The Signal Service in the European War of 1914 to 1918 – Chapter 12

«--	The Signal Service in the European War of 1914 to 1918 Chapter XII	--»
Chapter XI		Chapter XIII

Chapter XII.

FORWARD SIGNALS IN THE 1917 POSITION BATTLES

SYNOPSIS.

General Remarks. — Horses of Cable Sections a Problem. — Universality of Training Aimed at. — Motor Cyclists. — Visual in the Division. — Wireless in the Division. — R.A. Signals. — Brigade Signals. — Forward Signals in the Attack. — Pushing Forward the Buried Cable. — General Policy of Forward Signals Compared with S.S. 148. — Forward Lines. — Forward Visual. — Message-carrying Agencies. — Decentralization of the Pigeon Service. — Formation and Organization of the Messenger Dog Service. — Runners. — Forward Wireless. — Loop Sets. — Continuous Wave Wireless. — The History of the Power Buzzer and Amplifier.

In the previous chapter a picture has been given of the activities of Army and Corps Signal personnel in the 1917 battles and it has been shown how the responsibility of both D.D. Signals, Army and A.D. Signals, Corps had moved forward. This forward extension of supervision had also been accompanied by the use of Army and Corps personnel, in the shape of the area detachments and the heavy artillery signal units, well within the Divisional areas. The tendency was for the buried system, although serving units right up to the front line, to be gradually taken out of the hands of the Divisional signal officer who was relieved more and more of responsibility both for its administration and for the upkeep of the routes. This policy was never entirely achieved, however. To the close of position warfare, the Divisional Signal Company personnel remained responsible in great measure for the planning, and almost entirely for the execution, of the forward buries, in maintaining which the same unit also provided considerable help. During the whole duration of position warfare, one great drawback to the Divisional Signal Company had been a lack of modification to suit stationary conditions. Great changes in establishment had taken place; yet further changes, involving considerable additions in strength, were to take place in 1918. Nevertheless, the Divisional Signal Company was still primarily constituted and equipped for mobile warfare. In no respect did the peculiar consistency of policy of the higher signal command show to better advantage under most insistent pressure than in this refusal to reorganize the Divisional Signal Company's constitution. Almost every Divisional Signal Officer came to consider the horses of his cable sections as little less than a curse at times. Two-thirds of the companies at the least did not use their cable wagons for any purpose other than training until the retreat of March, 1918. One or two had had a reminder of their usefulness in March, 1917, during the German retreat. Half a dozen or so were to have a similar demonstration in November, 1917, at Cambrai. One here and there laid a line Iwo or three miles long by cable wagon in the deeper advances of the position battles of 1917.

In the meantime, all cable between Division and Brigade — the extent of the responsibilities of No. 1 section of a Divisional Signal Company — was either buried, laid by hand, or at most from a horse-drawn drum barrow. When advances were sufficiently deep and shelling i:uthcieDtly light to permit of the survival of a six-horsed cable wagon in the forward area, our artillery preparation and the enemy defences had in the vast majority of cases made the terrain impassable for such a vehicle. Roads destroyed beyond immediate repair defied wheeled transport; dense and broad belts of wire only partially uprooted barred the way, or the destruction of bridges over unfordable streams effectually put a limit to their activity.

And all this while the sappers of the section, without whose aid it was almost impossible to keep horses and vehicles clean and equipment in good order, were required for work on the position warfare line system. Cable section operators were employed in the advanced and rear Divisional signal offices or attached to Brigades to work the forward end of the Divisional circuits. Field linemen (the "cable hands" of the old nomenclature) and general duty pioneers were wanted to lay buried cable or to help maintain the system when laid. To crown all, there was no general officer of Division who did not know all about transport and horses, while there were many who were not versed in the technical details which usually distinguished an efficient from an inefficient signal company. To such commanders, the only way of judging the discipline and efficiency of the signal company, which was giving at least average satisfaction in its technical duties, was by frequent inspections of the company transport and personnel. It is not too much to say that, in the manifold activities of trench warfare, these recurring inspections were a principal source of trial to the Divisional Signal Company commander. With the resources of his company strained to the utmost to cope with a signal system almost beyond his power, he was constrained to employ more men upon his horses than he could conscientiously spare. Apart altogether from the inspections — which were very necessary if the horses were to receive their due share of attention — in fairness to the animals themselves, a considerable proportion of the company must devote most of their time to their welfare. In some divisions the representations of the O.C. Signals carried sufficient weight to enable him to draw upon certain infantry personnel to help out his drivers; in others, with the ingenuity which distinguished many of these sorely-taxed officers, he started classes in horsemastership, part of the virtue of which consisted in the pupils attaining a thorough proficiency in grooming. In most cases, however, it was left to the officers and men to struggle as best they could to keep the company efficient. This aim was attained in the majority of cases, but the struggle was a very-hard one and was one of the main reasons why the personnel of the Divisional Signal Company was never "in rest" in the true sense of the phrase. Whether in the line or not, horses had to be cared for, signal offices must be manned, and lines must be maintained. In addition, out of the line, after the first two or three days' clearing up, the whole of the surplus energy of all ranks was devoted to framing for mobile warfare.

In view of the circumstances set forth above, it is not surprising that many officers sent in from time to time strong recommendations as to the desirability of a considerable reduction in the percentage of horses in the company. These proposals were very seriously considered, particularly in 1917 and 1918, when the general man-power situation was causing anxiety. The decision very wisely reached, however, was that, though the percentage of men to horses might be improved by the addition of extra dismounted personnel, no great revision involving loss oi horses could be permitted. Officers' second chargers disappeared; a slight reduction was made in spare horses, etc. No considerable reduction was, however, permitted and the Divisional Signal Company entered upon the mobile warfare of 1918 unchanged in any of the essential features of its establishment as they affected the mobility of the unit.

At the date of the present review the company had been substantially increased from the pre-war establishment. It was, however, none too large for the work it had to do. The standard organization included a rear signal office and an advanced one at battle headquarters. All the operators available in headquarters and No. 1 section were required to man the normal shifts in these two offices and to provide two sounder operators for each Brigade. The Divisional end of the Corps circuit was manned by operators attached from the Corps, or the office personnel could not have been made to go round. Headquarter linemen were usually pooled with the linemen of No. 1 section and utilized turn and turn about upon the headquarter circuits, on the maintenance of the buries between Division and Brigades, or on night cable-laying parties.

