Measurement of changes in the dimensions of Solid Bodies. 445 
Arrangement of a Horizontal Pendulum for measuring minute 
distances. 
The apparatus constructed by me was almost gales! of 
brass; the dimensions were as follows: the height of the 
nearly. The whole was supported on a well braced md 
firmly constructed wooden stand, which was hung by its upper 
end on the brick aig! of the college: building, and again at- 
tached to them below, so as to prevent vibrations independent 
of the walls of the bulldog itself. 
An inspection of Zdllner’s apparatus shows that its form was 
such as to render the column CO, fi g. 1, 
quite liable to vibrations of its 0 at 
this I guarded against by the addition 
of three long an nd the same number 
short cross-braces, as shown in fig. 3: 
they were found to make a most marked 
change for the better, and gave the in- 
strument a stability which it previously 
did not possess. 
Again, the German astronomer pro- 
— no means for bringing his instru- 
ment to rest, and in his published ob- 
hoyatione “this actually never was the 
case, the extremes of the oscillations 
cna to the Totals and nature of the 
9 al wholly out of the question, it 
with olive-oil. By removing with the pliers successive por- 
tions of the wire, it was found that matters could be arranged 
so that the instrument came to rest after a couple of oscil- 
lations. The oil-box rested on a little table provided with 
rack-work to adjust its height ie remove it when necessary ; 
it is shown at T in fig. 3; ig cone aie were as follows : 
length 70 mm., breadth 30 cic: F depth 32mm. The wire was 
placed of course as near the center of the box as possible, 
