Subject: WII V Batch MSc dissertation abstracts Hello! I recently received a copy of the abstracts of the M.Sc. dissertations of the V batch of WII students. An interesting range of topics from habitat, community, pollination, to human ecology, using using field experiments (fertlise/cut/burn experiments in grasslands of north India and Nepal, and flower exclusion experiments), quantitative ecology, and social science approaches.=20 The taxa span hoolock gibbons and monal pheasants, butterflies and birds, plants and Indian gazelle to, of course, The Deer and the Tiger. In community ecology, I found the studies by Manoj Nair (butterflies in W Ghats) and R Jeyapaul (birds in Central India) particularly interesting.=20 The abstracts of all (11) studies are given below for your information.=20 Regards Shankar Raman --------------------------------------------------------------------- HABITAT OCCUPANCY BY WILD UNGULATES IN PENCH TIGER RESERVE, MADHYA PRADESH.=20 BY B. Bhaskar Acharya. SUPERVISOR: Dr. K. Sankar. SUMMARY This study on the habitat occupancy by wild ungulates was carried out from= =20 November 1996 to April 1997 in the Pench Tiger Reserve, Madhya Pradesh.=20 The spatial and seasonal difference in habitat occupancy of the major=20 forest types by the wild ungulates viz. chital (Axis axis), sambar (Cervus= =20 unicolor), gaur (Bos gaurus), nilgai (Boselaphus tragocamelus) were=20 assessed.=20 The study area was stratified into three major forest types based on=20 floristics and physiognomy as: Teak dominant forest, Anogeissus-Boswellia= =20 mixed forest and Miscellaneous forest. A total of 12 transects were laid=20 in three forest types. Data was collected on ungulate densities, dung=20 abundance and habitat parameters. The Line transect Method was used for=20 estimating ungulate densities. Dung abundance was quantified using a belt= =20 transect, at each 200 m point on the transects. Habitat parameters were=20 also quantified at these points. The relationship between habitat factors and the distribution of ungulates= =20 were examined using bivariate and multivariate analyses.=20 The results showed that chital densities were significantly lower in Teak= =20 dominant forest type in winter, and in the Anogeissus-Boswellia mixed=20 forest type in summer. Both Teak and Miscellaneous forest types showed=20 greater chital densities in summer than in winter. Sambar densities showed= =20 no significant differences between forest types both in winter and in=20 summer. Anogeissus-Boswellia mixed forest type showed a significant=20 increase in sambar density from winter to summer. Chital dung abundance showed no significant differences between forest=20 types, both in winter and in summer, while sambar dung abundance in the=20 Anogeissus-Boswellia mixed forest type was significantly higher than other= =20 forest types in both seasons. There were significant seasonal differences= =20 in chital dung abundance in all forest types, whereas sambar dung=20 abundance increased only in the miscellaneous forest type.=20 Since most of the habitat variables were interrelated, the variables were= =20 factorised and then regressed against dung. The regression showed that in= =20 winter, chital dung was negatively correlated with cover, and positively=20 correlated with grass. Tree factor was positively correlated both in=20 winter and summer. Slope was negatively correlated with chital dung in=20 summer. Grass height, which was significantly higher in winter, was=20 associated with cover only in winter.=20 For sambar, ground cover factor was positively correlated and tree factor= =20 was negatively correlated in winter. In summer, ground cover, tree factor= =20 and distance to water source factor were negatively correlated, and slope= =20 was positively correlated with sambar dung. Sambar dung abundance was=20 higher in the burnt areas, where the grass cover had been totally removed= =20 by fire. A maximum variability of about 16% (R=D7=3D0.1659) for sambar dung abundanc= e,=20 and about 13% (R=D7=3D0.1659) for chital dung abundance, could be explained= by=20 the different factors by regressing dung against factors. ----------------- MOVEMENT PATTERNS OF RADIO TAGGED TIGERS IN PANNA NATIONAL PARK, MADHYA=20 PRADESH BY ABI TAMIM F. VANAK SUPERVISOR DR. R.S. CHUNDAWAT SUMMARY I studied the movement patterns and home ranges of two radio tagged tigers= =20 (an adult male and an adult female with three cubs), in a dry deciduous=20 habitat. The