While linemen and operators were fully employed, the D.R.'s and the artificers — instrument repairer, harness maker, wheeler, etc. — were also without much spare time upon their hands. The motor cyclist despatch rider's work had to a great extent lost the romance which made it one of the most interesting aspects of the Signal Service during the 1914 campaigns. The danger of his work had markedly decreased, though the new German practice of putting down shell storms on the main roads behind our front was once more adding a considerable element of excitement to his routine journeys. He had, however, a great amount of hard work to get through each day. The normal D.R.L.S. kept the various reliefs continually at work, while the Divisional Staff had never overcome their liking for special D.R.'s. Time and again an unfortunate motor cyclist would be called from his well-earned rest to take a single packet which through thoughtlessness or forgetfulness had missed the regular post. Occasions were even recorded when special D.R.'s were utilized to take forward a commander's private mail. The practice of despatching "specials" on insufficient grounds, indeed, became so common and had such unfortunate effects on the working of the service that special legislation was required to deal with it.

Other portions of the headquarters of the Divisional Signal Company, as constituted in 1917, were the "Signallers and Despatch Riders" and Wireless Section. The former were still so few in number that for any position battle in which cyclist or mounted despatch riders were required, detachments of Corps cyclists or Divisional mounted troops had to be attached temporarily to the signal company. It was also — quite apart from the difficulty of the same men being regarded as despatch riders — usually impossible for more than one Divisional visual station to be manned by the personnel of the Company Headquarters. Unless circumstances were particularly favourable, this meant that an alternative route by visual between Division and Brigade was impossible without outside assistance. In a measure this difficulty was overcome by the appropriation of some portion of the Brigade pool of battalion signallers for this purpose, and in some Divisions the signal section of the pioneer battalion was utilized to eke out the very meagre visual establishment. By means of these men, a system of visual stations could be set up between Divisional and Brigade headquarters in any but the most enclosed country. Where the country was flat and very enclosed, however, it was not usually worth while to attempt an elaborate system. The buried cable routes were usually to be relied upon, and, if alternative communication was required, the wireless section was looked to to provide it.

The wireless section was destined to come into its own during the Cambrai offensive and the 1918 battles. It consisted, in its earliest form, of personnel for three trench wireless sets and for two power buzzer and amplifier sets. It was specifically laid down in the establishment that a proportion of the operators were available for work as office telegraphists. This was a regulation which was destined to be much insisted upon. One of the main obstacles to the development of wireless within the Division was the fact that the best wireless operators were seldom available for duty in that capacity. This trouble was not overcome until the appointment of a wireless officer gave to wireless within the Division an authorized champion.

At the present time — the campaign season of 1917 — wireless was used principally as an alternative for the buried cable system between Divisional and Battalion headquarters. In such a capacity it was often of great momentary value, especially in the Flanders area. A special use to which the trench sets were sometimes put was observation with the artillery. Here, however, they were at once superseded by the new continuous wave sets which had peculiar qualifications for this work.

In a bird's eye view of the Signal Service at this period of the war, it is necessary to imagine the personnel of the Divisional wireless section either at work in the signal office as telegraphists or, more legitimately, manning a trench set at or near Divisional headquarters -and one, or possibly two, others at convenient spots well forward along the main Divisional lines of communication, probably handy to Brigade headquarters. In addition, in a Division where wireless was being fully worked, two or more power buzzer amplifier stations might be in use, either between a particularly exposed battalion and its Brigade headquarters or between isolated posts and the forward Divisional wireless station. Forward of these, again, would be power buzzers installed at company headquarters or at other important strong-points and worked by battalion signallers.

The remaining personnel of a Divisional Signal Company was contained in a R.A. headquarters section, the two R.A. signal sub-sections, and the three Brigade sections (Nos. 2, 3, and 4 sections of the company).

The artillery signal system of a Division was usually run to all intents and purposes entirely separate from the infantry command system. Possibly the artillery exchange might be a different board in the same office, far more often the artillery ran a separate signal office closely adjacent to their headquarters which was itself often at a considerable distance from Divisional headquarters. The system was the special care of a captain of the Divisional Signal Company and circuits ran from the artillery signal office to the various artillery Brigade headquarters which had their own smaller exchanges. From here, again, lines radiated to batteries, to the Brigade wagon lines and the Brigade Staff, to the affiliated infantry Brigade, and to the neighbouring artillery Brigades. Usually, also, subsidiary circuits linked the ordinary Brigade or individual batteries to the local kite balloon and to other means of observation common to more than one formation. The subaltern with the R.A. signal sub-section was responsible for all the signal communication between Brigade headquarters, the Brigade O.P. or O.P. exchange, and the batteries; forward of the batteries the signal system of each was controlled by an officer of the battery and the work carried out by R.A. personnel.

A general supervision was, however, exercised by the artillery Brigade signal officer, who was held responsible for the technical efficiency of the whole system. An artillery sub-section officer was more important to his formation Staff than even the infantry Brigade signal officer to his. Artillery without good observation is comparatively inefficient. In modern warfare good observation is robbed of most of its value without equally good intercommunication. Fleeting targets could not be engaged at all without both being efficient. Good instances of this were frequently afforded in the autumn of 1918 by the firing of the German heavy artillery during their retreat. All firing for days together was done at random by the map. Shells, patently intended for a local objective such as a bridge across a canal or an important road junction, frequently fell with monotonous regularity two or three hundred yards from the intended target. Entire absence of observation was shown by a complete failure to correct such an error. Poor signal communication was shown when an error which obviously was visible from German occupied territory was rectified only after a considerable lapse of time.

No signal units suffered a greater proportion of casualties in 1917 and 1918 than the newly-formed artillery signal sub-sections. This was especially so with gas wounds. The German artillery systematically shelled the British battery areas at all times. Battery personnel, generally, could take shelter during the periods of heavy shelling. Signal offices must, however, be kept manned and linesmen were continually out in pairs upon the lines. It was seldom that many weeks passed without casualties occurring in such units as were in the line, and the total mounted up far in excess of that of any other class of R.E. signal personnel.

Casualties in the Divisional Signal Company, generally, were undoubtedly minimised by the general use of heavily-fortified signal offices and the concentration of the forward routes into deep buries. The latter were dug and the cables buried under cover of darkness. Breaks were not common and maintenance was therefore reduced to a minimum. After the gunner sub-sections, most casualties occurred in the Brigade sections. Indeed, during the actual offensives the casualties of these sections must have exceeded in number those suffered by the artillery signal units.

