400 Scientific Proceedings, Royal Dublin Society. 



alike, while that representing sea-water is widely different (see fig. 4). This 

 difference is evidently due to a difference in the saturation value, for 

 when the experimental results are expressed as percentages of saturation, 

 the curves from the different experiments are almost coincident, as shown 

 in fig. 5. 



The results of the experiments show that the bubble of air in passing up 

 the tube continually exposed fresh water-surface to the air, and at the same 

 time kept the water mixed. In order to confirm this view, a test experiment 

 was made with a tube 5 feet long. This was filled in the ordinary way, and 

 the inversions carried out, until the air-content was about 60-70 per cent, of 

 saturation. The water in the tube was then drawn into the pump in two 

 equal portions, each of which was analysed separately. One of the analyses 

 represents the air-content of the water in the upper half of the tube, and the 

 other that in the lower half. 



Totals, . . 0-776 cc. 0'7S8 cc. 



These figures show that the difference in air-content between any one 

 portion of the water in the tube and any other is not appreciable on 

 analysis ; hence the assumption that the water is well mixed seems to be a 

 reasonable one. 



In order to keep a check on the manometer readings, the air-content of 

 the water was determined after each experiment by boiling out in vacuo and 

 measuring the gases, using the apparatus described above. A comparison of 

 the results obtained by the different methods is given in Table 4. 



The area of the surface of the bubble was calculated from the measurements 

 of its length when at rest and in motion and the known internal diameter of 

 the tube. 



The volume of the bubble was 15 cc. Its length when in motion was 

 16-04 cms. and when at rest 13-30 cms. The diameter of the tube was 1*20 

 cms. ; hence its sectional area was 1-13 sq. cm. 



The total volume of the water supported around the bubble was the 

 amount which collected when the bubble burst, i.e., = (16-04 - 13 - 3) 

 1-13 cc. = 3-10 cc. 



This volume is made up of the portion round the hemispherical cap of the 

 bubble and the cylindrical shell below. 



