SWIMBLADDER STRUCTURE OF DEEP-SEA 



FISHES IN RELATION TO THEIR 



SYSTEMATICS AND BIOLOGY 



By N. B. Marshall 



British Museum (Natural History) 



(Plates I— III, Text-figures 1-47) 



" When one picks up a fish, one may be said, allegorically, to hold one of the knots in an endless web of 

 netting, of which the countless other knots represent other facts, whether of marine chemistry, physics 

 or geology, or other animals and plants. And just as one cannot make a fish-net until one has tied all the 

 knots in their proper positions, so one cannot hope to comprehend this web until one can see its inter- 

 nodes in their true relationship." bigelow (icno) 



INTRODUCTION 



IN both structure and function the teleost swimbladder is one of the most plastic of vertebrate 

 organs. While it is primarily a hydrostatic organ, making a fish weightless in water, it may also be 

 modified for respiratory, sensory and sound-producing activities (Jones and Marshall, 1953). Apart 

 from the respiratory aspect, this statement is equally true of many deep-sea fishes. And the exception 

 is understandable, for the use of the swimbladder as a lung is obviously limited to physostomatous 

 teleosts, those with a pneumatic duct joining the sac to the foregut. Virtually all deep-sea fishes 

 have a completely closed (physoclistous) 1 swimbladder. 



Cuvier and Valenciennes (1848) appear to have been the first to discover a swimbladder in a deep- 

 sea fish. Concerning the salmonoid genus Argentina, Valenciennes wrote as follows: 'La vessie 

 natatoire de l'Argentine a un autre caractere anatomique et physiologique fort interessant pour nos 

 etudes; elle ne communique pas avec le canal digestif; je n'ai pu du moins trouver de conduit pneu- 

 matique dans les trois individus d'especes differentes que j'ai disseques et dont les visceres etaient 

 cependant parfaitement conserves.' (In the following volume (1850) they record the presence of a 

 swimbladder in the hatchet fish, Argyropelecus hemigymnus.) Nearly forty years later, Giinther (1887) 

 made a similar observation in his ' Challenger ' Report : ' In none of the abyssal forms examined by me 

 have I found an open communication between the air bladder and the oesophagus, not even in those 

 which are referred to the Physostomatous division.' 



In defining the family Sternoptychidae, in which he included the deep-sea fishes, Sternoptyx, 

 Ichthyococcus, Maurolicus, Gonostoma and Chauliodus, Giinther (1864) stated that the swimbladder 

 was simple when present. But under the family Scopelidae (containing fishes now placed in the 

 order Iniomi) he remarked that the swimbladder was absent, a statement that is true with the notable 

 exception of the lantern fishes (Myctophidae). 



There are other observations on the structure of the swimbladder in Giinther's 'Challenger' 

 Report (1887) on deep-sea fishes. Further reference will be made to these later, but we may notice 

 here that the rete mirabile of Gonostoma denudatum was described as a 'conical muscular body' 

 attached to the posterior ventral part of the swimbladder. This confusion is hardly surprising, for the 

 gas-secreting complex (retia mirabilia and gas-glands) of deep-sea fishes is highly developed. The 

 large rete of G. denudatum contains thousands of closely associated arterial and venous capillaries 



1 See page 50. 



