PELAGIC AND BENTHIC FISHES, SWIMBLADDER, ECONOMY OF DEEP-SEA LIFE 99 



The evidence suggests that the swimbladder is perfectly functional in benthic species ranging 

 beyond a depth of 2000 m. Why has this organ been lost in the deeper-dwelling pelagic fishes? 



There are many differences between fishes living in the upper and lower reaches of the pelagic 

 environment. Clearly, the most relevant contrasts may be made within genera consisting of species 

 that inhabit both regions. Gonostoma is one such genus, and here G. denudatum will be contrasted 

 with G. bathyphilum. 



The populations of the first species are centred above 1000 m., while those of the second are con- 

 centrated well below this level. Certain differences between these species have already been con- 

 sidered (Marshall, 1954, 1955; Denton and Marshall, 1958). 



Gonostoma denudatum has a capacious, gas-filled swimbladder, but there is no trace of this organ 

 in bathyphilum (p. 13). Radiographs of the two fishes show the skeleton of the first species to be 

 more heavily ossified than that of the second. Furthermore, the flanks of denudatum are sheathed 

 with well-formed scales, while the skin of bathyphilum is naked. Correlated with the firmer skeleton, 

 the myotomes of denudatum are more compact and extensive than those of bathyphilum. This is 

 particularly evident along the post-anal part of the body, which is slimmer and less robust in bathy- 

 philum (see Text-fig. 42). 



By comparison with bathyphilum, the eyes and light organs of denudatum are highly developed. In 

 fishes of the same size, the diameter of the eye in the former is rather less than half that of the latter, 

 and there is a striking difference in the relative development of the dioptric parts (Marshall, 1954). 

 Reference to Text-fig. 42 will show the extraordinary contrast in the size of the photophores. 



Having proportionately larger eyes, it is not surprising that denudatum has larger optic lobes. But 

 this difference in emphasis extends to all parts of the brain (see Text-fig. 42). Particular attention may 

 be drawn to the corpus cerebellum, the volume of which in denudatum is about double that of bathy- 

 philum (Text-fig. 42). Now there seems to be a direct correlation between the degree of development 

 of this nervous centre and locomotor activity (see Healey, 1957; Kurepina and Pavolovsky, 1946), 

 which would suggest that G. denudatum is the more active species. As we have seen, the myotomes 

 of this species are better developed. 



Parallel differences are also found in the gill-system. In G. denudatum the four gill-arches bear 

 filaments along their entire length. In the specimen drawn in Text-fig. 42 the average length of the 

 filaments on the second arch is about 2 mm., and there are some six filaments to each millimetre of 

 the arch. The filaments bear about 60 lamellae per millimetre. 



Conditions are very different in G. bathyphilum. On the first gill-arch the filaments are confined to 

 the forward part of the lower arm, while on the other arches they occur along the entire extent of the 

 lower arms. The filaments are short and feathery (about 1 mm. long and three or four per millimetre 

 along the second arch). The number of lamellae on a i-mm. filament is about 20-25. While there is 

 the possibility that the skin of G. bathyphilum, being scaleless, might assist in respiration, there is 

 striking indication that the tissues are much regressed by comparison with those of G. denudatum. 



Finally, this economy of living substance in bathyphilum is also revealed in the size of the kidneys, 

 which in Gonostoma consist of paired anterior parts joined by thin bridges of tissue to an unpaired 

 posterior part. Comparison of a G. bathyphilum (standard length 77-5 mm.) with a G. denudatum 

 (standard length 81-5 mm.) showed that the kidneys of the latter have a more compact appearance 

 and are much larger than those of the latter. (The anterior parts in both species have much the same 

 length (7-0 mm.) but those of denudatum are more voluminous. The posterior parts of the kidneys of 

 denudatum are larger by a factor of about three.) 



Now in teleosts the simpler products of catabolism (ammonia and urea) tend to be excreted by the 

 gills, while the more complex products (creatine, creatinine and uric acid) are dealt with by the 



13-2 



