The first day’s fighting in the Pyrenean passes could not be called satisfactory either to Wellington or to Soult. The former had lost both the defiles in which he had intended to make his first stand, and had lost them in a very tiresome fashion—he thought that Maya might have been held at least for twenty-four hours, if there had been a divisional general on the spot to direct the defence: while Roncesvalles had not been forced, but abandoned by Cole, who could certainly have made a longer resistance, if only the orders sent to him had been obeyed. It was, above all things, necessary to gain time for the concentration of the army, and a precious day had been lost—and need not have been lost.

But Soult can have been no better pleased: time, to him also, meant everything; and the orders which he had issued to his lieutenants had presupposed an easy triumph by surprise in the early morning, with a forward march in the afternoon. Instead of this he had won by nightfall a bare foothold on the summit of each pass, after much fighting of an unsatisfactory sort. He, too, had lost a day; and it was only on the morrow that he discovered that both at Maya and at Roncesvalles the enemy had slipped away in the dark, leaving to him the power to debouch from the defiles.

Nevertheless, the Marshal sent on the morning of the 26th a very flamboyant message of victory to his master the Emperor, who then lay at Mayence. Both Maya and Roncesvalles had been forced, D’Erlon had captured five guns and many hundreds of prisoners at the former pass: he himself hoped to be at Pampeluna, and to have raised its siege, by the 27th. These news were sent on from Bayonne by semaphore to Paris and the Rhine, and reached Napoleon on August 1st. At the same time, and by the same rapid method of transmission, arrived General Rey’s report of his successful repulse of the assault of July 25th. It is worth while to turn away from solid history for a moment, in order to see how the Imperial editor of the Moniteur utilized this useful material for propaganda. He first wrote to Clarke the Minister of War: ‘We can now give the public some account of affairs in Spain. The Vittoria business and the King must not be mentioned. The first note which you must put in the Moniteur should run as follows—“His Majesty has named the Duke of Dalmatia as his lieutenant-general commanding his armies in Spain. The Marshal took up the command on July 12, and made immediate dispositions for marching against the besiegers of Pampeluna and St. Sebastian.” After that put in General Rey’s first letter about the events of the 25th-27th. You had better make some small additions to the number of prisoners and of guns captured, not for French consumption but to influence European opinion. As I am printing General Rey’s dispatch in the Frankfort Journal, and have made some changes of this sort in it, I send you a corrected copy so that it may appear in the Moniteur in identical terms.’

The Emperor’s second letter to his Foreign Minister, the Duke of Bassano, sent from Dresden three days later, is even more amusing. ‘You had better circulate the news that in consequence of Marshal Soult’s victory over the English on July 25, the siege of St. Sebastian has been raised, and 30 siege-guns and 200 waggons taken. The blockade of Pampeluna was raised on the 27th: General Hill, who was in command at that siege, could not carry off his wounded, and was obliged to burn part of his baggage. Twelve siege-guns (24-pounders) were captured there. Send this to Prague, Leipzig, and Frankfort[897].’

This ‘intelligent anticipation of the future,’ for utilization in the armistice-negotiations going on with Austria, could not have been bettered. Unfortunately there arrived next day another semaphore message from Soult of the night of the 26th-27th. The Emperor has to warn Caulaincourt that yesterday’s propaganda will not stand criticism. ‘I have just got another “telegraphic dispatch” sent on by the Empress from Mayence, giving another communication from Soult, written 24 hours after the last, in which he said he would be at Pampeluna on the 27th. The enemy lost many men and seven guns. But nothing decisive seems to have happened. I am impatient for more news, in order to be able to understand in detail Soult’s dispositions, and to form from them a general idea of the situation[898].’

Alas for human ingenuity! Soult’s next dispatch, of July 29, was not to be of the sort that craved for publicity in the Moniteur, even with the most judicious editing.

But to return from Dresden to Biscay, and from the head-quarters of the Emperor to those of the ‘Sepoy General’ whom he had at last begun to recognize as capable of ‘des projets très sensés.’

If only Wellington had been at his head-quarters at Lesaca at 11 o’clock on the morning of July 25th, and if William Stewart had been on the spot at Maya, and had sent early news of D’Erlon’s attack, many things might have happened differently. Wellington would have had a long afternoon before him to concert operations, and would have possessed information to guide him in drawing up his scheme. Unfortunately he was absent—as we have seen—and only received at 6 o’clock a second-hand report from Lord Dalhousie at Echalar, to the effect that fighting was going on at Maya, with the unfortunate addition that D’Erlon had been repulsed—a most inaccurate summary of what had happened. Later on in the evening, not before 10 p.m., came Cole’s first dispatch from Roncesvalles, to say that he and Byng were heavily engaged at 1 o’clock with a large French force, and were holding their own. On these scanty data Wellington felt that no conclusions could be drawn—he wrote to Graham that there must be a great mass of French troops not yet discovered, which would come into action on some other point on the 26th, and that his policy would depend on where that force appeared—he could only account for 30,000 of Soult’s men so far. He did not commit himself to any definite guess as to the undiscovered part of the Marshal’s plan, but from his other correspondence it is clear that he suspected an attempt to relieve St. Sebastian by an attack on the lower Bidassoa—a very possible solution of the problem, but not the correct one[899].

Awaiting further developments, Wellington issued no more orders on the night of the 25th, save one to the Conde de Abispal, directing him to send one of his two infantry divisions from in front of Pampeluna to join Picton and Cole, and to keep the fortress blockaded by the other. The force thus taken away would be replaced by Carlos de España’s division, which was marching up from Burgos, and due to arrive on the 26th. In this dispatch Wellington asked the Conde to direct Mina to send up his infantry from Saragossa, and told him that he was intending to order to the front the British heavy cavalry brigades, now cantoned along the Ebro. No other movements were settled that night; but Wellington was aware that during his absence his Quartermaster-General, George Murray, had directed Lord Dalhousie to have the 7th Division massed at Echalar, prepared to move at an hour’s notice, and Charles Alten at Vera to have the Light Division got into a similar readiness. Either would be able to march off at dawn.

Somewhere late in the night[900] Wellington received more news, which made the situation clearer but more unsatisfactory. The true story of the Maya fighting came in from two sources: Hill sent a dispatch dated from Elizondo at some hour after 6 p.m., to say that on getting back from the Alduides he had found Stewart unable to hold the Pass, and had bidden him to retire. Stewart, who was wounded and unable to write, sent a verbal message, which came in about the same time, reporting that Hill had directed him to fall back on Elizondo and Berueta. The officers who brought this information stated that the French were in great force, and that the 2nd Division had been much cut up. No more reports arrived from Cole, so that the result of the Roncesvalles fighting remained unknown.

