152 Transact ions. 



theory is insufficient to explain the fluctuations of the well with the 

 fluctuations of the barometer as observed at Lincoln. 



In the above calculation it was assumed that the gas-bubbles formed 

 remained in the water, but since the changes in pressure are very 

 gradual, since the water is always flowing upward, and since one-third 

 of the total liberation of gases takes place in the top 30 ft., it is 

 evident that the bubbles of gas must escape, and therefore cannot raise 

 the level of the water anything like the 1-8 in. calculated above, much 

 less raise it the 8 in. recorded by the observations. 



An explanation of the rise of the well with decrease of barometric 

 pressure more in accordance with the observed facts is as follows : 

 Water must continually be drawn away from the water-table at the 

 outcrop by the flow of water from the well, and more particularly the 

 flow at the lower outcrop of the stratum under the sea. Well-sinkers 

 find that the water runs in certain fairly defined streams in the water- 

 bearing strata, and at Islington is to be seen a very large and freely 

 moving underground stream running through the shingle at the bottom 

 of an open well 42 ft. deep. Small particles of sand have therefore 

 been removed from these strata, and the water can move freely ; 

 but the land over the water-table at the outcrop is not thus freed 

 from small particles, and, as the water is removed, the air has a diffi- 

 culty in following the water downwards, and so a partial vacuum is 

 set up over the water at the outcrop, after the manner of the pro- 

 duction of a Sprengel's vacuum. The water in the water-bearing stratum 

 and the water in the well-pipe now form the two arms of a water- 

 barometer, at the open end of which the observations are being taken. 

 Since the open end is being observed, the water goes up when the 

 mercurial barometer goes down ; since it is a water-barometer, it should 

 go up thirteen times as much as a mercurial barometer falls, but since 

 the vacuum at its closed end is not perfect its motion is not so great 

 as this. It goes up four times as much as the barometer goes down, 

 thus indicating that the vacuum over the water-table at the outcrop is 

 about one-third of a true vacuum — i.e., that the air-pressure amounts to 

 about 10 lb. instead of 15 lb. to the square inch.* 



The Evening Rise. 



That the well at the Museum in Christchurch is usually higher in 

 the evening than in the morning is noted both by Captain Hutton 

 (loc. cit.) and by Mr. Speight (lac. cit.). By both these writers it was 

 thought possible that this evening rise might be caused by the shutting- 

 off of other wells of the same stratum in the near neighbourhood in the 

 afternoon, although Mr. Speight is not inclined to accept this explana- 

 tion. That the shutting-ofl of adjacent wells causes any particular well 

 to rise is proved by Captain Hutton's observation that the Museum well 



* This explanation met with a great deal of adverse criticism at the meeting at which 

 the paper was read. Mr. Hogg and Mr. Page suggested that changes of aerial pressure 

 would be felt directly by the water in the open pipe, but only slowly by the water at 

 the outcrop, owing to the fact that the air superincumbent on the water there is entangled 

 among particles of soil. This, 1 find, is also King's explanation (" The Soil," p. 180). 

 Warrington (" The Physics of the Soil," p. 129) appears to prefer the explanation 

 attributed to King in the present paper in the section on " The Evening Rise " — viz.. 

 with a falling barometer the air in the soil expands, and the water filling the interstices 

 above the water-level h expelled, and causes a rise in the water-level of the soil. Either 

 of these explanations is perhaps sufficient to account for the fluctuations observed, but 

 I still regard my explanation as a possible, and even a probable, one. 



