Intelligence and Miscellaneous Articles. 899 



tional to its magnetism at 0°. With a magnet A, the ratio 



".. — — ° was found equal to — — , -r^TT"- and — — ■ in three series, 

 JV1 lU'o lU'b 1U"5 



for values of M varying nearly as 1, 2, and 4. With another 



magnet D, the same ratio has the values — , — -, and 7-- for suc- 



4*7 4*y O'U 



cessive magnetizations. 



The value of the ratio found constant for the same bar, whatever 

 may be the intensity of its magnetism, depends, as is well known, 

 on the form of the bar, on the nature of the air, and on the tem- 

 pering. The author's researches are not sufficient for the estima- 

 tion of the influence of these different elements. A single bar B 

 was annealed, when the ratio mentioned above became equal to 

 one-fifteenth to one- tenth of what it was before. This experiment 

 agrees with a slight diminution which the ratio undergoes in succes- 

 sive series, because the tempering also diminishes a little. With 

 one bar A, the ratio increased ; and it is to be observed that this 

 bar had been submitted to fourteen alterations of temperature before 

 the first series, from which it may be considered that the tempering 

 had become stationary. According to M. Wiedemann, when a mag- 

 net returns to the initial temperature, after having been submitted 

 to a high temperature, the diminution of the magnetic force is pro- 

 portional to the force itself, and an apparent non-agreement with 

 this rule may result from a change in the tempering. According to 

 M. Mauritius, the experiments made with the bar A contradict this. 



M — M' 

 In fact the author finds for the ratio — ^r= — -, which should be 



M 



constant according to M. Wiedemann, the values — , — , and — 



tjy 40 o 1 



after the first elevation of temperature in each of the three series 



with A, of which the tempering, as mentioned above, had become 



constant. 



Temporary Magnetism. — The method of measuring is the same, 

 and the bar, the temporary magnetism of which is to be studied, 

 is placed in an electro-magnetic coil arranged in the same way as 

 the permanent magnet in the preceding experiments. Each experi- 

 ment is made in the following manner. The current is passed 

 through the coil and the deviation measured ; the red-hot bar of 

 iron is suspended inside the coil ; and starting from this moment, 

 the position of the moveable magnet is observed, noting the moment 

 at which each observation is made. By this means the variation, 

 with the time, of the temporary magnetism of the bar of iron is 

 measured. Four different bars were employed — one of wrought 

 iron, two of cast iron, and one of cast steel. 



Results. — At a bright red heat none of the bars were magnetic. 

 Approximate determinations, by immersion of the bars in water, of 

 the temperature at which magnetism began to be manifested, gave 

 1000°. 



With the steel bar the increase of magnetic power takes place at 