male was radio tracked for five and a half months over two=20 seasons, while the female was tracked for three months during summer. A=20 total of 74 radio locations of the male and 51 locations of the female=20 were obtained. The male had a home range of 237 km=D7 and the female ranged= =20 over 16 km=D7 of area. Movement patterns of the radio tagged tigers were=20 determined by taking consecutive daily locations and averaging the=20 distance between two successive locations to get a rough measure of mean= =20 daily distance travelled. The structure and composition of a habitat play an important role in the=20 behaviour of all animals including large felids. Habitat parameters and=20 relative prey abundance were measured in grids used by the radio tagged=20 tigers to determine the relationship between tiger use and habitat=20 parameters of that grid. It was found that the male used areas with=20 higher cattle dung abundance than the female. The female used areas with=20 higher horizontal cover, grass height and chital dung. A total of 27=20 kills made by tigers were found. Of these, the male killed eight and the=20 female accounted for sixteen and three were made by other tigers. There=20 was a marked difference in the species composition of the kills found made= =20 by two animals. The female killed more sambar than other wild ungulate=20 species while the male subsisted mainly on livestock. =09While the male=20 occupies a home range half the size of the National Park, in direct=20 contrast the female's home range which is presently confined to a small=20 area which has a high wild prey abundance and good cover. This restriction= =20 in movement by the female may be due to her litter of three large cubs.=20 Sustaining them requires a high prey base and therefore she might be=20 confining herself to a resource patch.=20 ------------------ ECOLOGICAL IMPACTS OF PRESCRIBED BURNING IN CORBETT TIGER RESERVE. By: Sonali Ghosh Supervisor: Dr. Vinod B. Mathur SUMMARY This study was aimed at studying the ecological impacts of prescribed=20 burning by evaluating the current practices of burning being carried out= =20 in a typical grassland habitat in Corbett Tiger Reserve . The study=20 extended from November 1996 to April 1997. Twenty four experimental plots= =20 were laid to study the differences in soil and vegetation characteristics,= =20 with respect to burnt and unburnt areas in short ( Arundinella and=20 Vetiveria type) grasslands and tall (Saccharum-Themeda type) grasslands. The main objectives were to determine the impacts of prescribed burning=20 on soil and vegetation characteristics, distribution of animal and birds=20 and to evolve spatial strategies for the management of Hog deer( Cervus=20 porcinus) populations using Remote Sensing and Geographical Information=20 Technology. T-tests and ANOVA (analysis of variance) were applied on the various=20 soil and vegetation parameters to test the differences between the burnt=20 and unburnt areas. Short grasslands showed significant differences in soil available=20 phosphorus, above ground biomass, crude protein, percentage fibre, ash,=20 elemental potassium and sodium. The general increase in the soil=20 available phosphorus, above ground biomass, crude protein and the=20 decrease in the levels of percentage fibre, ash , elemental potassium and= =20 sodium in the short grassland burnt areas can be generally attributed to=20 the removal of litter and the addition of ash which recycles the minerals= =20 back into the system.=20 Tall grasslands showed significant differences in case of soil available=20 potassium and acid detergent fibre. Both the parameters decreased in case= =20 of the burnt areas. The tall grasslands are tussock-forming perennial=20 species which do not get burnt completely. There is not much removal of=20 litter which perhaps is the important factor for increasing the above=20 ground biomass. Samples collected from repeatedly clipped and burnt areas= =20 were tested for crude protein, percentage fibre, ash and lignin. They=20 showed significant differences in case of percentage fibre. The fibre=20 being the maximum in case of these areas.