308 Transactions, 



disastrous to the fish, for if discovered within an hour or so, wiped clean,, 

 and returned to the water it would recover completely. Three of these 

 fish will live in 1,000 c.c. of water from thirty-six to forty-eight hours or 

 more, and are apparently quite comfortable, even when the water is much 

 fouled with their own excreta. 



Sollmann states that 0-001 per cent, picrotoxin is fatal to fundulus in 

 five to nine hours. His specimens (5 cm.) were not much smaller than the 

 minnows used in the experiments recorded here. He used only 150 c.c. 

 of water to each fish, and admits that occasionally one of the controls would 

 die. It is probable that the amount of water used by Sollman was too 

 small, for in the experiments made by the present writer to compare the 

 effects of tutin and picrotoxin, it appeared that in the early experiments, 

 where a small quantity of water (200 c.c.) was used, death occurred in four 

 hours when the concentration was 0-001 ; whereas in the later experiments, 

 with three fish in 1,000 c.c. fluid, 0-001 was fatal to only one fish of the 

 three in twenty-four hours. This shows the importance, when using fish 

 in this way, of allowing a sufficient quantity of water. 



Symptoms. — The symptoms shown by the minnows when three were 

 put together in 1,000 c.c. of a lethal percentage (0-005) of tutin were as 

 follows : After two to four hours, with no symptoms, the fish began to 

 swim near the surface of the fluid, and became excited in their movements, 

 swimming vigorously about, and even leaping repeatedly out of the water. 

 At this stage they were frequently observed to emit bubbles of gas from the 

 mouth. This discharge of gas, and the inability to sink, point to a derange- 

 ment of the function of the swim-bladder. Later they lose their power 

 of maintaining the normal position, and swim about near the surface, 

 turned on the side. Now and again they recover the upright position, and 

 swim excitedly about for a minute or two, and then fall back on the 

 side. Later on they lie at an angle with the surface of the fluid, often with 

 the head down, and swim about feebly in this position ; and later still they 

 lie on the side on the bottom of the dish, and the gill-movements are 

 laboured. At intervals they spring up from this position, and swim round 

 on the side, while appearances suggesting convulsions are seen — viz., bend- 

 ing of the trunk, accompanied by shuddering movements. At death it was 

 invariably noticed that the gill-movements ceased before the heart stopped 

 beating. (For protocol of symptoms, see Exp. 108.) 



Table IV shows that under the above-mentioned experimental conditions 

 a percentage of 0-001 was necessary to cause symptoms to appear, but was 

 not necessarily fatal unless other deleterious conditions, such as too small 

 an amount of fluid, were present {e.g., Exps. 77, 78, 79 in Table IV show- 

 fatal results within six hours, but under improved conditions 0"0035 was 

 fatal to only one fish out of three) ; while a percentage of 0-005 caused 

 symptoms within a few hours, and was generally fatal to all three fish 

 within twenty - four hours. Between these limits various results were 

 obtained : thus in 0-0035 per cent, two of the three fish recovered and one 

 died, while in 0-003 per cent, and in 0-004 per cent, two of the three 

 died and one recovered. But in these experiments, Nos. 89, 91, and 95, 

 tin dishes were used, and the water became rusty before the end of the 

 experiments. 



In order to insure a lethal effect, the dose was therefore raised to 0-005 

 per cent., and experiments were made to test the effect of various reagents 

 in increasing or diminishing the toxic power of tutin. The effect of tutin 

 was also compared with that of picrotoxin. 



