Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
by the choice of plants the woody browse components of their diets, in that the quality of their diets is influenced by fibre, but secondary chemistry 1 However, giraffe in the 1 for 24 hours, including i feeding time to just three species of Acacia (Appendix Bl). It could be that giraffe Acacia diet, found that Colophoapermum unpalatable and almost VOX 1974 A). are specifically adapted for an but In on area adjacent to the KNP it was a mopane, semi-evergreen woody generally plant, made up of the giraffe diet in August (Hall-Martin, Giraffe in another area of the Transvaal lowveld were found to consistently ignore C. mopane due ibility of sms that giraffe )pe with npounds preferred Acacia and Combretum have I relatively high when they have ability has been reported for t rhinoceros (Oiceros Mcortiis) in Damaralandi South West Africa/Namibia ILoutiL et al ,, 1987). By way of contract with smaller browsers, there is circumstantial compounds produced spread their build-up of by plants, diets over and Freeland and Jansen numerous plants specific toxins, to prevent a However, giraffe in the mid- dry season (July, observed for 24 hours, including could be : 7: to the that giraffe availability of that preferred Acacia and Combret' giraffe have the faouitati' ability has been reported for the desert-dweliing black rhinoceros IDiceros bleornis) In Damnraland, South West although forbs species (e.g. by Sell that the and woody browse giraffe) to select high quality foods i1971) i but optimum diet missed by Deist (1974). This was preferred by all ungulates is (Owen-Smithi 1908; 4.5 SUMMARY Giraffe, kudu, impala and steenbok in the central KNP all depend to a large degree on woody browse in the dry season, However, proportions ef mean annual they were diets. Giraffe woody browse Kudu allocated fed almost of pods differ in the exclusively on and flowers when available. approximately a forbs and bulk, especially browse made up the in the dry season. Impala were mostly the wet about half the diet browse, Steenbok but third ot their feeding creepers, Woody grazers in abundant, to throughout the year, including relatively large proportions time to found woody browse, forbs, and grass in theii season, but in the mid- dry season was composed fed mainly of forbs on forbs depended heavily and woody when they on woody were brows-s and Acacia tortills pods In the dry season. The dry season convergence on woody browse was asso. 'ated with a tendency for giraffe, kudu and impala to spend of the proportionally more lower catena. overlap increased Although not to a time feedingin habitats Mean dietary peak in diversity and diet the late dry season. in the wet season, feeding preferences of relatively low, sensitive than and smaller giraffe browsers were to apparently plant less secondary compounds. Hence for ungulates it is clear that dietary tolerance varies more closely absolute diet quality does. with body size than CHAPTER 5 BROWSER - WOODY PLANT INTERACTIONS distribution, •bundance, and chemical noopesitif By Crawley's (1! in browsing around addressing surface water? two questions: Secondly, when giraffe 6.1 BROWSING, ACACIA TREES AND SOIL AT A SAVANNA WATERIIOLE 5.1.1 Introduction 6.1.1.1 Browning and plant compensation Mcitaughton (1984:881) proposed that "individual grazers obtain a foraging advantage because of the greater by membership forege yield per in a herd bite from grazing Inwns compared with lightly grazed vegetation." A grazing lawn is a patch of dense, closely grazed grass that is more productive and nutritious than tall, sparser grass growing on the lightly grazed periphery. McNaughton {op. clt.) suggests that the dense, highly branched canopy surface of heavily browsed woody plants is analogous browsing has production in Sweden, to been grazing to the Serengeti browsing {fletuia spp,) a found by noose can induce lawn, Indeed, stimulate (Pellow, (A2oes giraffe Acacia 19836) alces) shoot and on in birch compensatory growth which is more palatable than foliage on unbrowsed plants (Danell et. si., section I grazing 1985; Danell and Huss-Daneli, 1986). In this consider the lawn hypothesis Acacia nigrescens, applicability of to browsing HcNaughton's ruminants and one of the principal food plants of browsers In the central KNP. 4 :r In the Solerocai-ya birrea / Ac&ci& landscape of the central A. nigrescent trees but are KNP predominate on relatively less nigrescent i (Gertenbach, 1983), the basalt plains abundant near surface water (Codd, 1931), where their canopies are typically pruned by impala from below observations). This and giraffe from above (personal pruning results in a clear browse- line beneath a dense, highly branched canopy, which may be sculptured as is into a cone, spf ’■e, or hour-glass shape characteristic of (Sinclair and by giraffe Norton-Griffiths, 1979; Pellew, severe browsing 1984h; Coe and Coe, 2987). The browse-line is caused by impair which are mixed feeders turno.