CHAPTER VI

COMBAT OF THE COA: SIEGE OF ALMEIDA (JULY-AUGUST 1810)

On July 10th the French had entered Ciudad Rodrigo, but ten days more elapsed before they made any further advance. Masséna, who had returned to the front, was resolved to follow his master’s orders and to act ‘methodically.’ It was clearly incumbent on him to begin the siege of Almeida as soon as possible, and, as that place is only twenty-one miles from Ciudad Rodrigo, one long march would have placed him before its walls. But since he had only a few thousand rounds of ammunition left for his heavy guns, he refused to move on till all the available reserves were on their way from Salamanca to the front, and requisition for a further supply had been sent to Bayonne. He had also to do his best to scrape together more food, since the magazines that Ney had collected were nearly exhausted when Rodrigo fell. Moreover, 1,500 draught animals had died during the late siege, and it was necessary to replace them before the Great Park could move forward.

On July 21st, however, some convoys having come up from Salamanca, Masséna directed Ney to advance with the 6th Corps and to drive Craufurd back on to Almeida. The main point that he was directed to ascertain on this day was whether the English intended to make a stand at Fort Concepcion, the isolated Spanish work which faces Almeida on the frontier, beyond the Turones. This was a solid eighteenth-century fort, covering the bridge where the high-road passes the river. It had lately been repaired, and could have resisted a bombardment for some days. But it would have required a garrison of 1,000 men, and, since it lies in the midst of the plain, there would be little chance of relieving it, if once it were surrounded. Wellington, therefore, gave orders that it should be blown up whenever the French should advance in force toward Almeida, and that Craufurd should make no attempt to defend the line of the Turones, and should send back his infantry to Junca, a village about a mile outside the gates of Almeida[296], keeping his cavalry only to the front. On the 21st Ney advanced with the whole of Loison’s division, and Treillard’s cavalry brigade. Thereupon Craufurd, with some reluctance, retired and blew up Fort Concepcion as he went. The French advanced, skirmishing with the 14th Light Dragoons and the German Hussars, but finally halted at Val de Mula, four miles from Almeida. Craufurd established himself at Junca, only three miles from the enemy’s line of pickets. On the next evening he received a strong suggestion, if not quite an order, from his Chief to send his infantry across the Coa. ‘I am not desirous of engaging an affair beyond the Coa,’ wrote Wellington. ‘Under these circumstances, if you are not covered from the sun where you are, would it not be better that you should come to this side of it, with your infantry at least[297]?’ The tentative form of the note well marks the confidence that the Commander-in-Chief was wont to place in his subordinate’s judgement. This time that confidence was somewhat misplaced, for Craufurd tarried two days longer by the glacis of Almeida, and thereby risked a disaster.

[296] Wellington to Craufurd from Alverca, July 16.

[297] Wellington to Craufurd from Alverca, July 22, 8 p.m.

It must be remembered that Almeida is not on the Coa, but two miles from it, and that its guns, therefore, did not cover the one bridge over which Craufurd could make his retreat. Indeed, that bridge and the river also are invisible from Almeida. The fortress is slightly raised above the level of the rolling plain, which extends as far as Ciudad Rodrigo: the river flows in a deep bed, so much below the plateau as to be lost to sight. Its ravine is a sort of cañon which marks the end of the plains of Leon. It has often been remarked that Almeida’s value would have been doubled, if only it had been on the near side of the Coa, and commanded its bridge. But Portuguese kings had built and rebuilt the old fortress on its original site, with no regard for strategy. Craufurd, then, should have remembered that, if he were suddenly attacked in his camp outside the gates, he risked being thrown back into the town (the last thing he would wish), or being hustled down to the bridge and forced to pass his division across it in dangerous haste. But he had so often challenged, held back, and evaded Ney’s and Junot’s advanced guards, that he evidently considered that he was taking no very serious risk in staying where he was. He was, moreover, discharging a valuable function by keeping Almeida from being invested, as stores and munitions were still being poured into the place. The only peril was that he might be attacked both without warning and by overwhelming superiority of numbers, with the defile at his back. Neither of these misadventures had yet happened to him during the four and a half months while he had been defying the 6th and the 8th Corps along the banks of the Agueda. The French had never assailed him with much more than a division, nor had they ever pressed on him with headlong speed, so as to prevent an orderly retreat properly covered by a moderate rearguard.

