1893.] 



MICROSCOPICAL JOURNAL. 



25 



tubing in the form of a diaphragm, and held in place by the 

 overlapping india-rubber tui)ing. This prevents the possibility 

 of the narrow part of the tube being choketl up by any minute 

 particles. 



Fluid 2 thus enters into the mixing bottle at an extremely slow 

 rate of How. and l)ecomes completely ilitl'used. at first in 

 extremelv minute cpiantitv, through lUiid i. The tluid from the 

 mixinsr bottle is meanwhile entering bottle B at the same ex- 



^zy 



Mixiiij; I'.utllc, pp. -."4-5. 



Walcr Purifi'.r, p. 29. 



tremelv slow rate, and it is oi)vious that with two fluids that 

 readily mix, fluid i may be made to replace fluid 2 in bottle B 

 with the required excessive slowness and regularity. 



In the case of some of the liquids used in fixing and preserving, 

 it is not necessary to use such a precaution as this. We may 

 substitute saturated solution of corrosive sublimate for sea-water 

 without the least risk of damage to the most delicate structures, 

 the specific gravity of the tube being very near the same. 



