SECTION XXI: CHAPTER IV

THE LINES OF TORRES VEDRAS

We have hitherto, when speaking of Wellington’s immense scheme for fortifying the position on which he intended to bring his enemy to a standstill, refrained from entering into the details of his plan. It is now time to describe it in full, and to explain its design.

The character of the peninsula on which Lisbon stands lends itself sufficiently well to defence. At a first inspection the country-side offers a rather chaotic expanse of mountain and valley, whose general features are hard to seize from any one point. On further examination, it appears that the whole square mass of land between the Atlantic and the Tagus estuary is nothing more than a continuation of the ridge of the Serra de Monte Junta, the main mountain-chain of Estremadura. But from the backbone or central mass of the highland so many large spurs are thrown out to each side, and these are themselves so high and steep, that the whole peninsula seems more like a ganglion of mountains than a well-marked chain. The two chief joints or vertebrae in the backbone are the Monte Agra?a above Sobral, and the Cabe?a de Montechique six miles south of it, and these form the central points respectively of the first and second lines of defence which were finally laid out. Besides the outer defences there was in Wellington’s scheme, from the very start, an inner ring of works, covering only a small area on the sea-shore, at the southernmost point of the peninsula, to the west of Lisbon. This was merely intended to cover an embarkation, if by any unforeseen disaster the Lines themselves should be pierced.

It remains to speak of the system of defences in detail. In October 1809, Wellington’s plan had embraced no more than one continuous line of works from Alhandra on the Tagus to the[p. 420] mouth of the Rio S?o Louren?o on the Atlantic, with certain redoubts and fortified camps thrown out in front, at Torres Vedras, Monte Agra?a, and other points. These latter fortifications were not intended to be held in permanence; but it was hoped that they might defer and hinder the enemy’s attack on the main line in the rear. It was only the long delay in Masséna’s advance, which gave Wellington five or six months on which he had not counted, that led to the ultimate strengthening of the scattered outer works, and their conversion into a continuous whole, capable of turning back, instead of merely detaining for a time, the invading army. Indeed, all across the peninsula, designs that were slight, isolated, and provisional when first drawn up, were in the end enlarged, and perfected into wholly different structures. For the engineers, having unlimited labour at their disposal, and much more time than had been promised them, could turn their attention, after the essential works had been completed, to devising all manner of additional improvements and securities for the chosen position.

The construction of the Lines was entrusted to Colonel Fletcher, Wellington’s commanding engineer, who had as his chief assistant Major John Jones, the historian of the works, and in addition eleven British officers of the Royal Engineers, two from the King’s German Legion, and three from the Portuguese regular army. Wellington himself, after making one all-embracing survey of-the positions in Fletcher’s company in October 1809, and another in February 1810, left all the rest to his subordinate, and refrained from worrying him with matters of detail, being satisfied that his own intentions had been thoroughly well grasped. The labour available was, firstly, that of the Lisbon militia regiments, who were brought up by alternate pairs, and paid an extra 4d. a day for their services[478]; secondly, that of hired volunteers from the peasantry of the district, of whom from 5,000 to 7,000 were generally in hand; they received 1s., afterwards 1s. 8d. a day[479]; and lastly of a conscription from the whole of southern Estremadura, for a circuit of forty miles around. The forced labour was paid at the same rate as[p. 421] that freely hired. On the whole, only about ￡100,000 was paid out between November 1809 and September 1810—so that the Lines of Torres Vedras may be considered one of the cheapest investments in history. The militiamen and peasantry were worked in gangs of some 1,000 or 1,500 men, each in charge of an engineer officer, who had a few English and Portuguese military artificers as his assistants: only 150 such were available, so short were both armies of trained men. ‘In some districts a subaltern officer of engineers with a few English soldiers, utterly ignorant of the language, directed and controlled the labour of 1,500 peasantry, many of them compelled to work at a distance of forty miles from their homes, while their lands lay neglected. Nevertheless, during a year of this forced labour not a single instance of insubordination or riot occurred. The great quantity of work performed should, in justice to the Portuguese, be ascribed more to the regular habit of persevering labour in those employed than to the efficiency of the control exercised over them[480].... Indeed, it is but a tribute of justice to the Portuguese of Estremadura to state that, during many months of constant personal intercourse, both private and public, the labouring classes ever showed themselves respectful, industrious, docile, and obedient, while the governing classes in every public transaction evinced much intelligence, patriotism, good sense, and probity. Secrecy with respect to the extent and nature of the works was enjoined, and it is highly creditable to all concerned that hardly a vague paragraph concerning the Lines found its way into the public prints. The French invaders remained ignorant of the nature of the barrier rising against them, till they found our army arrayed on it so as to stop their further advance[481].’

