As you read these lines, I expect to be in Sweden. I will be on a brief sojourn in what is, by all accounts, a very dull, pedestrian, provincial town called Norrkopping, a place that the Swedes insist on calling "Knorrchirping". One of my British colleagues calls it the Dagenham of Sweden; another, equally chauvinistic, prefers the West Hartlepool of Scandinavia. Be that as it may, at least for the first time in my life, I shall have been to Sweden.
Funny people Swedes. The rumour is that from a date in the not too very distant future, the use of mercury thermometers to measure Swedish temperatures will be forbidden. Meteorologists will have to use electrical sensors, or some other means, instead. But one can understand the reasons why.
For all its silky sheen and the strange attractiveness of its paradoxical liquidity, mercury is not at all a pleasant substance: it is highly toxic. In older days, when compounds of mercury were used in the making of felt hats, the eccentricities of those exposed to their noxious fumes were epitomised in the Mad Hatter stereotype immortalised by Lewis Carroll.
It was also responsible - or so we have been told - for Oscar Wilde's uncharacteristic habit of covering his mouth as he delivered all those witty epigrams. Mercury was the accepted treatment at that time for an indelicate ailment that Wilde contracted in his 20s; it did not cure him, and as a side-effect, his teeth were unattractively discoloured.
Over the centuries, however, mercury has served us well as the working fluid in thermometers. In the early days of meteorology, it was obvious that the key to the challenge lay in the fact that a change in temperature can bring about a change in volume. In due course, meteorologists hit upon the idea of a bulb filled with a liquid, attached to a narrow tube of glass: the changing level of the fluid in the tube would reflect variations in the temperature.
The two most popular liquids used at first were alcohol and water, but both had disadvantages. Water solidifies at 0C, and a thermometer which uses it will not function in sub-zero temperatures. Alcohol, on the other hand, does not freeze until its temperature is well below minus 100C, but it boils at 78C, another obvious limitation.
Mercury, however, proved to be ideal; not only does it have a conveniently large range of liquidity, from minus 40C to plus 357 C, but it also reacts quickly and uniformly to any change in temperature. As an added bonus, the thin silver thread in the glass tube is easy to observe.