, the crew included a very senior W/T Officer, Col L F Blandy, to supervise the handling of the D/F equipment. The designnated navigator was Maj Towler who flew buried in the centre fuselage from where he had no sight of the ground throughout either flight.28 His only instruments were a compass and altimeter, no airspeed indicator having been provided. Using only the meagre information available to him plus radio bearings taken from Paris, Horsea Island, Chelmsford, Poldhu and, ironically, on the return journey, two from Nauen (near Berlin), Towler maintained a plot of the aircraft’s position, calculated the wind and passed headings for the pilot, Lt A Woodward, to steer. As verified by an independent observer, Maj W H D Acland, who was able to monitor progress by map-reading, the aircraft stayed close to its planned track on both occasions and arrived within two miles of its destinations.²⁹ Another officer who had come along to see fair play was Maj-Gen Brancker.

Too many struts! This 0/400, D9712, has been fitted with the additional struts necessary to carry the fixed D/F loops – see diagram on previous page. (J M Bruce/G S Leslie collection)

The flight to Paris was deemed to be the convincing demonstration that had been required and on 21 October, before the aircraft had even made its return journey, an amendment to the establishment of O/400 squadrons was issued to support the introduction of directional wireless. The cost, in both equipment and manpower, turned out to be quite substantial, so one can perhaps understand some of the earlier misgivings over the implications of this technique. Over and above its normal entitlement, a squadron operating D/F equipment was to be provided with a repair lorry and three tenders, two light and one heavy. Its establishment of personnel was amended to reflect no fewer than twenty-five additional men: a technical officer (wireless), seven wireless mechanics (one of them a sergeant), fourteen wireless operators (one of them a corporal) and a driver for each of the three vehicles.

The CFS Mk IV bomb sight.

At the same time arrangements were put in hand to restart training at Andover and negotiations were hastily started with the French, who had also been working on directional wireless, with a view to co-ordinating the use of the high power beacons available to the Allies. At a conference held in Paris on 24 October it was agreed that, when required, the designated stations would transmit for two three-minute periods in each hour.30 The minutes of the conference recorded that the French claimed to have been achieving an overall accuracy of +/-3°, which was comparable to British experience (but see Note 29). The minutes also noted that it was not foreseen that there would be any call to activate the system to support operations before 1919.

In the event, the Armistice intervened long before preparations to use the system could be completed. Three days later, in preparation for its disbandment, orders were issued directing the entire Independent Force to withdraw to western France.³¹ Among the units to be redeployed was No 97 Sqn which was to move from Xaffévillers to St Inglevert, where it would be subordinated to HQ RAF. HQ Independent Force wrote to the squadron’s prospective new ‘owners’ to let them know that one of its O/400s (presumably one of the two trials aircraft noted above) was ‘fitted with directional wireless apparatus’ and that ‘directional wireless observers and operators’ were attached to this unit. Technically, despite the Armistice, a state of war still existed with Germany and this was underlined by the fact that the letter also advised that the Air Ministry had ruled that ‘until further instructions, directional wireless is not to be employed on the enemy’s side of the lines.’³²

Thus, while the practicality of airborne D/F had been convincingly demonstrated during the last few weeks of the war, and a cadre of observers and technicians had been trained to handle the system, it is clear that the considerable potential of this technology had never been exploited operationally. Development work continued for a while and by December 1918 a set of moveable coils had been installed in a Felixstowe F.3 at the Isle of Grain. Andover was subsequently directed to build three more sets for the F.5s which were to be used by the Long Navigation Course when this was set up at Calshot in 1920. It has not been established whether these coils were ever installed but, whether they were or not, airborne D/F activity soon faded away.

Annex G. Bomb Sights, 1914-18.

The policy decision which had made observers responsible for bomb-aiming in the summer of 1918 (see pages 117-118) was, in part, a reflection of the success which had finally attended a prolonged, if uncoordinated, effort to develop an efficient bomb sight. Fertile minds within both the RNAS and the RFC had been at work since before the war, creating a variety of increasingly complex devices all of which were intended to introduce a degree of science into the art of bomb-aiming.

Personalities who made notable pre-war contributions to this process included Maj H Musgrave and Lt C G S Gould within the RFC’s Military Wing and Lt R H Clark-Hall and Sub-Lt J L Travers of the Naval Wing. Despite their efforts, when the war began the RFC had no effective means of aiming bombs while the RNAS had only a primitive ‘lever sight’. Before the end of 1914, however, 2/Lt R B Bourdillon had produced a promising design. After further development at Upavon, in collaboration with Henry Tizard and Lt G M Dobson, this emerged as the CFS Bomb Sight. Bourdillon’s basic concept was produced in a number of variants, the most numerous, the Mk IVb, being used by both the RFC and the RNAS until well into 1917.³³

In the meantime development continued with the RNAS introducing the Equal Distance Sight (which was used in conjunction with a special reversing stop-watch) in 1916, and the RFC devising a means of using the CFS principle in conjunction with a periscope on the RE7 and the Martinsyde G.100/102 in an attempt to overcome the problems involved in direct sighting. While these devices did permit a notional degree of accuracy to be achieved they were all quite difficult to use effectively and they all required a timed run to establish groundspeed – which, under combat conditions, was often extremely inconvenient, to say the least.

Lt Cdr H E Wimperis’ drift sights were the most common bomb-aiming devices in operational use during 1918. This is a Low Height Mk II on which the height adjustment was made by pinching the clamps and sliding the datum up and down a calibrated vertical scale.

To overcome this limitation Lt Cdr H E Wimperis developed his Drift Sight which began to enter service in substantial numbers during 1917. By adjusting three scales to reflect the aircraft’s height, its airspeed and (the reciprocal of) the wind velocity, the sight automatically produced the correct bombing angle. If time and the enemy permitted, it was also possible to use the sight to assess the local wind velocity, but in practice it was more usual to set the forecast wind which was available at briefing. Wimperis’ basic design was eventually produced in five sub-variants: the High Altitude Drift Sight (HAD) Mks I and Ia, and the Low Height Drift Sight (LH) Mks II, IIa and III. The Mks I, II and III were all calibrated in knots for use at sea (the Mk III being specifically calibrated for airships), and the Mks Ia and IIa in mph for use over land. Like all of the preceding designs, the family of Drift Sights was intended to be used while flying directly up- or downwind, which imposed a significant tactical limitation, although Wimperis’ sights could be used for cross-wind attacks at the expense of some theoretical accuracy.

With the exception of the periscopic variety (which never became widely available, apparently being abandoned after field trials), all of the sights described thus far were normally fitted on the outside of the fuselage, requiring the user, pre- 1918 usually the pilot, to grope about in the slipstream while making the necessary adjustments. Ultimately he had to sight the target by leaning out of the cockpit to position his eye directly above the sight, to ensure that the target tracked along the drift wires, while simultaneously doing his best to keep the aeroplane on an even keel by reference to spirit levels. To do this properly required a long and articulated neck and rather more arms than the average pilot possessed, these inadequacies having to be compensated for by a great deal of contortion.