After what must have been a very short and disturbed night’s rest, Wellington was in the saddle by 4 a.m. on the 26th, and preparing to ride up the Bastan to visit Hill, and to ascertain the exact measure of the mishap at Maya. Before departing from Lesaca he gave his first definite orders in view of the events of the previous day[901]. Maya being lost, the 7th Division must fall back from Echalar to Sumbilla, on the road to Santesteban: the Light Division must retire from Vera to the west bank of the Bidassoa, and be ready to march either towards Yanzi or towards Santesteban, as might be necessary. Longa’s Cantabrians were to block the hill road from the Bidassoa to Oyarzun. Graham was told to hurry on the embarkation of the siege-train from St. Sebastian. Hill was to hold on as long as he could to the position at Irurita, in order to keep touch with the 6th Division, which was directed to feel towards him, and to be ready to join him if necessary. It was to push two of its three brigades to Legasa, on the road from Santesteban along the upper Bidassoa, which would bring them within eight miles of Hill’s proposed line of defence at Irurita. The third brigade of the 6th Division was to stand fast at Santesteban, where it would be in touch with Dalhousie, when the latter should have reached Sumbilla.

All these orders, as is obvious, are concerned only with the measures necessary to stop D’Erlon’s advance. None of them have any reference to the action of the other French force at Roncesvalles. Till news should come up from Cole and Picton, it was impossible to realize what was going on at that front, or whether the enemy was making his main attack in that direction. There might be still (as Wellington had guessed three days back) a violent demonstration towards Pampeluna, intended to distract a real attempt to relieve St. Sebastian.

And this state of ignorance with regard to the southern theatre of operations was destined to last till late in the afternoon. Either Cole and Picton themselves, or the officers to whom they entrusted their dispatches, were sadly lacking in a sense of the value of the prompt delivery of news. Wellington rode along the Bidassoa for many a mile, till he came on Hill still holding the position of Irurita, and entirely unmolested by the French. There were now in line the sadly reduced remnant of the British brigades which had fought at Maya, and da Costa’s and Ashworth’s Portuguese, with the three 7th-Division battalions which had saved Stewart from disaster. The total made up about 9,000 bayonets. Hill estimated[902] D’Erlon’s force at 14,000 men—a miscalculation, for even after the losses at Maya there were still 18,000 French in line. The immediate result of the error, however, was beneficial rather than otherwise, for Wellington considered that Hill was in no particular danger, and let him stand, while he himself rode southward towards the lofty Col de Velate, to seek for intelligence from the Pampeluna front in person, since his lieutenants had vouchsafed him none. He reached Almandoz, near the crest of the Pass, in the afternoon, and resolved to establish his head-quarters there for the night, as it was conveniently central between the two halves of his army.

Soon after his arrival Wellington, being much vexed at receiving no news whatever from the south, resolved to send the 6th Division toward Pampeluna by the Col de Velate as a matter of precaution—they were to march to Olague in the valley of the Lanz. The 7th Division was to close in, to take up the ground where the 6th had been placed, and cover Hill’s left flank[903]. That haste in these movements was not considered a primary necessity, is shown by the fact that Pack and Dalhousie were told that they need not march till the morning of the 27th. For the enemy’s surprising quiescence at the head of the Maya pass had reassured Wellington as to any danger on this side. If D’Erlon, indeed, possessed no more than 14,000 men, Hill with the aid of the 7th Division could easily take care of him. And the Light Division might still be left near Lesaca, as a reserve for Graham in case any new mass of French troops should take the offensive on the Bidassoa.

D’Erlon’s conduct on the morning of the 26th was explicable to himself, though inexplicable to his enemy. He had been engaged in a most bitter fight, in which he had lost 2,000 men and more. Two British divisions, so he wrote to Soult, were in front of him—the 2nd and the 7th. For he had taken Barnes’s brigade for the whole of Dalhousie’s unit—the effect of its desperate charge almost justified him in the hypothesis. These troops had been forced to a strategic retreat, but by no means put out of action. They must have been joined, ere now, by the Portuguese column which Darmagnac had sighted on its approach to Ariscun. But there were also troops on his right, of whom he must beware: he knew that Vera and Echalar had been held in strength, and Graham might send reinforcements in that direction, and assemble a heavy force on his flank. Hence he resolved to discover how matters lay by reconnaissances, before committing himself to the march down into the Bastan and then up the Col de Velate which his orders prescribed.

‘In my position on the pass of Maya,’ he wrote, ‘I had on my right all the forces which the enemy had in line as far as St. Sebastian. I had to be prudent, in order not to expose myself to a check in the Bastan, in which the enemy was holding the strongest position. I therefore determined to leave Abbé and Maransin in the pass, with orders to send out reconnaissances towards Santesteban, Echalar, and Mount Atchiola. They would profit by the halt to distribute the half-ration of food which had just come up from Ainhoue. I sent Darmagnac down the road to Ariscun, with orders to push a vanguard to Elizondo, and to explore towards the passes of Ispegui and Berderis, to see if there were any hostile force still on my left.’

An advance of six miles to Elizondo, and that by a mere advanced guard, was all the movement that D’Erlon made this day. It was not till the afternoon that he learnt, by Abbé’s reconnaissances, that there were still allied troops on his right—apparently the Light Division opposite Vera, and the 7th at Sumbilla—while Darmagnac reported that the eastern passes were clear, but that Hill was lying across the road beyond Elizondo in great strength. In the evening D’Erlon heard that Soult had forced the pass of Roncesvalles, and was about to advance: this success, he deduced, would make the enemy in front of him give way, in fear that his positions might be taken from behind. So he thought himself justified in ordering a general advance for the morning of the 27th—though Maransin was still to remain for a day at Maya, lest any allied force might move up from the west against the pass. Thus it came that for the whole of July 26th Hill was unmolested, and Soult’s plan for a rapid concentration round Pampeluna became almost impossible to carry out. A whole day had been wasted by D’Erlon, though he was not without his extenuating circumstances.

