HEAT. 57 



If we select a substance of a different but also of a definite 

 structure, such as wood, we find that heat progresses through 

 it with more or less rapidity, according to its direction with 

 reference to the fibre of the wood : thus Decandolle and De 

 la Rive found that the conduction was better in a direction 

 parallel to the fibre than in one transverse to it ; and Dr. 

 Tyndall has added the fact, that the conduction is better in a 

 direction transverse to the fibres and layers of the wood than 

 when transverse to the fibre but parallel to the layers, though 

 in both these directions the conduction is inferior to that fol- 

 lowing the direction of the fibre. Thus, in the three possible 

 directions in which the structure of wood may be contem- 

 plated, we have three different degrees of progression for 

 heat. 



In the above examples we see, as we shall see farther on 

 with reference to all the so-called imponderables, that the 

 phenomena depend upon the molecular structure of the mat- 

 ter affected ; and although these facts are not absolutely in- 

 consistent with the theory which supposes them to be fluids 

 or entities, it will, I think, be found to be far more consistent 

 with that which views them as motion. Heat, which we are 

 at present considering, cannot be insulated: we cannot re- 

 move the heat from a substance and retain it as heat ; we 

 can only transmit it to another substance, either as heat or 

 as some other mode of force. We only know certain changes 

 of matter, for which changes heat is a generic name ; the 

 thing heat is unknown. 



Heat having been shown to be a force capable of pro- 

 ducing motion, and motion to be capable of producing the 

 other modes of force, it necessarily follows that heat is capa- 

 ble, mediately, of producing them ; I will, therefore, content 

 myself with enquiring how far heat is capable of immediately 

 producing the other modes of force. It will immediately 

 produce electricity, as shown in the beautiful experiments of 

 Seebeck, one of which I have already cited, which experi- 

 3* 



