507 
of Edinburgh, Session 1870 - 71 . 
Further, it was pointed out, that antagonism between any two sub¬ 
stances, in the sense of the lethal action of the one being preventible 
by the physiological action of the other, had not previously been 
shown to exist by any certain and satisfactory evidence. In the 
various instances where experiment seemed to indicate the existence 
of such an antagonism, sufficient proof was not given that the dose 
of the substance whose action appeared to be antagonised was certainly 
a lethal one. The conflicting opinions and doubts this fallacy has 
given origin to, have induced the author to follow a plan whereby 
it may be completely avoided. 
In the first place, the minimum fatal dose of physostigma for the 
species of animal employed was accurately determined by a number 
of preliminary experiments; so that the weight of the animal being- 
ascertained, it was an easy matter to be certain of the dose that 
could kill it. Then, in those experiments where an animal re¬ 
covered after the administration of a dose of atropia given in com¬ 
bination with a dose of physostigma, equal to or in excess of the 
minimum fatal, it was killed many days afterwards, and when the 
effects of the two substances had completely disappeared, by a dose 
of physostigma, equal to or less than that from which it had pre¬ 
viously recovered. Therefore, when the administration of atropia 
prevented an otherwise fatal dose of physostigma from causing death , 
a perfect demonstration was obtained of the power of atropia to pro¬ 
duce some physiological a.ction or actions that counteracted some other¬ 
wise lethal action or actions of physostigma. 
In the preliminary note referred to, it was suggested that, as 
both atropia and physostigma are capable of producing a number 
of different actions, several of which may not be mutually antago¬ 
nistic, and that, as both substances are capable of producing several 
actions of a similar kind, considerably less potent to cause death than 
those by which their fatal effects are usually induced, it would pro¬ 
bably be found that a region exists where the non-antagonised and 
the similar actions are present in sufficient degrees of activity to be 
themselves able to produce fatal results. This anticipation has 
proved to be correct. A large number of experiments have been 
made, by which the region of the successful antagonism of fatal 
doses of physostigma has been delined witli considerable exactness. 
The smallest and the largest doses of atropia that are able to pre- 
