The origins of modern photography go back to the ancient concept of the camera obscura, a device originally comprising a dark-room into which light was allowed to pass through a small hole, allowing an image of the scene outside to be projected on the opposite wall.
The idea was miniaturised by 16th-century artists as an aid to fathoming the mysteries of perspective, and the invention of lenses not long afterwards allowed the image to be sharply focused. But it was not until the 19th century that techniques evolved to preserve the images thus conjured up.
It was an art which allowed William A Bentley of Jericho, Vermont, to carve for himself a special niche in meteorology. In 1885, at the age of 20, he photographed a snowflake; then he photographed another, and another, until it became for him a passion that was to last for more than 40 years. Using his camera through a microscope, Bentley built up a library of 6,000 snowflake pictures.
There are billions of snowflakes in a shower of snow. Each is an aggregate of many hundreds of tiny hexagonal ice-crystals, even the largest of which are only a millimetre or two in diameter.
Bentley had to work fast, with only seconds, or at most minutes, available each time before the snow melted. Nowadays there are techniques to make such matters easier: collected samples are plunged into a cup of liquid nitrogen at more than 100 C below zero, and at this very low temperature the crystals can be coated with platinum, so as to have them retain their shape indefinitely.
But why, aside from the aesthetic passion which obviously prompted Bentley, would one wish to photograph a snowflake? As it happens, in countries subject to heavy falls of snow in winter-time, and floods in spring as a result of melting, the shape of the crystals making up the snowflake is of some importance.
Satellites can assess the area of snow quite accurately, and rough calculations of its water content can be made from the way in which the snow reflects the light.
But to make more accurate predictions of the amount of melt-water to expect in spring, it is desirable to have information about the size and structure of the individual ice-crystals, and this is best obtained by taking random samples and photographing them for further study.
William Bentley died 70 years ago tomorrow, on December 23rd, 1931. It was just a few days before a book featuring 2,000 of his "photomicrographs" was due to appear in shops, destined to become a meteorological classic.