9 8 DISCOVERY REPORTS 



Table 7. Length of retia and the depth range of some deep-sea fishes 



concentration of plankton is also found over the abyssal plain, such a fish could take full advantage 

 of this food as well as that on and in the sediments. Macrourids plough up the oozes, no doubt in 

 search of food, and they also take planktonic forms such as euphausiids, sergestids and copepods 

 (Marshall, 1954). Bathypteroid fishes, which have no swimbladder and come to rest on the oozes 

 after swimming, may be more limited in their choice. A radiograph of one of these fishes showed the 

 skeleton to be well ossified and, as the myotomes are compact and well-developed, they are un- 

 doubtedly heavier than their environment, which as Jones and Marshall (1953) point out, is no 

 handicap to a fish living on the bottom. But the more striking fact is that about two-thirds of the 

 fishes that range below 3000 m. have swimbladders. 



Besides its buoyant effect, the swimbladder can also be used to produce sounds. Certain species of 

 macrourids have special drumming muscles attached to the swimbladder, but sounds can be made 

 without such aid (Marshall, 1954). The gas-filled sac may act as a resonator to enhance the noises 

 produced by neighbouring structures, such as the grinding of the pharyngeal teeth. In the abyssal 

 darkness, sounds might be vitally important for communication between the sexes during the breeding 

 season. After studying certain sounds recorded in deep water, which were like those made by fishes, 

 Griffin (1955) suggested that the 'fish' might use the sounds for echolocation of the bottom. His 

 concluding remarks are interesting : ' Finally it must be pointed out that even though this recording 

 does reveal a fish-call plus its echo from the bottom, we have no direct evidence that the unknown fish 

 could hear such an echo, and still less that it would pay any attention if it did. Yet the "echo-fish" 

 could easily have heard these bottom echoes if it had an auditory sensitivity equal to that of any fish 

 adequately studied to date, and in the unlighted depths of the ocean echolocation could be as advan- 

 tageous to a fish as it is to a bat flying in darkness through the air.' 



By becoming closely associated with the inner ears the swimbladder can also act as a hydrophone. 

 Such association is known in one group of benthic deep-sea fishes, the Moridae, (Svetovidov, 1948) 

 a family containing about seventy species. In other fishes a connection between the ears and swim- 

 bladder is correlated with a wide frequency range and low auditory threshold (Jones and Marshall, 

 1953). At present we can only add that the deep ocean is by no means a silent world. 



It must now be evident that there is a striking difference between the development of the swim- 

 bladder in the pelagic and benthic fishes of the deep ocean. Taking a vertical section over the deep-sea 

 floor between depths of 2000 and 5000 m., at least half of the pelagic fishes with centres of concen- 

 tration in the upper reaches (between 200 and 1000 m.) have capacious swimbladders and the same 

 is true of the bottom-dwelling species. But this organ is regressed or absent in the pelagic fishes living 

 in the intervening waters below the 1000-m. level. 



