306 



Sulphate of iron. — It often produces, with tannic acids, grada- 

 tions of colour different from those obtained by the chloride of 

 iron, and has to be used with the same precautions. 



Glue serves to confirm the presence of the true tannic acids, by- 

 producing a dirty, flocky precipitate. Some substances, generally 

 accepted as iron-greening tannic acids, produce no turbidness 

 with glue. They are, therefore, either no true tannic acids, or 

 such as yield with glue compounds soluble in water. 



Tartarated antimony. — It precipitates some only of the tannic 

 acids, and serves therefore to distinguish such as differ in this 

 respect. 



Acetate of lead. — It produces almost always a flocky, moi*e or less 

 considerable precipitate, which is sometimes light, sometimes dark, 

 but in most cases of a dix'ty brownish or earth-colour, while the 

 liquid becomes light and eventually clear as water. The precipitate 

 may contain, besides traces of phosphoric and sulphuric acids, the 

 oxalic and tannic acids of the ethereous extracts, small poi*tions of 

 resin and of acids yielding insoluble or sparingly soluble compounds 

 with lead. No sulphate, phosphate, or oxalate was present in the 

 precipitate if it dissolves completely in acetic acid. 



Alkaline copper-sohition (sodio-cupric tartarate). — This test re- 

 quires an alkaline or at least neutral condition of the sample under 

 trial ; an acid condition must therefore be previously removed by 

 one or two drops of potash or soda-leys. After adding one to two 

 drops of the reagent, heat to the boiling-point ; a yellowish and 

 afterwards red turbidness of siib-oxyd of copper indicates sugar; 

 but as other matters (tannic, gallic acid, &c.), have a similar 

 reducing power, this experiment is not decisive for the presence of 

 sugar, unless coi'roborated by a sweet taste. But as the latter is 

 not easily perceived in the presence of other matters which might 

 have passed into the ethereous extract, and even less so with very 

 small quantities of it, the presence of sugar cannot here be decided 

 on, and must be referred to experiment in the course of the 

 analysis. Should more than mere traces of sugar be contained in 

 the svibstance under trial, most of it will pass into the alcoholic 

 extract (III.) 



Tannic acid. — It precipitates most of the alkaloids, but also 

 many indifferent bodies ; it therefore serves only to give collateral 

 evidence for the presence of the one or other of these bodies. 



After the application of the above tests has j^roved insufficient 

 for indicating the presence of an alkaloid, try again small quantities 

 of the liquid with the special alkaloid-tests, mentioned under B, a 

 (page 301), such as yield sparingly or not at all soluble precipitates 

 with alkaloids. If the result with all of them be positive, the pre- 