With the aim of producing a more ‘user friendly’ system the RFC introduced the Negative Lens Sight in 1917. In essence this amounted to a rectangular box set in a hole cut in the floor of the pilot’s cockpit. The upper face of the box was a lens, suitably shaped (plano-concave) to provide an extended view of the ground beneath and ahead of the aircraft. Longitudinal sighting wires were stretched centrally across the upper and lower faces of the box to establish an approximation of the vertical (so long as the wings were kept level) and to facilitate lining-up on the target. Three fixed wires were arranged laterally across the top of the box, serving as backsights – one for each of three predetermined bombing heights, 6,000, 10,000 and 15,000 feet, each of which was associated with a specific airspeed. The foresight was another lateral wire stretched across the bottom of the box, its fore-and-aft position being adjusted by setting the forecast windspeed for the upwind or downwind case.

The Negative Lens Sight. There were often three lateral backsight wires, rather than the two in this diagram.

This up-ended Bristol Fighter reveals two apertures cut in its undersides. The rectangular one in the centre section of the lower wing permitted the pilot to see the ground via his Negative Lens Sight; the circular cut out was for a camera. (J M Bruce/G S Leslie collection)

This sight was much easier and more comfortable to use, although it still required attacks to be made directly up- or downwind. It was not very accurate, however, as it had to be pre-set before take off, any subsequent adjustment of the wind speed or deviation from the three bombing altitudes/airspeeds for which the sight was calibrated having to made by eye. It also had a tendency to become obscured by oil leaking from the engine. The extent of this problem varied, depending upon the aircraft in which the sight was installed, but it would have ruled out its use on rotary-powered types in which oil spillage was actually a design feature and, in the case of the DH 9, the most numerous of the RAF’s bombers, the view through the sight was obscured when a 230 lb bomb was being carried on the centre-line station.

Although it was probably never used on operations, the best wartime bomb sights were later versions of the Course Setting Bomb Sight (CSBS) originally developed by H E Wimperis for the RNAS which he subsequently perfected for the RAF. This particular version is the Mk Ia which was intended to be used at altitudes of between 300 and 2,500 feet, the Mk II variants being calibrated for high level work up to 14,000 feet. By taking drift readings on different headings, the CSBS could also be used to find the local wind velocity at height. Theoretically, this wind could have been measured very precisely but, in practice, the results would have been degraded by the relative inaccuracy of the means available to measure other essential parameters, including the aircraft’s course (heading), its indicated airspeed, the air temperature and the atmospheric pressure (altitude). The Mk IX version of this series was still the standard bomb sight during the early years of WW II – see page 202.

Because of its limitations the Negative Lens Sight was soon relegated to being a back-up system, only to be used for bomb-aiming in extremis. On the other hand, the fact that it enabled the pilot to see directly beneath and, to some degree, ahead of the aircraft actually made it of more practical use for reconnaissance work than it was for bombing. While the sight was too crude to permit accurate assessment of the forward throw for bombing, it was perfectly adequate for ensuring that the aeroplane actually overflew the target and this made it very handy for photographic work. At one stage it had been intended to fit Negative Lens Sights in all corps types but it proved to be very difficult to install in the RE8. A remote installation was eventually devised, with the ‘view’ being presented to the pilot via a series of lenses, but this system was deemed to be overcomplicated and it was finally decided not to persevere with the RE8 programme.³⁴ Although it had a number of drawbacks, the Negative Lens Sight was a very useful adjunct to the range of tools available to a crew and they were produced in far greater numbers than any other type of sight.

The most efficient wartime aiming device, the Course Setting Bomb Sight, was also designed by Henry Wimperis. Development work was initiated by the RNAS in 1917, the project being taken over by the RAF in the following year. In its final form the integral compass was an aperiodic type and versions of the sight were calibrated for bombing, on any heading, at heights up to 14,000 feet with allowance being made for such refinements as the trail angle associated with the ballistic characteristics of each type of bomb. While it was relatively complicated, this complexity conferred a useful degree of flexibility and it could also be used for wind-finding using the principle of multiple drifts – the so-called ‘wind star’ method. Very few Course Setting Sights saw service before the Armistice but they became standard post-war RAF equipment, as did Wimperis’ earlier Drift Sights which also remained in use for many years.

Another avenue that was being explored was the exploitation of the principle of the pendulum in order to provide a level datum for sighting. The possibility of using electrically driven and/or wind-spun gyroscopes to achieve the same aim was also being investigated by 1916 and by mid-1918 two practical applications had emerged, the Horsley and the Gray Sights. The Gray Sight appears to have shown the greater promise and it was being actively developed for Handley Pages as the war ended.³⁵

In the event no stabilised sights ever reached the squadrons, but it is interesting to note that the Gray Sight was to have been installed in the rear cabin of the O/400, necessitating the provision of a remote steering indicator to enable the bomb-aimer to direct the pilot towards the target. This may have been the relatively simple device introduced by the RNAS in 1917 which permitted the observer to instruct the pilot to turn left or right or to fly straight ahead by illuminating one of three lamps on the instrument panel. This system had proved to be rather crude and insensitive, however, and it is more likely that the Gray Sight was intended to be used in conjunction with a steering indicator which enabled the observer to give the pilot a precise heading to maintain (see Note 28).

Fig G1. Production figures for WW I bomb sights.

The table at Figure G1 provides some indication of the production figures for wartime bomb sights of various models.³⁶

Annex H. Meteorology, 1914-18.

Apart from needing to know the direction and strength of the wind in order to navigate with any accuracy, other aspects of the weather were crucial to the planning and conduct of air operations. Although there was a sound understanding of the concept of air masses long before 1914, detailed knowledge of the upper air was still scanty and the familiar concept of frontal systems, essential to successful forecasting, did not materialise until the early 1920s.³⁷

Nevertheless, from the summer of 1915 periodic observations of actual conditions were reported to ‘Meteor’ (the Meteorological Service, RE) at GHQ in France and regular releases of balloons from half-a-dozen or more sites, permitted the winds aloft to be assessed by theodolite tracking. While it was sometimes possible to follow these balloons to 15,000 feet or even higher, the cloud base often intervened and they would be lost to view at much lower altitudes. To overcome this limitation, a means of measuring wind velocity was devised that involved an aeroplane flying through a series of anti-aircraft shell bursts fired at known intervals to establish wind speed, direction being read off the compass.

An FK8 of Meteor Flight taking off from Berck circa February 1919. The ‘comet’ marking was, presumably, as close as the unit could get to a ‘meteor’. (R Douglas)

To assist with this procedure, and to carry out upper air observations of temperature and humidity using aeroplanes, a Meteor Flight was set up at Berck early in 1918. Among its more significant contributions to the science of meteorology was a series of air-to-air cloud photographs taken by Capt C K M Douglas and Lt E H Sessions.

All of this current information, combined with a knowledge of what had happened in the recent past ‘upstream’ in the British Isles and/or in southern France, permitted Capt (later Lt-Col) E Gold to maintain a reasonable plot of the developing weather pattern. From this he was able to forecast ahead for several hours with reasonable confidence. He could not, however, guarantee to get it right every time.³⁸

It should be appreciated, incidentally, that aviators were not the only people who required meteorological information. The artillery needed to allow for the wind aloft in calculating ballistic trajectories, for instance, and the forecast surface wind was of absolutely critical importance to those concerned with gas warfare.

Contemporary diagram illustrating a means of establishing wind velocity above cloud.

Annex I. Knots v mph.

Most aircrew reading this book will probably take it for granted that, so far as speed and distance are concerned, aeroplanes have always worked in knots and nautical miles. This feels natural because the meridians on maps and charts are divided into minutes of latitude, each of which is one nautical mile, thus providing a convenient measuring scale. Furthermore, when using astro the elevation of a body is measured in minutes of arc, which equate to nautical miles, and the intercept that eventually has to be plotted on the chart is also measured in nautical miles. Finally, regardless of how the airspeed indicator was calibrated, the air mileage unit (AMU) was fed with knots which it turned into the nautical miles which it fed to the Air Position Indicator (API).

