chemical properties, penetrate, in a certain degree, the 
pores of the seed, and there unite with its carbon, 
after it has been brought into a state fitted for such 
combination by the spontaneous changes described 
above. All that we contend for, is> that the separa- 
tion of the carbonic matter from the other elements 
of the seed is not effected by the attractive force of 
the oxygen, but that the seed itself m\ist have previ- 
ously undergone changes, by which its carbon is re- 
duced to a state fitted to combine with the surround- 
ing air j and so far is oxygen from being necessary 
to abstract this carbon, that this substance is often 
given off by the seed where no oxygen gas is pre- 
sent. In ordinary cases, therefore, we believe the 
union of carbon and oxygen to take place at the sur- 
face, or, it may be, within the pores of the seed or 
plant ; but if oxygen be not supplied in due quanti- 
ty, to unite with and carry off the carbon, this sub- 
stance will escape, in combination with nitrogen or 
any gas that surrounds it, according to the laws of its 
affinity for it ; or, if no gas be present, it then passes 
off in an elastic form in union with oxygen (19.), deri- 
ved either from the decomposition of the seed, or of 
the water which it has previously imbibed. Even in 
hydrogen and nitrogen gases, carbon not only passes 
off in a simple form, but also in combination with 
oxygen (5.)> derived from one or other of these 
sources. 
640. Without attending to the state in which we 
have supposed the carbon to pass off from living bo- 
dies, some writers have imagined that we considered 
it to be afforded in a solid form. It has, however, 
been our aim throughout to establish the existence 