Brigade signals in position warfare consisted mainly in the manning of the Brigade signal office, the giving of assistance with the forward portions of the Divisional lines, and maintenance of trench cable and alternative wireless, power buzzer, and visual, to battalions and flank Brigades. The Brigade section is one of the few signal units which remained practically unchanged throughout the war. Its only rival, indeed, in this respect, was the cable section. Both changed to a certain extent as regards the qualifications of the men composing the units, but neither underwent radical changes in organization. Both the Brigade and the battalion signal section shared in the prevalent shifting forward of technical qualifications towards the front line. The battalion signaller of 1914 knew visual only — both Morse and semaphore — but nothing else. Telephones, if they existed at all, were unofficial. The height of position warfare in 1917 and 1918 saw battalion signallers biases as regards lines and the common "D3" buzzer; versed in the mysteries of the fullerphone; and familiar with the highly-technical power buzzer and loop wireless set. In addition, he was required to know how to handle pigeons, how to fire message-carrying rockets and how to use small buzzer exchanges. In a word, he was expected to have more qualifications than the old "telegraphists, field line" and "telegraphists, permanent line" of the pre-war classification had possessed.

This needs to be borne in mind when considering the evolution of forward signals, for it was one of the chief difficulties encountered in replacing casualties amongst battalion signallers.

It was impossible for the Brigade section, unchanged in numbers as it was, to keep efficient a normal Brigade signal system in the position battles. This was legislated for by the formation of the Brigade pool of battalion signallers already referred to. These men were specially trained to enable them to work the more technical alternative methods of forward signalling and were controlled by the Brigade signal officer. This fact helped very much to make possible the decisive step which characterized the forward signals of this year — that is, the concentration of signal activities along one main route down the centre of each Brigade area.

On a "peaceful" front in normal times the brigade signal office would be connected by sounder or fullerphone and by telephone with Division; by telephone and by fullerphone with each battalion and with lateral Brigades. As far as possible, lines would be accommodated in the buried system. Where this was not possible, trench cable would be used. Forward ot battalion headquarters, an emergency line system would be laid and connected up to "D3" telephones or fullerphones at company headquarters or the more important posts. Normally, this system would be tested through but not used except for purposes of artillery liaison or observation, or in emergency. The line system would be completed by laterals between company headquarters and to the battalions and companies to left and right. It would be reinforced by as complete a system of visual stations as was possible. In addition would be power buzzer and amplifier stations at most Brigade and some battalion headquarters and power buzzers at important company headquarters or at posts particularly likely to be cut off by shell fire or a sudden attack. At battalion headquarters and in each company area would be pigeon posts; normal routine traffic which had no particular urgency and yet could not be passed over the telephone was sent by runner. Buzzer was strictly prohibited forward of Brigade headquarters.

This normal Divisional intercommunication system was, of course, modified considerably during offensives. In three ways, in particular, distinct advances from the policy and practice of the previous year can be seen. These three were: —

(1) The pushing forward of the buried system across "No Man's Land."

(2) The adoption of the single Brigade line of signal communications and the formation of Brigade forward parties.

(3) The success of the power buzzer and amplifier and the initial use of loop wireless sets. (Under this heading, also, might be considered the first development of forward continuous wave wireless for use with the artillery) .

In the position battles of 1916, no attempt was made as an ordered policy to push forward the buried cable immediately after zero. Two considerations militated against the adoption of such a policy, the first being the shortage of labour, the second, the casualties which would certainly be inflicted on the working parties by the enemy's barrage. In 1917 both considerations were over-ruled aid in some cases the buried cable system was pushed forward well into the captured zone within a few hours of the commencement of the action. The best examples occurred in the advances opposite Arras and in the battle for the Vimy Ridge. Here, circumstances were particularly favourable, and on one occasion a short bury of 60 yards only had to be dug to connect up the British mined cable head with the extensive tunnels on the German side of the line. In such circumstances, the best results might be expected, but even without the use of such adventitious tunnels good work was done in pushing forward normal buries. One Corps reported the completion of 1,500 yards within 15 hours of zero, and a further 1,000 yards by zero plus 31 hours.The whole system was installed and connected up and working by the time quoted and gave good results. In other cases, results were not so good and the opinion of Divisional signal officers was by no means generally in favour of persevering with the new policy. It would seem that here, as in most other aspects of forward signalling, regard must be had to the particular circumstances of each individual case. The deciding factors in 1917 were the position in the Corps as regards the supply of labour, the weight of the enemy barrage on the country to be crossed, the depth and speed of the advance, and the presence or absence of aids on the German side of the line. Under any but the most favourable conditions, only the most carefully worked-out schemes could be expected to achieve a reasonable measure of success.

The weight of the enemy's shelling varied very much in different actions. If the attack was pressed home well and the enemy forced to withdraw his guns, his shelling was so reduced that for some time a ground line system could be maintained with ease. The interval between the advance and the normal forward extension of the buried cables was thus bridged with a sufficiently good temporary system. If, on the other hand, the enemy barrage was extraordinarily heavy and widely spread, buried cable was out of the question on account of the casualties involved. It was necessary to wait until our own guns had engaged the enemy batteries or the latter had shifted to individual targets. It was then often possible to pick a good way forward, along which the cable could be extended without undue loss. In the meantime, alternative methods must be resorted to and in such situations pigeons, visual, and wireless carried the essential traffic.

The depth of the advance was also a powerful determining factor. It was useless attempting to keep up with a deep advance by means of buried cable. Signal communication was required at once and must be extended forward by the most rapid and not by the slowest existing means. Fortunately, such deep advances implied relatively little enemy retaliation. In such a case it was therefore possible to follow up the advancing troops with poled cable and airline routes. Thus, for brief periods, comic airline and poled or treed cable crept up once more into the Divisional areas. Once the overground system was completed and working, a portion of the outdoor signal personnel was told off to maintain it, while the remainder, with the assistance of what working parties could be spared, worked steadily on the extension of the buried system.

The general policy of forward signals advocated in S.S. 148 was a great advance on anything that had previously been attempted. It was based upon the principle that on each Brigade front all efforts should be concentrated upon the establishment and maintenance of one main cable route reinforced by all available means of communication. This route was to run from the advanced Brigade headquarters to the Brigade forward station, that is, from the headquarters of the Brigadier before zero to that at which his office would be established if the attack was successful. Forward of the latter, smaller arteries would run to the various forward observation officers and to the battalion forward command posts. By means of this one main route all artillery and infantry traffic was to be passed. Should the advance be of such a depth that movement of batteries and consequent re-registration were involved, one forward observation officer from each group would be permitted to lay one line back to cable head to reduce the congestion on the artillery lines. All circuits were to be metallic and no lines were to be laid without the express permission of the infantry Brigade signal officer.