=20 Below ground biomass did not show any difference between the burnt and=20 unburnt areas in short and tall grasslands . The area has had a long fire= =20 history and the below ground biomass has perhaps stabilised over the=20 years and hence cannot be detected in such a short term study. Direct (road transects) and indirect (pellets) evidences were used to=20 determine the animal and bird distribution with respect to the burning=20 regime. Programme DISTANCE was used to estimate the animal densities while= =20 Menhinick=C6s Index was used to calculate the bird species richness ,=20 before and after the date of prescribed burning. The animal densities and= =20 birds species richness increased after burning, but this could not be=20 fully attributed to the burning regime. Other parameters such as change of= =20 seasons, also need to be taken into consideration . Finally, hog deer suitability map was generated using Remote sensing and=20 Geographical Information system and the ground data for various=20 ecological parameters.=20 A burning regime separated on a temporal and spatial scale is required for= =20 sustaining hog deer populations in Corbett Tiger Reserve. ------------------ EFFECTS OF GRAZING, UTILISATION AND MANAGEMENT ON THE GRASSLANDS OF ROYAL= =20 BARDIA NATIONAL PARK, NEPAL. Jhamak B. Karki Supervisor: Dr.Y.V.Jhala=20 SUMMARY I studied the temporal effect of grass harvest, burning, fertiliser, and=20 grazing, with a factorial experimental design, in three grasslands of=20 Royal Bardia National Park, Nepal. I tested the hypothesis of (i) nutrient= =20 depletion of grasslands by continuous harvest and burning and (ii)=20 differential use of areas by ungulates that were differentially managed in= =20 relation to forage quantity and quality.=20 Addition of Di-ammonium phosphate and urea did not result in a significant= =20 increase in green above ground biomass and forage quality. Cut-burnt plots= =20 had the highest nutrient quality as indexed by crude protein and lignin.=20 Plots that were only harvested had highest above ground biomass for the=20 first month following harvest, and were more intensively grazed by=20 ungulates during that time. Ungulates preferred to graze cut-burnt and=20 burnt areas more during the second and third month following the=20 treatment. Unmanaged tall grassland areas were also observed to have=20 nutritive and other wildlife values. A management strategy for the=20 grasslands of Bardia based on the above results is suggested. I compared the community structure, nutritive quality and above ground=20 biomass of "grazing lawns" (patches of short grass communities), with=20 neighbouring grasslands. Grazing lawns differed form neighbouring=20 grasslands in species composition and community structure. A 50 day=20 grazing free environment made areas of grazing lawns indistinguishable=20 from neighbouring grasslands in terms of above ground biomass. Grazing=20 lawns had more nutritive forage in terms of digestibility and crude=20 protein content. Sodium, magnesium and phosphorus were found to be below=20 the critical requirements of ungulates. These minerals were found in=20 higher concentrations in forages from grazing lawns. Grazing lawns seem to= =20 be maintained by continuous grazing and enriched by deposition of urine,=20 dung and by certain plant species not found elsewhere in the neighbouring= =20 grasslands. =20 --------------------- POLLINATOR VISITATION AND REPRODUCTIVE SUCCESS IN TWO SPECIES OF MANGROVE= =20 PLANTS, IN BHITARKANIKA WILDLIFE SANCTUARY, ORISSA. by Shalini Pandit under the supervision of B.C.Choudhury SUMMARY Visitation patterns to the flowers of Sonneratia caseolaris and Aegiceras= =20 corniculatum were investigated between December 1996 and April 1997, in=20 the mangrove forests of Bhitarkanika Wildlife Sanctuary, Orissa. The=20 objectives of the study were to determine the pollination effectiveness of= =20 the different categories of visitors by quantifying their rate of=20 visitation. The effect of environmental variables on visitation was=20 examined, and the importance of the visitors to the reproductive success=20 of the plant was investigated by conducting exclusion experiments (i.e.,=20 bagging flowers to prevent visitation). Nectar was analysed for volume and= =20 sucrose content in S. caseolaris, and the impact of predation on the=20 reproductive success of the plant was examined.=20 The results of the study indicated that the flowers of both plant species= =20 attracted a wide array of visitors and did not show a specialised=20 relationship with any one visitor species/category. Different categories=20 of visitors were seen to vary in their pollination effectiveness for the=20 two plant species. Environmental variables such as temperature, sun=20 intensity and wind velocity were seen to influence the visitation of=20 Lepidoptera to the greatest extent, and Hymenoptera to a lesser extent.