( 2G ) 



in tlic water, is especially visible in some species, as the 

 Polj/acanthas cupanus, which dart up suddenly to the 

 siu'face, descending again as rapidly to the depths of the 

 water. I instituted a considerable number of experiments 

 (see Proc, Zoological Society of London, May lith, IHUS, 

 ji. 27'i)) to investigate this question. Some live specimens of 

 Ophlocephalus gacJiua were placed in a globe, which was 

 filled two-thirds full of fresh water. A diaphragm of line 

 net was then stretched lightly across the inside of this globe, 

 one inch below the surface of the water, thus elfectually pre- 

 venting them from ascending to the surface to obtain a direct 

 supply of atmospheric air ; death invariably ensued in a 

 longer or shorter time, geiicrally in accordance with whether 

 they remained quiet or conlinucd excited. A bandage stitched 

 tightly around the gill openings, whilst it prevented their 

 being iised for respiratory purposes, did not appear to cause 

 any inconvenience so long as they could inhale atmospheric air 

 direct, and this although it was not removed for 21 hours. 

 But it must not be considered that these fish are entirely 

 prevented from decarbonising their blood if they are unable 

 to obtain atmospheric air direct, as, although some died within 

 the first 40 minutes, others lived 7 and one 17 hours whilst 

 below the diaphragm. In wet grass, at the end of 3 hours, 

 those placed there were found as lively as -when first 

 put there : one in a dry cloth lived for 3 hours and 25 

 minutes. The Anahas scandcns are kept four or five days 

 alive by the fishermen in Calcutta in earthen pots destitute 

 of water, \ising daily what they require, the fish continuing 

 as lively as when captured. In Burma the fishermen 

 appear to be practically acquainted with the fact of some fish, 

 especially Ophiocephal'uhe, being air-breathers ; thus, after 

 nearly all the water has been removed from the tank to be 

 fished, leaving only about five feet of slimy mud, through 

 which their baml)oo net (gijaii) has been drawn, they are 

 aware that many fine fish still remain. A large cloth is 

 spread over tiic mud and left there two or three days, on remov- 

 ing which, the fish are seen stupefied and easily taken, their 

 blood having become carbonised from a deficiency of oxygen, 

 due to want of air for breathing. In short, my experiments, 

 I think, proved that these fishes died when deprived of access 

 to atmospheric air, not from any deleterious i)roperties either 

 in the water or apparatus employed, but due to being unable 

 to decarbonise their blood solely from the water, serial respi- 

 ration being indispensable. Also that they can live out 



