Oscillations of the Horizontal Needle. 11 
amination, that comparatively slight differences in the flexibility 
of the silk, change in some instances the rate of vibration. 'To 
avoid this source of inconvenience, it is desirable to prepare a 
great number of suspensions from the unspun fibre, so as to be 
without the least torsion, and as nearly alike as possible, and sub- 
mit each to experiment on the same bar. Their identity as to 
effect may then be determined. Each silk should be gently 
drawn through between the fingers, slightly smeared with some 
extremely fine oil, and finally wound upon a small cork, and laid 
aside in a box for use. 
The bar is easily hung to the silk by means of the hook 
above described, and a fine loop tied in the extremity of the silk. 
A similar method is employed, in attaching the opposite end to 
the suspension point. 
20. The possibility of change in the length of a vibrating 
bar, in consequence of variations in temperature, is very apparent. 
To arrive at the requisite correction, it seems desirable to make 
the particular magnet an immediate subject of experiment. This 
may be managed by means of the instrument above described, 
Fig. 2, with advantage. 
The bar being suspended in the exhausted receiver, and its 
rate of vibration determined in a very small are, under a given 
temperature, the lower portion of the receiver is then enclosed 
in a small oven of copper 0, Fig. 6, and a stream of heated air 
made to circulate about the glass, by means of a lamp placed un- 
dcr the funnel 7. When the temperature has been raised as 
many degrees as is considered requisite, the oven is removed, 
and the rate of vibration again taken as before ; and this process 
may be repeated through a range of temperature sufficient for 
our purpose, that is to say, up to 200 degrees of Fahrenheit. For 
the sake of greater accuracy, an additional and very delicate ther- 
mometer may be placed near the magnet. 
21. I have also employed water with advantage, for the pur- 
pose of observing the effects of heat on the needle. An outer 
B2 