Wellington meanwhile received at Almandoz, probably at about 8 p.m., the long-expected news from the South. They were, as we know, most unsatisfactory: Cole reported from Linzoain, on the Roncesvalles-Pampeluna road, that he had been driven out of the pass by an army of 35,000 men or more, that he had not yet been joined by the 3rd Division, and was still retreating towards Zubiri, where he understood that Picton would meet him and take over the command. His view of the situation was shown by a remark that if he had not been superseded, and had been compelled to retreat past Pampeluna, he supposed that the road towards Vittoria would have been the right one to take[904]. This most exasperating dispatch only reached Wellington that night by mere chance. The officer bearing it was going to Lesaca, having no knowledge that Army Head-Quarters had left that place: at Lanz he happened to meet the cavalry brigadier Long, whose squadrons were keeping up the line of communication between the two halves of the Army. Hearing from the aide-de-camp of the sort of news that the letter contained, Long opened it and made a copy of it, which he sent to Sir Rowland Hill, before permitting the bearer to go on. Hill received the transcript at Berueta at 6 p.m., and very wisely forwarded it to Wellington at Almandoz. The original was carried on by Cole’s messenger to Santesteban, and did not reach Wellington that night.

Thanks to Long’s and Hill’s intelligent action, the Commander-in-Chief could grasp the whole unpleasant situation at 8 p.m. on the 26th. He sent orders to Picton at once, telling him that the enemy must at all costs be detained: that considering the force at his disposal, he ought to be able to check Soult for some time in front of Zubiri: that he would be joined at once by one of O’Donnell’s divisions from the Pampeluna blockading force, and shortly by reinforcements coming from the Bastan (the 6th Division). Wellington himself was intending to ride over to the right wing by the next afternoon. Till he should arrive, Picton must send reports every few hours[905]. Unfortunately, Cole and Picton had got things into an even worse state than could have been expected. Just as Wellington was drafting these orders for an obstinate rearguard action, they were at 8.30 p.m. preparing to evacuate the Zubiri position, and setting out on a night march for Pampeluna[906].

To explain this move we must go back to the state of affairs at Roncesvalles on the very foggy morning of July 26th. Cole, Byng, and Morillo had abandoned, as we have already seen[907], their position on Altobiscar and the Linduz under cover of the night, and had all fallen into the Pampeluna road, Ross’s brigade descending from the heights by the Mendichuri pass, the other three brigades and Morillo moving by the chaussée past the Abbey and Burguete. Anson’s brigade formed the rearguard, not having been engaged on the previous day. Morillo’s outlying battalion at the Foundry of Orbaiceta safely joined in by a hill path. Campbell’s Portuguese retired by the way that they had come, along the Path of Atalosti, but instead of returning to the Alduides followed a mule track to Eugui in the upper valley of the Arga.

Cole’s long column, after completing its night march, took a much-needed rest for many hours along the high road near Viscarret. It saw nothing of the French till the early afternoon, when an exploring party of chasseurs ran into the rearguard of Anson’s brigade.

What had Soult been doing between early dawn, when his outposts ascertained that there was no enemy in front of them, and three o’clock in the afternoon, when his cavalry rediscovered Cole? To our surprise we find that he had been attempting to repeat his error of the preceding day—that of sending a whole army corps along a rugged mule track, similar to the one on which Reille’s column had been blocked by Ross’s brigade. His original order on the 25th had been that Reille, after seizing the Linduz, should turn along the ‘crest of the mountains’, occupy the Atalosti defile, and push ever westward till he could threaten the Col de Velate, the main line of communication between the two sections of Wellington’s army. One would have supposed that the events of the 25th on the Linduz, where one British brigade had checked for a whole day Reille’s column of 17,000 men in Indian file, would have taught him the impracticability of such plans. But (as Soult’s malevolent critic, quoted already above, observed) when the Marshal had once got his plan drawn up on paper it was like the laws of the Medes and Persians, and must not be altered[908].

While Clausel was directed to use the chaussée and pursue Cole along the Pampeluna road past Roncesvalles, Burguete, and Espinal, Reille was once more ordered[909] ‘to follow the crest of the mountains to the right, and to try to take in the rear the hostile corps which has been holding the pass of Maya against Count D’Erlon.’ The itinerary seems insane: there was a mule track and no more, and Soult proposed to engage upon it a column of 17,000 men, with a front of one file and a depth of at least six miles, allowing for the battalion- and brigade-intervals. The crest was not a flat plateau, but an interminable series of ups and downs, often steep and stony, occasionally wooded. Campbell’s brigade had traversed part of it on the 25th, but to move a brigade on a fine day is a different thing from moving an army corps in a fog.

Reille obeyed orders, though the fog was lying as densely upon the mountains as on the preceding night. Apparently Soult had supposed that it would lift at dawn—but it did not till midday. Lamartinière’s division was left to guard the Linduz and the debouch of the Atalosti path: Foy’s, followed by Maucune’s, tried to keep to the crest, with the most absurd results. It was supposed to be guided by French-Basque peasants (smugglers, no doubt) who were reputed to know the ground. After going no more than a mile or two in the fog, the guides, at a confusion of tracks in the middle of a wood, came to a standstill, and talked volubly to Foy in unintelligible Basque. Whether they had lost their way, or were giving advice, the General could not quite discover. In despair he allowed the leading battalion to take the most obvious track. They had got completely off the Atalosti path, and after two miles of downhill marching found themselves on the chaussée not far from Espinal, with the rear of Clausel’s corps defiling past them[910]. It would still have been possible to stop the column, for only one brigade had reached the foot of the mountain, and Maucune and Lamartinière were still on the crest. But Reille took upon himself the responsibility of overriding his commander’s impracticable directions, and ordered Foy to go on, and the rest to follow, and to fall in to the rear of Clausel’s impedimenta. ‘Il est fort dangereux dans les hautes montagnes de s’engager sans guides et en brouillard,’ as he very truly observed. Justifying himself in a letter of that night to Soult, he wrote that if it were absolutely necessary to get on the crest-path again, it could be done by turning up the Arga valley at Zubiri, and following it to Eugui, from which there were tracks both to the Col de Velate and to Irurita.[911]

Thus ended Soult’s impracticable scheme for seizing the Col de Velate by marching three divisions along a precipitous mule track. Even if there had been no fog, it is hard to believe that anything could have come of it, as Campbell’s Portuguese would have been found at Eugui well on in the day, and after Reille’s column would have been much fatigued. Any show of resistance, even by one brigade, would have checked Foy, and compelled Reille to deploy—an interminable affair, as the fight on the Linduz upon the preceding afternoon had sufficiently demonstrated. But to try this manœuvre in a dense fog was insane, and Reille was quite right to throw it up.

