PENDULUM METHODS 159 



the thickness of the lenses themselves (which is at least 

 60 mm. inside width for high water pressures and 130 mm. 

 outside). Therefore a new form of fixing device for 

 simultaneous centering was adopted by Erlinghagen in 1906 

 in cooperation with Professor Klingenberg of the General 

 Electric Company in Berlin, as shown in Figs. 98 and 100. 



The borehole magnet was made by having an I-shaped 

 bronze frame, between the webs of which on each end a n- 

 shaped iron was placed enveloped by a magnetic coil. 

 The legs of the iron were beveled (Fig. 99) corresponding to 

 the internal diameter of the borehole. The coils have to be 

 absolutely watertight. The coil was wound with enameled 

 wire and the bearing spots repeatedly insulated from one 

 another and the whole placed in a zinc case and waxed up. 

 The neck has a soldered bridge through which the winding 

 wire is carried well insulated. The construction has been 

 tested for hours under a pressure of 9 atm. 



Figure 98 shows the centering device where we have three 

 link-arm borne steel rolls pressed outward together by a 

 strong central steel spring, from which it swings down to 

 the bottom of the apparatus in fixed links. Above, it is 

 movable up and down by a hnked ring and a movable 

 center bolt. There are three of these centering devices, 

 one to center the upper magnet, another the lower magnet 

 and the other to hold the measuring apparatus properly in 

 the middle. The measuring apparatus (Fig. 100) has a 

 powerful frame of three steel rings connected by two longi- 

 tudinal ribs having, in the upper part, a glass encased clock- 

 work. Under this a roll paper 50 mm. wide winds from roll 

 Ti over the cork-lined plate j) on to roll r2 with uniform veloc- 

 ity, only roll r2 being clockwork driven. As the angular 

 velocity of the clock is always the same, that of the paper 

 increases the more paper is wound on to r-i giving a uni- 

 formly accelerating motion. To control the time points 

 of the measurements the paper must move uniformly and 

 this is done by means of the string drive s on roll r^ which 

 has a slipping arrangement. Under the paper strip moves 

 the point of the universally suspended spring pendulum. 



