BETWEEN SURFACES MOVING- AT LOW SPEEDS. 
513 
oscillate about the line joining these two bearing points. Its period can be altered by 
shifting up or down the bob e. At f a small cradle is soldered to the pendulum-rod, 
and carries a fine glass siphon (g), the shorter end of which dips into a box ( h ) containing 
ink (aniline blue dissolved in water). The longer end ( i ) of the siphon is bent in the 
manner shown in fig. 1, and stands at a distance of rather less than one tenth of an inch 
from a strip of paper (E E) two and a half inches broad which is fastened round the circum- 
ference of the disk, and projects about three quarters of an inch on each side of it. The 
lateral stiffness of the paper makes it assume a cylindrical form when placed round the 
disk, and the ends are fastened together so as to make the cylinder complete. The 
breadth of the paper is somewhat greater than the maximum amplitude of oscillation of 
the point i of the siphon fixed to the pendulum. In order to make the ink run through 
the siphon and be deposited on the strip of paper, the ink is continually maintained in 
a state of electrification. This is effected as follows : — The ink-box (h) stands on an insu- 
lating rod of vulcanite (k), and has fixed to it a small horizontal brass plate (m). At a 
short distance above m is the point of a brass rod (l) which slides up and down in a V- 
groove cut in the side of n, another rod of vulcanite forming the bracket to which the 
knife-edge of the pendulum is secured. The rod l is pressed into the notch on n by the 
spring o. To the top of the rod l a wire is fastened which leads to an inductive electrical 
machine. The machine which we employed was identical in construction with the 
“ mouse-mill ” which is used to electrify the ink in the siphon recorder. (See Sir W. 
Thomson’s ‘Electrostatics and Magnetism,’ and ‘Journal of the Society of Telegraph 
Engineers,’ vol. v. 1877.) The plate m becomes electrified by aerial convection from 
the point of the rod l. The rate of electrification of the ink may be varied by raising 
or lowering the rod l. 
When the siphon becomes electrified to a certain extent, the attraction so developed 
between its point i and the paper strip is sufficient to cause the long limb of the siphon 
to bend until the point i, or rather the particle of ink projecting from it, just touches 
the paper. When this takes place a very small drop of ink is deposited on the paper, 
and the siphon, ink-holder, and plate m are instantaneously diselectrified. The point i 
then recedes from the paper, drawn back by the elasticity of the long limb, until the 
electrification (continuously communicated by the rod l) is again sufficient to cause an 
advance of i towards the paper, when another drop of ink is deposited, and so on. The 
point of the siphon is by these means kept in a state of rapid vibration towards and 
from the paper, every advance being accompanied by the deposit of a particle of ink. 
The rate of electrification is adjusted (by moving the rod l up or down) so that the 
time taken to recharge the siphon, inkholder, and plate corresponds to the period of 
vibration due to the elasticity of the glass. If the electrification be too rapid the point 
i will be checked in its recession from the paper before the completion of its semiperiod 
of free vibration. If, on the other hand, the electrification be too slow, the impulse 
given by the new accumulation of electricity following each discharge will not come soon 
enough. The adjustment of the rate of electrification is a matter requiring some 