It may be a little surprising, therefore, to find that, while naval and maritime aircraft always had their ASIs calibrated in knots, until 1945 the bulk of the Air Force worked in mph. This involved some duplication and, albeit minor, inconvenience. For example, Pilots Notes for certain aero-planes had to address both cases 39 and, for the benefit of navigators, it was also necessary to print a statute miles scale on plotting charts.

Attention was first drawn to this somewhat anomalous situation, which was mirrored by the Americans – the US Navy worked in knots while the USAAF used mph – in June 1942.⁴⁰ Apart from knots being more appropriate/practical for Bomber, Coastal and Transport Commands (and there was no fundamental disadvantage for Fighter Command), standardisation would also avoid duplication in stock holdings and the potential for accidents arising from a pilot’s using the ‘wrong’ number.

A Meteorological Officer about to release a clutch of balloons to permit the upper winds to be assessed by theodolite tracking. This picture has been associated with No 143 Sqn, which suggests that it may have been taken at Detling, but the same procedure would have applied in France. (J M Bruce/G S Leslie collection)

The decision to adopt nautical miles/knots was eventually announced in April 1945.⁴¹ Bomber Command were the first to implement the new policy. Having accumulated adequate stocks of new ASIs, the change was implemented with effect from 1 April when all aircraft had their instruments changed more or less overnight. Coastal Command already complied with the new policy, of course, with Transport and Flying Training Commands soon following suit. The new policy was publicised via Tee Emm in a short essay that also explained that Fighter Command and the Tactical Air Force on the continent would not be able to adopt the new procedure immediately, as it would be incompatible with the complex ground network that supported their operations.42

Section of a wartime plotting chart featuring a supplementary statute miles scale.

Fig J1. The conditions governing the award of RAF Air Navigation Certificates as originally promulgated and as modified in 1938.

Annex J. The relationship between the RAF’s Air Navigation Certificates and Warrants and civil Air Navigation Licences.

In order to embody the resolutions adopted by the International Convention for Air Navigation (aka the Paris Convention) of 1919, the British Government passed its Air Navigation Act, 1920. The regulations that followed from this required all civil pilots to hold one of three licences. The Class A Licence covered private pilots; commercial pilots required a Class B Licence or a Master Pilot’s Certificate, the latter being based on the Class B but reflecting extensive practical flying experience and additional formal certification as an air navigator. There were two forms of Navigator’s Licence, First and Second Class, the main difference being the number of flying hours logged. In essence, any civil aircraft carrying passengers or goods for hire or reward on an international flight of more than 625 miles over the high seas or uninhabited terrain or at night was required to have on board the holder of a First Class Navigator’s Licence. For flights of between 100 and 625 miles a Second Class Licence sufficed.

In practice there were very few (if any) dedicated civil navigators between the wars, it being the normal practice for the holder of a Class B pilot’s licence to be dual-qualified by gaining, in addition, a First or Second Class Navigator’s Licence. Provision of suitable courses of instruction for civil pilots had been the original reason for the establishment of the navigation school run by Air Service Training and of The Imperial School of Air Navigation, the facilities of both of which would be exploited by the RAF in 1937 (see Note 60 to Chapter 16).

A little confusingly, in the wake of the civil Navigator’s Licences, the RAF had followed suit by introducing its own series of three Air Navigation Certificates in 1924. The conditions governing the award of each of these are summarised in the second column of Figure J1. All certificates were issued on the authority of the Air Ministry, specifically the Air Member for Personnel, in response to applications made ‘through the usual channels’ and supported by navigation log books and/or other documentary evidence.⁴³

As described elsewhere (see Note 76 to Chapter 15) there were a number of changes among the units conducting postgraduate training in navigation during the 1930s and in 1938 the regulations associated with Air Navigation Certificates were revised and updated to reflect contemporary practice.⁴⁴ Under the amended arrangements, which are summarised in the third column of Figure J1, 2nd Class Certificates were now issued by the COs of the School of Air Navigation at Manston or the School of General Reconnaissance at Thorney Island, but applications for a 1st Class or Air Master’s Certificate still had to be submitted to the Air Ministry along with flying log books, astro sight books and any other supporting documentation. It is not known who authorised any early certificates that may have been issued by the Air Ministry but by 1941 they were being personally endorsed by the Air Member for Training.

Two points arise. First, when the RAF introduced its certificates, air navigation was still perceived to be the exclusive preserve of the pilot, because there were virtually no observers in the 1920s and there would certainly have been no room for corporals on any of the courses being run at the SAN and elsewhere in 1938. During the inter-war years, therefore, the only people able to qualify for any of these certificates would have been pilots and, while 2nd Class certificates may have been relatively commonplace in 1938-39, in that most officers attending the relevant courses would have been issued with one, there will have been few 1st Class and even fewer Air Master’s Certificates. Why? Because of the flying hours proviso. The contemporary RAF definition of what passed for flight time ‘as navigator’ would probably have been fairly generous but, even so, it is doubtful whether many pilots, with the possible exception of some flying boat pilots, would have been able to produce convincing evidence to support a claim for 200 hours, let alone 600 (which would have represented something in excess of two years’ worth of flying) of flight time ‘as navigator’.

Since pilots seem unlikely to have claimed many of the higher grade certificates, most of those that were eventually issued during WW II will have gone to observers. But the Specialist Navigation Course was open only to officers and there were no commissioned observers until April 1940 so it would have been 1941 before many (any?) of them would have been selected to attend the course which, by that time, was being run at Port Albert in Canada.45

The second point that needs to be made is that it should be appreciated that these RAF certificates were an entirely internal affair. As KR 366, and the later KR 382, made very clear: ‘The possession of an air navigator’s certificate does not entitle the holder to navigate any civil aircraft …’ To do that one would have had to have held a First or Second Class Navigator’s Licence. These were also issued by the Air Ministry, but specifically by the Department of the Director-General of Civil Aviation.

In 1938 the military side of the Ministry succeeded in obtaining a waiver that exempted RAF pilots who had obtained adequate marks on the Short Navigation Course or the General Reconnaissance Course.⁴⁶ from having to sit most of the examination papers that candidates applying for a Second Class civil ticket were obliged to pass.⁴⁷ The exceptions embraced a paper dealing with various aspects of national and international air law, an oral examination on the civil radio organisation and its associated procedures, and practical tests in signalling by both visual and aural means (semaphore and Morse by Aldis lamp and W/T).

Meanwhile, since August 1938, direct entrant observers had been attending formal courses at Civil Air Navigation Schools where the syllabus covered much the same subjects as those examined for a civil licence. Following the elevation of observers to sergeant rank in January 1939 negotiations began with the aim of having the exemptions, recently granted to RAF pilots who had been formally trained as navigators, extended to embrace observers.⁴⁸ This was probably a little premature because the Department of Civil Aviation was hardly likely to acknowledge the navigational skills of military observers until the RAF was prepared to issue them with its in-house Air Navigation Certificates, and this it did not actually do until 1941.

