QUARTZ FIBRES AND ITS TKMPKRATURE COEFFICIENT. 



401) 



the end of the steel rod E (fig. 2). This passes through the torsion head D, and its 

 length is sufficient to allow of its end heing lowered so as to be well outside the 

 jacket, for the purposes of fixing the fibre in position and of measuring its length. 

 The fibres were attached to the brass pin and to the vibrator sometimes by shellac, 

 sometimes by silvering, electroplating with copper and soldering, and sometimes by 



means of a tooth -stopping cement, suggested to 

 me as a suitable material by Mr. C. V. BOYS, and 

 supplied by Messrs. C. ASH, Limited, of Golden 

 Square, W. This last method has the advantage 

 that it does not involve heating the fibre, and 

 thereby possibly altering its rigidity. It was found 

 to give a suspension free from zero changes, and in 

 every respect as good as shellac. I had hoped that 

 it would withstand the effects of heat, but the 

 cement always cracked away from the metal at 

 about 50 C. 



The jacket H consists of two concentric tubes. 

 The outer one is of brass, 3 inches in diameter, and 

 is covered externally with layers of cotton wool, 

 around which a piece of sheet asbestos is wrapped 

 to protect it from fire. The inner tube is of copper, 

 i inch in thickness and 1 inch in internal diameter. 

 This has two side tubes, each inch in diameter. 

 One of these is at the middle of the jacket, and 

 through this the thermometer T, used to indicate the 

 temperature of the jacket, projects. The second 

 side tube, which is not shown in the figure, is at 

 right angles to the first and nearer to the lower end 

 of the jacket. The outer end of this tulw is covered 

 with a piece of optically worked glass, and through 

 it the mirror on the vibrator can be seen. 



In order to set the vibrator in oscillation without 

 swinging it like a pendulum bob, the following 

 device, already described in a previous paper,* was 

 adopted. Two fine glass jets were arranged close to the vibrator and in such positions 

 that when a stream of air was blown through them it would produce a couple on the 

 vibrator and so rotate it These jets (B, fig. 2) are joined at the bottom of the jacket 

 to a glass tube, which passes out through a cork and is connected to 3 or 4 yards of 

 fine " composition " tubing passing along the wall of the laboratory and ending on a 

 small table beside the observing telescope. Here it is joined to a piece of indiarubber 



* F. HORTON, 'Phil. Trans.,' A, rol. 204, p. 1, 1904. 

 VOL. OCIV. A. 3 O 



B 



Fig. 2. 



