For Cornell University biologist John P. Berry, knowing the punch line to the joke, "Where does an 800-pound gorilla eat?" is not enough. Certainly, the mountain gorillas he studies in Uganda's Bwindi impenetrable forest eat wherever they want. Whatever, too.

But why is the key question in the field of zoopharmacognosy, the study of animals that use plants for medicine. Berry, a graduate student in the laboratory of noted Cornell phytochemist Eloy Rodriguez, knows enough about Gorilla gorilla beringei's dietary preferences to open Gorilla My Dreams Cafe. He is focusing on a fruit with anti-bacterial properties that even the savvy apes may not understand.

"When you watch gorillas eat in habitats with wide plant diversity, you see why we say they're living in a salad bowl," said Berry, recently returned to Ithaca after another observation and plant-collecting trip to the Bwindi National Park. "They eat practically everything in reach -- leaves, stems, bark, fruit -- and they barely move. When you come back the next day, they're eating a few meters away." Working with trackers, who are hired to observe and protect the mountain gorillas, Berry has documented mountain gorillas eating more than 40 different types of plants.

"They eat some pretty unappealing stuff, too, like pith and rotten wood," said Berry, who has personally sampled much of the gorilla fare. "But so far I haven't found the kind of self-medication that we see in chimps and Aspilia," he said, referring to the first- and best-known example of zoopharmacognosy.

Eloy Rodriguez, now the Perkins Professor of Environmental Studies at Cornell, and Richard Wrangham of Harvard University, in 1985 showed that thiarubrine was the active chemical (against intestinal nematode parasites) in the Aspilia leaves that wild chimpanzees seek when they have upset stomachs. The queasy chimps pick Aspilia leaves and roll them around in their mouths, rather than chewing, before swallowing the capsule-like leaves whole. They swallow as many as 15-35 Aspilia leaves at a sitting, particularly in the rainy season when the rate of parasitic infection is highest.

Furthermore, Aspilia also may have some anti-bacterial effect, Berry, Rodriguez and Marcus McFerren noted in a 1995 article for the American Society of Plant Physiologists' publication, Phytochemicals and Health. When Berry found mountain gorillas eating fruit of the wild ginger plant called Aframomum, which has anti-bacterial properties, the question became: Why?

The finger-sized, urn-shaped Aframomum fruits grow a few inches above the ground, on a plant whose distinctively shaped fronds reach 9 feet high. The bright red, sweet fruit of one Aframomum species is a favorite of Ugandan children and is sold in the marketplace and at roadside stands. But mountain gorillas don't need to go that far; they gather and swallow the fruit along with their other salad items.

Back in his Ithaca laboratory, Berry found that some species of Aframomum fruit prevent the growth of bacteria, including some strains of E. coli and Pseudomonas. Could it be that gorillas -- and human kids, too -- are getting a does of medicine along with their sweet snack? If so, that might add another natural, plant-based treatment to the medicine chest -- one of the goals of zoopharmacognosy.

But before prescribing Aframomum fruit for bacterial infection, Berry acknowledges that he must answer some basic questions: "We need to get a better handle on the taxonomy and learn which species are growing where, which ones the gorillas are eating and which are eaten by humans," he said. "We need more tests of activity against pathogenic bacteria. I'd like to know how this fruit affects the gorillas' micro flora -- the 'good' digestive bacteria in their gut -- and whether their micro flora have developed a resistance to the fruit chemicals. And I have to get more information about the gorillas' feeding behavior."

It's not that he hasn't tried. Berry himself tastes everything he sees gorillas eat. Or almost everything. He can report that gorilla preferences cover a diverse range of tastes, from "sweet," "astringent" and "tasteless" to mouth-numbingly "bitter." But with caution that comes from an extensive knowledge of phytochemistry, Berry doesn't graze willy-nilly through the gorilla salad bar. Some of the alkaloid-laden plants that gorillas crave, he notes, are poisonous to humans.

The easy solution would be to bring a mountain gorilla into the Cornell phytochemical lab and serve him lunch, but that's not possible with one of the most endangered mammal species. The so-called Kyagurilu Group of mountain gorillas that Berry studies in the Bwindi National Park is down to just 13 individuals, including only one silverback (adult male) after deaths last year. But tests of the Aframomum fruit with lowland gorillas in American zoos may be possible, Berry hopes.

The Cornell gorilla-diet study is conducted with assistance from the Institute of Tropical Forest Conservation (ITFC), which oversees gorilla habituation and conservation programs in Bwindi, and from ITFC Director Simon Jennings. Research is conducted with the cooperation of the Uganda National Parks (UNP) and UNP Director Eric Edroma. Berry works from Mbarara University's Ruhija field station, which is directed by Dr. Frederick Kayanja.

Safe of harm, mountain gorillas potentially have a life expectancy nearly equal to humans' -- 40 to 50 years -- and their health is generally good. What role their low-fat, extremely high-fiber diet plays in their health is another question for the phytochemists.

If nothing else, all that fiber makes mountain gorillas fairly easy to find in the impenetrable forest. One might -- to be polite to an 800-pound gorilla -- say they are, well, especially flatulent.

"Let's just say," Berry notes, "you tend to hear and smell them before you see them."