In October of that year (1941) the Air Ministry restated what it intended the issue of a 2nd Class RAF Certificate to signify. In short, it identified the holder as a competent practical navigator, while a 1st Class ticket implied additional competence in the theory of navigation. Those entitled to hold a 2nd Class Certificate now included the following.⁴⁹

a. A pilot or observer who had completed a Short Navigation Instructors Course (ie the current ‘sn’ course).

b. A pilot who had completed a course at a School of General Reconnaissance.

c. An observer or pilot who had completed a ‘special navigation course’ in an overseas RAF command, provided that the syllabus had been approved by the Air Ministry.⁵⁰

d. A pilot or observer who had been certified as being competent in astro or had been certified by his CO as having flown ten operational sorties, each of at least three hours’ duration, while performing reliably as navigator.

The most significant innovation, however, was that, in addition to the above, an observer who had performed adequately in basic training could also now hold a 2nd Class RAF Certificate. It was no coincidence that this announcement was made just nine days after the opening of No 1 Elementary Air Observers School (see page 236), which had effectively extended both the length and the breadth of basic navigation training. Thereafter, any observer who had attained at least 60% in each navigational subject, and an overall average of 70%, on that course, and on his subsequent course at an Air Observers School, was now automatically entitled to a 2nd Class ticket.

An RAF Air Navigator’s Certificate First Class as issued in 1942; the original was about 7½ inches square.

Within a mere three weeks of the RAF’s having finally acknowledged the capabilities of its observers, the Department of Civil Aviation had amended its list of candidates who were exempted from sitting several of the examination papers required to obtain a civil Second Class Navigator’s Licence. The list now embraced applicants who had completed one of the following.51

a. A pre-war Short (ie the ten week) Navigation Course at the SAN (ie the ‘sn’).

b. A short wartime Navigation Instructors Course at the SAN (ie the ‘sn’).

c. The Navigation Reconnaissance Course at a School of General Reconnaissance.

d. The courses at an Elementary Air Observers School and an Air Observers School.

The usual provisos about pass marks remained and exams still had to be passed in aviation law, civil procedures and signalling, but the point was that any competent RAF observer could now acquire a civil licence with the relative ease that had previously been confined to pilots.

Another milestone had been passed on the long road towards recognition of the professional RAF navigator.

That said, this sudden burst of autumnal activity in 1941 was no coincidence; it was the culmination of a lengthy exchange of inter-departmental correspondence which had involved a certain amount of pragmatism. The problem that had needed to be solved was that, unlike pre-war days, it was not a forgone conclusion that all holders of a wartime Class B civil pilot’s licence would also be a licensed navigator, and this had presented BOAC with crewing difficulties on some of its routes.

Indeed, in order to sustain a commercial network at all it had been necessary to second a number of RAF pilots to the Corporation but none of these men were licensed as navigators.⁵² To realise the potential represented by these underqualified military pilots they would either have to be trained as navigators themselves or be accompanied by already-qualified military navigators. The second option would be much quicker, and cheaper, but – even in wartime – it would be necessary to abide by international law and to fly as navigators in civil-registered aeroplanes the RAF’s observers would need to hold a civil licence. This was the catalyst that had provoked the co-ordinated shifts in policy that had been introduced in such short order during October 1941.

Although there would subsequently be some tinkering with the details of what was covered by a wartime civil licence, these provisions sufficed until 1945 when it became necessary to reinstate full peacetime standards, which, because of the considerable advances in both aids and techniques, were far more stringent than they had been before the war. It was anticipated that there would be a postwar boom in civil aviation and that the RAF would inevitably become the major provider of aircrew for ‘UK Ltd’. This provoked a requirement for a means of endorsing the capabilities of RAF navigators that would be recognised by the civil licensing authority. This was necessary because, while the RAF’s certificates may have appeared to read across to civil aviation (and many folk believed that they did), this had never actually been the case.

In practice, the RAF recognised degrees of navigational expertise by the application of suitable annotations (‘N’ and ‘sn’) both to appointments and to individuals, not by the issue of its certificates. Similarly, the Department (now the Ministry) of Civil Aviation granted its exemptions from examinations on the basis of RAF courses passed, not the possession of an RAF certificate. Thus, while the issue of an RAF certificate may have had some intrinsic significance to the holder, it was a non-negotiable document and thus had no ‘market value’ whatsoever, and, as KR 382 (and KR 366 before it) had always made very clear, it specifically did not read across to a civil licence.

With effect from September 1945, therefore, the RAF ceased to issue certificates and replaced them with 1st and 2nd Class Navigation Warrants. In essence, there was not a great deal of difference; as before a 1st Class Warrant implied a satisfactory performance on the ‘Spec N’ course plus at least 500 hours of practical experience whereas a 2nd Class one recognised much the same selection of courses as had been required during the war and at least 300 hours of experience. What was significant, however, was that the civil authorities now recognised the validity of the RAF’s warrants, whereas it had never recognised its previous certificates. While this could almost be dismissed as semantics, it did have a crucial practical implication in that (once the applicant had passed the necessary supplementary civil examination papers) the holder of a 1st/2nd Class RAF Navigation Warrant was automatically entitled to a 1st/2nd Class Civil Aircraft Navigator’s Licence.53

An RAF Second Class Navigation Warrant; the original was about 8½ inches wide by 7 inches deep.

It soon became apparent that the regulations were a little too lax in that they had failed to impose any time constraint. The civil authorities were entitled to assume that anyone issued with one of their licences on the basis of an RAF warrant would be ‘up to date and in practice’. This implied that candidates should immediately apply for their warrants as soon as they had passed the relevant course(s), and/or accumulated the necessary flying experience. This requirement was spelled out in August 1946, although, to allow for those who had qualified but had yet to apply, it was not made mandatory until December.⁵⁴

In order to reflect changes in the nature and content of post-war post-graduate training, the nominated courses and flying hours required to qualify for the issue of a 1st Class Warrant were revised in 1947.⁵⁵ The same Order indicated that a similar revision of the preconditions for a 2nd Class Warrant would be published in due course. This does not appear to have been followed through, however, and it would seem that the issue of post-war warrants soon became superfluous, possibly as early as 1951 when the Order that had introduced them, and which represented the authority against which they had been issued, was cancelled.⁵⁶ What is certain is that they had lapsed by 1953, when the 3rd Edition of, now, Queen’s Regulations, appeared, this contained no reference to Navigation Warrants and lacked a para 382.

That said, despite subsequent re-brandings (as the Ministry of Transport and Civil Aviation, and then the Ministry of Aviation) the civil licensing authorities continued to grant exemptions from certain examinations until, at least, the mid-1960s, although this was now confined to applicants who had obtained a minimum of a distinguished pass on the Staff Navigation Course.⁵⁷ By the end of that decade, however, the licensing concession had effectively been rendered redundant because employment opportunities for ex-RAF navigators were rapidly dwindling and the market was increasingly oversubscribed by the older hands who already held their civil tickets. While the Service would continue to employ substantial numbers of navigators for another thirty years or so, a 1970s airliner no longer required one because his place had been taken by inertial platforms and computers which were much cheaper to maintain, far more accurate and did not answer back or harass the female cabin crew.

An example of the five ‘crowned’ single-winged flying badges introduced by the RCAF for its non-pilot air crew categories in 1943 to replace the RAF-style badges that it had used until then. (John Gowan)

Annex K. The flying badges sponsored by the RCAF and other Commonwealth air forces during WW II.

