Why X-rays? X-rays and visible light are both electromagnetic radiation and have a wave-like nature. But the wavelength of visible light is more than a thousand times longer than the distances within molecules such as the length of the bonds that join atoms together or the separation of the bases in a nucleic acid chain. To get information about this kind of detail, we have to 'probe' the molecules with radiation whose wavelength is of a similar order to the dimensions involved and X-rays are just right for this purpose - they have wavelengths of about 1-2 Angstrom units (0.1 - 0.2 x 10-9 m). X-rays falling on a molecule are scattered by the atoms in it (strictly speaking by the atoms' electrons).

The diagram shows a beam of X-rays falling on a molecule (such as a protein molecule) and the rays that have been scattered in a particular direction by some of the atoms in the molecule. You can see that the incoming beam is made up from rays that are all in step (in phase) with each other, but the scattered rays are no longer in step, having travelled different distances. The result is that the rays tend to cancel each other out, as the peaks of one ray fall near the troughs of another, and so on. The strength of the scattered beam is therefore sensitive to the spatial arrangement of the atoms in the molecule.

Why crystals? A single molecule is very small and would scatter the X-rays very weakly. In a crystal, molecules are lined up in a regular way, like soldiers in a platoon, and their scattering of X-rays adds up like the sound of the footsteps of the marching soldiers.

The molecules in some important biological materials are lined up naturally. These include hair, wool and silk, fibrous protein molecules that, because of their inner regularity and resultant strength were used for textiles before artificial materials such as nylon and terylene became available - even now they remain widely used, often in combination with the newer substances. The consequence of their internal regularity is that they form fibres and, when a beam of X-rays falls upon a fibre, it is scattered strongly only in those directions in which the contributions from different parts of the molecules add up. The scattering pattern can be recorded on a piece of photographic film placed behind the specimen, and the positions of the strong spots can then be measured, giving information about the regularities within the fibre. Astbury, one of the British pioneers of Biophysics working in Leeds, found that the pattern given by hair or wool changed when the fibre was stretched, showing that the molecules themselves had been pulled out from a tightly-wound, compact, form into an elongated shape.

DNA is a very long molecule and does not naturally form long, thin fibres. But it is possible to extract DNA from cells in the form of a viscous gel; if a needle is dipped into the gel and slowly wound up, it drags out a DNA fibre in which many molecules are lined up parallel to each other. The X-ray patterns given by DNA fibres show a pair of strong arcs along their vertical axis; Astbury realised that their position indicated a very regular periodicity of 3.4 along the axis of the fibre and that this figure was similar to the thickness of the DNA bases; he therefore suggested that the bases were stacked on top of each other "like a pile of pennies". He was quite right, but the well-known double helix structure had to await much better X-ray pictures (obtained by Wilkins, Franklin and colleagues at King's College, London) and the realisation by Crick and Watson (in Cambridge) that the bases were in pairs, joining two backbones running in opposite directions.

Globular proteins (such as enzymes, hormones and antibodies), unlike the elongated protein molecules that nature has designed to form fibres, are folded up into compact shapes. Fortunately, it is often possible to form crystals of them, in which the molecules are regularly arranged in a 3-dimensional lattice with water filling the spaces between the protein molecules.

X-rays falling on such crystals are scattered in a very regular way; the regular spacings between the spots just tell us about the distances between the molecules in the crystal lattice, but the variations in strength of the spots give information about the arrangement of the atoms within each molecule. Unfortunately, this is only half of the information required to determine the atom positions completely, because each X-ray beam falling on the film has two properties - its amplitude (which we can measure) and its phase relative to all the other beams (which we cannot). Various cunning methods have been found to overcome this problem, one being to modify the protein molecules in the crystal at one or two positions by adding marker atoms which scatter X-rays strongly.

Kendrew (working in Cambridge) and his colleagues were the first to elucidate the 3- dimensional structure of a globular protein, that of myoglobin, the protein that stores oxygen in muscle tissues. Shortly afterwards, Perutz and his team (also in Cambridge) found the structure of haemoglobin, which carries oxygen in the erythrocytes in the blood. Each haemoglobin molecule comprises two pairs of protein chains, each kind of chain being very similar to a myoglobin chain. This was the first evidence from 3-dimensional structures of the way in which complicated molecules have evolved from simpler ones, so giving rise to novel properties; in the case of haemoglobin, it is vitally important that the protein is capable of picking up oxygen molecules in the lungs and releasing them in the muscle tissues, where they are required as fuel, and its ability to do this is a result of a subtle re-arrangement of its four chains in response to small changes in oxygen pressure and pH (acidity).

The third protein of known 3-dimensional structure was an enzyme, lysozyme, studied by Phillips, Blake, Johnson, North and their colleagues in London. Lysozyme is a protein with anti-bacterial properties (found in tears and other fluids) and in this case the studies of the enzyme allowed the mechanism of its action to be worked out - the enzyme grips its substrate molecule like a pair of pliers so that active chemical groups on the protein are just in the right place to attack the substrate.

This pioneering research on biological molecules was almost all carried out in Britain, but activity is now international and many hundreds of structures are known, including those of substances of great complexity such as the proteins that control the activity of DNA and even viruses such as the common cold virus shown here.