CERTAIN ALKALOIDAL PERIODIDES. 267 



These authors show that a large part of the dissolved iodine is in the form 

 of the tri-iodide of potassium, while smaller quantities are present as the penta- 

 iodide. They state: "This result evidently indicates that a small amount of poly- 

 iodide higher than KI3 is present. If we consider it to be only KI5, then we have: 



CSl5=0.080 and C2l3=0.582. 



[The total concentration of the potassium salt present in other forms than 

 potassium iodide is 0.662 mols per liter.] 



"Thus we reach a conclusion similar to Laurie, that a large part of the dissolved 

 iodine in the normal potassium iodide solution saturated ^vith iodine is present 

 as tri-iodide. A similar treatment of the remaining measurements shows that 

 the amount of higher poly-iodide decreases as the potassium iodide concentration 

 is decreased, and becomes negligible in the neighborhood of the 0.1 nonnal 

 solution." 



This work is also significant in connection witli the results obtained 

 by precipitating the alkaloidal periodides with a solution of iodine in 

 potassium iodide in which the concentration of potassium iodide varies 

 greatly. (See Table II.) The decrease in the amount of iodine enter- 

 ing into the precipitate with the increase in concentration of potassium 

 iodide may be due to the solubility of the alkaloidal periodide in potas- 

 sium iodide solution; or to the increased reluctance of the poly-iodide to 

 give up the iodine to the alkaloid. 



Throughout this paper the term free iodine is used. This should 

 be read to mean the iodine present in the form of poly-iodides and 

 reacting with thiosulphate solution. If it is true, as stated by Bray and 

 MacKay, that the hydrolj'tic reaction Ij -f H^O ^ HIO -\- H^ + I- is 

 almost completely driven back in the presence of the dilute acid, and 

 also an iodide, it is evident that it can be ignored in a consideration of 

 the formation of the alkaloidal periodides in this work. 



THE ACTION OF WAGNER's REAGENT WITH SOLUTIONS OF THE FREE 



ALKALOIDS. 



The question of the action of Wagner's reagent upon aqueous solu- 

 tions of the free alkaloids, so far as I am aware, has been very little 

 studied. It is known that certain alkaloids, notably caffeine, are not 

 precipitated under these conditions and Gomberg -* states that others, 

 including moi-phine, strychnine, and atropine, are so precipitated. 



EXPERIMENTAL. 

 MOBPHIXE. 



A number of qualitative tests demonstrated that Wagner's reagent will pre- 

 cipitate morphine from very dilute solutions. 



. "76id. (1898), 18, 335. 



