The latent energy remaining in metal after cold working is measured by observing the energy absorbed during plastic deformation and the heat released. The study shows that more cold work can be done on a metal in torsion than in direct tension. As the amount of cold work increases, the proportion of energy absorbed decreases. For copper, the latent energy required to reach saturation at 15°C is slightly more than 14 calories per gram. Using compression instead of torsion allows more cold work to be done on copper, with compressive stress increasing until the total cold work is equivalent to 15 calories per gram. The absorption of latent energy and the increase in strength both cease when the same amount of cold work has been applied, suggesting that the strength of pure metals depends on the amount of cold work latent in them. When a metal is cold worked, most of the work done is converted into heat, but a portion remains latent and is associated with changes in the metal's physical properties. This latent heat must be released before the metal reaches its melting point, and when dissolved, it appears as a heat of solution. The latent energy of cold working can be measured either when energy is added to the metal or when it is released. In the former case, the work done and the heat evolved during plastic deformation are measured. The difference is the latent energy of cold working. This method was used by Farren and Taylor and by Hort, who found that 5.5 to 13.5% of the work done remains latent in the metal. Attempts to measure the latent energy of cold working when it is released have yielded inconsistent results. In Farren and Taylor's experiments, the cold work was done by stretching a metal rod, but only a limited amount of work could be done before the rod broke. For copper, the maximum latent energy measured was equivalent to a temperature rise of only 0.83°C. To release the latent energy of copper, it is necessary to raise its temperature to 500°C or more. The latent energy would appear as the difference between the heat required to raise the temperature of two equal specimens to 500°C, one of which had been cold worked and the other not. Measuring this difference is equivalent to measuring the difference between the specific heat of the two metals, which is only 1 part in 600. It is difficult to achieve such accuracy in heat measurements, but by using metals that have undergone more severe cold working, greater latent heat can be obtained, and the accuracy of the measurements of the energy released on heating can be increased. It is well known that more cold work can be done on a metal rod by twisting it than by stretching it. The amount of work done in direct extension depends on the load-extension curve. The condition for fracture by instability due to the formation of a local "neck" is given by a specific equation. The criterion for fracture can be found by plottingThe latent energy remaining in metal after cold working is measured by observing the energy absorbed during plastic deformation and the heat released. The study shows that more cold work can be done on a metal in torsion than in direct tension. As the amount of cold work increases, the proportion of energy absorbed decreases. For copper, the latent energy required to reach saturation at 15°C is slightly more than 14 calories per gram. Using compression instead of torsion allows more cold work to be done on copper, with compressive stress increasing until the total cold work is equivalent to 15 calories per gram. The absorption of latent energy and the increase in strength both cease when the same amount of cold work has been applied, suggesting that the strength of pure metals depends on the amount of cold work latent in them. When a metal is cold worked, most of the work done is converted into heat, but a portion remains latent and is associated with changes in the metal's physical properties. This latent heat must be released before the metal reaches its melting point, and when dissolved, it appears as a heat of solution. The latent energy of cold working can be measured either when energy is added to the metal or when it is released. In the former case, the work done and the heat evolved during plastic deformation are measured. The difference is the latent energy of cold working. This method was used by Farren and Taylor and by Hort, who found that 5.5 to 13.5% of the work done remains latent in the metal. Attempts to measure the latent energy of cold working when it is released have yielded inconsistent results. In Farren and Taylor's experiments, the cold work was done by stretching a metal rod, but only a limited amount of work could be done before the rod broke. For copper, the maximum latent energy measured was equivalent to a temperature rise of only 0.83°C. To release the latent energy of copper, it is necessary to raise its temperature to 500°C or more. The latent energy would appear as the difference between the heat required to raise the temperature of two equal specimens to 500°C, one of which had been cold worked and the other not. Measuring this difference is equivalent to measuring the difference between the specific heat of the two metals, which is only 1 part in 600. It is difficult to achieve such accuracy in heat measurements, but by using metals that have undergone more severe cold working, greater latent heat can be obtained, and the accuracy of the measurements of the energy released on heating can be increased. It is well known that more cold work can be done on a metal rod by twisting it than by stretching it. The amount of work done in direct extension depends on the load-extension curve. The condition for fracture by instability due to the formation of a local "neck" is given by a specific equation. The criterion for fracture can be found by plotting