The scheme had been matured with care and was based upon the experience of the previous fighting. In theory, it looked to be perfect,, but in practice several defects were found to prevent its general adoption without considerable modification. The grouping of all signal responsibility along one line, and the implied obligation to Staffs to select their new command stations along that line, was a novel feature which was destined to revolutionize forward signals in position as well as in mobile warfare. This was still further insisted upon in the successor to this first intercommunication manual and was the more easily fallen in with by the Staffs because it was from the first patent to all concerned that it was the most economical and most efficient method yet adopted. No longer did the signal officer fritter away his energies on quite unnecessary lateral spurs between the main route and the various headquarters he served. His duty was complete with the completion of the backbone of the communication system, though he often assisted the signal personnel of junior formations to lay lines from their command posts to his forward report centre. If Staffs could not comply with orders and settle down along the immediate line of advance, their signal communications would be more precarious according as their distance from the line increased. In so far as this principle was concerned, no fault could be found with the new policy.

In its details, however, there was much to cavil at. The decision to allow none but metallic circuits was a mistake. It was an ideal quite impossible of realization. In 1915, 1916 and 1917, one great stand-by of the forward signal officer was his laddered lines. These were earthed lines in the fullest sense of the word. Thirty, forty, fifty breaks had occurred without destroying the circuits. Often they were the only means of line communication left. The same troubles were experienced in laying and maintaining the forward line system which have already been referred to in describing the 1916 battles. At Ypres and Passchendaele, especially, lines of any description were often impossible to maintain. Behind the newly captured defences on the Passchendaele Ridge, everything had been shot away with the exception of one main road which an army of men laboured to preserve. What lines were left by the shells were not likely to survive the attentions of the labourers. In such circumstances, metallic circuits were impossible and the attempt to keep "them through would have been a farce.

Another feature in which the scheme as laid down in S.S. 148 failed, was the inadequate provision for artillery traffic. Nearly all the important traffic in a position battle had reference in one way or another to artillery. Normally, it was upon this arm that the General Officer commanding the formation relied for the series of situation reports which enabled him to form his mind-pictures of the progress of the action.

One pair only was provided for the artillery on a Brigade front, and stringent regulations debarred the forward observation officers from laying other Lines on their own initiative. The system broke down hopelessly in this respect and after the first attempt was modified in one of two ways. In some cases, the artillery ignored the Brigade forward centre and each forward observation officer laid lines of his own to buried cable head and endeavoured to maintain them as best he could. In others, an extra forward station was improvised and the two systems ran parallel to each other down the length of the Brigade area. Of the two, the latter method was, of course, the better organized and therefore the most successful. When the two systems were connected by lateral junctions in front and in rear, a strong line system resulted which was very efficient and amply sufficient for the needs of the infantry Brigades and their allied artillery formations. In yet other cases, the need was met more simply by the addition of a third line or pair to the main route. By this means and by the allowance of a certain amount of latitude to unfavourably situated artillery F.O.O's, a close approximation to the original plan was worked with success.

As regards the detail of the Brigade routes, it was soon realized that the fullest latitude must be allowed to the discretion of individual commanders. No situation resembled any other in its entirety. A scheme devised for the crater-pitted, mud-strewn flats of Flanders was quite unsuited to the hilly and wooded country further south and vice versa. A system devised for a hurried advance over a comparatively broad zone, must be decisively and quickly modified if the attack failed and the intended lines were telescoped to one-half or one-third the length for which the original organization was intended. This was, indeed, one of the serious faults of the new system. There was a tendency to crowd so many means along one route and to allow for so many relay posts in case of an unexpected depth of advance, that resources were severely strained to find the necessary personnel. Signal officers became seriously concerned as to the advisability of persisting with so many alternative methods of signalling. Towards the end of the year there was a disposition to rely upon lines and visual or on fines and wireless only, instead of dividing up the available men between all three means.

The Brigade signal section of one officer and 24 N.C.O.'s and men could not man the proposed forward route. The difficulty was overcome by calling upon the battalions for visual signallers and runners, and upon the Division and the Corps for wireless and power buzzer operators and linemen.

Later on, with further training, battalion signallers were also to operate the forward wireless sets, but this stage had not yet come within the realm of practical politics.

The main feature of the Brigade signal system during the position battles of 1917 was thus the single chain of stations somewhere along the middle of the area of operations. Along this line were cables — metallic if possible, but usually earthed — runner relay posts, pigeon. aeroplane, and visual stations, wireless, and power buzzer-amplifier stations. The number of stations varied, of course — within the limits possible to the Signal personnel available — according as the advance was shallow or deep.Protection was attempted, and posts were usually selected in the best available shelter. In the south they were frequently sited in dug-outs, connecting possibly with above ground machine-gun emplacements. In the north, use was made of some of the "pill-boxes" which were the characteristic feature of the German defence system in the low-lying swampy country where deep dug-outs were impossible. The occupation of such sites agreed very well with the general scheme of the advance, for they were usually favourably situated with regard to possible command posts.

It was in such a central combined system that the universality of training aimed at by the Signal Service was of superlative use. In many cases, when part of the complement of a forward station or relay post had been disabled or killed, the station would have been rendered useless but for the presence of men who could make shift to use signalling apparatus for which they were not specially trained. Many an orderly rendered useful service in repairing lines, man\' a lineman took his turn at the key and proved a serviceable substitute for a wounded operator. The fact that forward parties were on some occasions involved in the fighting has already been emphasized. .S. 148 laid down the principle that "The Brigade forward party must move forward soon enough to escape the hostile barrage, but must avoid becoming involved in the fighting. Usually it should follow the last wave of the attacking infantry." Here, again, individual discretion was necessary. Mopping-up was oftentimes a lengthy operation and it was not always possible to proceed according to schedule. Many signal officers decided to hold back the forward party until the first objective had been reached.

The composition of the forward parties also varied greatly according to the situation. As a general rule, the principles laid down held good and signal commanders soon learnt how best to vary them to suit particular sets of circumstances. For instance, if the attack were not pressed home, the collection of the whole of the forward party on a shortened line of communication could only lead to congestion and unnecessary casualties. Arrangements must be made for the speedy evacuation of a proportion of the men to positions further to the rear where they could be of more use and in less danger. Every advance, and almost every formation taking part in an advance, presented its own special features. The experience of each battle was collated and utilized as a means of further improving the general policy. S.S. 191 was the crystallization of the experience of 1917, just as S.S. 148, on which the policy of 1917 was based, was the outcome of the earlier experiences of 1916.

There remains the consideration of the individual means of forward signalling: lines, visual, power buzzer and amplifier, loop wireless sets, pigeons and runners, etc. In some we shall see little change from the methods already described in the previous year, while others became of much greater importance in 1917. All were of considerable value except, perhaps, the loop wireless sets, which were new and in the use of which few signallers had yet been trained.

