106 Dr T. Anderson on a Bemarkahle Compound 



digest it for some time at a moderate temperature, after 

 each addition of water and acid. 



y f 5'255 grains of teriodide of codeine gave 

 * 15*590 iodide of silver. 



55-32 



These results correspond closely with the formula Cgg Hg^ 

 NOg + 13, of which the following are the calculated results 

 compared with the mean of analysis : — 



Mean. Calculation. 



10-000 680-3 



The relation of teriodide of codeine to nitrate of silver is 

 of a very remarkable kind. It is instantaneously decom- 

 posed by solution of the nitrate, and this reaction appeared 

 at first sight likely to afford the most accurate and commo- 

 dious method of determining the iodine, a method which has 

 already been applied by Pelletier to the analysis of the com- 

 pound of iodine and strichnine. Experiment, however, proved 

 it to be entirely inapplicable, as the whole of the iodine is not 

 obtained in the state of iodide of silver, but always a much 

 less amount, which also appears to be constant, although not 

 bearing any satisfactory atomic relation to the whole quan- 

 tity. Three different determinations of the iodine were made 

 by this method with the following results : — 



Per cent. 



5*590 grs. teriodide of codeine gave 4*45 grs. iodide of silver = 43-01 



5-304 4-24 =4319 



7-449 5-795 =42-04 



If we suppose three equivalents of teriodide of codeine to 

 yield seven equivalents of iodide of silver, we ought to obtain 

 43*6 per cent, of iodine in this state. Assuming this to be 

 the case, which, however, has not much probability in its 



