Ernst fJJayr Library of Comparative Zoology Harvard University

MC2

LIBRARY

JUL 3 0 2013

HARVARD

UNIVERSITY

JOURNAL

OF THE

BOMBAY NATURAL HISTORY SOCIETY

APRIL 2010 VOL. 107(1)

JOURNAL OF THE BOMBAY NATURAL HISTORY SOCIETY

Hornbill House, Shaheed Bhagat Singh Marg, Mumbai 400 001 .

Executive Editor

Asad R. Rahmani, Ph. D.

Bombay Natural History Society, Mumbai

Copy and Production Editor Vibhuti Dedhia, M. Sc.

Editorial Board

Ajith Kumar, Ph. D.

National Centre for Biological Sciences, GKVK Campus, Hebbal, Bengaluru

Aasheesh Pittie, B Com.

Bird Watchers Society of Andhra Pradesh, Hyderabad

C.R. Babu, Ph. D.

Professor, Centre for Environmental Management of Degraded Ecosystems, University of Delhi, New Delhi

M.K. Chandrashekaran, Ph. D., D. Sc.

Professor, Jawaharlal Nehru Centre for Advanced Scientific Research, Bengaluru

Anwaruddin Choudhury, Ph. D., D. Sc.

The Rhino Foundation for Nature, Guwahati

Indraneil Das, D. Phil.

Institute of Biodiversity and Environmental Conservation, Universiti Malaysia, Sarawak, Malaysia

Y.V. Jhala. Ph. D.

Wildlife Institute of India, Dehradun

K. Ullas Karanth, Ph. D

Wildlife Conservation Society - India Program, Bengaluru, Karnataka

T.C. Narendran, Ph. D., D. Sc.

Professor, Department of Zoology,

University of Calicut, Kerala

G.S. Rawat, Ph. D.

Wildlife Institute of India, Dehradun

K. Rema Devi, Ph. D.

Zoological Survey of India, Chennai

J.S. Singh, Ph. D.

Professor, Banaras Hindu University Varanasi

S. Subramanya, Ph. D.

University of Agricultural Sciences, GKVK, Hebbal, Bengaluru

R. Sukumar, Ph. D.

Professor, Centre for Ecological Sciences, Indian Institute of Science, Bengaluru

Romulus Whitaker, B. Sc.

Madras Reptile Park and Crocodile Bank Trust, Tamil Nadu

S.R. Yadav, Ph. D.

Shivaji University, Kolhapur

Senior Consultant Editor J.C. Daniel, M. Sc.

Consultant Editors Raghunandan Chundawat, Ph. D.

Wildlife Conservation Society, Bengaluru

Nigel Collar, Ph. D.

BirdLife International, UK

Rhys Green, Ph. D.

Royal Society for Protection of Birds, UK

Qamar Qureshi, M. Phil.

Wildlife Institute of India, Dehradun

T.J. Roberts, Ph. D.

World Wildlife Fund - Pakistan

Editorial Assistant: Sonali V. Vadhavkar, M. Sc. Layout and Typesetting: V. Gopi Naidu

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage and retrieval system, without permission in writing from the Bombay Natural History Society (BNHS). Enquiries concerning reproduction outside the scope of the above should be addressed to the Honorary Secretary, BNHS at the address given above.

VOLUME 107(1): APRIL 2010 CONTENTS

HARVAnn

editorial . UNtVETn 1

FEEDING ECOLOGY OF THE ASIAN ELEPHANT ELEPHAS MAXIMUS LINNAEUS IN THE NILGIRI BIOSPHERE RESERVE, SOUTHERN INDIA

N. Baskaran, M. Balasubramanian, S. Swaminathan and Ajay A. Desai . 3

AN ANNOTATED AND ILLUSTRATED CHECKLIST OF THE OPISTHOBRANCH FAUNA OF GULF OF KUTCH, GUJARAT, INDIA, WITH 21 NEW RECORDS FOR GUJARAT AND 13 NEW RECORDS FROM INDIA: PART 1

Deepak Apte, Vishal Bhave and Dishant Parasharya . 14

FISH DIVERSITY, PRODUCTION POTENTIALAND COMMERCIAL FISHERIES OF RAMSAGAR RESERVOIR, DATIA, MADHYA PRADESH, INDIA

R.K. Garg, R.J. Rao and D.N. Saksena . 24

DEMOGRAPHY OF CAPTIVE ASIAN ELEPHANTS ELEPHAS MAXIMUS LINNAEUS IN THREE MANAGEMENT SYSTEMS IN TAMIL NADU, INDIA

V. Vanitha, K. Thiyagesan and N. Baskaran . 30

GERMINATION RATE OF MESQUITE PROSOPIS JUUFLORA SEEDS PASSED THROUGH GUT OF THE INDIAN WILD ASS EQUUS HEMIONUS KHUR IN SALT DESERT OF INDIA

Bitapi C. Sinha, S.P. Goyal and P.R. Krausman . 38

LIFE HISTORY OF ATTACUS ATLAS L. (LEPIDOPTERA: SATURNIIDAE) ON LITSEA MONOPETALA JUSS.

