The Andamooka opal field, some 600 km north of Adelaide, South Australia, was discovered early in the 1930's. My uncle, Ralph W. Segnit, was a geologist with the South Australian Department of Mines at that time, and was sent north to investigate the discovery.

It turned out to be a wonderful opal field, and, on his return, he presented me, as a child of nine or ten years of age, with some samples of precious opal.

Thirty years later, I was working in the then Division of Building Research of the Commonwealth Scientific and Industrial Research Organization, Australia, and I had occasion to resurrect these old opal samples.

I arranged the testing of a number of mineral samples, including one of my old opal specimens, in the Differential Thermal Analysis equipment.

DTA, amongst other things, detects changes in the sample, such as loss of water, or change in the crystal structure, by rapid changes in the temperature of the sample during slow and even heating. As opal contains several percent of water, I naturally expected the instrument to react to this loss on the graph produced. It did not.

This seemed to be a false result, so the experiment was repeated, but with the same result. So came the change of direction in what had started out as a survey rather than a research project. A few other opal samples which were available were tested; some gave the expected result, some did not.

On a visit to the Department of Geology at The University of Adelaide, I contacted a colleague, Dr. J.B. Jones, to beg a few more samples of opal. Dr. Jones had scores of opal samples on the benches in his laboratory. He had been studying opal samples by X-ray diffraction (XRD), and had been finding unexplained differences between opal samples. It was natural now that we pool resources and try to iron out our problems together.

After examining a large number of samples in various ways, we realised without doubt that there were two or three distinct types of opal. Nevertheless, there were still some unexplained characteristics, especially regarding precious opal. By a process of elimination, we realised that the difference must be due to something special about the fine microstructure, but what could it be? We found no solution by examination in the optical microscope, which showed few clear structures.

In the meantime we had obtained the help of Dr. J.V. Sanders, one of Australia's foremost electron microscopists, to examine the fine microstructure of many of these opals. We obtained many surprising results. However, we still had the problem of unexplained characteristics of precious opal. Armed with one of those trusty specimens given to me when a child, I once again visited Sanders.

The next day came an excited call saying that he had found in my sample a completely new and unexpected structure.

Better examples of precious opal were then exhumed from collections, as well as samples of potch (opal, from opal fields, showing no colour play). Sanders obtained beautiful electron micrographs showing tiny spheres of silica stacked regularly together like ball-bearings in a box.

It was immediately clear that it was this regular structure which was the basic cause of the colours in precious opal. The structure formed a three dimensional diffraction grating which 'reflected' (actually 'diffracted') only one wavelength (colour) of light from the white light falling on the opal. The diffracted colour would vary according to the size of the spheres and the angle of incidence of the light. These results were first announced in the Annual Report of the CSIRO Division of Building Research for 1963 (R1633).

Once again, one of those coincidences which occur from time to time in scientific research happened. Professor E. Baier, of the University of Mainz in Germany, had studied the optical properties of precious opal using a polarising microscope; he published this work in 1932 (R1465). From this work he deduced that the colour of precious opal was due to diffraction of light, but could not establish the true underlying structure which gave rise to the phenomenon.

However, in the early part of the 1960's, he collaborated with Dr J. Pense, also from Mainz, who produced electron micrographs showing the periodicity of the microspheres of silica in gem opal (R1561).

So the story of the colour of opal seemed to have come to an end, with numerous scientists having made positive contributions to the problem over a period of more than 30 years.

But not so. About this time I happened upon an old edition of the Encyclopedia Britannica, about 1870 vintage, in a church meeting room. On looking up the article on opal, I found that the colour of precious opal was caused by the diffraction of light from a regular periodic structure in the stone. This had been observed and essentially described by Sir David Brewster, the Scottish scientist famous for his work on wave motion and the properties of light, in 1845! (Brewster's paper is reproduced here)

It seemed impossible that Brewster could have seen this periodicity in an optical microscope, but next day, knowing what to look for, I could confirm the observation. An expert from the firm of Carl Zeiss later told me that in the 19th century, special microscope objectives having only a small and curved field of view but with very high resolution at the centre, were made especially for such people as Brewster.

We must admire the pertinacity and thoroughness of the 19th century scientists. They had, certainly in our terms, no sophisticated equipment, but worked meticulously, and gave much careful and intuitive thought to the results they obtained.