762 REPORT— 1897. 



The overhead trolley wire system therefore appears likely to come into still 

 greater use than has already been the case, if only on the g^'ound of economy ; but 

 in view of its admitted defects, the aiitlior has worked out an alternative method 

 which avoids the erection of trolley wires ahmg the streets above the tracks. 



4. On a Neio Method of Pleasuring Hysteresis in Iron. 



By J. L. W. Gill, B.A.Sc, of McGill University, Montreal. 



[Communicated by Professor Callexdar, M.A., F.R.S.] 



When a specimen of iron is passed to and fro through a naagnetic field without 

 any motion of rotation, the direction of the field being reversed each time the 

 specimen passes out of the field, the iron passes through a complete magnetic 

 cycle for each cycle of motion, and a definite amount of energy is lost, due to 

 hysteresis in the iron. Since energy is supplied only in the form of mechanical 

 work upon the specimen, the hysteresis loss is, by the law of the conservation of 

 energy, numerically equal to the resultant mechanical work expended. 



The instrimrent described below is based upon the above principle, and its 

 function is to measure the work so expended. 



The magnetic field is obtained by the use of a solenoid wound on a brass tube. 

 This solenoid is arranged vertically and has a vertical motion, the ends of the 

 solenoid being fitted with collars, which slide on two rigid vertical rods. An 

 arm is fastened rigidly to the solenoid; and extends out on one side. To this arm 

 is fastened a cord, which passes over a grooved pulley vertically above. A 

 balance-weight is attached to the other end of the cord. By rotating the gi-ooved 

 pulley the solenoid may be moved up and down, and will remain in any desired 

 position. The specimen to be tested is placed in a stirrup, which is sufficiently 

 small to pass through the solenoid, and is suspended by a helical spring, the point 

 of suspension being vertically above the centre of the solenoid. Another helical 

 spring extends from the bottom of the stirrup to a point vertically below. Tliis 

 serves to keep the stirrup steady. The stiiTup is suspended so that when the 

 solenoid is in its lowest position the specimen is out of the magnetic field, being 

 above the solenoid. As the solenoid is moved up the stirrup and specimen pass 

 through it, and when the solenoid is in its highest position the specimen is prac- 

 tically out of the field. If the solenoid be moved once up and down, the field 

 being reversed when the specimen is out of it, the specimen passes through a 

 •complete magnetic cycle, provided the specimen has been once through the field 

 and is initially in that particular cyclic state. 



As the solenoid is moved up, the specimen is attracted down, the force of 

 ;attraction increasing until it reaches a maximum when about one-half of the 

 specimen is inside the solenoid. The attraction then decreases and becomes zero 

 when tlie specimen is in the centre of the solenoid. Up to this point work is 

 being done by the magnetic force. As the solenoid is moved up to its highest 

 position the specimen is attracted upward, and work is done against the magnetic 

 force ; the attracting force becomes a maximum when the specimen is about one- 

 half out of the solenoid on the lower side, and becomes zero when the solenoid is 

 in its highest position. The maximum force in the second half of the motion is 

 greater than the maximum force in the first half. The work done in the second 

 half of the motion is also greater than that done in the first half, the difference 

 being the work expended in taking the specimen through half a cycle. When the 

 field is reversed and the solenoid moved down, the action is similar to that which 

 takes place when the solenoid is moved up, and the resultant work done will be 

 the same, provided the specimen is homogeneous. 



The resultant work done on the specimen may be determined by observing the 

 attracting force when the solenoid is in different positions, and then drawing a 

 -distance-force curve. The integrral of this curve gives the resultant work done on 

 the specimen. The force at different points can be determined by calibrating the 

 ■springs which support the stirrup, and then observing the extension of these 

 springs. The author has determined the hysteresis loss in different specimens at 



