Mr. G. J. Stoney on Crookes's Radiometer. 305 



tism, like the residual, increase with each successive impulse, 

 remain constant, or diminish? 



In order to contribute to a correct judgment on the degree 

 of accuracy which an answer to such questions concerning the 

 temporary magnetism can possess, I would remark that the 

 latter, as is well known, may fluctuate, always within certain 

 but, with soft iron, tolerably wide limits, even when the cur- 

 rent is constant. 



Nevertheless, on the ground of numerous experiments, I 

 believe I may assert that The temporary magnetism of a steel 

 bar diminishes on the bar being repeatedly magnetized by a con- 

 stant current, but so that the total magnetism remains unaltered ; 

 therefore what is gained in residual, is lost in temporary, mag- 

 netism. 



The second case has been already treated in § 11. We 

 there had the same ellipsoid in different magnetic states, with 

 different RM, submitted to the same force ; and we inquired 

 how the magnetization-function behaves with an increasing 

 permanent magnetization. To this question we could make 

 no definite answer ; k, and with it the temporary magnetism, 

 in part remained constant, in part increased, and in part even 

 diminished. 



From this it is now evident that an answer to the third case, 

 given by the last preceding section, cannot but be extremely 

 difficult. In fact, on making some experiments, I obtained 

 contradictory results, sometimes greater, sometimes smaller 

 temporary moments, while the residual magnetism, in accord- 

 ance with the proposition advanced in § 14, always proved 

 greater. 



Gottingen, September 1875. 



XXXV. On Crookes's Radiometer. — Part II.* By G. John- 

 stone Stoney, M.A., P.R.S., Secretary to the Queen's 

 University in Ireland. 



To the Editors of the Philosophical Magazine and Journal. 

 GENTLEMEN Hampstead, March 18, 1876. 



MANY persons have told me that they have found it diffi- 

 cult to understand the explanation that I offered in the 

 March Number of the Philosophical Magazine, of the pres- 

 sure which Mr. Crookes has discovered. I have therefore en- 

 deavoured to present a picture of the mechanism by which I 

 conceive the pressure to be produced, in a form which will, I 



* Being the substance of part of a communication made to the Physical 

 Society, March 11, 1876. 



Phil. Mag. S. 5. Vol. 1. No. 4. April 1876. Y 



