But, as has been pointed out recently[276], Wellington himself may be quoted in favour of this theory. In a letter written to Canning in 1820 concerning quite another matter, he remarked, ‘I believe that it has always been understood that the defenders of a fortress stormed have no claim to quarter, and the practice which prevailed during the last century of surrendering fortresses when a breach was opened, and the counterscarp blown in, was founded on this understanding. Of late years the French availed themselves of the humanity of modern warfare, and made a new regulation that a breach should stand one assault at least. The consequence of this regulation of Bonaparte’s was the loss to me of the flower of my army, in the assaults on Ciudad Rodrigo and Badajoz. I should have thought myself justified in putting both garrisons to the sword, and if I had done so at the first, it is probable that I should have saved 5,000 men at the second. I mention this to show you that the practice which refuses quarter to a garrison that stands an assault is not a _useless_ effusion of blood.’ [276] By Colonel Callwell, in an article in _Blackwood’s Magazine_ for September 1913. Comparatively few of the garrisons of Rodrigo and Badajoz were shot down, and those all in hot blood in the moment after the walls were carried. Suchet’s army was much more pitiless at Tarragona, where a great part of the Spanish garrison was deliberately hunted down and slaughtered. But there was, of course, a much more bitter feeling between French and Spaniards than between English and French.