The total frontage of the southern and stronger series of lines, those which Wellington originally planned as his line of defence, was twenty-two miles from sea to sea. The outer and northern series of works, which was originally only a supplement and outer bulwark to the other, was longer, extending to twenty-nine miles, for it crosses the peninsula in a diagonal fashion and not on the shortest possible line that could be drawn. Lastly,[p. 422] the small interior line round St. Julian’s and Oyeras, which was prepared as the embarking-place of the army in the event of defeat, has a circumference of about two miles. In all, therefore, fifty-three miles of defences were planned—a stupendous work, far exceeding, when its elaborate details are studied, anything that had been constructed in modern times in the way of field-fortification.

It must be remembered that the character of the Lines in no way resembles that of our own great Roman wall from Tyne to Solway, of the wall of China, or of any other long continuous stretch of masonry. It is only on a few points that works of any great length are to be found. The Lines are in essence a series of closed earthworks, dotted along the commanding points of the two ranges of hills which Wellington chose as his first and second fronts of resistance. Some few of the earthworks rose to the dignity of fortified camps, armed with many scores of guns. The majority of them were small redoubts, constructed to hold three to six guns and garrisons of two or three hundred men only. But even the smallest of them were individually formidable from their structure: the normal ditch was 16 feet wide and 12 feet deep, the parapets 8 to 14 feet thick, and all were properly fitted with banquettes. When it is remembered that they were well palisaded, and had outer hindrances of abattis, chevaux de frise, and trous-de-loup scattered in front, it is clear that they were forts requiring a regular attack, not mere lines of trench and mound. The strength of the whole series was that they were placed in scientific fashion, so as to cross fires over all the ground on which an attacking force was likely to present itself. No practicable point of assault could be found on which advancing columns would not be cut up by flanking fire for a very long distance, before they drew near to their objective. Immense pains had been taken to make the more exposed sections of the country-side into one vast glacis. Mounds which might have given cover had been removed to the last stone, hollow roads filled up, houses pulled down, olive-groves and vineyards stubbed up to the roots, so as to give a perfectly smooth and featureless ascent up to the line of redoubts. Greatly to Wellington’s credit (as may be incidentally remarked) compensation was paid on a liberal scale to all owners[p. 423] of dwellings, mills, fruit-trees, &c., for the havoc made by these necessary pieces of demolition. The result was a complete clearance of cover. ‘We have spared neither house, garden, vineyard, olive-trees, woods, or private property of any description,’ wrote the officer in charge of the works to his chief at the end of the preparations: ‘the only blind to the fire of the works now standing anywhere is that beautiful avenue of old trees in the pass of Torres Vedras. The Juiz da Fora and the inhabitants pleaded with me so hard for the latest moment, lest they might be cut down unnecessarily, that I have consented to defer it till the day before the troops march in. As I have trustworthy men with axes in readiness on the spot, there is no doubt of their being felled in time. The pine woods on the Torres heights are already down, and formed into abattis[482].’

