278 Rev. C. Trotter, On the molecular [May 11, 



viscous solid must be, so to speak, more or less " diagrammatic." 

 In the present state of our ignorance as to the nature of the 

 action between the molecules of a solid it would appear to be 

 premature either to affirm or to deny that there is a distinct line 

 of demarcation between what I have called the " solid " and what 

 I have called the "liquid " parts of the mass. It might be more 

 accurate to speak of the "more solid" and "less solid" portions; 

 all that is involved in my supposition is that while some "elements" 

 of the mass are in a state of aggregation which is essentially that 

 of a solid others are in a state which more nearly resembles that 

 of a liquid, and that we may pass from one extreme condition to 

 the other in so small a space that the whole appears sensibly 

 homogeneous when examined in any ordinary way. When I 

 speak hereafter of the " solid " and " liquid " portions of the mass 

 I must be understood to make use of the words subject to the 

 foregoing qualifying remarks. 



Now whenever we have a portion of any substance in contact 

 with another portion of the same substance in a different state of 

 aggregation we usually assume that there is at the common sur- 

 face a condition of "mobile equilibrium;" that an interchange of 

 molecules is going on between the portion in one state and the 

 portion in the other state. The most familiar instance of this 

 is when a volatile liquid is in contact with its own vapour in a 

 confined space and at a constant temperature, but there is no 

 reason to doubt that a similar state of things exist when a 

 volatile solid is in contact with its own vapour or a fusible solid 

 is in contact with its own liquid under analogous conditions. 



In our supposed viscous body the relations between the "solid" 

 portions and the contiguous " liquid " portions throughout the 

 whole mass will be similar to those which obtain at the common 

 surface of an ordinary fusible solid and its own liquid, and the 

 most natural assumption seems to be that at each common sur- 

 face throughout the mass there is a condition of " mobile equi- 

 librium," so that an individual molecule which at one time formed 

 part of a "liquid" portion of the mass may become attached to 

 and form part of a "solid" portion and vice versa. Subject to 

 the same reservation as applies to the use of the words " solid " 

 and " liquid " we may call those molecules which at any time form 

 part of a " solid " element " attached " molecules and those which 

 form part of a "liquid" element "unattached." 



The peculiar characteristic of a viscous solid, viz. its gradual 

 change of shape under the influence of external forces which are 

 insufficient to produce a sensible change of shape in a very short 

 time, follows as a natural consequence from the preceding sup- 

 position as to its molecular structure. 



The first effect of axi external force is to produce a slight and 



