556-611, published 1 45 1889 Proc. R. Soc. Lond. Email alerting service here the article or click the top right-hand corner of article -sign up in the box at when new articles cite this Receive free email alerts http://rspl.royalsocietypublishing.org/subscriptions , go to: Proc. R. Soc. Lond. To subscribe to on July 20, 2018 http://rspl.royalsocietypublishing.org/ Downloaded from on July 20, 2018 http://rspl.royalsocietypublishing.org/ Downloaded from Mr. J.N. Shoolbred, the Secretary of both

more »

... mmittees, was instructed to intrust the curves to me, in order that they might be submitted to harmonic analysis. He afterwards was so good as to obtain from Mr. Druce the continuation of the Dover curves. As the reduction of the whole series of curves would have been very expensive, it was determined that only the curves for 1883-4-5 should be treated ; these years were selected because there was reason to suppose that the curves were more accurate than the earlier ones. To meet the expense of the reduction, Sir William Thomson obtained £50 from the Royal Society Grant, and this sum was afterwards handed to me. The amount would, however, have been altogether insufficient if Major Baird had not interested himself in the matter, and introduced me to Mr. E. Connor, of the Tidal Department of the Survey of India. Mr. Connor then generously offered to devote his spare time to the work, and undertook the superintendence of the native computers at Poona. The reductions of three years of Dover curves, and of the same three of Ostend curves, have been made with all the thoroughness and care of the Indian work. The computations themselves are now in my hands, and the curves have been returned to Mr. Shoolbred. The tidal record was frequently interrupted at Dover, for there are 34 days wanting in 1883, 57 days in 1884, and 72 days in 1885. The gaps are onlv of a few days at a time, except from September 24 to October 26, 1885. The zero of the Dover gauge is said to be 8'67 feet below the Ordnance datum, and therefore 1133 feet above the " international datum," which is stated in the British Association Report (1879) on Levels to be 20*00 feet below English Ordnance datum. The reduction of the tide curves shows that the mean sea level at Dover was, in 1883, 0'52 foot; in 1884, 0*46 foot; and in 1885, 0*21 foot above Ordnance datum. The French Nivellement General is 2*625 feet below Atlantic M.S.L., and 1*992 foot below Ordnance datum. Hence Atlantic M.S.L. is 0*633 foot above Ordnance datum. Thus Dover M.S.L. was, in 1883,0*11 foot; in 1884, 0*17 foot; and in 1885, 0*42 foot below Atlantic M.S.L. It appears from the Ostend curves that Ostend M.S.L. was, in 1883, 0*25 foot; in 1884, 0*37 foot; and in 1885, 0*21 foot above Ordnance datum, and therefore in 1883, 0*38 foot; in 1884, 0*26 foot; and in 1885, 0*42 foot below Atlantic M.S.L.' Thus Ostend M.S.L. was below Dover M.S.L. by 0*27 foot in 1883 ; by 0*09 foot in 1884; and they were the same in 1885. By reference to the Atlantic M.S.L. we see that by far the larger part of these remarkable oscillations depends on Dover. But it is nearly incredible that the sea at Dover should have been Harmonic Analysis o f Tidal Observations. 557 on July 20, 2018 http://rspl.royalsocietypublishing.org/ Downloaded from * Captain Wharton, R.N., is of opinion that the situation of Dover is such that the tides are likely to be irregular there. I cannot, however, believe that this affords a sufficient explanation of the irregularity of the results.