HuTCHiNSOX — On the Estimation of Oxygen in Water. 145- 



•accurate methods. The process adopted by Schutzenberger for check- 

 ing the results appears to have been, submitting the liquid to the 

 action of the mercury pump for fifteen or twenty minutes. I do not 

 think this is sufficient, for, in expelling the mixed gases from a water 

 by boiling, I have found that traces of gas are given off even after 

 a considerable period. 



One great drawback is the considerable amount of change the' 

 hyposulphite solution undergoes, even when excluded from the air 

 and kept in the dark. The following give the volume ratios between 

 it and the indigo, as taken on different days, showing the extent of 

 this change : — 



Oct. 23rd. — 25 cub. cents, indigo = 7 cub. cents, hyposulphite. 

 ,, 25th. — 25 ,, ,, ,, = 7"84 ,, „ ,, 



,, 27th. — 25 ,, ,, ,, = 9"8 ,, ,, ,, 



Nov. 1st.— 25 „ ,, ,, = 13 ,, „ ,, 



„ 6th.— 25 „ „ „ =20-75 „ „ 



These investigations were carried out in the Chemical Laboratories 

 of the Eoyal College of Science, under the supervision of Professor- 

 Galloway. 



