110 AN EMPIRICAL STUDY OB^ GYRATING BODIES. 



fests itself (1) at the ends of the axle, and (2) at the 

 point of support and in the resistance of the air. The 

 sole effect of the first is to reduce the rate of rotation ; 

 consequently, if sufficient energy is constantly added to 

 replace what is lost in this way, this rotation will remain 

 constant. 



This is accomplished by the electro- magnetic gyros- 

 copes invented by Mr. George M. Hopkins, of New York, 

 figured and described in the Scientific American, vol. 

 xxxviii, p. 335. This ingenious instrument gives a contin- 

 uous rotation which, for a short time, may be regarded 

 as uniform ; but, owing to the pressure of the connect- 

 ing wire against the standard, a disturbing element is in- 

 troduced, which greatly retards the gyroscope, and, 

 consequently, on the principle stated under the twelfth 1 

 experiment, causes it to fall abnormally fast. 



To obviate this, I devised the instrument shown in fig. 

 4, where it is represented as suspended from the ceiling 

 by a fine cord. From a to m there is metallic connection, 

 b andh being mercury cups, except at d and c, where the 

 electrical continuity is broken by short silk cords. These 

 compel the current to pass to the circular frame, and 

 thence around the magnet seen in the diagram. On the 

 opposide side, out of sight, is a current breaker, by 

 which the soft iron is made, and unmade, a magnet twice 

 in each revolution. 



By this arrangement, the only resistance to horizontal 

 rotation comes from the torsion of the string. This is 

 easily obviated by twisting the string so that its torsion 

 is a slightly accelerating force, and therefore, so far as 

 it goes, tending to lift up the loaded end. The pins d 

 and g are in line as nearly as possible with the centre of 

 gravity. This arrangement gives great freedom of move- 

 ment vertically as well as horizontally. In line with the 

 axis are two hooks, one at each end, to which a weight 



1 Retarding the gyroscope or pulling it backward causes it go down. 



94: 