=20 Visitation by birds was found to be independent of the environment=20 variables. The territorial behaviour of purple-rumped sunbirds at the S.=20 caseolaris site was seen to reduce visitation of other birds and of bees=20 to the flowers of this species.=20 Results of the bagging set-ups indicated that there was no difference in=20 the pollinator effectiveness of the nocturnal and diurnal visitors.=20 Reproductive success was not pollinator-limited in either of the two plant= =20 species. In S. caseolaris, predation pressure was the single most=20 important factor limiting fruit-set, while in A. corniculatum, fruit-set=20 is probably resource-limited. =20 -------------- IMPACT OF TEAK PLANTATIONS ON FOREST BUTTERFLY COMMUNITIES IN=20 PARAMBIKULAM, SOUTHERN WESTERN GHATS, KERALA. MANOJ.V.NAIR SUPERVISOR : DR. A.J.T. JOHNSINGH=20 I studied butterfly communities in natural and man-modified habitats at=20 Parambikulam Wildlife Sanctuary, Kerala, from December 1996 to April 1997= =20 to understand the impact of a long history of habitat conversion on forest= =20 butterfly communities. Community attributes were investigated in evergreen= =20 and moist-deciduous forests (EGF and MDF respectively, together comprising= =20 natural habitats) and a gradient of teak plantations (TKPs) of different=20 ages (14,30,54 and 74 year old TKPs, together comprising altered=20 habitats). Regular line transect and time-effort methods showed that=20 butterfly species richness, abundance and diversity were significantly=20 higher in evergreen forests. Interestingly, moist-deciduous forest was=20 found to be poorer in all these attributes, when compared to the two=20 younger TKPs. In an overall teak age gradient,the trend showed a clear=20 negative correlation of all community attributes with increasing age of=20 teak (14 > 30 > 54 > 74). The two younger plantations, particularly the=20 14 year old one, were untended ones with good secondary growth, and were=20 ecologically and structurally more akin to natural forests than to=20 classical monocultures, and was probably the reason for the high levels of= =20 richness and diversity.=20 A drastic seasonal shift in butterfly abundance and consequent changes in= =20 community composition was observed across cool and dry seasons in both=20 natural and modified habitats. In all strata except EGF,abundance and=20 species richness were significantly higher in the cool season than dry=20 season, while EGF showed exactly an opposite pattern. It seems that flower= =20 abundance might be playing an important role in determining butterfly=20 abundances and showed high correlation with butterfly abundance and=20 richness in all strata. Among the TKPs, mean and cumulative plant species= =20 richness was also negatively correlated with the age of the plantation. = =20 =20 Small scale adult movement was observed within and between spatially=20 proximate habitats across seasons. Data from transects, supplemented by ad= =20 libitum observations seem to indicate that there is a movement of=20 generalist open-country species from MDF and TKPs to EGF during the dry=20 season, while there may be a reverse movement of specialist evergreen=20 species towards other habitats during cool season and brief spells of cool= =20 weather following rains. An analysis of guilds showed that EGF fares much= =20 better vis a vis other strata. Unique species, habitat specialists,=20 uncommon and rare species and endemic species are all higher in, and in=20 some cases restricted, to EGF.=20 To sum up, habitat conversion from natural forests to TKPs has definitely= =20 had a negative influence on forest butterfly communities. It is likely=20 that generalist species of natural forests, particularly MDF might not=20 have been impacted much; perhaps in some cases, even benefitted. Evergreen= =20 species, on the other hand seem to have taken the brunt of habitat=20 alteration. EGFs assume highest conservation priority, when viewed with=20 regard to communities, guilds or individual species. Old growth TKPs with= =20 a long history of successive silvicultural practices,though structurally=20 similar to natural forests, support very poor levels of butterfly=20 diversity. Younger untended ones, in contrast is significantly richer,=20 albeit composed largely of generalist species. This implies that phasing=20 out of some old plantations along with active intervention in others,=20 might help in supporting higher levels of butterfly diversity. However, it= =20 is highly unlikely that levels of diversity similar to natural forests=20 (EGF in particular), can ever be supported by managed forests.It thus=20 follows that, if we are to conserve the available range of butterfly=20 diversity in the Western Ghats, conserving natural vegetation, in=20 particular, wet evergreen forests is a critical pre-requisite.