It was not necessary, or indeed possible, to slope into a glacis the whole of the ground in front of each of the lines of defences. In many places other methods of making it impassable were used. At the north-western front of the first line, between Torres Vedras and the sea, for nearly six miles, a long marsh had been created: the river Zizandre had been dammed up, and had filled the whole of the narrow bottom in which it flows. ‘It has overflowed its banks, and in a short time more than half the valley has become so complete a bog that no reward can induce any of the peasantry to pass over it[483],’ wrote the officer who had carried out the experiment. Nor was it possible for the enemy to attempt to drain the bog, for four[484] redoubts furnished with heavy guns, and placed on dominating points of the hillside, commanded the bottom so completely that it was impossible for any party to approach it with safety. Yet the redoubts were out of the range of field-guns on the slopes beyond the Zizandre: only guns of position could have touched them, and Masséna had none such with him. Two[p. 424] similar inundations on a smaller scale had been caused at the other end of the Lines, by damming up the Alhandra and Alverca streams, each of which spread out in a marsh a mile broad, reaching to the foot of the heights above the Tagus, and could only be passed on the narrow paved high-road from Santarem to Lisbon.

In other places a very different method of making the Lines unapproachable had been adopted. Where the heights were very steep, but not absolutely inaccessible—a dangerous thing to the defence, for here ‘dead ground,’ unsearchable by the cannon of the redoubts above, must almost necessarily occur,—the slope had been cut or blasted away in bands, so as to make absolute precipices on a small scale. At one point above Alhandra[485] this was done on a front of full 2,000 yards. Even this was not the last precaution taken: at several places ravines ran deep into the line, and up them columns, more or less under cover, might possibly have penetrated. Such ravines, therefore, were stuffed, at chosen points, by a broad abattis or entanglement, mainly composed of olive-trees with all their chief boughs remaining, dragged together and interlaced for a depth of many yards. Such a structure could not be crawled through, nor could it be hewn down without an infinite waste of time and labour; nor, on the other hand, did it afford any cover, since grape or musketry could play perfectly well through it. The chief of these traps was that laid across the long ravine above the village of Arruda, down the bottom of which flows one of the winter torrents which fall eastward into the Tagus.

It was fortunate that Portugal was a well-wooded country: there are regions where it would be impossible to procure the immense amount of timber that was lavished on the accessories of the redoubts. All, as has been already mentioned, were palisaded; many had in addition abattis or entanglements thrown up in front of them, some way down the hillside, so as to detain the advancing enemy under fire as long as possible.

The works were divided into eight sections, the first line[p. 425] composed of four, the second line of three, while the eighth consisted of the inner retrenchment for purposes of embarkation, at the extreme southern point of the peninsula. Of the outer or secondary line the three sections were:

(1) A front of five miles from the Tagus at Alhandra along the crest of a steep but not very lofty ridge, as far as the great ravine that overlooks the village of Arruda. This front was elaborately fortified, as it blocks the great road, in the flat by the waterside, which forms the easiest approach to Lisbon from the north. In the five miles there were ultimately constructed no less than 23 redoubts with 96 guns. Two thousand yards of hillside in one place had been scarped into a precipice; a mile by the side of the Tagus had been inundated. The one considerable gap in the line, the ravine at the head of the valley of Calandriz, had been choked by one of the great abattis above described. The redoubts required a garrison of 6,000 men.

(2) The second section, from the ravine above Arruda to the left of the steep Monte Agra?a, formed somewhat of a salient angle: it had a front in all of some four and a half miles, which included the most lofty and defensible part of the backbone range of the Lisbon peninsula. One of the four great paved roads entering the capital from the north, however, passes over the shoulder of these heights, and they were therefore very heavily fortified from the first, the large redoubt for 1,600 men, on the top of Monte Agra?a being one of the original outer works ordered for construction in Wellington’s earliest notes of October 1809. There were in all seven redoubts mounting 55 guns and requiring a garrison of 3,000 men on this fraction of the lines.

