and of Carbonates by the Light of the Sun. 167 



maximum effect, yet nothing in the shape of absolute mea- 

 sures of quantities can be obtained. When however gas can 

 be collected and its volume determined, as in the voltameter 

 and in the experiments just described, the results possess a 

 degree of exactness which enables us to draw from them de- 

 finite conclusions. 



Let us now proceed to determine the constitution of the 

 gaseous mixture given off during their decompositions. It is 

 not pure oxygen, as has often been supposed and often dis- 

 proved, but a mixture of oxygen, nitrogen and carbonic acid. 

 It is mainly to the ratio of the two former that attention has 

 to be directed, the amount of the latter is always variable in 

 different trials. Before proceeding to this there are certain ob- 

 servations to be premised, the results of which, though familiar 

 to chemists accustomed to gaseous analysis, deserve a place 

 here, for they seem to be wholly overlooked in many of the 

 experiments connected with the so-called respiration, but 

 rather digestion, of plants recorded in the books of botany. 



When gas of any kind is confined over water in the pneu- 

 matic trough, its constitution is undergoing incessant change. 

 A portion of it dissolves more or less slowly in the water, and 

 in exchange it receives from the water gas which is always 

 dissolved therein. If two jars, filled with different gases, stand 

 side by side on the shelf, each is incessantly disturbing the 

 constitution of the other, nor does this disturbance cease until 

 the contents of both jars are chemically the same. There are 

 some beautiful experiments of easy repetition which serve to 

 show how rapidly gases and vapours can thus percolate through 

 fluids. Take a pint bottle and pass through its cork, which 

 ought to fit it very loosely, a glass tube a foot long, drawn 

 narrow at its upper end. Into the bottle put a few drops of 

 water of ammonia. Dip the wide end of the tube into a so- 

 lution of soap, and introduce it into the interior of the bottle, 

 adjusting it in such a position by the cork, that when air is 

 blown in at the narrow end, the soap bubble which expands 

 at the wide end may occupy the middle of the bottle. Placing 

 the lips on the narrow end, blow a bubble an inch or more 

 in diameter, and, without loss of time, cautiously draw back 

 again the air from the interior of the bubble into the mouth. 

 A strong ammoniacal taste is at once perceived. Now it is 

 obvious that this ammonia must have passed with very great 

 rapidity through the bubble. 



A still more instructive experiment may be easily made. 

 Take a three-ounce bottle with a wide neck, close the mouth 

 of it by a film of soap water, by passing the moistened finger 

 over it. Place it under a jar of protoxide of nitrogen. In- 



