C. B. WARRING. 161 



gyrostat to a wire yoke, and then suspend from the ceil- 

 ing, as fig. 35. 



Set the wheel in rapid motion, and leave the machine 

 to itself. It will keep its vertical position, although 

 swung back and forth and roughly handled. 



The explanation is the same as when the gyrostat 

 stands on a plate of glass. There is the same rapid re- 

 versal of positions and consequent play of forces as was 

 seen in fig. 33. 



STABILITY. 



By stability I understand that condition or property 

 by which a body, if disturbed from its position, returns 

 to it unless forcibly prevented. The body is said to be 

 stable. A weight hanging below the point of support is 

 stable. A rod balanced on its lower end is unstable. 

 A body, therefore, in a stable position, will never leave it. 



If these definitions are correct, and if Thomson & 

 Tait use the word in that sense, their statements on pp. 

 397 and 398 appear not to bear the test of experiment. 



Out of their seven "stabilities," only one (viz., No. 1, 

 p. 399), is stable in the above sense. 



If we define stability to be nothing more than resist- 

 ance to acceleration of motion, then their seven "sta- 

 bilities' 1 are justified by experiment ; i. e., if you push 

 the gyrostats out of position faster than they are going 

 of themselves, you will feel a resistance. 



There will not, however, be a going back to their posi- 

 tion, except in case of the stilted gyrostat, where ap- 

 parently there is an exception. It does go back to its 

 first position, but does not stop there, as a. stable body 

 would. It goes beyond and then back, each time in- 

 creasing the amplitude of the arc, till it falls. This can 

 hardly be called stability. 



GYROSTATIC TIGHT ROPE PERFORMANCE. 



There is a form of the gyroscope, fig. 36, which is often 



145 



