Figure by means of the Pendulum. 455 



The outcome of all this will be that when observations from 

 opposite sides of the earth are compared we shall be able to 

 decide whether the axis of the earth actually shifts its position, 

 or whether changes of latitudes are due to transfers of molten 

 matter below the crust. 



If the results at Berlin and Honolulu show opposite phases 

 at the same time, we should expect the latitude to be stationary 

 at Washington, because this point is one-quarter way around 

 the globe, or midway between the other two stations. The 

 conclusion from this would be that there is a real motion of 

 the pole and not a transfer of material inside the earth. 



There is a decided maximum and minimum within twelve 

 months with a larger maximum and minimum in a "five-year 

 period. In addition to this the Greenwich observations show 

 a long period of inequality extending over sixty years. The 

 cause of the short period movements has been ascribed to the 

 interference of the motion of the axis of inertia with that of 

 the ten-monthly period of the axis of rotation. It can certainly 

 be assumed that the sun and moon produce atmospheric tidal 

 effects changing with the seasons, and it is also known that the 

 shifting of a mass of water covering yL- the earth's surface 

 and being 0*10 meter thick would cause the axis to move //# 16 

 — a quantity which is quite measurable in all latitude work of 

 precision. As this depth of water corresponds to an atmos- 

 pheric pressure of about - 007 meters it is evident that exten- 

 sive changes in the density of the air may produce a slight 

 change in the position of the earth's axis; so that it would 

 seem well worth while to measure the force of gravity from 

 time to time at the same place in order to detect changes that 

 would most probably be produced by changes of latitude. 



Practical Methods. 



When we come to the actual field work, again several 

 methods present themselves for our consideration. The ulti- 

 mate end of the observations is to find how long it takes the 

 pendulum to make an oscillation at a given temperature and 

 atmospheric pressure. Most of the slight corrections necessary 

 to reduce the different experiments to the same conditions, and 

 thus make them comparable, may for our present purpose be 

 passed over in silence. The reduction to an infinitely small 

 arc involving only simple geometrical considerations is easily 

 disposed of. The influence of the temperature and density of 

 the air requires special treatment, and has been made the sub- 

 ject of careful study by all pendulum observers. The air not 

 only has a buoyant effect on the oscillating body, but by reason 

 of its viscosity adheres to the pendulum and is drawn along 



