Recent experiments (Dittrich et al. (Proc. R. Soc. Lond. B 251, 195 (1
993))) suggest that pigeon perception of wasp mimicry by hoverflies is
similar to that of humans and of computer-based image matching. Howev
er, the relations are nonlinear and may explain why some species are a
bundant despite their being poor mimics to the human eye. We suggest t
hat these discrepancies between pigeon and human categorization may li
e in the differences between avian and primate colour vision. As pigeo
n categorization and computer image analysis were both assessed by usi
ng colour slides designed for human vision, they lacked the natural co
lour information available to wild birds, in particular that from ultr
aviolet (uv) wavelengths.