=20 --------------- FOOD HABITS AND FORAGING BEHAVIOUR OF INDIAN GAZELLE (GAZELLA BENNETI) IN= =20 RAJASTHAN DESERT. BY: GANESH KODOTH SUPERVISOR: DR. S.P. GOYAL SUMMARY The Indian gazelle inhabiting the arid regions is studied with respect=20 to their food habits and foraging behaviour. In arid areas the selection= =20 of a habitat by an animal is mostly governed by spatial and temporal=20 distribution and abundance of food items and the food requirements of that= =20 species. This becomes critical especially in an semi-agricultural=20 landscape where the ecosystem is altered without any consideration for the= =20 animal. The study was conducted in a habitat close to human habitation=20 around Jodhpur after the harvest. This area is dominated by desert=20 vegetation such as Crotalaria burhia. The study is conducted for a period= =20 of six months encompassing two seasons- winter (December 15, 1996 -=20 February 20, 1997) and summer (March 20, 1997 - May 10, 1997). The intensive study area was differentiated into six habitat types based=20 on the structural and floristic composition of the vegetation-viz, crop=20 fields, fallow lands, scrub area dominated by Zizyphus, and hedges of=20 Maytenus-Capparis-Zizyphus among the agricultural fields. Differential=20 habitat selection by gazelle was observed during summer and winter mainly= =20 due to selection of diet based on seasonal nutritional requirements. Diet selection is looked in terms of use-availability and plant chemistry.= =20 Based on a relative preference ratio, the habitat and food types were=20 ranked, and the observed trends are explained in terms of the existing=20 theories on diet selection. Diet composition varies with the age and sex= =20 classes of the animal viz. breeding males, sub-adult males and adult=20 females. In winter and summer differences in habitat use was exhibited by= =20 gazelle, also there was a difference in proportion of the food species in= =20 the diet between two seasons. The scrub area was dominated by vegetation= =20 having higher crude protein and tannin, was used more than available in=20 order to supplement their diet (especially in the case of females) and in= =20 summer the crop fields dominated by Crotelaria burhia were used more by=20 all categories of animals. Females fed on a high protein and less tannin=20 diets which is pronounced in winter, sub-adults also fed on a diet having= =20 higher protein and tannin, while males were not biased towards a high=20 quality food both in winter and summer. These differences were influenced= =20 by the spatial distribution, of food species, plant chemistry , and the=20 differential use of food species among different age and sex classes=20 during the two seasons. Protein and tannins along with moisture seem to=20 govern diet selection rather than just being a function of availability=20 during winter and summer. Spatial distribution of food plants along the=20 habitat types (in the current land-use pattern) does influence habitat use= =20 by gazelle. Selection and proportion of the time spent on food species=20 during summer and winter suggest that Indian gazelle is a browser.=20 -------------- WINTER HABITAT USE BY MONAL PHEASANT (LOPHOPHORUS IMPEJANUS) IN KEDARNATH= =20 WILDLIFE SANCTUARY, WESTERN HIMALAYA. BY: R. Suresh Kumar. Supervisors: Dr. G. S. Rawat & Dr. S. Sathyakumar. SUMMARY I studied the winter habitat use by Monal pheasant (Lophophorus=20 impejanus), in Kedarnath Wildlife Sanctuary, Western Himalaya, from=20 November 1996 to April 1997. The study period included three seasons:=20 autumn (November-December), winter (January-February-March), and spring=20 (April). The objectives of the study were to quantify availability and=20 utilization of the different habitats, which were named after prominent=20 vegetation types in the three seasons, identify habitat variables=20 influencing monal habitat use, and determine the sex ratio, group size and= =20 group composition of monal. Existing trails and