PHOTORECEPTION 205 



length of edges of black against white. This finding suggested that 

 recognition and discrimination is based merely upon the degree of 

 transitory stimulation produced in the compound eye. To test this 

 hypothesis, Wolf and Zerrahn-Wolf mapped on translucent co- 

 ordinate paper the points of intersection of the axes of the ommatidia 

 with a place located at the same distance from the eye as were patterns 

 used in testing. They then placed the co-ordinate paper over a pattern 

 and counted the number of ommatidia whose visual fields would be 

 black and the number whose fields would be white. They also counted 

 the number of ommatidia in which there would be a black-white shift 

 when the pattern was moved over a unit distance in any direction. 

 Calculations could then be made of the relationships between the area 

 of a pattern and the degree of subdivision necessary to provide a con- 

 stant transition value. Predictions of discrimination based on these 

 calculations were realized in field tests. Other lines of evidence were in 

 agreement. For example, bees conditioned to flickering fields of equal 

 size but different flicker frequencies exhibited choices that were directly 

 proportional to the flicker frequency. These findings are in agreement 

 with observations on the natural behaviour of bees in the field (Wolf, 

 1933 c). The honeybee reacts much more effectively to moving flowers 

 than to stationary ones. 



All of these experiments suggested that movement is of greater sig- 

 nificance in form perception than the ability to resolve stationary 

 patterns (cf. Exner, 1891). In its response to flicker the insect eye 

 exhibits many of the characteristics of the human eye. It obeys Talbot's 

 law (the brightness of a fused light produced by flashing it on and off" is 

 visually equivalent to the product of the actual source of illumination 

 and the proportion of the time it is on) and the Ferry-Porter law 

 (critical frequency, that is, the frequency at which a flashing light fuses, 

 is proportional to the logarithm of intensity) as does the human eye 

 (Wolf and Zerrahn-Wolf, 1935 b; Wolf, 1933 b). It is, however, more 

 eflflcient at detecting flicker than is the eye of man. Some of the early 

 values obtained for insects by optomotor methods were : 55 per second 

 for the bee (Wolf, 1933 b); i 60 per second for dragonfly larvae 

 (Salzle, 1932; Crozier, Wolf, and Zerrahn-Wolf, 1937). More highly 

 developed behavioural techniques indicated that maximum values are 

 actually much higher. The critical flicker frequency for Calliphora is 

 about 265 per second, depending upon the number of ommatidia 

 stimulated. The maximum of 265 per second drops to 60-165 per 

 second when only one to four ommatidia are stimulated. A similar 

 relationship holds for the honeybee (Autrum and Stoecker, 1950). 