-er (Fairall seldom found with a more than a few water (Young, 1972; this maintained by many impala from woody high rate of water and Klein, 1984), and in the KNP are plants in giraffe exerting kilometres from surface study), The browse-line each taking a little is browse a localised area. The reason for such a high browsing pressure at waterholes is rather less apparent however. Unlike grass, woody foliage water content through the savanna browsers water (Taylor, Giraffe, kudu, available, but high preformed which enables to be largely independent of surface 1969; Western, and steenbok not during droughts). retains a dry season, by This is 1976; do drink necessity Louw, when (except evident from 1984). water is obviously the fact that during the out of two dry tjosona of this study, at least two the four kudu cow groups monitored by radio telemetry (Chapter 3) did not visit surface water once. Neither did the two giraffe were often radio-collared steenbok. Although often observed observed feeding drinking, they around waterholes were also and then moving off without drinking. During a dry season in the East African Amboaeli ecosystem, the highest density of giraffe was found 30-36 km from water, while zebra and wildebeest were concentrated 5-10 km from water (Western, 1976). Since browsers ore largely water-independent it is to be expected that they should actively avoid waterholes. The central of lions KNP supports in Africa distribution of one of the highest censities (± 13 adults per 100 km2 ), and the pride core areas is closely related to the distribution of waterholes (Smuts, 1982). Lions are attracted to waterholes by mainly wildebeest by lions and zebra, as they my observations a high biomass of prey, which are often ambushed come to drink (Schaller, 1972). From it is clear that kudu in the central KNP are sensitive to the predation risk associated with waterholsg, which they do actively avoid, This does not app'y for giraffe however. sculptured A. nigrescens To tl.a canopies are contrary, dense evidence of the fact that giraffe are actually attracted to waterholes, Because giraffe to visit risk a high predation threat in order waterholea that from, especially explanation is Since the in the they do not need wet season, that they to the most drink likely are attracted there by food. browse (A. nigrescens foliage) that giraffe spend most time feeding on at waterholea is equally (or more) abundant elsewhere, this suggests that browse at waterholea might Hence this be more palatable investigation hypothesis that was than designed severe browsing elsewhere. to test the by giraffe and impala results in compensatory growth of enhanced palatability in A . nigrescens patches that are frequented by both species. 5.1,1.2 Browsing and Acacia savanna e/yasBica In African feeding semi-arid by savanna concentrations ecosystems, of selective indigenous large herbivores can profoundly influence vegetation dynamics (Walker, 1986). overstocking In areas causes of subsistence encroachment of paatoralism, less preferred woody plants, soil erosion, and dea: '.vLcation (Walker et al ■, 1981; du 1986). The dynamics is 196B[ Sweet and Mphinyane, browsing on aomi-arid savanna Inextricably linked with tho influences of Boil moisture, 1882; Pellew, impact of Toifc, influence of grazing and 1983a), and fire (Walker and Noy-Meir, in conservation elephants (Loxodonta areas Africans) is a the major 'ogetation dynamics, Large browsing and that, is the concern of this herbivory could food plants lead to replacement of preferred woody by better defended species concentration points, such as at waberholes. herbivore In boreal forests it has been found that herbivore-driven changes in woody plant community with reduced Chapin, rates of 1988). In possibility of this cycling section I be associated (Bryant and consider the a similar effect occurring in a savanna ecosystem, where 5.1.2 structure may nutrient, herbivores concentrate around surface Methods i. Experimental design Three sites were chosen on the upper catena on basaltic clay soil in the Sclerocarya birrea / Acacia nigrescene savanna landscape of the central KNP (described by Gertenbaeh, 19B3). One site was at a seasonal watorhole that had formed in a shallow depression enlarged by wallowing elephant and buffalo. The heavily grazed immediately about this waterhole Acacia and tcrtilis trees, Five both species severely mature trees at this individually labelled. and was pruned trampled area vegetated with A. nigrescena were chosen at random site, each and tree from was On each tree ten growing shoots were marked with individually numbered plastic tags, which were tied with string approximately 60 cm in from the canopy surface, The tags were distributed all round the canopy from the surface, The branches) distal using a length against a lower browse-line length of shoot to each of the upper all side- tag was string ruler, Tagging to (including which was done precisely measured was compared then in mid- September, just before the spring flush, A second site, the A. nigr-iacena control site, was in a monospecific stand nlgrescens trees 3 km from of mature, lightly browsed A. 7 km from the waterhole and more than alternative water. The herb layer at this site, which supported a relatively low density of large herbivores, consisted mainly of tall, lightly grazed Themada triandra grass. Here, five trees were chosen at random, and way (and A third individually marked and taggeo in the same during the site, the monospecific stand ranger station, herbivores visited due to the nevertheless same week) as at the waterhole A. tortilis control site, was in a of the 4 km A. tortilis from this site disturbance of representative the at relatively human of Tshokwane Waterhole, activity, the Large infrequently It was surrounding A , tortilis savanna vegetation, which is heavily grazed rampled by large herbivoj growing season), The proportion of tagged shoots showing signs of recent lite, during proportions of browsed shoots pej -vr Methods of analysis were those