178 Trans. Acad. Sci. of St. Lowis. 
enclosing these suspension fibers, was a lateral brace 
formed of two bars of hard rubber, about 30 cm. in 
length. The ends of these bars formed a right angle 
against which the vertical suspension cylinder was 
pulled. These rubber bars formed the ends of two long 
bars of wood attached to the tops of instrument cases 
along adjoining sides of the room. The bars were also 
supported by a framed structure. Silk cords attached 
to the cases ran along the two bars and were tied to the 
suspension cylinder. Weights distributed along these 
cords served to hold the cylinder against the ends of the 
rubber bars. Vibrations of the suspension cylinder were 
thus wholly eliminated. The torsion head forming the 
top of the cylinder was provided with a rod which ex- 
tended radially outward to a distance of about 20 em. 
By this means the torsion head could be turned in order 
to put the suspended masses into vibration. The top of 
the suspension cylinder had a similar rod clamped to it 
having an upward projecting stop, by means of which 
the torsion head could be returned to the original posi- 
tion. Change in the temperature of the air within the 
shield resulting in a change in volume was provided for 
by the outflow or inflow around the torsion head. 
The enclosing case was of the general form used by 
Cavendish. In the early part of the year it was formed 
wholly of metal. It rested upon insulating blocks of 
rubber, which were mounted upon two long pieces of tim- 
ber having a cross section of 4x4 inches. These timbers 
rested upon a massive table, which was on a floor of re- 
inforced conerete within a building having granite walls. 
At the central part of the metal shield the dimensions 
were 5x5 inches. A narrow slit on one side covered by 
a small plate of glass sealed to the shield, served for ob- 
serving the movement of the suspended masses by means 
of a telescope and scale. At the outer ends of the shield, 
the lateral dimension was about 15 inches. This form 
was chosen in order that the suspended masses and the 
brass tube upon which they were mounted could be in- 
