SECT. 4] LIGHT AND ANIMAL LIFE 461 



There is no certain knowledge of the depths at which animals can see by 

 daylight. Observers in a bathyscaphe could detect daylight when looking 

 horizontally and could see light reflected by objects down to 600 m (Dietz, 

 1959). The fully dark-adapted human eye with a pupil of about 0.5 cm^ can 

 detect about 10"^ ^W/cm^ of retina (A 510 m[ji) when looking at 5-sec flashes of 

 a uniform circular field subtending 47° at the eye. The area of retina covered 

 by such a broad field will be 1.5 cm 2 so that, at the pupil, the flux will be 

 3 X 10-8 [jt,W/cm2 (data taken from Pirenne, 1956). Recent direct measurements 

 by Clarke (unpublished data) show that, in the very clear waters of the Brown- 

 son Deep, 50 miles north of Puerto Rico, daylight will be reduced to this 

 intensity at a depth of about 880 m 1 [a depth slightly less than this was given 

 by a small extrapolation of measurements obtained in the Mediterranean off 

 Monaco (Clarke and Breslau, 1959)]. 



In the eye of a deep-sea fish, the aperture is greater, the eye media are more 

 transparent, and a greater fraction of the light striking the retina is absorbed 

 by the retinal photosensitive pigments than in the human eye. There are, 

 therefore, good reasons for believing that deep-sea fish will be able to respond 

 to lower intensities of light than we can. Denton and Warren (1957) have dis- 

 cussed this problem and suggest that deep-sea fish might be expected to be 

 between approximately 10 and 100 times more sensitive than man. If a fish 

 were 100 times more sensitive than man, Clarke's data shows that it would, in 

 the Brownson Deep, be able to detect daylight (in the absence of luminescent 

 fight) at a depth 120 m greater than man. If the water below 880 m had been 

 as clear as the clearest stratum measured, the extra depth would have been 

 200 m. 



In the human, visual acuity rises very rapidly as the light intensity is in- 

 creased above absolute threshold (Pirenne quoted by Pirenne and Denton, 

 1952). When looking at a broad field of hght only a few times brighter than the 

 absolute threshold of vision, the fingers of a hand held in front of the eye can 

 easily be seen. A fish wifl, therefore, probably begin to discriminate surrounding 

 objects quite well by the light of day only a short distance above the depth at 

 which it can just detect daylight. 



If the two eyes have the same aperture, then, in detecting the broad field of 

 penetrating daylight, the illumination on the retina of a small fish (or cepha- 

 lopod) eye will be equal to that on the retina of a fish with a big eye. The big 

 eye may be able to sum the responses of light over a larger area of retina, yet 

 the advantage of a big eye may not be very great in these circumstances. To 

 detect a small flashing light, a large eye which can collect light over a large 

 area of pupil and concentrate it on to a small area of retina is clearly better 

 than a small eye, just as a large telescope is better than a small one at detecting 

 distant stars. In the sea we do find animals, particularly squid, with very large 

 eyes indeed. The eye of one giant squid measured 37 cm in diameter (Cooke 



1 The estimate of 950 m by Denton and Warren (1957) for the absolute limit to which 

 the human eye could detect daylight is too high, being based on the assumption of too high 

 a value for the transparency of oceanic water. 



