206 PROCESS OF DEVKLOPEMENT. CttAP. IV. 



itself at all ; which circumstance M. Mirbel adduces 

 as an additional argument in favour of his hypothe- 

 sis, though to me it seems an argument against it. 

 For if the air produces such violent effects upon the 

 trunk and branches of woody plants, why should it 

 not produce similar effects upon other plants, or upon 

 other parts of the same plant ? and why is the epi- 

 dermis of the leaf and fruit incapable of being again 

 regenerated if accidentally destroyed ? Till a satis- 

 Whose factory answer can be given to these questions it is 

 ^P^iJ! impossible to admit the hypothesis of M. Mirbel. 

 sible. But so far is the action of the external air from 

 being the cause and origin of the epidermis, that it 

 is even detrimental to its formation. For the repro- 

 duction of a part that has been destroyed, in cases 

 capable of reproduction, is always more easily 

 effected if the wound is covered closely up. And 

 hence it is extremely improbable that the epidermis 

 is merely a modification of the external surface of 

 the parenchyma effected by the influence and action 

 of the air; if rather it is not evidently an organ 

 formed by the agency of the vital principle, even 

 while the plant is yet in embryo, for the very pur- 

 pose of protecting it from injury when it shall have 

 been exposed to the air in the process of vegetation. 

 Its growth or dcvelopement is accordingly found to 

 keep pace with that of the plant which it invests as a 

 sheath, extending in all its dimensions and accom- 

 modating itself with wonderful facility to the ex- 

 pansion of the interior parts ; as may be seen in large 



