22 



pearance *. This experiment we repeated by pass- 

 ing up half a dozen soaked peas into a graduated 

 tube filled with mercury, and immersed in a jar of 

 the same fluid. By the following day two-tenths of 

 a cubic inch of air were collected in the upper part 

 of the tube, and the experiment was continued till 

 the bulk of air was increased to 1.6 cubic inch, be- 

 yond which it did not proceed. Lime water was 

 then passed into the tube, and with the aid of a lit- 

 tle agitation, all the air except a small bubble was 

 rapidly attracted by it. The peas did not exhibit 

 the slightest sign of germination. These facts show, 

 that the production of this acid during the growth 

 of seeds is altogether different, both in its formation 

 and the consequences which attend it, from that pro- 

 duction of it which takes place where no living ac- 

 tion exists. Where this acid is formed independent 

 of germination, no change is effected in the seed, 

 and the bulk of air is increased : where it is pro- 

 duced by that process, a part of the air disappears ; 

 the evolution of the seed goes on ; and the quantity 

 of the air is somewhat diminished. Hence, therefore, 

 we conclude with De Saussure, that in germination 

 the seed does not form carbonic acid from its own 

 substance, but furnishes only one of the constituent 

 parts of it, namely, the carbon t ; and farther, that 

 when it does form this acid, independent of oxygen 

 gas, it is only under a state of decomposition, (5.), 



* Experiments on Sugar, loc. cit. 



f Journal de Physique, torn. xlv. p. 98. 



