Light comes in through here, falls down onto a plate here. It is reflected sideways where there’s a prism. That prism sets up the first dispersion. We can forget all about this for the moment. We come to the other prism here, which recombines them and then sends them back into the middle of the instrument again. Now these levers here control a prism in each branch of the dispersion, which enables you to select a special wavelength range. Having recombined them they fall here on a sector wheel and the job of the sector wheel is to allow one wavelength to come past at a time. Whereupon we come to a detector in the middle which then feeds out a signal of the strength of what it’s seeing to this meter. And so essentially one setting of these levers selects a pair of wavelengths and then, by moving the prisms, you select another pair of wavelengths. So that’s a total of four wavelengths in all on which we’re going to measure. The signal is then rectified, fed to this, and if the wavelengths are equal there’s no deflection of the pointer. If they vary enormously in strength the thing would oscillate at huge and it’s your job to reduce the oscillation and bring it down. You can’t really keep it still, I mean to keep it still is nonsensical. Essentially what you want to do is make a very small movement which means the thing oscillates gently across the null point. And a measure of the width of how much you’re having to do for that is very crudely related to the intensity of light. If it’s very strong light then a very small movement will do it for you. If you’re working under very extreme light conditions you may even have to, in order to see anything happening, to have quite a large movement of this. In the old days when you had the smoked plate on it was obvious ‘cause you left a [laughs] a trace on the smoke so when the light was strong you get a very small fine trace, when you’re working with very weak light you get a massively wide one. But you can still read it pretty accurately off the smoked plate. And it’s those readings which with a lot of work you eventually turn into a measurement of ozone. Well one is always reading in various places about the recovery of the ozone hole, and yes it will continue if we wait long enough for all the chemicals which we put up there to disappear. The trouble is some of those have immensely long timescales, but I like to think that perhaps if the ozone’s hole, which will almost certainly still be visible long after I’m gone, that it might just remind people that looking after the Earth is a rather difficult and important thing.