142 



SCIENCE 



[N. S. Vol. L. No. 1284 



ing whicli ignores the objectionable slow 

 wanderings of the heavy mass, and yet records 

 displacements directly, retaining the flat mag- 

 nification curve of the best seismographs, and 

 permitting the use of very long-period instru- 

 ments. The essential feature of this record- 

 ing system is that the (apparent) motion of 

 the heavy mass is transmitted through a vis- 

 cous medium to a light system having a 

 shorter period of its own : the viscous coupling 

 amounting to direct connection for the seismic 

 motion of a period up to that of the pendulum, 

 but permitting the light system to avoid the 

 very slow changes. 



This method of recording may be explained 

 by describing the system as applied to a cer- 

 tain horizontal pendulimi at the College of 

 Hawaii. The adaptation to vertical-motion 

 instrmnents will suggest itself. 



Attached to the seventy-pound mass of a 

 suitably damped horizontal pendulum is a 

 horizontal cylindrical vessel of heavy oil, the 

 axis of the cylinder being in the direction of 

 motion of the pendulum. The cylinder is two 

 inches in diameter, and has an opening in the 

 form of a longitudinal slit one half inch 

 wide at the top, around which there is a rim, 

 to allow the surface of the oil to be somewhat 

 higher than the top of the cylinder. In the 

 cylinder, immersed in the oil, is the mass 

 (about one pound) of a second horizontal 

 pendulum. This mass is itself cylindrical and 

 forms a piston within the larger vessel, though 

 not touching it. The axis of rotation of this 

 light pendulum coincides approximately with 

 that of the heavy pendulum, but is sufficiently 

 inclined to give a free period short enough 

 to allow of registration, and it is the motion 

 of this light pendulum which is recorded. 

 It will be seen that for ordinary seismic 

 motion the two pendulums formi a single mass, 

 but that the oil can flow so as to allow the 

 light i)endulum to retain approximately its 

 ovm equilibrium position. 



Registration is accomplished photographic- 

 ally as follows: A piece of no. 36 nickel wire 

 two inches long is soldered at its ends to 

 pieces of galvanometer suspension ribbon A A 

 (see figure) each three inches long, and 



stretched betv/een spring supports in a vertical 

 position. At its middle this nickel wire passes 

 through the hole of a watch jewel J of suit- 

 able size held by an arm fastened to the sup- 

 port, preventing transverse vibration. Fixed 

 to the nickel wire just below the jewel is an 

 arm of wire, one fourth inch long, holding at 

 its other end a similar jewel. Through the 

 hole of this jewel passes a short piece of no. 36 



a 



r '■!3>'t 1 



X 



(iM 



"0(7 



wire K attached to the end of an almninum 

 wire which is itself attached to the light pen- 

 dulimi by a flexible connection. The motion 

 of the x)endulum is thus transmitted to the 

 mirror M which is cemented to the nickel 

 wire at another point. The rocking of the 

 mirror is recorded in the usual way on 

 bromide paper. 



This system succeeds with a magnification 

 of 75 on a pendulum moving sometimes half 

 an inch in the course of the twenty-four hours, 

 with a lateral drum-motion of an eighth inch 

 per (hourly) revolution. 



Ahnold Eomberg 

 ' College of Hawaii, 

 Honolulu, Hawaii 



SCIENCE 



A Weekly Journal devoted to the Advancement of 

 Science, publishing the official notices and pro- 

 ceedings of the Americem Association for 

 the Advancement of Science 



Published every Friday by 



THE SaENCE PRESS 



LANCASTER, PA. GARRISON, N. Y. 



NEW YORK, N. Y. 



Entered in tbe poit-afficc at Lancutet. Pa., as lecond cUm natts 