IN NORTH-EAST INDIA

B.N. Sarkar, B.C. Chutia, J. Ghose and A. Barah . 42

NEW DESCRIPTIONS

RECORD OF THE GENUS SCHIZOPRYMNUS FOERSTER (HYMENOPTERA: BRACONIDAE) FROM INDIA,

WITH DESCRIPTIONS OF TWO NEW SPECIES

Zubair Ahmad and Zaheer Ahmed . 45

MISCELLANEOUS NOTES

MAMMALS

1. A note on distribution range of Hanuman Langur Semnopithecus entellus (Dufresne) and Rhesus Macaque Macaca mulatta (Zimmermann) in Rajasthan

Satish Kumar Sharma . 48

2. Sight record of the Indian Wolf Canis lupus pallipes in the

river Gandak floodplains

Sushant Dey, Viveksheel Sagar, Subhasis Dey and

Sunil K. Choudhary . 51

3. Wildlife mortality from vehicular traffic in Sriharikota Island, southern India

S. Sivakumar and Ranjit Manakadan . 53

BIRDS

4. Factors causing nest losses in the Painted Stork Mycterla

leucocephala: a review of some Indian studies

Abdul Jamil Urfi . 55

5. Partial albinism in Black Ibis Pseudibls papillosa

Rajesh C. Senma and Chirag A. Acharya . 58

6. First record: selection of an electric pole as a roosting site by Black Ibis in North Gujarat region

Rajesh C. Senma and Chirag A. Acharya . 59

7. Occurrence of the Great Indian Bustard Ardeotis nigriceps in Bikaner region of the Thar Desert Partap Singh, D.R. Saharan, Jitendar Solanki and S.P. Mehra . 59

8. Addition to the avifauna of the Indian subcontinent - “White-faced” Plover Charadrius dealbatusUom Andaman and Nicobar Islands, India

Nikhil Bhopale . 60

9. First record of the Hume's Leaf-warbler Phylloscopus humel from Kachchh, Gujarat, India

Nikhil Bhopale . 61

FISH

10. An observational note on Gangetic Latia Crossocheilus latius latius in Khoh river, Uttarakhand, India

Vidyadhar Atkore . 62

INSECTS

11. A new record of larval host plant of Tawny Coster Acraea violae (Fabricius)

Rudra Prasad Das, Arjan Basu Roy, Radhanath Polley

and Goutam Saha . 63

12. A checklist of ants of Thirunelli in Wayanad, Kerala K.A. Karmaly, S. Sumesh, T.P. Rabeesh and Lambert Kishore . 64

OTHER INVERTEBRATES

1 3. First report on the occurrence of an economically important Spiral nematode Helicotylenchus multicinctus Cobb, from Goa

I.K. Pai and H.S. Gaur . 68

14. Scolopendra hardwickei (Newport, 1844) feeding on Oligodon taeniolatus (Jerdon, 1853) in the scrub jungles of Pondicherry, southern India

Utpal Smart, Prakash Patel and Pradeep Pattanayak ...

15. Architecture of abutting surfaces of the shells of acorn barnacles

A. A. Karande and M. Udhayakumar .

BOTANY

1 6. Andrachne telephioides L. (Phyllanthaceae) - an addition to the flora of peninsular India

68 M.M. Sardesai and S.Y. Chavan .

Cover Photograph: Tiger

Panthera tigris

70 By Sachin Rai

ACKNOWLEDGEMENT

We are grateful to the Ministry of Science and Technology, Govt of India,

FOR ENHANCED FINANCIAL SUPPORT FOR THE PUBLICATION OF THE JOURNAL.

73

11

Editorial

Are we saving tigers for Chinese consumers?

Tiger is perhaps the most famous animal in the world.

Everything about the tiger is written in superlative terms - its beauty, grace, aura, grandeur, strength, ecological role, iconic role, and even aphrodisiac potency of its body parts. Nothing is mundane about the tiger. Even some tiger conservationists consider themselves above all other conservationists. Earlier every tiger shot was a life-long memory of a hunter, now every tiger sighting by a tourist is a conversation topic among family and friends. Poaching of tigers makes the front-page in daily newspapers. There are more books on tiger than any other Indian animal. There are tourist agencies that survive solely on tiger tourism. Such is the aura of this grand animal, and rightly so. For me, the tiger is a spirit, literally and figuratively, of Indian conservation movement - a flagship species. It is the animal which inspires many of us to save our wilderness.

