There are 119 drugs of known structure that are still extracted from higher plants and used globally in allopathic medicine. About 74% of these were discovered by chemists who were attempting to identify the chemical substances in the plants that were responsible for their medical uses by humans. These 119 plant-derived drugs are produced commercially from less than 90 species of higher plants. Since there are at least 250,000 species of higher plants on earth, it is logical to presume that many more useful drugs will be found in the plant kingdom if the search for these entities is carried out in a logical and systematic manner. The first and most important stage in a drug development programme using plants as the starting material should be the collection and analysis of information on the use(s) of the plant(s) by various indigenous cultures. Ethnobotany, ethnomedicine, folk medicine and traditional medicine can provide information that is useful as a 'pre-screen' to select plants for experimental pharmacological studies. Examples are given to illustrate how data from ethnomedicine can be analysed with the aim of selecting a reasonable number of plants to be tested in bioassay systems that are believed to predict the action of these drugs in humans. The ultimate goal of ethnopharmacology should be to identify drugs to alleviate human illness via a thorough analysis of plants alleged to be useful in human cultures throughout the world. Problems and prospects involved in attaining this goal are discussed.