Now, however, Ney, untrammelled by any other operation, had his whole corps concentrated behind Val de Mula, and having learnt of the defile that lay in Craufurd’s rear, thought that he might be hurled into it and crushed or caught. Before dawn he arrayed his whole 24,000 men in one broad and deep column. Two cavalry brigades, Lamotte’s 3rd Hussars and 15th Chasseurs, and Gardanne’s 15th and 25th Dragoons, were in front. Then came the thirteen battalions of Loison’s division, in a line of columns; behind them was Mermet with eleven battalions more, while three regiments of Marchand’s division (the fourth was garrisoning Ciudad Rodrigo) formed the reserve. In a grey morning, following a night of bitter rain, the French horsemen rode at the British cavalry pickets, and sent them flying helter-skelter across the three miles of rolling ground that lay in front of the Light Division. On hearing the fire of the carbines Craufurd’s men turned out with their accustomed celerity, and in a very short space were aligned to the right of Almeida, with their flank only 800 yards from the glacis, and their front covered by a series of high stone walls bounding suburban fields. There would have been just enough time to get cavalry, guns, and impedimenta across the bridge of the Coa if the General had started off at once. But, not realizing the fearful strength that lay behind the French cavalry advance, he resolved to treat himself to a rearguard action, and not to go till he was pushed. On a survey of the ground it is easy to understand the temptation, for it would be hard to find a prettier battlefield for a detaining force, if only the enemy were in no more than moderate strength. A long double-headed spur runs down from the high plateau on which Almeida stands to the Coa. Successive points of it can be held one after the other, and it is crossed by many stone walls giving good cover for skirmishers. With his left covered by the fire of the fortress, and his right ‘refused’ and trending back towards the river, Craufurd waited to be attacked, intending to give the leading French brigade a lesson. There was a delay of more than an hour before the French infantry was up, but when the assault came it was overwhelming. Craufurd’s line of three British and two Portuguese battalions[298] was suddenly assaulted by Loison’s thirteen, who came on at the _pas de charge_ ‘yelling, with drums beating, and the officers, like mountebanks, running ahead with their hats on their swords, capering like madmen and crying as they turned to wave on their men, “Allons, enfants de la Patrie, le premier qui s’avançera, Napoléon le recompensera[299].”’ The rolling fire of the British stopped the first rush, when suddenly a French cavalry regiment, the 3rd Hussars, charged across the interval between Craufurd’s left and the walls of Almeida, braving the fire of the ramparts in the most gallant style. Some fell, but the gunners were flurried at this unexpected development, and fired wildly, so that the hussars swept down unchecked on the extreme flank of the Light Division, where a company of the 95th Rifles was annihilated[300], and began riding along the rear of the line and rolling it up. They were luckily checked for a moment by a stone wall, but Craufurd saw that he must retreat at once, since he was turned on the side where he had thought that he was safest. The cavalry and guns were ordered to gallop for the bridge, the Caçadores to follow them, and the rest of the infantry to fall back in échelon from the left, defending each enclosure and fold of the hillside as long as possible. But it is hard to make an orderly retreat when a foe with twofold strength in his fighting line is pressing hard. Moreover, the road to the bridge has an unfortunate peculiarity; instead of making straight for its goal it overshoots it, in order to descend the slope at an easy point, and then comes back along the river bank for a quarter of a mile. The cavalry and guns, forced to keep to the road because the hillside was too steep for them, had to cover two sides of a triangle with a sharp turn at the apex, which delayed them terribly. To add to the trouble an artillery caisson was upset at a sharp turn, and took much trouble to right and send forward. Thus it chanced that the covering infantry were driven down close to the bridge before the Caçadores and the last of the guns had crossed the river. The retreat of the three British battalions had been most perilous; at one moment a wing of the 43rd found themselves checked by a vineyard wall ten feet high, while the French were pressing hard on their rear. They only escaped by shoving a long part of it over by sheer strength--fortunately, like all other walls in this part of Portugal, it was made of dry flat stones without mortar. Finally the 43rd, the Rifles, and part of the 52nd were massed on a long knoll covered with pine-trees, which lies above the bridge and completely masks it against an attack from above. While they held firm, Craufurd ranged the guns and the Caçadores on the slopes upon the other side, so as to command the passage when the rest of the troops should have to cross. He then began to withdraw the 43rd, and part of that battalion had already crossed the water, when five companies of the 52nd, which had occupied the extreme right wing of the division, were seen hastening along the river bank some way above the bridge. They had held out a little too long on the slopes above, and seemed likely to be cut off, for the French, noting their position, made a vigorous effort, and carried the knoll which protected their line of retreat to the point of passage. This was a desperate crisis, but such was the splendid courage and initiative of the regimental officers of the Light Division that the disaster was averted. At one point Beckwith, colonel of the Rifles, at another Major McLeod of the 43rd, called on the disordered mass of men, who had been driven back to the bridge head, to charge again and save the 52nd. The soldiers grasped the situation, cheered and followed; they recaptured the knoll that they had just lost, and held it for ten minutes more, gaining time for the companies of the 52nd to pass behind and cross the bridge. ‘No one present,’ wrote an eye-witness, ‘can fail to remember the gallantry of Major McLeod. How either he or his horse escaped being blown to atoms, while in this daring manner he charged on horseback at the head of some 200 skirmishers of the 43rd and 95th mixed together, and headed them in making a dash at the line of French infantry, whom we dislodged, I am at a loss to imagine. It was one of those extraordinary escapes which tend to implant in the mind some faith in the doctrine of fatality[301].’

[298] The 43rd on the left, the two Caçador battalions in the centre, the 52nd on the right, while the Rifles were partly dispersed along the front, partly with the 43rd.

[299] Simmons’s _Journal of a British Rifleman_, p. 77.

[300] Of this, O’Hare’s Company of the 1/95th, sixty-seven strong, an officer and eleven men were killed or wounded and forty-five were taken prisoners.

[301] Leach’s _Reminiscences_, pp. 149-50.

The moment that the 52nd were safe, the troops on the knoll evacuated it, and crossed the bridge behind them at full speed, while the French reoccupied the wooded eminence. If Ney had been wise he would have stopped at this moment, and have contented himself with having driven in the Light Division with a loss of 300 men, while his own troops had suffered comparatively little. But, carried away by the excitement of victory, he resolved to storm the bridge, thinking that the British troops were too much shaken and disordered to make another stand, even in a strong position. There were plenty of examples in recent French military history, from Lodi to Ebersberg, where passages had been forced under difficulties as great. Accordingly he ordered the 66th, the leading regiment of Loison’s division, to push on and cross the river. This was a dire mistake: Craufurd already had the Caçadores in position behind stone walls a little above the bridge, and Ross’s guns placed across the road so as to sweep it from end to end. The British battalions were no sooner across the river than they began to string themselves out behind the rocks and walls, which lie in a sort of small amphitheatre on the slope commanding the passage. The bridge, a two-arched structure seventy yards long, crosses the Coa diagonally, at a point where it is narrowed down between rocks, and flows very fiercely: it was flooded at this moment from the rain of the previous night, and was swelling still, for a tropical storm had just begun and raged at intervals throughout the afternoon. The cavalry, useless at the bridge, was sent up-stream to watch some difficult fords near Alveirenos.

The French 66th, ordered by the Marshal to carry the bridge, formed its grenadiers on the knoll, to lead the column, and then charged at the passage. But the leading company was mown down, before it had got half way across, by a concentrated musketry salvo from the hillside in front, and the enfilading fire of the guns from the right. The column broke, and the men recoiled and dispersed among the rocks and trees by the bank, from whence they opened a fierce but ineffective fire upon the well-sheltered British battalions. Ney, who had now lost his temper, ordered up a _bataillon d’élite_ of light infantry[302] which had distinguished itself at the siege of Ciudad Rodrigo, and told his aide de camp Sprünglin to take the command and cross at all costs[303]. There ensued a most gallant effort and a hideous butchery. The Chasseurs flung themselves at the bridge, and pushed on till it was absolutely blocked by the bodies of the killed and the wounded, and till they themselves had been almost literally exterminated, for out of a battalion of little more than 300 men 90 were killed and 147 wounded in less than ten minutes. A few survivors actually crossed the bridge, and threw themselves down among the rocks at its western end, where they took shelter from the British fire in a little corner of dead ground, but could of course make no further attempt to advance.