The whole interest, therefore, of the French operation on July 26th turns on the doing of Clausel’s column. It advanced very cautiously down the slopes to the Abbey of Roncesvalles, discovering no trace of the enemy save a few abandoned wounded. Having reached the upland valley of Burguete, Clausel sent out cavalry patrols, and found, after much searching in the fog, that Cole had gone off with his whole force towards Espinal. His rearguard was discovered bivouacking along the road beyond that village. When it sighted the French it retired towards Viscarret. Clausel then ordered his infantry to pursue, but they were far to the rear and it was only about 3 p.m. when Taupin’s division came into touch with the light companies of Anson’s brigade, just as they were falling back on the whole 4th Division, drawn up in a favourable position on heights behind the Erro river, near the village of Linzoain. The day had at last become clear and fine. The 31st Léger, leading the French column, exchanged a lively fusillade with the light companies, while a squadron of chasseurs tried a charge on their flank. But both were driven off, and Clausel halted when he saw Cole waiting for him in order of battle. It was not till he had brought up and deployed two divisions that he ventured to press the Allied front, and nothing serious happened till after 4 o’clock.

Meanwhile Picton had come up from the rear, and joined Cole at Linzoain: the head of his troops had reached Zubiri only three miles behind. The arrival of the truculent general, looking even more eccentric than usual, for he was wearing a tall round civilian hat above a blue undress frock-coat, and was using a furled umbrella by way of riding whip, was taken by the 4th-Division soldiers as a sign of battle[912]. ‘Here comes old Tommy: now, boys, make up your minds for a fight’ passed down the ranks[913]. But, oddly enough, this was about the only day in Sir Thomas’s military career when he did not take a fair risk. He certainly came up in a bellicose mood, for he ordered Ross’s brigade to be ready to move forward when the 3rd Division should have come up to support it. But after riding to the front, and holding a long talk with Cole, he agreed with the latter that it would be dangerous to fight on ground which could be turned on both flanks, with an enemy who was known to have 35,000 men in hand. Only part of the French were up—Reille’s divisions after their stroll in the fog were far to the rear behind Clausel—so it would be possible to hold on till night, and slip away in the dark. Picton wrote to Wellington to report his decision, and does not seem in his dispatch to have realized in the least that he was contravening the whole spirit of his commander’s instructions of July 23rd with reference to the ‘stopping of the enemy’s progress towards Pampeluna in the event of the passes being given up[914].’ He merely stated that he had received these instructions too late to make it possible for him to reach Roncesvalles, or to join Cole before the latter had evacuated his positions[915]. As there was no favourable ground between the Erro river and the immediate vicinity of Pampeluna, on which a smaller force could make an effectual stand against a much larger one, he had determined to retire at once, and proposed to ‘take up a position at as short a distance as practicable from Pampeluna’—by which he meant the heights of San Cristobal, only two or three miles out from that fortress. He was thus intending to give up without further fighting ten miles of most difficult hilly country, where the enemy could be checked for a time at every successive ridge—though, no doubt, all the positions could be turned one after the other by long flank détours. But the net result was that Picton gave Soult a clear road on the 27th, and allowed him to arrive in front of Pampeluna on that day, whereas the least show of resistance between Zubiri and the debouch into the plains at Huarte, would have forced the French to deploy and waste time, and they could not have reached the open country till the 28th. This is sufficiently proved by the extreme difficulty which Soult found in conducting his march, even when he was not opposed.

So determined was Picton not to fight on the Erro river, or on the Arga, that he did not bring up his own division from Zubiri, but let it stand, only three miles behind the line on which Cole kept up a mild detaining action during the late afternoon hours of the 26th. Soult attacked with great caution, and more by way of flank movements than by frontal pressure. By evening Cole had drawn back one mile, and had 168 casualties, all but four of them in Anson’s brigade[916]. Those of the French can hardly have been more numerous: they seem all to have been in Taupin’s division[917].

On the afternoon of the 26th Picton had nearly 19,000 men at his disposition[918], Soult had somewhat less, since Reille’s column was so far to the rear that it could not get up before dark. There was no wonder, therefore, that the enemy made no resolute attack; and it can only be said that the Marshal was acting very wisely, for the French force on the ground was not sufficient to move the opposing body, until Reille should have come up; and Cole and Picton had resolved not to give way before dark. But when the fires of the French, shining for many miles on each side of the road, showed that they had settled down for the night, Cole drew off his division, and retired on Zubiri, where he passed through Picton’s troops, who were to take over the rearguard duty, as they were fresh and well rested. Campbell’s Portuguese dropped into the line of march from Eugui, by orders issued to them that afternoon, and by 11 p.m. the whole corps was in march for Pampeluna. Its departure had passed wholly unnoticed by the enemy. Meanwhile, Wellington’s aide-de-camp, riding through the night from Almandoz, with orders to Picton to maintain the ground which he was abandoning, can only have met the column when it was drawing near its destination.

It was quite early in the morning, though the sun was well up, when the head of the retreating column reached the village of Zabaldica, where the valley of the Arga begins to open out into the plain of Pampeluna, between the last flanking heights which constrict it. In front was the very ill-chosen position which Picton intended to hold, along a line of hills which are quite separate from the main block of the mountains, and stretch isolated in the lowland for some five miles north-west and south-east. These are the hill of Huarte on the right, parted from the mountains by the valley of the Egues river; the hill of San Miguel in the centre, on the other side of the high road and of the Arga river, and on the left the very long ridge of San Cristobal, separated from San Miguel by the Ulzama river, which flows all along its front.

Now these hills are strong posts in themselves, each with a good glacis of slope in its front; the gaps between them are stopped by the large villages of Villaba and Huarte, both susceptible of obstinate defence; and the two flanking hills are covered in front by river-beds—though fordable ones. But they are far too close to Pampeluna, which is but one single mile from San Cristobal: the guns of the fortress actually commanded at a range of only 1,200 yards, the sole road of communication along the rear of the position. Cassan’s garrison of 3,000 men was not large enough to furnish men for any large sortie—though he made a vigorous sally against O’Donnell’s blockading division on the 27th, and destroyed some of its trenches[919]. But no army should fight with a hostile fortress less than two miles in its rear, and commanding its line of retreat: it is surprising that such an old soldier as Picton chose this ground—presumably he was seduced by the fine position for both infantry and guns which it shows looking towards the enemy’s road of arrival.

