12 REPORT — 1901. 



whether there is valid ground for believinof that all matter is made up of 

 discrete parts ; and secondly, whether we can have any knowledge of the 

 constitution or properties which those parts possess. 



The Coarse-grainedness of Matter. 



Matter in bulk appears to be continuous. Such substances as water 

 or air appear to the ordinary observer to be perfectly uniform in all their 

 properties and qualities, in all their parts. 



The hasty conclusion that these bodies are really uniform is, never- 

 theless, unthinkable. 



In the first place the phenomena of diffusion afford conclusive proof 

 that matter when apparently quiescent is in fact in a state of internal 

 commotion. I need not recapitulate the familiar evidence to prove that 

 gases and many liquids when placed in communication interpenetrate or 

 diffuse into each other ; or that air, in contact with a surface of water, 

 gradually becomes laden with water vapour, while the atmospheric gases 

 in turn mingle with the water. Such phenomena are not exhibited by 

 liquids and gases alone, nor by solids at high temperatures only. Sir W. 

 Roberts Austen has placed pieces of gold and lead in contact at a tem- 

 perature of 18° C. After four years the gold had travelled into the lead 

 to such an extent that not only were the two metals united, but, on 

 analysis, appreciable quantities of the gold were detected even at a dis- 

 tance of more than 5 millimetres from the common surface, while within a 

 distance of three-quarters of a millimetre from the surface gold had 

 penetrated into the lead to the extent of 1 oz. 6 dwts. per ton, an amount 

 which could have been pi'ofitably extracted. 



Whether it is or is not possible to devise any other intelligible account 

 of the cause of such phenomena, it is certain that a simple and adequate 

 explanation is found in the hypothesis that matter consists of discrete 

 parts in a state of motion, which can penetrate into the spaces between 

 the corresponding parts of surrounding bodies. 



The hypothesis thus framed is also the only one which affords a rational 

 explanation of other simple and well known facts. If matter is regarded 

 as a continuous medium the phenomena of expansion are unintelligible. 

 There is, apparently, no limit to the expansion of matter, or, to fix our 

 attention on one kind of matter, let us say to the expansion of a gas ; but 

 it is inconceivable that a continuous material which fills or is present in 

 every part of a given space could also be present in every part of a space 

 a million times as great. Such a statement might be made of a mathe- 

 matical abstraction ; it cannot be true of any real substance or thing. 

 If, however, matter consists of discrete particles, separated from each 

 other either by empty space or by something different from themselves, 

 we can at once understand that expansion and contraction may be nothing 

 more than the mutual separation or approach of these particles. 



Again, no clear mental picture can be formed of the phenomena of 



