On the left the decisive crossing at Pombeiro had been made by the Guards’ brigade of the 1st Division[205]: Reynier made no attempt to support the single battalion at the ford, but called it in, and drew up in battle order on the Serra de Moita some distance above the river. There was no fighting here, and after dusk the 2nd Corps decamped, taking the road Galliges-Chamusca-Gouvea-Celorico, which runs parallel to and above the Alva; the 6th Corps fell in behind the 2nd; the 8th, which had already got further to the east during the march of the preceding night, led the column and was beyond Galliges in the afternoon[206].

On the night of the 18th-19th and the following morning the whole French army made a most protracted and fatiguing march, which cost them many stragglers and much material. This was the longest stage which Masséna made during the whole retreat, more than twenty miles of mountain road being covered in one stretch. On the evening of the 19th the head of the 8th Corps was at Pinhanços, that of the 2nd at Çaragoça and Sandomil, that of the 6th at Chamusca. The British cavalry in pursuit picked up an immense number of prisoners this day, mainly small parties of foragers belonging to the 6th and 8th Corps, who had been marauding in the direction of the Mondego when their regiments received sudden orders to march: they returned towards their camps to fall into the hands of the British cavalry. Arentschildt’s brigade alone took 200 this day[207], and the total according to Wellington’s dispatch was 600 men[208]. Among them were an aide-de-camp of Loison’s and several other officers. Some droves of oxen were captured from the foragers and from the rear of the 6th Corps, and this was the only food which the British vanguard got that day, for once more, as several diarists ruefully note, ‘biscuit was out.’ The pursuers state that ‘quantities of tumbrils, carts, waggons and other articles were abandoned by the enemy’[209], and no doubt were right, though Masséna wrote that evening from Maceira, ‘Nous n’avons pas laissé en arrière un malade, un blessé, ni la moindre voiture d’artillerie ou de bagages’[210]. But his dispatches frequently contained what he knew that Napoleon would desire to see rather than the truth. Considering the difficulties of the retreat, and the skilful way in which he had conducted it, he might have been contented with his actual achievements, and need not have padded out his reports with ‘terminological inexactitudes’. It must be confessed that his aide-de-camp Fririon played a good second to him when he wrote in his journal that between the 15th and 31st of March the Army of Portugal left behind only twenty prisoners and fifty men lost while marauding[211].

On the morning of the 20th March Wellington’s army was all across the Alva, but only the cavalry and the Light, 3rd, and 6th Divisions continued the pursuit. The small convoy received on the 17th was exhausted, and what little food remained was made over to the three divisions named above, while the 1st, 5th, and Pack’s and Ashworth’s Portuguese halted for five days at Moita and the neighbouring villages, till the first train of provisions sent up from Coimbra should arrive. Of the troops which could still be fed, the Light Division pushed on as far as Galliges that night, the 3rd and 6th were behind them. The French, thanks to their desperate night march, were already a long way ahead, and it was late on the 20th before the cavalry overtook, near Cea, their extreme rearguard, two battalions of Heudelet’s division and a hussar regiment. General Slade, whose brigade was leading, refused to attack them until infantry and guns came up to his aid, and the French slipped away before Elder’s Caçadores and Bull’s battery reached him[212]. ‘This was the second day without bread, and the third without corn for the horses, and we had marched nearly five leagues,’ explains a cavalry diarist[213]. The pursuers were as harassed as the pursued, and could go no further, but Slade was thought to have shown weakness. He picked up a good many stragglers and sick, and found the road strewn with broken vehicles and dying mules.

Meanwhile Masséna had divided his army into two columns at the fork of the roads near Maceira. The 2nd Corps, taking, on the right hand, the southern and more hilly route, reached Gouvea. The 8th Corps, taking the northern fork, got to Villacortes, and sent its cavalry on ahead to the bridge of Fornos, which they found broken and guarded by Trant’s Militia. The 6th Corps, following the 8th, lay at Pinhanços. Since leaving the Ponte de Murcella the French were a little less pressed for food: the district through which they were now passing was fertile, and had not been raided before; most of the peasantry had returned to it during the winter. Hasty plundering on both sides of the road brought in a certain amount of food, especially in the way of cattle, which had been sent up into the hills, but were often discovered by skilled marauders. There was continual bickering with the Ordenança, one party of whom, only 300 strong, tried, with more courage than discretion, to defend the village of Penalva against the advanced guard of the 2nd Corps. All along the road the pursuing Light Division found the dead bodies of peasantry, mixed with those of the French sick who had fallen by the way.

On the 21st the 8th Corps reached Celorico, where Drouet was found in position with Conroux’s division of the 9th Corps. The 6th Corps reached Carapichina and Cortiça. The 2nd, turning off at Villacortes on the Guarda road, had a most distressing mountain march to Villamonte. The pursuing infantry of the Light and 3rd Divisions reached Pinhanços and Maceira, with the cavalry five miles in front, at the convent of Vinho. They had now dropped fifteen miles behind the French rear, and were quite out of touch with it, but continued to pick up stragglers—200 men are said to have been taken on that day[214]. Like the enemy, they were now commencing to get a little food from the country. ‘The peasants, who had all fled to the mountains on the enemy’s retreat, on seeing us come down, baked bread for the troops, and gave us whatever they had left. They had suffered a good deal, all the principal houses had been burnt, and those left a good deal destroyed, but the [French] troops had not been able to discover all the corn, &c., concealed[215].’