Forward Lines remained unchanged to any great extent during 1917. They were laid on a more systematic plan, the main object of the Brigade forward parties being the establishment of two metallic (?) circuits between the Brigade forward station and cable head. Here, once more, is seen the rather academic nature of the first intercommunication manual. It was definitely laid down that the circuits should be metallic. It was also definitely ordered that detachment "A" of the forward party should proceed forward to the appointed spot, establish a signal office, and then commence to lay back two circuits to meet detachment "B" which was laying forward from cable head. Both of these instructions were impracticable. The inadvisability of the use of metallic circuits has already been referred to; the folly of attempting to lay back towards cable head was amply demonstrated on many occasions. In the event, the policy followed was to lay earthed lines forward, either to the site selected or, if this was not recognisable or not accessible, to a suitable spot as near as possible to the site ordered in operation orders. The spot selected was then marked conspicuously with a blue and white pennant or other mark. If distances were long, the lines were looped in at fairly frequent intervals to other friendly dug-outs or shelter points of one description or another. Here the runner relay posts were set up and these also were conspicuously marked.

These earthed lines were replaced by metallic circuits as soon as shelling permitted, but this was not always possible. It depended primarily on the weight of our own artillery and the depth of our advance, and these two factors varied, of course, with every attack. Good examples of this variation were afforded in the third Battle of Ypres. After the success of the first attack, and the capture of the ridges, the British artillery for a time dominated the position and poured such a devastating hail of fire upon the enemy's lines that his retaliation was comparatively ineffective. Work became possible by day for the first time in this sector and a good cable system was completed forward of buried cable head. This was the more fortunate as it was some time before it proved possible to advance the buried cable head into the captured territory. This favourable position lasted for two or three days, when the attack was completely held up by the malign results of the heavy rain which characterized the beginning of August in this sector. The shell-torn ground became quite impassable. The enemy guns from their new positions on comparatively untouched ground once more began to assert a local superiority. They were reinforced daily, as the enemy realized the gravity of the threat to his defences. The maintenance of ground lines became almost impossible. Lines were blown to bits almost as soon as they were laid. Even laddered cable could not survive the storm of shell. To add to the troubles of the signal units the ground was flooded by the choked and overflowing streams. The shell-holes which pitted the greater part of the country were filled with foul gas-saturated water from 2 to 15 ft. deep. The pillboxes in which the signal offices and relay posts had been set up had not escaped the attention of our own artillery. They were now subjected to an equally fierce bombardment by the enemy and many of them were shot to pieces, overset, or crumbling under the repeated impact of the heavy shells. It was in these circumstances that methods of signalling which did not involve Hues or human passage scored most freely. The power buzzer with its short earth leads, the wireless set with its inconspicuous and easily-erected aerial, the pigeon with its free flight through untainted air, all played a significant part in bridging the gap between the different headquarters whose cement roofs stood out above the rain-soaked, mud-drenched country.

In the battles of 1917, as in those of the two previous years, there was a constant struggle for mastery between "air" and "ground" signalling. In normal circumstances in stationary or slow-moving warfare, "ground" signalling was quickest and most reliable. When conditions became abnormal or movement rapid, the situation was often reversed and "air" signalling established a momentary superiority. In fact, just as in Arctic exploration, the difference between "winter" and "summer" sledging can be best tested by whether a man is most happy pulling hard in the traces of his sledge or lying at ease in his sleeping bag, so in position warfare, the normality or abnormality of the conditions could often be accurately gauged by the relative efficiency of these two classes of signal methods. "Air" signalling of all descriptions, whether pigeons, wireless, or visual were employed, involved a loss of time and a lack of reliability which was only compensated by a gain in invulnerability and a decrease in the loss of life involved.

To return to the general consideration of the forward lines of the 1917 offensives, the characteristics peculiar to these battles were more in the way of special organization to improve maintenance of lines than in methods of laying. On some occasions hues were pushed forward through underground saps until the enemy's barrage was passed. Breakdown parties were often utilized right forward along the Divisional routes, the men making their headquarters at or near a Brigade signal office. At other times, when buries were impossible to any great extent, and the situation was fairly stable, special efforts were made to obtain small working parties and to bury short lengths of the routes. Instances of such local buries were those laid over the Pilkem Ridges during the Ypres battle — one on each Brigade front. The work was most unpleasant. It had to be done in a great hurry by night and under heavy shelling and the casualties in the working parties were heavy.

Laddered lines were, however, still the normal expression of signal ingenuity, when shelling was heavy. The use of three parallel cables, 15 yards apart and laddered at intervals of 50 yards, was common and' such lines were kept through in spite of as many as 75 breaks in two or three thousand yards.

Another innovation adopted by some Divisions was the appointment of an officer to control all forward lines. The breakdown parties and the maintenance personnel acted under his orders and careful organization often resulted in the fairly continuous maintenance of lines right up to the artillery observation posts in front. The amount of traffic dealt with on such an improvised system during one of the later offensives is shown by the figures of a Division in the Passchendaele fighting. On zero day, besides continuous telephone calls, 1,010 urgent operation messages passed over the Divisional lines. Imagine a complete interruption of communication for 24 hours, and it can be seen that a break-down of the Signal Service would have made all the difference between success and failure. Such a break-down, however, never occurred. Always alternative methods worked to a greater or less extent.

Methods of laying lines were much as in the Somme offensive. It was seldom that the country to be traversed was clear enough to permit of cable wagons being used even in the deepest of the advances. Most lines were laid by hand or at best from a light cable cart or from a horse-drawn or man-drawn barrow. Man pack cable-laying apparatus had been designed and issued and was extremely useful. Brigade sections which had had their pack horses withdrawn did not greatly miss these animals now, though in the later battles at Cambrai and during the March retreat their loss was to be severely felt.

When lines were continually broken, as often happened forward of cable head, other means were employed. Other means must,, indeed, always be available from Divisional headquarters forward and especially on a Brigade forward signal route. The least technical of the means employed were runners, pigeons, and messenger dogs; slightly more technical were visual, aeroplane signalling, Very lights and S.O.S. rockets. Most technical of all, and therefore the source of most trouble to the unspecialized forward personnel, were the wireless and earth induction sets.

Of these three groups, the first and second presented few special features of interest this year except as regards their organization along the main forward route.