India has come a long way from the bad old days of tiger shooting as ‘sport’ to tiger tourism as a growing business. We even have a school of art based solely on tiger paintings. Instead of an ugly rug of a tiger skin or a decaying ‘trophy’ of a tiger head in some decrepit house of an aging former rajah or nawab, tiger paintings now proudly adore art galleries and board rooms of corporates. Visiting a tiger reserve is a fashion statement.

India has 39 tiger reserves scattered all over the tiger’s range, covering about 40,000 sq. km of forest. Unfortunately, almost 50 per cent are in very bad shape, but they can be recovered with proper management. According to tiger experts, a male tiger requires about 100-160 sq. km territory and a tigress requires about 40-60 sq. km, which means we should have about 400 adult tigers and about 800 adult tigresses only in the tiger reserves. We also have about 100,000 sq. km forest, which can support tigers, may be in lower densities. In well-protected areas, such as the Corbett National Park, there are 20 adult and subadult tigers of both sexes per 100 sq. km. Similarly, in Kaziranga National Park the density is 26 tigers per 100 sq. km. Therefore, ideally

India should have 2,400 to 3,000 tigers, perhaps more, as in good protected areas (e.g. Corbett, Kaziranga, Bandhavgarh, Ranthambore) tigers can live in much higher densities. We have less than 50 per cent of the rough estimated figures. Thanks to mismanagement, lack of funds and administrative support, and extensive poaching we have vast empty forests where the tiger and its prey have almost gone. Besides, our forests face constant threats of livestock overgrazing, encroachment, and mining.

While mining, livestock grazing and encroachments can be stopped by strong administrative and legal measures, the invidious threat of poaching is much more difficult to control, particularly when the tiger moves out of the protective cover of a tiger reserve or a national park. As long as there is demand in China for tiger parts, tigers will be poached. With 60 per cent of world’s tigers in India, we have become the biggest supplier of tiger parts to the growing Chinese market.

BINGOS (big international NGOs), donors and tiger conservationists frequently go through the ritual meetings and conferences where the issue of protecting tiger through training of staff, giving them more guns and boots (!), getting stakeholders support etc. are discussed on the well- trodden lines, but not many are willing to take up China. As long as we have demand of tiger parts in China, all wild tigers of the world will be under constant threat.

We may have a million children writing to the Prime Minister of India to save the tiger, a retinue of celebrities endorsing tiger protection, large hoardings appealing to save our national animal, but a poacher is not going to listen to this; for him a dead tiger is money. The higher-up you are in illegal tiger trade, the more money you make. And as long as there are people willing to give any amount of money to have tiger-penis soup for purportedly aphrodisiac properties, as long as there are people willing to wear a tiger nail around their neck for good omen, and as long as there are people who consider tiger meat, fat and bones as cure-all, tiger poaching will continue.

Before tiger hunting was banned in 1969, we used to have about 30 shikar companies exclusively for the so-called sport hunting of tiger in India. As one of India’s greatest living conservationists. Dr. M.K. Ranjitsinh tells that in his younger days, when one saw a tiger, it was shot, what else would one do? Now, when we see a tiger, we still shoot, but with a camera. When we Indians can change our way of living in one generation, from tiger hunters to tiger lovers, why can’t the Chinese stop using tiger parts? When they have death penalty for killing a Giant Panda - their conservation symbol - why can't they protect the tiger in their own country and stop smuggling of tiger parts from other tiger-range countries? 1 remember the old slogan of WildAid, “When the

Buying Stops, the Killing Can Too”. This is the basic issue of tiger conservation. When the main problem lies in China, the solution also lies there. We have to see that consumption of tiger parts is stopped in China and other countries through strong legislative and administrative actions and national and international pressure - otherwise we will continue breeding tigers in India, spending crores of rupees and with great sacrifice by the local people (e.g. shifting villages), for the Chinese market. 'Guns and boots’ and well-intentioned petitions cannot save tiger as a free-ranging wild animal in India.

