The launch of a space shuttle is an impressive event. It is impressively big, impressively noisy, and impressively expensive. Most of us love fireworks, and I have never met anyone who did not enjoy watching this fireworks display on the grandest scale.

During the first few minutes of ascent, energy is used at a rate sufficient to power the whole United States. Is this a necessary display of size and power? We are in the habit of thinking that sending something into space requires a vast and powerful rocket, but could we be wrong?

We could, and we are. Most of our preconceived ideas about rocket launches go back to the early days of the "space race" between the United States and the Soviet Union, and in the 1950s and 1960s the name of the game was placing humans into orbit.

There were good psychological and practical reasons for wanting to send astronauts and cosmonauts. First, the public is always far more interested in men than in machines; and second, the computers of the early days of the space program were big, primitive and limited in what they could do. People, by contrast, possessed - and possess - far more versatility than any computer, and can perform an endless variety of tasks.

On the other hand, people come in more-or-less standard sizes. They also need to eat, drink and breathe. Once you decide that humans are necessary in space, you have no alternative to big rockets; but today's applications satellites, for communications, weather observation, and resource mapping, neither need nor want a human presence.

As computers become smarter and more powerful, they are also shrinking in size and weight. The personal computer in your home today is faster and has far more storage than anything in the Apollo program spacecraft. Other electronics, for observing instruments and for returning data to the ground, are becoming microminiaturized. Payloads can weigh less. So how big - or how small - can a useful rocket be?

We are in the process of finding out. Miniature thrust chambers for rocket engines have been built, each one smaller and lighter than a dime. A group of about a hundred of these should be able to launch into orbit something about the size of a Coke can. That's more than big enough to house a powerful computer, plus an array of instruments. One of these "microsats" could well become an earth resources or weather observing station.

We are at the very beginning of thinking small in space. How small might we go?

Since our experience with space vehicles is limited, let us draw an analogy with aircraft. Today's aircraft, like today's spacecraft, are designed to carry people. Suppose, however, that we just want a flying machine that can carry a small payload (maybe a few grams, enough for a powerful computer).

How small and light can it be? We don't have a final answer to that question, although today a jet engine no bigger than a shirt button is being built at MIT. However, Nature provides us with an upper limit on overall size. Swallows, weighing just a few ounces, every year migrate thousands of miles without refueling. We should be able to do at least this well.

And if you want to think really small, look at what the swallows eat: flying insects, each with its own on-board navigation and observing instruments. Imagine a swarm of space midges, all launched on a rocket no bigger than a waste paper basket, each one observing the Earth or the sky and returning their coordinated observations back to the ground.

Imagination should become reality in less than half a century.

"Borderlands of Science" is syndicated by: