190 



PRESSURE. 



in order to show the curves for twelve months and for all twelve stations on plate II 

 the scale used for plate I has had to be diminished, but the relationship between pressure 

 and time remains the same on the two plates. 



One or two striking results are at once apparent. The curves for Seychelles and Bombay 

 show that for all practical purposes surges do not exist in the tropics. Also the curves 

 for Adelaide, Greenwich and Irkoutski are obviously intermediate between those for the tropica 

 and the circumpolar stations. 



Further discussion of the curves can only be undertaken by means of a similar statistical 

 investigation to the one used for the pressure waves. As with the waves, surges of greater 

 variation than -2" (.5 m.m.) will be selected and their average length and depth calculated. 

 It is quite obvious that one year is far too short a period to give satisfactory results, but a 

 seneral oversight can be obtained from the following table : — 



Table 110. 

 Pressure surges. March 1902-February 1903. 



The surges at Hut Point were very much larger during the year chosen for the above 

 table than in any other of the four years for which we have observations. I have therefore 

 added in brackets the mean values obtained from all the observations and these should be 

 taken into account in comparing the stations. 



It is not easy to draw any very simple relationship from the numbers, and it is question- 

 able whether one year's observations are sufficient to give satisfactory results, but the following 

 inferences may probably be correct. 



In the southern hemisphere the surges are deeper in high than in low latitudes, also 

 the length of the surges is, on the whole, greater near the continents than over the open 

 ocean. 



