6 EXPERIMENTS WITH THE DISPLACEMENT INTERFEROMETER. 



For convenience in observation, it is necessary that the horizontal pendu- 

 lum be damped. A water damper, as well as an oil damper, was used without 

 apparent disadvantage, care being taken to keep the long copper trough in 

 which the vane dips full of water. Soldered parts must be covered with a 

 varnish, for instance of wax and resin, as there will otherwise be slow galvanic 

 corrosion and a precipitate. 



3. Observations. The observations were carried out for experimental pur- 

 poses only, because an adequate lens and plate-glass mirrors were not at hand, 

 and because the hill on which the laboratory is built is in a continual state 

 of tremor, due to the heavy car freightage, both on the surface of the hill 

 and through it. The pier, moreover, was not protected and insulated to an 

 extent needed in refined seismological work. Hence the chief purpose of 

 these experiments is to indicate the deviations to be expected prior to the 

 interferometer work of the next section. In the preliminary experiments, 

 the pendulum was used without a damper and changes of inclination of the 

 horizontal pendulum of a = 0.4 second per day, or even i second in several 

 successive days, were not infrequent. On other days the pendulum was rela- 

 tively fixed. The succession of points was quite regular, and maxima and 

 minima frequent. 



Observations of a more definite character were taken between September 

 26 and December 3, 1913. They were computed throughout and charted. 

 They, however, have little more than the local interest specified, and I will 

 therefore merely give an example of part of the results in the graph, fig. 5. 

 By equation (6) above, where x is the distance apart of the two images of 

 the slit and D the intervening space between the plane of these images and 

 the fixed mirror N nearest the lamp, 



6 being the angular deflection of the pendulum. Since I? = 2,000 cm., 



0=125X1 o- 6 x radians. 



Furthermore, a = <p6, where <p is the inclination of the line drawn through the 

 points of the pivots to the vertical, and a the change of inclination of the 

 pier to the vertical corresponding to 0. Hence, 



a = 



The value of <p found below in the work with the interferometer is ^ = 0.0108 

 radian, so that a is a little larger than i per cent of 6. Thus 



a * ' 3 5 X io~ 6 x radians = 2&x second, nearly. 



In fig. 5, the values of a in seconds are given from November 4 to November 

 26. The apparatus was interfered with from time to time, as shown at a, 

 and new modifications were introduced. The water damper was used through- 

 out, so that the pendulum did not vibrate. 



