_Effect of Stretching. Hartmann’s Experiment._—When a cylindrical rod of metal, held firmly at either end—a test-piece, as it is called in metallurgy—is pulled sufficiently hard, it often elongates considerably, part of the elongation disappearing as soon as the strain ceases, and the other part remaining. The total elongation is thus the sum of an “elastic elongation,” which is temporary, and a “permanent elongation.” If we continue the stretching, there appears at some point of the rod a local extension with contraction of sectional area. It is here that the rod will break. But in place of continuing the stretching, Mr. Hartmann suspends it. He stops, as if to give the “metal-being” time to rally. During this delay it would seem that the molecules hasten to the menaced point to reinforce and harden the weak part. In fact the metal, which was soft at other points, at this spot looks like tempered metal. It is no longer extensible. When the experimenter begins the stretching again after this rest, and after the narrowed bar has been rolled and become cylindrical again, the local extension and sectional contraction is forced to occur at another point. If another rest is given at this point the metal will also become hardened. If we repeat the experiment a sufficient number of times, we shall find a total transformation of the rod, which becomes hardened throughout its entire extent. It will break rather than elongate if the stretching is sufficiently severe. _Nickel Steels—their “Heroic” Resistance._—Nickel steels present this phenomena in an exaggerated degree. The alternation of operations which we have just described, bringing the various parts of an ordinary steel rod into a tempered state, is not necessary with nickel steel. The effect is produced in the course of a single trial. As soon as there is any tendency to contraction the alloy hardens at that precise place; the contraction is hardly noticeable; the movement is stopped at this point to attack another weak point, stops there again and attacks a third, and so on; and, finally, the paradoxical fact appears that a rod of metal which was in a soft state and could be considerably elongated has now become throughout its whole extent as hard, brittle, and inextensible as tempered steel. It is in connection with this point that M. C. E. Guillaume spoke of “heroic resistance to rupture.” It would seem, in fact, as if the ferro-nickel bar had reinforced each weak point as it was threatened. It is only at the end of these efforts that the inevitable catastrophe occurs.