described by Bryant _et £j (1985) and outlined in this thesis in Chapter 4. Sc '1 samples were collected beneath each marked tree at each site by scraping away surface litter and removing a spade-full places (at of soil canopy, 1 o well mix.J depth of of the ± 10 cm compass) at four beneath the out from the stem. These four samples were and one sub-sample was placed in a labelled paper packet. of soil to a each point This resulted collected from each site. These were temperature in were analysed air-dried the USA tree species and stored until analysed. for total ammonium nitrogen, in five replicate samples beneath each nitrogen, at Soil nitrate at low samples nitrogen, total phosphorus, and water soluble phosphorus. Total nitrogen and phosphorus were analysed by the methods described outlined in this thesis by Bryant _et si in Chapter 4. (1985) and Soil nitrate (NOa), ammonium (NHO, and water soluble phosphorus (P) were assayed 1982): NOg by standard agronomic and NH< were extracted soil with 75 ml of 2-normal KC1; extracted overnight distilled deionized extracts were autoanalyser. from water 2-3 g (7 ml methods (Black, for 1 hr from 15 g water soluble P was soil Hz0/g with soil). double These analysed colorimetrically on a Technicon iv. Assumptions This experiment ■ was designed to compare browsing ! and i local soil nutrients available to specific trees of two ' pressure, net shoot Acacia species It is leaf chemistry, across a gradient of herbivore density. | not assumed that measurements from each site are | representative of the extension. the regions experimental pseudoreplication differences design (Hurlbert, (i.e. differences) were in which they occur, as would then 1984). excluding flawed Intrinsic by site herbivory-induced , f in the I and soil type, and in the same position | at the the 1 top of control sites) catenary choosing sites j { same landscape furthest distance minimized by be drainage between sites was sequence. 10 km (the The two and so rainfall differences are assumed to be negligible. 6.1.3 ► Results (Table 6.1) i . Browsing pressure The overall conclusion of the Student-Newman-Keuls (SNK) multiple range test (a » 0.06) was as follows: BPven ^ BPv a s + BPo an « BPo a e , .here BP is browsing pressure on tree samples denoted Tibia 5.1, Flint i M soil data frgi mte r h o l i M d control sites, firubli f w a u l H 1} 8.6,1 per v i r U b l a per site A, E ortlh Control Browsed shoots' Het sboot e Uenslon <,»]gresceits D iff 0,60 mt a r h o l s I1.M Diff ( Uaterhole 30.TS Oiff ) Control J.M ns leaf tannin" Issf total H (t in mssI leaf total P H dry iiesl Soil total N Soil nitrate N 18.6061 (1581 Z.OO Soil JMOBiUJ! « Soil total P Sofl soluble f luir'l ‘Ciflpiritors Inolcite differences deteryined by the Student-Hems-Keiilf m l t l p l e rinse test (e : 0,06), 'Pereantise of w r W shoots v U h l n eech saeple of trees that i h o m d signs of recent browsing, avaraged over « quarterly Inspections, ’A b s o rbmce value obtained In tue proanthocjmldlfi assey. the waterhole heavily site than two control sil tortilla 11, Shoot where NSE is net previously described), ranked from the right. Sample 46. Although shoots died still alive after The comparison was greater shows that net annual in A. tortllis waterhole and control sites, annual ahoot extension did shoot extension than A. nigrascens at both Within each species, net not differ significantly between waterhole and control sites. This important, as shows despite almost 10 it times more waterhole site per shoot achieved a equivalent to achieve this, browsed A. nitfrescens exceptionally high finding is being browsed severely, A. nlgrescens order to and to that at the net annual growth increment that at the control site. In compensatory growth in heavily trees must have been to compensate for continual removal achieve a net annual growth increment per shoot equivalent to that of lightly browsed trees. iii. Leaf chemistry (Table 6.2) Leaves on heavily browsed A. nigpescens trees at the waterhole were significantly lower in condensed tannin, and higher in total nitrogen and total phosphorus, than leaves on any of the other tree samples. Iv, Soil chemistry (Table 5.3) a). Soil total A. nlgrescens nitrogen control A, tortills control intermediate. was site situ. The highest and lowest at the at the waterhole site was TabU 6.2. Leaf chemistry comparisons across Acacia species and sampling sites. Leaf chemistry variable grouping1 Sample2 Condensed tannin A. nigrescens control tortilla control tortilis waterhole A. nigrescens waterhole nitrogen A. nigrescens waterhole nigrescens control tortilla waterhole A. tortilla control phosphorus b b A. nigrescens waterhole nigrescens control tortilis waterhole A. tortilis control ‘Samples with the same letter are not significantly different {o = 0,06). *Ranked by mean value of each chemical variable, with the sample with the highest mean value at the top in each case. nigrescens waterhole value of each cheuical variable, with the highest mean value at the top in each caae. I f b). Soil total phosphorus was waterhole and lowest at site. A. nigrescens The intermediate. This the was highest at A. tortilis control the same the control site was conclusion for nitrate nitrogen and ammonium nitrogen. c ). Water soluble phosphorus was higher in soil at the waterhole than at the two control sites. d ). In site all comparisons, had e). At the lowest the waterhole beneath the of site, soil A. nigrescens significantly A. tortilla levels different soil nutrient