No. 453 ] AERATING LIQUID SOLUTIONS. 659 



tube is light. Although by this device the bubbles are not 

 broken up as much as they can be by using a piece of sponge 

 forced into the end of the tube, this method has ihc dec idcd 

 advantage of requiring much less pressure, and Ihcrc is also 

 much less danger of clogging the tube than if a sponge is us^-d. 



In making the aerating constriction the tube should he healed, 

 without being drawn out, until the walls fall in. as this causes 

 them to become much thicker and consequently sliongei than 

 they otherwise would be. 



It is much better to use a capillary siphon than a larL;er tube 

 drawn out to a capillary ending ; first, because a capillary siphon 

 never needs to be filled, for when the supply jar is filled t«j w ithin 

 about I cm. of the top capillary attraction will fill the siphon 

 and start it running ; secondly, because a large tube drawn out 

 at one end is much more readily clogged than a small one of 

 equal diameter throughout. 



The constantdevel siphon [F) should be unquestionably large 

 enough to drain off the solution as fast as it comes into the cul- 

 ture jar, but not much larger ; neither should its outer end be 

 more than a few centimeters lower than its highest jooint, 

 for if it is the decrease in pressure at the highest point causes 

 sufficient increase in vapor pressure of dissolved gases to cause 

 them to be thrown out of solution, to collect at this point, and 

 thus to clog the siphon. 



The principles of operation in the second ai)])aratus, tliat, 

 represented in Figure 2, are in all res]:)ects similar to those 

 discussed with reference to the apjmratus rei)resented in I-igure 

 I. The upper end of the aerating tube, the eai)i]lary siphon, 

 and the supply jar are the same as rei)resentetl in h'iguie i. 



The chamber G, which .serves to separate the air from the 

 liquid, can be readily made by cutting off a test tube. Jt should 

 be about 1.5 cm. in diameter and 4 or 5 cm. long. It must of 

 course be air-tight. Ordinary corks soaked in paraffin will serve 

 very well to close the ends. 



The tube should be at least 5 mm. in diameter, slightly 

 larger than the aerating tube proper, /. large enough to i)re- 

 vent air being forced through it with the solution. Its length 

 must be such that its outer end is somewhat farther above its 



