If shell guns were of small value, what was suitable? Were the old spherical solid shot still capable of beating the defence? A serious effort was made in this country to bring them to do it. The Armstrong rifled breech-loading guns recently adopted had been proving defective and indifferent on service; a return was wisely made to muzzle-loading; and it was in question also to revert to spherical shot and shell. Spherical shot of hardest steel were tried by the _Excellent_, in the hope that they would penetrate 4½-inch plates. Experimental guns were also made, in 1864, to discharge 100-pound balls with charges of 25 pounds of powder; guns so heavy (6½ tons) that it was doubted at the time whether they could be efficiently worked on the broadside of a rolling ship. Should not increased power be obtained by persevering with rifled guns? The advantages possessed by the rifled gun in ranging power, accuracy, capacity of shell, were admitted; nevertheless the navy as a whole favoured the smooth-bore, with its simplicity, rapidity of fire, strength, and greater initial velocity, and thought that, at close ranges, the 100-pounder 6½-ton smooth-bore gun was the best and most suitable weapon for the service. But the rifled gun was advancing rapidly. “By May, 1864, the 7-inch muzzle-loading rifled shunt gun of 6½ tons had been tried in the _Excellent_, and had a good deal shaken the position of the smooth-bore. Captain Key reported that it was more than equal to naval requirements.... It was admirably adapted for the naval service.”[166] This fired a projectile 115 pounds in weight. By June of the following year the target of 9-inch plate representing the side of the _Hercules_ had beaten the latest Armstrong achievement, a 12½-ton 300-pounder. And on this pretext, and judging the defensive power of the whole ship by the defensive power of the thickest patch of its armour, a still more powerful gun was demanded for the navy by the inventor and by the press: a 25-ton 600-pounder.