NATURAL PHILOSOPHY. 215 



arranged. But we need not confine ourselves to the material theory 



^j 



of heat. Assuming the hypothesis which is now gaining ground, that 

 heat, instead of being an agent apart from ordinary matter, consists 

 in a motion of the material particles ; the conclusion is equally proba- 

 ble that the transmission of the motion must be influenced by the 

 manner in which the particles are arranged. Does experimental sci- 

 ence furnish us with any corroboration of this inference ? It does. 

 More than twenty years ago, MM. De la Rive and De Candolle 

 proved that heat is transmitted through wood with a velocity almost 

 twice as great along the fibre as across it. This result has been re- 



C 1 ^5 



cently expanded, and it has been proved that this substance possesses 

 three axes of calorific conduction ; the first and greatest axis being 

 parallel to the fibre and to the ligneous layers ; while the third axis, 

 which marks the direction in which the greatest resistance is offered 

 to the passage of the heat, is perpendicular to the fibre and parallel 

 to the layers. 



But it is the modification of the magnetic force by the peculiarities 

 of aggregation, which forms the subject of the evening's lecture. 

 What has been stated regarding heat applies with equal force to mag- 

 netism. The observed magnetic phenomena are of a composite char- 

 acter. The action of a magnetic mass is the resultant action of its 

 molecules, and will be influenced by the manner in which they are 

 aggregated. The fundamental phenomena of magnetism are too well 

 known to render it necessary to dwell upon them for an instant. A 

 small bar of iron was suspended in the magnetic field ; it set its 

 length parallel to the line joining the poles. Should we be justified 

 from this experiment in concluding that a magnetic mass will always 

 set its longest dimension axial ? No. A second magnetic bar, equal 

 in size to the former, was suspended between the poles ; it set its 

 length at right angles to the line joining the poles. Whence this de- 

 portment ? We find the reason of it in the mechanical structure of 

 the bar : it is composed of magnetic plates, transverse to its length ; 

 these plates set from pole to pole, and hence the length of the bar 

 equatorial. But let us proceed from this coarse experiment to one 

 more delicate, where nature herself has imposed the conditions of ag- 

 gregation. A plate taken from a mass of shale, picked up a few 

 weeks ago in the coal district of Blackburn, was suspended between 

 the poles ; although strongly magnetic, it set its longest dimension at 

 right angles to the line joining the poles. This deportment was at 

 once explained by reference to the structure of the mass ; it also 

 though apparently compact, was composed of layers transverse to its 

 length ; these layers set from pole to pole, and hence the length equa- 

 torial. Let us ascend to a case still more refined. A crystal of sul- 

 phate of nickel was suspended between the poles, and on exciting the 

 magnet a certain determinate position was taken up by the crystal. 

 The substance was magnetic, still its shortest dimension set from pole 

 to pole. The crystal was removed from the magnetic field and the 

 edge of a penknife placed along the line which set axial ; a slight 

 pressure split the crystal and disclosed two beautiful surfaces of cleav- 



