It will be seen that Captain E. W. Furse, when he picked up the enemy, first observed Marwitz's cavalry corps which crossed the Oise at Thourotte on the morning of the 31st. He then saw part of the German Third Corps, which, after spending the night in Roye, moved on the 31st through Lassigny, crossed the Oise at Ribecourt, and in the evening reached Attichy on the Aisne. The remainder of the Third Corps moved on Noyon and at night reached Vic on the Aisne. These movements on Noyon and Ribecourt differed in direction from the previous movements of the German left wing.

The reports supplement and confirm one another. Captain D. Le G. Pitcher, of No. 4 Squadron, had gone up with Captain A. H. L. Soames soon after 7.0 a.m., and had returned at 8.40 a.m. with the news of a column stretching from Roye to Chevincourt. This information was at once telephoned from the aerodrome at Senlis to General Headquarters. The movements of some of the other formations of the German First Army were also seen to have changed direction. Lieutenant C. G. Hosking and Lieutenant K. P. Atkinson on a B.E. of No. 4 Squadron flew over Roye and Lassigny, confirming the report of movements in that area. Then turning west they passed over various columns moving in a southerly direction until they reached the road that follows along the east bank of the river Avre from Amiens through Montdidier, and here they found part of the German Second Corps. The head of the main body was in Montdidier at 2.0 p.m., and its tail was in La Neuville. Flying south along the road they found the advanced guard of the column at Le Ployron.

All these air reports left little doubt as to the enemy's movements, and the operation orders sent out by General Headquarters from Dammartin-en-Goële at 8.50 p.m. on the 31st of August gave the information that the enemy appeared to have completed his westerly movement, and that large columns were advancing in a general southerly or south-easterly direction on Noyon-Compiègne. Sir John French directed that the retirement should be continued on the following day in a south-westerly direction.

Air reconnaissances of the 1st of September, whilst confirming the news of von Kluck's wheel in a south-easterly direction, also reported heavy columns as having reached Villers-Cotterets and Crépy-en-Valois. To withdraw the British out of reach of a night attack Sir John French decided to continue the retreat earlier than he had intended. The corps commanders were ordered to get clear by a night march. We know now from von Kluck's own statement that, perturbed at leaving the British army on his flank, he determined to make another effort to catch them up. He therefore ordered his corps to turn south to settle with the British. So on the 1st of September he was again in pursuit of the British, but the British were slipping from his grasp. There was fighting on this day, which held up the pursuit, and by the evening the German army had made an average advance of no more than ten miles.

Von Kluck persisted on the following day, but in vain. The British escaped towards the Marne. 'A chance of dealing a decisive blow', he says, 'against the British Army was now no longer to be hoped for, and it was therefore decided to move the two Corps on the left wing, the Third and Ninth, in the general direction of Château-Thierry against the flank of the French retreating from Braisne-Fismes on Château-Thierry-Dormans in front of the Second Army.'

The air reports which came in on the 3rd of September showed much of this further change of plan. Long columns were seen marching almost due east towards the Ourcq and later in the day other columns were nearing the Marne. Some had already crossed the Marne at Château-Thierry, whilst others were making for crossings west of that town. At 4.35 p.m. General Headquarters sent out the following telegram:

'Present information leads to the belief that the enemy is moving from west to east and that no immediate attack is intended. Unless the situation again changes troops will remain in their present billets. The Commander-in-Chief is most anxious that the Army should have a complete rest to-morrow. No digging or other operations except those necessary for protection will be undertaken unless special orders are issued.'

Pilots who went out soon after dawn on the morning of the 4th found a thick mist over the river Marne extending to the depth of a mile on either bank, but various columns were seen stirring out of bivouacs on the north of the river and there were other movements well to the south of the river. At 12.20 p.m. Lieutenant R. P. Mills saw movements between Bellot and Rebais and artillery in action on the high ground one mile south-east of Bellot. In the afternoon there came fuller reports of movements towards the Petit Morin. The situation as traced at Royal Flying Corps headquarters on the night of the 4th from observations made during the day is very accurate. It shows that the German Ninth Corps, which had secured the crossings at Château-Thierry on the previous evening, had progressed to near Montmirail; that the Third and Fourth Corps had got well clear of the Marne and were about and across the Petit Morin; and that the Second Corps and Marwitz's cavalry were held up at the Marne east of Meaux.

Von Kluck had marched into a bag between the Fifth French Army on the Marne and the newly formed Sixth French Army advancing to the Ourcq. Just at this time the German Supreme Command seems to have become aware of the danger threatening the German armies on the right wing. On the night of the 4th of September orders had been sent out from German First Army headquarters at La Ferté Milon, detailing the movements to be made on the following day. These movements had already begun when at 7.15 a.m. on the 5th fresh instructions arrived from the Supreme Command ordering the First and Second Armies to remain facing the eastern front of Paris; the First Army between the Oise and the Marne, occupying the Marne crossings west of Château-Thierry, and the Second Army between the Marne and the Seine, occupying the Seine crossings from Nogent to Méry. This led, says von Kluck, to 'the difficult backwards wheel' of the First Army, and to what he calls 'the important events that occurred during the second week of September'—events known to history as the battle of the Marne. Von Kluck allowed the original movements ordered for the 5th to be carried out, and, he says, 'the conclusion of this advance marked the culminating point of the operations of the First Army'. On this same day General Joffre told Sir John French that he intended to take the offensive forthwith as the conditions seemed favourable, and on the morning of the 6th this offensive opened.

The main work of the Royal Flying Corps throughout the days of the retreat was reconnaissance, and enough has been said of their reports to show that Sir John French was well served by his new arm. He had been warned before the battle of Mons, not only of the heavy movement on his front but of the enveloping attempt on his flank, and throughout the retreat he was punctually informed of von Kluck's enveloping efforts. The change of direction made on the 31st of August was immediately seen and reported. Von Kluck's renewed pursuit of the British on the two following days did not escape observation. Finally, the German swerve to the left on the 3rd of September was closely followed from the air. These are the main conclusions that come from a study of the air reports of those days. General Headquarters were perhaps at first a little shy of trusting the air reports, but they realized their value during the retreat, and paid more and more attention to them—an attention which found practical results in the operation orders issued. The Royal Flying Corps played their part in helping the British army to escape. Further, they were making themselves, and were improving in skill every day. The lessons learned on the retreat from Mons bore their full fruit at a later period, when the officers of the original squadrons held the command of those Flying Corps units which operated in the mobile campaigns of distant theatres of the war.

