From George Gabriel Stokes   7 June 1861

Lensfield Cottage Cambridge | 7 June 1861

My dear D^r Tyndall,

I returned home last night after an absence of about 10 days. I would have written again in answer to your further questions and remarks¹ only that I thought I should have met you at the R.S. and should have been able to speak about it.

You asked me one explicit question which I have not yet answered.

I do not recollect my conversation with you at the Phil Club. I first heard of Foucault’s experiment² from Foucault himself whom I met at dinner at Dr Arnots³ when he came to London to receive the Copley Medal.⁴ I was greatly struck with it. He told me that no one in Paris seemed to think anything of it except Babinet.⁵ Believing the line to be due to soda⁶ I asked him about the poles and he said that they were of carbon and that there had been various salts on the poles. In a conversation shortly afterwards with Thomson I informed him of Foucault’s exper^t. When he asked me how I accounted for the appearance of a bright and dark line in the same place I gave the explanation you will find in the note in the Phil Mag Vol. 19 p. 196⁷ and said I believed it was due to sodium. He asked me if there were any other examples and I mentioned an observation of Sir David Brewster (Rep. Brit. Ass.)⁸ who found bright lines in the spectrum of burning potassium (or rather deflagrating nitre) agreeing exactly with certain dark lines in the red in the solar spectrum. Thomson if I recollect right at once jumped to the conclusion that we should find all or most of the solar lines coincident with bright lines in artificial flames.⁹ I was not prepared to leap so far but thought that many of the non-atmospheric lines might be due to gaseous substances in the solar atmosphere having the selective absorbing power of NO₄, I, Br &c I never published anything on the subject nor did I look on my remark as more than a (to me) obvious inference from a very remarkable discovery due to Foucault.

I agree with you that the coincidence in the n^os for radiation and conduction is too remarkable to be accidental. I am not however satisfied with the explanation¹⁰ but we shall I hope have time to talk about this.

Yours very truly | G.G. Stokes.

RI MS JT/1/S/224

RI MS JT/1/TYP/4/1373–3a