Figure by means of the Pendulum. 459 



ience on the Canada frontier and too short for accuracy on the 

 Gulf of Mexico. A compromise was made between the several 

 conditions and the above mentioned interval of 6 m 15 s chosen. 

 It so happens that the effect on the coincidence period is about 

 the same, whether we pass from the latitude of Washington to 

 the Gulf of Mexico, or carry the pendulum from the sea level 

 to the top of Pike's Peak ; the height in the latter case having 

 the same effect as the change in latitude in the former. 



In the new work proposed by the Coast and Geodetic Sur- 

 vey all experiments are to be made at a given atmospheric 

 pressure. This will be about a mean value of those actually 

 found in practice. Air will therefore be forced into the cham- 

 ber at mountain stations, and pumped out at the lower ones. 

 It being decided to swing the pendulums in an enclosed space, 

 the interesting question came up, how close can the pendulum 

 be placed to the walls of the chamber, without influencing the 

 time of oscillation ; or in other words, how small can we make 

 the box and still have the pendulum swing just as it would in 

 the open air. Experiments were made with boxes of different 

 sizes and shapes. The result generally stated was that the 

 effect of the sides of the chamber only began to be felt when 

 they were within about one inch of the moving body, and that 

 what is known as " skin friction " is more effective than im- 

 pact friction. That is to say that proximity of the wall to the 

 side of the pendulum has more influence than nearness in front 

 or back of it. The viscosity of the air is indeed an important 

 factor in the investigation, as it is well known that besides the 

 buoyant effect of the air it adheres to the pendulum and is 

 drawn along with it. This influence has been studied both 

 theoretically and practically by Stokes, Green, Peirce and 

 others, and has furnished some fine examples of mathematical 

 analysis. 



The temperature of the pendulum is found by means of 

 another pendulum of exactly the same shape, size and material, 

 except that the knife edge is of hard rubber instead of agate. 

 This auxiliary pendulum is suspended inside the receiver. To 

 it is attached a thermometer, whose bulb is encased in filings 

 of the metal imbedded in the stem of the temperature pendu- 

 lum. The presence of this auxiliary instrument has no appre- 

 ciable effect on the period of the swinging pendulum, either 

 from air disturbance or from the vibratory movement of the 

 support. This fact was carefully determined by experiment. 



In the last cruise of the IT. g. Man-of- War Pensacola, an 

 officer of the Coast and Geodetic Survey was sent to determine 

 the force of gravity at some stations in Africa, and on some 

 islands of the North and South Atlantic. The computations 

 have just been completed, and the results are in conformity 



