SECT. 4] SOUND SCATTERING KV MARINE ORGANISMS 535 



and warm bodies of water in the region of the (lulf Stream. It was not possible 

 to say in these cases, however, that the same animal populations were repre- 

 sented in the two water types. 



The change in ambient temperature experienced by a layer during the course 

 of its diurnal migration may be considerable. A layer studied by Clarke and 

 Backus (1956) in deep water south of New England moved up in the evening 

 from a depth- of 345 m, where the temperature was about 7°C, to near 40 m, 

 w'here the temperature was about 17.5°C, thus experiencing a change of more 

 than lO'^C in a period of about 1| h. If this had an effect on the activity rates 

 of the scatterers, as it should, it was not manifested in rate of the vertical 

 migration, which was, as is usual, virtually uniform throughout the post-sunset 

 period. 



C. Swim-bladder Function and Deep Scattering Layers 



One objection which has been continually raised to the hypothesis that 

 fishes with gas-filled swim-bladders are the principal constituents of deep 

 scattering layers is the difficulty with which the contents of the bladder would 

 be managed during the extensive vertical migration, especially if the hydro- 

 static function of the bladder is to be realized, i.e. if the fish is to maintain 

 neutral buoyancy at both deep and shallow level. During the evening ascent 

 gas must be continually absorbed. (The swim-bladder in these bathj^pelagic 

 fishes is of the closed or physoclistous type ; that is, there is no opening between 

 the bladder and the gut which might allow •the fish to vent gas.) During the 

 morning descent gas must be rapidly generated to fill the shrinking chamber. 

 That such fishes be capable of absorbing and generating gas at change-of-depth 

 rates observed for migrating scattering layers has been questioned on physio- 

 logical grounds (Kanwisher and Ebeling, 1957). However, Marshall (1960), in 

 an extensive and fascinating account of the swim-bladder in bathypelagic 

 fishes, offers ample anatomical and biological argument for the use of swim- 

 bladder as a hydrostatic organ in vertical migrants of the upper 1000 m of the 

 ocean. In such fishes both the gas-absorbing and gas-generating structures of 

 the swim-bladder are very elaborately developed. 



The alternative in such a vertically migrating fish is to be at neutral buoyancy 

 at near-surface level and simply to let the swim-bladder gas compress as the 

 fish migrates to a depth where the animal would be negatively buoyant. In 

 certain eastern Pacific myctophids, Kanwisher and Ebeling {op. cit.) observed 

 (from the composition of gas in the bladder) that oxygen had been secreted at 

 depth. These authors ask how the fish could have secreted the right amount of 

 gas at depth for achieving neutral buoyancy at shallow level. That this is what 

 these fishes do seems improbable and it is more parsimonious to reason that 

 adjustments in the volume of gas are made during the upward migration. 



We have determined the relationship between peak sound-scattering fre- 

 quency and depth during the vertical migration of deep scattering layers in 

 three instances (see p. 525). The relationship that would obtain were the swim- 

 bladder and its gas passively responding to changes in ambient pressure 



