312 



T ransactions. 



tions three fish were placed, and they were observed at frequent intervals 

 during the day- In B (control), symptoms began in three hours and three- 

 quarters in a fish which died in twelve hours ; and at eight hours in a fish 

 which died in eight hours and a half. The third fish showed slight symp- 

 toms when last seen at night, twelve hours after the experiment was begun, 

 and was apparently quite well the next day. After thirty hours this fish 

 was removed, the fluid filtered, and three other fish placed in the solution. 

 This second group of fishes was affected as follows : When last seen at 

 night, two hours after the experiment was begun, they seemed unduly 

 excited, and next morning all three were dead — i.e., within sixteen hours 

 of the commencement of the experiment (Exp. 98). (In order to test 

 whether a second use of the solution would influence the result, a control 

 of 1,000 c.c. water that had been used the day before was also filtered, and 

 three fishes placed in it. These fish were quite unaffected at the end of the 

 experiment, so that one may disregard the fact that the control solution 

 mentioned above was used twice.) 



Table VI. 

 Effect of Hydrolysis by 0'2 per Cetit. HCl on the Toxicity of Tutin. 



Now, with regard to the hydrolysed tutin (Exp. 94), to which the foregoing 

 served as control experiments : In the first set of three fishes, one showed 



