Arthur C. Parlett

“ The holding of a perfect bridge hand is considered a news item because the number of highly shuffled, random, chaotic arrangements are hundreds of millions of times greater than the perfectly ordered hand. Yet the probabilities in this case are based upon a deck of only fifty-two cards. Because of the unimaginable number of molecules involved, Dodge has said: "Any arrangement in which the hot molecules would predominate to permit an observable difference in temperature between any macroscopic portions of the gas is so improbable as to be nonexistent." ”

“ Air is certainly a well mixed substance, and the number of molecules in a unit volume is fantastically large. A tremendous aggregation of gaseous molecules in completely random movement simply must arrive at the most probable configuration of complete randomness in short order. Yet the device Monsieur Ranque patented in 1934 [1] apparently effects a separation of hot molecules from cold with consequent loss of entropy from the system as a whole. ”

“ This loss of capacity for spontaneous change is measured by its gain in entropy, indeed can serve as a basic definition of entropy. It will be noted that no loss of energy is involved; only the degree of availability of the energy to do useful work has been lessened. ”

“ Friction with the inner surface of the tube heats up the gas next it. That is, gas drawn off at the axis has been cooled while that drawn off at the periphery near the outer wall has been warmed. Adiabatic expansion of air from ten atmospheres and room temperature to one atmosphere will cool the air by about 135°C., so there is plenty of available cooling. [5]
And thus the charming picture of the busy little demon sitting inside the tube with a tennis racquet in his hand, batting the molecules to right and to left as they come in, saying to himself, "There's a hot one, smack it to the right! ”

“ The action is strikingly similar to that of the demon first conceived by Clerk Maxwell.
Clerk Maxwell, remarkable physicist of the last century, recognized the statistical nature of temperature. In a gas of a certain observed temperature, a large percentage of molecules will have a kinetic energy content greatly higher than that corresponding to the thermometer reading, and a large percentage will have a much lower kinetic energy content. The temperature reading is merely the average value. ”