TRANSACTIONS OF SECTION G. 763 



different iuductiona by tliis method, the extension of the springs being observed 

 with the aid of a microscope. 



To make this method practical, a simple integrating apparatus is attached to 

 the instrument above described, by which the work done is integrated automati- 

 cally. A glass disc is connected rigidly to the pulley which moves the solenoid, 

 so that when the pulley rotates, the glass disc rotates in its own plane, which is 

 vertical, the axis of rotation passing through its centre. The motion of the glass 

 disc is therefore proportional to the motion of the solenoid. An arm is fastened 

 to the stirrup which supports the specimen, and extends up to the glass disc. 

 This arm supports a graduated steel disc, which is free to rotate in its own plane 

 about a vertical axis through its centre. This steel disc presses lightly on the 

 glass disc, the point of contact being at the centre of the glass disc. AVhen the 

 solenoid is moved up, the specimen is attracted down, taking the stirrup with it. 

 This causes the steel disc to recede from the centre of the glass disc. A motion of 

 rotation is then communicated to it by the glass disc, the speed of rotation 

 depending on its distance from the centre ; since its distance from the centre at 

 any instant is proportional to the attracting force, and the motion of the glass 

 disc is proportional to the motion of the solenoid, the speed of rotation of the disc 

 is proportional to the work being done at that instant. The total amount of 

 rotation is therefore proportional to the total work done. Consequently all that 

 is necessary to test a specimen with this instrument is to place the specimen in the 

 stirrup, move the solenoid up and down to get the specimen in a cyclic state, then 

 take it through a cycle, and observe the amount of rotation communicated to the 

 steel disc. This is a direct measure of the work expended. 



The constant of the instrument is determined by placing a known weight in 

 the stirrup, and observing the amount of rotation communicated to the disc when 

 the solenoid is moved through a known distance. 



The specimen may be taken through a number of cycles and the readings 

 allowed to accumulate. Tlie average of a number of cycles is thus obtained. 



5. A New Method of Investigating the Variation of the Magnetic Qualifies 

 of Iron icith Temperatxtre. By F. H. Pitcher, M.A.Sc., Demonstrator 

 of Physics, McGill University, Montreal, 



[Communicated by Professor H. L. Callendar, M.A., F.R.S.] 



Owing to the apparent lack of exact knowledge on the subject of the variation 

 of hysteresis in iron with temperature, and as it is of some importance in the work- 

 ing of transformers, it was thought well to investigate the subject further. 



At the same time it was intended to repeat the experiments of Hopkinson and 

 others on magnetism at high temperatures, by a different method and with higher 

 fields. For this purpose a new method, devised by Professor Callendar, was 

 employed. 



Description of the Method. 



The specimen of iron in question was in the form of a wire, and was tested by 

 the direct magnetometric method in-the broad side-on position. The first intention 

 was to insulate the specimen in a small and Tery thin platinum tube, heated by 

 having a current passed through it. The temperature of the specimen was to be 

 inferred from tlie resistance of the platinum tube over tlie length occupied by the 

 iron wire specimen. The resistance was to be measured by the fall of potential 

 between the terminals of very fine platinum wire leads, attached to the platinum 

 tube at the ends of the above length. 



In this way, if the tube were made considerably longer than the wire, the 

 middle portion occupied by the wire would be very uniformly heated, and very 

 exact values of the mean temperature could be obtained. The slack of the 

 platinum tube when heated was arranged to be taken up by copper springs. To 

 prevent oxidation of the iron wire specimen, and at the same time to promote a 