Apparently Cole had a better eye for ground than Picton, for as they were riding together between Zabaldica and Huarte, he pointed out to his senior the advantage that would be gained by throwing forward the left wing of the army to a position much more remote from Pampeluna, the hill of Oricain or Sorauren, which faces the San Cristobal ridge from the other side of the Ulzama river[920]. This height is the last roll of the mountains, but almost separated from their main massif: it is only joined to the next summit by a high col at its right centre. For the rest of its length it is separated from its neighbour-height by a well-marked ravine. Its flanks are guarded by the beds of the Arga to the right and the Ulzama to the left. It is well under two miles long, about 1,000 feet high, and except at the Col has a very formidable front of steep slopes, covered with gorse and scattered bushes. The whole formed a strong and self-contained position, whose weak point was that it was rather too much in advance of the Huarte-San Miguel heights, which trend away southward, so that when the army was drawn out its right was much ‘refused,’ and its left very much thrown forward. It was also inconvenient that the access to the crest from the rear was bad, a steep climb by sheep-tracks from Oricain or Arre, up which all food or munitions would have to be brought. From the north there was a slightly better path to the summit from Sorauren, leading up to the small pilgrimage chapel of San Salvador on the left end of the crest. But this would be of more use to the assailants than to the defenders of the heights. Between the Col and the river Arga, and close above the village of Zabaldica, there was a spur or under-feature of the main position, which formed a sort of outwork or flank protection to it. At the moment when the retreating army was passing on towards Huarte, this spur was being held by two Spanish battalions, part of the division which O’Donnell, by Wellington’s orders, had detached to reinforce Picton. It was perhaps the sight of this small force in a very good position which suggested to Cole that the right policy was to prolong his line in continuation of it, across the Col and as far as the chapel above Sorauren.

Having allowed Cole to take up his new advanced position, Picton drew out the 3rd Division on the hill to the right of Huarte, with its flank eastwards covered by four brigades of cavalry, which had come up by Wellington’s orders from their cantonments on the Ebro[921]. Morillo’s Spaniards continued the line westward along the Cerro de San Miguel, as far as Villaba: from thence the San Cristobal ridge was occupied by the greater part of the division which O’Donnell had drawn from the blockading lines—all, in fact, save the two battalions in advance on the hill by Zabaldica. Later in the day two battalions more were added from the besieging force, for Carlos de España’s division from Castile had arrived, and relieved part of the troops which had hitherto been observing Pampeluna. Byng’s brigade was told off to support the 4th Division, and took post on the rear of the summit of the Oricain hill, half a mile behind Cole. The actual fighting line on the left was composed of Anson’s brigade on the Col, next to the Spaniards on the lower spur, of Campbell’s Portuguese upon the central stretch of the heights (except one battalion which was sent to support the Spaniards below)[922], and of Ross’s brigade holding the left. Stubbs’s Portuguese were in rear of Campbell’s, except the 7th Caçadores, which was detached to the front and held the ground about the chapel of San Salvador. The divisional battery (Sympher’s of the K.G.L.) was placed far down the right side of the hill, below and behind Byng’s brigade, in a position from which it could sweep the high road from Zabaldica to Arleta. Cole’s tactical dispositions were in the complete Wellingtonian style, with the light companies and caçadores thrown out some way down the slope, far in advance of the main force, whose battalions were drawn back well behind the sky-line, so as to be invisible till the last moment to enemies storming the hill. Soult followed up the retreating Allies at such a slow pace that the whole of Picton’s troops were settling into their ground before the enemy came in sight[923].

The slow advance of the French was due to the accumulation of such a large force in a narrow valley provided with only one road. The Marshal made an attempt to relieve the congestion, by ordering that the chaussée should be left to Clausel and to the cavalry and impedimenta in his rear, while Reille’s divisions should move on the east bank of the Arga by local paths between the villages. The excellent intention of securing room for both columns to move freely had no good result. Clausel arrived in front of Zabaldica by 9 a.m.[924] But Reille was nowhere in sight. His report fully explains his absence: he had obeyed orders by turning up into the hills a mile and a half beyond the village of Erro. ‘This direction rendered the march of the three divisions extremely slow and difficult. They found no road, and had to tramp through brushwood, climb steep slopes, or to follow tracks obliterated by recent rain. At last Count Reille took the decision to abandon the high ground. The 1st Division (Foy) dropped down to the village of Alzuza on the extreme left. The 7th Division (Maucune) re-descended into the valley of the Arga, a little above Iroz, where it bivouacked. The 9th Division (Lamartinière) also came down into the valley opposite Larrasoana, and kept along the high road to Iroz,’ where it fell in with the rear of Clausel’s column late in the day. The only result of Soult’s precaution had been to put Reille out of the game on the 27th, just as on the 26th.

For the whole of the morning hours, therefore, Soult had only Clausel’s corps at his disposition, a fact which accounts for the unenterprising character of his action. But that the 27th was a very slack day on the French side was not Clausel’s fault. On arriving at Zabaldica, and discovering that the heights of Oricain were held in strength, he did not wait for the Marshal’s orders, but began to form a line of battle parallel to Cole’s front, along the mountain opposite. Halting Conroux’s division on the high road in face of the hill held by the Spaniards, he pushed Taupin’s and Vandermaesen’s divisions up the slopes, with cavalry detachments feeling the way in front of them, till they had lined the whole ridge, and their right was overlooking Sorauren and the valley of the Ulzama. He then sent down to ask the Marshal’s leave to attack, saying that he could see from the summit behind his line large baggage trains moving away along the Vittoria road in the plain of Pampeluna, and bodies of troops in motion northward[925]—the enemy was about to raise the siege, and was only offering a rearguard action in order to cover the retreat of his impedimenta. If pressed he would give way at once[926].

Soult did not believe this, and very rightly; but being pressed by repeated messages he mounted up to the heights behind Clausel’s front at 11 a.m.: if he had chanced to notice it, he was just in time to see a solitary horseman ride up the north-western slope of the hill of Oricain, and to hear the whole of the Allied troops aligned opposite him burst out into a storm of tempestuous cheering. Wellington had come upon the ground. Soult heard the noise, but (as his dispatches show) did not guess its precise cause. He thought that reinforcements had just come up for Cole.

The story of Wellington’s eventful ride from Almandoz to Sorauren is a very interesting one. Much irritated at receiving no further news from Picton, he had mounted at sunrise and ridden over the Col de Velate, taking with him only George Murray, his Quartermaster-General, his Military Secretary Fitzroy Somerset, and three or four other officers: the bulk of the head-quarters staff was to follow at leisure. On arriving at Lanz, the first village on the south side of the pass, they heard rumours of Picton’s continued retreat, though they do not seem to have met the aide-de-camp whom he had sent off on the preceding night to report it. This news was so unexpected and vexatious that Wellington halted for a moment, to send back orders to Hill to the effect that it was conceivable that affairs might go badly on the Pampeluna front. If so, the whole right or southern wing of the army might have to swing back to the line Yrurzun-Tolosa, and Hill would have to direct his own two divisions, and also Dalhousie, and Pack, with all the artillery and baggage, to fall back westward on Lizaso and Lecumberri, instead of coming over the Col de Velate towards Pampeluna. The Light Division, too, might have to leave the neighbourhood of the Bastan, and to retire to Zubieta on the Oyarzun-Lecumberri road, in order to keep up the touch between the main army and Graham’s force in front of St. Sebastian. The latter general, however, was not to move, unless matters went very badly indeed, as the blockade of St. Sebastian must be kept up till the last possible minute. But previous orders were to stand, unless and until the Commander-in-Chief should send new ones: in particular Pack and the 6th Division were expected at Olague, and the batteries of Silveira’s division and the Light Division might come on to Lanz, as there was an artillery road from Olague by which they might be turned off eastward if it became necessary[927].

