90 



REPOETS ON THE STATE OP SCIENCE. — 1919. 



ship in the neighbourhood of which the Destroyer manoeuvred.- It was 

 dining these experiments that a new difficulty appeared which seems to 

 add materially to the task of determining gravity over the ocean. It 

 certainly adds to the difficulty of determining the effects of vibration. 



The preliminary experiments showed that the vibration on a Destroyer 

 in calm water does not greatly increase the difficulty of reading the instru- 

 ment. Even when the ship was running at 22 knots in the sheltered waters 

 off Spithead it was not difficult to obtain readings which were consistent 

 to O'l mb., and the writer believes that on many occasions the readings 

 were consistent to 0'05 mb., even with the ordinary vernier type of scale. 

 Had the dial instrument previously employed by the writer been available, 

 the readings coidd have been made with at least as great accuracy, and 

 certainly much more quickly and easily. 



August 29, 1919, was the first day of the trials, which took place off 

 Sj)ithead ; a fresh breeze was blowing from almost due west, which gradually 

 strengthened to about 20 or 25 knots. It was found that the barometer 

 in the chart-room suffered small fluctuations according to the direction 

 in which the ship was heading ; this was at first attributed to the gravita- 

 tional change due to the E.-W. motion of the ship, but, as the wind freshened, 

 the fluctuations became much more marked, and of an order of magnitude 

 which ruled this effect out of account as the main cause. Moreover, the 

 ship 's aneroid, also in the chart-house, showed similar fluctuations. It was 

 evident that the changes of pressure were real, and that they were due to 

 the eddy motion of the wind about the ship's hull. Going West the baro- 

 meters in the chart-house showed invariably a reduced pressure, indicating 

 a suction effect as the ship met the wind with a relative velocity of about 

 45 knots. 



Typical readings in the chart-room are shown below, the aneroid 

 referred to was that belonging to the ship : — 



The aneroid scale is very much in error, but the changes recorded by it 

 are certainly real. Two features are obvious : (1) the fall going west 

 against the wind, and the rise going east, the difference sometimes 

 amounting to 1-3 mb., and (2) the gradual change in the reading of the 

 mercury barometer during each run. The lag is due to the constriction 

 in the barometer tube ; the lag of the aneroid was scarcely appreciable. 



Closing the port-holes and the door of the chart-house made little 

 difference to the readings, the fluctuations being just as marked as when 

 they were all open. In the ward -room the barometer showed fluctuations 

 in the same sense, a fall going west and a rise going east ; but the change 

 was far less, amounting to 0-6 mb. as a maximum, of which 0-2 mb. is 

 probably due to the gravitational effect of the E.-W. motion. Subsequent 

 experiments with a specially sensitive aneroid confirmed the existence of 

 these fluctuations, which are superposed upon any general atmospheric 

 change of pressure. Few cabins could be more favourably situated for 

 avoiding eddies than the ward-room, as its only opening to the deck is 



