248 ■ Mr Jolin Eattray on the 



with its lower extremitj^ under tlie surface of the mercury 

 in the mercurial trougli. A reading of the amount of gas 

 present, was now taken, the temperature of the room and 

 the height of the harometer at the time being at the same 

 time recorded. Potash and pyrogallic acid were tiien added 

 in turn, and after allowing adequate time for absorption of 

 the oxygen to take place a second reading of the remaining 

 volume was made, the temperature and pressure being noted 

 as previously. In all cases this direct method was employed 

 in analysing the gases. 



In reading the volumes of these gases, the level of the 

 centre or summit the downw^ardly convex meniscus has been 

 taken, and the observations were always made in the outer 

 room of the floating laboratory, at a distance from all 

 artificial heat, and in a place secluded from the action of sun- 

 light. The mercury in the trough, after becoming unsuitable 

 for work, owing to the presence of pyrogallate and water, 

 was cleaned by being passed through filter paper. 



Ttisimportant to note that the presence of micro-organisms 

 of a vegetable or animal nature in the sea water inside the 

 apparatus exercises another disturbing influence on the 

 gaseous products ; but, as already stated, this factor has not 

 been found to cause an accumulation of gas in the empty 

 apparatus, and its influence, though real, is not included in 

 the annexed calculations. 



The corrections of the volumes of the gases for tempera- 

 ture and pressure have been calculated from the following 

 formula : — 



V' = . 



Vx(B-&-T) 



760 X (1 + 0-003665 0' 



where V' — corrected volume, V = ob3erved volume, B = 

 height of barometer at time of reading, & = difference between 

 height of mercury in trough and in eudiometer, ^=tempera- 

 ture in degrees centigrade, T = tension of acpieous vapour in 

 mm. of mercury at temperature t. With regard to the 

 value of h in the above equation, it is to be noted that the 

 watery substance in the eudiometer have been reduced to 

 mm. of mercury by regarding ^ of the volume of these 

 substances as giving their mercurial equivalent. This 

 factor is of importance, as the height of the column of 



