MOUNTAIN LAUREL, A POISONOUS PLANT. 31 



body, anclromedorubrin. This andromeclotoxin gave no precipitates 

 with alkaloidal reagents and did not reduce Fehling's solution, but 

 gave certain color reactions and produced distinct and characteristic 

 symptoms on injection into guinea pigs and frogs — severe retching, 

 which, however, is absent after large doses; disturbances of respira- 

 tion; convulsions, and paralysis. 



The characteristic color reactions are as follows : 



(1) Concentrated sulphuric acid gives a dark reddish brown color 

 and dissolves it gradually. On warming it becomes a deeper red. 

 On dilution with water this passes into a light mulberry-red, disap- 

 pearing with alkalis returning with H 2 S0 4 . 



(2) Evaporation with dilute H 2 S0 4 (1:5) on water bath causes 

 rose-red color. ' If the body is pure there is no odor, but if not pure 

 there is an odor of ericinol, due to the decomposition of ericolin, an 

 attached body. 



(3) Dilute HC1 acts the same as H 2 S0 2 , only it gives a color more 

 nearly violet-red. 



(4) Dilute phosphoric acid, 25 per cent, gives a mulberry-red color. 

 By this means 0.0000005 gram of andromedotoxin can be recognized. 



Especial emphasis is laid on the second and fourth reactions. 

 Plugge also extended his studies to other members of this group. The 

 first of these to be examined was Andromeda polifolia. He made the 

 plausible claim that because the chloroform " shaking " from the aque- 

 ous extract freed from extraneous matter by lead gave the same color 

 reactions and produced the same action in animals as that from 

 Pieris japonica it contained the same active principle, but he admitted 

 that this body was much less active than the andromedotoxin obtained 

 from Pieris japonica. He explained this weakness as being due to 

 the fact that the plant was grown on a different soil and that the 

 active principle varied from, that of Pieris japonica much as the 

 active principle of the various species of Digitalis and Aconitum 

 vary. 5 



In using this method with LeUcothoe catesbaei c Plugge noted that 

 the first six or seven chloroform shakings gave an amorphous residue, 

 while the seventh to the ninth gave a white crystalline body from 

 which he obtained the characteristic physiological action and color 

 reactions of andromedotoxin. 



From C hamaedaphne calyculata he obtained some crystalline de- 

 posit, but in this case from the lack of material was unable to 



a Plugge, P. C. Vorkommen von Andromedotoxin in verschied. Ericaceen. 

 Arch. d. Pharm., vol. 223, p. 906, 1885. 



& Plugge, P. C. Ueber d. Vorkommen d. Andromedotoxin in Andromeda Poli- 

 folia. Arch. d. Pharm., vol. 221, p. 818, 1883. 



c Plugge, P. C. Vorkommen von Andromedotoxin in verschied. • Ericaceen, 

 Arch. d. Pharm., vol. 223, p. 905, 1885. 



32893— Bui. 121— 08- 3 



