ON SElSMOLOGlCAt INVESTIGATION^ 53 



ttle case of the pendulum there are really only four independent evalu- 

 ations, and the values of K 2 and of P might have been omitted as far as 

 concerns the provision of a means of comparison between the pendulum 

 and the tide. 



' A fortnight is much too short a period of observation to afford 

 trustworthy values for the deflections of the pendulum, and therefore we 

 should not place implicit reliance on the exact numerical values obtained. 



' The phase of M 2 for the pendulum is virtually identical with that of 

 the tide, but this exactness of coincidenie is probably to some extent 

 accidental. The high tide, so to say, for the solar tide S 2 , differs in phase 

 from that of the water by 36° or Ih. 12m., and the amplitude is 

 considerably greater relatively to M 2 than is the corresponding ratio for 

 the sea. 



' The phases of the diurnal sea-tides at Hilbre Island are very 

 abnormal, for whereas it might have been expected that they should all 

 come out nearly the same, the phases of K x and O differ by 147°. The 

 result is, however, derived from so many years of observation that it is 

 certainly correct and is, moreover, confirmed by the tidal constants for 

 Liverpool. In the case of the pendulum we observe a similar abnor- 

 mality, for the phases of K x and O differ by 109°. It is, however, 

 remarkable that these tides are almost inverted with reference to the 

 sea-tides. One may conjecture that there are perhaps nodal lines for 

 these tides at some short distance out to sea, and that the bulk of the 

 sea which produces the flexure is in the opposite phase from that which 

 gives the visible tide at Hilbre Island and Liverpool. The amplitudes 

 of K x and O are also very discordant, both in absolute amount and 

 between themselves. In the sea Kj and have nearly the same ampli- 

 tude, but with the pendulum that of K x is three times as great as that 

 of O. This would result if the supposed node of K\ were nearer the 

 shore than that for O, because if this were so there would be a larger 

 weight of water, oscillating in a phase opposite to that of the sea in 

 shore, to produce flexure in the case of K x than in that of 0. How- 

 ever, the series is much too short to justify any confidence in such 

 conjectures. 



' The last column gives the relative importance of the tide-generating 

 forces for the several tides, and it will be seen that the force for K x is 

 much larger and that for somewhat larger than that for M 2 . We 

 see that both in the sea and in the case of the pendulum there is an 

 enormous reduction of amplitude for diurnal tides as compared with the 

 semidiurnal ones, but the reduction is markedly less for the pendulum. 

 If these values should be confirmed, we may perhaps suspect that the 

 direct lunisolar tide-generating force is rendering itself evident in the 

 K x tide, and such a conjecture would accord with the phase of K x 

 approaching 360° without the intervention of the nodal line at sea 

 suggested above. However, as already pointed out, it is too soon to 

 draw any conclusions with confidence. ' 



"Whatever may have yet to come from this new departure in obaerva» 

 tions bearing upon Earth Physics, the work already accomplished is 

 suggestive of certain conclusions. 



We see that an observatory near to a shore line, in consequence of 