Their work during the retreat was done under difficulties. There were alarms at Compiègne of Uhlans seen in the vicinity of the aerodrome, and a guard was provided from the Camerons. Major B. H. Barrington-Kennett remarks on the difficulty of defending a Flying Corps camp from attack by cavalry. It would seem advisable, he says, when camped in an open aerodrome to park the aeroplanes inside a laager formed by lorries and cars. The head-lights of the cars would lighten a good field of fire, and would probably, if switched on at the approach of cavalry, cause the horses to stampede. The Royal Flying Corps, he adds, should be armed and practised with machine-guns and rifles, so that they may protect themselves without asking for an escort.

At Juilly on the 1st of September there was another alarm. The country to the north was thickly wooded, and German cavalry, which proved later to be those escaped from the affair at Néry, were reported within a few miles, with no British troops between. General Headquarters at Dammartin-en-Goële, some two miles away, hastily took their departure, and the Royal Flying Corps transport was sent off at once to Serris. But the aeroplanes could not leave, for already it was dark. The suggestion was made that the aeroplanes should fly off in the dark, but fortunately, says Major C. J. Burke, this was not attempted. The Flying Corps stood to arms to defend itself. A sunken road running east and west past the aerodrome was occupied, rifles and ammunition were served out to the mechanics, and machine-guns were set in position. After a time a troop of North Irish Horse arrived, to aid in the defence. All night watch was kept, but the German cavalry did not appear. In the morning, for the first time since the beginning of the retreat, there was no ground mist, and the machines got away at once.

The history of the retreat is made up of incidents like this. Some of the flying officers have kindly communicated their memories and impressions. 'The extraordinary part about the retreat', says Wing Commander P. B. Joubert de la Ferté, 'was the contrasts that one experienced from day to day; one night sleeping under a hedge in a thunder-storm; the next in a comfortable private house; the third in the most modern type of hotel with every luxury and convenience, the whole forming a picture the impression of which has lasted throughout the war.... One curious thing was, unless one was brought down or left behind near the firing line one never came up against the actual unpleasantnesses of war, and it was not until the advance to the Aisne started that those of us who had not been on ground duty, or unlucky, saw any signs of fighting other than from the air. What we saw during the advance confirmed our impressions from the air as to the unspeakableness of the Hun in his methods of dealing with the civilian population. I saw half a dozen villages on fire during the first day of the battle, twenty miles west of Mons, where by no possible means could there have been any armed resistance to the passage of the Huns. It was simply frightfulness on the part of the Uhlans, and what we saw later on the ground at Pezarches, Coulommiers, and La Fère was a clear indication of wilful and unnecessary destruction of private property. The sight of a draper's shop with every window smashed, every shelf emptied, and the contents thrown into the street was quite a common one.'

Major F. G. Small says, speaking of the 27th of August: 'The retreat continued to Compiègne Forest, Huns pressing our troops all the while. On returning from late reconnaissances in the dusk, it was most interesting to watch the local fighting in the roads between their vanguards and our rearguards. The spreading of fires all over the country around Compiègne Forest was a more curious sight than even the later trench offensive, the fires spreading like long flaming worms along the main road, as the Huns fired each village they went through. The northern portion of Compiègne Forest was blazing at this date.'

The speed of the retreat caused some embarrassments. On the 31st of August, while the Flying Corps occupied Senlis racecourse, two officers, belonging to Nos. 4 and 5 Squadrons, motored to Paris to get some aircraft spares, and returning in the evening found the Germans in occupation. In the dusk they were mistaken for German officers and drove their car right up to the cottages which a few hours earlier had been the headquarters of the Flying Corps. Aviation teaches quick resource; the officers managed to escape.

The pilots were not down-hearted. At Compiègne, where they were billeted in a school, Major Baring records that they were in tearing spirits. Besides their main duty of observation from the air, they rendered other occasional services. 'The usual orders on the retreat', says Wing Commander L. A. Strange, 'were dawn reconnaissances, dropping hand-grenades and petrol bombs on the enemy, and when it was impossible to notify pilots of the next aerodrome, the orders were to fly approximately twenty miles south and look out for the remainder of the machines on the ground, if machines had left the last aerodromes.'

During the retreat the dropping of bombs was still in an early experimental stage. There were some mildly successful exploits. About dusk on the 1st of September an unnamed officer of the Flying Corps, flying over the woods north of Villers-Cotterets, noticed two columns of the enemy's cavalry converging at the angle of cross-roads. He dropped two bombs, which caused confusion and a stampede. There was no bomb-dropping gear in use at this time, but small hand-grenades were carried in the pockets, and larger bombs were slung or tied about the person. The first experience of German bombs was at Compiègne on the 29th of August; while the Flying Corps were stationed there a German machine flew over the aerodrome and dropped three small bombs, which did no harm. On our side there was no time during the retreat for experiment with new devices; it was not until the Germans took up fixed positions on the Aisne that the inventive powers of the Flying Corps got to work on the devising of bombing gear, the improving of artillery observation, and the mounting of machine-guns.

The retreat also witnessed the beginnings of fighting in the air. The first German machine to be seen by the British appeared over the aerodrome at Maubeuge on the 22nd of August. There are various accounts of this. Major C. J. Burke in his diary says: 'At about 2.25 p.m. an Albatross biplane passed over the town. Major Longcroft with Captain Dawes as passenger, Lieutenant Dawes with Major Burke as passenger, on B.E.'s, gave chase. The gun machine piloted by Lieutenant Strange also went out. The machine (Albatross) had far too long a start, and got into a rain cloud.' Wing Commander L. A. Strange says: 'Chased a German Albatross machine for forty-five minutes, Lieutenant Penn-Gaskell observer, with Lewis gun. Was unable to get higher than 3,500 feet, while the Albatross was at about 5,000 feet. Observed no effect from the fire. As a result of this received orders to discard Lewis gun and mounting, and transfer the controls from rear seat to the front seat, the passenger to carry rifle in the back seat.' Major J. T. B. McCudden says: 'About the 22nd August a strange aeroplane flew over us at about 4,000 feet, and the aerodrome look-out reported it to be a German machine, the first we had seen in the War. We all turned out armed with rifles, and about six machines got ready to go up in pursuit.... All the machines which went up were loaded with hand-grenades, as the intention then was to bring a hostile aeroplane down by dropping bombs on it. The German easily got away, although it looked at one time as if Captain Longcroft would be able to intercept him on a B.E. 2 a. About half an hour after the German had departed a Henri Farman of No. 5 Squadron, fitted with a machine-gun, was still climbing steadily over the aerodrome at about 1,000 feet in a strenuous endeavour to catch the Boche.'

No. 5 Squadron from the first had been zealous in experimenting with machine-guns. Experience of fighting in the air, which began with this adventure, soon taught how enormous is the advantage, whether for attack or escape, given by superiority in height.