On the 22nd the French army had reached the end of the main retreat. Two corps were concentrated in and near Celorico[216], the other (Reynier) reached Guarda, where it found Claparéde’s division of Drouet’s corps, which had been there for some weeks. They were now only three marches from Almeida and four from Ciudad Rodrigo; communication with these two places was open, for Drouet and the 9th Corps had now come into touch again with the Commander-in-Chief, and were available for keeping the roads safe. The English had been outmarched, and Masséna might have retired to Almeida and Rodrigo practically unmolested, by a good chaussée. But this was not to be the end of the campaign, as every one in the French army, save its Commander-in-Chief, fervently desired. For generals and rank and file alike were tired out, and yearned for a cessation of mountain marches, and a rest in well-provisioned cantonments, before they should be called upon for another effort. But another and a most unexpected episode was to take place before the weary columns reached Rodrigo. Masséna made one more attempt to ‘save his face,’ and to avoid being thrust over the Spanish frontier, along the road on to which his adversary had forced him. This led to a fortnight more of manœuvres in the mountains, and to the combats of Guarda and Sabugal, after which the French Marshal had to pocket his pride and acquiesce in the inevitable. These operations are of a character so different from those which preceded them that they must be treated as a separate story. The retreat from Santarem really ended at Celorico. The events between March 22nd and April 4th must be dealt with apart, since they practically amounted to a belated attempt on the part of Masséna to seize the offensive once more, and to shift the scene of war without his adversary’s consent. The project failed, and (as we shall see) was hopeless from the first.

Before dealing with it, a few remarks on the general character of the operations that had taken place between March 9th and March 22nd must be made. When reading in succession two narratives of Masséna’s retreat from Santarem to Celorico, one by an English and one by a French eye-witness, it is often difficult to realize that the two writers are describing the same series of operations. Most of the French conceive the retreat to have been a series of triumphant rearguard actions, in which their army got off practically unmolested, under cover of the skilful operations of a small covering force. On the other hand, the English tell the tale as if the whole French army was easily hunted out of Portugal by inferior numbers, foiled in repeated attempts to occupy a permanent position in the valley of the Mondego, and finally thrust back in a direction which it did not intend to take. Where the French tell of nothing but an orderly retreat and small losses, the English speak of the capture of hundreds of prisoners, and the destruction of the whole baggage-train of the retiring army.

To a certain extent the mental attitudes of both sets of narrators can be understood and justified by the impartial student. When a rearguard action takes place, the covering force left behind by the retiring army always thinks of itself as being opposed to the whole pursuing host. It considers itself to be braving the assault of immensely superior forces, and if it holds its own for a time and gets off without crushing losses, is well satisfied with its own conduct. This is both justifiable and comprehensible. But to the pursuer the same rearguard action presents an entirely different moral aspect. The few squadrons and battalions forming the head of his advance find themselves suddenly brought to a check by a considerable hostile force arrayed in a formidable position. They are forced to halt till their supports begin to come up, from five, ten, or fifteen miles in the rear. Then, when a body of reserve has begun to accumulate behind them, they launch themselves upon the enemy, who gives ground after more or less fighting and retires. The leading brigade or division of the pursuers considers that it has victoriously driven the whole hostile army from its strong position, and is no less satisfied with itself than is the force to which it has been opposed. In short, all rearguard actions begin with a check to the pursuers; they all end with the retreat of the defenders. This is their necessary course; both parties, if their generals play the game properly, may be content with themselves; the one has gained time for the escape of the main body of the retreating host, the other has cleared the way for the progress of the pursuing army. The only method in which their relative merits can be tested, is by asking whether the rearguard detained the advanced guard as long as might have been expected, inflicted disproportionate losses upon it, and got away with the minimum of loss to itself, or whether, on the other hand, the advanced guard evicted its opponents from their position at a rapid rate, with small sacrifices, and with considerable punishment inflicted on the enemy. The mere facts that the rearguard held back the pursuers for some hours, and that the advanced guard ultimately carried the enemy’s position, are obligatory incidents of such fights. When we come to examine the details of the combats which took place between the 11th and 19th of March, we shall neither hold with the French narrators that for many days Ney and two divisions of the 6th Corps fought and held back Wellington’s whole army, nor with the English narrators that the Light and 3rd Divisions, unassisted by their comrades, hunted 40,000 French out of the valley of the lower Mondego. Yet it is true that of forty-four French infantry officers, who were killed or wounded in Portugal between these dates, no less than thirty-seven belonged to the divisions of Marchand and Mermet, while, similarly, of twenty-nine British and Portuguese officers hit in the same period, no less than nineteen belonged to the Light Division and eight more to the 3rd. Clearly therefore Ney did not contend with the whole British army, but only with its two leading divisions, and those two divisions did not contend with the whole French army, but only with the two units which formed its rearguard. Fine writing on both sides, as to struggles against overwhelming odds, must be disregarded. Still more so may the exaggerated estimates as to loss inflicted on the enemy which are to be found in the narratives of most of the first-hand writers. What they are worth may be guessed from Marbot’s statement that the British lost 1,000 men at Redinha—Noël raises this liberal estimate to 1,800[217]—the real casualty list being 240. Similarly Grattan tells us that the French had 1,000 men hors de combat at Foz do Arouce[218], when their actual loss seems to have been about 250.