[302] The _Chasseurs de la Siège_ formed of picked marksmen from all the regiments of the 6th Corps.

[303] That Ney himself was the person responsible for this mad adventure seems proved by the journal of Sprünglin, who writes ‘À midi je reçus de M. le Maréchal lui-même l’ordre d’emporter _à tout prix_ le pont de la Coa, d’où deux compagnies de Grenadiers venaient d’être repoussés. J’avais 300 hommes; je formai mon bataillon en colonne et abordai les Anglais à la baïonnette, et au cri de _Vive l’Empereur_. Le pont fut emporté, mais j’eus 4 officiers et 86 soldats tués, et 3 officiers et 144 soldats blessés. Le 25 le bataillon, étant détruit, fut dissous.’ That the bridge was ‘emporté’ in any other sense than that a score or so of survivors got to the other side, and then returned, is of course untrue. Sprünglin, p. 439.

Ney, irritated beyond measure, now bade a mounted officer sound for a ford at a spot above the bridge, where the river spreads out into a broad reach. But horse and man were killed by a volley from the British side, and floated down the swollen stream[304]. Finding the river impracticable, the Marshal again ordered the 66th to go forward: this third attack, delivered without the dash and determination of the first two, was beaten back with little trouble. The firing then died down, and during one of the fierce rainstorms of the late afternoon the few chasseurs who had crossed the bridge ran back and escaped to their own bank. Craufurd held the position that he had occupied till midnight, and then retired on Pinhel. He had lost 333 men only[305], and was fortunate therein, for half his division might have been destroyed if the officers had shown less intelligence and the men less pluck. The French had 527 casualties, four-fifths of them in the mad attempt to force the bridge, in which the colonel of the 66th and fifteen of his officers had fallen, and the battalion of Chasseurs had been practically exterminated[306]. Ney forwarded an honest chronicle of the day’s doings to his chief, which Masséna wrote up, and sent to the Emperor turned into a work of fancy, in which he declared that he had destroyed 1,200 of Craufurd’s men (whom he estimated at 2,000 horse and 8,000 infantry, double their real strength), taken 300 prisoners, a colour, and two guns. Making no mention of the complete check that Loison’s division had suffered at the bridge, he stated that ‘the Imperial troops have shown once again this day that there is no position which can resist their intrepidity.’ He added foolish gossip, ‘Their Estafete-Mor (chief Portuguese courier) has been captured with all his dispatches, in which are several of the 25th and 26th instant, which declare that the English army is in complete rout, that its deplorable state cannot be exaggerated, that the English have never been in such a hot corner, that they have lost sixty officers, of whom they buried twenty-four on the battlefield, about 400 dead and 700 wounded[307].’ Apparently these ‘dispatches’ are an invention of Masséna’s own. It is incredible that any British officer can have written such stuff after a combat of which every man present was particularly proud, and in which the losses had been incredibly small, considering the risks that had been run. Four officers, not twenty-four, had been killed, and one made prisoner. Instead of being in ‘complete rout’ the Light Division had retired at leisure and unmolested, without leaving even a wounded man or a single cart behind.

[304] For an interesting description of this incident, see George Napier’s autobiography, p. 131.

[305] Thirty-six killed, 189 wounded, 83 missing. See Tables in Appendix.

[306] Martinien’s invaluable lists show 7 officers killed and 17 wounded, which at the normal rate of 22 men per officer, exactly corresponds to the actual loss of 117 killed and 410 wounded (Koch, vii. 118).

[307] It is a curious fact that in the draft of Masséna’s dispatch in the _Archives du Ministère de la Guerre_, we actually catch him in the act of falsifying returns. There is first written ‘Nous leur avons pris 100 hommes et deux pièces de canon. Notre perte a été de près de 500 hommes tant tués que blessés.’ Then the figures 100 are scratched out and above is inserted ‘un drapeau et 400 hommes,’ while for the French loss 500 is scratched out and 300 inserted. Ney, whose dispatch was lying before Masséna, had honestly written that Craufurd ‘a été chassé de sa position avec une perte considérable de tués et de blessés, nous lui avons fait en outre une centaine de prisonniers.’ Ney reported also a loss of about 500 men, which Masséna deliberately cut down to 300. Belmas (iii. 379) has replaced the genuine figures in his reprint of Masséna’s dispatch, though both the draft in the _Archives_ and the original publication in the _Moniteur_ give the falsifications. Masséna says nought of the check at the bridge, though Ney honestly wrote ‘au delà du Coa, une réserve qu’il avait lui permis de se reconnaître, et il continue sa retraite sur Pinhel la nuit du 24.’ As to the guns captured, it was perfectly true that some cannon were taken that day, but not in fighting, nor from Craufurd. The governor of Almeida was mounting two small guns (4-pounders) on a windmill some way outside the glacis. They had not been got up to their position, but were lying below--removed from their carriages, in order to be slung up more easily on to the roof. The mill was abandoned when Ney came up, and the dismounted cannon fell into his hands. He said not a word of them, any more than he did of the imaginary flag alleged by Masséna to have been captured. But the Prince of Essling brought in both, to please the imperial palate, which yearned for British flags and guns. His dispatch, published some weeks later in the _Moniteur_, came into Craufurd’s hands in November, and provoked him to write a vindication of his conduct, and a contradiction of ‘the false assertions contained in Marshal Masséna’s report of an action which was not only highly honourable to the Light Division, but positively terminated in its favour, notwithstanding the extraordinary disparity of numbers. For a corps of 4,000 men performed, in the face of an army of 24,000, one of the most difficult operations of war,--a retreat from a broken and extensive position over one narrow defile, and defended during the whole day the first defensible position that was to be found in the neighbourhood of the place where the action commenced.’ For the whole letter see Alex. Craufurd’s _Life of Craufurd_, pp. 140-1.

