434 Proceedings of Royal Society of Edinburgh. [sess. 
that, if the image were broadened one-tenth part, the duration of 
the radiation must have been about 
i x tV x to x 5T?T x TT =TToW second. 
We may, therefore, conclude that the duration of the Rontgen 
rays produced by each discharge cannot be greater than T qoW 
second, and shall use this as an upper limit. 
This number is less than any given by those who have pre- 
viously determined the duration of the Rontgen rays. 
A. Roiti (Rendiconti della R. Acc. dei Lincei , vol. v. p. 243), 
who seems to have been the first to attempt to do this accurately, 
used a Ruhmkorff coil with a rotating interrupter to make and 
break the current in the primary circuit. On the interrupter he 
mounted a photographic plate, and in front of it had a screen of 
lead with a slit in it. The Rontgen rays passing through the slit 
produced a photograph of it on the plate, which appeared broadened 
into a small sector of a circle. From the amount of broadening, 
Roiti concluded that the duration of a discharge was about 
second. 
Dr E. Trouton ( Report of the British Association , Liverpool, 
1896) adopted the method of rotating a zinc-toothed wheel between 
the Rontgen lamp and the photographic plate. A photograph of 
the moving teeth was obtained by making one interruption of the 
primary current in the induction coil, and so allowing one discharge 
to pass. The departure from sharpness of outline of the photo- 
graph of the teeth indicated the time that the radiation lasted. 
The results obtained in this way varied from g~ to xo.Voo’ 
second. 
M. Colardeau ( Edairage electrique , vol. viii. p. 112), using a 
similar method, found the duration to be about T oVo second. 
H. Morize ( Comptes rendus , vol. cxxvii. p. 546), like Roiti, 
makes the photographic plate revolve behind a slit in a metal 
screen. The plate is fixed to one end of the axle of an electro- 
motor, and at the other end is a toothed wheel which interrupts 
an electric current, and makes a contact which registers itself as 
well as the mean seconds of an electric chronometer on the band of a 
Breguet chronograph. He is thus able to determine the speed of 
the motor at any moment. From the broadening of the image of 
