562 Prof. G. H. Darwin. 



Singapore and Hongkong. 



I have no information about these observations. The results were, 

 however, kindly placed at my disposal for this collection by Mr. 

 Roberts. They were given me in the form which was used before the 

 publication of the E/eport of 1883 to the British Association, and I 

 am responsible for the redaction to the standard form. 



Mr. Roberts performed the reductions of the observations himself, 

 and has published the tide tables for the two ports on behalf of the 

 Governments of the two colonies. He proposes to write a paper on 

 these tides, which will doubtless give the information which is here 

 wanting. 



Indian Stations. 



Major Baird and Mr. Connor have sent me for publication the 

 values of the constants at a large number of stations in India. 



I have divided them into two groups. The first of these comprises 

 stations for which results were published in the paper by Major Baird 

 and myself in the ' Proceedings of the Royal Society.' Many years of 

 observation are thus added to the previous ones, and the mean values 

 of the constants given below include the values given in our paper of 

 ] 885. The station at Karachi is especially valuable for tidal theory, 

 since we now have results for nearly a whole lunar cycle of nineteen 

 years. The second group comprises a number of ports, for which 

 the constants have been only hitherto published in the prefaces to 

 the Indian Tide Tables.* 



The constants for certain tides initialled 2N, MN, MK, 2MK are 

 now given for the first time.t The first of these, 2N, is the elliptic 

 semidiurnal tide of the second order. It appeared from the develop- 

 ment of the equilibrium theory that it might be easily sensible, and 

 the values now given prove that this is the case. The other three, 

 MN", MK, 2MK, are shallow water tides arising from the interference 

 of the principal lunar tide M 2 , 1st, with the larger elliptic tide N, 

 2ndly, with the luni-solar diurnal tide K l5 and 3rdly, with the lunar 

 diurnal tide 0. The two latter of these, viz., MK and 2MK, also arise 

 from the interference of M 4 with 0, and from M 4 with K x . The 

 values appear to be all fairly consistent from year to year at the 

 riverain stations, but at other places they are obviously quite without 

 significance. 



Mean Sea Levels. 



In our previous paper we did not give the mean sea levels, as deter- 

 mined from each year of observation. 



* Published by authority of the Government of India. 



f See introduction to our previous paper on the " Kesults of Harmonic 

 Analysis." 