Asad R. Rahmani

2

J. Bombay Nat. Hist. Soc., 107 (1), Jan-Apr 2010

Journal of the Bombay Natural History Society, 107(1), Jan-Apr 2010

3-13

FEEDING ECOLOGY OF THE ASIAN ELEPHANT ELEPHAS MAXIMUS LINNAEUS IN THE NILGIRI BIOSPHERE RESERVE, SOUTHERN INDIA

N. Baskaran12, M. Balasubramanian13, S. Swaminathan4 and Ajay A. Desai15

'Bombay Natural History Society, Hombill House, Dr. Salim Ali Chowk, S B. Singh Road, Mumbai 400 001, Maharashtra, India. 2Present address: Asian Nature Conservation Foundation, Innovation Centre, First Floor, Indian Institute of Science,

Bengaluru 560 012, Karnataka, India. Email: nagarajan.baskaran@gmail.com

^Present address: The Periyar Foundation, Thekkady 685 536, Kerala, India. Email: wildbala@gmail.com 4No. 5, Perumal Kovil Street. Porayar 609 307, Tamil Nadu, India.

584 BC Camp, Belgaum 590 001, Karnataka, India. Email: ajaydesai.l@gmail.com

We studied the activity patterns and feeding ecology of Asian Elephants Elephas maximus in deciduous and dry thorn forests of the Nilgiri Biosphere Reserve, southern India. Over 20,000 instantaneous scan samplings on elephants revealed that 60% of the daylight hours were devoted to feeding. Feeding patterns were strongly bimodal, with peaks in the morning and evening. Elephants spent less time feeding during the dry season than in the wet season, both in dry deciduous and dry thorn forests. Feeding decreased with increasing ambient temperature and its influence is more pronounced during the dry season in all the habitats. The time spent on feeding was less in dry thorn (53%) than in dry deciduous forests (68%), attributed to higher ambient temperatures coupled with poor shade availability and higher human disturbances in dry thorn forest. The diet of elephants constituted more species of browse (59) than grass (29), but grass formed the bulk of the annual diet (84.6%) than browse (15.4%). Elephants fed on more diverse food plants during the dry than the two wet seasons, and in the dry thorn than dry deciduous forests, which is discussed in the light of availability of grass biomass. The proportion of browsing was significantly more during the dry season in dry thorn forest, coinciding with poor availability of grass. These observations indicate that grass forms the principal diet of elephants in this area.

Key words: Asian elephant, Elephas maximus, activity, feeding, seasonal variation, temperature, browse, grass biomass

INTRODUCTION

Both living species of proboscideans, the Asian Elephant Elephas maximus and African Elephant Loxodonta africana, are well adapted to living in diverse habitats by exploiting a wide spectrum of plant species. Their physiological adaptations, like the large prehensile trunk, dentition and digestive system, which help to collect and process vast quantities of diverse plant food required to compensate for an extremely poor digestive ability and the nutritional demands of the elephant’s large body mass, are undoubtedly critical to the survival of the species (Sukumar 2003). However, such physiological adaptations alone are unlikely to be sufficient, especially in tropical ecosystems, which show large spatio-temporal variance in climate, and food quality and quantity. Additional behavioural adaptations may also be necessary for both the species to efficiently exploit the highly changing heterogeneous tropical environments.

The Nilgiri Biosphere Reserve (NBR) in southern India, along with its adjoining contiguous areas in the Western and Eastern Ghats, supports the largest elephant population in Asia (Daniel el al. 1995). The Reserve encompasses a wide range of habitats ranging from semi-evergreen to tropical dry thorn forests and shows distinct seasonality - dry versus wet - making it an ideal system to study the effects of the spatial and environmental factors on the activity and feeding

behaviour of the Asian Elephant. This paper documents the seasonal influences of ambient temperature and the availability of grass on the activity pattern and feeding behaviour of elephants in the tropical deciduous and dry thorn forests of NBR. Though the study was carried out over a decade back ( 1992-95), the findings are still important as there exist no detailed published data on the feeding ecology of elephants from optimal habitats (like Mudumalai, Bandipur, Nagarahole and Wayanad) of NBR. which support the major population of elephants in southern India. Additionally, it would provide baseline data to know the impact of the recent changes taking place on the vegetation physiognomy of elephant habitats due to proliferation of exotic weeds like Lantana camara and Eupatorium odoratum and the reported decline of preferred food plant species (Sivaganesan and Sathyanarayana 1995), and their impact on elephant feeding.

