512 



Dr. P. F. Frankland. 



[June 10, 



Method of Experiment. 



Tubes prepared as above are easily transportable in cylindrical 

 cardboard boxes, the exterior of which should be coated white to 

 prevent heating if they have to be exposed to sunshine. The tube is 

 strapped to a small horizontal table supported by an ordinary port- 



Fig. 2. 



able camera stand, and the end of the small tube passing through the 

 cork is connected by flexible tubing with an aspirator, which, so as to 

 render it as portable as possible, consists of two bottles or strong flasks, 

 each of rather more than 1 litre capacity, arranged as in the figure to 

 form a reversible syphon. A measured litre of water is poured into 

 one of these flasks, and by syphoning this into the second, a litre of 

 air is made to pass through the experimental tube. The rate of flow 

 is regulated by a screw clamp, and by alternately connecting the end 

 of the tube with the two syphon flasks any desired volume of air may 

 be drawn through the apparatus. 



The experiments of Hesse show that the rate of aspiration should 

 not exceed 1 litre in two or three minutes, and that when this precau- 

 tion is observed the organisms present in the air are almost wholly 

 deposited in the first two-thirds of the tube, the remaining third being 

 either wholly free or practically so. In my experiments I have almost 