Signalling to aeroplanes did not occupy nearly as prominent place as in the Somme offensive. The methods employed were again the ground sheet, the ground panel and the aeroplane dropping ground. The panel was cumbersome and slow and still unpopular; the latter was the most popular method and the most efficient. By this means valuable information was brought back by the aeroplane observers and a series of precis of the situation delivered regularly to the General Staff at Division and Corps. Signalling to kite balloons and aeroplanes by visual was again tried without success. Visual, generally, was much employed, but all other methods paled to insignificance beside the efficiency of the Lucas lamp. Flags were little used, discs were used only by very forward units, and by these less than in 1916, other types of lamps than the Lucas were practically obsolescent, and were superseded as fast as the Lucas lamps came to hand. Very lights and rockets were laid down as part of the equipment of the normal Brigade forward party and were specially useful for the purpose of notifying sudden emergencies and for showing up the position of the front line and of particular posts. They were used also, as in the previous year, for acknowledging the receipt of D.D.D.D. visual messages.

One special feature of the use of visual was the greater prevalence of two-way working. This was made possible by the smothering of enemy observation stations by the British artillery fire and by the depth of the advances. Another innovation was the use of the pill-boxes in the Ypres district as the sites of visual stations. These little fortresses had deep embrasures and doors which still further decreased the dispersion of the rays of the Lucas lamps. Two-way working in the very face of the enemy was possible in many cases by judicious use of suitable posts.

Pigeons, messenger dogs and runners are a separate group of methods of signalling all involving the element of life. In all three cases, but of course much more in the last, sentiment must enter to some extent into the question of their use. In the case of the two former, it took the shape of a tendency to make pets of what should strictly have been treated as mechanical means of conveying messages. In the case of the third, the human element introduced a variety of complications, both good and bad.

The use of pigeons for forward signalling had undergone Uttle change in the last year. The service itself had undergone a most vital change through its decentralization. All forward signal officers recognized this as the dawn of a new era of usefulness to the pigeon.

No longer would all arrangements have to be made with a remote and inaccessible officer at G.H.Q. whose duties necessarily involved many and repeated journeys to all parts of the French and British fronts. His responsibilities were largely delegated to pigeon officers at the Army and the Corps, and, in many cases. Divisions now had their own lofts allotted to them by the Corps. The roundabout nature of the itinerary of pigeon messages was thus decreased and the "times" improved in like measure. General supervision was still exercised by the O.C. Carrier Pigeon Service at G.H.Q., who acted as adviser to the D.D. Signals of the Armies and the A.D. Signals of the Corps in all matters appertaining to the organization of their own pigeon services.

The value of pigeons had been so demonstrated during stationary warfare that the service had reached very important figures. In one Corps alone, in the Second Army offensive of June, 1917, 532 birds were issued on a single day and 92 operation messages received back. The approximate number of pigeons in use during the offensive on the Somme in 1917 was 12,000 birds.

Times were also much improved, averaging 25 minutes for a message from the front line to the formation headquarters for which it was intended. The speed of the pigeon remained constant, but delivery was hastened by constant attention to the minutest points of organization of the forward service and particularly of the means of despatching the message from the loft to headquarters.

Casualties to pigeons were comparatively high during the battles and in the German offensive on the Belgian coast a mobile loft was hit by a shell and destroyed. At the battle of Messines, also, quite ten per cent, of the birds in action were killed, over 50 being destroyed by a shell which made a direct hit upon a distributing station. On the other hand, their value was demonstrated again and again and the demand was still in excess of the supply. Several hundred messages were passed by this means in every battle, and a reflection of the efficiency and utility of the service was seen in July, 1917, when a further increase of establishment to 120 horse-drawn lofts and six motor mobile lofts was authorized.

Novel features of the employment of the birds in this year were their use by artillery officers for observation purposes, and by the crews of tanks. An unusual incident of training occurred when, in preparation for a coast offensive which never materialized, motor despatch boats were allotted for training a proportion of the birds out to sea.

The pigeon service at this period of the war had far outstripped the forward wireless service in its practical utility . No better example of the mutability of the fortunes of the various branches of the Signal Service can, however, be seen, than the reversal in the importance of these two methods of signalling which took place in the following year.

The use of messenger dogs in the British Army is an interesting instance of the evolution of an entirely new method of forward signalling to meet a particular type of warfare. In September, 1917, definite information was received from German prisoners of the utilization of these animals for message-carrying purposes, and it was rather earlier than this — in the Messines offensive — that the first experiments with "liaison dogs," as they were then called, was carried out by an artillery commander. These initial ventures were so successful that the project was taken up officially and a War Dog School was formed at Shoeburyness under an officer who had long made a hobby and business of training dogs for war purposes and who was an acknowledged expert on this subject. At first the employment of the dogs was permitted to depend upon the will of the individual units who expressed a desire to have them as an extra means of communication, but this haphazard organization could not long survive. The relations between the human and the canine species, though very gratifying from the point of view of civilization, were too amicable for the best results to be obtained without great insistence upon dog discipline. The tendency to make pets of the animals, and the lack of appreciation both of their capabilities and of their limitations, much decreased the value of the dogs as reliable means of liaison. Orders were therefore issued in November, 1917, for the centralization of the messenger dog service under the O.C. Carrier Pigeon Service as the senior representative of the most nearly allied branch of the Signal Service. The messenger dogs already with units in the field were withdrawn to a central kennel at Etaples, re-sorted and re-trained. Sectional kennels were then formed at certain Corps headquarters and the messenger dog service re-started on a regular basis.

Reports as to the utility of the dogs varied considerably in tenor during the next few months; but they were sufficiently successful to permit of the retention of the service. The animals were speedy — averaging in one Division a mile in seven minutes — but their eyes were badly affected by gas and they were reported as somewhat unreliable under heavy shelling. On the other hand, they frequently did valuable service in situations where runners would have been exposed to great risk. They had one or two advantages over pigeons; they could be used at night, and their training to a new area took one week only as opposed to the three or four weeks necessary to habituate pigeons to a new district. Training was carried out both at home and abroad. At home, the dogs were given a thorough training at the Shoeburyness War Dog School. Here they were accustomed to the noise and smoke of gunfire, and certain conscientious objectors were sorted out and consigned to a handy lethal chamber. On arrival in France, however, it was found— as was often the case, too, with hastily trained Signal Service personnel — that they had forgotten most of what they knew. The Central kennel at Etaples was therefore surrounded with bomb pits and the explosion of the bombs carefully synchronized with the feeding of the dogs, in order to associate the noise with the greatest of their daily pleasures. Thorough training in message-carrying was also carried out at both places, and the new service was in full working order by the spring of 1918. Its usefulness was, however, seriously curtailed by the general aspect of the war in this year. The exploitation of the messenger dog to its full extent is reserved for the intercommunication service in future wars, if position warfare should recur before the present signal prac'ice has been altered out of all recognition.

