22 KANSAS ACADEMY OF SCIENCE. 



The residual powder, after treatment with alcohol, was treated with other solv- 

 ents, according to the method of Dragendorff, with the view of ascertaining whether 

 any alkaloid was left behind in the drug after treating with alcohol. The result was 

 negative. 



Inferring from these experiments that alkaloid was present in the drug, that it 

 could be extracted by an alcoholic solvent, an alcoholic fluid extract was then made, 

 each cubic centimeter of the preparation representing one gram of the drug. This 

 was made in quite a large quantity for the purpose of assay. 



Assay. — 10 c. c. of the fluid extract was put into a separator, and a few drops of 

 solution of ammonia added; to this 10 c. c. of chloroform were added, agitated with 

 10 c. c. of water, and again agitated. Thechloroformic layer, which was easily sepa- 

 rated, was drawn off into a second separator, containing 5 c. c. of water to which had 

 been added a few drops of dilute sulphuric acid, agitated, and when the chloro- 

 formic layer had separated, it was drawn off into a third separator, containing .5 c. c. 

 of acidulated water, again agitated, and allowed to stand, which allowed the chloro- 

 formic layer to separate, which was then drawn off. To the contents of the first' 

 meanwhile, were added 10 c. c. of fresh chloroform, which was passed into the sev- 

 eral percolators as before. The acidulated aqueous fluids were mixed and rendered 

 alkaline with ammonia, thus separating the alkaloid in a tolerably pure state. The 

 neutralized fluid was washed with three successive portions of chloroform: 10, 10, 

 and 10 c. c, respectively. The chloroformic solution evaporated. The somewhat crys- 

 talline residue was dissolved in acidulated water, and the solution divided into two 

 portions, each portion representing 5 c. c. of the fluid extract. This was triturated 

 with solution of potassio-mercuric iodide (Mayer's reagent), NI^q, which required 

 .7 CO. of this reagent. According to Dragendorff, 1 c. c. of Mayer's reagent, NIj,,, 

 precipitates .00625 of atropine. Estimating the amount of this unknown alkaloid 

 from the above data, we have .00625 multiplied by .7 = .004375, which, multiplied by 

 20 to obtain the percentage, gives us .087 per cent. The 5 c. c, corresponding to the 

 one just examined, was tested gravimetrically, by adding Mayer's reagent in excess 

 and weighing the precipitate. The precipitate, when dried, weighed .015, which' 

 when multiplied by .449, (see Lyon's Pharmaceutical Assaying, page 56,) yielded 

 .094 per cent. 



For the purpose of confirmation, still another process of assay was tried, a proc- 

 ess recently introduced by Prof. J. U. Lloyd, of Cincinnati: 5 c. c. of the fluid ex- 

 tract were placed in a porcelain mortar, with 1 c. c. of solution of perchloride of iron 

 and bicarbonate of soda added, until a stiff magma was produced; then this magma 

 was extracted by repeated trituration of chloroform. This excludes tannates, gums' 

 albuminates, mineral salts, most vegetable acids and salts, and most coloring mat- 

 ters. Chloroform extracts the chloroform; the alkaloids mixed with the wax, some 

 resins, chlorophyll, and fats. The alkaloid chloroform was extracted by rotating in 

 a bulb separator with repeated portions of dilute sulphuric acid (one part in 50 of 

 water). The aqueous acid solution was extracted in a separator by stronger ether 

 in repeated portions, the ether separating the chlorophyll. The acid solution was 

 neutralized with ammonia and rotated in a bulb separator with chloroform, thus ab- 

 stracting the alkaloid in a tolerably pure state. The chloroformic solution was 

 evaporated in a tarred dish and the residue weighed. This residue weighed .058' 

 which, multiplied by 20= l.KJO, the percentage of alkaloid, estimated by this so-called 

 easy process. 



In presenting the foregoing, the authors are conscious of the fact that any chem- 

 ical assay is open to numerous objections unless it places before us, in a definite, 

 crystallized form, the active principle that we seek to estimate, and this supple- 