On getting five miles farther down the road, Wellington halted for another moment at Olague, to leave word that the 6th Division[928], when it arrived, was to hold that place till further notice, and especially to look out for a possible movement of the French across the hills from Eugui, which must be blocked at all costs. Pack must turn all wheeled transport, batteries, convoys, &c., arriving from the Col de Velate off the high road to Pampeluna, and send them westward by the side road Olague-Lizaso, at which last-named village everything must wait for further orders. The closest and most frequent communication must be kept up with Hill’s corps, which would be wanting to use this same road. Finally, Pack, after resting his division and giving it its noonday meal, must be ready to march again at a moment’s notice in the afternoon[929].

Three miles farther down the road, at Ostiz, Wellington found waiting for him General Long, with some of the squadrons of his Light Dragoons, who were dispersed all along the lines of communication, keeping touch with all divisions. Long gave the alarming information that Picton had abandoned the Linzoain and Zubiri positions during the previous night, and was now in the immediate neighbourhood of Pampeluna, where he was intending to fight on the San Cristobal heights. The French were known to be in pursuit, and a collision might occur at any moment—indeed might have occurred already, but no firing had yet been heard.

Ostiz is only four miles from Sorauren and six from Villaba; there was probably time to reach the fighting ground before an action might commence. Wellington directed his Quartermaster-General to stop behind, and make preparations for turning all troops off the Lanz valley road on to the Lizaso road, if he should receive further orders in the next hour—everything depended on what was going on six miles away. He then went off at racing speed down the chaussée, gradually dropping behind him all his staff except Fitzroy Somerset—their horses could not keep up with his thoroughbred. Turning the corner half a mile from Sorauren, he suddenly came on the whole panorama of battle. Cole’s line was visible on the right-hand heights stretching away from the Chapel of San Salvador to the Col above Zabaldica. On the opposite mountain Taupin’s and Vandermaesen’s divisions were moving along the crest towards Sorauren and the Ulzama valley: cavalry vedettes were pushing ahead of them all over the slopes, looking for paths or British outposts. They were only a mile away at most. There was just time, and no more, to join Cole and take over the direction of affairs. Wellington put on full speed till he reached Sorauren bridge, and then (with his usual cool-blooded calculation of risks and moments) dismounted, and wrote a short order to Murray in pencil, using the cap-stone of the bridge end as his table. While he was writing the thirteen hurried lines, he was much distracted by well-intentioned peasants, who flocked around him with shouts that the French were coming down into the other end of the village. But the dispatch reads clearly enough. Murray is informed that the high road is blocked by the presence of the French at Sorauren; all troops, therefore, must turn off on to the side-road Olague-Lizaso, both Pack and the artillery, and also Hill’s corps. The latter must march at once from the Bastan, and get across the Col de Velate by nightfall if it could, leaving a rearguard to hold the pass against any possible pursuit by D’Erlon. The 7th Division near Santesteban should also come across by the Puerto de Arraiz to Lizaso. Orders for the further movements of all troops would be sent to Lizaso as soon as possible[930].

Fitzroy Somerset dashed out of the village at its northern end with the completed dispatch, just as the French chasseurs came exploring into its other end. He was not seen, or at least not pursued, and Murray received the orders, which made Lizaso the concentration point of all the central divisions of the army in half an hour, and set to work to amplify them and to forward them to their destinations. As Wellington very truly observed, several hours were gained by sending back Somerset by the straight road, and in particular the 6th Division was able to reach Lizaso by dark, and to get a good rest for the march of the next day to the battlefield[931].

Meanwhile, Wellington, now all alone, rode up the steep track which rises from Sorauren to the pilgrimage-chapel on the height above, and was presently among the skirmishing line of O’Toole’s Caçadores, who were holding that corner of the front. His familiar but unobtrusive silhouette—the short frock-coat, small plumeless cocked hat fitting down tight over the great Roman nose, and wiry thoroughbred—was at once recognized—the Portuguese set up the cry of ‘Douro,’ with which they were wont to greet him—recalling the first victory in which English and Portuguese co-operated, and also his first title of nobility: the noise swelled into the hoarse cheers of the British soldier as it passed up the line towards the Col. The 4th Division, which had been grumbling bitterly since it had been on the retreat, suddenly felt the atmosphere change. ‘I never can forget the joy which beamed in every countenance when his Lordship’s presence became known. It diffused a feeling of confidence throughout all ranks. No more dispiriting murmurs on the awkwardness of our situation: now we began to talk of driving the French over the frontier as a matter of course[932].’ Wellington halted in front of Ross’s brigade, and for a long time studied the French movements through a telescope. He easily made out Soult himself, who was conferring with Clausel and other staff-officers in a conspicuous group. Napier says that he observed that the Marshal would have heard the cheering, and would try to make out what it was about, before taking any serious step: ‘that will give time for the 6th Division to arrive, and I shall beat him.’ There is no corroboration for this story, though it may be true: Napier was in England on July 28th, and speaks only from hearsay. But an eye-witness present on the spot says that General Ross, as his chief continued to focus the French staff, ventured the remark that ‘this time Soult certainly meditates an attack,’ to which Wellington, with the glass still to his eye, replied, ‘It is just probable that I shall attack him[933]. And this thought seems to be corroborated by a remark which Wellington made six days after to Judge Larpent, to the effect that he should and could have done more on the 27th[934]. The French army was but half arrived—only Clausel’s three divisions were up, and Reille was at 11 a.m. still miles away. The enemy was tempting Providence by marching across the Allied front, just as Marmont had done at Salamanca a year before.

Whether Soult was stopped from an early attack, by guessing that the cheers on Cole’s front implied the arrival of Wellington, seems more than doubtful. There is, of course, no trace of such an idea in the French dispatches—naturally it would not have been mentioned. But the one personal narrative which we have from a member of the group of Soult’s staff-officers whom Wellington was eyeing, is to the effect that Clausel was at the moment trying to persuade the Marshal to attack at once, though only half his force was up, that Soult utterly refused to do so, spread out his maps, and finally took his lunch and a nap after it. ‘Clausel meanwhile leaning against an oak was literally beating his forehead with rage, muttering “Who could go to sleep at such a moment”[935].’

