MALACOPTERVGII APODES. 551 



circle which is formed by the two hands, may be very much 

 enlarged without any sensible diminution in the force of the 

 commotion. Twenty-seven persons holding each other by the 

 hands, and composing a chain, the two ends of which were 

 in contact with to two points on the surface of the fish's 

 body, felt a very smart shock simultaneously. 



It depends upon the will of the animal to produce com- 

 motions more or less strong. It is often even necessary that 

 it should be, as it it were, progressively roused and animated. 

 In general the first shocks are more feeble ; they become 

 stronger and stronger, in proportion as the irritation and agi- 

 tation of the animal increases. Finally, they are terrible, as 

 observers remark, when thegymnotus is worked up to a pitch 

 of rage. 



When the fish has thus emitted reiterated shocks around, 

 it appears to be exhausted, and a repose is necessary for the 

 renewal of its power. One would think that this time was 

 employed to recharge its fulminating organs with a fresh 

 supply of the torporific fluid. " In America," says M. de 

 Humboldt, " they take advantage of this interval to catch these 

 fish with but little risk. They force wild horses to enter the 

 ponds inhabited by the electric eels. These unfortunate 

 quadrupeds receive the first discharges ; stunned and over- 

 thrown, they disappear under the water, and the fishermen 

 take possession of their assailants, either with nets or with 

 the harpoon. The combat is finished in about a quarter of 

 an hour." 



The Indians assured M. de Humboldt, that on putting 

 horses two days in succession into a pool full of these electric 

 eels, no horse was killed on the second day. This is another 

 proof of the necessity of repose to enable these fish to accu- 

 mulate a new quantity of the electric fluid. 



A phenomenon extremely worthy of attention, presented 

 by the same fish, is the following : we are assured that some 