canopies to control nutrients levels were those not beneath A. tortiiis canopies in all comparisons, v, Variation in leaf nutrients with respect to variations in soil nutrients and browsing pressure a). In both Acacia species leaf X and P levels were not significantly correlated with soil N and P levels (P > 0.05 in all cases). b). Browsing pressure on A. nigrescens Was strongly positively correlated with leaf total nitrogen nitrogen content (percent dry ge percentage of marked ahoota that tor trees at control (triangles! and .rr nm) in the proanthooyai content (Fig- 5.1; r = 0.80, P negatively correlated 0.01, n = 10), and with leaf condensed tannin content (Fig. 5.2; r = -0.70, P < 0.05, n = 10). c). Browsing pressure on was A. tortllis not correlated with any of the leaf chemistry variables assayed (P > 0.05 In all cases), 6.1.4 Discussion 6.1.4.1 In a at Browsing and plant compensation mixed stand a seasonal significantly of Acacia tortilis and A. nlgrescens waterhole, higher on A. tortilis. Browsing also very To the waterhole, pressure on the on A. nigrescens was waterhole tissue was than lost than to at a browsers, growth occurred in A, nigrescens at ao that not significantly control trees. at replace compensatory shoot foliage on pressure much higher control site. browsing A, nlgrescens net annual shoot extension wee less than that among lightly browsed Furthermore, the nutritional quality of severely browsed significantly higher condensed tannin) <4. nlgrescens trees was (higher in nutrients and lower in than that of foliage on lightly browsed trees. The difference between A. tortilis and A. nlgrescens in expected from the hit Firstly, it phosphoi A. nigresoens leaf nutrients should also have been higher at the waterhole than at the control sites. This was not the case, nitrogen at the significantly from site, which all three however, In waterhole that fact, A. tortilts leaf site at the did not differ control A. tartllis had the lowest soil nutrient levels out of sites. Finally, cannot account waterhole for the and soil control A. nigreacens leaf nutrient differences significant difference between sites with respect to condensed tannin levels, as no such difference was detected for A. tortilis. An alternative induces a hypothesis is that physiological response severe browsing in some woody plants that increases palatability, leading to a feedback loop of further browsing. The mechanics of this feedback loop (refer to Fig. 5.31 are outlined as follows: Browsing ruminants even steenbok during feeding 1985; personal pruning (step response quite such aa often bite giraffe, kudu, off (Dunham, 1980; or damage Pellew, 1984ej observations). This 1 ) which different impala and shoot is ends Cooper, equivalent to induces a plant physiological to that induced by severe defoliation such as by insects (Dane11 and Huss-Danell, 1985; Bryant et al,, in press). Pruning the shoot system of mature woody plants reverses ageing (step 2), reducing between-shoot competition for nutrients Figure 5.3. Cauee-and-eftect model of woody plant chemistry changes induced by severe browsing, showing positive effects on palatability that Induce further browsing. Arrows between boxes indicate positive effects unless marked (-). Small arrows within boxes indicate directions of change. (Moorby and Wareing, 1963). concentrations of 3)i stimulating appears that The result nutrients in growth (step the pruning is increased remaining shoots {step 4), Prom this study effect, widely to the it employed in advantage of orchard management, also works browsing ruminants. Evidence is compensatory growth of shoots and significantly higher levels of leaf nitrogen and phosphorus relative to in severely less pruned pruned A. nigreacens A. nigreacens and A, tortllia (Table B.l). Furthermore, carbohydrate demands incurred by compensatory limitation growth (step (step 6) metabolite synthesis of 5) cause substrate carbon-based (Bryant et secondary a!., 1983; Danell and Huss-Danell, 1686; Bryant et al., 1987; Bryant et al,, in leaf press)i Hence heavily browsed significantly condenses A, nigreacens at reduced browsed plants at the (halved) tannin .he levels waterhole relative to in were lightly control site (Table 1). Reduced chemical defence and increased shoot nutrients (steps 6 and 3) increase the palatability (step 7) of most woody plants (Bryant et al., 1983; Danell and Huss-Danell, 1986; Cooper and Owen-Smith, 1986; Cooper et al., 1988; Bryant et thereby al., in attracting press; this increased study, Chapter browsing 4), pressure (feedback loop to step 1). For the above cause-and-effect system to be initiated and maintained, it requires browsing pressure to become focused on a suggest that which localised this focus ioipala are requirements, height range. of attracted and A. nigrescens to patch where the 4. nigrescens. I is provided by waterfioles, to for they upper limit Even moderate their water severely browse of their feeding browsing by giraffe would then raise the browsing pressure on these patches above the "background" produce the level. pruning palatability, then If this effect was sufficient leading more giraffe to to increased would be attracted to these patches and the browsing feedback loop would come into effect. Stimulation of been Acacia shoot production by browsing has demonstrated giraffe (Pellew, 1987). Pellew induced by Acacia browse quality. Hence considered analagous to the which modify also Chapter is to in a 6, section consider the that this could be agents browsing not production, by browsing individual graaers (Teague, indicates that shoot concentrated feeding in grazers, with goats Dyer, 1980; McNaughton, 1986). this study stimulates experiments domestic plant-frowth-promoting: al., 1972; Evidence from promotes controlled (op. oit.) suggests salivary (Reardon et only in 1983b) and the but effects ruminants could also of be effects of gregariousness the sward to the bene.it of herd (McNaughton, 1984; see 6,9). The next important step degree to which A, nigroscens can continue to compensate for tissue loss, and by so doing maintain the browsing system. 