104 Mitchell on the Penetrativencss of Fluids. 



agents chemically attractive of a particular gas, when placed be- 

 neath the membrane, would promote its permeation. In fact, lime 

 water and solution of baryta were rapidly carbonated by the trans- 

 mission of carbonic acid, and sulphuretted hydrogen almost instantly 

 precipitated the lead of the acetate placed in solution on the opposite 

 side of the membrane, which became black on the side of the solu- 

 tion. A neater mode of performing this experiment is the following: 

 Inject by means of a gum elastic bottle and pipe, into a very small 

 bag of gum elastic, stretched until fully transparent, a solution of 

 the substance to be acted on. Carefully tied, washed, and dried, the 

 bag is to be passed up through mercury into a receiver holding the 

 gas, which for solution of baryta should be carbonic acid, and for 

 that of acetate of lead, hydrosulphuric acid. In a few moments, in 

 the former case, a white coat is seen to completely line the internal 

 surface of the bag, and in a few minutes to fall down and accumu- 

 late at the bottom of it. In the latter case, the inner coat is dyed 

 indelibly black. In either case, if water be alone placed in the bag, 

 it will absorb a considerable quantity of either of these gases, and 

 their presence may be ascertained by the usual tests. 



If any suspicion had arisen that the gases escaped or entered by 

 the route of the space included under ligature, it was dissipated by 

 all the experiments mentioned in the last section ; inasmuch as in 

 the first experiment, that with the lamp-glass, the gas was seen to 

 stud beautifully the under surface of the membrane, standing on it 

 in minute drops or bubbles, mistaken at first for water. In the 

 experiments with baryta and lead in bags, the whole surface was 

 covered, the precipitation taking place only there. Especially was 

 it manifest in the last experiment, where the inner surface was 

 stained black, while the solution remained clear and colourless. The 

 gas, therefore, penetrates through every part of the membrane. 



Being desirous of ascertaining more accurately the relative facilities 

 of transmission, I solicited the assistance of my friend and pupil, 

 Professor J. K. FINLEY, to whose patience, skill, and delicate 

 manipulation, I owe much of the certainty of the following experi- 

 ments. 



Having constructed a syphon of glass with one limb three inches 

 long, and the other ten or twelve inches, the open end of the short 

 leg was enlarged and formed into the shape of a funnel, over which 

 finally was firmly tied a piece of thin gum elaslic. By inverting 

 this syphon and pouring into its longer limb some clean mercury, a 

 portion of common air was shut up in the short leg, and was in 

 communication with the membrane. Over this end, in the mercu- 