(3) Quite different in character was the front of eight miles from the left of Monte Agra?a to the pass of Runa, overlooking the upper valley of the Zizandre and the village of Sobral. The fortification of this line had not entered into Wellington’s original plan, and there were only two redoubts upon it when Masséna appeared before it in October 1810. Such defence as there was consisted in the fact that the dominating Monte Agra?a redoubts overlooked it on the right, and that the two small works just mentioned commanded the main high-road from Sobral to Cabe?a de Montechique, which goes through its centre.[p. 426] But there was a clear possibility that the enemy might make a push up the valley and the high-road, by the village of Zibreira, and this was indeed the most probable point of attack in the whole 29 miles of front for the enemy to select. When, at the last moment, the British Commander-in-Chief determined to hold the outer lines, and not merely to fall back after having used them for a temporary defence, he had to cram this point with troops, and to construct new works upon it as quickly as possible. Four divisions, therefore, more than 20,000 men, were concentrated here. Wellington’s own head quarters were established at the hamlet of Pero Negro, on the slope above the high-road, and a very large redoubt was thrown up on the Portello hill, above Zibreira, with several smaller ones further to the right, to connect it with the Monte Agra?a works. Sobral, the village at the foot of the heights, was held as an outpost, but abandoned when Masséna pushed forward to the front, as it was too far advanced to the north to be treated as an integral part of the position. But the French, when they had carried Sobral with difficulty, looked at the main line behind it, and refused to attempt any further advance. The hillside was as formidable as the Bussaco heights from which they had only recently been repulsed: it was full of troops and growing in strength every moment as the earthworks continued to arise.

(4) The fourth section of the outer or northern front was that from the gorge of the Zizandre (or the pass of Runa, as it is sometimes called) to the sea. It was about twelve miles long, but of this space six miles and more was covered by the impassable bog formed by the obstructed Zizandre, and another mile was formed by the formidable entrenched camp of San Vincente, above the town of Torres Vedras, the most complete and self-sufficing of all the works in the peninsula. This stronghold lay outside the main line, beyond the river, covering the bridge and the paved chaussée from Leiria to Lisbon, the only carriage-road on the western side of the Lines[486]. It was one of[p. 427] the earliest of the fortifications commenced by Wellington’s engineers, having been started on November 8, 1809, and was placed in such a conspicuous point, and planned on such a large scale, that it attracted public attention more than any other part of the works, and gave its name to the whole in popular parlance[487]. The whole front on both sides of Torres Vedras and its great fort was so strong and inaccessible as to offer little temptation to the invader to select it as a point of serious attack, all the more so because troops brought opposite to it would be completely cut off from any left in front of the eastern and central part of the lines. For the geography of the peninsula at this point is peculiar: north of the gorge of the Zizandre the great backbone range, the Serra de Barregudo and the Serra de Monte Junta, extends for a distance of fifteen miles, without being crossed by a single road practicable for horses, much less for wheeled vehicles. There are nothing but goat-tracks across the heights. If, therefore, any considerable body of troops had been sent to observe or contain the western section of the lines, it would have been separated by two days’ march from the rest of the army, and liable to be crushed, ere succoured, by the defenders of Lisbon, who had good cross-roads across the peninsula, by which they could transfer themselves from point to point under the protection of their works. As a matter of fact, nothing but flying parties of French horse ever appeared in this direction. Masséna had not troops to spare for any secondary attack, more especially for one on such an unpromising part of the Lines. Wellington had foreseen this when he distributed his field army behind the various sections of the front: to support the garrisons of the twelve miles of redoubts about Torres Vedras he only placed one division, while there were three behind the eastern section, and more than four in the partially entrenched central part[488].

Passing on to the second line of defence, from Quintella on the Tagus to the mouth of the Rio S?o Louren?o on the[p. 428] Atlantic, we find three sections of defence, which, unlike those of the outer line, were all completed by September 1810, and had no central gap.

