237 
the ultimate molecules are thrown into a state of agitation, or 
rather of greater agitation than before, and that it is this state of 
agitation that constitutes thermometric heat. According to the 
molecular theory of gases, which is in great measure due to 
Maxwell himself, and which has now received such remarkable 
confirmations that it may be considered pretty well established, 
in a gas the molecules are for the most part free, provided at 
least the gas be not under a very high pressure, and are 
moving about with very high velocities, and occasionally coming 
in contact with one another, or, what comes to much the same, 
so close as powerfully to affect each other's motion. The 
velocity is not the same for the different molecules, and if it- 
were it would not remain so, for as they came casually into 
collision some would be so struck as to be made to move faster, 
and others so as to be made to move more slowly ; it is only a 
sort of average state of agitation that remains permanently 
unchanged so long as the condition of the gas remaius the 
same. 
Suppose now our imaginary vessel divided into two by a 
thin partition, and suppose this partition pierced with a vast 
number of very minute holes, each large enough to let through 
one molecule at a time, but not much larger than that. 
Imagine each little hole closed by a sliding shutter, and sup- 
pose each shutter presided over by a minute intelligent creature, 
that Maxwell called a demon. Suppose it were wished to 
have one, call it the right hand, compartment of the vessel 
filled with warmer and the left hand compartment filled 
with cooler gas. This might be effected by the demons by 
suitably opening or closing the shutters. When a demon 
saw a quickly-moving molecule approaching his hole from left 
to right, or a slowly-moving one approaching it from right 
to left, he would open the shutter to let it through. When 
he saw a slowly-moving molecule approaching the hole from 
left to right, or a quickly-moving one approaching it from 
right to left, he would shut the shutter to stop it. Thus 
after a time the right-hand compartment would be filled with 
molecules which on the whole were moving more rapidly, 
and the left-hand compartment with molecules which, on the 
whole, were moving less rapidly than the average. If the 
limits of speed which decided whether they should shut or 
open the shutters for the molecules moving to right or left 
were properly chosen by the demons, the pressure would 
be the same on both sides of the partition, and if the 
partition were then conceived to be away, no alteration would 
take place until the molecules had had time to diffuse among 
one another. Meanwhile, without any change in the total 