While Canadian pilots wore badges featuring an ‘RCAF’ monogram, all other Canadian air crew originally had to make do with RAF emblems, some locally produced versions of these reportedly tending to be rather larger than the official British pattern. Resentment at having to wear British badges grew to such an extent that the Canadian Air Council was obliged to spend much of 1942 pondering the problem, this process including serious consideration of the possibility of introducing twin-winged national badges for all categories. It was eventually decided to stick with single wings but to add a crown and to incorporate the letters RCAF within the lower arc of the laurel wreath. Although they were never approved, some twin-winged non-pilot badges were manufactured and some are said to have been worn in early 1943, although the authorities soon stamped them out.

The definitive single-winged Canadian badges were introduced by an Air Regulation of 26 March 1943, the details of which were promulgated via AMO A.1291/1943 of 16 December for the benefit of RCAF personnel serving under the aegis of the RAF. There were five patterns: ‘N’ for all categories of navigator, ‘B’ for air bombers, ‘E’ for flight engineers, ‘AG’ for air gunners and wireless mechanics (air gunner), and ‘WAG’ for WOp/AGs.

The latter represented something of a victory for popular opinion, as an ‘RAF-style’, ie crownless, ‘WAG’ badge had been introduced in Canada, quite unofficially, in 1942. The RCAF had rapidly outlawed these, but the point had obviously been made. It is not impossible, of course, that some RAF WOp/AGs may also have gained access to examples of the illicit Canadian ‘WAG’ badge. Even after the definitive selection of ‘crowned’ RCAF badges had been introduced, non-standard variations continued to appear, ‘NW’s and ‘NB’s for Navs(W) and Navs(B), for instance, neither of which are believed to have been formally authorised, although both appear to have been tolerated.

For the record it should be noted that the Canadians offered similar ranges of crowned badges incorporating RAF, RAAF and RNZAF monograms within the lower arc of the wreath (the latter being woven in blue for Australians). These were little used, however, and most of those that may have been issued on graduation were soon replaced by the appropriate national emblems. That is to say that British, New Zealand and Southern Rhodesian air crew wore standard RAF pattern badges, while Australians wore RAF-style emblems with the wreath woven in blue (and, uniquely, with the observer’s ‘O’ set within the wreath).

As a footnote, the reader should be aware that the authentic-looking, RAF-style ‘WAG’ badges which may be purchased today are replicas produced for the ‘collectors market’; no British ‘WAG’ badge was ever approved by the Air Council.

Annex L. The process of approving a new aircrew badge.

A long-standing protocol requires that certain innovations to do with military dress should receive the formal approval of the Sovereign, the submission being made by the Secretary of State. This includes the introduction of new badges and the War Office dutifully followed this procedure in 1912 when the design of the badge to be worn by military pilots of the newly created RFC was submitted to the Palace to receive the formal endorsement of the King.58

References to the approval of badges for all later categories of RAF aircrew have been dealt with as they occurred within the main narrative,⁵⁹ but for convenience and ease of reference, they are summarised at Figure L1. What is immediately apparent is that the ‘AE’ of 1956-57 appears to have been the last RAF flying badge to have been introduced in accordance with the long-established practice.

At the time of writing, despite extensive research among records in the public domain, enquiries with the Royal Archives at Windsor Castle and the Secretariat at Buckingham Palace, and a submission under the terms of the Freedom of Information Act (FOI) in 2011, no evidence has emerged to establish that the next badge, the ‘QM’ of 1962 (or any later badge) was ever submitted to the Palace for approval. Despite this omission, the introduction of the ‘QM’ was, at least, announced by a formal Departmental edict, as is specifically required by Queen’s Regulations.⁶⁰ That was also the case with the ‘LM’ of 1970 but thereafter, even that proviso appears to have been ignored.

This was confirmed by the FOI submission noted above which had also requested the dates and reference numbers of the Defence Council regulations associated with the (at the time) four badges (‘FC’, ‘AT’, ‘IA’ and WSO) that had been introduced since 1983. This elicited the response that ‘such documentation does not exist.’⁶¹ In other words, apart from not having been approved by the Sovereign, all of these badges were also unsupported by the publication of the necessary Defence Council Instruction (DCI). Although, somewhat confusingly, because three of these badges are worn by personnel who are not aircrew, they are included in the listing of flying badges at QR 206. The exception is the W SO badge, which is worn by aircrew, of course, but which also appears to have been introduced without formal sanction. All of which is rather sad.

Since the absence of proof is not proof of absence, the possibility must remain that the contentious ‘FC’ and ‘AT’ badges (see pages 327-329) were properly sanctioned. There is, however, persuasive anecdotal evidence to suggest that this was not the case. In 1983, when the ‘FC’ first appeared, responsibility for the award of flying badges was vested in the Air Officer Training at HQ Support Command and his feathers were somewhat ruffled when it was learned that AOC 11 Gp had been pinning an unknown variety of badge on chests at Lossiemouth. Furthermore, when this author wrote to the MOD in 1999 requesting that the specific authority for the introduction of these badges be identified, the specialist officers concerned were unable to cite a relevant DCI, or a Queen’s Order or, for the ‘AT’, even an entry in QR 206 (the ‘FC’ had been tacked on in April 1998).

As to the ‘IA’, having queried the ‘such documentation does not exist’ response and suggested a direction in which the search might be more productively pursued, the MOD was eventually able to report that the introduction of the ‘IA’ badge had been authorised by CAS in February 2003 in response to a recommendation of the Air Force Board Standing Committee in the context of the contemporary Aircrew Structures Review Paper.⁶² Unfortunately, that did not meet the essential criterion laid down in QR 206(2) which requires that the authority to wear a flying badge must be in accordance with ‘regulations prescribed from time to time by the Defence Council’ – not CAS. There was no related DCI and (unsurprisingly) there was no evidence to indicate that the badge had been submitted to the Palace.

In the light of all this, it is interesting to consider the way in which the most recent (at the time of writing) ‘flying’ badge was handled. When it was decided to introduce an emblem to be worn by personnel who, while they do not actually fly themselves, control remotely piloted air systems (RPAS) – drones – it would seem that, having neglected to follow the correct procedure since the 1960s, the RAF no longer had any idea how to go about this. Following another FOI enquiry, correspondence between this writer, HQ Air Command and the College of Arms revealed that a most innovative approach had been adopted.

It transpired that, for some reason, in 2004 (after it had been worn for 86 years) the RAF had asked the Inspector of RAF Badges, Garter King of Arms, to produce a definitive painting of the pilots badge. There may be some reason why he was not asked to produce similar images of all the other flying badges, but this question has not been pursued. When the requirement for the RPAS badge surfaced in 2012, the staffs asked the Inspector to produce a painting of it, like the one that he had produced in 2004. Garter agreed to do so. Air Command then persuaded itself that this had also constituted formal ‘approval’ and the Inspector did not demur. When this writer subsequently queried this with the College of Arms, however, the Inspector’s office clarified its position by stating that its undertaking had been confined to producing the painting, which fell some way short of ‘approving’ its introduction. When this was pointed out to Air Command, however, it was adamant that it had followed the correct procedure, ie the one that it had just made up.

Fig L1. Table indicating the decline, over time, in the degree of formal endorsement associated with the badges worn by RAF personnel who fly – and, more recently, of some who do not.

