A seemingly contradictory result

I mentioned in my previous post that I would describe my favourite example of a seemly contradictory result in science. (The following is an edited excerpt from an article I wrote a couple of years ago.)

G. I. Taylor in 1909 under to tutelage of J.J Thompson conducted an experiment in which a double slit in front of a photographic plate was illumined by a very dim light source – equivalent to a candle burning at a distance of one mile away. The intensity was set to ensure that the hypothesized photons would pass through the slits one at a time; thus if the usual diffraction pattern was observed it would require the photons to ‘interfere with themselves’ – a truly immoral thought. After a three month exposure – conveniently timed so that Taylor could partake in a sailing trip he had planned – the usual diffraction pattern was observed. He concluded that light must be fundamentally wavelike as the idea that particles could interfere with themselves to generate a diffraction pattern was inconceivable. Not until 1987 when Robert Austin and Lyman Page conducted experiments in which they recorded single photons striking a surface behind a double slit was it irrefutably confirmed that Taylor had misinterpreted his results, in fact photons do “interfere with themselves”.

Three slit diffraction recorded using a single photon counting CCD camera: (A) after 0.033s exposure, (B) after 1s exposure and (C) after 100s exposure; work by Robert Austin and Lyman Page.