SNAKEROOT is the source of reserpine, used in tranquilizers and to treat high blood pressure

ROSY PERIWINKLE has yielded two cancer drugs: vincristine and vinblastine

This time he brought along his wife and four young children. The family settled on the island of Savai'i in the isolated village of Falealupo, the westernmost point of Western Samoa, one of the world's poorest countries (average annual per capita income: $100). Here, far from many of the Western influences of neighboring American Samoa, Cox felt he could learn about the plants and the healers who use them before both vanished.

Major technological advances in screening processes have helped Cox and other ethnobotanists immensely. Pharmacologists must analyze between 10,000 and 17,000 chemical compounds before finding one with the potential to be tested for efficacy in humans. Until recently, animal testing and clinical trials of a single drug required an average 12 years of research and cost up to $300 million. But initial screening can now be done in a matter of days without using animals. Molecular biologists are able to isolate enzymes that can trigger human diseases, then expose those enzymes to a plant's chemical compounds. If a plant extract blocks the action of a particular enzyme--say, one that promotes a skin inflammation--they know the plant has drug potential. By extracting specific chemicals from the leaves, roots or bark with a series of solvents and testing each sample individually, scientists can determine which of the plant's thousands of compounds actually blocks the enzyme.

As a result of these advances, about 100 U.S. companies are searching out plants. Drug companies and scientific institutions are collaborating on field research all over the globe, racing to study as many natural substances as possible before they, or the native people who use them, disappear. Some work with the handful of ethnobotanists like Cox to ferret out drug candidates based on their knowledge of indigenous peoples. Others use a broad-brush approach, mass-collecting plants whose chemical compounds might contribute to new drugs.

One of the most extensive prospecting efforts is the National Cancer Institute's, which is focusing on screening plants for compounds active against the AIDS virus and nine major types of cancer. Since 1986, the NCI has received samplings of thousands of different species from ethnobotanists as well as such institutions as the New York Botanical Garden, the Missouri Botanical Garden and the University of Illinois at Chicago.

In contrast to random collecting, Cox feels, ethnobotanical field research provides a far more streamlined way of locating plants that have medical potential. "Indigenous people have been testing plants on people for thousands of years," says Cox. More important, healers may alert ethnobotanists to nuances that random collecting could miss. Take Homalanthus nutans, a rain-forest tree whose bark Samoans have used for centuries as a cure for hepatitis. Cox quickly found that he could not just casually go into the forest and gather the bark because 1) there are two varieties of the tree, and the bark of only one is effective, and 2) only trees of a certain size produce the desired extract.

After Cox collected the proper bark samples, he sent them to the NCI in the mid-1980s for testing. One isolate, called prostratin, appeared to inhibit growth of the AIDS virus, at least in the test tube, leading the NCI to patent it. If prostratin should ever be developed and approved by the Food and Drug Administration, both the Western Samoan government and the citizens of Falealupo could be in for a windfall under a royalty arrangement that Cox worked out between both entities and the NCI.

Cox has located three other medically promising plants. Two of the plants, used by Samoans to control skin inflammations, are being investigated by a pharmaceutical firm. The third doubles the life span of infection-fighting T lymphocytes in the test tube; its effect in the human body is not yet known. Cox's family has already benefited from the anti-inflammatories. After his infant daughter Hillary came down with a skin infection that did not respond to Western ointments, a healer ground up some leaves; the resulting greenish goo made the infection disappear. When Cox's son Paul Matthew was stung by wasps, healers rubbed bark on the wounds, and the swelling vanished.