Since Air Command was content that its locally devised, and somewhat inadequate, process was all that was required to ‘approve’ the RPAS badge, there had, of course, been no question of submitting it to the Palace. Another curious aspect of this correspondence was that, as with the first FOI enquiry, the second had sought details of the Defence Council regulation authorising the introduction of the new badge. In the course of these exchanges, Air Command had been advised that the appropriate document in the past would have been an AMO or its successor, a DCI; today, therefore, one would have expected this to have been one of their current equivalents, Defence Instructions and Notifications (DIN). Nevertheless, echoing the response received in 2011, Air Command stated, again categorically, that this information was ‘not held by the Ministry of Defence.’63 Surprisingly, in view of that statement (or perhaps not), a subsequent appeal to the MOD revealed that the introduction of the RPAS badge had indeed been announced by a DIN.⁶⁴ All of which was, once again, rather sad.

Although this book is primarily concerned with aircrew of the RAF and its immediate forebears, some references have been made to the other Services where appropriate. In the specific context of badges, it is worth pointing out that the RAF is (was?) not alone in being obliged to seek the Sovereign’s approval when a new emblem is required. Although their procedures appear to be rather less formal than those adopted by the Air Ministry, in that they did not result in a ‘King’s/Queen’s Order’, it was also Admiralty and War Office practice to submit new badges to the Palace for approval.⁶⁵ Examples include the badge for commissioned FAA air observers, ⁶⁶ and the badge for army air observation post and glider pilots.⁶⁷

________________________

¹      The other members of Moore-Brabazon’s original pioneering team had been Lt C D M Campbell, F/Sgt F C V Laws and 2/AM Gorse.

²      AIR1/724/91/6/1.

³      Late in 1918 a new system of nomenclature was introduced whereby plate cameras were classified as P types, the Type LB, for instance, becoming the P7. Cameras using film became F types and gun cameras G types. Designations derived from this system, eg the G45 and the ubiquitous F24 and F95, remained in use to the end of the century.

⁴      For the rationale underpinning this practice, see Chapter 12, Note 19.

⁵      It is a little difficult to understand why the RFC had persevered for so long with its old one-way W/T because the Germans had introduced twoway wireless as standard equipment at the beginning of 1917. Nevertheless, this seems not to have impressed the British and the RAF remained an (almost) exclusively air-to-ground-only organisation until the end of the war.

⁶      AIR1/835/204/5/254. HQ RAF DRO No 732 of 24 August 1918 reproduced an Air Ministry amendment of 26 July to establishments AF/F/17, 18, 19 and 20 (among others) covering the provision of additional personnel for squadrons equipped with radio telephony.

⁷      Although the end result would be the same, it is a misnomer to refer to synchronising systems as ‘interrupter gears’. As the name implies, the CC system synchronised the firing of the gun to permit the bullet to pass through the arc of the propeller; it did not stop the gun from firing.

⁸      Viscount Tiverton was very influential in developing the practical aspects of the doctrine of strategic bombing. Adopting a rigorously scientific approach, he applied mathematical analysis and statistical method to many problems. By such means he was able, for instance, to identify the most critical sectors of the enemy’s war industries, eg the production of chemicals for use in the manufacture of explosives. By establishing, by experiment, both the ballistic characteristics of bombs and the actual, rather than the theoretical, accuracy of bombing, he was then able to determine the number of sorties which would be required to disable a particular target complex. If a serious attempt had ever been made to apply Tiverton’s calculated force levels to his target list it is arguable that the Independent Force, small as it was, just might have been able to damage vulnerable choke points within Germany’s war economy to an extent sufficient to have had some tangible impact on her capacity to continue fighting. In the event, much of the effort of Trenchard’s bomber force was dissipated in attacks against tactical targets (mostly aerodromes and railways) and the concept of strategic air warfare remained an untested theory. Like so many of the innovations which were essential to the application of air power that were developed during WW I, Tiverton’s pioneering work in the field of what would later come to be known as ‘Operational Research’, was largely forgotten and had to be reinvented a quarter of a century later.

⁹      An early, perhaps the first, serious attempt to use astro in the air had been made as early as 1913 when, flying from Eastbourne, F B Fowler had taken an experienced maritime navigator, a Mr Rainey of the Royal Mail Steam Packet Company, on a flight sufficiently far out to sea to be out of sight of land. Using a chronometer and a nautical sextant, Rainey claimed to have established his position within a (quite remarkable) quarter of a mile; see R Dallas Brett, History of British Aviation, 1908-1914, Vol II, p27 (republished by Air Research 1988).

¹⁰    Celestial, or ‘astro’, navigation depends fundamentally upon the precise measurement of the angle between the horizontal (ie the horizon at sea level), the observer and the heavenly body – this angle being known as the body’s ‘altitude’. As a rule of thumb, an error of one minute of arc in measured altitude will result in an error of one nautical mile in the calculated position. As soon as the observer is raised above the surface of the Earth, however, problems begin to multiply. Because of his elevated position, the observer’s ‘horizon’ is effectively depressed, since he can see further round the curvature of the Earth. The higher he is, the further away is the horizon and the more likely it is to become indistinct in haze and/or to be obscured by cloud (the horizon is 3 miles away for a 6-foot man standing on a beach but for an observer in an aeroplane flying at 10,000 ft it is 122 miles away). Assuming that the horizon can be seen clearly, however, the error caused by measuring altitude from this depressed ‘false’ datum can be calculated precisely, so long as the observer’s height above sea level is known accurately. Unfortunately, contemporary barometric instruments did not permit height above the surface to be measured with much precision. An alternative approach to measuring the body’s altitude ‘upwards’ from the horizon would be to measure its co-altitude ‘downwards’ from the vertical but a practical means of defining the local vertical would not emerge until the 1920s.

Some idea of what all this meant in practice can be gained from the experience of Maj G H Cooke, navigator of the R.34 on its transatlantic crossings in 1919. Only three of seventeen astronomical observations taken during the westbound flight could be made with reference to a clearly defined sea horizon; the remainder had to be estimated against a cloud horizon and were thus inherently inaccurate. Furthermore, with no information available on the surface pressure, the ship’s altimeters became highly unreliable, so it was not possible to apply a height correction with any confidence. At one stage, by innovative use of his sextant to measure actual height above sea level, Cooke calculated that the ship’s altimeters were nearly 1,000 feet in error, this assessment being validated shortly afterwards when a sea level pressure reading was obtained by W/T from a nearby surface vessel. Cooke later estimated that in mid-Atlantic he had probably known the whereabouts of his craft to within about 50 miles. This information has been extracted from R.34 by E M Maitland (1921).

¹¹    Dead (a corruption of the original ‘Deduced’) Reckoning is the method of determining position by keeping an account (or ‘reckoning’) of the distances covered on each of the courses steered since leaving a known position, the ‘Point of Departure’.

¹²    The initials RAF stood for Royal Aircraft Factory in this instance.

¹³    Like the nautical compasses from which they had been derived, early aircraft compasses were designed to detect and respond to the Earth’s magnetic field, but only in the horizontal plane. When in banked flight, however, the compass was also affected by the vertical component of the Earth’s field and, depending upon its orientation, it could behave in a disturbingly erratic fashion as it attempted to indicate the resultant of the two forces, ie to align itself with the Earth’s actual magnetic field and thus to reflect the angle of dip as well as the direction of the pole. The phenomenon was known as ‘northerly turning error’ because the symptoms were at their worst on northerly headings.

¹⁴    Not long after completing his work, Capt Keith Lucas was killed while flying a BE2c which collided with another being flown by 2/Lt G P L Jacques near Farnborough on 5 October 1916.

¹⁵    AIR1/391/15/231/32. AO 438 dated 19 May 1917 announced the creation of the Air Compass Branch of the RFC at 47 Victoria St.

