ECOLOGICAL CHEMISTRY OF PRIMATE FOOD CHOICE: I. THE INFLUENCE OF PLANT CHEMISTRY ON RHESUS MONKEY FOOD CHOICE; II. COMPARATIVE METHODS IN ECOLOGICAL CHEMISTRY (TANNINS, NUTRITION, PHYTOCHEMISTRY)

摘要

Primates must select a nutritionally balanced diet while minimizing their intake of plant secondary compounds which are toxic or inhibit digestion. In previous studies, the most important chemical factors appeared to be selection for protein and avoidance of tannins. Feeding behavior of wild rhesus monkeys (Macaca mulatta) in the Himalayan foothills of Northern Pakistan was quantified by researchers from Yale University. Food and nonfood plant items were identified and collected. Chemical analysis of these items showed that the major foods of the rhesus were low in tannins and high in nutrients. Some of these important food items (for example Trifolium sp.) contained most essential amino acids in amounts sufficient to fulfill the amino acid needs of primates. Foods of the rhesus were significantly lower in tannins than non-food items. Protein binding by tannins (astringency) of plant extracts (n = 91) were highly positively correlated with ellagitannin, gallotannin and extractable condensed tannin contents for samples preserved dry (n = 54) but not for samples preserved in alcohol (n = 37). Among a subset of samples (n = 38), foods of the rhesus were significantly higher in glucose than nonfood items although total soluble carbohydrates were not different between eaten and uneaten samples. Protein, water, fiber, lignin and alkaloid content apparently did not influence food choice. Certain items (notably 3 Araceae samples) were not eaten although they were high in proteins and amino acids and apparently contained no typical tannins, alkaloids or cyanogenic glycosides. When these 3 items were removed from the data set, free amino acids were significantly higher in eaten than unpreferred items.;A new method of protein analysis of plant samples was developed in which samples were acid hydrolyzed and the amino acids released measured with ninhydrin. Protein content as measured by this method was highly positively correlated with total protein estimated by nitrogen content. Unlike other conventional methods of measurement of protein content, the ninhydrin method does not react with tannins and can be used with dried plant samples.