Measure for measure, the kilogram doesn't square up

As spring arrives, the National Institute of Standards and Technology urges us to "hug a weights and measures official." That's not as quixotic as it might seem.

These officials make sure that a pound of potatoes is a pound of potatoes and a quart of milk is really a quart when we buy them. Without that assurance, thrifty shopping would be a dubious exercise. But there's more at stake than market honesty, for all is not well in weights and measures land.

The definition of the standard kilogram, against which all other standards of mass and weight are calibrated, is fundamentally flawed. Getting that definition right is a challenge that has tried the patience and ingenuity of metric scientists for decades.

Scientists use just seven basic units to define all the other quantities we use - quantities such as speed, density, or electric power. All of those basic units except the kilogram are themselves defined in terms of natural properties that are beyond the reach of human manipulation.

For example, the standard second (time) is defined as a specific number of vibrations of a type of radiation emitted by atoms of caesium-133. The standard meter (length), in turn, is defined as the length of the path light travels in a vacuum during a specific fraction of a second.

Not so the kilogram. This orphan of the basic unit family is simply the mass of a small platinum-iridium alloy cylinder locked away in a vault maintained by the International Bureau of Weights & Measures in Sèvres, France, near Paris.

Official bodies around the world have copies of that cylinder. The one at the National Institute of Standards and Technology (NIST) in the United States stands behind the scale that weighs out your pound of potatoes.

Embarrassingly, the last time the copies were brought to Sèvres for a check up in the 1980s, officials found that some copies had gained about 20 parts per billion in weight compared to the master cylinder since the previous checkup in the 1940s. That may not matter when you're baking a cake, but it implies that the master cylinder itself may be an inconstant standard.

No one knows what causes the weight changes. But the uncertainty is intolerable when precision in research and some manufacturing now demands accuracy to a few parts per billion.

Several efforts in several different countries are under way to redefine the kilogram in terms of basic physical quantities such as counting the actual number of atoms of a specific substance in a kilogram or the electromagnetic force that balances a kilogram mass against gravity.

A project of the latter type at the NIST laboratories in Gaithersburg, Md., hopes eventually to define mass in terms of electrical units.

So far, none of these redefinition projects has borne fruit. They require exquisite precision of measurement and control of experimental conditions. The slightest contamination, tiny vibrations, or other external influences - even changes in weather - can ruin results.

You've got to hand it to scientists who are willing to devote many years to such painstaking - but fundamentally important - research.

They deserve a hug.