7 
6 
^Mayer's reagent), as well as those methods which employ varions 
modifications of Wagner’s reagent (a solution of iodine in potassium 
iodide ) as the precipitant, chief among which we may place Kippen- 
herger’s® process. To class (2) we may assign all the acidimetric or 
alkalimetric processes, including Gordiu’s ^ method. 
^Wien we compare the merits of the methods of class (1) with those 
of the acidimetric or alkalimetric processes iucluded under (2), we find 
them to be largely in favor of the latter. For not only do the former 
lack the advantage of bemg based upon a more or less general charac- 
teristic property of the alkaloids, namely the alkahne nature of most 
alkaloids toward certain indicators, as is the case with the latter, 
but they also possess serious inherent defects. Thus Linde pomts 
out among other things, that the precipitate formed with Mayer’s 
reagent has no definite composition, that an aUowance has to be made 
for its solubfiity. and that the end poiut of the reaction is affected by 
the concentration of the alkaloid in the solution titrated: while ac- 
cording to Scholtz"^ the method of Kippenberger is so untrustworthy 
as to be entirely inapphcable. 
The acidimetric or alkalimetric methods are thus left to us as the 
only ones wherein we may look for a satisfactory volumetric method 
for the estimation of the vegetable alkaloids. With more or less success 
such methods have been worked out for most of the more common 
alkaloids through the employment of a mmiber of various indicators. 
Thus, according to Kuppenberger.^ satisfactory results may be ob- 
tained, with lacmoid as indicator, in titrations of atropine, mori^hine, 
veratrine. codeine, cocaine, nicotine, coniine, papaverine, or nar- 
cotine: with iodeosin as indicator, in titrations of thebaine, codeine, 
emetine, or coniine: with cochineal as indicator, in titrations of 
coniine, morphine, thebaine. emetine, brucine, or peUetierine: with 
azolitmin as indicator, in titrations of aconitine, stryclmine, or qui- 
nine: while in the U. S. Phcirmacopoeia (1905) hcnmatoxyhn or 
iodeosin are recommended for use in various alkaloidal assays. 
A closer examination of the literature, however, shows that there 
exists considerable difference of opinion, and even contradiction, as 
to the suitability of the various indicators which have been proposed 
for use in the estimation of the vegetable alkaloids. Thus Kippen- 
berger’s statement cited above, in which he recommends azohtmin 
in titrations of quinine, is contradicted by Ranmistedt,^" who denies 
the applicability of azolitmin as indicator in quinine titrations: the 
a Zeit. anal. Chem., 35, 10-27 and 422-471 (1^96;. 
&Ber., 32, 2S71-2S76 (1899 ). 
- eArcli. Pharm., 237, 172-185 (1899;>. 
Arch. Pharm., 237, 71-80 (1899). 
«Zeit. anal. Chem., 39, 201-229 Q900). 
/ Apoth. Zeit., 22, 1117 (1907). 