Runners, again, as in 1916, were the last resource of the forward commander and his signal officer. Their use was limited by the need for the economy of personnel, by the heavy casualties amongst their ranks, and by the slowness of this method of conveying information. Forward of battalion headquarters, however, runners — and possibly visual — were the only reliable means of signalling in anything approaching moving warfare. In the very forward areas, runners held their own throughout, and they often formed a valuable adjunct to the Brigade forward route. Whenever feasible, other means were employed and the signal officer who was reduced to the use of runners only, was compelled to confess to himself that for the time being he was a failure. His main preoccupation in such a situation was to organize some alternative means aid take as many of his runners as possible off the route as quickly as he could substitute them by some other means of conveying the information.

When defining the principal advances in signal practice during the battles of 1917, prominence was given to the success which attended the use of the power buzzer and amplifier stations and to two fresh developments of army wireless — the initial attempts to use special "forward" sets, and the use of continuous wave wireless with the artillery. Even yet, forward wireless had not gained its rightful place as an essential means of communication, though in many situations the wireless stations had dealt with traffic of vital importance. Month after month, propaganda work had been zealously carried on, and the education of the Signal Service in the use and management of forward wireless sets had been steadily pursued. The decentralization policy, coming as it did at a critical period in the history of signals, much improved the chances of wireless making good. It occurred too late to have its full effect upon the position battles of this year, but was destined to act as a potent influence for the popularisation of this form of signalling. Its complete effect was seen in 1918 when, at the time of the March German offensive, wireless in the Divisions was at last in a position to utilize to the full every opportunity given to it. Then, and only then, was seen the wisdom of the policy persisted in in spite of every obstacle.

Meanwhile, in 1917, two principal developments marked the progress of wireless telegraphy. The first was the invention of special "forward" sets to supersede the trench set in work for which the latter was not fitted. The British field wireless set (trench set. spark, 50 watts) was eminently fitted for its normal work as a station for work at Brigade headquarters or Divisional headquarters. In stationary warfare, when time and labour were available, it had often been used with good effect in special situations well forward of either. It could not, however, be called a "mobile" set in the strict sense of the word. It required a party of at least three men to carry it forward, and the accumulators used with it were bulky, easily damaged, and required frequent renewal.

The set could not be used with any likelihood of great success as a subsidiary part of the Brigade forward route in a sustained offensive. In addition, its aerial was fairly conspicuous and Limited its use in exposed positions near an ill-defined front line. A conspicuous wireless set was a special mark for enemy attention and as such was anything but popular with headquarter staffs. There was a distinct element of nervousness in their attitude towards it and this was strengthened by the spread of rumours as to the extreme accuracy with which the enemy could locate such a set by means of directional apparatus. Something much less obvious both to the enemy and to our own Staff was required. This need was met by the invention of the "loop" wireless set.

The essential advantages of the new sets were their lightness and compactness, inconspicuous aerials, and short wave lengths. They were designed to work on one or two low fixed wave lengths (65 and 80 metres) far below those in use by the Royal Flying Corps, the trench sets, or the direction finding sets. They were made in four distinct parts, a front transmitter and receiver and a rear transmitter and receiver. Special canvas "carriers" were issued with the sets and much improved their portability. The front sets were intended for erection well forward — possibly even within direct view of the enemy — and had "ground" aerials for the receiver and a small brass loop fixed to a bayonet for the transmitting aerial. This set could be erected and worked entirely under shelter, except for the projecting ends of the ground aerials which were invisible at a distance of a few yards. The rear set, on the other hand, had a low horizontal aerial supported on tripods. This was less inconspicuous, but much more efficient than the forward aerials. The general range of the sets was from 2,000 to 3,oot> yards so that they were well adapted for use between the main stations along the Brigade forward route and also between battalion forward command posts and the Brigade forward station. Three considerations reduced the practical value of the sets. The first was a certain "individuality" about the earlier sets which was the cause of a rather disconcerting uncertainty as to their efficiency. The second was an unduly great dependence upon favourable circumstances and surroundings for their success. The third and deciding factor was a shortage of supply which prevented the universal training of forward signallers in the use of the sets until the opportunity for their employment was almost past. Even in 1917, individual signal units — notably those of the Canadian and Australian Corps, often more hospitable to fresh departures in signalling than the Imperial troops — made conspicuous successes with the new sets. On individual occasions they were the only means of signal communication between forward units for as long as 24 hours at a time. For a short time in the summer of 1918 they almost completely ousted the power buzzer and amplifier sets from the position these had won as a result of their brilliant successes in the 1917 battles. Then, however, mobile warfare set in and the loop sets, in company with other innovations of the position warfare period, were relegated to the bottom of the transport wagons and to relative obscurity.

The trench wireless set itself underwent little change during this year. Slight technical alterations resulted in a more robust and efficient set, but in its essentials it remained unchanged, awaiting its opportunity in the future. For a short time, these sets achieved a certain popularity with the artillery who were beginning to turn their attention to wireless for observation purposes. From this position, however, spark wireless was quickly driven by its half-brother, continuous wave wireless, which was just beginning a rapid and extraordinary career as applied to Army purposes.

Its great advantages over spark wireless were an extreme selectivity, an extraordinary efficiency which gave altogether disproportionate ranges for the same power, and, bound up with the latter trait, a consequent inconspicuousness of aerial which was a particularly great asset at forward observation posts.

These advantages were, however, for some time, counter-balanced by equally great disadvantages, though fortunately disadvantages which could be gradually overcome. The sets were extremely delicate, so delicate in fact that until 1917 none had been evolved which were capable of surviving the exigeant conditions of service in the field. The work also was technical in the highest degree, for the most simple sets required delicate and careful adjustment. On the early stations, the C.W. operator needed to be considerably more than a good manipulator of the key. There were literally dozens of ways in which the sets might go wrong, and some of these faults, at least, he must be capable of diagnosing and repairing. These earlier sets were gradually improved upon, and then C.W. wireless came further and further within the sphere of practical politics,. though until the Armistice its use in the forward area was practically confined to the artillery.

In June, 1917, two continuous wave wireless sets were in constant communication between a Corps H.A. headquarters at Ecorves and a Corps observation post at Bailleul — a distance of 12,000 yards. The forward aerial was 3 ft. in height and 25 yards in length; the height of the rear aerial at such a distance from the front line is, of course, immaterial. For some time when telephone communication was unsatisfactory the stations were working continuously and handled from 50 to 60 messages daily. This was a landmark in the history of Army wireless but it was soon paralleled by similar exploits in other Corps and other Armies. For a time, some Armies reported satisfactorily on the new wireless, others unfavourably; it was largely a question of the personality and technical bias of the wireless officer with the Army. Continuous wave wireless had, however, come to stay, and in December, 1917, orders were sent home for 882 sets, chiefly for the use of the artillery. At the same time, consideration was being given to the use of similar sets for anti-aircraft signal communication in which, as already indicated, direct communication was essential. Sound-ranging sections also put in a claim, and, indeed, there sprang up at once several ways in which this type of signalling might become of paramount importance.

