THE PERCEPTION OF COLOUR 



631 



from coloured pieces of paper on large grey backgrounds, Revesz (1921) 

 showed that the phenomenon of simultaneous contrast could be 

 elicited as in man. It would seem, indeed, that while the behaviour 

 of Birds is largely determined by vision, they rely more upon the 

 discrimination of hue than of luminosity, and respond more consistently 

 to clues involving colour than those depending on form (Jones, 1954). 



The appreciation of " warning colours " displayed by insects illustrates 

 the biological value of colour vision to the bird in its feeding habits. In experi- 

 ments with the swallow, Hirundo, Swymierton described vividly how one bird 

 would watch another intently, observing its reaction to a new test -insect of a 

 particular colour as if with the intention of profiting thereby by avoiding the 

 unpleasant experience of eating a distasteful species. In the same way birds 

 such as the domestic hen can be trained to tasks involving the discrimination 

 of colour as well as of size and form, both singly and in combination (see Altevogt, 

 1951 ; Thorpe, 1956). 



The colour vision of Birds must be considerably modified by the 

 presence of coloured oil-droplets, a circumstance which must also 

 apply to other species similarly equipped.^ Initially, owing to the 

 inferior quality of the earlier lenses in the microscopes employed in 

 histological work, droplets of a much larger range of spectral hues — - 

 green, blue and violet — were described in the retina of birds, and 

 Krause (1863) put forward the theory that this coloured mosaic 

 represented a peripheral mechanism whereby colour vision could be 

 determined in the avian eye by the absorption of all wave-lengths 

 except one by a particular drojDlet so that different cones were stimu- 

 lated only by a single narrow sjjectral band of light. This theory held 

 the field for many years. Convincing arguments, however, can be 

 advanced against it for oil -droplets are by no means necessary for 

 colour vision : fishes (and man) have colour vision and no coloured 

 droplets ; lizards have a colour-system of considerable complexity and 

 only yellow droplets ; the fovea of birds with its excellent appreciation 

 of colour has yellow droplets only ; and in the periphery of the avian 

 retina the colours of the droplets bear no relation to the spectral range 

 of the bird. It is much more likely, as Walls and Judd (1933) suggested, 

 that these droplets, whatever their colour, act as filters with the triple 

 function of increasing contrast, reducing glare and lessening chromatic 

 aberration — that they are, in fact, an aid to visual acuity. 



The yellow droplet at the avian fovea, cutting off the sjjectrum at 515-520 

 m(x like a yellow-tinted sjaectacle, allows the transmission of many hues but 



> It may be useful to summarize the occurrence of oil-droplets in the visual elements 

 of Vertebrates at this point. They are found in the rods of Lepidosiren; in the cones 

 of Chondrosteans, the coelacanth and Protopterus (all colourless), diurnal Anurans 

 (yellow), lizards (yellow ; some nocturnal types colourless or none), Chelonians 

 (orange, yellow, red), <S'jD/ieHodoM (colourless or pale yellow), Birds (red, orange, yellow, 

 occasionally green or colourless), the platypus (colourless), and Marsupials (except some 

 Didelphyidse) (colourless). 



Swallow 