5.1.4.2 Browsing and Acacia savanna dynamics At the waterhole abundant than site, in the A. nigrescens savanna, A. nigreacens was pruning the previous that compensatory severe being competitors, such more browsers on the palatability of Here I to a feedback propose eventually replaced as on than section it was suggested browsing A, nigreacens was severe A. nigreacens, leading further browsing. of by growth enhances severely browsed cycle A. tortilis significantly more A. toctilts. In loop of where surrounding Sclerocarya bircea / by that less A, tortilis, this results in preferred where large herbivores concentrate around surface water. Whether herbivory so at may actually what level, Wiegert, 1976; HcNaughton, found that are issues HcNeughton, 1986; Belsky, even under Serengeti ecosystem, grazer-adapted definitions of (1984b) benefit plants, and if of hot debate (Owen and 1983; 1987). very 1986! (1979) intense productivity grasses Belsky, McNaughton to grazing was which in maintained the by conventional overgrazing may be inapplicable, Pellew suggests that Acacia species are able to withstand severe browsing as the result of evolutionary e herblvory depresses plant fecundity (Crawley, ippressed by severe browsing (Coe and Coe, 19B7). defended against browsing Istructurally leaves (particularly palatable indehiscent A. tortilla), pods. By and produce contrast A . nlgrescens and Owen-Smith, 1986; Pellew, 19846), while consumption of pods facilitates seed dispersal and reduces the risk of infestation by seed-boring bruchid beetles (reviewed by Coo and Coe, 1: -7), It is not surprising then that density where sakened (Codd, the grass layer has be< Mphinyane, 1986). the KNP Indeed, one (Wolhuter, D. uinerea in u948) of the first rangers in ascribes t.he spread of the southern KNP to population increases among ruminants (particularly irapala, but alio kudu and giraffe) after the rinderpest epizootic and subsequent protection from hunting. An alternative that to the A. tortilis herbivory hypothesis and nutrient-rich soils, and because of dung. is not replace surface water This investigation are P, clnerea could adapted A. nigeescens be for near local enrichment of soils by supported by however,which found the present that soils at the d . tortilis control site had the lowest nutrient levels of all three sites nitrogen control was site alternative sampled. Furthermore, In fact than could higher at be the that A. nigrescens where surface, such as near waterhole and control sites catena though, the and the at the total waterhole. A second out-corapetes A. tortills water-table is near riverlines. A. tortiiis were at soil X . nigrescens the top the at the of the small seasonal waterhole would have bed an insignificant influence on soil moisture in the surrounding elephants are near su. "ice expect area. Thirdly, it could be argued that the prime agents of vegetation change water. This aey be so, but then one would A. tortllis A. nigresaens, because to be for affected example in as Lake much as Hanyara National Park, Tanzania, it is elephant damage to A. toccllls that is causing most concern (Laws, 1970; Mwaiyosi, 1987). Finally it must be heavy grazing and trampling recognised that remove the fuel load near surface water. Consequently reduced burning could cause species changes of herbivory, in woody vegetation as a distal effect but the effects of reduced burning and herbivory are difficult to separate. Considering possible herbivory remains establishment of near surface severe selective A. tortilla f D. cinerea thorn-scrub water in the central KNP. Associated with this thorn-scrub such as alternatives, the most likely factor promoting the i/aytenus (Coddi 1961; are various sclerophyllous evergreens seneg&lensis van Wyk, These species 1984; and Euclea personal divinorun observations). are generally low-preference food plants to browsins ruminants (Chapter 4). I suggest that this woody plant association could develop where selective browsing reduces the fitness of species A. nigrescons. such changes occur where as in the selective inherently herbivory sclerophyllous evergreens litter produced by favours encroachment chemically (Bryant and small amount mineralises slowly vegetation boreal forests of North America, slow-growing The relatively Similar severe preferred these (Lewis and of defended Chapin, 1986). of fibrous, tanniniferous plants decomposes and Starkey, 19681 Swift et i. a.1., .979), thus impeding nutrient cycling (Bryant and Chapin, 1996). A potentially important investigation is finding that total of the present soil nitrogen was highest at the lightly browsed and graaed A. nSgresaens control site, intermediate at the waterbole site, and lowest at the control A. tortiiis Nwaswitsontso River, severe berbivory be replicated site. the latter Being near the site had a history of and trampling. Sampling would have to using more sites to determine if total soil nutrient