very Small Portions of Time. 317 



dark spot which does not vanish until the surface has been 

 polished anew. The electric spark endures only for an immea- 

 surable instant of time ; the propagation of the electricity through 

 a wire of the length which comes into action here is also effected 

 in a space of time vanishingly small. If, therefore, in the path 

 of the ball a net be placed composed of insulated wires, which are 

 alternately put in connexion with the external and internal coating 

 of a Ley den jar, the rotating cylinder with the point being also 

 included in the circuit, then the discharge will take place at 

 the moment when the ball by passing through the net esta- 

 blishes the metallic connexion between its different wires. At 

 the same moment the mark is made upon the surface of the 

 cylinder. For one and the same ball any number whatever of 

 wire nets may be made use of and corresponding marks obtained. 

 This method permits of our availing ourselves of the entire 

 accuracy of which the rotating cylinder is capable. If we omit 

 taking into account the displacement of the net by the com- 

 pressed air which precedes the ball, the time during which the 

 ball passes over the space of half a line might be measured. 



Further, the graduation and velocity of the cylinder stated 

 above are by no means near the limits to which both might be 

 extended ; we can, indeed, form no judgement of the extent to 

 which the accuracy of such measurements might be carried. 

 Another process, indeed, depending upon a different application 

 of the same principle, has led to far more astonishing results. 

 In the case of the phenomena of light whose duration is to be 

 determined, the difficulty is greatly diminished by making use 

 of a rotating mirror instead of a cylinder. When we look at a 

 mirror and slowly change the inclination of its surface, the images 

 of external objects seem at the same time to move, and moreover 

 with a velocity equal to double of that of the mirror. They move 

 in a circle, the centre of which is in the axis of the mirror, and 

 whose radius is equal to the distance of the object from the said 

 axis. For the measurement of time, the image thus moving in 

 a circle takes the place of the rotating cylindrical surface. Its 

 momentary position may be observed with a telescope With the 

 same exactitude as the mark upon the cylinder surface with the 

 microscope. As, however, it moves in a circle of far greater 

 radius, the accuracy of the measurement, when the telescope is a 

 good one, increases in the proportion in which the radius of the 

 apparent path exceeds that of the steel cylinder. The advantage 

 is, that without increasing the mass of the moving body, by 

 which the mechanical difficulties of construction would be greatly 

 augmented, the velocity, by means of which we convert the time 

 into apparent space, increases in such a great ratio. Supposing, 

 for example, two electric sparks to occur successively at the same 