STUDY AREA

Nilgiri Biosphere Reserve ( 12° 15'- 10" 45' N;76°0'-77° 15' E), spread over an area of 5,520 sq. km is situated at the junction of three southern states — Tamil Nadu, Karnataka and Kerala. It has an undulating terrain with an average elevation of 1 ,000 m above msl. Rivers such as Nugu. Moyar and Bhavani, and most of their tributaries, are perennial and drain the area. The Reserve has a diverse climate due to its

FEEDING ECOLOGY OF THE ASIAN ELEPHANT IN THE NILGIRI BIOSPHERE RESERVE

varied reliefs and topography. The temperature ranges from 7°C in December to 37°C in April, and receives rainfall both from the Southwest (May to August) and Northeast (September to December) monsoons. The mean annual rainfall varies from 600 (in the eastern side) to 2,000 mm (in the western side). The dry season is from January to April. Corresponding to the gradient in rainfall, the vegetation varies from southern tropical dry thorn forest in the east to moist deciduous forest in the west with dry deciduous forest in between the two forest types (Champion and Seth 1 968). NBR along with its adjoining natural habitats has remarkable faunal diversity and is well-known for supporting the largest population of Asian elephants with an estimated population of 5,750 individuals (Project Elephant 2007). Overgrazing by domestic cattle and firewood collection are serious problems in the eastern fringes of NBR (Baskaran etal. 2004).

METHODS

Grass biomass

The abundance of grass, in terms of biomass, was estimated twice in a season for three seasons from stratified transects of one to two kilometres in dry deciduous (7 transects of total length of 10 km) and dry thorn forest (6 transects of total length of 10 km). The grass biomass could not be assessed in moist deciduous forest due to inadequate manpower. At 200 m intervals along these transects, two 1 sq. m quadrats were placed at a 5 m distance on either side of the transect. All the grass species were clipped at the ground level from each quadrat and weighed to estimate the grass biomass (wet weight). The biomass estimates using dry weight is more appropriate than wet weight method, due to varied water content in plant samples in different season. However, given the manpower and infrastructure facilities, dry weight method could not be used. Mean grass biomass for grazed and un¬ grazed (by domestic cattle) areas for each habitat was also estimated, as there were remarkable differences in grazing pressure across habitats. All transect were restricted to areas where direct observations on feeding of elephants was carried out.

Activity and feeding behaviour

Observations were made on elephant clans and bulls using instantaneous scan sampling method (Altmann 1974). Using radio-collared elephant clans and bull, a minimum of two clans and a bull were observed for a period of 2 days/ month. Non-collared elephant clans and bulls were also observed, especially during months when radio-collared elephants were not recorded within a habitat. Daylight hours from 06:00 to 18:00 hrs were divided into 1 2 one-hour blocks

for sampling and an attempt was made to sample each one- hour block at least once a month. Scan sampling was made at 15-minute intervals (four scans per hour) presuming that this interval would rule out over-sampling of any particular behaviour. Observations were made on foot (ground) or from a tree, depending on the topography, wind direction and visibility. Care was taken to ensure that the target animal or target group did not detect the observer’s presence. During the sampling, animals were systematically scanned and information such as age, sex and activity (feeding, resting, moving and others) were recorded. If the animal was feeding, data on plant species eaten was also recorded. Additionally, the ambient temperature was recorded at every 30-minute intervals using digital thermometer at the observation site.

Data analyses

The frequency of activities and plant species eaten was estimated season-wise for each habitat. The data blocks in the morning (06:00-08:00 hrs) and evening ( 16:00-18:00 hrs) were less compared to other sample blocks primarily due to delay in radio-locating the animals because of weather conditions (mist, rain, etc.) and the remoteness of certain areas. Since the activity of elephants changes according to daylight hours (McKay 1973), any bias in observation at particular hours of the day would result in over- or under-estimation of a particular activity. To standardize such bias, the percentages of various activities/hour was derived from observed hourly- pooled data, and from this percentage, the mean time spent on various activities (weighted average) was calculated for the season. Data on activity pattern and grass, and browse ratio collected from the radio-collared tuskless bull, a habitual crop raider, were not included into the analysis, as its activities and feeding habits were skewed due to crop raiding behaviour. However, its data on food species eaten were included into the analysis mainly to capture the wide spectrum of food species eaten by elephants in this area. All the data were analyzed using non-parametric statistical tests and analyses were done using 'Statistical Package for Social Studies’ (Norusis 1990). Kruskal- Wallis’ one-way ANOVA and the Man-Whitney U tests were used to test the differences in activity pattern. Chi-square analysis was used to test the differences in the selected browsing and grazing plant species. The relationship between ambient temperature and activities (feeding and resting) was tested using Spearman Rank Correlation.