¹⁶    AMWO 1051 of 18 September 1919 laid down the flying standards to be achieved by an officer, required to qualify as a pilot, in order to accept the grant of a permanent commission in the post-war air force.

¹⁷    The eleven landing grounds nominated as being available for night operations as at 20 June 1917 were:

Formation	Primary	Emergency
I Bde	Treizennes	Auchel
II Bde	Bailleul	Abeele
III Bde	La Bellevue	Izel-le-Hameau
IV Bde	Estrées-en-Chausée	Longavesnes
V Bde	Abeele	Droglandt
9th Wg	Boisdinghem	Liettres

¹⁸    While they had not been positioned on designated aerodromes, twelve of the active lighthouses were annotated as being located at sites suitable for use as emergency landing grounds by night-flying Handley Pages, FEs and (apart from two cases) Camels.

¹⁹    In fact exactly the same indication would be obtained if the aeroplane were pointing directly away from the transmitter, the appropriate direction having to be established by common sense.

²⁰    This decision highlights a problem associated with all ‘secret weapons’. Unless they can be employed to overwhelming effect, commanders are often reluctant to use them for fear of their being duplicated and used against their own forces. Similar prudent hesitation delayed the operational deployment of ‘Window’, the magnetron and a number of other devices and techniques during WW II.

²¹    AIR1/399/15/231/40. Memo A/n.4/143935 dated 10 February 1918, covering a report of a trial flight made from Cranwell on 27 January.

²²    Ibid. DAO letter BM/61/18 (0.1a) dated 15 February 1918.

²³    Ibid. DAO letter 87/9981 (0.1T) dated 14 February 1918.

²⁴    Ibid. DAO letter BM/61/18 (O.1a) dated 23 February 1918.

²⁵    AIR2/97/B.847. Lt-Col Grenfell’s memorandum was prompted by a conference held on 30 September 1918, to discuss matters relating to the Independent Force. The Chairman, Maj-Gen Trenchard, had observed that, since progress with the D/F equipment was slow, the system was unlikely to become operational before the following spring and he had recommended that training be cut back.

²⁶    AIR1/1129/204/5/2165. A note detailing overseas drafts includes reference to Capt Greenwood and a team of six tradesmen who were to leave for France on 29 September 1918 in order to ‘fit wireless to HP machines at No 3 ASD’, ie the depot at Courban serving the Independent Force,

²⁷    This would appear to have reflected a significant change in policy as it is known that officers of both the US Army and Navy had spent some two months at Cranwell in early 1918 observing British D/F trials work.

²⁸    Some O/400s were being fitted with a telephonic intercommunication system before the war ended to permit crew members in the rear cabin to speak to the pilot. Whether C9694 was one of these aeroplanes is uncertain but it would seem quite likely.

Another form of intercommunication between crew stations was a ‘Steering Indicator’ which was to be an integral element of the fully developed airborne D/F installation. It presented the pilot with a permanent indication of the heading (course) that he was required to maintain through a light display on his instrument panel which was remotely controlled from the navigator’s station. Since this device is known to have been in use as early as March 1918 it would seem more than likely that it would have been available to Towler in the following October. It was, incidentally, also intended to use the same light display to pass steering demands to the pilot when using the Gray gyro-stabilised bomb sight which, had it ever entered service, was to have been installed in the rear cabin.

²⁹    AIR2/97/B.847. The report on this flight includes a typewritten copy of the navigator’s logs and a pair of charts showing the activity undertaken during each flight (the charts are most unlikely to be those which were actually used in the air, however, as they do not appear to have had any coffee spilled on them …) It is interesting to note that the bearings logged are to the astonishing accuracy of a quarter of a degree. The preserved charts are Mercators projections on which a straight line represents a rhumb line and, since a radio bearing follows a great circle, it would have been necessary to have applied conversion angle to some of the bearings taken in order to plot these accurately. There is no indication that this was actually done, but the trials team were certainly well aware of the need for this refinement. It is suspected that the necessary calculation will have been made ‘in the margin’, the bearing actually logged being the observed reading +/- conversion angle, the magnitude of which could well have been assessed to within a fraction of a degree, hence the apparent accuracy of the result.

In reality, while the D/F system itself may have worked to the claimed tolerance of +/-3°, the overall accuracy of the system would have been considerably degraded by the errors arising from the aircraft’s compass, the accuracy of which would rarely have been better than +/-5o. Another form of error to which radio D/F systems are inherently vulnerable is ‘night effect’. The ability to transmit over extended ranges at night by ‘bouncing’ signals off the ionosphere was being exploited long before the war but, prior to the 1920s, it is doubtful whether it was appreciated that the simultaneous reception of the, out-of-phase, direct and reflected signals gave rise to anomalies in the apparent bearing of the transmission. But, whether it had been appreciated or not, unlike the complications arising from plotting on a Mercators chart, there was no easy answer to the problem of night effect, not least because it is variable in its sense, its magnitude and its duration. While they cannot be entirely eliminated, means were later devised to minimise the inaccuracies caused by night effect but these were certainly not available during WW I.

While it would have served well enough as a homing device, since airborne D/F was subject to so many errors, it is questionable whether the system really would have been as valuable to night bomber crews in 1918-19 as its advocates claimed.

³⁰    The agreed transmission sequence was as tabulated below:

³¹    AIR1/1988/204/273/128. HQ Independent Force Order No 44 of 14 November 1918. The run-down was remarkably rapid. Maj-Gen Trenchard relinquished command on 20 November, leaving Brig-Gen C L Courtney to finish tidying up. HQ Independent Force ceased to function at 2359hrs on the 25th.

³²    AIR1/1996/204/273/234. Brig-Gen E B Gordon’s letter IFG 92/1/20 dated 13 November 1918.

³³    Robert Benedict Bourdillon was to spend two years working on bomb-sights as an EO, mostly at the RFC Experimental Station (originally No 37 Sqn) at Orfordness. He would later be decorated with the MC while flying with No 27 Sqn – which just might explain the effort devoted to developing a periscopic sight for the Martinsyde G.100.

³⁴    AIR1/946/204/5/1000. In view of the technical problems which were being encountered, HQ RFC letter CRFC 1638/53 OB1 of 7 March 1918 withdrew the requirement for the installation of the Negative Lens Sight in the RE8.

³⁵    In Cross & Cockade International Journal, Vol 24, No 2 (1993) Harry Clarke contributed a very interesting and lengthy article on the wartime career of Capt G McKerrow, an Experimental Officer at Orfordness. McKerrow spent much of his time on bomb sight development work, including operational trials in the field, and Clarke’s account is recommended as further reading.

³⁶    AIR1/724/91/6/1. A survey of bomb sight development, compiled by Capt H Batsford shortly after the war, had appended to it a table (reproduced at Figure G1), which purported to show the annual production figures for various types of sight. The figures are far too neat to be totally convincing but they presumably provide a fair indication of the pattern of production, although the total for Negative Lens Sights is astonishingly high. If production really did run to 19,750 units someone must have seriously miscalculated when placing the order. It is generally accepted that the RAF had about 22,000 aeroplanes on charge when the war ended but the majority of these were single-seat fighters, trainers, flying boats, airships and sundry obsolete types, none of which would have been fitted with a lens sight – 10,000 would have been more than sufficient to equip every aeroplane that needed one and to provide a substantial reserve stock.

Unless it proves possible to verify Batsford’s statistics, however (which now seems unlikely), they would appear to remain the best available indication of wartime bomb sight production.

