1874.] RESPIRATION OF FRESHWATER FISHES. 313 



The following experiments were undertaken with the view of ex- 

 tending our knowledge in this direction, by determining what other 

 species are compound breathers, and in what degree the direct use 

 of atmospheric air is necessary for their existence. 



The manner of conducting the experiments was similar to that 

 adopted by Mr. Francis Day *, to whose investigations these notes 

 may be regarded as addenda. 



The specimens experimented upon were recently taken aud quite 

 uninjured. The temperature of the water during the period occupied 

 in making the following observations varied very slightly, from about 

 87°to91°Fahr. ' ° J 



Exp. 1. — A specimen of Plotosus canius was placed in a large 

 cylindrical glass jar (of equal diameter throughout) nearly filled with 

 water, and having a metal plate, pierced throughout with large 

 openings, fixed 1 inch beneath the surface of the water. The 

 metal plate was fixed in this position immediately after the fish had 

 been observed to rise to the surface of the water. Twelve minutes 

 afterwards the fish again rose and pushed violently against the plate, 

 but not succeeding in getting to the surface it sank slowly to the 

 bottom of the jar. Having remained thirteen minutes at the bottom 

 rapidly moving the gill- covers, it made a second attempt to reach 

 the surface, and dashed so strongly against the plate as to force it 

 upwards, and so obtained access to the air. On sinking down 

 again, a large number of air-bubbles passed upwards from the gill- 

 openings. 



The fish was then placed for the night in a large earthenware 

 basin, with a specimen of Ophiocephalus punctatus about the same 

 size. Next morning the latter had eaten the greater portion of the 

 barbels of the former, and had torn away the skin of the abdomen. 

 The greater portion of the anal fin was detached from the body and 

 eaten, but the fish seemed as lively as before. Two other fish of 

 the same species had leaped out of another basin, and were found 

 dead on the floor. The Plotosus was again placed in the same glass 

 jar, having a diaphragm of net fastened' 1| inch below the surface of 

 the water. It remained inactive for nearly half an hour, then sought 

 the surface, but being repulsed by the net struggled violently against 

 it, during which time several air-bubbles passed from the gill-openings 

 to the surface of the water. It died in one hour from the time of 

 being first placed in the jar, and in half an hour from first com- 

 mencing its struggles. 



Exp. 2. — Placed four specimens ofMacrones tengara at 10.40 a.m. 

 in the glass jar, the netted diaphragm being fixed as in Exp. 1. In 

 another jar without a diaphragm two specimens were also placed. 

 At 3 p.m. the respiratory action of the fish in the jar with the 

 diaphragm became much excited; at 3.40 the largest individual, 

 which had been for some time swimming about, rose to the net and 

 pushed against it several times; at 4.40 all the fish had almost 



* Loc. tit. p. 278, exp. 1 & 4. 

 Proc. Zool. Soc— 1874, No. XXI. 21 



