Ecological Archives E096-185-A1

Nitin Sekar, Chia-Lo Lee, and Raman Sukumar. 2015. In the elephant's seed shadow: the prospects of domestic bovids as replacement dispersers of three tropical Asian trees. Ecology 96:20932105.

Appendix A. Feeding trials.

D. indica (chalta) fruit

The details of feeding trials varied amongst seasons. Four elephants, four buffaloes, and four cattle were used for D. indica feeding trial in 2012. The elephants included one male and three females (aged c.11 to 55 years). The four buffaloes included an 8-year old female, a seven-year-old female, and two four-year-old males. The cattle included two 10-year old females and two 5-year old males; one of the males was said to be of a local Duars breed, whereas the others were of a Bhutanese variety, but no documentation was available. Two other cows and another buffalo were also offered D. indica fruit, but refused to eat them. Feeding trials with bovids were conducted between feeding trials for elephants in an effort to control for the time of D. indica fruit harvest.

D. indica fruits (both the freshly fallen, hard and older, soft varieties) were collected from under fourteen trees in Buxa. Each fruit was assigned a number and marked by making small holes in the outer layer of the fruit with a sharp stick shortly prior to the feeding trial; the mass of each fruit was recorded. Fruits from the same trees were used for feeding trials of all animals.

Prior to the trial, animals were not given access to D. indica fruits. Their dung was monitored for D. indica seeds until seeds had not been found for 3 days (Campos-Arceiz 2008, Campos-Arceiz et al. 2012); after this we presumed that the animal’s system no longer contained any D. indica seeds. D. indica fruits were offered to each animal over the course of an hour, after which animals were not allowed to consume D. indica fruits until completion of the trial. Many of the bovids struggled to consume D. indica fruit; two of the cattle and one of the small male buffaloes were unable to eat the harder fruit and ate exclusively the older, softer D. indica fruit. The numbers of fruits consumed and rejected were noted; we also tried to count the number of seeds dropped by each animal while feeding, though this was not always possible with the elephants.

Elephants were accompanied during their patrols, baths, and food collection trips to ensure they did not eat any other D. indica fruit and to collect dung deposited during outings. Elephants did not share space with other individuals, so there was no risk of confusing the dung of different animals. Bovids were restrained for the duration of the trial by their owners, who similarly restrain their animals at times of illness or pregnancy or to prevent crop raiding, thus also preventing mixing of their dung.

Each dung pile was filtered for seeds by washing it with water using a 2-mm wire mesh; the number of seeds in each pile was then recorded and seeds were stored in labeled envelopes for germination trials (see below). For elephants, dung deposited after the feeding trial was searched for seeds until seeds were not detected for 36 hours; for bovids, dung was examined for seeds until seeds were not detected for at least three days and up to a week.

During the trial, elephants maintained their regular diet of banana plants and leaves, rice, and lentils fed to them by their mahouts, and various other items (bread, sweets, fruits) by villagers. They also opportunistically grazed and browsed vegetation while working in the reserve. Buffalo and cattle were fed hay from rice paddy, corn or wheat meal with salt, and browse harvested from some thirteen tree species in and around the village.


A. chaplasha (lator) fruit

Feeding trials for A. chaplasha fruit were conducted over three years. In 2010, trials were conducted with two elephants and one cow; in 2011, with one elephant and four cows; and in 2012 with three elephants, three buffaloes, and three cows (including one of the cows from 2011 for a second trial). The elephants included one male (c. 20 years old) and 5 females (aged c. 13–60 years old), including one very old female who supposedly had few if any remaining teeth. The buffaloes included two young males (c. 4 years old) and a 7-year old female. The cattle included seven female cattle of reproductive age and a male less than five years old; two of the females were considered of the local Duars breed, whereas the other six cattle were said to be Bhutanese breed.

Fruit were drawn from 3, 6, and 11 trees in 2010, 2011, and 2012, respectively. Since whole A. chaplasha fruit was not easily found on the ground, fruit were often harvested from the canopy. A. chaplasha fruit ripened and rotted quite quickly; thus, whichever fruits were ripened were fed to animals on the day of the feeding trial. Thus, fruit allocation to individuals was not random for A. chaplasha as it was with the other fruit species. In addition, due to logistical limitations in 2012, bovid feeding trials were begun three weeks before elephant feeding trials for logistical reasons.

Since A. chaplasha seeds are substantially bigger, searches for seeds in elephant and bovid dung was sometimes conducted by hand, though filters were used most of the time. Different plants were harvested from the jungle to feed bovids and elephants during the feeding trials, in keeping with the season. Otherwise, methods were the same as those used for D. indica feeding trials.


C. arborea (kumbhi) fruit

Feeding trials with C. arborea fruit were conducted in 2012 at the same time as that year’s A. chaplasha feeding trials. Three female elephants (c. 13–56 years old), three buffaloes (same as for A. chaplasha), and three female cattle of reproductive age were utilized for C. arborea feeding trials. Fruit for C. arborea trials were drawn from 11 trees. Since C. arborea trees generally ran out of fruit before many A. chaplasha trees’ fruit had reached maturation, some C. arborea fruit had to be refrigerated before the trials, particularly the elephant feeding trials, which were later. Fruit from each tree were essentially randomly allocated amongst elephant, bovid, and control groups. Otherwise, C. arborea feeding trials followed the same methods as those described for A. chaplasha above.

[Back to E096-185]