116 SCIENTIFIC AND TECHNICAL LABOURS TO 1850. 



moveable nut on the screw in the same direction. 

 The nut must then move right or left on the screw 

 as long as it turns quicker or slower than the screw. 

 and can thus perfectly regulate the pace of the engine, 

 immediately ceasing to move, when the velocity of the 

 engine is precisely equal to that of the circular pen- 

 dulum. The differential regulator (or chronometric 

 governor, as brother William, who practically elabo- 

 rated and mainly perfected it. afterwards called it in 

 England), constructed on this principle, has certainly 

 not been largely introduced into practical engineering. 

 It is neither so simple nor so cheap as the Watt- 

 regulator, which in later years has been considerably 

 improved, but the differential movement, which we 

 carried out in the most varied forms, has proved an 

 exceedingly fertile element of construction. 



My occupation with the problem of the exact 

 measurement of the velocity of projectiles, imperfectly 

 solved by Leonhardt's ingenious clock, caused me to 

 perceive that only a method, in which no masses 

 had to be set in motion and brought to rest, could 



o 



lead to the goal. Thus I came to employ the electric 

 spark for the solution of the problem. My proposal 

 consisted in causing electric sparks to pass on to a 

 rapidly and uniformly rotating polished steel-cylinder 

 from a fine point approximated as far as possible to 

 its periphery, and in calculating, from the interval 

 between the marks produced by these sparks and the 

 known number of revolutions of the cylinder, the 

 velocity of the ball, which at particular stages of its 



