172 Prof. Stewart and Mr. W. Stroud 



on a 



with any accuracy, the experiment was quite negative in result. 

 I now, however, took a glass tube, fused the end, and blew out 

 the bubble till it burst. A piece of the exceedingly thin film 

 thus obtained was subjected to the action of the magnet, and 

 most strenuous endeavours were made to detect its action upon 

 the beam of polarized light. This action must have been ex- 

 ceedingly slight, though there is some evidence, which it is 

 not necessary to give, that it was detected. The glass, how- 

 ever, was coated as the first piece had been, and again with its 

 coating subjected to the action of the magnet. The trial was, 

 for certain reasons, rather unsatisfactory ; and although no 

 rotation of the plane of polarization was now detected, I do 

 not think this fact can be taken as evidence that the effect of 

 the nickel had counterbalanced the effect of the glass. Both 

 these experiments on direct transmission have been, we may 

 say, quite negative ; but these details are given as marking 

 out a line of research which will probably be some time 

 resumed. 



An endeavour has also been made to detect a possible rota- 

 tional effect due to reflection from silver when strongly mag- 

 netized*. For this purpose two strips of silver upon glass 

 were used; and these strips were fastened one upon each pole 

 of the magnet, the silvered surfaces being turned toward each 

 other and as nearly parallel as practicable. The poles being 

 brought near together and the light being let in between the 

 silvered surfaces at a large incidence, it was possible to obtain 

 twenty or thirty successive reflections before the beam emerged 

 toward the analyzing Nicol. Certain difficulties were intro- 

 duced by this arrangement; but in spite of these I think that, 

 if the action of silver had been one tenth as strong as that of 

 iron, the effect would have been detected. No such effect was 

 observed. 



XIX. On the Results obtained from a Modification of Bun- 

 sen's Calorimeter. By Prof . B. Stewart and W. Stroud*) - . 



ADESCBXPTION of this instrument was brought before 

 this Society on June 26, 1880, and afterwards ap- 

 peared in the Proceedings of the Physical Society, vol. iv. 

 p. 52, and Philosophical Magazine, vol. x. p. 171. The 

 results obtained at that time were not very good, owing 



* In this experiment I had the very efficient assistance of Mr. Arthur 

 W. Wheeler, Fellow in Physics, whose untimely death is so deeply 

 lamented. 



t Communicated by the Physical Society, having been read at the 

 meeting- on June 26, 1881. 



