GENETIC ENGINEERING. Biologists have come a long way in learning to alter genes. But for the most part, they're like budding mechanics who aren't sure how using a particular tool will ultimately affect the performance of a car. They have a lot to learn before genetic engineering becomes practical for industry and agriculture.

TO the biologist, it's a breakthrough in understanding organic life. To environmentalists, it's a potential hazard of unknown proportions. To industrial chemists, it's a nifty way to get useful products. But to much of the general public, genetic engineering is a vague notion of high-tech manipulation of plants and animals. Its results may affect people, but the subject often seems too sophisticated to understand. Too sophisticated to understand? Citizens of Cambridge, Mass., didn't find it too hard to grasp when they helped set guidelines for gene-splicing research at local universities a decade ago. Their work has become a model for citizen involvement in regulating potentially hazardous research.

``The public . . . can assimilate an astonishing amount of technical information if they feel that it's necessary to protect themselves in a dispute,'' Robert C. Forney, an executive vice-president of E. I. du Pont de Nemours & Co., told a recent American Chemical Society meeting.

To begin with, experts say that genetic engineering is not as advanced as some of its boosters and critics may lead you to believe. It has taken biologists a decade just to master the handling of the tools for genetic manipulation. They're able to alter certain bacteria, among the simplest of organisms. But they're far from redesigning higher animals, let alone human beings.

In fact, biologists trying to adapt genetic engineering for practical use are a bit like budding auto mechanics who have learned to rejigger an engine without knowing how the change would affect the car's performance.

Their growing skill has enabled them to decipher the so-called genetic code -- rules by which the instructions that determine the form and functions of an organism are recorded in the chemical structure of a molecule called DNA (deoxyribonucleic acid). They have learned how living cells translate those instructions into blueprints that specify the exact structure of proteins, a basic chemical of life. Proteins are the source of nutrition and energy for living organisms and often serve as catalysts for making other chemicals that a plant or animal cell needs.

One of the biologist's most important tools for tinkering with genes is a set of chemical ``scissors'' called restriction enzymes. Genetic engineers use them to snip a gene (the basic unit of genetic information) or complex of genes out of a DNA molecule and put it somewhere else. But biologists can't insert genes into another organism directly. So they rely on two agents that can naturally carry genetic material into cells such as bacteria, which are are simple to work with and reproduce rapidly. The genes are usually inserted into a virus that naturally infects bacteria. Or it could be inserted into a little circular molecule of DNA called a plasmid, which occurs naturally in bacterial cells. The virus or plasmid carries the foreign DNA into bacteria, which then acquire new properties.

Genetic engineers use their ability to cut and paste bits of DNA in several ways.

They may give bacteria the capacity to make copious amounts of specific proteins, usually for use as drugs. They may use a bit of DNA as a probe to ``fish'' for similar DNA pieces in an unknown DNA mixture. This helps them locate and read the genetic instructions from the DNA of different organisms -- a process called gene sequencing. Scientists may even use their manipulative abilities to form new genes by linking up DNA sequences of their own design.

These and other techniques are developing rapidly. For example, genetic designers have been limited in using restriction enzymes, because they cut DNA only at specific points. Last March, Waclaw Szybalski of the University of Wisconsin at Madison reported a technique that allows engineers to snip DNA more or less where they like. It's a difficult, fairly expensive operation. But its development shows that what had seemed a basic limitation can be overcome.

Another difficulty has been the cumbersome manual methods for sequencing DNA. They take time and effort and thus slow down the work.

But in June, Lloyd Smith, Leroy E. Hood, Jane Sanders, and Robert Kaiser of the California Institute of Technology announced development of a machine to read off the genes automatically.

Referring to the insights physicists have gained by probing inside the atom, Dr. Hood observed: ``Just as the revolution in physics occurred some 20 years ago with the development of accelerators which could smash atoms, so has the revolution in biology occurred with machines which allow us to analyze and sequence proteins and genes.''

Now biologists are challenged to put that growing knowledge to practical use. Genetic engineers can give bacteria the ability to make proteins valuable to industry or medicine -- say rennin to coagulate milk in making cheese, or human insulin. But they don't know the rules that determine a protein molecule's shape, which enables it to perform a desired function.

Thus, while researchers can specify a protein's chemical makeup, they can't always ensure that it will snap into the shape they need.

Also, giving redesigned bacteria an ability they don't normally have to make large amounts of a protein makes them less fit to survive. So it's hard to keep a new strain of bacteria working day after day in industrial vats when natural evolution would tend to eliminate it.

Industrial chemists Harold G. Monbouquette and Davis F. Ollis of North Carolina State University explain in the current issue of the industry journal Chemtech: ``Genetic tampering with the well-evolved biochemical machinery of a living cell often results in microorganisms that are at a severe disadvantage. . . . Those few cells that somehow revert to the wild type have a natural advantage . . . [and] may become the dominant cell population.''

Such renegade cells gobble up a manufacturer's raw materials and give no useful product in return. The economic success of the new biotechnology will depend heavily on learning how to ensure the stability of redesigned microbes.

So far, the United States Patent Office has received some 6,000 applications involving the new biology. Many detail ways to get unprecedented amounts of unusually pure chemicals. It's essentially ``better mousetrap'' technology rather than the basis for an industrial revolution. Massive industrial use of genetic engineering is still beyond the horizon. On Oct. 2: Prospects for genetic products.