(1) There was over a mile of impassable inundation at the eastern end, between Quintella and Alverca. Above the first-named village was an isolated hill, which was all fortress, for no less than six redoubts had been placed upon it, to enfilade the high-road across the inundated lower ground. Then came the Serra de Serves, three miles of lofty and difficult hills, which had been scarped into almost perfect inaccessibility. In a sudden dip west of this range was the pass of Bucellas, through which runs one of the three great high-roads that enter Lisbon. It was easily defensible, as it lies between two high and steep mountain-sides, and is only a couple of hundred yards broad. Redoubts were placed so as to rake it from end to end, and to flank it on both sides. The chaussée itself was blocked with successive abattis, and the viaduct leading up to it was mined.

(2) The second section of the inner line extended from the pass of Bucellas to the Park of Mafra, a front of over six miles. The eastern part of this was formed by the towering heights of the Cabe?a de Montechique, the most dominating mountain-summit in the whole peninsula, almost steep enough to defend itself without fortification; but three redoubts nevertheless had been reared upon its summit. But from the pass of Montechique, at the left side of the summit, down to Mafra the ground was less well marked, and here the chaussée from Sobral and Zibreira crossed the range. Much fortification, therefore, was lavished on these four miles, along which there were nine strong redoubts, connected with each other in the rear by a military road passing along the southern crest of the heights. There was a second and formidable ridge behind this line, where further defence could be offered in the unlikely event of the enemy forcing his way up the high-road.

(3) From Mafra to the sea, nearly ten miles, there was for the most part a well-marked line of heights protected in front by the ravine of the river of S?o Louren?o, a deep, rugged, and in many cases inaccessible cleft, only crossed by a single road, that from Torres Vedras to Mafra. Nevertheless, six redoubts were reared, to cover this, and the few other points where the ravine[p. 429] was passable. The eastern part of this section, that along the wall of the Royal Park (Tapada) of Mafra, was its weakest portion, and for two miles at this point the British engineers set all their ingenuity to work. The outlying heights called the Serra de Chypre, in front of the park, were covered by four redoubts, and turned into a first defence. The wall of the Tapada itself was loopholed and furnished with a banquette. The important road which passes its foot was obstructed with cuts, enfiladed by the artillery of several works, and stockaded at more than one point. There was another group of redoubts along the south end of the Torres Vedras road, at the village of Morugueira; and finally Mafra town, in the rear of all, was turned into a defensive post by means of trenches and barricades. Altogether, what was by nature the weakest point in the southern lines was made by art one of the strongest. This too, in spite of the fact that, being approachable only from Torres Vedras, it was on the whole not a probable front on which to expect an attack.

A mere mention must suffice for the eighth section of the defensive works, the semicircle at St. Julian’s and Oyeras which was intended to protect the embarkation of the army if the worst should come. It was strongly entrenched, and could be held by a very few battalions, while the rest were utilizing the numerous and solid piers alongside of which the fleet of transports was to be moored.

Having described the Lines, it remains that we should describe the garrison set to guard them, detailing separately each element, regular and irregular. The forces at the disposition of Wellington were materially increased at the moment of his arrival within the Lines. On October 8 he found at his disposition a brigade of three battalions newly arrived at Lisbon, the 1/50th, 1/71st, and 1/92nd, all old Corunna regiments which had served in the Walcheren expedition, and were still none too healthy from their long sojourn in the deadly marshes of Zeeland. There had also landed about the same time the 94th regiment, and the Brunswick Oels Light Infantry, a foreign battalion raised from the refugees who had fought under the Duke of Brunswick in the abortive North German insurrection of 1809. Moreover, two battalions—the[p. 430] 2/30th and 2/44th—had just been sent to Lisbon from Cadiz, where General Graham now thought that the British contingent was larger than was absolutely necessary. The 1/4th and 1/23rd came out a little later, and do not appear in the fortnighty ‘general state’ of the army till November 15. Thus the army was swelled by nine battalions, or some 6,500 men[489]. No cavalry, however, had arrived.

