113 APPEKBIX. 



a reddish -yellow colour, non- crystalline, and frothing considerably 

 with welter. In warm water it dissolved, forming a clear solution, 

 which became turbid on cooling. An attempt was made to decolourise 

 the aqueous solution by agitation with purified animal charcoal, 

 but very little colouring a^atter was thus removed. As neutral salts, 

 as NaCl, MgSO*, gave a white curdy precipitate from the aqueous 

 solution of the extract, an attempt was made to separate the 

 saponin -like principle by saturating the w^at#ry solution with 

 MgSO* ; it was found, however, that the flocks agglutinated together, 

 forming a sticky mass, and filtration was impossible. Baryta water 

 was next used for separating the principle. With this object the 

 amylic alcohol extract was dissolved iu water and excess of aqueous 

 barium hydrate added. The turbid mixture was then filtered (filtrate 

 A), the precipitate was washed with baiyta water and transferred to 

 a beaker, water added, and CO^ passed for a considerable time. The 

 turbid mixture was then evaporated to dryness on a water-bath, 

 and exhausted with rectified spirit, the filtered alcoholic solution 

 was evaporated to dryness, and left a scaly, friable, shining residue, 

 w^hich afforded the following reactions: — With concentrated H'SO* 

 a yellow coloration, changing to red. Concentrated HNO^ yellow. 

 In concentrated HOI it dissolved freely, forming a faint pinkish 

 coloured solution, the colour deepening on the application of heat, and 

 a few flocks separating. In strong acetic acid it was also readily 

 soluble, forming a colourless solution, no change being induced by the 

 subsequent addition of potassic dichromate. When heated with 

 aqueous phosphoric acid it did not yield a clear solution, no colour 

 developed, and no odour. With acj[ueous ammonia it was sparingly 

 soluble ; no precipitate with acetic acid ; the ammoniacal ■ solution 

 frothed on agitation. Boiled with dilute HCI, it afforded a solution 

 which reduced alkaline copper. The amount of principle precipitated 

 by baryta was small, and though this principle afforded some of the 

 reactions of saponin, it seemed probable that the greater part was 

 still present in the filtrate. 



A fresh portion of the original amylic alcohol extract was 

 dissolved in water, and treated with lead acetate, which afforded a 

 white curdy precipitate, after separation of lead by H^S, yielded 



with 



some 



of the reactions of saponin. Tlie amount of extractive yielded . 

 was, however, small, and it appeared to us that probably both the 