What is certain is that Soult’s dispatch to Paris says that he had discovered that Wellington had arrived in the afternoon, which would seem to show that he did not know by personal observation that he had come up at 11 a.m. He says, also, that at the moment of his own arrival the Allies had 30,000 men in line, including all the blockading troops from Pampeluna, which rendered it necessary to make a thorough examination of their position, and to get up all the divisions[936]. To discover the exact ground on which the enemy intended to fight, it proved necessary to make demonstrations or partial attacks.

Two of these were executed: the spur held by the two Spanish battalions, above Zabaldica, was so close in to the French position, that it was thought worth while to make an attempt to occupy it. A regiment of Conroux’s division was sent up from the village to storm it, but was handsomely repulsed, when near the top of the slope, by a charge in line of these corps (Principe and Pravia of O’Donnell’s Andalusian Reserve). Clausel’s report says that his men ‘took the hill but could not keep it,’ but many British eye-witnesses on the slope above say that the summit was never reached. Clausel estimates the loss of his regiment at 100 men, Soult gives the more liberal estimate of 200. The repulse showed plainly enough that the whole hill of Oricain was to form part of the Allies’ position, and that they intended to fight for every inch of it.

Late in the afternoon Soult directed Foy, who had now reached Alzuza on the extreme French left, to demonstrate against the heights of Huarte, so as to discover the end of the British line, and the strength in which it was held. Foy sent forward two regiments in column down the slope toward the Egues river, while showing the rest of his troops on the hill behind. When the French got within cannon shot, Picton brought up the whole 3rd Division to the crest, and R. Hill’s and Ponsonby’s heavy cavalry showed themselves in front of the village of Gorraiz, covering his flank. The divisional battery fired a few rounds at the columns, which at once swerved and retired in haste. Foy could report that the Huarte heights were held in strength, and by all arms.

This was the last incident of the day—shortly afterwards a heavy thunderstorm swept down from the Pyrenees, darkened the twilight, and drenched both armies. The same thing had happened on the eve of Salamanca—and was to happen again on the eve of Waterloo.

Far more important than the trifling skirmishes of the afternoon were the arrangements for the morrow—orders and dispatches, which the rival commanders were evolving during the evening hours. Soult felt himself still blocked in the valley of the Arga and wanted to deploy—was it possible to get his long column of guns and cavalry out of the defile—and if so, how? Was there a possibility of extending to the left beyond Foy’s present position, or to the right beyond Sorauren, the limit of Clausel’s occupation? Or must the Oricain heights be captured at all costs before the army could get into a proper order of battle? Could any immediate help be expected from D’Erlon, whose tiresome letter of the 26th had come to hand, showing that he was no farther forward than Elizondo on that night? He was not doing his best to occupy the enemy in his front, and there was a danger that Hill might arrive at Sorauren long before the corps that had been set the task of occupying his attention.

There would appear to have been something like a council of war at the French head-quarters in the evening, in which Clausel, Reille, Gazan, the Chief of the Staff, and probably other officers took part. Clausel wished to extend the line northward up the valley of the Ulzama, and to turn the flank of the whole Oricain position. The objection to this was that the transport and guns could not follow over the mule-tracks which must be used; they were blocked in the Arga valley as long as the Allies held the heights commanding the main road. Moreover, reports had come in that British troops were descending on to Sorauren by the Lanz valley road, who would take in flank any attempt of Clausel to turn the Oricain position. (This seems to refer to the arrival of the 6th Division at Olague, but Long’s cavalry was also visible up the chaussée.) And an extension of the French right to the heights westward would make the whole line of battle very long and weak: how could Reille’s three divisions take over the whole ground from Sorauren to Alzuza? After much discussion Soult decided in favour of a concentric attack on the whole Oricain position by five of his six divisions, while only Foy should remain out on the left, observing and containing the British force on the Huarte hills. Clausel’s three divisions should attack Cole’s line from Sorauren to the Col, two of Reille’s divisions should co-operate, by assailing the Col and the Spaniards’ Hill south of it. Some guns should be got to the front if possible, and Foy should be lent some cavalry, who must climb over the hills to cover his flank. Details would have to be settled on the morrow, also the transference of troops, who could not move in the dark over steep and unknown ground. Yet there was an uneasy feeling that too much time had been lost already—Soult notes a rumour that Wellington had announced the approach of four more British divisions. Even the arrival of D’Erlon could not compensate for this, and it did not look as if D’Erlon was likely to appear early on the 28th.

Wellington moved the troops who were on the ground very little that evening—only relieving the Portuguese battalion which was in reserve on the hill above Zabaldica by a British regiment of Anson’s brigade—the 40th Foot—and sending two of O’Donnell’s battalions to Ollocarizqueta, to watch the mountain road west of Sorauren, by which it was conceivable that Clausel might try to turn the hill of Oricain. He had got news that French troops had been seen beyond Sorauren, and did not want to have them prying too close behind this flank, or observing the paths by which he intended to bring up his reserves on that side.

For his main attention that afternoon was devoted to drawing up the orders for the divisions coming from the Bastan, which he had promised to send to Murray, in the note that Fitzroy Somerset bore from the bridge of Sorauren. At 4 o’clock he sent off the all-important dispatch[937]. It will be remembered that his last orders provided for the successive arrival at Lizaso of the 6th and 7th Divisions, of the whole corps of Hill, and of the artillery, and divisional baggage trains of all the troops.

Pack was to start from Lizaso at dawn, and to move as rapidly as possible along the country road by Marcalain to Ollocarizqueta, which goes along the back of the hills that form the right bank of the valley of the Ulzama. At the last-named village (which he would find held by two Spanish battalions) he would be only five miles from Cole’s flank, and in a position to strengthen or cover it. A supplementary note[938] warned Pack that if any French were found anywhere on the way he must not let himself be turned off; even if it came to leaving the road and taking to mountain tracks, the 6th Division must arrive at its destination. The artillery and the reserve of infantry ammunition were to follow Pack, unless serious opposition were offered to him, in which case they must return to Lizaso, and from thence turn on to Yrurzun.