RESULTS

Overall time activity pattern

Overall, during daylight hours, elephants showed two peaks in feeding, one in the morning (06:00-09:00 hrs) and

4

1 Bombay Nat. Hist. Soc., 107 (1), Jan-Apr 2010

FEEDING ECOLOGY OF THE ASIAN ELEPHANT IN THE NILGIRI BIOSPHERE RESERVE

another in the evening ( 1 5:00- 1 7:00 hrs) (Fig. 1 a). Time spent on resting was more around midday than in mornings and evenings. Elephants frequently engaged in other activities such as mud-bath, sand-bath, salt-licking and play during 14:00-16:00 hrs. As the temperature increased from morning with a peak between 12:00 and 13:00 hrs, resting became more common. However, comparisons of feeding and resting with ambient temperature, with pooled data over habitats and seasons, showed no significant correlation. Overall, the activity budget revealed that elephants spent 60% of the daylight hours (06:00-18:00 hrs) on feeding and 20% on resting. Time spent on moving was 14% and 6% on other activities

Seasonal difference in time activity in different habitats

Dry deciduous forest: During the dry season, elephants showed a bimodal feeding activity with a peak each at 07:00 hrs and 18:00 hrs in dry deciduous forests (Fig. lb). Elephants mainly rested during midday between 1 1 :00 and 14:00 hrs. Feeding decreased significantly with increasing ambient temperature (r = - 0.7671, df= 12, P = 0.01), while resting increased positively (r = 0.8581, df = 12, P = 0.01 ). Movement was mostly restricted to the mornings and evenings. Unlike the dry season, elephants spent a minimum of 50% of time on feeding in all the hours of day during the first wet season, and resting being considerably less (Fig. 1 c). Feeding and resting showed no significant correlation with temperature during the first wet season, as the ambient

Table 1 : Time spent (%) in various activities by elephants in the different habitats in Nilgiri Biosphere Reserve

Habitat and	Season	Annual
Activity	Dry	First	Second
		wet	wet
Dry deciduous	(n = 4603)	(n = 3310) (n = 3203)	(n= 11,116)
Feeding	59.55	72.25	72.16	67.99
Moving	11.54	11.06	12.06	11.55
Resting	24.49	12.02	10.57	15.69
Others	4.42	4.66	5.2	4.76
Moist deciduous	in 00 n e	(n = 221)	(n = 0)	(n= 256)
Feeding	22.2	60.0	-	41.10
Moving	33.12	26.37	-	29.74
Resting	31.21	9.23	-	20.22
Others	13.46	4.4	-	8.93
Dry thorn	(n = 2715)	(n = 819)	( n = 5562)	(n = 9096)
Feeding	47.08	57.63	52.35	52.35
Moving	17.21	12.14	15.46	14.94
Resting	29.55	13.77	24.26	22.33
Others	6.16	16.45	7.92	10.18

temperature during this season was relatively lower than the dry season. During the second wet season, the pattern of elephant activities observed was similar to the first wet season (Fig. Id), but resting positively increased with temperature (r = 0.5874, df - 12, P = 0.04), as ambient temperature increased gradually in this season unlike the first wet season.

Activity budget data show that in dry deciduous forest, elephants spent a major part (68%) of the annual daylight hours feeding (Table 1 ). However, time spent on feeding and resting varied among the three seasons. During the dry season, elephants fed for significantly less time than the first (M-W U = 14475, P = 0.01 ) and the second (M-WU= 14503, P = 0.01) wet seasons. Time spent on resting was significantly more during the dry season than the first (M-W U = 15402, P = 0.01) and second (M-W U=14864.5, P = 0.01) wet seasons.

Moist deciduous forest: In moist deciduous forest, the activity pattern shown (Fig. le) for the first wet season was based on a small number of observations (n = 221) collected over a short period of three days in a disturbed area around human settlements, and may therefore not accurately represent a picture for the entire season. Similarly, as the observations made on elephants were limited during dry season ( n = 35) and nil during second wet season, the time activity pattern of elephants could not be constructed.

Dry thorn forest: The pattern of elephant feeding and resting observed in thorn forest during the dry season was similar to the pattern observed in dry deciduous forest (Fig. lf-h), but there was a sharp rise in time spent on movement between 11:00 and 12:00 hrs. The peak temperature recorded during midday hours coincided with peak resting time. Resting increased positively with temperature (r= 0.7273, df = 11, P = 0.01), while feeding decreased (r = - 0.7091, df = 11, P = 0.01). During the first and second wet seasons, the activities observed among elephants were similar, except for an unusually longer time (>55%) spent in resting in the morning hours (06:00- 07:00 hrs) observed during second wet season (November and December), which is similar to that observed in the early dry season (January). No significant correlation was observed between ambient temperature and feeding, and resting during first and second wet seasons.