Wellington used these new arrivals to form a new 6th Division of infantry, and to complete to full strength the 5th Division, which had hitherto possessed only one British brigade. He did not, however, keep the lately landed units together: acting on the principle which he always followed, of mixing veteran acclimatized battalions with new arrivals, he formed the new 6th Division by adding Campbell’s brigade, taken from Cole’s 4th Division, to two Portuguese regiments the 8th and the Lusitanian Legion, both of which had been hitherto attached to the 5th Division, and had served with Leith at Bussaco. In the 4th Division, Campbell’s brigade was replaced by that of Pakenham, taken from Spencer’s 1st Division, while compensation was made to Spencer, by giving him the newly landed 1/50th, 1/71st, and 1/92nd, as a new brigade under Erskine. The 5th Division under Leith got three more of the fresh arrivals, the 1/4th, 2/30th, and 2/44th as its second brigade. The 94th was given to Picton’s second brigade,—which had hitherto consisted of only 2? battalions,—to raise it to average brigade-strength. The Brunswick Oels J?gers, being a light corps, were partly divided up into separate companies and told off to different brigades (as the 5/60th, a similar unit, had already been), though the head quarters and six companies joined Pakenham’s brigade in the 4th Division. But when the 1/23rd, the last of the reinforcements, came out, it also joined Pakenham, while the Brunswickers were transferred to the Light Division—where they did not long abide[490].

Even after allowing for the trifling losses at Bussaco, the British field army in Portugal was now far larger than it had[p. 431] ever been before, the gross total of troops in the Lines amounting to 42,000 men, of whom about 7,000 were sick or detached, and 35,000 were present under arms. This figure does not include the two battalions of marines who guarded St. Julian’s and the lines around it at the mouth of the Tagus.

Of Portuguese regulars, Wellington had now under his hand the 24,000 men who had fought at Bussaco, plus the 1,400 cavalry under Fane and the brigade under Bradford (now consisting of five battalions)[491] which had been guarding the position behind the Alva on the day of the battle, together with the reserve artillery of Lisbon. The total made 27,500 men, of whom 24,500 were with the colours and 3,000 sick in hospital.

Of militia there were three brigades and four isolated units more within the Lines—the Southern Beira brigade of Lecor, three regiments[492], the Northern Estremaduran brigade of Miranda, also three regiments[493], and the Lisbon local brigade of five regiments[494] with two stray units from the north[495], and two from the south[496]. The numbers of all the regiments ran very low, owing to the way in which they had been neglected and under-fed by their government, since they were called out nine months before; many had died, and far more had deserted. The thirteen corps did not between them supply more than 8,200 men present under arms, with 1,000 sick in hospital. In addition there were 3,200 artillerymen improvised from the ranks of the infantry militia or the Ordenan?a, making altogether 12,400 troops of the ‘second line.’

As to the Ordenan?a who had taken refuge in and about Lisbon with their families, when the whole population of south[p. 432]western Beira and northern Estremadura retired within the Lines, it is impossible to obtain any figures, save that they supplied the bulk of the 3,000 volunteer artillerymen just mentioned above, and that the picked men of the Ordenan?a of the capital itself had been organized into two battalions of ‘Atiradores Nacionales’ of about 450 men each. The whole may have amounted to any number from 20,000 up to 40,000 men, of whom about two-thirds were armed with muskets, the rest, those from the remoter districts, having still nothing better than pikes. As most of them were scattered with their families in the villages where they had taken refuge, or the camps of huts which they had formed in sheltered situations, they could hardly be considered to be in a state of mobilization, and certainly were of no use either for garrisoning forts or for employment in the line of battle.
Map of the Lines of Torres Vedras

