ON THE DETERMINATION OF GRAVITY AT SEA. 83 



Determination of Gravity at Sea. — Second Report of the Com- 

 mittee, consisting of Professor A. E. H. Love (Chairman)_ 

 Professor W. G. Duffield (Secretary), Mr. T. W. Chaundy, 

 Sir H. Darwin, Professor A. S. Eddington, Major E. 0. 

 Henrk'I, Professor A. Schuster, and Professor H. H. 

 Turner. 



[Plates I. and II.] 



I. Report upon Hecker's New Method of Measuring Gravity at 

 Sea with Apparatus of the Enclosed Mercury Barometer 

 Type. By Professor W. G. Duffield. 



Two methods used by the writer in 1914, during the voyage of the 

 British Association to Australia, have been described in i2o?/. (Soc. Proc. A., 

 vol. xcii., p. 505, and in the British Association Report, Newcastle, 1916. 

 It remains to discuss the third method of attack. The apparatus employed 

 was lent to the writer in June 1914 by Professor Hecker, of Strassburg, 

 who was anxious to have it tested at sea, and welcomed the opportunity 

 which the voyage of the British Association aflforded- Although less 

 successful than either of the other methods emjjloyed, it will be useful to 

 put upon record an account of the apparatus, the manner of making 

 the experiments, and the causes of failure, in order that subsequent 

 workers may benefit from the experience gained during the voyages. 



The apparatus (fig. 1) was of the enclosed barometer type arranged for 

 photographic records. The air sealed in the barometer cistern was main- 

 tained at as constant a temperature as possible in order that its pressure 

 might vary but slightly during the voyage. This pressure, is equal to pgh, 

 where p is the density and h the height of the mercury column. It is clear 

 that any variation in g will occasion a change in the level of the mercury, 

 and that the column will be shorter if gravity increases and longer if it 

 diminishes, provided that isothermal conditions are maintained ; con- 

 versely, a measure of the displacement of the mercury is a measure of the 

 change of gravity. Four similar barometers were provided, each with a 

 constriction to prevent excessive ' piunping ' of the mercury surface at 

 sea ; each was mounted in a metal sheath and provided with two ther- 

 mometers, one, graduated in 0-01° C, for measuring the temperature of 

 the air in the reservoir, and another, graduated in 1-0° C, for obtaining 

 the stem temperature. 



The barometers were mounted in pairs at the ends of a wooden light- 

 tight box, through the centre of which passed a photographic film operated 

 by clockwork (fig. 2). An 8-volt lamp illuminated each meicury surface 

 and a separate lens focussed the image of each meniscus upon the moving 