It was not, however, until the 25th of August that an enemy machine was brought down by a British aeroplane. Sir John French in his first dispatch, dated the 7th of September 1914, alludes to the earliest combats. His tribute must be quoted in full: 'I wish particularly to bring to your Lordships' notice the admirable work done by the Royal Flying Corps under Sir David Henderson. Their skill, energy, and perseverance have been beyond all praise. They have furnished me with the most complete and accurate information, which has been of incalculable value in the conduct of operations. Fired at constantly both by friend and foe, and not hesitating to fly in every kind of weather, they have remained undaunted throughout. Further, by actually fighting in the air, they have succeeded in destroying five of the enemy's machines.'

Unfortunately during the retreat combat reports were not made out, so that there is no account in the war diaries of the actual fighting. Some of the fights are mentioned. On the 25th of August three machines of No. 2 Squadron chased an enemy monoplane. It was forced to land; Lieutenant H. D. Harvey-Kelly and Lieutenant W. H. C. Mansfield landed near it and continued the chase on foot, but the Germans escaped into a wood. When some trophies had been taken from the machine it was burnt. Another German machine was forced to descend on the same day near Le Quesnoy, where it was captured. Aeroplanes at this time had no special armament; officers carried revolvers and sometimes a carbine; but the confidence and determination with which they attacked did the work of a machine-gun, and brought the enemy down. In one instance, a little later on, a British pilot and observer, who were destitute of ammunition, succeeded by manœuvring boldly above a German machine in bringing it to the ground and taking it captive.

On the afternoon of the 5th of September neither the German Supreme Command (which had its headquarters at Luxembourg) nor the staff of the German First Army had any idea that an offensive of the whole French army was imminent. The Supreme Command was expecting a decisive victory in the east against the Verdun-Nancy-St.-Dié defences. They believed that the German First and Second Armies could easily hold the weak French forces around Paris until this decision should be achieved, and they did not know how great a part of the French strength had been transferred from the east to the west. From the 5th to the 9th of September they issued no orders to their First and Second Armies, who were left to fight out the decisive battle of the war, without their help and almost without their knowledge, against superior forces.

General Joffre's 'Instruction' for the offensive on the 6th was brought to British General Headquarters by a French staff officer at 3 a.m. on the morning of the 5th. Unfortunately, orders to the British army to continue the retreat in accordance with General Joffre's previous instructions had already been given to the corps commanders. The Second Corps had moved off before midnight and the First and Third Corps a little later. Consequently the British army by the end of the day was some twelve to fifteen miles farther back than the French Commander-in-Chief expected, and although its subsequent advance across the Marne had a decisive effect, the hard fighting of the battle was borne by the French army on the Ourcq. During the 5th, General Maunoury, commanding the Sixth French Army on the British left, and later on General Joffre himself, visited Sir John French, and all arrangements for the morrow's offensive were discussed. Sir John French's operation orders issued at 5.15 p.m. on the 5th of September directed the army to advance eastward with a view to attacking. The preliminary movement of the British army, a wheel to the east, pivoting on its right, was to be completed by the right wing at 9 a.m. and by the left wing at 10 a.m. on the 6th.

On the early morning of the 6th Sir John French gave verbal instructions that the Royal Flying Corps were to send aeroplanes to report for reconnaissance direct to the First and Second Corps. The officer commanding No. 5 Squadron, with three machines, was to report for tactical reconnaissance direct to Sir Douglas Haig at Chaubuisson farm, one and a half miles east of Fontenay; and the officer commanding No. 3 Squadron, also with three machines, was to report direct to Sir Horace Smith-Dorrien at Combreux Château, near Tournan. With each detachment was to go a wireless aeroplane from No. 4 Squadron to keep Royal Flying Corps headquarters informed by wireless. The machines were to return to headquarters at night. This was the beginning of the decentralization of the Royal Flying Corps, whereby certain squadrons, which came to be called corps squadrons, were attached to the corps commands. The German air service from the beginning had been thus organized. With the German First Army headquarters there was one aeroplane section for long-distance strategic reconnaissance and each of the corps, with the exception of the Fourth Reserve Corps, had its own section for tactical work. From Maubeuge to the Marne the squadrons of the Royal Flying Corps had been kept together under the immediate control of General Henderson. The experiment of detaching machines to report direct to the First and Second Corps worked well on the 6th of September, and Sir John French gave orders that this arrangement was to continue.

Aeroplanes which were sent out on the morning of the 6th brought information of confused movements of the German First Army. On the British front a certain amount of movement northwards was seen in the afternoon. Of the progress of the battle on his flanks Sir John French had little knowledge. Aeroplanes were sent up to reconnoitre the position. One which flew over the area of the Sixth French Army west of the Ourcq saw at about five o'clock a good deal of movement and shells bursting in the area Étavigny-Marcilly-May-en-Multien. Another machine which flew along the line of the Fifth French Army on the British right came back with the information that at four o'clock fighting was going on south of Esternay and north of Villiers-St.-Georges. By seven o'clock that evening Sir John French had no definite news of the progress of the French armies on his wings save what was contained in these air reports, and the orders which he issued stated simply that all troops should be ready to move at any time after 8.0 a.m. on the morrow.

Early on the 7th the situation became clearer; a general retirement of the Germans on the British front was in progress. Sir John French had issued orders at 8.0 a.m. for the advance to be continued in the direction of Rebais, the army to move in echelon from the right and to attack the enemy wherever met. Aeroplanes on morning reconnaissance returned soon after the army began to move with information of early activity behind the German lines and general movement northwards. Later in the morning columns were seen moving in a north-westerly direction towards the Ourcq. These reconnaissances seemed to show that von Kluck was hurriedly withdrawing two of his corps—the Second and the Fourth—to reinforce his right wing across the Ourcq.

Early reconnaissances on the 8th told of congested movement over the bridge at La Ferté-sous-Jouarre, south of which masses of troops were awaiting their turn to cross. But the British advance was necessarily slow. The country was well suited to rearguard actions and skilful use was made of the ground by the German machine-gunners. By the evening the British had forced the passage of the Petit Morin, but they spent the night south of the Marne. Meantime, as air reports showed, von Kluck's right was heavily engaged by Maunoury's Sixth Army to the west of the Ourcq. On the night of the 8th General Joffre, taking advantage of the withdrawal of the two German corps from the British front, ordered that Maunoury's army should hold the enemy troops on the right bank of the Ourcq, whilst the British on the following day should advance across the Marne between Nogent l'Artaud and La Ferté-sous-Jouarre against the left and rear of the enemy on the Ourcq. The Marne with its steep wooded sides was well suited to rearguard actions and a stubborn resistance was expected. But air observers who came in early on the morning of the 9th brought back the news of enemy columns formed up facing in a northerly direction. Some were already on the move, and it became apparent that the enemy intended no determined, but only a delaying stand on the Marne.

