Fitzgerald — On Limits to Velocity of Motion of Engines. 161 



March number of the Philosophical Magazine this year, called atten- 

 tion to the way in which the velocity of flow of a gas into a vacuum 

 is limited, and this limits the velocity of motion of the piston in 

 an engine. He has, however, omitted to notice that there is a 

 greater velocity than the velocity of sound with which a gas can 

 move into a vacuum, namely, at the rate at which its particles 

 are moving. This only comes into effect when the space is so 

 small compared with the free path that we cannot deal with the 

 molecules, as making an indefinite number of encounters on their 

 way across the vessel. In the case, for instance, of a piston in 

 a vessel full of a gas moving suddenly from rest, with a velocity 

 equal to that of the average velocity of the molecules of the gas, 

 which is greater than the velocity of sound in the gas, it is 

 evident that all the molecules that were just on the point of 

 striking the piston would follow it up, and that those that 

 happened to be moving normally to it would keep following it 

 up, and so would be diffusing into this vacuum, at a greater rate 

 than the velocity of sound in the gas. This leads to a diffusion 

 velocity of energy in a vacuum small compared with the free 

 path, quite different from the velocity of sound, and upon which 

 evidently radiometer action depend. It is this that would ulti- 

 mately limit the rate at which the piston could be moved by the 

 gas. I have explained this at my lectures on the Theory of 

 Steam Engines for some years back. Steam may also be used 

 kinetically, as in Griff ard's injector, and Hero's engine; and in 

 these cases velocity of motion is limited by the velocity of flow 

 of the steam. 



In the case of most of these engines that transform kinetic 

 energy into work, it is to be remarked that when moving slowly 

 there is a very small power produced at the expense of a great 

 expenditure. For example, in Hero's engine and engines of this 

 type, if the steam runs out freely without moving the engine, 

 there is certainly the maximum pressure tending to move the 

 parts, but no power is produced, even though a great deal of 

 steam is being employed. It is not the same with pressure 

 engines, like ordinary steam engines. They may be worked 

 slowly, and the power produced is proportional to the steam 

 employed. The same distinction holds in the case of water 

 engines working pistons and turbines. In the case of the kinetic 



SCIEN. PEOC. E.D.S. — VOL. V. PT. III. jS T 



