ON EAETH TREMORS. 317 



meteorological phenomena as in the two former cases. In consequence 

 the effect of variability of the daily period is much better eliminated, and 

 it makes little difference if, in computing the hmar terms, one takea 

 account of it or not. 



The following expression was deduced by a simple transformation 

 from the original valaes of a, b, after expressing each of the eight series 

 as an harmonic function of C. The form in which it is given is, of course, 

 entirely empirical. Thus the changes in the coefficients, which here 

 depend on C, in reality probably depend on the declination of the moon. 

 All uncertain terms I have suppressed, and only retain those the exist- 

 ence of which is demanded by the observations. 



f 9-5 (± 1-12) +41-9 (± 2-43) cos (C- 268°-7) "] 



<^ +20-8 (± 2-25) cos (2C-164°-1) } cos (t-C-UrS) » 



L + 24-5 ( ± 2-43) cos (30 - 260°-4) J 



+ {45-0 (± 0-95) + 10'5 (±1-97) cos (C-194°-8)} cos (2i-2C-341°-3) 

 + { 9-1 (± 0-50) + 12-1 l± 071) cos (20- 50°-9)} cos (4i-4O-332°-0) 



In comparing the above expression with the lunar term computed by the 

 fii'st simpler method, which is 



45-4 cos (2<-20-345°7), 



it is seen that it contains a nearly identical term. Considering this alone, 

 we find that it represents a semi-diurnal deflection of the plumb-line, the 

 eastern excursion of which takes place on an average a little more than 

 half an hour before the meridian passage of the moon, or about 1^ hour 

 before high water. The observatory lies on the west side of the Fahde- 

 busen, at a distance of about 200 m. from high- water mark, whilst the 

 station for tidal observations is a good deal more to the south, at the 

 entrance of the harbour. Accordingly the semi-diurnal terna might be ex- 

 plained in a pretty satisfactory way by the pressure of the tides, which on 

 an average rise to a height of 3^ m. at Wilhelmshaven. This would certainly 

 be a very strong effect of tidal depression compared with the theoretical 

 results obtained by Professor Darwin in his researches ; bat as the soil is 

 of a marshy character, and apparently very elastic, such an effect would 

 not appear impossible. 



The present results, however, seem to tell against such an explana- 

 tion ; for, if it were the tides that produce by their pressure the lunar 

 terms, a close connection between these and the form of the tides ought 

 to be established, which is not the case. In fact, the very large term of 

 the first order, which is perhaps the best determined of all, and which is 

 all the more remarkable, because, according to our computation, it changes 

 the sign, reminds one of the corresponding lunar tide, which, though 

 playing an important part in many places in the world, is comparatively 

 small on the coasts of Germany. 



It seems advisable to postpone a discussion of the above facts until 

 farther observations undertaken at some suitable place shall have placed 

 them beyond doubt. Perhaps few places in the world would offer a better 



' Expressed in units of 0"-0028. The numbers in brackets are approximate values 

 of the probable errors of the coefficients. 