Captain D. Le G. Pitcher, piloted by Lieutenant G. W. Mapplebeck, whilst reconnoitring near Château-Thierry about 12.45 p.m. saw large bodies of enemy troops in the neighbourhood of Château-Thierry and infantry moving on Domptin. This position was west of the British Third Infantry Brigade, and the whole of the First Corps was ordered to halt until the situation should be cleared up. The First Corps did not move forward again until 3.0 p.m. By the evening of the 9th the First and Second Corps were across the Marne, but the Third Corps on the left had been held up and was mostly south of the river. General Maunoury had had a hard fight, but by the late afternoon the Germans, pressed by the advance of the British across the Marne, had begun to retire in a north-easterly direction. Captain R. A. Boger, piloted by Captain R. Grey, brought this welcome news direct to the General Officer Commanding the Third Corps (Lieutenant-General W. P. Pulteney) at 5.0 p.m. He had seen long columns moving north-east from Lizy through Ocquerre on to Coulombs. This was believed to be von Kluck's Fifth Division. Other observers came in with similar information. By the evening of the 9th the retirement of the enemy was general from the Ourcq to Verdun. The battle of the Marne was won. The German armies retired, with no very great disorder, to strong positions along the heights of the northern bank of the river Aisne. Paris was saved; for the first time for over a hundred years an invading Prussian army had been turned and driven back; but the war was yet to come.

During the battle the Royal Flying Corps had been active over the enemy, and, as has been shown, reported his movements fully day by day. The machines which worked direct with the corps had supplied much useful tactical information, which was passed on direct to the corps commanders as soon as the machines landed. The observers usually reported by word of mouth, and so were able to convey a full and true impression. They reported which river-bridges were broken and which intact, and they dropped messages on to the advanced British infantry in places, warning them of danger ahead. They sometimes located for corps commanders the head of the leading troops of their corps. After a three days' stay at Melun, the headquarters of the Flying Corps moved on the 7th of September to Touquin—the first move forward since the retreat from Mons. At Pezarches, about a mile away, a field was chosen for an aerodrome. Fighting had taken place there, and small one-man trenches had to be filled in before any machine could land. On the 9th of September headquarters moved forward again to Coulommiers, and on the 12th to Fère-en-Tardenois, which place became the headquarters for the battle of the Aisne. Here the squadrons were established at Saponay, some two miles to the north-west. For many long months and years the Flying Corps was not again to be employed in a war of movement against a powerful European army, so that the work they did from the time when they arrived at Maubeuge to the time when they settled at Fère-en-Tardenois has a unique value. The French Commander-in-Chief paid tribute to their skill. His message ran: 'Please express most particularly to Marshal French my thanks for the services rendered to us every day by the English Flying Corps. The precision, exactitude, and regularity of the news brought in by them are evidence of their perfect organization and also of the perfect training of pilots and observers.'

The weather during the early part of the Marne battle had been excellent for flying. The air had been still and the heat tropical. On the 9th of September, the critical day of the battle, the weather broke, and for the next few days there were violent storms and heavy rains which greatly impeded air work of any sort. The worst of these storms occurred on the night of the 12th of September, when the squadrons had newly arrived at Saponay. Four machines of No. 5 Squadron were completely wrecked, and others damaged. Lieutenant L. A. Strange saved his Henri Farman machine, which had made a forced landing, by pushing it up against a haystack, laying a ladder over the front skids, and piling large paving-stones on the ladder, using hay twisted into ropes for tying down the machine. A diary of No. 3 Squadron records that when the machines of that squadron arrived at Saponay, about five hours before the transport, 'a terrible storm was raging, and before anything could be done to make the machines more secure the wind shifted, and about half the total number of machines were over on their backs. One Henri Farman went up about thirty feet in the air and crashed on top of another Henri Farman in a hopeless tangle. B.E.'s of No. 2 Squadron were blowing across the aerodrome, and when daylight arrived and the storm abated, the aerodrome presented a pitiful sight. The Royal Flying Corps in the field had probably not more than ten machines serviceable that morning.... Hangars were not yet issued.' The protection of machines from accidents like this became comparatively easy when the line of battle was stabilized and fixed aerodromes were made.

On Sunday, the 13th of September, the Allied armies had crossed the Aisne, but were held up by the enemy line of defence which, ran along the heights from east of Compiègne to north of Rheims. There was dogged fighting, with attacks and counter-attacks, but little or no progress was made. The Germans had regained the initiative, and the British army was forced to dig itself in along the line of battle. On the 18th of September General Joffre changed his plans and began to push forces up on the Allied left in order to envelop the German right flank. To give this movement a chance the enemy had to be held on the front, and the cavalry were called on to take their turn in the trenches—a duty which before long became very familiar to them. But the Germans extended and reinforced their line for a similar outflanking movement. These enveloping attempts did not cease until the opposing armies were ranged along a line of trenches stretching from the Swiss frontier to the coast of Belgium.

During the battle of the Aisne, from the 12th to the 15th of September, the British forced the passage of the river and captured the Aisne heights including the Chemin des Dames. Thereafter fighting degenerated into a sullen trench warfare, culminating on the 26th of September and the two following days in a series of fierce attacks by the Germans. These attacks were repulsed and were not again renewed.

On the 12th of September Lieutenant L. Dawes and Lieutenant W. R. Freeman, of No. 2 Squadron, had a notable adventure. They left in the morning to carry out an aerial reconnaissance to St.-Quentin. A little south of Anizy-le-Château, between Soissons and Laon, their machine began to rock and vibrate in the air, as if the tail were loose. They planed down at once, and landed in a small field, finishing up in a wood, where they damaged their undercarriage, wings, and airscrew. Large German columns were on the roads on both sides of them, within about two hundred yards. Taking only a biscuit and some tubes of beef extract with them, they hid in another wood close by. Some German cavalry came up to the machine, and then went all round the first wood, but found nothing, and after an hour and a half went away. The two officers lay hid until the evening, and then started in the direction of the Aisne, some eight miles distant. During the night they passed several German pickets, but the war was young, the spirit of exhilaration still prevailed in the German army, and the pickets were making so much noise that they passed unobserved. At 3.0 in the morning they reached the Aisne, where they lay down and slept. At 6.0 they were wakened by the firing of a gun close by, and realized that they were in front of the German position. German cavalry patrolled the road in front of them, and they were under heavy shell-fire from the British. They swam the Aisne, dried their clothes in a house by the canal, and then walked to the British guns, which were still in action. They were given food by the Third Cavalry Brigade, and were taken back on a supply column to rejoin their squadron after an absence of more than two days. It might be supposed that their troubles were now at an end, but they had yet to face their squadron commander, Major Burke, who sternly rebuked them for violating the order that no two pilots should fly together in the same machine.