Wellington was justly displeased with Craufurd for accepting this wholly unnecessary combat: if the Light Division had been withdrawn behind the Coa on the 22nd, as he had advised, no danger would have been incurred, and the bridge might have been defended without the preliminary retreat to the water’s edge. Yet so great was the confidence in which Craufurd was held by Wellington, that their correspondence shows no break of cordiality or tension of relations during the ensuing days[308], though unofficially the divisional general was aware that the Commander-in-Chief had disapproved his action, and felt the blame that was unspoken in the keenest fashion[309]. There was another British general involved in a serious degree of culpability on the 24th: this was Picton, who hearing at his post of Pinhel the firing in the morning, rode up to the bridge of the Coa; there he met Craufurd, who was just preparing to resist Ney’s attempt to cross the river. Picton was asked to bring up the 3rd Division in support, which could have been done in less than three hours, but roughly refused, saying apparently that Craufurd might get out of his own scrape. The generals parted after an exchange of some hard words, and Picton rode back to order his division to get ready to retreat, having committed one of the greatest military sins, that of refusing to support a comrade in the moment of danger, because he did not choose to compromise his own troops[310].

[308] See the letter to Craufurd in the _Dispatches_, dated July 26 and 27. His letter to Lord Liverpool of July 25 offers, indeed, excuses for Craufurd. But in that to Henry Wellesley of July 27, and still more in that to his relative Pole of July 31, he expresses vexation. ‘I had positively forbidden the foolish affairs in which Craufurd involved his outposts, ... and repeated my injunction that he should not engage in an affair on the right of the river.... You will say in this case, “Why not accuse Craufurd?” I answer, “Because if I am to be hanged for it, I cannot accuse a man who I believe has meant well, and whose error was one of judgement, not of intention.”’

[309] See _Craufurd’s Life_, pp. 149-50.

[310] This interview was denied by Robinson in his _Life of Picton_ (i. 294) on the mere allegation of some of Picton’s staff that they had not heard of it, or been present at it. But the evidence of William Campbell, Craufurd’s brigade-major, brought forward by Napier at Robinson’s challenge, is conclusive. See Napier, vi. pp. 418-19, for the ‘fiery looks and violent rejoinders’ witnessed by Campbell. Picton had been specially ordered to support Craufurd if necessary. See _Wellington Dispatches_, v. pp. 535 and 547.

Having cleared the country-side beyond the Coa by pressing back the Light Division, and having ascertained by a reconnaissance that Picton had evacuated Pinhel on the night of the 25th, Masséna was able to sit down to besiege Almeida at his leisure. The investment was assigned to Ney and the 6th Corps, while Junot and the 8th Corps were brought up from the Agueda, and placed in the villages behind and to the right of the besieged place, so as to be able to support Ney at a few hours’ notice. The extreme steepness of the banks of the Coa during its whole course rendered it most unlikely that Wellington would attempt the relief of Almeida by a direct advance. He would have had to force a passage, and the Coa, unlike the Agueda, has very few fords. Its only two bridges, that opposite Almeida, and that higher up at Castello Bom, were held in force by the 6th Corps. The siege however might not improbably prove long. Almeida was in far better repair than Ciudad Rodrigo, and had less defects. The little town is situated on the culminating knoll of an undulating plateau, a very slight eminence, but one which was not commanded by any higher ground as Rodrigo was by the two Tesons. The outline of the place is almost circular, and exactly fits the round knoll on which it stands. It has six bastions, with demi-lunes and a covered way. There is a dry ditch cut in the solid rock, for Almeida lies on a bare granite plateau, with only two or three feet of earth covering the hard stratum below. It was well armed with over 100 guns, forty of which were 18-pounders or still heavier. It had casemates completely proof against bomb fire, and large enough to cover the whole garrison. This, as has been already said, consisted of one regular regiment, the 24th of the Line over 1,200 strong, and the three militia regiments of Arganil, Trancoso, and Vizeu--in all some 4,000 infantry, with a squadron of the 11th cavalry regiment and 400 gunners. The governor was William Cox--an English colonel and a Portuguese brigadier; he had with him five other English officers, all the rest of the garrison being Portuguese. There was an ample store both of food and of ammunition, which Wellington had been pouring in ever since the siege of Ciudad Rodrigo began. Only two serious defects existed in the place: the first was that its glacis was too low, and left exposed an unduly large portion of the walls[311]. The second, a far worse fault, was that the grand magazine was established in a rather flimsy mediaeval castle in the centre of the town, and was not nearly so well protected as could have been desired. Nevertheless Wellington calculated that Almeida should hold out at least as long as Ciudad Rodrigo, and had some hope that its siege would detain Masséna so much that the autumn rains would set in before he had taken the place, in which case the invasion of Portugal would assume a character of difficulty which it was far from presenting in August or September.

[311] This came from the extreme hardness of the soil, which induced the builders of the 18th-century enceinte to put less earth into the glacis than was needed, since it had to be scraped up and carried from a great distance, owing to the fact that the coating of soil all around is so thin above the rock.

For some days after the investment of Almeida had been completed the 6th Corps remained quiescent, and made no attempt to break ground in front of the place. Ney was waiting for the Grand Park and the train, which had now started in detachments from Ciudad Rodrigo, but were advancing very slowly on account of the lack of draught animals. For a moment Wellington thought it possible that the enemy was about to mask Almeida, and to advance into Portugal with his main army without delay[312]. This hypothesis received some support from the facts that Junot had moved up from the Agueda, and that Reynier had shown the head of a column beyond the Pass of Perales. This last appeared a most significant movement; for if the 2nd Corps was about to march up from the Tagus to join Masséna, the deduction was that it was required to join in a general invasion, since it was clear that it was not needed for the mere siege of Almeida. Wellington accordingly wrote to urge Hill to keep a most vigilant eye on Reynier, and to be ready to move up to the Mondego the moment that it was certain that his opponent had passed the Sierra de Gata and linked himself to the main French army. As a matter of fact there was, as yet, no danger from Reynier. The advance of one of his flanking detachments to Navas Frias beyond the Pass of Perales, and a raid made upon Penamacor on July 31 and upon Monsanto on August 1, by another, were pure matters of foraging and reconnaissance. Reynier had no orders to move up his whole force to join Masséna, and was only amusing himself by demonstrations. His actions became most puzzling to Wellington when, a few days later, he called back all the troops that had moved northward, and concentrated his force at Zarza la Mayor, on the road to Castello Branco, so as to threaten once more to invade Central Portugal by the line of the Tagus. This was no device of his own, but the result of a dispatch from Masséna dated July 27, ordering him to keep more to the south for the present, to threaten Abrantes, and to afford Hill no chance of joining Wellington.