transects (eight in=20 number) were used to quantify the above mentioned parameters. Eight different vegetation types were identified in the study area. They=20 are : Oak-Rhododendron Forest (ORF), Oak-Rhododendron Degraded forest=20 (ORD), Oak-Rhododendron-Lyonia (ORL), Maple-Oak-Rhododendron forest (MOR),= =20 Fir forest (FIR), Scattered Tree and Scrub (STS), Alpine Scrub (AS),=20 Alpine Meadows and Rocks (AMR), and a separate category `cliffs'. During=20 the entire study period monal showed preference for ORF. Within this=20 vegetation type, they were seen mostly close to the streams. Most of the=20 monal sightings (66.7%) during autumn were between 2900 m and 3200 m=20 altitudes. Monal showed movement to slightly lower altitudes (to 2800 m)=20 during peak winter. During late March, the snow had started to melt and a= =20 gradual movement of monal to higher altitudes was noticed. Monal showed strong preference for dense ORF with high litter cover during= =20 autumn and winter. At the onset of spring, there was a shift in the=20 habitat use and they showed preference for cliffs. The males used such=20 areas for displaying to the females. Other habitat variables such as=20 bamboo cover, canopy cover, and presence of snow played an important role= =20 in the choice of habitat by monal. The mean group size did not vary significantly across the seasons. During= =20 autumn, male and female monal were in loose groups i.e., females had=20 tendency to form small groups, whereas males remained more or less=20 solitary. After the first snow, distinct group formations were seen. The=20 groups were categorised into an all-male group of seven or eight=20 individuals, all-female groups of 10 to 12 individuals, mixed groups, and= =20 solitary males. The females remained in groups throughout the study=20 period, while the males remained in groups only for a short span and only= =20 during peak winter. Monal started to move to the higher reaches during=20 late March, and from then on, solitary males were quite often encountered. --------------- A STUDY ON BIRD COMMUNITIES-HABITAT STRUCTURE RELATIONSHIPS IN PENCH=20 NATIONAL PARK, M.P. BY : R.JAYAPAL SUPERVISOR : Mr. QAMAR QURESHI SUMMARY There has been a number of studies on bird communities and their=20 relationships with habitat features of floristics and physiognomy. In=20 particular, the spatial distribution of bird communities along the axis of= =20 structural variables of a habitat has received a wider attention in the=20 light of recent findings on landscape behaviour.=20 Most of these works relate to either temperate or neotropical forests and= =20 our knowledge of community dynamics of tropical deciduous forests is very= =20 limited, for there has been hardly any comprehensive study in the tropics.= =20 The present study was done in the deciduous forests of the Pench National= =20 Park, Central India (M.P) between November, 1996 and April, 1997 covering= =20 winter (November,1996-January,1997) and summer (March,1997-April,1997). Seven distinct habitat types were identified for the study based on=20 floristic and structural diversity. Seasonal data on 27 habitat structural= =20 variables were collected from bird-centred sampling plots. Point-count=20 census technique was employed to study the bird population. Principal Components Analysis (PCA) was used to determine the patterns of= =20 inter-relationships among the habitat structural variables. Canonical=20 Correspondence Analysis (CCA) was done to understand the relationships of= =20 habitat components to the distribution of bird communities.=20 For guild analysis, two different sets of criteria were used to identify=20 the guilds among the bird populations viz., food type and foraging=20 behaviour. Twelve guilds were recognized based on the food type and, six=20 among the insectivores with respect to foraging behaviour. The results of the PCA showed that habitat structural variables of=20 tropical deciduous forests have a marked pattern of inter-relationships on= =20 two axes of landscape behaviour, viz., heterogeneity and complexity. But=20 they failed to reveal any such segregation vis-a-vis bird community=20 structure, as the scale of measurement did not address all the members of= =20 the community. The spatial distribution pattern of bird communities in=20 relation to habitat architecture was shown by CCA to be not very=20 tenacious; the interactions of other extraneous factors like floristics,=20 habitat dynamics, phenophases of vegetation, disturbance, or interactions= =20 within the community may influence the habitat selection process. There was no linear correlation observed between Foliage Height Diversity= =20 and Bird Species Diversity; possible explanations and alternative views=20 have been discussed briefly.