The work of observation now entered a new phase. When armies are in fixed positions movement behind the front and along the lines of communication does not greatly vary from day to day. The Flying Corps were employed to map out the enemy's chief railheads, his aerodromes (which were surprisingly numerous), his camps, and his dumps. They began also to observe the positions of enemy batteries in order to range them for our own artillery, and they made some attempts to take photographs from the air of the enemy trenches and lines of communication.

Maubeuge had fallen on the 7th of September and, in addition to the Seventh Reserve Corps and other troops, the siege artillery which had been used to reduce Maubeuge was brought down to the Aisne, and the British guns were outranged and outnumbered. The spotting of hostile batteries became an operation of the first importance, and the Flying Corps quickly rose to its opportunities. When trench warfare began, the aeroplanes attached to corps commands took up artillery officers daily from each division over the German batteries. The positions of these batteries were noted on maps, and the maps were sent in every day to the divisional artillery commander, who allotted the targets to his batteries. When any part of the British lines was shelled, information was obtained from the air and orders were given to those of our batteries which could best reply, to concentrate on the enemy's guns. The wireless machines of No. 4 Squadron had been attached to the army corps direct during the battle of the Marne, but their opportunities had been few. On the Aisne they were first used to observe for the artillery. Two pioneers of wireless telegraphy are associated in work and in memory with these early attempts at wireless co-operation with the artillery—Lieutenants Lewis and James. Donald Swain Lewis had joined the Royal Engineers in 1904, and, after qualifying as a pilot in May 1912, had transferred to the Royal Flying Corps in December 1913. By example and precept he had done all that he could before the war to adapt wireless telegraphy to the uses of the Flying Corps and to convince others of its necessity. Before the battle of the Aisne ended he had won his victory. He was in the habit of going out alone in a B.E. machine, piloting the machine and operating the wireless at the same time. A brother-officer noted of him in a diary: 'Lewis, R.E., came in from spotting with his machine shot full of holes; I believe he likes it!' Later on in the war, at home and in the field, he continued his work. In April 1915 he was appointed to command No. 3 Squadron, in succession to Major J. M. Salmond, and did much to maintain and advance the great reputation of that pioneer among squadrons. After a spell at home during the winter of 1915-16, he returned to France in February 1916, to command the Second Wing, co-operating with the Second Army in the Ypres salient. By this time he held the rank of lieutenant-colonel, but he continued to fly over the enemy lines. On the 10th of April 1916, flying a Morane parasol, east of Wytschaete, with Captain A. W. Gale, an officer of the Trench Mortars, as passenger, he was brought down by a direct hit from the enemy's anti-aircraft guns. He had been showing Captain Gale some of the objectives on which the trench mortar fire had been directed during the week, and was killed in action while he was carrying out the duties of that artillery observation which he had done so much to perfect.

Baron Trevenen James had been a mathematical scholar and head of his House at Harrow; in 1907 he passed into Woolwich, and two years later was commissioned in the Royal Engineers. He was early interested in aviation; in June 1912, after only three days' practice, he obtained the Royal Aero Club certificate at Hendon, flying a Howard Wright biplane. In April 1913 he joined the Military Wing of the Royal Flying Corps, and was at once employed in carrying out experiments with wireless. In December 1913 he was joined by Lieutenant Lewis, and the two became famous for the theory and practice of their craft. On the outbreak of war Lieutenant James was attached to No. 4 Squadron for wireless duties; when in September 1914 the headquarters wireless telegraphy unit was formed, under the command of Major H. Musgrave, at Fère-en-Tardenois, Lieutenant James was attached to it, and shared with Lieutenant Lewis the duty of reporting by wireless over the fire of the enemy guns. Like Lieutenant Lewis, he was subsequently killed in the air. On the 13th of July 1915 his commanding officer reports: 'He was observing from the aeroplane alone, as he generally did. He was ranging a battery, and was being heavily shelled. His machine was hit by a shell, and was seen to dive to the ground from a great height. The Germans dropped a note from one of their machines saying that he was dead when he fell.... He met the end I am sure he would have wished for—if it had to be—suddenly, alone, and doing his duty.'

These two, then, Lieutenants Lewis and James, had been untiring in their enthusiasm and perseverance during the years before the war. On the Aisne their reward was granted them. 'I wish to express', says General Sir Horace Smith-Dorrien in a telegram dated the 27th of September 1914, 'my great admiration for the splendid work the Royal Flying Corps is doing for my Corps from day to day. Nothing prevents them from obtaining the required information, and they frequently return with rifle or shrapnel bullets through their aeroplane or even their clothing, without considering such, to them, ordinary incidents as worth mentioning. To-day I watched for a long time an aeroplane observing for the six-inch howitzers for the Third Division. It was, at times, smothered with hostile anti-aircraft guns, but, nothing daunted, it continued for hours through a wireless installation to observe the fire and indeed to control the Battery with most satisfactory results. I am not mentioning names, as to do so, where all are daily showing such heroic and efficient work, would be invidious.' Lieutenants Lewis and James are now beyond the voices of envy, and their names may fitly be recorded in the memory of their country.

One of the earliest of the messages sent down by wireless from the air is dated the 24th of September 1914. It is worthy of full quotation:

The later signals directing artillery fire were not so full of colour as these early messages. They consisted of code letters. The clock code for signalling the results of artillery fire was first used in 1915 and afterwards generally throughout the war. The target was taken as the centre of a clock and imaginary lines were circumscribed around it at distances of 10, 25, 50, 100, 200, 300, 400, and 500 yards. These lines were lettered Y, Z, A, B, C, D, E, F, respectively. Twelve o'clock was always taken as true north from the target and the remaining hours accordingly. An observer noted the fall of the rounds with reference to the imaginary circles and clock-hours and signalled the result, for instance, as Y 4, or C 6. A direct hit was O.K, and there were other signals. Messages from the battery or any other ground station were signalled to the observer in the aeroplane by means of white strips which were laid out on the ground to form the letters of a code.

During the battle of the Aisne, the wireless machines were few in number and other methods of signalling were mostly in use. On the 15th of September Captain L. E. O. Charlton fired Very lights over enemy guns previously observed. On the 24th of September 'Lieutenant Allen and two others with aeroplanes indicated targets and observed fire, communication being by flash signals'. Sometimes the pilots returned and landed to report on gun positions. But when once the gunners had profited by the superior accuracy and speed of report by wireless, they were hungry for more machines. On the 23rd of September the commander of the Second Corps telegraphed to General Headquarters: 'I hope that you will be able to spare the wireless aeroplane and receiving set to Third Division again to-morrow. The results were so good yesterday that it seems a pity not to keep it with the Division, which has got accustomed to its uses and is in a position to benefit even more largely by the experience gained.' The answer was that the machine had been damaged by anti-aircraft fire, but would be ready again shortly. A wireless aeroplane was as popular as an opera-singer, and the headquarters wireless section soon developed into No. 9 Squadron of the Royal Flying Corps. The attitude of the gunners may be well seen in an entry made in the war diary of No. 3 Siege Battery, dated the 23rd of January 1915—'Airman' (Captain Cherry) 'reported for co-operation (lamp only, alas!).'

