204 MK. i\ DAT ON AMPHIBIOUS AND 



which normally rose to the surface, apparently for the purpose 

 of respiring atmospheric air direct, were unaffected. 



If, as has been and still is maintained, these accessory organs 

 or sacs are for the purpose of retaining water to moisten the gills* 

 whilst these fishes are wandering on the land, how comes it that 

 when placed in a situation where such, a requirement is not 

 needed (as in a globe of water under a net) they die ? On 

 the other hand, I have kept these fishes hours, have seen them 

 kept alive days, with merely the addition of a sprinkling now 

 and then of v.ater. In this moist condition they are lively, 

 and the earthen pot in which they are contained has nume- 

 rous mucus-covered bubbles present and more being constantly 

 formed by the respiration of these fishes. 



From the foregoing I came to the conclusion -.—first, that in 

 the fishes which died when deprived of direct access to atmospheric 

 air death was not occasioned by any deleterious properties either 

 in the water or apparatus employed, but simply because they were 

 unable to sufiiciently decarbonize their blood on the air present 

 solely in the water ; secondly, that they were able to live in mois- 

 ture (out of water) for lengthened periods ; lastly, that the cavity 

 or receptacle admitted to exist in such fishes (as are here termed 

 amphibious) is not for the purpose of retaining water for moist- 

 ening the gills, but that its walls have a highly vascular covering, 

 and air is passed there for the purpose of respiration, whilst it 

 seems probable that the air, having been so employed, is ejected 

 by the mouth and not swallowed. 



I had intended forming an artificial opening through the muscles 

 of the back into the respiratory air-sac of the Saccolranclius fos- 

 silis ; but injuries caused by its pectoral spine assume such a phleg- 

 monous character that I desisted, after one of my native servants 

 had nearly lost his arm from a wound inflicted by one of those 

 fishes, which the fishermen had previously refused to handle. 



However, in looking over a paper by Dr. Taylor (' Gleanings in 

 Science,' 1830, p. 170), I found the following observations respect- 

 ing this fish : — " If a quill, open at both ends, be introduced 

 through an incision at the side of the spine into one of these 

 canals (air-sacs), bubbles of air will arise from it " — thus show- 

 ing that this cavity is for the reception of air, not of water. 

 Likewise, amongst the Ophiocephalidse, if they are held firmly 

 under water, then tvu'ned over on their backs and the gills opened, 

 bubbles of air arise. 



* If the braiichiie become dry, respiration is prechided. 



