The large, hawk like bird lives in large caves in tropical, South American rain forests. It is called an oilbird, because native peoples rendered cooking oil from the fate of its nestlings. We see Roberto Roca, a Venezuelan biologist who heads the zoology section of the Nature Conservancy’s Latin American Science Program during the period of 1985 to 1988. Roca recently completed a study of the oil bird’s feeding habits in northern Venezuela. I will now examine his study in some detail.

Multiple observations set the stages for Roca’s work. Before he began his studies, others had observed that oilbirds roost and nest in large colonies in caves and eat the fruits of rain forest trees. At dusk, the adults fly out of the caves, returning later with fruit, at least some of which they feed to their nestlings. The fruits eaten by oilbirds contain large seeds that the birds regurgitate. Seeds dropped by oilbirds pile up in the caves.

These observations triggered several questions for Roca: How far do oilbirds fly in search of food? Do they eat most of their own food in the forest, regurgitating the seeds there, or bring it back to the caves before eating? If the adults drop large numbers of seeds in the forest, do oilbirds play a role in reforesting areas where trees have been destroyed?

With the Help of Analogies and Inferences, Roca’s Questions Led to Testable Hypotheses, Predictions, and Tests. Roca’s questions soon led him to his main hypothesis: those oilbirds play an important role in reseeding the rain forest. However, because the birds search for food in remote mountain areas at night, Roca could not test this hypothesis directly; he could not simply go out and see whether oilbirds drop seeds in the forest. Roca dealt with this difficulty by generating related hypothesis that were more easily testable. He arrived at these hypotheses by using inferences and analogies, ways of reasoning that often play important roles in science. For example, by analogy with birds what he knew about many other animals, Roca expected that oilbird behaviour would probably tend to conserve energy. From this assumption, he inferred that adult oilbirds would tend to fly the shortest distance possible to obtain fruit. Thus, as long as enough fruit was available near the cave, the adult birds would pick fruit there and probably eat the fruit and drop the seeds in the cave. But from knowledge of the amounts of food that most birds need, Roca inferred that the oilbirds would use up the fruit supply near their cave early in the season, and would have to travel farther and farther for food as the nesting season wore on. He reasoned that eventually the adult birds would be flying so far that, to meet their own energy needs, they would have to eat fruit and drop seeds while out in the forest. This train of thinking led Roca to two testable hypotheses related to the question of whether oilbirds reseed rain forests.

Roca hypothesized that adult oilbirds tend to fly farther from their cave to obtain food late in the nesting season than they do early in the nesting season. To test this hypotheses, he captured eleven adult oilbirds and attached tiny radio transmitters to their hacks. After releasing the birds, Roca climbed a nearby mountain, set up a receiver, and tracked the birds by radio. As he had predicted, the distances the birds flew increased, and by the time the nestlings were four months old, the adults were traveling 100 kilometers (60 miles) from the cave to collect fruit each night.

Roca’s second hypothesis was that adult oilbirds regurgitate many seeds in the forest. He predicted that if this hypothesis were true, most of the seeds piled up in the cave were dropped there by nestlings, not by adults.

To test this hypothesis, Roca performed an experiment in an oilbird cave. He suspended wire trays under rock ledges where there were solitary nests, placing some trays under fully occupied nests and some under nests with adults but no nestlings. Seeds accumulated rapidly under nests with nestlings, but none appeared under nests with out nestlings. These adults supported his second hypothesis and led Roca to make some calculations. He determined the energy (caloric) content of the fruits the oilbird eats. A comparison of the fruits’ energy content with would have to eat at least 50 fruits a day just to sustain itself during the breeding season, suggesting that the bird would indeed drop many seeds in the forest. Roca estimated that 10,000 adults (the population of only one of 20 caves he studied) disperse at least 9 million seeds (almost 13 tons) each month during the breeding season- enough to reseed thousands of square kilometers of nearby rain forest destroyed by burning and logging. In 1989, the Venezuelan government, impressed by Roca’s findings, established a 400-km sq (250-mi sq) preserve for oilbird nesting grounds.

Roberto Roca’s studies are a real world example of the thought processes that contact observations, questions, hypothesis, and deductive reasoning in scientific work. Roca’s studies also illustrate two key elements of the scientific process that we have not yet examined: the use of controls in experiments and the repetition of tests.

Roca’s study of seed piles in a cave was a controlled experiment. In performing a controlled experiment, a scientist actually carries out two parallel tests, one called experimental, the other the control. Ideally the experimental test differs from the control test by only a single factor, called the variable. In Roca’s experiment, the variable was the presence of nestlings in the nest. For his experimental tests, he placed seed trays under nests with nestlings. For his controls, he placed trays under nests with adults only. Controls make it possible to draw clear-cut conclusions from the results of experiments. Without controls, Roca would not have been able to tell whether adults, nestlings, or both were producing the seed piles in the caves.

Repetition of tests is an essential element of the process of science. What if Roca had placed seed trays under only two nests, one with nestlings and one without? It would be risky to generalize from the results of such an experiment. For example, the two adults in one of these nests might have had different habits from most oilbirds. We can draw firm conclusions only after repeating experiments and obtaining consistent results. In Roca’s case, he obtained consistent results from seed trays under ten nests.

It is important to emphasize that science is an adaptable process, not a rigid, prescribed method. For example, in the use of controls, many biological studies have to be performed where clear-out (single-variable) controls are not practical. Such situations often arise when experiments must be conducted in nature rather than in a laboratory.

Roca’s oilbird study illustrates the adaptability of the scientific process to the varied conditions that a field researcher often encounters. In his seed-collection experiment, for instance, there were actually there variables- three differences between the experimental nests and the control nests. In addition to the presence of absence of nestlings, the nests deferred in age and in the presence or absence of eggs. The nests without nestlings were newer and contained eggs. Multiple variables like these can make results difficult to interpret, but they do not necessarily invalidate scientific tests. In Roca’s case, the fact that no seeds were dropped from nests without nestlings did show that nestlings are the chief, if not the only, source of seeds in the cave.

It is characteristic of scientific work that hypotheses are not proved. Roca did not prove that oilbirds drop seeds in rain forests, instead, he amassed evidence to support his hypothesis. Roca’s results and conclusion, like all scientific knowledge, are considered tentative and open to challenge, reinterpretation, and refinement. Science is a social activity with a self-correcting mechanism. A succession of scientists working on the same or similar problems build on what has been learned earlier. Contemporary scientists share information through publications, meetings, and personal communication. Scientists also critique one another’s work and often check one another’s results and conclusions by repeating observations and experiments. Science is a powerful means of gaining information and understanding, but it does not deal in absolute truths.

An important part of understanding science is realizing is limitations. Science does not deal with hypotheses that are not testable, such as those involving supernatural causes. More over, many questions are unapproachable by science because they are not concerned with the natural world. For example, a scientist would not attempt to answer such questions as, what is the purpose of life? Or What is the nature of the human spirit? The quest for answers to questions like these lies in the realms of philosophy and religion.