362 Dr. W. E. Adeney on the 



clipped at different depths, from 10 to 200 mm., below the 

 surface. 



Slow currents of air were employed, at the rate of 1000 c.c. 

 per hour, with the smaller bore inlet-tubes, and at somewhat 

 quicker rates with the tubes of the larger bore. 



In a few cases, the surfaces of the water-columns were 

 agitated by mechanically rotated glass stirrers, just dipping 

 below the surface of the water. In these cases the experi- 

 mental tube employed was of wider bore to allow of stirrers 

 of a gridiron shape being used. 



When the surfaces of the water-columns were left un- 

 broken, the experimental tubes were loosely covered to 

 protect their contents from dust. 



The height of the columns of water were in nearly all expe- 

 riments about 1800 mm. In all experiments the surfaces of 

 the water were freely exposed to the air. Observations of 

 barometric pressures have not been thought necessary. 



At the conclusion of each experiment, layers of the water 

 of 100 to 200 mm. thick were drawn from different depths 

 from the surface, and stored in tubes standing over mercury 

 until they could be analysed. 



The dissolved gases in these samples were extracted by 

 boiling with a little sulphuric acid in vacuo. The apparatus 

 employed for this purpose, and for the analysis of the gases, 

 has already been described by the author in his memoir on 

 Dissolved Gases and Fermentative Changes*. 



Of the results of the analysis of the dissolved gases, only 

 those for the nitrogen are recorded in this communication. 

 These are expressed in c.c. at 0° C. and 760 mm. bar. per litre. 



Description and Eesults of Experiments. 



Experiments ivith Unbroken Surfaces. 



No. 1. — Two tubes, each 1950 mm. long and 18 mm. bore, 

 were filled to the height of 1800 mm. from the bottom, 

 one with sea- water and one with distilled water, both being 

 nearly nitrogen free, and were immersed in the water- 

 jacket and kept undisturbed for 28 days. The tempe- 

 rature of the room varied between 16°'5 and 10 c C, that 

 of the jacket between 12°'9 and 8°"0 C. During the last 

 week of the experiment the temperature of the room 

 varied between 12° and 10° and that of the jacket between 

 9° and 8° C. 



* Tians. Royal Dublin Society, to], y. Part 2 (1895). 



