BY ETHEL C. PINKERTOX. 153 



column of gas passes through a straight tube for a long distance 

 in the form of a spike when there is no resistance in the tube, 

 the tube employed was lightly packed with glass wool for three 

 inches. The brass tube used in the experiment had a bore of 

 20 mm., and was 5 metres long. It was perfectly straight. A 

 brass mouthpiece, 5 cm. long and of 15 mm. bore, was soldered at 

 one end of it, and a coiled rubber-tube, 2 metres Ion 



&5 



was 



attached to the opposite end, the end of the rubber-tubing 

 dipping into a vessel of water. The total volume of the brass 

 and rubber tubes was approximately 2,200 c.c. Just beyond the 

 mouthpiece and at intervals of 25 cm., capillary brass tubes 

 were securely soldered with silver into the main tube. These 

 served for the withdrawal of samples. The gas to be used in 

 the experiment was placed in a strong air-tight vessel. This 

 vessel was fitted with a cork through which passed two bent 

 glass tubes, by means of which the vessel could be connected to 

 the water-supply, and also to the mouthpiece of the brass tube. 

 The vessel was completely filled with water, which w^as displaced 

 by a mixture of air and carbon dioxide. The rubber tubing, 

 connecting tlie vessel to the mouthpiece of the brass tube, was 

 tightly clamped with a strong spring clip, and the interior of the 

 Aessel was connected with the water-supply ho that the gas in 

 t4iie vessel w^as strongly compressed. When the pressure was 

 sufficient, the connection with the water-supply was clamped off. 

 The clamp connecting the gas with the tube was released for a 

 fraction of a second, and at once retightened. As the far end 

 of the tube was under water, no air could be sucked in as a 

 result of any ch.aiige in the volume of the air in the tube. The 

 volume of the vessel containing the gaseous mixture in the first 

 set of experiments was three litres, which was compressed to half 

 its volume by the water-pressure. Later, a larger vessel holding 

 11,750 c.c. was employed. In this, the amount of compression 

 of the gaseous mixture was varied. In the later series, the end 

 of the hose was so placed that the volume of gas passed through 

 the tube, could be measured. The results of the first sets of ex- 

 periments are recorded in Table iii. 



OS 



LIBRARY zo\ 



