280 



Dr. J. L. W. Thudichum. 



[Mar. 18, 



When such an indolent bead of potassic phosphate which has given 

 a, feeble potassium line in red, is dipped into syrupy phosphoric acid 

 and again ignited, the continuous spectrum of volatilising phosphoric 

 acid is again at first obtained, the aerated gas flame to the naked eye 

 is enveloped in a greenish-yellow mantle, and in the spectroscope the 

 continuous spectrum of phosphoric acid only, stretching far into the 

 blue, and without a trace of the potassium line in red, is seen. Gradu- 

 ally the bead flows again as a transparent glass, and bhen the red 

 potassium line reappears in the spectroscope though without ever 

 reaching the vivacity of the line produced by volatile salts. If now 

 the bead have added to it a particle of a chloride — for example, zinc 

 chloride — and be again ignited, a lively potassium spectrum will be 

 seen in the spectroscope, i.e., the potassium goes away as chloride. 

 That double decomposition with the separation of the products takes 

 little more time than the volatilisation of a bead of potassic chloride of 

 about the same size. 



Evidently then, if it were intended to estimate from the spectro- 

 scopic signs the concentration of a given potassic solution, it would 

 be necessary to take into account not only the intensity of the 

 potassium line produced during the volatilisation of the solution, but 

 also the length of time during which that line is perceived. If solu- 

 tions equally concentrated of potassic chloride and potassic phosphate 

 are similarly examined before the spectroscope, the phosphatic solu- 

 tion is found to give the feebler band, but to give it for the longest 

 time ; and no spectroscopic observation with regard to such a solution 

 can pretend to any validity for purposes of quantitative comparison 

 unless the observer be equally attentive to the duration as to the in- 

 tensity of the phenomenon. 



Referring now, in the light of these considerations, to Professor 

 Roscoe's letter, I wish first of all to point out that his observation, 

 when made with the carbonised mass from 1 grm. of substance 

 corroborated my assertion that the substance contains potassium. 

 And I may add, by the way, that in- my paper on which Dr. Gamgee 

 comments to the Royal Society, I have given sixteen quantitative 

 analyses of " protagon " and its congeners with regard to the quantity 

 of potassium contained in each ; having made each determination of 

 potassium by ordinary gravimetric methods, and having weighed the 

 potassium as platinochloride. 



Secondly, however, Professor Roscoe estimates that the quantity 

 of potash contained in a gram of Dr. Gamgee's " protagon twice 

 recrystallised " did not exceed %\ mgrm., an estimate which he 

 appears to found on the faintness of the potassium line (a.) obtained 

 by him from that substance as compared " with a dilute solution of a 

 potassium salt." This inference I think it necessary to check by the 

 considerations I have before stated. Professor Roscoe states that he 