Enlarge  LINES OF TORRES VEDRAS

Lastly, in calculating the forces which Wellington accumulated within the Lines, we must mention the two Spanish divisions from the Army of Estremadura. Hearing that all was quiet for the moment on the frontier of Andalusia, the British Commander had asked the Marquis of La Romana, whether, in accordance with a promise made so long ago as July, he could spare any troops to assist in the holding back of the main French army of invasion. The Marquis, with a liberality of which the Cadiz Regency would have disapproved, if its leave had been asked, replied that he would bring up his two reserve divisions. Leaving Ballasteros on the Andalusian border, and another division under Imaz at Badajoz, in addition to the garrison and Madden’s Portuguese cavalry, he marched for Aldea Gallega and Lisbon with the troops of La Carrera[497] and Charles O’Donnell[498], about 8,000 men. On October 25th he had arrived at the cantonments behind Mafra, on the second line of defence, which his ally had requested him to occupy. Wellington defended the bringing up of these troops by the plea ‘that he did not think himself justified in not bringing into his positions all the force which was at his disposal’[499]. But it is doubtful whether the[p. 433] advantage of getting 8,000 Spanish troops within the Lines justified the danger incurred in Estremadura, when it was possible that Soult might send out Mortier at any moment to attack the depleted army that covered the approach to Badajoz. Napoleon thought that he should have done so, and when he heard of the arrival of La Romana at Lisbon, wrote to censure the Duke of Dalmatia in the fiercest strain[500]. ‘It was a shame and a scandal that he had retired to Seville: the 5th Corps had orders to be always at La Romana’s heels, and to prevent him from moving into Portugal, so that the news of its return to Seville roused the Emperor’s surprise and anger.’ There can be no doubt that Napoleon did well to be angry. The balance of affairs in Andalusia tended to stand at an equipoise precisely because La Romana’s army was strong enough to keep the 5th Corps employed. When 8,000 men had been withdrawn by the Marquis to the Lisbon lines, Mortier was in a position to sweep all before him as far as the gates of Badajoz, or to execute a raid into the Alemtejo if that course seemed preferable. But Soult did not send his lieutenant on this errand on his own initiative, but waited till he received direct orders to do so from Paris. By that time it was too late, and neither the disaster of the Gebora nor even the fall of Badajoz had any influence on the course of events in Portugal. Masséna was forced to retreat before a single patrol from the Army of Andalusia had got into touch with his outposts. What might have happened if Soult had launched his blow at Badajoz in October, and had appeared on the left bank of the Lower Tagus in December, it is impossible to say. Probably Wellington would have found some means of averting disaster, but it is unquestionable that his task of defence would have been made far more difficult.

For the full realization of the meaning of the Lines of Torres Vedras there are two general facts which must be remembered. Firstly, they were garrisoned by troops which formed no part of the field army. Wellington’s sixty thousand regulars were not frittered away in the garrisoning of redoubts, but were held in masses behind the lines, ready to reinforce any threatened point, and to deliver a pitched battle in the open, if the head of the[p. 434] French army were thrust through the defences at some weak section. The generals of the seventeenth and eighteenth centuries, who so often built lines, and were so easily evicted from them, suffered disaster because they drew out their armies in one attenuated thread, and were therefore weak at every point, and always inferior to the assailant at the place where he made his assault. Wellington’s army was (with the exception of the 3rd Division at Torres Vedras) gathered in two solid masses, one facing Sobral, on the heights between Monte Agra?a and Runa, the second and smaller behind Alhandra. The one could reach the other in half a day’s march, for the roads behind and parallel to the lines had been put in good repair.

The whole of this vast system of redoubts was to be held by the troops of the second line, and by them only. There were altogether some 20,000 men of the second line in the fortifications, composed of (1) the 8,000 (afterwards raised to 11,000) militia infantry. (2) Of about 800 Portuguese regular artillery, aided by over 2,000 gunners picked from the militia and Ordenan?a, trained by the regulars and incorporated with them. (3) Of some 250 British artillerymen from the batteries which had been lying in reserve at Lisbon. (4) Of picked companies of the Lisbon Ordenan?a (atiradores) drilled into a state of discipline not much worse than that of the militia. (5) Of the landing force of 2,000 British marines, partly from the fleet, partly brought specially from England to garrison the proposed lines of embarkation at St. Julian’s. (6) Of the dép?ts, convalescents and recruits of the eight Line regiments of infantry raised from Lisbon and Southern Estremadura—about 4,000 strong.