levels are always higher in A. nigrescent savanna than around savanna. Nevertheless exact reverse (1984) for woterholea the pattern of that or in A. tortilis I describe described by is Hatten and the Smart the effects of long-term exclusion of large herbivores on soil nutrient National Park, destruction Uganda. of trees status in Murchison Falls Weir of (1971) found certain extensive species around artificial pans in Hankie National Park, Rhodesia (nov Zimbabwe). flung accumulated at these pane, where soil phosphorus levels were high, but the phosphorus levels declined steadily outwards for the woodland from ffier'e fringes, and study, the approximately 4 km to then increased again, Apart influence of African large herbivores on nutrient distribution, redistribution and cycling has On purely received little attention (Gumming, 1982). theoretical grounds, Botkin et al, (1981) suggested that large herbivores may be important in the maintenance of but this are a nutrient is mainly susceptible pool near the soil surface, in high rainfall areas where soils to leaching. These authors also acknowledge that cycling of nitrogen through herbivores could in fact increase nutrient losses via volatilization of ammonia from urine and faeces. I suggest that severe selective herbivory in ariu and semi-arid ecosystems could retard nutrient cycling near surface water, resulting in a local soil decline in fertility. The process by which this could occur (refer to Fig. 5.4) is as follows: Where browsing heavily on growth of ruminants concentrate A. nigresaens, which enhanced browsing (this fall from removal of and compaction, important represents nitrogen, the herb trampling, '.'his as a the organic through nitrification replenishes (Hatten and in mineralizable which ammonia-nitrogen which decline and attracting further browsing reduces litter A. nigrescens. Litter reduced by grazing palatability, study). Severe they impact most produces compensatory production is further layer through also litter matter of of ammonification the plant-available nifcrate-nitrogen soil production content resource severe causes is soil organic and pool of respectively Smart, 1984). A. nigrescens is replaced by Enhanced A side d g r e s c m U ls tib llit* S e n re se lK tiM T tro n in j depletion of herbaceous Deduced ra te of n irtr/e st C fd fn j? ^ r v lif e n tlo n o f thorn-scruh and aderophyilous evergreens Notwithstanding the limited a significant infit •lent cycling ir particular attention ig The diet of giraffe in the (Hall-Martin, 1974a; t. ;udy period during which flowering >red 7, and scores for other months relative to this, This method provided a convenient 1 •elative flowering index" . feeding on giraffe, kudu, impala, in Chapter 4. January 1985 to For this December 5.2.3 Results i. Phenology of flowering A. tortilis Flowering in A. nigresi short period in the July to li September 1981 tmber 1986, a t. tortilis v less ret ular, peaking : : varying per; Flowering in 1986 (Fig. both species 6,6), probably I on rainfall; from 1986 was weaker than of lower rainfall i during the wet season of 1985/86 relative to 1984/86 (439 mm c.f. 809 mm). 11. Flower consumption by browsers Giraffe were the only woody plants constituted an browsers for which flowers important food of resource, During September in both years over which feeding data were (1986 collected A. nirfreseens flowers were available feeding time and 198G), in large giraffe consumed quantities while they (Figi 5.5). This accounted for 23,6% of (out of 11,6 hours of observation time spread over 6 days), recorded during September 1986 and IP.3% in September1986 (out 6 days). The lowervalue for 198$ corresponds with the weaker flowering A. tortills flower when flowers for 4.2* A. nigresoens consumption did were available of giraffe 1S86 (17.5 of of 16.5 hours spread over that year. always occur (Fig, 6.5). feeding time hour» spread not over 6 It accounted recorded in January days), and 29.SX in March 1&86 (out of 6.3 hours spread over 2 days). Other flowers making giraffe diet and Combretum Flowers making Maytenus a significant were Acacia contribution gerrardii in to the the wet season hereroense in the dry season (Fig. 6.6), a minor senegalensia, paniculatum, ,c hotia mespiJiformis, and contribution to the diet were M. heteropkylla, Combretua brachypetaliaj Dlospyros Kigelia africana (Appendix Bl). These plants bloom in the c during September and accounted for 3.4% of feeding ti for that flowers month, averaged eaten sengalenals, by kudu over the two years. Ott incJ M. heterophyi and Combcetua mossambtcense (Appendix B2). Impala and steenhok were flowers of any woody not observed attractive re, ntal to trong feeding on the plaiits, these being mainly above If the reproductive selection for food giraffe resource feeding success of defence to was flowering against flower Consequently, giraffe might stand nibbling at A, t< is unlikely that much damage is done m s "bottle-brush") inflores (Hall-Martin, 1974a), jpinesoent although thi stipules sarly all the species with capitate well armed with apinescent stipules. have pals sith capitate •phological if species inflorescences have differences between with capitate and bright the spicate 4 furrows were not found on any pollen grains from African species with spicate Inflorescences. Acacia pollination is attributed mainly to insects, although the breeding systems and pollinating agents of many African Coe and species remain unexplained (Ross, 1979; Coe, 1981). The pollen grains are