3. RESULTS

Food composition

The GGS found to be feeding on different plant parts including the leaves, seeds, flowers and sap of different trees and other vegetation (Table 1).
A total of 1314 minutes (in 10 months) in 42 encounters of feeding observation were taken during the study. A total of 30 different species of plants per se , which included 22 tree species, four climbers, one liana, one paraphyte, one shrub and one cactus species, formed the diet. The maximum duration of feeding was observed on Bauhinia racemosa (19.79%), followed by Tamarindus indica (14.08%),Nothopegia beddomei (9.89%), Strychnos potatorum (7.23%) and Terminalia arjuna  (6.47%; Table 1).
The feeding varied by the food availability in various months. The GGS fed on Terminalia arjuna (January, April and September) andTamarindus indica (May, November and December) in three different months, followed by Bauhinia racemose (February and March), Ficus macrocarpa (February and November) and Syzygium cumini(August, May) in two different months. However, the other species were found fed only once during the study period. The climbers that formed part of the diet were Derris brevipes, Diplocyclos palmatus andCayratia trifolia . The GGS also used shrubs like Hibiscus rosa-chinensis , cacti - Euphorbia trigona, and liana -Entada rheedii for feeding. These species come under 18 different families. Among them, the most preferred family was Fabaceae, with eight species found to be part of the GGS diet. This was followed by Moraceae, four species, and Anacardiaceae, two species.

3.2 Contribution of different vegetation parts

The total feeding incidence across the whole study period was found to be 62. Out of it, 30 incidences were on leaf feeding (48.39%), followed by seed feeding (27.32%), flower feeding (16.13 %), and sap-feeding (8.06 %) (Figure 2. A). The duration of feeding on different items is on seeds (52.12 %), followed by leaves (32.55 %), flowers (10.58 %) and sap (4.76 %) (Figure 2. B).

3.3 Diurnal variation in feeding habit

The feeding recorded for the whole study period was 1314 minutes, including 915 feeding bouts. 42% of the total feeding duration was contributed by forenoon observation and the rest 58% by afternoon observation, sampling hours are equal for both times of the day. when the feeding bout took into consideration forenoon hours, 321 bouts, while the afternoon had around twice the value, 594 bouts. However, the GGS was found to be inactive in a food source for some time before commencing the feeding or after feeding, before going elsewhere in search of a new source or for other activities. In some cases, the GGS was found to be inactive or resting at a food source between periods of intensive feeding. The GGS were not significantly different in the duration of feeding between the forenoon and afternoon (U = 189.5 p =0.319) as well as in the feeding bout (U = 205.50 p =0.542).

3.4 Seasonal variation in diet

The feeding incidence on all food items was maximum in the summer, followed by the North-East and South-West monsoon. Univariate analysis was done in the absence of a correlation between different variables under consideration; there was no effect of season and time of the day on the duration of feeding and the feeding bouts. To check if the animal prefers food items with respect to different seasons, 2 test for association was performed. However, no significant association ( 2= 6.32, p= 0.70), could be observed.
The canonical correspondence analysis (CCA) (Figure 3) to check the association of species and season is also done for selected species based on maximum incidences and duration of feeding and found no significant association. The food species choice of the GGS was found not affected by the seasons, but it is based on the availability of various food resources across different seasons.

Feeding technique

The GGS were found to be highly selective in picking the food items. The selection of food items appears to be done by the smell of the items (Figure 4A). The GGS was found to be handling its food with both mouth and forelimbs (Figure 4B). The GGS goes to the tip of the branch or other areas of the crown of the tree according to the availability of food sources and cuts the pod, fruit, leaves or flower with its mouth (Figure 4C). Occasionally the food item was brought to the mouth with the help of the forelimbs. Holding the fruit in the mouth, the squirrel then moves to the thick horizontal branch to sit firmly, sometimes the feeding started at the place of harvest itself. The squirrel holds the branch with the help of the claws of the hind limbs and the tail hanging down, which gives further balance to the body (Figure 4D). It then gnaws the epicarp of the fruit to extract the seed to be consumed. During feeding, the forelimbs were used effectively to manipulate food items, whether long pods of Bauhinia racemosa or small fruit of