10 Rev. 0. Ft-sher — Convection Currents in Earth's Interior. 



rotation around its mean position, it would not explain the remarkable 

 irregularity both in amplitude aud period with which its wanderings 

 are affected.' Professor Milne is of opinion that some connection 

 exists between these irregularities and tlie frequency of large earth- 

 quakes.- Might not convection currents, bj' slightly altering the 

 distribution of mass within the earth in an irregular manner, account 

 for these irregular wanderings of the pole ? 



(5) Observations upon the force of gravity in India have revealed 

 the remarkable result that changes appear to have taken place within 

 the last thirty years. At Dehra Dun, at the foot of the Himalayas, 

 there has been found a notable increase of gravity corresponding to 

 four swings of the second's pendulum in twenty-four hours. ^' If the 

 two sets of observations that show this are to be depended on, we 

 have here an absolute proof that there has been a change of density 

 and consequently movement in the substratum of that neighbourhood ; 

 and it is noticeable that, in the interval between the two dates, 

 a severe earthquake was experienced at that place. 



(6) Making use of the best data newly available, I have calculated 

 by a mathematical method that, on the hypothesis that the crust of 

 the earth is being formed by congelation out of a liquid subjacent 

 magma, the age of the world would not be greater than about eight 

 million years unless there are convection currents in the interior.'* 

 Dr. See, Professor of Mathematics to the U.S.A. Navy, has accepted 

 my calculation, aud also the conclusion at which I have arrived 

 regarding the corresponding age of the world, and, not believing in 

 the existence of convection currents, he has been led to argue that 

 geological phenomena do not require a longer period for their 

 accomplishment than about eight million years.* Few geologists, 

 I think, will be willing to accept so short a time. It follows that 

 we have the opinion of an accomplished mathematician to endorse 

 the result of my calculations, which, on any reasonable estimate of 

 the world's age, go far to prove the existence of convection currents. 



The two principal objections to the hypothesis of a liquid interior 

 appear to be (1) the existence of a fortnightly tide, and (2) the two 

 .phases of the preliminary tremors of a world-shaking earthquake. 

 The argument for rigidity from the fortnightly tide is so extremely 

 technical and refined that it is not for the ordinary geologist to 

 follow it.'' It is based on the average of observations taken in India, 

 Britain, and France;^ but, confining our attention to India, it seems 

 doubtful whether that tide can be detected.* 



The argument for rigidity from the different velocities of the two 

 phases of the preliminary tremors of an earthquake assumes that 



1 Nature, February 27, 1S96, also May 12, 1898. 



- British Association Report, 1903 and 1906. 



3 Colonel Burrard, Phil. Trans., vol. civ, A 394, p. 294. 



^ Physics of the Earth's Crust, 2nd ed.. Appendix, ch. xxvii. 



" Vrot. American Phil. Sec, 1907, vol. xlvi, p. 221. 



^ Thomson & Tait's Natural Philosophy, 1883, pt. ii, p. 442. 



' Knott's Physics of Earthquakes, p. 104. 



^ Physics of ihc Enrth's Crust, Appendix to 2nd ed., p. 34. 



