( 1284 ) 
10% 0.348 cos (2t—170°.4) 
11 0.334 cos (2t—175°.1) 
12 0.400 cos (2t—177°.3) 
1 0.346 cos (2t—176°.5) 
2 0.414 cos (2t—178°.1) 
3 0.392 cos (2i—184°.2) 
4 0.420 cos (2t—183°.0) 
5 0.367 cos (2t—192°.4) 
For the N—S component of the O-tide, we found, taking together 
two consecutive hours : 
10 and 11 hour 0.123 cos (t—199°.2) 
Rien Sars 0409 eos (£—2295.D) 
Der Fo 5, 09 reds: (E20 
4, 5 0.375 cos ¢—205°4) 
The differences between these values are very small especially for 
the M,-tide; for the O-tide the amplitudes show considerable fluc- 
tuations; but the agreement of the arguments clearly indicates that 
the differences between the experimental and theoretical values are 
due to an external periodical disturbance. 
We shall presently see that they are caused by the watertides in 
the Indian Ocean and in the Java sea. | 
If, namely, we assume the amplitude of the undisturbed gravita- 
tion-tide M, to be equal to °/, of its theoretical value — an assumption 
ewhich cannot lead to an appreciable error considering the small 
value of this tide in proportion to the disturbing force — if, further, 
its argument is assumed to be equal to the theoretical argument, we 
find for the disturbing force : 
0"-00772 cos (2t—2°.3) 
If there were no retardation and, therefore, high water coincided 
with the moment of culmination of the fictive star, the tides would 
be represented for the longitude of Batavia by the expression : 
R cos Qt=315-5) 
and the disturbance could be explained if the kappanumber of the 
ocean-tide is assumed to be: 
| North of Batavia 2°.3—315°.5 = 46°.8 
South „ »  2°.38—815°.5 + 180° = 226°. 
Applying the same reasoning to the O-tide (where, of course, the 
accuracy is less owing to the greater value of the amplitude if */, 
of its theoretical value is assumed) we find for the disturbing force : 
000765 cos (t — 75.°7). 
As, for kappa = 0, the watertide is: 
