9 o DISCOVERY REPORTS 



Perhaps the figure for the oxygen consumption is too low. Job (1955) found that a 5-g. Salvelinns 

 fontinalis will take up 1 ml./hr., and from his graphs the figure for a 10-g. fish would be about 1-2 ml. 

 But the oxygen requirements of an alpine char are not necessarily those of a lantern fish. However, 

 it is clear that lantern fishes are very active. Summing up his bathyscaphe observations, Peres (1958) 

 wrote the following concerning these fishes: 'Les individus sont presque sans cesse en mouvement, 

 mais dans les directions les plus variees.' Furthermore, they have a relatively large expanse of gill 

 surface. Using Gray's (1954) procedure, I have measured the gill-surface of two Myctophum puncta- 

 tum, both with a weight of about 5 g. The number of gill-lamellae per millimetre of gill-filament is 

 about 50 and the gill area per gram of body weight is 600-700 mm. 2 Comparing this figure with those 

 got by Gray and those listed by Fry (1957), this lantern fish can be put among the active species with 

 a relatively large gill surface. 1 



There is a further aspect. The catches of the 'Michael Sars' led Hjort (1935) to conclude that the 

 twilight depths of the warm Atlantic were inhabited by a Lilliputian fauna of bathypelagic fishes. 2 

 This is particularly true of the species with a swimbladder. The adults of most gonostomatids, 

 sternoptychids, Astronesthes spp., myctophids and melamphaids range between 20 and 150 mm. in 

 standard length and weigh from about 0-5 to io-o g. 



Now the oxygen consumption of fishes does not follow the surface rule, nor is it proportionate to 

 weight. When the logarithm of the rate of respiration is plotted against the logarithm of weight, the 

 slope of the curve is less than unity (usually from o-66 to 0-9). The rate of uptake of oxygen falls 

 between surface-area and weight dependence (Fry, 1957). 



Thus, the smaller the species, the greater its oxygen consumption per unit of body weight. This 

 may not mean that extra oxygen would be available for gas secretion, but coupled with the consider- 

 able expanse of gill surface in such fishes as the myctophids and Astronesthes spp., the idea is not to be 

 dismissed. At all events it is striking that the only predatory fishes with swimbladders which cross 

 the upper thermocline are the species of Astronesthes? and there is a marked tendency for the adults 

 of this genus to be smaller than those of other predatory stomiatoids (Chauliodontidae, Stomiatidae, 

 Melanostomiatidae, Idiacanthidae and Malacosteidae), which as we have seen, have no swimbladder. 



Lastly, hydrostatic pressure has an effect on the rate of metabolism. Fontaine (1930) experimented 

 on Pleuronectes platessa, Ammodytes lanceolatns, Gobius minutus, and Anguilla anguilla, finding that an 

 increase in pressure from the atmospheric level to 100 kg. /cm. 2 (1 kg./cm. 2 = 0-968 atm.) enhanced 

 the rate of oxygen consumption by 35-88 per cent. The smaller the fish, the greater was this per- 

 centage increase. Using the first species, Fontaine showed that the rate of oxygen consumption 

 steadily rose as the pressure was increased to 125 kg./cm. 2 , but thereafter fell rapidly, the fish being 

 killed at a pressure of 150 kg./cm. 2 . 



A myctophid or astronesthid diving to its daytime level might pass from a pressure of a few 

 atmospheres to one of about forty or fifty. There is thus a possibility that their rate of oxygen uptake 

 would tend to increase with depth. However, it is not certain that deep-water fishes would be in- 

 fluenced in the same way as shallow-water species, although Anguilla anguilla spends part of its life 

 as a deep-sea fish. The question can only be decided by comparable experiments on deep-water 

 species. 



But even supposing that the oxygen consumption of a 10-g. deep-sea fish (say, a large myctophid) 



1 Astronesthes niger, which is also a thermocline-crosser also has about 50 gill-lamellae per millimetre of gill-filament. 

 Argyropelecus aadeatus, presumably a partial migrator, has about 40. 



2 While nets give a limited impression of the size range of a species, bathyscaphe observations do not controvert Hjort's 

 view. 



3 Astronesthes spp. seem to have a decided liking for myctophids (Murray and Hjort, 1912; Beebe and Vander Pyl, 1944). 