Data on activity budget showed that annually, elephants in thorn forest devoted significantly less time for feeding and more time for resting compared to dry deciduous forest (Table 1 ). On a seasonal basis, elephants in thorn forest also spent significantly less time on feeding (M-W U = 3838, P = 0.03) and more on resting during the dry season than the first wet season (M-W U = 2936, P = 0.01). The time spent on various activities did not vary much between the dry and

J. Bombay Nat. Hist. Soc., 107 (1), Jan-Apr 2010

5

FEEDING ECOLOGY OF THE ASIAN ELEPHANT IN THE NILGIRI BIOSPHERE RESERVE

(a) Overall (n = 20,468)

34

o

30 *

=

26 * Cj

77

“ S

Cj

18 ^

Hours

DUD Feeding I I Moving ^HRestiug ^3 Others -Q-Temperature

(c) Dry deciduous: first wet season (n =3310)

34 u'

o

30 «’

26 * Oj

V ffu

-- s

a j

18 *-

Horns

DIIIIIFeedmg I I Moving IB Resting ^3 Otliers Temperature

100

5 60

eu

" 40

a>

■ 20

(b) Dry deciduous: dry season (« = 4603)

ri m

9 10 11 12 13 14 15 16 lr

Hours

(Ml Feeding CZ] Moving Ml Resting E3 Others — j-Temperature

(d) Dry deciduous: second wet season (n = 3203)

34 o

o

30 Z

26 « a>

77

e

a>

18

llllllllFepdino I I Moving

Hours

IRestme

I Others —O-Temperature

(e) Moist deciduous: first wet season (« = 221)

(f) Dry thorn: dry season (« = 2715)

28

o

26 * 24 *

22 “

20

11 12 13 14 15 16

Hours

EIID Feeding I I Moving Hi Resting CS Others — j-Temperature

(g) Dn thorn: first wet season (« = 819)

34 o

O

30 Z

26 «

0>

-- g a>

18

Hours

mnn Feeding I 1 Moving ■■Resting ^3 Others — Q-Temperature

_ 100

£ 80

5 60 eu

4/1 40

3 20

	i	1	J	n	1		H		4
	1		1	it	J	II	11

34 cj

o

30 Z

s

26 «

11 5 18

6 7

9 10 11 12 13 14 15 16 17

Hours

IIMD Feeding I I Moving HH Resting i— i Others —o-Temperature

(h) Dry thorn: second wet season (« = 5562)

_100 £ 80 S 60 J* 40 s 20

E-h

0

l	■i	\\	ft	1 1	M		: I
	l'	\	1 1

24 *

=

20 «

Of

16 g 0>

12 M

9 10 11 12 13 14 15 16 17

Hours

mnn Feeding I I Movin g HI Resting E3 Others -3-Temperature

Fig. 1 : Season-wise diurnal activity pattern of elephants in different habitats of Nilgiri Biosphere Reserve

6

J. Bombay Nat. Hist. Soc., 107 (1), Jan-Apr 2010

FEEDING ECOLOGY OF THE ASIAN ELEPHANT IN THE NILGIRI BIOSPHERE RESERVE

second wet seasons, but there were significant variations in resting (M-W U = 4581, P = 0.01) and moving (M-W U = 5500, P = 0.01 ) between the wet seasons.

Grass biomass

In dry deciduous forest, mean grass biomass varied significantly across the three seasons (K-W-/2 = 32.1122, P = 0.0001 ) (Table 2). The biomass was significantly higher during the second wet season (921 gm/m2) as compared to the dry (573.9 gm/m2, M-W U = 1 838.5, P = 0.000 1 ) and the first wet (618.1 gm/m2, M-W U = 3033, f’ = 0.0014) seasons, and in the first wet season as compared to the dry season (M-W U=2039. 5, P- 0.0002). Similarly, in thorn forest, grass biomass varied significantly across the three seasons (K-W- X2 = 1 02.46, P = 0.000 1 ), and was significantly higher during the second wet season (524.1 gm/m2) than the dry ( 1 56.9 gm/m2. M-WU = 781.5,/> = 0.000 1 ) and the first wet (405 gm/m2. M-W U = 2263.5. P = 0.003) seasons. The grass biomass in the first wet season was also significantly more than in the dry season (M-W U= 1088, P = 0.0001 ). Sampling was not carried out in moist deciduous forest due to manpower constraints as mentioned under methods. The observed variation in biomass between dry and wet seasons could marginally be due to variation in water content in grass samples.