[312] Wellington to Hill, Alverca, July 27, ‘There is not the smallest appearance of the enemy’s intending to attack Almeida, and I conclude that as soon as they have got together their force, they will make a dash at us, and endeavour to make our retreat as difficult as possible.’

Reynier’s feints meanwhile had given Hill some trouble; the appearance of a northward move on the part of his adversary had caused the British general to make ready for a parallel march on Fundão and Guarda, so as to connect himself with his chief. He transferred his head quarters first from Castello Branco to Sarzedas, and then from Sarzedas to Atalaya, at the foot of the pass that leads to the Mondego valley, intending to cross the mountains the moment that Reynier had passed over the Perales defiles with his main body. But seeing the 2nd Corps unexpectedly turning back and concentrating at Zarza, Hill also retraced his steps, and lay at Sarzedas again from August 3rd till September 21st, with his advanced guard at Castello Branco and his cavalry well out to the front along the Spanish frontier, watching every movement of the 2nd Corps. During this time of waiting the Portuguese cavalry of his division had two small but successful engagements with Reynier’s horse, of whom they cut up a squadron on the 3rd of August near Penamacor and another on the 22nd at Ladoeiro, when two officers and sixty men of the Hanoverian _Chasseurs à Cheval_ were killed or taken[313].

[313] For details of this combat see Foy’s observations on p. 97 of his _Vie Militaire_, ed. Girod de L’Ain.

Wellington’s doubts as to Masséna’s intentions in the first days of August were provoked not merely by the movements of the 2nd Corps, but by a demonstration made on an entirely new front by General Serras, the officer who had been left with an unattached division to hold the plains of Leon, when Junot and the 8th Corps went off to join the main army on the Agueda. In obedience to Masséna’s orders, on July 27 Serras collected at Benavente as much of his division as could be spared from garrison duty, and moved forward to threaten the frontier of the Tras-os-Montes, far to the north of Portugal. He advanced with some 5,000 men as far as Puebla de Senabria, from which on July 29 he drove out a small Spanish force under General Taboada--the weak brigade which Echevarria had formerly commanded. Silveira immediately collected all the Portuguese militia of his district at Braganza, and prepared to defend the frontier. But Serras unexpectedly turned back, left a battalion of the 2nd Swiss Regiment and a squadron of horse in Puebla de Senabria, and returned to Zamora. The moment that he was gone Silveira and Taboada united their forces, attacked this small detached force, routed it, and shut it up in the town on August 4. It was forced to surrender some six days later, about 20 officers and 350 men, all that remained of 600, being made prisoners. Serras, who had hurried back when he heard of Silveira’s offensive movement, was too late by twelve hours to save his men, and found Puebla de Senabria empty, for the allies had gone off with their prisoners and taken to the mountains. He then retired to Benavente, and Taboada reoccupied Puebla de Senabria, where he was not again disturbed. Serras soon after was drawn away to the north-east by the demands of Bonnet, whose communications with Santander had once more been cut by Porlier’s roving Asturian bands. He called on his colleague to attack this partisan force in the rear, and while Serras was hunting it at Potes and Alba, in the Cantabrian Hills, Northern Portugal and Galicia were left undisturbed in September[314].

[314] For a narrative of these obscure campaigns see Schaller’s _Souvenirs d’un officier Fribourgeois_, pp. 29-37.

While glancing at the subsidiary operations in this remote corner of Spain, it may be worth while to note, as a proof of the slight hold which Bonnet and Serras possessed on their allotted districts, that on June 7 Mahy threatened Astorga, while the Asturian bands of Colonel Barcena, eluding Bonnet, came down into the plains by the Pass of Pajares and surprised Leon[315]. They got into the town by escalade at night, held it for two days, and only evacuated it when Serras came up in strength on June 9. Provoked at this bold adventure, Bonnet made his last attempt to conquer Western Asturias, and so to destroy the indefatigable and evasive partisans in his front. He forced his way across the Narcea and the Navia, and his vanguard had reached Castropol, on the Galician border, upon July 5, when he heard to his disgust that the enemy had slipped behind him. Barcena was threatening his base at Oviedo, while Porlier’s band, carried round by English ships, had landed near Llanes and cut the communication with Santander. These clever moves brought Bonnet back in haste: he evacuated Western Asturias, called up Serras to his aid, and was engaged in August and September in the hunt after Porlier which we have already mentioned[316].

[315] See ibid., pp. 32-3.

[316] For a narrative of these interesting but obscure movements, see Schepeler, iii. 596-9. It is impossible to give a full account of them here, but necessary to mention them, to show the Sisyphean character of Bonnet’s task.

But to return to the main focus of the war in the North. On August 15th Ney’s troops, having at last received the siege-train and a good supply of munitions from Ciudad Rodrigo and Salamanca, broke ground in front of Almeida. Wellington was much relieved at the news, as it was now clear that Masséna was about to besiege the place, and not to mask it and march forward into Portugal. The front which the engineers of the 6th Corps had chosen for attack was that facing the bastion of San Pedro on the south-east front of the town. The first parallel was drawn at a distance of only 500 yards from the walls; it was found very difficult to complete, owing to the shallowness of the earth, and had to be built with gabions and sandbags rather than to be excavated in the rocky subsoil. In many places outcrops of stone came to the surface, and had actually to be blasted away by the sappers, in order to allow of a trench of the shallowest sort being formed. It was clear that the construction of approaches towards the town would present the greatest difficulties, since there was little earth in which to burrow. Between the 17th and the 24th no less than eleven batteries were constructed along the first parallel. They were armed with more than fifty heavy guns, for there was artillery in abundance; in addition to the old siege-train many of the Spanish guns taken in Ciudad Rodrigo had been brought forward. The Portuguese kept up a vigorous but not very destructive fire all the time; but on the 24th they succeeded in preventing the commencement of a second parallel, driving out the workmen before they could cover themselves in the stony ground. At six o’clock on the morning of the 26th August the batteries were all completed and opened fire. Several quarters of the town were in flames before the afternoon, and the guns on the three bastions attacked were unable to hold their own against the converging fire directed on them. But no serious damage had been done to the defences, and the governor was undismayed. At seven o’clock in the evening, however, a fearful disaster occurred--one in its own way unparalleled in magnitude during the whole Peninsular War. The door of the great magazine in the castle had been opened, in order to allow of the sending out of a convoy of powder to the southern ramparts, where the artillery had been hard at work all day. A leaky barrel was handed out, which left a trail of powder behind it along the ground; it was being fixed to the saddle of a pack-ass when a French bomb fell in the courtyard of the castle. In bursting, the bomb chanced to ignite the train; the spark ran along it and exploded another barrel at the door of the magazine, which was still open[317]. This mischance fired the whole store, and in two seconds the castle, the cathedral at its side, and the whole central portion of the town had been blasted out of existence. ‘The earth trembled,’ wrote a French eye-witness, ‘and we saw an immense whirlwind of fire and smoke rise from the middle of the place. It was like the bursting of a volcano--one of the things that I can never forget after twenty-six years. Enormous blocks of stone were hurled into the trenches, where they killed and wounded some of our men. Guns of heavy calibre were lifted from the ramparts and hurled down far outside them. When the smoke cleared off, a great part of Almeida had disappeared, and the rest was a heap of débris[318].’ Five hundred of the garrison perished, including nearly every man of the two hundred artillerymen who were serving the guns on the front of attack. Some inhabitants were killed, but not many, for the majority had taken refuge in the casemates when the bombardment began that morning. It was the unfortunate soldiers who were manning the walls that suffered.