³⁷    The first frontal analysis of a weather system over the UK was not carried out until 1925.

³⁸    After the war Ernie Gold DSO FRS was to become a prominent figure in the Meteorological Office and an internationally acknowledged authority in his field.

³⁹    For example the second, December 1943, edition of AP1721A, Pilots Notes for the Beaufighter Mk I, provided speeds in both mph and knots for the benefit of pilots flying the Mk IF in Fighter Command and the Mk IC in Coastal Command.

⁴⁰    AVIA15/3120. Unreferenced memo by RDNav, dated 13 June 1942.

⁴¹    AMO A.365 of 12 April.

⁴²    Article entitled ‘Knots Per Hour’ in Tee Emm for June 1945.

⁴³    AP 958. KRs&ACIs First (1924) Edition, Chap IX, Sect I, para 366. Apart from being renumbered as para 382 and no longer specifying precisely who, at the Air Ministry, was to endorse certificates, these requirements were republished without further amendment in the extensively revised Second (1928) Edition of KRs&ACIs.

⁴⁴    KR 382 as published in the 1938 annual reprint of the Second (1928) Edition of KRs&ACIs which incorporated Amendment Lists 1-80.

⁴⁵    It is known that the Air Navigator’s Certificate, First Class issued to Flt Lt M M Wallenstein, who was an observer, was dated 1 August 1942. By the time that he graduated from the Specialist Navigation Course at least 142 other RAF (plus a handful of Commonwealth) officers had passed wartime courses. To this total could be added a further twenty who had attended the last two pre-war courses, ie those that had graduated since 1938 when the revised regulations governing the award of RAF certificates had been introduced (and there will have been even earlier graduates, from as far back as the early 1920s, who could probably have claimed them in arrears). Assuming that most students would have done tolerably well, and with a pool of well in excess of 180 to draw on, one could reasonably have expected the serial number of Wallenstein’s certificate to have been about No 150; it was actually only No 48.

There appear to be two possible explanations for this shortfall. First, all of the pre-1941 ‘Spec N’ students, and many of the post-1941 intake, will have been pilots. Regardless of how well they had done on the course, however, few of these men are likely to have been able to produce evidence to prove that they had flown the requisite 200 hours as navigator, rendering them ineligible to claim a certificate. Secondly, and perhaps more significantly, to obtain a First Class (or a Master’s) Certificate one had to send the necessary documentation to the Air Ministry. It is thought that this would have represented a substantial practical disincentive and it seems likely that, while some officers stationed in or near London may have taken the trouble to claim their certificates, those stationed more remotely – especially those serving overseas – may not have bothered. It is stressed that the above is entirely conjectural and is merely an attempt to make sense of the evidence represented by the Serial Number of just one (Wallenstein’s) certificate.

All of that having been said, in his Track Made Good (2006), Tim Coyle records that Air Master Navigator’s Certificate No 11, dated 10 April 1941 and duly signed by AMT (Air Mshl Garrod), was issued to a pilot, Sqn Ldr J A Cohen, while he was serving in the UK with No 10 Sqn RAAF. Cohen (who, having changed his name by deed poll in 1947, later became Gp Capt Sir Richard Kingsland) was a graduate of No 1 RAAF ‘Spec N’ Course which had been run at Point Cook between July 1938 and January 1939, and, as a flying boat pilot, he may well have been able to log 600 hours as navigator.

⁴⁶    As a temporary measure, this concession also applied to pilots attending the courses then being run on behalf of the RAF by Air Service Training and The Imperial School of Air Navigation (see page 157).

⁴⁷    AMO A.349/1938 of 15 September.

⁴⁸    AVIA2/1526. HQ Reserve Command letter RC/5/Air dated 18 July 1939 sets out the case.

⁴⁹    AMO A.822/1941 of 9 October. This Order is specifically concerned with the 2nd Class Certificate. While it makes only passing reference to the 1st Class Certificate, and does not mention the Air Master’s Certificate at all, these continued to be available under the terms of the 1938 iteration of KR 382, ie as reflected in the right hand column of Figure J1. Interestingly, despite the significant amendments to the conditions governing the award of the 2nd Class Certificate introduced in 1941, the text of KR 382 in the 1942 annual reprint KRs&ACIs was still unchanged from that published in 1938.

⁵⁰    It is not clear what this proviso was meant to imply, but it may have referred to one of the RAF’s ‘transferred schools’ in Canada, No 31 ANS at Port Albert (Ontario), which, as the lineal descendant of the pre-war SAN had continued to run wartime Specialist Navigation Courses. No 31 ANS retained this commitment until a much revised and extended syllabus was introduced at the end of 1942, at which point the ‘Spec N’ course became the exclusive responsibility of the CNS at Cranage (see page 210.

⁵¹    AMO A.891/1941 of 30 October.

⁵²    AVIA2/1668. A return of British Airways (sic) pilots dated 29 July 1942, for instance, shows a total of 171, of whom 155 held 2nd Class Navigator’s Licences, plus sixty-six seconded from the RAF and another four from the ATA. Since none of the seconded men were licensed navigators it followed that some 36% of the available pilots were of limited potential to a civil airline.

⁵³    AMO A.925/1945 of 20 September.

⁵⁴    AMO A.670/1946 of 1 August.

⁵⁵    AMO A.456/1947 of 5 June.

⁵⁶    AMO N.106/1951 of 25 January.

⁵⁷    For example, AMO A.331/1958 of 26 November, AMO A.316/1963 of 13 November and DCI S.203/1966 of 16 November all provided details of the extent of the exemption that could be granted with respect to a variety of commercial licences.

⁵⁸    AIR2/3. The memorandum submitted to the Palace by the Secretary of State for War, J E B Seely, on 24 August 1912 and bearing HM King George V’s manuscript ‘Appd. G.R.I.’, is preserved on this file.

⁵⁹    For examples of some of the correspondence associated with the introduction of the ‘AG’, ‘RO’, ‘N’, ‘B’, ‘E’, ‘S’ and ‘AE’ badges (which were variously sanctioned by King’s/Queen’s Orders 392, 439, 480, 521 and 767) see, for instance, AIR2/8369 and AIR2/18211.

⁶⁰    The 3rd (1953) edition of Queen’s Regulations states, at QR 206(2), that ‘Flying personnel are not to wear any of the badges listed in clause 1 unless authority for them to do so has been granted in accordance with regulations prescribed from time to time by the Air Council.’ At the time of writing (2013) that proviso still appears in the current edition of Queen’s Regulations, still at paragraph 206(2), but it now cites the Defence Council.

⁶¹    HQ Air Command letter FOI 171359-009 dated 18 August 2011.

⁶²    HQ Air Command letter RB115/2011 dated 29 September 2011.

⁶³    HQ Air Command letter FOI08-04-2013-164223-015 dated 7 May 2013.

⁶⁴    2012DIN01-259 of December 2012.

⁶⁵    AIR1/143/15/40/318. When the Admiralty was considering introducing a badge for graded RNAS personnel the Head of Naval Law advised on 2 December 1913 that ‘it is considered that the King’s pleasure, which is taken in all alterations of Naval uniform […] should be obtained.’

⁶⁶    ADM1/11844. An official stamp, dated 9 July 1942, on the minute sheet of Naval Law file NL 21093/41 records the approval of HM King George VI.

⁶⁷    WO32/9873. Buckingham Palace memo 54/Gen/9432, dated 16 February 1942, to the Secretary of State for War, David Margesson, conveys the approval of HM King George VI.