The photography was a mere beginning. On the 15th of September Lieutenant G. F. Pretyman took five photographs of the enemy positions; these were developed later on the ground, and were the forerunners of that immense photographic map of the western front in thousands of sections, constantly renewed and corrected, which played so great a part in the later stages of the war. Some other experiments had no later history. Steel darts called 'flêchettes', about five inches long and three-eighths of an inch in diameter, were dropped over enemy horse-lines and troops by No. 3 Squadron. A canister holding about 250 of these darts was fixed under the fuselage; by the pulling of a wire the bottom of the tin was opened and the darts were released. To do any harm these darts had to score a direct hit on some living object, so that a whole canister of them was probably a less formidable weapon than a bomb. Even on a battle-field life is sparsely distributed on the ground.

There was hardly any fighting in the air during the battle of the Aisne, and reconnaissance machines were not attacked by other aeroplanes. They were fired at from the ground by anti-aircraft artillery. The anti-aircraft guns got their name of 'Archies' from a light-hearted British pilot, who when he was fired at in the air quoted a popular music-hall refrain—'Archibald, certainly not!' The Germans used kite balloons for observation. In the attempt to drop a bomb on one of these Lieutenant G. W. Mapplebeck was attacked, on the 22nd of September, by a German Albatross, and was wounded in the leg. He was the first of our pilots to be wounded in the air from an enemy aeroplane—a long list it was to be.

The Royal Flying Corps were few indeed in comparison with the air forces opposed to them, but they were full of zeal and initiative. On the 19th of September they received a valued compliment from the French General Staff, who asked the British Commander-in-Chief to permit them to carry out reconnaissances along the front of the Fifth French Army. This was already being done, but Sir David Henderson promised to take measures to make the reconnaissance more complete.

In the battle of the Aisne the British forces were co-operating with General Maunoury's Sixth French Army on their left. The so-called race for the sea was, in fact, a race for the flank of the opposing army. On the 20th of September De Castelnau's army formed up on the left of Maunoury and at first made some progress, but was pushed back by the reinforced army of General von Bülow, and was held on a line extending from Ribecourt on the Oise to Albert. On the 30th of September General Maud'huy's army came into position on the left of De Castelnau, along a line extending from Albert to Lens, while at the same time cavalry and territorials occupied Lille and Douai on the German right. This army in its turn was opposed by the German Sixth Army sent up from Metz, which pushed the French behind Arras, occupied Lens and Douai, and began to shell Lille. General Maud'huy could do no more than fight to hold his ground till another army should come to his relief on his left. For this purpose the British army was shifted from the Aisne to its natural position in defence of the Channel ports, and came into action along a line extending northwards from La Bassée. The actual line was fixed by a series of fierce engagements culminating in the battles of Ypres, 1914.

The Allied plan was to hold the French and Belgian coast and to take the offensive in the north. With this purpose in view the Seventh Division of the British army and the Third Cavalry Division, both of which came under the command of Sir Henry Rawlinson, were disembarked, from the 6th of October onward, at Zeebrugge and Ostend. But Antwerp was taken by the Germans on the 9th of October, and the first business of this famous force was to cover the Belgian retreat along the coast. The German Fourth Army was being rapidly pushed forward into Belgium; Lille capitulated on the 13th of October; Zeebrugge and Ostend were occupied by the Germans on the 15th. Still the idea of a counter-offensive was not abandoned, and the works and defences of Zeebrugge were left intact in the hope of its speedy recapture. On the night of the 1st of October the British army had begun to move northwards from the Aisne. By the 9th of October the British Second Corps had detrained at Abbeville and received orders to march on Béthune; on the 12th the Third Corps began detraining and concentrating at St.-Omer and Hazebrouck, and subsequently moved up to Bailleul and Armentières. A week later, on the 19th, the First Corps under Sir Douglas Haig detrained at Hazebrouck and moved on Ypres. General Headquarters left Fère-en-Tardenois on the 8th of October and after a five-days' stay at Abbeville established themselves at St.-Omer.

The Royal Flying Corps had moved north with the British Expeditionary Force, from Fère-en-Tardenois by way of Abbeville, to St.-Omer, where they were established by the 12th of October. No. 2 Squadron remained behind for a few days, to carry on with Sir Douglas Haig's corps on the Aisne, but joined up at St.-Omer on the 17th of October. In addition to the four original squadrons, No. 6 Squadron, newly arrived from England under Major J. H. W. Becke, came under Brigadier-General Henderson's orders on the 16th of October. This squadron had been stationed at South Farnborough as a reserve for the squadrons in the field. When General Rawlinson's force was sent to Ostend, to attempt the relief of Antwerp, Lord Kitchener said, 'I want a squadron to go with it'. He ordered that No. 6 Squadron should be ready in forty-eight hours. The squadron was hastily completed; some pilots and machines were obtained from the Central Flying School; some machines were bought from private firms; equipment, tools, and the like were collected at night; and on the 7th of October the squadron flew to Bruges and began at once to carry out reconnaissances. On the following day they flew to Ostend, and, their transport having arrived, were concentrated on the racecourse. Five days later they retired to Dunkirk, and by the 16th of October were established at Poperinghe, where they came under the orders of headquarters at St.-Omer.

A good deal of reconnaissance was carried on by the squadrons during the northward move of the army. On the 29th of September unusual and heavy movement in a northerly and north-westerly direction had been observed behind the enemy lines on the Aisne. On the 1st of October air reconnaissances showed that the trenches in front of the British First Army Corps were unoccupied or very lightly held, and during the next few days there were many indications that one or two German army corps were being withdrawn to the north. Meantime the enemy took more trouble than usual to interfere with our aircraft, and employed an increased number of anti-aircraft guns. In the north our strategic reconnaissances were not so successful, and the formidable enemy movement against the Ypres line developed undetected. Not many aeroplanes were available at this time for the wider sort of strategic reconnaissance. Nos. 2, 3, and 5 Squadrons had been attached, by an order issued on the 1st of October, to the First, Second, and Third Army Corps respectively, while No. 4 Squadron was detailed for strategical reconnaissance. The General Officers Commanding army corps had learned the value of aeroplanes and demanded their assistance. Much of the country over which they were operating in Northern France and Flanders was flat and enclosed, unsuitable either way for cavalry reconnaissance.