[317] This version of the cause of the disaster is given by Soriano da Luz (iii. 73) from the mouth of an artillery officer (one José Moreira) who had it from the only man in the castle-yard who escaped. This soldier, seeing the train fired, jumped into an oven-hole which lay behind him, and chanced not to be killed.

[318] Sprünglin’s _Journal_, pp. 444-5.

[Illustration: COMBAT OF THE COA. JULY 24TH, 1810.]

Fearing that the French might seize the moment for an escalade, General Cox ran to the ramparts and, assisted by a Portuguese artillery officer, loaded and fired into the trenches some of the few guns on the south front which were not disabled. He turned out the whole garrison, and kept them under arms that night, lying behind the walls in expectation of an assault which never came. The morning light enabled him to realize the full extent of the disaster; the bastions and curtains had suffered little, the shell, so to speak, of the town was still intact, and the casemates had stood firm, but everything within the enceinte was wrecked. Only five houses in the place had kept their roofs: the castle was a deep hole, like the crater of a volcano: the streets were absolutely blocked with ruins, so that there was no going from place to place save along the ramparts.

There were still 4,000 men under arms; but the officer commanding the artillery reported that thirty-nine barrels of powder, and a few hundred rounds in the small expense-magazines on the ramparts, were all that had escaped the explosion. That is to say, there was not powder in the place to keep up a reply for one day to the batteries of the besieger. The infantry had 600,000 cartridges in their regimental stores (150 rounds per man) but that was of no use for the heavy guns. Moreover, more than half the gunners had perished in the disaster of the previous night--only 200 were left to man nearly 100 guns that were still serviceable. It was clear that Almeida was doomed, since it could not defend itself without powder: but there was a chance that Wellington, whose outposts must have heard and seen the explosion, might think it worth while to dash forward and endeavour to save the garrison during the next twenty-four hours. Therefore Cox resolved to protract his resistance as long as was possible, to give his chief the option of fighting if he should so please. But the defence could not be prolonged for more than a day or two at the most.

At nine on the morning of the 27th Masséna sent in his aide de camp Pelet to demand the surrender of the fortress. Cox had him blindfolded, and taken into a casemate for their interview, so that he might not be able to judge of the awful effects of the explosion. The usual haggling followed--the French officer threatened that the place should be escaladed at once, and the garrison put to the sword. The governor replied that his walls were intact, that he could still defend himself, and that the ‘deplorable accident’ had not appreciably diminished his resisting power[319]. But he finally consented to send out an officer to the French camp to negotiate for terms. All this was merely done to gain time, and the semaphore on the western ramparts was signalling desperate messages to Wellington all the morning.

[319] There is a good account of this interview in Sprünglin’s _Journal_, p. 445, the diarist having accompanied Pelet into the town.

Cox’s attempt to gain time was fruitless, for a reason that he had not foreseen. The garrison was hopelessly demoralized, knew that it must surrender, and did not see why it should expose itself to another day’s bombardment for a lost cause. During the conference in the casemate General d’Alorna and other Portuguese officers on Masséna’s staff came out of the trenches, and boldly presented themselves at the foot of the walls, calling to their compatriots above and beseeching them to accept the good terms offered, and not to risk their lives for Wellington, who would abandon them just as he had abandoned Herrasti at Ciudad Rodrigo. The officers on the ramparts ought to have driven the renegades away, by shots if necessary; but, far from doing so, they entered into long conversation with them and approved their arguments. D’Alorna recognized some old acquaintances among the regulars, and pledged his word to them that an assault was imminent, and that they were doomed if they made any resistance. What was still more unlucky for Cox was that the officer whom he sent out to the French camp to treat, Major Barreiros of the artillery, was one of those who were most convinced that further defence was fruitless; he divulged the hopeless state of the place to the Marshal, and bade him press his attack without fear, for the garrison would not fight. He himself remained at the French head quarters, and did not return to Almeida. Masséna, therefore, sent back a blank refusal of all Cox’s demands and conditions, and ordered the bombardment to recommence at seven in the evening, while approaches were thrown out from the second parallel towards the ramparts. A feeble musketry fire alone replied.

The renewal of the bombardment speedily brought matters inside the place to a head. A deputation of Portuguese officers, headed by Bernardo Da Costa, the second in command, visited Cox and informed him that further resistance was madness, and that if he did not at once hoist the white flag they would open the gates to the enemy. The Governor was forced to yield, and capitulated at eleven o’clock on the night of the 27th. Masséna granted him the terms that the regular troops should be sent as prisoners to France, while the three militia regiments should be allowed to disperse to their homes, on giving their parole not to serve again during the war.

