M. Tresca has lately presented to the Academy of Sciences some very interesting experiments on the development and distribution of heat produced by a blow of the steam hammer in the process of forging. The method used was as follows: The bar was carefully polished on both sides, and this polished part covered with a thin layer of wax. It was then placed on an anvil and struck by a monkey of known weight, P, falling from a height, H. The faces of the monkey and anvil were exactly alike, and care was taken that the whole work, T = PH, should be expended upon the bar. A single blow was enough to melt the wax over a certain zone; and this indicated clearly how much of the lateral faces had been raised by the shock to the temperature of melting wax. The form of this melted part could be made to differ considerably, but approximated to that of an equilateral hyperbola. Let A be the area of this zone, b the width of the bar, d the density, C the heat capacity, and t-t0 the excess of temperature of melting wax over the temperature of the air. Then, assuming that the area, A, is the base of a horizontal prism, which is everywhere heated to the temperature, t, the heating effect produced will be expressed by Ab x d x C(t-t0) Multiplying this by 425, or Joule's equivalent for the metrical system, the energy developed in heat is given by T1 = 425 AbdC(t-t0). Dividing T1 by T, we obtain the ratio which the energy developed in heat bears to the total energy of the blow. With regard to the form of the zone of melting, it was found always to extend round the edges of the indent produced in the bar by the blow. We are speaking for the present of cases where the faces of the monkey and anvil were sharp. On the sides of the bar the zone took the form of a sort of cross with curved arms, the arms being thinner or thicker according to the greater or less energy of the shock. These forms are shown in Figs. 1 to 6. It will be seen that these zones correspond to the zones of greatest sliding in the deformation of a bar forged with a sharp edged hammer, showing in fact that it is the mechanical work done in this sliding which is afterward transformed into heat. [Illustration] With regard to the ratio, above mentioned, between the heat developed and the energy of the blow, it is very much greater than had been expected when the other sources of loss were taken into consideration. In some cases it reached 80 per cent., and in a table given the limits vary for an iron bar between 68.4 per cent. with an energy of 40 kilogram-meters, and 83.6 per cent. with an energy of 90 kilogram-meters. With copper the energy is nearly constant at 70 per cent. It will be seen that the proportion is less when the energy is less, and it also diminishes with the section of the bar. This is no doubt due to the fact that the heat is then conducted away more rapidly. On the whole, the results are summed up by M. Tresca as follows: