220 On the Existence of a Hoot for any Equation, 



coating of oxide, was gained, on the other hand, by the re- 

 duction of the alkaline stratum; and the consequence was a 

 singular steadiness of action, the needle continuing for an 

 hour within a degree or two of its maximum deflection. 



This durability of action offers the means of a new class of 

 voltaic researches, on which I am not yet prepared to make 

 any report. 



23. From multiplied experiments of the kind above quoted, 

 it may be inferred that the metals partake of the electrical 

 character of the liquids in contact with them ; their electric 

 condition being exalted if that liquid is similar, and depressed 

 if it is of the opposite kind. And hence the fact observed by 

 Morichini, that an addition to the quantity of copper increases 

 the power of a voltaic pair, the charge being always nega- 

 tive, and therefore homo-electric with copper. 



24. Whatever be the nature of electricity, it would seem to 

 be connected with material particles by something analogous 

 to affinity; inducing bodies which are naturally positive, to 

 withdraw positive electricity from those which are naturally 

 negative, when brought into mutual contact. In what way 

 they do this I am unable to conceive in a manner consistent 

 at once with the phenomena, with chemical analogy, and with 

 probability ; nor are experimental indications very easily 

 found. I may perhaps venture the surmise, that repulsion is 

 the stimulative, attraction the suppressive, principle of voltaic 

 agency. 



[To be continued.] 



XXXV. Further Demonstration of the Existence of a real 

 or imaginary Hoot for any proposed Equation. By R. 

 Murphy, Esq., M.A., Fellow of Caius College, Cambridge. 



To the Editors of the Philosophical Magazine and Journal. 



Gentlemen, 



TN your Number for January, you favoured me with the 

 * insertion of a simple demonstration relative to the exist- 

 ence of a real or imaginary root for any proposed equation. 

 I beg leave to reproduce that proof in a more distinct point 

 of view in the present Number. 



When the equation f (x) = is of odd dimensions, it is 

 known from the simplest principles that there exists a real 

 root. 



When the function f (x) is of even dimensions, put 

 p 4. q a/ _ i for x ; where p and q are real quantities, the re- 

 sult R will evidently be of the form P + Q V - 1. Where 