On the morning of the 28th the garrison marched out, still 4,000 strong, its total loss during the siege having been some 600--nearly all destroyed in the explosion. The French had lost fifty-eight killed and 320 wounded during the operations. The capitulation was no sooner ratified than it was violated: instead of dismissing the militia and marching off the regulars towards France, Masséna kept them together, and set the renegades d’Alorna and Pamplona to tempt them to enter the French service. The officers were promised confirmation of their rank, the men were invited to compare the relative advantages of prison and of joining the victorious side and keeping their liberty. The arguments of the traitors seemed to prevail; almost the whole of the regulars and 600 of the militia signified their consent to enlist with the enemy. The rest of the militia were turned loose, but d’Alorna was able to organize a brigade of three battalions to serve the Emperor as the ‘Second Portuguese Legion[320].’ But the intentions of these docile recruits were quite other than Masséna had supposed. They had changed their allegiance merely in order to escape being sent to France, and while left unguarded during the next three days, absconded in bands of 200 or 300 at a time, officers and all, and kept presenting themselves at Silveira’s and Wellington’s outposts, for a week. The French, undeceived too late, disarmed the few men remaining in the camp, who were packed off to France, to rejoin Cox and the half-dozen officers who had loyally refused to accept d’Alorna’s offers[321]. Wellington had been somewhat alarmed when the first news of the adhesion of the garrison to the French cause reached him, fearing that it implied serious disaffection in the whole Portuguese army[322]. He was soon undeceived on this point, as the troops gradually streamed in to his camp[323]. But he was then seized with grave doubts as to whether he could, consistently with military honour, accept the service of these perjured but patriotic people. ‘It was well enough for the private men, but highly disgraceful to the character of the officers[324],’ he observed, and was pondering what should be done, and proposing to cashier the officers, when he received a proclamation from the Regency approving the conduct of the deserters and restoring them to their place in the army. Finally he resolved that, since Masséna had obviously broken the capitulation by his

## action, it might be held that it was not binding on the prisoners,

and ordered the 24th Regiment to be re-formed at its head quarters at Braganza. The militia he dismissed to their homes, on the scruple that the French had let some, though not all of them, go free after the capitulation[325]. There remained the problem of what was to be done with the Portuguese officers who had played a treacherous part on the 27th, especially Barreiros, the negotiator who had betrayed the state of the town to Masséna, and Da Costa, who had headed the mutinous deputation to Cox, which forced him to surrender. Finally, their names were included with those of the officers who had served on Masséna’s Portuguese staff during the campaign, in a great indictment placed before a special commission on traitors (called a _Junta de Inconfidencia_), which sat at Lisbon during the autumn. All from d’Alorna downwards were declared guilty of high treason, and condemned to death on December 22, 1810, but only two were caught and executed--João de Mascarenhas, one of d’Alorna’s aides de camp, and Da Costa, the lieutenant-governor of Almeida. The former was captured by the Ordenança while carrying Masséna’s dispatches in 1810, and the latter was apprehended in 1812; Mascarenhas died by the garotte, Da Costa was shot. Of the others, some never returned to Portugal, the others were pardoned at various dates between 1816 and 1820[326].

[320] The First Portuguese Legion, which served against Austria in 1809, was composed of the troops drafted out of the Peninsula by Junot in 1808 during his domination at Lisbon.

[321] D’Urban’s diary reports that 450 men and 18 officers of the 24th of the Line came in between the 2nd and 4th of September to Silveira’s outposts; a still larger number reached Wellington’s.

[322] D’Urban has most gloomy remarks on the subject in his diary, under the date Aug. 30.

[323] To Chas. Stuart, from Celorico, Aug. 31.

[324] To Chas. Stuart, from Celorico, Sept. 11.

[325] Wellington to Masséna, Sept. 24. ‘Votre excellence s’est engagée que les officiers et les soldats de la milice retourneraient chez eux: malgré cet engagement vous en avez retenu 7 officiers et 200 soldats de chaque régiment, pour en faire un corps de pionniers. La capitulation d’Almeida est donc nulle, et je suis en droit d’en faire ce que je voudrais. Mais je puis vous assurer qu’il n’y a pas un seul soldat de la milice qui était en Almeida au service.’

[326] For details of all this, including the curious terms of the Portuguese sentence for high treason, see Soriano da Luz, iii. 80-109, and 719-22. The attempts to exculpate Barreiros seem inadequate. Da Costa was shot, not for treason, but for cowardice and mutiny.

During the siege of Almeida, the British army had been held in a position somewhat less advanced and more concentrated than that which it had occupied in July. Wellington had brought back his head quarters from Alverca to Celorico, where he had the Light Division under his hand. A few miles behind, on the high-road running down the south bank of the Mondego, was the 1st Division, at Villa Cortes. Picton and the 3rd Division had been drawn back from Pinhel to Carapichina, but Cole and the 4th remained firm at Guarda. The Portuguese brigades of Coleman and A. Campbell were at Pinhanços, that of Pack at Jegua. The whole of the cavalry had come up from the rear to join the brigade that had recently operated under Craufurd’s orders. They now lay in a thick line of six regiments from in front of Guarda, through Alverca and Freixadas to Lamegal. Thus the whole army was concentrated on a short front of fifteen miles, covering the watershed between the Coa and the Mondego, and the bifurcation of the roads which start from Celorico, down the two banks of the last-named river. The French held back, close to Almeida, with a strong advance guard at Pinhel, and occasionally raided the low country towards the Douro, in the direction of Villanova de Fosboa and Castel Rodrigo. About August 19 Wellington moved forward a day’s march, the front of his infantry columns being pushed up to Alverca and Freixadas, on a false rumour that Masséna was leaving the 6th Corps unsupported at Almeida, and had drawn back Junot into the plains of Leon. If this had been the case, Wellington intended to make a push to relieve the besieged fortress. But he soon discovered that the report was baseless, and that the 8th Corps was still on the Azava and the Agueda, wherefore he halted, and was still lying twelve miles in front of Celorico when the noise of the explosion at Almeida, and the cessation of fire on the next day, betrayed the fact that the place had fallen, and that there was no longer any reason for maintaining a forward position. On the night of the 28th, therefore, the whole army was drawn back once more to the strong line between Guarda and Celorico, and arrangements were made for a further retreat, in case the French should follow up the capture of Almeida by an instant and general advance. Masséna seemed at first likely to make this move: on September 2 a brigade of infantry and 1,200 horse drove in the British cavalry outposts to Maçal de Chão only five miles in front of Celorico. Looking upon this as the commencement of the serious invasion of Portugal, Wellington sent back his infantry to Villa Cortes, Pinhanços, and Moita, far down the high-road on the south of the Mondego, and bade Cole draw in the 4th Division from Guarda to San Martinho, under the north side of the Serra da Estrella. Only cavalry were left at, and in front of, Celorico and Guarda. This retreat shows that Wellington was fully convinced that the French would advance along the high-road to the south of the Mondego, where he intended to stand at bay on the Alva, behind the entrenchments of Ponte de Murcella.

