It is often mentioned that elephants are generalist feeders because they consume a wide variety of plants and plant parts. While this is true when we compare the elephant to a specialized grazer or a frugivorous animal, the elephant also exercises considerable choice in the type of plant it eats both on a short-term as well as on a seasonal basis. Feeding behavior in the short-to-medium term of days or a few weeks essentially involves food selection within one or a few vegetation communities, while the seasonal changes in diet may involve complete shifts to different vegetation communities.
The selection of dietary items obviously depends to a large degree on what is available. In the process of consuming an ideal diet from a natural environment, an elephant has to select from a changing mosaic of different plant species, phenological stages, structural types, chemical compositions, relative or absolute abundances, and dispersion patterns. Because a particular plant item is consumed frequently or in large quantity, it does not necessarily mean that this is a "preferred" food plant. A positive or a negative preference for a particular plant species or part has to be scored in relation to its availability vis-a-vis other plant species. Some observers have looked at the selection of browse in a woodland habitat, while others have evaluated preferences among a variety of herbaceous plants, including grasses in grassland habitat. It is not an easy task to determine feeding preference in habitats that offer a mixture of browse, grass, and forbs as there are practical difficulties in quantifying the relative availability of these plant types.
In one of the early studies of elephant feeding, Peter Guy followed elephants at Sengwa in Zimbabwe, recording in detail the plants they ate and quantities consumed. In addition, he estimated tree densities and canopy volumes to measure availability of potential food plants. Using these data, he calculated three indices of plant selectivity by elephants—one based on tree canopy volumes, another based on tree densities, and the third combining the two measures plus also factoring in plant availability below 6 m height (the maximum reach of an elephant).
Guy evaluated the feeding habits of elephants for selection in two major woodland types, Colophospermum mopane woodland and Acacia tortilis-Grewia riparian community. In the mopane woodland, there was positive selection for only 2 (both Combretum species) of 11 species with a greater than 10% frequency (frequency is defined as percentage occurrence of a species in a sample of sites). The rest were either eaten in proportion to their availability or negatively selected. Several species with very high frequencies and abundances, such as Boscia matabelensis and Commiphora pyracanthoides, were not eaten at all. In the riparian community, 6 out of 17 species with high frequency were positively selected, 4 were not selected, and the rest were negatively selected. Here again, several abundant species were not consumed. Grewia flavescens, a species positively selected in the riparian community, was negatively selected in the mopane woodland. Overall, he concluded that "the majority of species in the woodlands are eaten in quantities proportional to their occurrence" (p. 59), but that elephants also selected for certain plants and avoided others.
While Peter Guy did not look closely at seasonal changes in the elephant's diet, another study in Kidepo Valley National Park, Uganda, published in the same year by C. R. Field and I. C. Ross, documented distinct seasonal differences in plant species and types consumed. Over a period of 3 years (19691971), when their study covered distinct dry and wet months, the most striking observation was a seasonal shift in the proportions of browse and grass in the diet (table 5.2). During the dry months, browsing on trees and shrubs constituted 59% of feeding observations (this went up to 71% when herbs were included as browse). Grazing, on the other hand, increased, on average, from 29% during the dry months to 57% during the wet months.
Studying elephants at Ruaha in Tanzania during the mid-1970s, Richard Barnes observed that grasses comprised 62%-74% of the diet during the wet months, while browse leaves, woody material, and bark constituted 77%-91% of the diet of bulls and 89%-99% of the diet for cows during the dry months.
My study of Asian elephant feeding habits during 1981-1983 in the Bili-girirangans (table 5.2) and later during 1988-1989 in the Nilgiris of southern India showed strikingly similar patterns to the Kidepo study. Within a particular habitat type, the choice of browse or grass was influenced by the availability of these plant types. Thus, as much as 90% of feeding in browse-rich habitats during the dry season (January-April) was on browse. In this habitat, grazing on short grasses picked up only during the second wet season (September-December) when the grasses reached half a meter in height. In deciduous
Proportion of browse and grass during different seasons in the diet of elephants as determined by direct field observations.
African Elephant Asian Elephant
(Kidepo, Uganda) (Biligirirangans, India)
Dry Wet Wet
Plant Type Dry Season Wet Season Season Season 1 Season 2
Sources: Field and Ross (1976) and Sukumar (1985, 1989a).
Elephants in the Nilgiris, India, spent a higher proportion of time grazing during the wet season (see text).
woodlands, about three-fourths of feeding during the first wet season (May-August) was on tall grasses. In grasslands, elephants obviously grazed most of the time they spent there. When data from all habitat types, correcting for seasonal utilization, were pooled, the importance of browse in the dry season diet was very clear. Adult elephants spent, on average, 75% of their time browsing during January-February; this figure dropped below 70% during MarchApril, the later part of the dry season. During the early wet months of May-June, grazing by adults peaked at about 68%, reducing only slightly over the next 2 months. Feeding preferences during the second wet season of September-December were roughly intermediate between the other two seasons.
In seasonal habitats, the switch between a predominantly grass diet during the wet season and a browse diet during the dry period by elephants thus has been established by several observers. Even at Amboseli, where woodlands have been largely transformed to grassland, Keith Lindsay observed that elephants had a distinct tendency to go for browse during the dry period even though their diets otherwise comprised mostly grass and aquatic plants from swamps. This seasonal dietary shift could occur either within a given habitat, as seen in some of the African studies, or through change from one habitat type to another, as seen in parts of southern India. Either way, elephants exercise considerable selectivity for plant types to optimize their energy-nutrient intake over the year, an aspect I discuss below.
Was this article helpful?