Manchester Memoirs, Vol. xlvii. (1902), No. 3. 3 



paraffin wax to protect the glass from the action of the 

 alkaline solution. 



As I had employed Pettenkofer's method for a long 

 series of determinations of carbon dioxide in air,* in which 

 unprotected glass vessels were used, I thought it worth 

 while to re-investigate the subject, and endeavour to 

 settle the disputed point. 



That glass, exposed for a long time to the action of 

 baryta, is acted on, cannot be doubted. I possess a flask 

 in which baryta water was kept for at least fifteen years, 

 which has, to a great extent, lost its transparency, while 

 its original smooth surface is roughened. The question at 

 issue, therefore, is not whether glass is attacked by baryta 

 water, but whether, in the short period required for a 

 Pettenkofer test, there is any loss in strength of the 

 baryta solution which can affect the determination of 

 carbon dioxide. 



In my first experiments lime and baryta water alone 

 were used and their action tested, not only on the bottles but 

 also on finely divided silica and on powdered glass. Three 

 bottles of green glass (such as " Winchester " quarts are 

 usuallymadeof)of35oc.c.capacity were thoroughly cleaned, 

 rinsed with clear lime water of known strength, and then 

 half filled with the solution. To the first, nothing else was 

 added, to the second, 01 gramme of finely divided silica, 

 and to the third, i gramme of powdered glass of the same 

 quality as the bottles used. A similar set of three bottles 

 was prepared, but containing baryta instead of lime water. 

 After standing a week, with occasional shaking, 10 c.c. of 

 the clear liquid from each bottle were removed by means 

 of a pipette, and titrated with half deci-normal hydro- 

 chloric acid. Phenol-phthalein was used as indicator. 

 The tests were repeated at intervals, and the results are 

 recorded in the following table : — 



* The Air of Rooms, Taylor, Garnett, Evans & Co., Manchester. 



