25 
was contributed by Barger and Shaw° in 1904. Their investigations 
were carried out upon tinctures of digitalis prepared according to 
the British Pharmacopoeia and obtained from various commercial 
sources. For the estimation of the digitoxin content they used 
Keller’s method with slight modifications. For the biological experi- 
ments male frogs (Rana temporaria), weighing between 20 and 35 
grams, were used almost exclusively, but comparative experiments 
were limited to those of nearly the same weight. The experiments 
were carried out in the early summer. 
The tinctures were modified by evaporating a known amount over 
a hot-water bath and suspending the residue in hot water. The 
digitoxin employed in some of these experiments was dissolved in 
alcohol and later diluted with water so as to give it an alcoholic 
strength of 10 per cent, the digitoxin being injected in suspension. 
In all cases the drugs were injected into the dorsal lymph sac, the 
volume for decisive experiments being 1 c.c. The animals after injec- 
tion were kept in a moist atmosphere and the time noted when the 
heart stopped, which always occurred within three hours, if at all. 
After an estimation by chemical means of the digitoxin content in 
nine tinctures, they estimated the toxic value of the same tincture 
by the biological method described. The results obtained by the 
latter method show less variation between the tinctures than is 
usually found, as the strongest was only one and one-half times as 
toxic as the weakest. 
They show that, on account of the slight toxic strength of digitalin 
and digitalein, a supposition exists that the toxicity of a tincture runs 
parallel to the digitoxin content (by Keller’s method), but this is not 
correct, as they found that the ratio o‘f toxicity of tinctures to the 
toxicity of the digitoxin content in them was as 1 to 3.5, 5, 3.7, 
5.5, 6.9, 4.1, 4.1, 5, and 4.2. 
The writers then estimated physiologically the toxicity of the 
water soluble active constituents, and this, added to the toxicity of 
the digitoxin, was compared to the total toxicity of the tincture. 
The results show that about 50 per cent of the tincture toxicity was 
still unaccounted for even when both water-soluble and insoluble 
constituents were considered. 
In order to find the cause of this discrepancy they prepared a tinc- 
ture from chaff and hay and added to this 0.04 per cent crystallized 
digitoxin. By biological methods 0.0375 per cent was demonstrated, 
while by Keller’s method only 0.01 per cent was found. The cause of 
the difficulty was shown by further experiments to depend upon the 
fact that much of the digitoxin adheres to the resin which sepa- 
rates out when the alcohol is evaporated off from the tincture, this 
being lost in the chemical method of estimation, but appearing in the 
physiological. 
« Barger and Shaw, Pharm. J. & Tr., London, 1904, XIX, 249. 