Dalhousie and Hill had already had orders to march on Lizaso, the one by the Puerto de Arraiz, the other by the Velate. Both had a long march, but Wellington hoped that they might reach Lizaso during the night of the 27th-28th. If the men were not over-fatigued, all three divisions should follow Pack to Ollocarizqueta, after being given a suitable time of rest. All impedimenta likely to hinder rapid marching, and the wounded from the Maya fight, were to be directed from Lizaso to Yrurzun. Murray had already, acting on the orders of the Sorauren Bridge dispatch of 11 a.m., sent the route for Lizaso to Hill and Dalhousie, adding some precautions of his own, to the effect that they should leave small rearguards at the passes, to detain D’Erlon if he should come up. Moreover, he had spread out the 1st Hussars of the K.G.L. along the roads, to keep up touch between all the divisions on the move, and also between Lizaso and the Light Division, now at Zubieta. The only addition made by Wellington’s new orders was the all-important one that everything of the fighting sort that came to Lizaso was to march for the main army, but all baggage for Yrurzun.

Of the arrangements thus made, that for Pack worked perfectly—he was at Ollocarizqueta with the 6th Division before 10 a.m. on the morning of the 28th. But Hill and Dalhousie were detained in the passes by the storm of the evening of the 27th, and only reached Lizaso so late on the 28th that they were of no use in the fighting of that day. As Wellington observed to Larpent[939], they could both have got into the battle of Sorauren if they had been given a longer march on the 26th, but ignorance of Cole’s and Picton’s retreat had prevented him from starting them early enough. However, the 6th Division alone sufficed to settle the matter.

The morning of the 28th was fine and bright—the storm of the preceding night seemed to have cleared the clouds from the hills, and the eye could range freely over a very wide landscape. Not only was the battle front of each army visible to the other, but from the high crest on or behind each position, a good deal could be made out of what was going on in the rear. This was more the case with the French line than with the British: Wellington’s usual plan of keeping his main force behind the sky-line had great efficacy on such lofty ground as that of the heights of Oricain.

Soult spent the long hours of the July morning in moving his troops to the positions which had been selected on the previous night. Conroux’s division, abandoning Zabaldica, marched over the hills in the rear of Clausel’s other divisions, and occupied Sorauren village—relieving there some of Taupin’s battalions, which shifted a little to their left. To replace Conroux at Zabaldica Lamartinière came up from Iroz, and deployed one brigade (Gauthier’s) in front of the Spaniards’ Hill, while the other (Menne’s) remained in reserve beside the Arga, with two regiments across the high road and the third on the slopes east of the river. Maucune’s division, starting up the slopes from the rear of Iroz, took post on the heights immediately opposite the Col, with its right touching the left of Vandermaesen. Two regiments were in front line exactly in face of the Col, a third somewhat farther to the left, keeping touch with Lamartinière’s brigade near Zabaldica. The other brigade (Montfort’s) was placed in reserve farther up the mountain[940]. Pierre Soult’s light cavalry regiments disentangled themselves from the long cavalry and artillery column in the rear, and picked their way up the heights southward till they reached Foy’s position, from which they extended themselves to the left, till they reached as far as the village of Elcano. Four howitzers were chosen out of the batteries in the rear, and brought forward to the front of Zabaldica, a position in which it was hoped that their high-trajectory fire might reach effectively the Allies on the Col and the Spanish Hill. These were the only French cannon used that day, except some mule-guns employed by Clausel on the side of Sorauren.

The greater part of these movements were perfectly visible to Wellington, though some of them were intermittently screened from sight when the marching columns dipped into dead ground, or were passing through thickets. They took up so many hours that he wrote to Graham at 10.30 that it looked as if Soult was not inclined to attack. Meanwhile, the only British force on the move was the 6th Division, which had been marching since daylight, and arrived about 10 a.m. at Ollocarizqueta. On getting news of its presence, Wellington ordered it to move by a country road which comes across a crack in the hills west of the Ulzama not far from the village of Oricain, and after reaching that river to keep its Portuguese brigade on the western bank, its two British brigades on the eastern, and so to advance till it should arrive facing Sorauren, prolonging Cole’s left in the lower ground, and covering him from any attempt to turn his flank. About noon Pack’s leading troops began to appear—these were Madden’s Portuguese on the left of the division, working along the hillside west of the Ulzama.

Now the ‘zero’ hour for the general attack on the whole Allied front had been fixed by Soult for 1 o’clock, but long before that hour Clausel was informed by an exploring officer, whom he had sent to the hills on the other side of the Ulzama, beyond his right, that a heavy British column was visible coming from the direction of Marcalain, obviously to join the British left[941]. Now the French general had been thinking of turning Cole’s flank on this side, and had actually got troops some little way up the Ulzama valley beyond Sorauren. But hearing of Pack’s approach and of his strength, he determined that he must now turn his attention to fending off this move against his own flank, and that the effort would absorb no small portion of his troops. Conroux’s division had been originally intended for the attack of the north-west corner of the Oricain heights, to the right (French) of the chapel, while Taupin was to make the chapel itself his objective; Vandermaesen was to strike at the centre of Cole’s position, and Maucune to attack the Col. But Clausel determined that, in order to make the general frontal attack feasible, he must at once stop the forward march of Pack, and keep him at a distance, while the other divisions made their great stroke. Accordingly he called in his right wing detachments, and having concentrated Conroux’s strong division (7,000 bayonets) he ordered it, at 12.30, to deploy across the valley of the Ulzama below Sorauren and advance in two lines of brigades against the approaching British column. This movement—probably a necessary one—brought on the first clash of battle, half an hour before the appointed time for the general movement. Clausel at the same moment sent word to the other divisions to attack at once; but they were not quite ready, the fixed hour not having yet been reached.

But Conroux pushed up the valley farther than was prudent, for when he was some half a mile beyond Sorauren, he found himself encompassed on three sides—Madden’s Portuguese pushing forward along the hill on the other side of the river, turned in on his right flank, some of Ross’s skirmishers on the slopes west of the chapel of San Salvador came down and began to fire upon his left flank, while the main body of the 6th Division—Stirling’s brigade deployed in front line, Lambert’s supporting in column—met him face to face in the low ground. The concentric fire was too heavy to be stood, and Conroux had to give back, fighting fiercely, till he had the support of the village of Sorauren at his back: there the battle stood still, for Pack had orders to cover Cole’s flank, not to attack the enemy’s[942].

Meanwhile, the general frontal attack was being delivered by the other French divisions, starting from the right, as each got forward on receiving Clausel’s orders to anticipate the fixed hour and advance at once. Hence the assault was made in échelon of brigades from the right, Taupin’s two brigades being a little ahead of Vandermaesen’s, and Vandermaesen’s perceptibly earlier on the hill than the brigade of Maucune which attacked at the Col. The separate assault by Lamartinière on the Spaniards’ Hill is definitely stated by that general to have been made at 12 o’clock—earlier, apparently, than the others—but he is contradicted by Reille, in command on this front, who says that it started vers une heure. This is much more likely to be correct.