From George Gabriel Stokes   6 Jan. 1862

Lensfield Cottage | Cambridge | 6 Jan. 1862

Dear Prof^r Tyndall,

Some of Graham’s results,¹ especially the great difference in the rate of transpiration he obtained with a long capillary of tapering bore discharging into a vacuum ascending as the narrow or wide end of the bore was foremost, (P.T. for 1846 p. 598)² lead me to think that the resistance in his tubes was not sensibly proportional to the first power of the velocity (mean velocity throughout the section at that point) but that a part depending on the square of the velocity was also very sensible, though the former was predominant. Now if the observed results were deemed to be produced by the first power of the vel[ocit]^y by an unknown function µ of the density, the really existing term depending on the square of the vel[ocit]^y being most important at the end of the tube near the receiver, where the density is less and the vel[ocit]^y greater, would cause us to exaggerate the value of µ at a low density.

Til more experiments shall have been made, directed to this particular end, I don’t think we can say anything as to the variation of µ with ρ.

Yours very truly | G.G. Stokes

RI MS JT/1/TYP/4/1384