932 



SCIENCE 



[N. S. Vol. XXXIX. No. 1017 



perfectly with, the observed ones. But the 

 computed tides for the N.-S. pipe had to be 

 reduced to about one half of the computed 

 values to give the best agreement with the 

 observations. Sample curves for the E.-W. 

 line are given in Fig. 1, and the final mean 

 values of the semi-daily lunar tide for both 

 lines are given in Figs. 2 and 3. The values 

 of the ratios finally adopted were 



E.-W 0.710 



N.-S 0.523 



These figures mean that the water tide in 

 the E.-W. pipe was only 71 per cent, of what 

 it would have been if the earth were perfectly 

 rigid, i. e., there is an E.-W. tide in the solid 

 earth which is 290/710 as large as the water 

 tide. In a N.-S. direction there is an earth 

 tide 477/523 as great as the water tide. That 

 is, the earth yields more readily to distorting 

 forces in the N.-S. direction than in the 

 E.-W. direction. Twice each day the surface 

 of the solid earth rises and falls about a foot. 



This result is rather surprising, as there ap- 

 pears to be no simple reason why the earth 

 should be more rigid in an E.-W. than in an 

 N.-S. direction. The result was, however, 

 foreshadowed by Sehweydar, in discussing 

 the observations of Hecker at Potsdam. A. E. 

 H. Love has suggested that the efiiect may be 

 due to the distortion caused by the weight of 

 the ocean tides against the edges of the conti- 

 nents. 



The times of high tides along the seacoasts 

 with reference to the moon's position are of 

 course very irregular, but if these enormous 

 masses of water were flung against the coasts 

 at the time they were due to rise in obedience 

 to the moon's attraction, there can be little 

 doubt that the efliect would be to reduce ap- 

 preciably the east and west earth tides, and 

 thus to increase the apparent rigidity of the 

 earth in this direction. But the ocean tides 

 are so very complex that there appears to be 

 little prospect of making accurate calculations 

 on this point at present. 



Added significance is given to the possibil- 

 ity that the ocean tides are responsible for the 

 difference between the E.-W. and N.-S. rigid- 

 ity by a consideration of the daily, instead of 



the semi-daily, tide. In reducing the observa- 

 tions Professor Michelson found that if he 

 considered the tide of period 25.812 hours, in- 

 stead of the semi-daily tide, the ratios of ob- 

 served to computed amplitudes were 



E.-W 0.72 



N.-S 0.66 



The agreement here is much better than for 

 the semi-daily tides, but as Professor Michel- 

 son has said, if the Love-Schweydar theory is 

 correct, we would have expected a consider- 

 able increase in the E.-W. ratio, when the dis- 

 turbing effect of the ocean tides was elimi- 

 nated. 



Of no less importance than the ratio of the 

 observed to the computed amplitudes is the 

 question of their agreement in phase. It was 

 found that the maxima and minima of the ob- 

 served and computed tides occurred almost 

 exactly simultaneously. The difference in 

 phase furnishes a measure of the plasticity 

 or viscosity of the earth. If there is no diiler- 

 ence in phase between the observed and com- 

 puted tides, the distortion of the earth does 

 not lag behind the tidal forces; i. e., for any 

 given value of the force the distortion does 

 not tend to increase with the time. If there 

 were a lag in the earth tide, it would throw 

 the observed water tides ahead of the com- 

 puted ones. This is rendered obvious at once 

 by an appropriate graph. In the reduction of 

 the observations. Professor Michelson found 

 that the N.-S. tides were behind the computed 

 ones, in the mean, by about 24 seconds. This 

 lag could have no physical meaning, and the 

 amount is well within the limits of observa- 

 tional error. In the E.-W. pipe the reduction 

 shows an acceleration of 3.6 minutes. This 

 may be a real acceleration, but the quantity is 

 not far from the limit of accuracy of the ob- 

 servations, and it is too small to be regarded 

 as highly accurate. 



The agreement of phase between the ob- 

 served and computed waves is of great im- 

 portance for the theories of planetary evolu- 

 tion. Darwin's theory of planetary fission, 

 i. e., the breaking off of the moon from the 

 earth and its subsequent withdrawal to its 

 present distance, is based on the assumption 



