190 AN EMPIRICAL STUDY OF GYRATING BODIES. 



Put the instrument in the position of fig. 45, and let it 

 rotate horizontally. The bottom will stay up and the 

 square keep its form for some time. Now stop the ro- 

 tation by applying the lingers to the upper brass piece, 

 b b, which is easily done without disturbing the "bal- 

 ance ;" it will instantly become limp, and take the form 

 shown in fig. 44. Leave it a few moments and it will be- 

 gin to go backwards, but there will be no vibration 

 whatever. 



As to whether, in a Motionless instrument, the " oscil- 

 lation would continue forever," lean here add nothing 

 to what was said under the gyroscope ; and if that es- 

 tablishes the descent in case of a f rictionless instrument, 

 it follows, of course, that the vibration also would cease. 



Experiment shows that the "gyrostats" arranged 

 as in the gyrostatic balance, do not act against each 

 other. It remains to show why they do not. 



Since only one pair of gyrostats is of any use in 

 the balance. I will use only one pair. Those shown in 

 fig. 46 weigh about two pounds. They are suspended 

 from the ceiling by a hook and cord. At h they are 

 connected by a broad, strong hinge, permitting great 

 freedom of vertical motion, but none of lateral. If, 

 now, both gyroscopes are set going at once, and in the 

 same direction as indicated by the arrows, we shall see 

 the whole system rotate around an axis passing through 

 the upper and lower hooks, instead of acting against 

 each other, and so preventing "any moment of mo- 

 mentum as a whole." 



The reason is easily seen if we remember that the 

 gyration of the end — the end that falls — is always in the 

 same direction as the under side of the wheel. There- 

 fore, h, in connection with the upper gyroscope, will 

 tend to move away from the observer. And p, the fall- 

 ing, or free end of the lower gyroscope, will tend to 

 move towards him. But a tendency for p to rotate in 



174 



