But the molecular theory of gases was still prominently before his mind.
In 1862, writing[32] to H. R. Droop, he says:—
“Some time ago, when investigating Bernoulli’s theory of gases, I was surprised to find that the internal friction of a gas (if it depends on the collision of particles) should be independent of the density.
“Stokes has been examining Graham’s experiments on the rate of flow of gases through fine tubes, and he finds that the friction, if independent of density, accounts for Graham’s results; but, if taken proportional to density, differs from those results very much. This seems rather a curious result, and an additional phenomenon, explained by the ‘collision of particles’ theory of gases. Still one phenomenon goes against that theory—the relation between specific heat at constant pressure and at constant volume, which is in air = 1·408, while it ought to be 1·333.”
And again[33] in the same year, 21st April, 1862, to Lewis Campbell:—
“Herr Clausius of Zürich, one of the heat philosophers, has been working at the theory of gases being little bodies flying about, and has found some cases in which he and I don’t tally. So I am working it out again. Several experimental results have turned up lately rather confirmatory than otherwise of that theory.
“I hope you enjoy the absence of pupils. I find the division of them into smaller classes is a great help to me and to them; but the total oblivion of them for definite intervals is a necessary condition for doing them justice at the proper time.”
The experiments on the viscosity of gases, which formed the Bakerian Lecture to the Royal Society read on February 8th, 1866, were the outcome of this work. His house in 8, Palace Gardens, Kensington, contained a large garret running the complete length.
“To maintain the proper temperature a large fire was for some days kept up in the room in the midst of very hot weather. Kettles were kept on the fire and large quantities of steam allowed to flow into the room. Mrs. Maxwell acted as stoker, which was very exhausting work when maintained for several consecutive hours. After this the room was kept cool for subsequent experiments by the employment of a considerable amount of ice.”
Next year, May, 1866, was read his paper on the “Dynamical Theory of Gases,” in which errors in his former papers, which had been pointed out by Clausius, were corrected.