The main point of interest at this moment was the movements of the French 2nd Corps, which still lay in cantonments at Zarza and Coria in front of Hill. Guarda being no longer held in force, it was clear that Hill could not safely join the main army by the road Atalaya-Fundão-Guarda, if Reynier moved up by the Pass of Perales to Alfayates and Sabugal. Wellington began an anxious daily correspondence with Hill, giving him a new line of march by Sobreira Formosa, Villa d’el Rei and Espinhal, for his junction, but ordering him to be sure that he did not move till Reynier had thoroughly committed himself to the transference of his whole force to the north of the Sierra de Gata. For it would be disastrous if feints should induce the British detaining force to leave Villa Velha and Abrantes uncovered, and Reynier should turn out to have selected them as his objective, and to be meditating an invasion along the Castello Branco line[327]. It was even possible, though not likely, that Masséna might bring up the 6th and 8th Corps to join Reynier, instead of bringing Reynier across the mountains to join them[328]. But every contingency had to be provided against, the unlikely ones as well as the likely. As a corollary to Hill’s march, that of Leith had also to be arranged; he must wait at Thomar till it was certain that the 2nd Corps had moved, but the moment that certainty was obtained, must march for Ponte de Murcella, and join the main army, if the French had gone north, but support Hill on the Castello Branco road if they had taken the other, and less probable, course.

[327] See Wellington to Hill of Aug. 31, Sept. 1, Sept. 4, Sept. 6. The Commander-in-Chief was much worried by a false rumour that Reynier was already in force at Sabugal on Aug. 31, and then by an equally false one that the whole 2nd Corps had marched south towards the Tagus, and was about to cross it near Alcantara (see the letter to La Romana of Sept. 6). As a matter of fact, Reynier made no definite move from Zarza till Sept. 10, though he had made feints, in both the directions indicated, with small forces.

[328] That this possibility was in Wellington’s mind is shown by the letter to La Romana of Sept. 6, from Gouvea, in which he writes, ‘Vous aurez appris les mouvements du corps de Regnier de la part du Général Hill. Ou l’ennemi va faire le mouvement sur notre droite (dont je vous ai écrit) ou il va faire le siège de Badajoz. On dit que du canon a passé d’Almeida à Sabugal, et de là vers Regnier, mais je ne sais pas si c’est vrai, ou si c’est du canon de siège.... Vous savez ce qu’il faut faire si on se met entre nous deux, en passant le Tage à Villa Velha, ou au-dessous de la jonction.’

NOTE ON ALMEIDA AND THE BRIDGE OF THE COA

The small circular town of Almeida has never recovered from the disaster of 1810. The population does not fill up the area within the walls: open spaces are frequent, and some of the more important buildings--especially the old palace of the governor--stand in ruins. Others show solid seventeenth- or eighteenth-century masonry on the ground floor, and flimsy modern repairs above, where the upper stories were blown away by the explosion. The cathedral has never been properly rebuilt, and is a mere fragment. The railway passes twelve miles south of Almeida, so that the place has had no chance of recovery, and remains in a state of decay. The walls stand just as they were left after Wellington’s hasty repairs in 1811. The vast bomb-proof shelters repeatedly mentioned in narratives of the siege are still visible, damp but intact. The surrounding country-side is a low, rolling, treeless upland--the edge of the vast plains of Leon. It contrasts very strongly with the hilly and picturesque scenery that is reached when once the Coa has been passed.

From Almeida the ground slopes down sharply to the place of Craufurd’s celebrated skirmish. The town is not visible after the first mile of the descent towards the deep-sunk gorge through which the Coa cuts its way. The high-road is very bad for artillery, being steep, filled with great stones, and in many places shut in by high banks, which tower above it and make it narrow. The sharp turn at the end, where it descends to the river with a sudden twist, must have been specially tiresome to a force with cavalry and guns, compelled to a hasty retreat. All the slopes about the road are cut up into small fields by high walls of undressed stone, without mortar, such as are seen on Cotswold. The bridge is not visible till the traveller approaching from Almeida has got down to the level of the river: it is completely masked by the high fir-clad knoll described in the text, so long as he is descending the slope above. From the point where the road swerves aside, to avoid this knoll, there is a rough goat-track down to the still invisible bridge, but this is not available for guns, horses, or _formed_ infantry, only for men scrambling individually.

The two-arched bridge is seventy yards long; it crosses the Coa diagonally, with a curious twist in the middle, where there is a little monument recording the reparation of the structure by John VI, in the days after the war was over. There is no mention of Craufurd’s fight in the inscription--only a laudation of the King. The bridge crosses the river at a sort of gorge--the place where the rocks on the two sides come nearest. Hence the stream runs under it very fiercely, being constricted to far less than its normal breadth. Up-stream the channel broadens and the passage looks much less formidable, but for some distance on each side of the bridge the river is very rapid, darting between rocks and boulders. The little corner where the few French who passed the bridge found a small angle of dead-ground can be easily identified. It was just to their right after crossing. All the rest of the ground on the west bank could be thoroughly searched by the British guns, which were placed a few hundred yards up the road on the left hand, as well as by the fire of the infantry ensconced among rocks and boulders above the bridge.

Of the French attacking force during the early part of the skirmish, those who were on their left, nearest the river and opposite the 52nd, had far easier ground to cover than those on the right, opposite the 43rd. It is not so high or rough, and less cut up by stone walls. Hence the stress on the 52nd ought to have been the heavier--yet they lost only 22 men to the 129 of the 43rd. The damage to the latter must have been caused partly when the cavalry got in among them, just as the retreat began, partly when they stormed the knoll to cover the retreat of the 52nd.

The scene of the fight is most picturesque on a small scale--one of the prettiest corners in Portugal, all rock, fir-trees, and rushing water.--[Notes made on the spot on April 14, 1906.]

SECTION XX

OPERATIONS IN THE EAST AND SOUTH OF SPAIN DURING THE SPRING AND SUMMER OF 1810

##