Tete mates 2 In my last Editorial {Perception 8(6) 607-608, 1979) I considered how mechanical toys, and dumb computers, can appear lifelike as we project our understanding into things without understanding. (Is this the basis of theologies?) I shall look now at perception in the absence of understanding-or at least perception where there is doubt. We commonly doubt other people's observations, but when should we doubt our own? I am not considering fleeting half-seen things, or things seen in poor light, or while one is ill or tired; but rather very surprising things, seen in the best possible conditions and for as long as one likes. Can one then doubt oneself? Of course this will depend somewhat upon what we mean by 'seeing', for all perceiving is full of interpretation, based on assumptions and knowledge which may be far from adequate. This inadequacy shows up dramatically in unfamiliar situations, and especially for technical observations made when we lack adequate background knowledge. Thus one trusts the naturalist's observations more than one's own, in his field; and if I were an engineer he might well prefer mine to his, for engines ancient and modern. Antique dealers, dentists, doctors, and zoologists all have to learn to see specially, and what they see is no doubt often affected by what they should be seeing. It can require, indeed, genius to make quite new observations. It took many nights, in July 1610, for Galileo to see that his four newly found 'Galilean moons' revolve round Jupiter as a parent body; and he never came to see that Saturn's ring encircles it, and is not handles or ears as he saw, or thought he saw. The first is hardly surprising, for he had to infer the motions of the star-like objects, that they are orbiting moons, from successive positions, but the second difficulty is highly surprising to an observer with our knowledge of Saturn. It is difficult for us to regain Galileo's perception. Perhaps most working scientists are faced with the problem of deciding whether what they perceive is sufficiently reliable for publication. This doubt can occur in spite of photography, and automatic recording devices, and it can occur even when several observers are present. Obvious examples are alleged paranormal phenomena; but doubts occur for normal investigations under laboratory conditions, when background knowledge is inadequate. Here is an example. In the summer of 1972 we were investigating, for the second time, the highly unusual eye of a copepod Copilia quadrata, at the Stazione Zoologica in Naples. Each of the two eyes has a large anterior lens, and, deep in the extremely transparent body, a smaller pear-shaped lens attached to a single 'rod' photoreceptor. This inner lens and receptor move together, across the image plane of the anterior lens, and there is only a single optic fibre-so here we see a scanning eye. With our knowledgebase, we see a mechanical television-type eye in nature: though the superb observer, Sellig Exner, who described Copilia in 1891, was necessarily blind to this pe...