CROOKES' RADIOMETER 81 



auspices of the British Association ; and Tait sent in two short Reports in 

 1869 and 1871. Most of the ' veteran ' students had a turn at the bars during 

 the seventies and eighties ; and Tail's paper on the application of Angstrom's 

 method of sending waves of heat along the bar (Proc. R. S. E, Vol. vni), 

 was based on observations made by A. L. MacLeish and C. E. Greig 1 in the 

 early part of the year 1873. The harmonic analysis is fully worked out so as 

 to give the amplitudes and phases of the temperature oscillations at each 

 chosen point ; but the final calculation of the conductivities is not given. In 

 fact the simple and solvable form of the equation of conduction did not apply 

 even to a rough approximation. Tait therefore fell back upon Forbes' method, 

 and in 1878 he published a detailed account of his investigations, the main 

 purpose of which was to extend to other metals what Forbes had done for iron. 

 An important supplement to this memoir appeared in 1887 by (Professor) 

 Crichton Mitchell, who as an advanced student went over the whole ground 

 again, the one difference being that all the bars were now nickel plated. 

 Their surface conditions were thus rendered more nearly identical than in the 

 first set of experiments. One of the final conclusions come to was that 



"We cannot yet state positively that there is any metal whose conductivity 

 becomes less as its temperature rises ; and thus the long sought analogy between 

 thermal and electric conductivity is not likely to be realised." 



Early in 1875 Tait and Dewar made together a series of well planned 

 experiments on the phenomena of Crookes' radiometer. They gave a 

 demonstration of these before the Royal Society of Edinburgh on July 5, 1875 ; 

 but unfortunately no authoritative account of them was ever published. In 

 Nature of July 15, 1875, a report of the communication was given under the 

 title " Charcoal Vacua " which does not bring out clearly the real significance 

 of certain parts of Tait and Dewar's investigations. The following quotation 

 from Lord Kelvin's obituary notice read before the Royal Society of Edinburgh 

 puts the question in a clearer light : 



" In a communication on ' Charcoal Vacua ' to the Royal Society of Edinburgh 

 of July 5, 1875, imperfectly reported in Nature of July 15 of that year, the true 

 dynamical explanation of one of the most interesting and suggestive of all the 

 scientific wonders of the nineteenth century, Crookes' radiometer, was clearly given. 

 The phenomenon to be explained is that in highly rarefied air a disc of pith or 

 cork or other substance of small thermal conductivity, blackened on one side, and 

 illuminated by light on all sides, even the cool light of a wholly clouded sky, 



1 Dr A. L. MacLeish is now a physician resident in Los Angeles : the Rev. C. E. Greig 

 is a pastor in Paris. 



T. II 