The areas under cattle grazing had significantly lower grass biomass in the dry deciduous forest during the dry season (un-grazed = 725 gm/m2 and grazed =188 gm/m2, M-W U = 220, P - 0.0002) and in second wet season (ungrazed = 1019 gm/m2 and grazed = 520 gm/m2, M-W U = 388.5, P = 0.0016). However, the influence of grazing was statistically insignificant in dry deciduous during the first wet season (un-grazed = 677gm/nr and grazed = 600 gm/m2, M-W U = 51 1 , P - 0.10), and in all the seasons in dry thorn forest (dry season ungrazed = 190 gm/m2 and grazed = 152 gm/m2, M-W U = 318.5, P = 0.23; first wet: ungrazed = 420 gm/m2 and grazed = 390 gm/m2, M-W U = 426.5, P = 0.34; second wet season ungrazed

= 528 gm/m2 and grazed = 480 gm/m2, M-W U = 238.5, P = 0.69).

Browse and grass ratio in the diet

Out of 10,743 feeding observations (viz., 7,003 in dry deciduous, 153 in moist deciduous and 3,587 in dry thorn forest), grazing and browsing constituted 84.6% and 15.4%, respectively. Grass dominated the diet of elephants during all the seasons in dry deciduous and dry thorn forests, indicating the importance of grass in the diet of elephants in this region. Browsing was more during the dry season in dry deciduous (15.1%) and dry thorn (47. 1%) forests than during the wet seasons (Table 2). The percentage of grazing and browsing varied significantly across seasons in dry deciduous (X2 = 148.64, df - 2, P - 0.00001) and dry thorn forests (yf- 554.24, df= 2, P = 0.00001). Elephants fed significantly more on grass and less on browse in dry deciduous than in dry thorn forest in all the seasons (dry season - X2 - 459.43, df = 1, P = 0.00001; first wet season - %2 = 6.37, df - 1, P = 0.01 and second wet season - %2 = 65.71, df - 1, P = 0.00001), indicating the importance of grass in dry deciduous forest.

Species composition in the diet

Overall, 83 plant species eaten by elephants were recorded from 11,186 feeding scans. Feeding scan observations (n = 443) made on the habitual crop raiding bull were also included in this analysis to know the diversity of food plants eaten by elephants. Of the 83 plant species, 59 were browse species (trees, shrubs, herbs and bamboo), and the rest (24) were grass species (Appendix 1). Among the 24 grass species, six constituted more than 75% of the total diet ( Themeda cyrnbaria 39.5%, Heteropogon contortus 13.4%, Themeda triandra 10.9%, Bothriochloa sp. 7.3%, Aristida adscensionis 2.4% and Cymbopogon flexuosus 2.3%). Among the 59 browse species, Acacia intsia , bamboo spp. and Kydia calycina were the most important, and contributed 5.4, 4.4 and 1 .8%, respectively to the total diet.

Table 2: Grass biomass (gm/sq. m) and grass: browse ratio in the diet of elephants in dry deciduous and dry thorn forests of Nilgiri Biosphere Reserve (grass biomass not assessed in moist deciduous forest due to inadequate manpower)

Season		Dry deciduous	Moist deciduous		Dry thorn
	Grass biomass/m2 (n = 254)	Grass: browse ratio (n = 7003)	Grass: browse ratio (n= 153)	Grass biomass/m2 Grass: browse ratio (n= 251) (n = 3587)
Dry	573.9		85: 15	78: 22	156.9	53: 47
First Wet	618.1		92: 8	3169	405.0	89: 11
Second Wet	921.0		95: 5	-	524.1	88: 12
Annual	720.2		91: 9	54: 46	352.0	74: 26

J. Bombay Nat. Hist. Soc., 107 (1), Jan-Apr 2010

7

FEEDING ECOLOGY OF THE ASIAN ELEPHANT IN THE NILGIRI BIOSPHERE RESERVE

In dry deciduous forest, 36 species of food plants were recorded from 7,003 feeding observations (Appendix 1 ). The number of grass species eaten (13) was less than browse species (23). The tall grass T. cyrnbaria alone contributed 62.8% of the diet and T. triandra 17.1%, other grass species formed <5%. Bamboo (4.4%) and K. calycina (2.9%) were the two major browse species (Table 3). Seasonal use of these food plants varied considerably, but the tall grass T. cyrnbaria was always the principal diet during all the seasons (Table 3). The proportion of the top four species (T. cyrnbaria, T. Triandra, bamboo and C. flexuosus) and the rest of the browse and grass species (pooled separately as other browse

Table 3: Major food species eaten (%) by elephants in different habitats in Nilgiri Biosphere Reserve

Food species

Season

Annual

Dry

First

wet

Second

wet

Dry deciduous

(n = 2510)

(n = 2236) (n = 2257)

(n =