252 ANNUAL EEPORT SMITHSONIAN INSTITUTION, 1916. 



and this feeling did not vanish until Newcomb pointed out that 

 Euler had made a certain assumption regarding the interior of the 

 earth that had in the meantime been universally discarded. His 

 period of 10 months applies in fact only to a perfectly rigid and 

 unyielding earth. Newcomb showed that if the earth yields to defor- 

 mation to the same extent as though it were composed throughout of 

 steel, then Euler's period would be lengthened to about 14 months. 

 Here we have the first dependable determination of the rigidity of 

 the earth, a result that has since been confirmed in several ways, par- 

 ticularly by a measurement of " bodily tides " in the earth. 



The 14-month term (or the modified Eulerian term as it is now 

 called) has been under accurate observation for a quarter of a cen- 

 tury. The period can probably (though not certainly) be regarded 

 as constant. This is what we should expect, for a change in this 

 period would call for a sensible alteration in the distribution of the 

 material within the earth, or a change in the rigidity of the earth. 

 The amplitude of this term presents a very puzzling problem. Its 

 usual value is about 0".27, but twice in recent years it has jumped 

 to about 0".40. Such a change could be accounted for by supposing 

 that the earth had received a severe blow or a succession of milder 

 blows tending in the same direction. We are. reminded that both 

 Milne and Helmert have suggested that there might be a direct con- 

 nection between latitude variations and earthquakes. This sugges- 

 tion was originally made by Milne very early in this century when 

 the astronomical data necessary to test it were still very meager. It 

 is to be hoped that the question will be taken up again in the light 

 of the information that has been added during the past 10 or 12 

 years. 



Though the Eulerian term is the largest part of the latitude vari- 

 ation, it is by no means the only important one. We have next an 

 annual term with a maximum amplitude of about 0".20. We may 

 say with some confidence that this term is seasonal and meteorologi- 

 cal in its origin, but at present no more definite statement would be 

 warranted. It was early suggested that ocean currents might cause 

 this variation. These currents would have to vary greatly with the 

 season, either in the volume or the speed of the flow, or in its direc- 

 tion; for an unvarying current would merely modify the Eulerian 

 term once for all .and would leave the latitude variations otherwise 

 unchanged. A similar suggestion has been made with regard to air 

 currents, and appeal has also been made to unequal deposits of snow 

 and ice on two opposite hemispheres of the earth to account for the 

 annual term. It seems to me that these explanations have not been 

 subjected to the critical numerical tests that are possible and desir- 

 able. The meteorological data are doubtless competent to enable us 