Probably the most striking success of all the alternative methods of communication in 1917 fell to the lot of the previously despised power buzzer and amplifier. The first suggestion of the utilization of earth induction telegraphy is found in the early experiments which were conducted at the Signal Service Training Centre at about the time in 1915when the overhearing problem first became acute. Signals were transmitted through the earth by means of buzzers and were picked up on an ordinary telephone receiver connected to widely separated earths. The first practical apparatus used in the British Army was contrived at Vermelles in the area held by the dismounted cavalry Division in January, 1916. Here, two French earth conduction transmitters were installed in the forward trenches. They were hand-driven alternators giving a high note which was picked up in special listening sets which had just been introduced, also borrowed from the French Army. These represented the first type of power buzzer, but they were not persisted with for several reasons. They were unwieldy and there was difficulty in working them in the confined space of the forward dug-outs; but, above all, they were unfavourably reported upon on account of the likelihood of their being overheard. The overhearing problem was so acute at the time that all means of signalling which frankly depended upon overhearing for their success could not be expected to find favour.

It was, perhaps, natural that the French should take a prominent part in the development of this type of signalling. Their engineers were the inventors of the original valve listening set. It was on their front that intensive warfare reached its first great climax in the battles at Vimy and at Verdun. Their idea of forward signals also differed considerably from the British, their system being directed much more towards getting back information from the front line and less towards the conveyance forward of orders and instructions. In other words, the Staff, dealing with a fully-trained army of long standing, felt they could afford to leave more discretion to the forward commanders than could the Staff of the less mature British armies. One-way working from front to rear, therefore, presented many more attractions to the French Signal Service than to the British.

It was from the French, also, that the next step in the development of earth induction telegraphy came. This was the introduction to the British Army of the French "Parleur," which was the earlier equivalent of the British power buzzer. The possibilities of this large buzzer, worked by accumulators and with short earth leads, 100 to 150 yards in length, were at once seen and in 1916 it was generally adopted throughout our own Army. A proportion of the available listening sets were set aside to receive the messages sent by this means from selected forward spots and in the battles on the Somme and the Ancre these instruments gave valuable, if inconsistent, results. They were on the whole unfavourably reported upon, however, for several reasons. The buzzers themselves were fairly robust instruments, but the accumulators needed a certain amount of care and had to be carried in an upright position. The signaller stumbling along in the dusk across shell-pitted, wire-infester country, frequently arrived at his destination with his power buzzer in order, but with accumulators quite dry, having distributed the acid between his clothes, his other equipment, and the ground over which he had passed.

In addition, geological considerations intimately affected the efficiency of the sets. Earth conductivity varied with geological strata, with the physiographic relief of the country, and with the amount of shelling to which the ground had beer exposed. No two situations were alike. Carefully co-ordinated experiments were necessary before general rules for the use of the sets could be issued. Again, the earths, although less vulnerable than lines, were still vulnerable to a certain extent, even when they were laid in trenches. Finally, the British staffs and British officers generally had an ineradicable dislike to one-way signalling and would only use it as a last resource, while the signaller himself, sitting at his key and tapping out "S.O.S. K23 C4, 5" or some equivalent shibboleth, felt no conviction that his message was getting through. Only occasionally did the shriek and whine of friendly shells, passing overhead, point the moral of his message and give him pleasing proof of his value in the military machine.

The winter of 1916-1917 was, therefore, devoted to experiment with the new apparatus. In December, 1916, an order was issued by certain armies that ten first-class signallers in every Brigade should be withdrawn from infantry and trained with the power buzzer. At the same time, the protection of the stations was improved and it was found that the best results were obtained if the forward earths were buried to a depth of 6 ft. This could easily be accomplished with comparatively little labour by an army accustomed to burying cable on a large scale, and, with the power buzzer established in a deep dug-out and the earths buried 6 ft. deep, the station was practically shell-proof. Another expression of the same policy showed itself in the increase in length of the rear earths to 300 or 350 yards, thus making possible the use of still shorter bases — 60 or 70 yards — for the forward station.

The difficulty of one-way working was also overcome to some extent by the invention of special amplifier sets and their provision in large quantities. The opening of the 1917 campaigns thus found the earth induction set in a much stronger position than it had held in the previous year.

This was soon reflected in the reports of signal officers and of regimental officers, too. Again and again, it was recorded that, when all other communication had gone, power buzzer and amplifier sets had bridged the gap until better conditions returned. Most of the occasions were for a few hours only and the number of messages handled was of course relatively small. At Bullecourt, however, in May, 1917, during the attacks on the Hindenburg line, the classical instance occurred which illustrated at its best the value of this means of signalling under favourable circumstances. For several days, the shelling had been so intense that no ground or air communication was possible between the Australians in the Hindenburg line each of Bullecourt and their supports. Wireless aerials would not stand up; lines were blasted off the face of the earth, runners left the trenches only to be blown to pieces, the supply of pigeons could not be maintained. deep down in the tunnels and caverns beneath the fine, however, were installed two power buzzers and amplifiers. With their bases securely hidden in the tunnels and with the sets themselves under many feet of headcover, they remained inviolate. For several days after the line was occupied these sets kept up constant communication with other sets further to the rear, and gave warning of several impending counter-attacks which were successfully repulsed. At the same time, lateral communication with the 7th Division to the west of Bullecourt was kept up, though the village itself was at that time in the enemy possession. It was a striking instance of the triumph of a highly specialized form of intercommunication in the peculiar situation for which it was designed.

After this and other less dramatic but useful successes, the power buzzer achieved some degree of popularity and was used to a great extent as part of the normal intercommunication system in position warfare. In 1918 it was further improved by the issue of a combined set which made the two-way station more compact and portable. During this year, however, the "forward" wireless sets replaced it in the affections of many units, including the Australian Divisions with which it had scored its most notable success. With the passing of position warfare the earth induction set became so much lumber to be carried with the signal companies, but in its day it had been of vital use on many occasions and had well justified the time, money, and labour spent on its evolution, and the personnel employed in working it.

Footnotes

«--	The Signal Service in the European War of 1914 to 1918 Chapter XII	--»
Chapter XI		Chapter XIII