PHYSIOLOGICAL ACTIONS OF DRUGS ON THE SECRETION OF BILE. 141 
action—if any—on this portion of the intestine. It must of course be borne in 
mind, that when a drug is placed in the duodenum directly, and a certain effect 
on the liver ensues, it by no means follows that the same effect will accrue, if 
the drug be placed in the stomach and thus come in contact with the gastric 
juice. But the general harmony of the results of our injecting substances into 
the duodenum, with those observed in man when the drugs are taken by the 
mouth, convinces us that our method is reliable. In only one instance indeed— 
that of calomel—did it seem probable that its having escaped the influence of 
the gastric juice was vitiating the result, for the hydrochloric acid of the juice 
can convert calomel into corrosive sublimate, and we have discovered that while 
calomel does not, corrosive sublimate does stimulate the liver. A discussion of 
that case will be found under the action of mercury, and we think it the only 
one that needs special consideration. 
SECRETION OF BILE IN A CURARISED FAsTine Doa. 
It was of course necessary—as a preliminary step—to observe the amount 
of bile secreted in the course of a day by a dog that had fasted about eighteen 
hours, and to which nothing but curara was administered. The solution of 
“curara employed in all the experiments was a filtered aqueous solution, every 
minim of which contained one milligramme of the poison. The solution was 
always injected into the jugular vein. 
In all the woodcuts the numbers under the abscissa indicate the hours during 
which the secretion of bile was observed, while those to the left of the ordinate 
indicate in cubic centimetres the amount of bile which flowed from the cannula; 
the dots in the curve indicate the quantities of bile collected every quarter of 
an hour. The vertical dotted lines that cross the curves in the illustrations 
indicate that something was given to the animal. In all such experiments the 
amount of bile first collected is usually considerably larger than that at 
subsequent periods. This apparently results from the sudden diminution in 
the resistance to the exit of the bile consequent upon opening the duct. The 
first one or two collections 
are therefore not reliable in- 
dices of secretion, and they 
are consequently omitted 
from some of the charts. 
Experiment 1. Dog that 
had fasted eighteen hours. _ : : 
Ps é Fig. 1.—Secretion of bile by a fasting dog with nothing but curara ad- 
Weight 76 kilogrammes. ministered. a, 20 mill. ; , 2 mill.; cand d, 4 mill; ef, g, 3 mill. 
—Tw enty milligrammes of curara injected into jugular vein. 
curara were injected into jugular vein (at a, fig. 1). The abdomen was then 
VOL. XXIX. PART I. 20 
al z 3 4 5 6 7 8 hours 
