92 



DISPLACEMENT INTERFEROMETRY BY 



to a ring rr, capable of revolving with friction around the tube of the objective 

 and of being fixed for any angle in altitude of the rod. The horizontal clamp 

 g, adjustably attached to the rod/, supports the spring s. This passes through 

 a vertical crevice in g and is fixed by the vertical set-screw p and the two 

 oblique adjusting strips k on either side of g. These pieces, k, are clamped by 

 the screws at h, and serve as holders of the two coaxial set-screws at i on either 

 side of sk, so that the objective A may 

 be centered relative to the tube Ee for 

 any oblique position of /, which must be 

 normal to the fringes. The vibration 

 may frequently be considerably changed 

 by sliding / and 5 in g and reclamping the 

 system. If the objective is to be fixed, 

 a screw at n may be depressed for the 

 purpose. The vibration may be started 

 and stimulated from time to time manu- 

 ally. It has not, thus far, been necessary 

 to add an electromagnetic vibrator. 



To find the fringes a spectro-telescope 

 will usually first have to be used. This 

 is then replaced by the apparatus in figure 

 67, for observation. If not too far dis- 

 placed, the fringes of a vibrating system may thereafter be found by the 

 vibration telescope even when they can not be seen with the fixed telescope, 

 as they overlap during vibration. A fine slit, so long as it supplies sufficient 

 light, gives the sharpest wave-curves. The angle of altitude of the fringes 

 is of no consequence, since / is set in altitude normal to them ; but it is of 

 advantage to rotate the slit also, until its image in the telescope is nearly 

 normal to the fringes. 



Under all circumstances the two spots of light representing the slit, if 

 caught objectively on a screen at T, must be nearly coincident, horizontally 

 and vertically, when the rays at T are parallel. If these spots are too far 

 apart the fringes will be very small or even absent. A search for them by 

 any method is then useless. To meet this preliminary but essential condi- 

 tion, the spots are first made coincident by rotating N (horizontal and ver- 

 tical axes) and thereafter rotating N' until the images coincide in the tele- 

 scope at T. One or two adjustments of this kind usually suffice, since per- 

 fect coincidence is non-essential and in fact undesirable, because the fringes 

 are then too large for convenience. 



54. Observations. The use of two telephones soon showed itself to be un- 

 suitable for the present purposes ; for the diaphragms oscillate not merely fore 

 and aft, as is here desirable, but locally around horizontal and vertical axes 

 as well, particularly when the vibration is relatively intense. Hence the 

 coincident slit-images of the silent telephone periodically separate or pass 