=20 The response of bird communities to floristics was found to be=20 inconsistent across habitat types as positive linear relationship was=20 detected only for Teak-dominant forests in winter, and for=20 Anogeissus-Boswellia forests and Cleistanthus collinus woodland in summer. Guild compositions of various habitat types were not found to be=20 significantly different from each other though the seasonal change was=20 consistent across all the habitats. ------------- EXTRACTION OF NON-TIMBER FOREST PRODUCE FROM SELECTED TREE SPECIES IN=20 BETUL FOREST DIVISION AND ITS IMPACTS ON THE POPULATION STRUCTURE OF THESE= =20 SPECIES. BY: ANUPAMA KOLIYAL SUPERVISORS: MR. V.B. SAWARKAR & DR. P.K. MATHUR SUMMARY This study, carried out from November 1996 to May 1997 in North Betul=20 Forest Division (NBFD), aimed to investigate the impacts of NTFP=20 collection on the population structure of three extracted species namely,= =20 Buchanania lanzan, Emblica officinalis and Madhuca indica. Population=20 structure, particularly the proportion of the younger recruitment classes= =20 in the extracted site, was compared to that in Satpura National Park, an=20 ecologically similar but undisturbed area, a perfect natural `control'=20 site. Regeneration of all three were negligible in the harvested site=20 when compared to that in the unharvested site. Interestingly, Timber=20 Forest Products, viz. Tectona grandis, Bauhinia racemosa, and Chloroxylon= =20 sweitenia, which are harvested for parts other than the fruit/seed, showed= =20 good regeneration in both sites. =20 The second component of the study viz. the socio-economic investigation of= =20 the dependent people showed that they ranked other occupations like Forest= =20 Department jobs and agriculture above NTFP collection and realised that=20 collection had a negative impact on the forests. It seems that it is only= =20 the want of other income generating opportunities that forces them to=20 continue extraction. Again this offers scope for conservation measures to= =20 be implemented. Though a short study like this is limited in its scope for= =20 providing long-term solutions, it clearly demonstrates that NTFP=20 collection, particularly of Buchanania lanzan, Emblica officinalis and=20 Madhuca indica in dry deciduous forests of central India does have a=20 significant impact on the forests in general and these species in=20 particular. -------------- FOOD SELECTION AND RANGING IN HOOLOCK GIBBON (Hylobates hoolock Harlan=20 1834) IN BORAJAN RESERVED FOREST, ASSAM BY KASHMIRA KAKATI SUPERVISOR - DR. S.SATHYAKUMAR SUMMARY=09 This study on food selection and ranging in the hoolock gibbon (Hylobates= =20 hoolock Harlan 1834) was carried out between December 1996 and April 1997= =20 in Borajan Reserved Forest, a disturbed and fragmented patch of remnant=20 tropical evergreen forest in Upper Assam. =09 The hoolock gibbon is a primarily frugivorous species, confined to=20 evergreen closed-canopy forests of north-east India, Bangladesh and Burma.= =20 This ape is highly endangered in its entire range. Threats to the=20 continued existence of the hoolock are from large-scale destruction of its= =20 habitat and hunting.=20 Two troops of gibbons were followed intensively for four months and=20 behavioural data was collected by continuous focal animal sampling. Data=20 was also collected ad libitum on a third troop. Data on the structural=20 components of the vegetation and on the availability of food resources in= =20 the intensive study area was collected systematically. =09 The results of the study showed that food availability influenced the=20 diet, movement patterns, home range sizes and behaviour of hoolock=20 gibbons. Gibbons were selective in their diet using only 42 species of=20 trees and vines out of more than 200 species present. Figs constituted=20 important keystone resources for the gibbons. In months of low fruit=20 availability the gibbons had a predominantly folivorous diet. Home ranges= =20 were small (10.4 ha and 5.4 ha) and habitat degradation represented the=20 greatest threat to the gibbons in Borajan RF. --------------------------------------------------------------------------