too large to be dispersed by wind (Coetzee, conspicuous aggregations some species, all) African and male of insects most African low pod;flower 1955), and ! on the flowers of ; Acacias have ratio (Rosa, a remarkably , 1979). Many (and possibly ; Acacia inflorescences consist of bisexual flowers in (op. cit.) suggests varying proportions, that sterile or male and Ross flowers are possibly the functional equivalents of petals, with the production of vast quantities of blossom creating light of the above, A. nigresoens flowers may seem. The removal he fertilized reproductive success flower mass and necks. telemetry on j | of flowers that are unlikely to j have little of the though, giraffe large quantities of pollen Giraffe are on adult ^ is much leas destructive than it anyway will feeding ) a visual attractant to pollinating agents, In : despite plant. By are by giraffe effect on the 1 feeding in the ; certain to collect highly mobile, and using radio females : in the hair of their beads I have recorded daily movements of up to 20 km, while one radio-collared bull i itudy. Chapter 4), flowering that giraffe (personal observations). >. By contrast it would by pollen di t.:.1 f ungulates, preventing 6.3 SUMMARY 6.3.1 At a Browsing and waterholes seasonal vaterhcle, A- nigresaens trees were more heavily A. ■ b'-owsed than trees. tills ! nlsrescens had Jpvels of browsed condensed tannin that intershoot compensatory severe browse generating a finer-leafed heavily for by browsed browsers lightly It is reduces promoting demands carbon-based attracts biowsing - of site. nutrients, The result that suggest that foliage control Carbohydrate reduce metabolite synthesis. palatable a pruning growth. growth than at competition compensatory and of higher Irvels of nutrients and lower A. nigreacena suggested loop, I thornier Foliage of secondary is a patch of highly further browsing, compensatory growth feedback this is analogous to the grazing lawn phenomenon. In the felerocarya birrea landscape of the central d . iortilia and Dichrostachys cinerea heavily grazed water. against These / xcscin nigreacena savanna JINP, there is a tendency for and trnmple< patches species herbivory, propagation, and selective browsing arc have hence physically to dominate around well he rbivore-adapted are successful reduces the in surface defended means where of severe fitness of competitors i such as A. nigreacens. Associated 0. cinerea thornscrub are evergreens. I quantity of litter produced impede suggest nutrient water points. that cycling, relatively lower with A. tortiJis various the / sclerophyllous lower quality and by this association could and ultimately soil fertility in the lead to vicinity of The management implications of this need consideration in the planning of artificial waterpoints. 6.3.2 Giraffe and Acacia nigrescens flowers Flowers of A. nigresceas are an important food resource to giraffe in the late necessarily deleterious dry season. This is not to the plant, however. Most or all African Acacias have very low pod:flower ratios, as many flowers are sterile. Such factors as inflorescence colour and structure, pollen characteristics, and thorn structure indicate could be that one pollinated by flowering in A. nlgrescens, and between A. nigcesoens suggest that species. group of ungulates. and giraffe Atrican Acacias The phenology of the close association in the central KNP, giraffe could be a pollen vector for this CHAPTER 0 FINAL DISCUSSION 6.1 Overview The questions address the posed at issues of the beginning of this thesis how syntopic browsing ruminants use shared food resources, how preferences for habitats and food plants differ, and. some of the implications of resource use on resource conceptual framework addressed was guild. In use by abundance and within which that of quality. these issues resource partitioning The were within a this chapter I synthesise my findings on the browsers of habitats and food plants, and discuss thase with respect to the resource partitioning hypotheses presented conclusions on in the introduction, I present my resource browsing ruminant partitioning guild, consider the within the Importance of interspecific competition to the guild’s structure, and compare the assess the browsing and grazing guilds. Finally, I value of the guild approach and suggest how future research could build on the results of this 6.2 The use of habitats My findings steenbok confirm that use communities) seasonal cycle for ungulates giraffe, different with different kudu, habitats impala and (vegetation intensities over the as a whole (section 3.3.3), as proposed by Lamprey (1963) and McNaughton and >ith respect of forba and woody forage types that they Hofmann and Stewart 11972) ruminants (“selectors separated African browsing of juicy, concentrated herbage”) into (a) "tree and shrub foliage eaters" and (b) "fruit and dicot (tree, shrub or forb) foliage selectors", The former ^roup group included giraffe included steenbok, duiker which was and kudu and the latter klipspringer, classified intermediate feeder. the basis and with This classification of s'--»-ach but impala not as an was made on structure and field observations published . ■. My findings on dist composition (section 4. al inconsistencies with Hofmann and Stewart’s classification. giraffe allocated time to Of the four species studied, the highest fruits and proportion flowers, which of feeding would warrant its inclusion in group (b) above. Steenbok hardly grazed at all (see also Cohen, 