362 Kreider — Oxygen in Air and in Aqueous Solution. 



volume of moist air which 100^""' of air at 0° and 760°^^ would 

 occupy under the given conditions ; this standard being very 

 carefully determined. By m^ans of a T-tube this standard 

 burette was placed between and in connection with the burette 

 in which the volume of air to be analyzed was measured, and 

 a movable reservoir of mercury. Both burettes were firmly 

 fastened to a movable iron rod and the zero marks accurately 

 adjusted to the same level. By drawing into the measuring 

 burette a volume of air greater than that required — for which 

 purpose a small bulb was attached to the lower end of the 

 burette, and then by raising the reservoir of mercury, com- 

 pressing the air in the standard tube to the 100^™^ mark, at 

 the same time allowing the excess of air to escape from the 

 measuring burette, exactly 100'^'''^ of air under the standard 

 conditions of temperature and pressure was obtained. To 

 facilitate the adjustment, two strips of wood were fastened to 

 the rubber connection by means of screw pinch-cocks in such 

 a way that by closing one pinch-cock the flow of mercury from 

 the reservoir could be shut off, and then by gradually tighten- 

 ing the other pinch-cock the mercury would be forced out of 

 the rubber, and thus an easy and accurate adjustment to the 

 2QQcm3 j^^j.]^ ]3g secured. 



The apparatus in which the action of the oxygen upon 

 hydriodic acid was effected consisted of a 300°"^^ bulb pipette, 

 both ends of which were cut off short and sealed to glass 

 stop-cocks. The tube from one of the stop-cocks was cut off 

 short after being tapered and restricted so as to hold a rubber 

 connector tightly, while the tube from the other stop-cock 

 was left sufficiently long to reach to the bottom of a 500°°^* 

 Erlenmeyer beaker. These tubes are preferably of about 3°^"' 

 bore, since for the several connections all air may be expelled 

 from tubes of this size by displacement with water. In order 

 to expel all air from the flask, instead of passing a current of 

 carbon dioxide as was done in the determination of perchlor- 

 ates, time was saved by first filling the flask with water, which 

 was then displaced by pure carbon dioxide (prepared as 

 described below) and the flask subsequently exhausted, which 

 was accomplished instantaneously by the device described in 

 the article on perchlorates. The required amounts of potas- 

 sium iodide solution, hydrochloric acid and nitric oxide were 

 drawn in in the order named, after which the measured volume 

 of air was gradually admitted while the bulb was constantly 

 agitated so as to keep the hydriodic acid continually renewed 

 along the surface of the bulb. The shaking was continued 

 for a minute or two until the action was completed, when a 

 dilute solution of potassium bicarbonate was admitted. The 



