Putting aside the question of calibre and volume of the bullets we are concerned with, I believe the most important variations as serious effects of ricochet depend on the relative thickness and the composition of the mantles. Fig. 26 illustrates the relative thickness of the mantles in the Krag-Joergensen, Mauser, Lee-Metford, and Guedes bullets. Given an equal degree of force and velocity on the part of the bullet at the moment of impact, the assumption is justifiable that the thinner mantles would tear or burst more readily in direct ratio to their relative thinness. I believe this assumption to be borne out by my own experience of the common deformities that occurred; but the great relative frequency with which Mauser bullets came under my observation, and the difficulty of forming any estimate of the velocity and force retained by any particular bullet at the moment of impact, make it impossible for me to express myself with the confidence which I should wish. [Illustration: FIG. 27.--Normal Mauser Bullet] The second condition which influences the nature and degree of the deformities depends on the relative tenacity or brittleness peculiar to the metal employed in the manufacture of the mantles. In the case of the Lee-Metford this consists of an alloy of 80 parts of nickel with 20 of copper. The Krag-Joergensen and Mauser are ensheathed in steel plated with cupro-nickel, and the Guedes has a plain steel envelope coated with wax. Both as a result of experience in the field gained from ricochet bullets, and in the hospitals from bullets which had undergone deformation within the body, I am under the firm impression that the thin nickel-plated steel envelope of the Mauser bullet splits more readily than the thicker and more tenacious cupro-nickel envelope of the Lee-Metford, that the direction of the ruptures is more purely longitudinal, and the fissuring itself more extensive and complete. I append below a series of deformities observed in Mauser bullets, some of which were collected on the field of battle, but all of which were familiar to me in bullets removed from the bodies of patients, except the complete disc shape shown in fig. 29. They correspond with specimens of which I made sketches at the time of removal from the body, but which I had not the heart to retain in view of the natural wish of the patients to keep them as mementoes of their wounds. [Illustration: FIG. 28.--Four common types of lateral Mauser Ricochet Bullets.