Long-distance work was done chiefly from headquarters. On the 3rd of October, when the situation at Antwerp had become critical, Lieutenant-Colonel F. H. Sykes flew direct to Bruges from Fère-en-Tardenois, with a message from Sir John French to the Belgian Chief of Staff at Antwerp. On the following day he returned and reported that the Germans had broken through the south-eastern sector of the outer defences of Antwerp, that the Belgians were awaiting help, and that they might possibly hold out for two or three weeks. In forwarding the report to Lord Kitchener Sir John French adds, 'The relief of Antwerp I regard as my first objective'. This mission was followed by others, and a few days later Sir John French speaks of reports which he is receiving daily by air from General Rawlinson.

Meantime a squadron of the Royal Naval Air Service, as shall be told in the next chapter, had been operating for some weeks from Ostend and Dunkirk with French territorial forces. The French territorials were hastily embodied troops taken from civilian life and were not of much use for a fight against odds. When the Seventh Division was landed at Ostend and Zeebrugge during the first week of October, and the improvised British Naval Division arrived at Antwerp, the situation was already out of hand. The British army was small; it had helped to save Paris, and now paid the price in the loss of the Belgian coast. The Seventh Division occupied Ghent, and after covering the retreat of the Belgian army, which halted along the line of the Yser, from Dixmude to Nieuport, fell back by way of Roulers to a position east of Ypres. When the whole British force came into line, it held a front of some thirty-five miles, with Maud'huy's Tenth French Army on its right across the Béthune-Lille road. On its left the line was held, from a point north of Ypres to the sea, by the Belgian army, assisted by four French cavalry divisions under General De Mitry.

The German army had failed to take Paris; all its efforts were now concentrated on the seizure of the Channel ports, and its pressure on the defending line was like the pressure of a great rising head of waters against the gates of a lock. The glory of the defence belongs to the infantry. The men who flew above them could only watch them and help them with eyes. The infantry were often unconscious of this help; they disliked seeing hostile observers above them and often fired on aeroplanes with very little distinction made between friends and foes. On the 26th of October Major G. H. Raleigh, of No. 4 Squadron, reports an artillery reconnaissance as follows: 'Hosking and Crean did a tactical reconnaissance early, but were unable to locate batteries owing to clouds. They went up later and did it. The clouds were low, so it was arranged that they should fly over one of our batteries to observe for ranging. The machine came down in flames and was completely demolished. Pilot and passenger had both been wounded by our own infantry fire when at a height of about a thousand feet with the large Union Jack plainly visible.'

Wing Commander W. D. Beatty tells how, before this time, the disadvantage of the Union Jack marking on the planes was becoming evident. The officer in command of an aviation camp at Paris had pointed out to him that, at a height, only the red cross of the Union Jack was clearly visible, and that it was mistaken by the French for the German marking. A suggestion was made that the British should adopt the French circular marking. The mishap of the 26th of October hastened the adoption of this suggestion, and thereafter the French target was painted on British aeroplanes, with the alteration only of blue for red and red for blue, to preserve national distinctions.

Commanding officers sometimes complained that our machines were little in evidence. The aeroplane observers, operating over enemy territory, reported to their own command, and their reports, forwarded to the proper quarters, took effect in the orders issued by Headquarters, so that crucial improvements were sometimes made in our dispositions, by information obtained from the air, though the infantry had seen no machine in the air above them. The use of machines for more local needs, such as artillery ranging, hastened the recognition of the services rendered by the Flying Corps, and brought it into closer touch with the other arms. Photographic cameras and fittings were still very imperfect, and photography from the air was not much practised, but sketch-maps of enemy trenches and gun-pits, as located by air reconnaissances, were issued by Headquarters during the battle of Ypres. Good work was done in directing the fire of the artillery, and the few wireless machines were much in demand. A telegram sent on the 28th of October from Sir Horace Smith-Dorrien to the Royal Flying Corps headquarters runs: 'Can you send us a second machine, with wireless installation, for use to-morrow? The aeroplane now working without wireless with Fifth Division has more to do than it can accomplish owing to observation being required for French artillery as well as our own.' But the wireless machine was required by the First Corps, at the northern end of the line, and a machine without wireless was sent instead.

The deadly and effective method of directing artillery fire on hostile batteries by means of wireless telegraphy played a great part in winning the war, but for the first battle of Ypres the wireless machines were not ready in quantity. The penalty which had to be paid for this unreadiness was heavy. Precious shells, which were all too few, had to be expended for ranging purposes. On the 4th of November Lord Kitchener wired to Sir John French: 'I have been talking to David Henderson about giving more observation to artillery by aeroplanes. As this saves the ranging ammunition, which is worth anything to us, please insist upon it.' Failing wireless, other methods of ranging had to be employed. These methods had been set forth in an official paper issued on the 28th of October. The aeroplane flies at any convenient height and when it is exactly above the target it fires a Very light. The battery range-finders, who have been following its course, then take its range and another observer with the battery takes its angle of elevation. These two observations are sufficient to determine the horizontal distance between the battery and the target. It was sometimes found difficult to take the range of an aeroplane, at a given moment, with an ordinary range-finder, and an alternative method of ranging is suggested. By this method the aeroplane flies at a prearranged height, and, as before, fires a light exactly over the target. But this method also is liable to error, for an aeroplane determines its height by the use of a barometer, and barometers are only approximately accurate for this purpose. So it was suggested that the two methods should be combined: the aeroplane should endeavour to fly at a fixed height, and the range-finders should, if possible, also make their calculations. These methods cannot attain to the accuracy of wireless, but they were found in practice to give fairly good results. They were not quickly or generally adopted; many battery commanders continued to prefer the reports of their trained ground observers to the indications given from the air. When wireless machines were increased in number, artillery observation from the air came into its own. In a report dated the 5th of February 1915, Brigadier-General Stokes, commander of the 27th Divisional Artillery, lays stress on the enormous advantages of wireless. He says that the 116th Heavy Battery of the Royal Garrison Artillery, which had at its disposal an aeroplane equipped with a lamp, had succeeded in registering only three targets in fifteen days, whereas the 130th Howitzer Battery, which had a share in the services of a wireless aeroplane, had registered eight targets in seven days. The disadvantages of the older and cruder method are many; a thin mist which does not prevent the aeroplane from observing the target is enough to prevent signalling to the battery; the lamp is difficult to use on a rough day, and difficult to read against the sun; the aeroplane has to be kept under continual observation by the battery. To get better value out of our artillery, the general concludes, the wireless service must be largely increased.

