What is the fastest event that can be measured?

—R. MITCHELL, MELBOURNE, AUSTRALIA

Scott A. Diddams and Thomas R. O'Brian of the Time and Frequency Division of the National Institute of Standards and Technology offer an explanation:

The answer depends on how one interprets the word “measured.” Both the accurate measurement of fleeting events and the recording or inference of such occurrences are of interest. So we suggest rephrasing the original query into two new ones: “What are the shortest time spans that can be measured with a particular accuracy?” and “What are the briefest happenings that can be recorded or inferred?”

Currently cesium atomic-fountain clocks are the best way to measure time with a certain accuracy. These “clocks” are actually frequency standards rather than timekeeping devices, and they achieve the defined cesium standard frequency with the exceptional precision of about one part in 10. Put another way, in 30 million years of continuous operation, they would neither gain nor lose more than a second. Yet these frequency standards are rather “noisy,” and to achieve such impressive results requires averaging many thousands of separate frequency measurements over a period of about one day. So fountain clocks are not generally useful for timing short-duration events.

Among the shortest-period events that can be directly created, controlled and measured are quick bursts of laser light. These pulses occur on timescales of femtoseconds (10−15 second) and, more recently, attoseconds (10−18 second). Femtosecond pulses can function in a manner similar to the flash on a camera used to “freeze” events that are too fast for the eye to register. A source of femtosecond-order light pulses is a mode-locked laser, in which many optical waves cooperate to produce a pulse. It is yet another problem, however, to accurately measure that pulse's duration. No photodetectors or electronics are fast enough, so scientists commonly employ correlation techniques that translate a temporal measurement into a distance measurement. To date, pulses as short as a few hundred attoseconds have been generated and measured.

Now let us consider the second question, regarding the most transitory episodes that we can record or infer. Events as short as about 10−25 second have been indirectly inferred in extremely energetic collisions in the largest particle accelerators. For example, the mean lifetime of the top quark, the most massive elementary particle so far observed, has been inferred to be about 0.4 yoctosecond (0.4 × 10−24 second).

Why is normal blood pressure less than 120/80? Why don't these numbers change with height?

Jeffrey A. Cutler, senior scientific adviser for the National Heart, Lung, and Blood Institute at the National Institutes of Health, responds:

The origin of the designation of 120/80 as the threshold for “normal” blood pressure is unknown. (The top number is the systolic pressure, which is the pressure in the arteries while the heart is pumping; the bottom is the diastolic pressure, a measure of pressure in the arteries while the heart is resting and refilling with blood.) It may have come from data available early in the 20th century from life insurance physical exams. In any case, epidemiological studies confirm that the risk of a heart attack or stroke begins to increase in adults when the systolic is 115 or greater or the diastolic is 75 or more. The risk steadily increases with higher and higher readings, so the traditional 120/80 level remains a reasonable guideline for getting a doctor's attention, with the main goal of preventing your pressure from continuing to rise in subsequent years.

Blood pressure does in fact increase with height, ensuring that the brain, located at the uppermost point of the circulatory system for most of the day, gets sufficient blood flow and oxygen. But the effect is fairly small, which is why the 120/80 figure is not adjusted for taller people.

For a complete text of these and other answers from scientists in diverse fields, visit www.sciam.com/askexpert