Before coming to these two conclusions Delvigne had made important efforts to render the spherical ball as efficient as possible. He had, in particular, proposed to make that part of the barrel near the breech which formed the powder-chamber of slightly smaller diameter than the rest of the barrel; so that a spherical ball, rammed down on it, became indented against its ledge and flattened sufficiently to fill the rifling grooves. By this device quick loading was obtained and the accuracy of aim, it was found, was doubled. Certain practical disadvantages, however, were associated with it: the chamber fouled rapidly, and the ball was frequently distorted and jagged by over-ramming. So in ’33 the Delvigne system, as it was called, was modified by the wrapping of the ball in a greased patch and the attaching of the patch to a “sabot” or wad of wood which was interposed between the ball and the shoulders of the powder-chamber. Rifles thus loaded did good work in Algeria in ’38. In the meantime Delvigne, admittedly inspired by the writings of Robins, was urging on the authorities the superiority of the elongated ball. He was insistent on the advantages which would accrue from augmenting the mass of the projectile while at the same time making it present to the air during flight its smallest surface. The shape he proposed was that of the present-day rifle bullet, considerably shortened: a bullet with a flat base, cylindrical sides and ogival head, somewhat resembling the form which had been proposed by Sir Isaac Newton as a “solid of least resistance.” After a succession of disappointments and refusals, the inventor had the satisfaction of seeing his bullet accepted. Its advantages over the spherical ball had been made manifest on the proving-ground. It was accepted in combination with the _carabine à tige_, a rifle invented by a Colonel Thouvenin, in which the Delvigne shouldered chamber was replaced by a small central pillar or anvil, projecting from the breech-end of the bore, against which the bullet was rammed. The powder, when poured into the barrel, collected in the annular space around the pillar. By this arrangement the necessity for the sabot was obviated and the charge of powder, protected by the pillar, was not in danger of being crushed or mealed. In ’46 the new bullet proved its high accuracy and ranging power on active service in Algeria. But the pillar was found liable to bend and distort; and the difficulty in keeping the space round it free from fouling proved to be another of its inherent disadvantages. [Illustration: “CARABINE À TIGE”] [Illustration: MINIÉ BULLET]