Reconnaissance from the air was much impeded, during the second half of October, by low clouds and bad weather, but enough was observed to give some forecast of the tremendous attack that was impending. The Germans outnumbered the British three or four times, and threw their whole weight, now against one part, and now against another, of the thin line of infantry fighting in mud and water. Those who would judge the battle will find no escape from the dilemma; either the British defence, maintained for thirty-four days, from the 19th of October to the 21st of November, against an army which esteemed itself the best army in the world, must be given a high and honourable place among the great military achievements of history, or the German army was disgraced by its defeat. But the German army was a good army, and was not disgraced. The Germans themselves respected their enemy, on the ground and in the air. On the 21st of November, at the close of the battle of Ypres, two German second lieutenants of the air corps, called Fribenius and Hahn, were taken prisoner near Neuve-Chapelle, and were examined. They said that the performances of British aeroplanes had caused instructions to be issued that a British aeroplane was to be attacked whenever encountered. British aeroplanes, they said, were easily distinguishable from others, for they always showed fight at once. What prisoners say under examination is not evidence, but this early tribute to the fighting quality of the Royal Flying Corps is repeated in many later testimonies.

The crisis of the battle of Ypres came on the 31st of October, when the line of the First Division was broken and the left flank of the Seventh Division exposed, at Gheluvelt, some six miles east of Ypres. The counter-attack by the First Guards Brigade and the famous bayonet charge of the Second Worcestershire Regiment retook Gheluvelt, and re-established the line. The last act of the long agony came on the 11th of November, when a great attack was delivered all along the line. The place of honour on the Ypres-Menin road was given to two brigades of the Prussian Guard Corps, who had been brought up from Arras for the purpose. The First Division of the British army met this attack at its heaviest point of impact, and by the close of the day the Prussians had gained five hundred yards of ground at the cost of enormous losses. The story of the battle belongs to military history; the loss and profit account can be summarized in two facts. The First Brigade, which met the Prussian spearhead, was taken back into reserve on the following day. It had gone into the battle four thousand five hundred strong; on the 12th of November there remained, of the First Scots Guards, one officer and sixty-nine men; of the Black Watch, one officer and a hundred and nine men; of the Cameron Highlanders, three officers and a hundred and forty men; of the First Coldstream Guards, no officers and a hundred and fifty men. This is not a list of the surrendered remnant of an army: it is a list of some of the victors of Ypres. The other fact is no less significant; after a week of fighting the German attack fainted and died, and when the next great assault upon the Ypres salient was delivered, in April 1915, it was led not by the Prussian Guard but by clouds of poison-gas.

No extraordinary or signal services were rendered by the Flying Corps during the crises of the battle. The weather was bad, and on some days flying was impossible. Yet by every flight knowledge was increased. When the British troops arrived in Flanders and were sent at once into the battle, the country in front of them was unknown. The dispositions of the enemy forces were not even guessed at. Then by the aid of the Flying Corps the enemy's batteries were mapped out, his trench lines observed and noted, his railheads and his roads watched for signs of movement. The reports received just before the battle do not, it is true, indicate the whole volume of movement that was coming towards the Ypres area. The newly raised reserve corps which formed part of the German Fourth Army, the transport of which to the western front began on the 10th of October, were not definitely seen from the air until just before the battle. But observers' reports did indicate that many troops were moving on the Ypres front, and once battle was joined enemy movements were fully reported on.

When at the end of October the Belgian army mortgaged great tracts of their ground for many years by opening the canal sluices and letting in the sea, the Germans were enabled to divert the Third Reserve Corps southwards. The movements of troops from this area were observed by the Royal Flying Corps, and General Headquarters on the 1st of November issued this summary: 'The coast road from Ostend to Nieuport was reported clear this morning, and there are indications generally of a transference of troops from the north of Dixmude southwards.' Again, when the attack on Ypres had failed and died away, the Germans transferred many troops from the western to the eastern front; these movements also were seen by the Royal Flying Corps, who reported on the 20th of November an abnormal amount of rolling stock at various stations behind the German front. 'The rolling stock formerly parked on the Ostend-Thourout and Ostend-Roulers lines has evidently been broken up', says General Headquarters Intelligence Summary for the 20th of November, 'and distributed to a number of stations along the Lys and in the area immediately north of it, which would be suitable points of entrainment for the forces in that district.... This redistribution of the rolling stock, together with the apparent reduction in motor transport, would seem to point to some important movement away from this immediate theatre being in contemplation.' Air reports for the following day proved that much movement eastwards had already taken place.

Throughout the battle tactical reconnaissances had been maintained to a depth of from fifteen to twenty miles behind the German lines. There were some few fights in the air, and a little bombing, but observation was still the principal duty of the Royal Flying Corps. They were greatly privileged; at a time when our people at home knew nothing of what the army was doing, they, and they alone, witnessed the battle of Ypres.

They would gladly have done more. Many of them had been infantry officers, and were eager to lend a hand to the infantry in that heroic struggle, but they lacked the means. Not until the summer of 1916 were they able, by organized attacks from the air, to help to determine the fortunes of a battle.

With the close of the battle there came a lull in the fighting. This lull continued throughout the dark and damp of the first winter, and the interest of the war in the air shifts to the preparations which were being pressed forward at home for renewing the war during 1915 on a larger scale and with better material.

One incident which occurred just after the battle of Ypres shall here be narrated; it serves to illustrate how the air work of the Germans may sometimes have been impeded by a certain defect of sympathy in the German officer class. German two-seater machines were commonly piloted by non-commissioned officers, who took their orders from the officer in the observer's seat. On the 22nd of November Lieutenants L. A. Strange and F. G. Small, of No. 5 Squadron, were returning from a reconnaissance, flying at a height of about seven thousand feet. Their machine, an Avro, with an 80 horse-power Gnome engine, carried a Lewis gun, which had been mounted by them, against orders, on rope tackle of their own devising, just above the observer's seat. In the air they met a new German Albatross with a 100 horse-power Mercedes engine. They showed fight at once. Diving from a height of five hundred feet above the German machine, and at right angles to its line of flight, they turned underneath it and flew along with it, a little in front and less than a hundred feet below. From this position, which they maintained while both machines made two complete turns in the air, they were able to empty two drums of ammunition into the German machine. After the second drum the German pilot lost his nerve, and the machine side-slipped away and down, landing behind our lines, close to Neuve-Église. There were twenty bullet-holes in the German machine, but the pilot and observer were both uninjured. The British officers landed close by, to claim their prisoners. The German observer, a commissioned officer, took little notice of them; as soon as his machine landed he jumped out of it, and dragging the partner of his dangers and triumphs out of the pilot's seat, knocked him down, and began to kick him heavily about the body. If ever a collection of incidents shall be made, under the title 'How the War was Lost and Won', to illustrate the causes of things, this little drama will deserve a place in it.