In all, therefore, there were about 20,000 men, mostly troops of secondary quality, or 28,000 if the Spanish auxiliaries are counted, ready to man the Lines, without a man being withdrawn from the ranks of the field army. The outer lines were calculated to require about 18,000 men for the redoubts, the inner ones 14,000, but clearly both did not require to be manned at once. If the outer line were broken, the garrison-troops from the intact parts of it could fall back on the second. Meanwhile the field army would be engaging any French columns that might have broken through,[p. 435] and there would be ample time to arrange for the manning of the second and stronger front. But it must be repeated once more that it was not on the passive defence of the redoubts by their garrisons that Wellington reckoned for success, but on the fighting of the field army, who would tackle the columns of attack that had committed themselves to the assault of the section—whichever it might be—that Masséna might select as his objective. All criticism based on general principles concerning the weakness of long extended lines falls to the ground, when it is remembered that Wellington had his army massed for a pitched battle in and behind his defences, not strung out on an interminable front.

The last point on which stress must be laid is that the most careful arrangements for the transmission of orders and intelligence from end to end of the Lines had been made. There were five signal-stations, with semaphores worked by seamen on (1) the redoubt No. 30 near the Atlantic, (2) the great redoubt of Torres Vedras, (3) the Monte de Socorro above Wellington’s head quarters at Pero Negro, (4) the summit of the Monte Agra?a, (5) the hill behind Alhandra on the Tagus. After some practice it was found that a message could be sent from one end to the other of the 29 miles in seven minutes, and from No. 3, the head quarters semaphore, to either end of the Lines in four minutes. There was a similar line of four semaphores on the second, or main, series of defences. Military roads had been opened behind both the fronts, so that troops could be moved along the shortest possible line. On the other hand, it was fortunate that there existed no cross-road from sea to sea outside the Lines, which could be of any practical use to the invader. The only route of this sort, that from Alemquer by Sobral to Runa, was commanded for the whole length from Sobral to Runa by the British heights, whose foot it hugs, while from Sobral to Alemquer it is separated from the Lines by the steep and pathless ridge of Galaria, across which nothing on wheels could pass. Nevertheless, here lay the invaders’ best chance—corps placed on this road, and screened by the ridge, could be moved for some distance to left or right unseen from the Lines. The road, however, was bad, rocky, and narrow: it is marked as the ‘Cal?ada Arruinada’ or ‘ruined road’ in contemporary maps. The other[p. 436] paved road in this direction, that from Sobral to Arruda and Alhandra, passed through the line of ground occupied by the British at two points, and was under fire from the redoubts at short range for the rest of its course: it was absolutely impracticable.

It only remains to be added that the navy had been utilized for auxiliary service: not only were its marines under orders to man the St. Julian’s lines, but its seamen had fitted out all the gunboats in the Lisbon arsenal. A flotilla of great strength infested the Tagus estuary, and by the fire of its heavy guns prevented the French from approaching the shore, or endeavouring to build boats at the mouths of its creeks. If any attacks had been made upon either of the extreme ends of the Lines, the columns delivering them would have been under fire from the sea throughout their operations. But, as we shall see, the French never contemplated this: the one temptation which Masséna felt was to assault, far inland, the gap in front of Sobral between the Monte Agra?a and the Serra de Socorro. And there, as we shall see, at the critical moment, prudence got the better of ambition, and the invader turned back foiled. The high-water mark of French conquest in Europe was reached on the knoll by Sobral on the wet and gusty 14th of October, 1810.