1976), feeding on forbs, foliage and Acacia tortilis pods, which warrant itn inclusion (with giraffe) in woody would also gr-up (b) of Hofmann and Stewart’s classification. The above examples separating ungulate illustrate species the into difficulties categories on of the basis of diet. If browser species are to be classified by diet, I suggest that they be graded on the basis of the relative proportions of herbaceous and woody browse that they ualleri) consume. For feed almost example, gerenuk exclusively on llitocranius woody browse i (Leuthold, 1970), to giraffe as woody and so than are are trophically more similar kudu, which feed on forbs as well browse. Block rhinoceros also feed on both forbs and woody browse (Goddard, 1970), and so in terms of their food requirements they are more similar to kudu than to giraffe and gerenuk. Such similarities and distinctions were Stewart, who placed giraffe, some category did not not considered ("tree and consider black by kudu and shrub foliage rhinoceros as Hofmann and gerenuk in the eaters"), and this species is not a ruminant. 6.3.2 Seasonality Among kudu, impala and steenbok, the proportion of woody browse in the diet rose sharply in the dry season to compensate for a reduced abundance of green herbaceous forage types (section 4.3,2), The proportion of forbs in the season, but tissue of also rose in the dry contributed mainly by desiccated dead or dormant forbs that kudu and steenbok ignored. The is probably surface impala diet this was ability of impala to use this forage type related to water, while the dependence steenbok and of impala on kudu depend on preformed water in their diets. Dietary diversity from the late wet and overlap season to among species increased a peak in the late dry hypothesis adhered Schoener (1982, probability over to that Smith et al. (1978) finding indicates interspecific intraspecific because if by 1986a), this competition competition for interspecific competition food. ai a li prevail This : was prevalent i most intense in the lean season, and then e lity food types, specifically found that fruits in the wet separate f 1978: but baboon on marula {Scl&roaarya observations). However, amon (Appendix 8) was ficient for this dietary overlap. convergence to result in probability of relatively higher r I having relatively higner preferences for certain shrub species. Wet season steenbok feeding preferences were strongly condensed tannin of kudu, negatively levels in iropala related to and leaf their food plants (section 4,3,0.1), This relationship fell away in the dry season when the choice of negative relationship and feeding conforms savanna ruminants tannin with (van level previous Woven, 1964; Owen-Smith, 1985). However, I found no such relationship to of giraffe. apply during either season in the case A relationship accessibility-related feeding types was reduced. A between condensed preference observations on Cooper and woody forage preferences could have factors for been masked by influencing some of the giraffe woody plants assayed (section 4,4.2.3). Nevertheless giraffe clearly tolerated relatively high levels of dietary tannins during certain periods of the dry season, when they fed entirely on tannini.erous Acacia trees foliage, (Appendix Bl). with proline-rich almost Deactivation salivary proteins all of from tannins (Mehansho et a J ,, 1987) has been suggested as a means by which deer avoid the digestion-inhibiting effects of food plants al,, 1987b}. It could well be (Robbins et tannins in their that the same applies for savanna browsers, and giraffe in particular, {op, cit.) to as proline-rich impart viscosity proteins are reported to saliva, and giraffe i. saliva is particularly viaoous (section <1.4.2.2). f.4 The influence of body size on resource use A dominant theme running Investigation is that through the resource-use results of this dit rences arc related to body size differences among species. This is not a new finding. asserted that resources Huxley (1942} coexisting species usually differ in and Lack using the body size. (1947) same food Hutchinson (19S9) concluded that for two such species the ratio of linear dimensions be approximately of the suggested, indicates to permit larger to the smaller should 1.3, This empirical ratio, Hutchinson the kind two species of difference neceaaary to coexist at. the same level of the food web. Scores of examples of Rule" have now been Indeed, Owen-Smith ungulates are root of conformity published with (see "Hutchinson's Greene, 1987), (1985) showed that African browsing discretelr graded in size, with the cube the average body sine ratio between successive species being this grading and leaf Istem very near 1.3. Owen-Smith suggested uhat reflects the distribution of leaf sizes ratios prevailing among woody plants and forbs, However, there is now evidence that Hutchinson’s rule is 1.3 based on a statistical artefact. A ratio near will result from comparing the linear dimensions of many objects (even some inanimate ones) in nature that are log-normally with standard 1987). Why distributed, and occur in assemblages deviations less sympatric species cannot be is the cause and Greenough, species unclear, as in many competition alone cases (reviewed by Pagel 1987). Irrespective of the way in which are graded differences exist important than one IEadie et al., this size distribution should prevail among in was size, found consequences for the to fact have the resource use that size following among the browser species studied: 6.4.1 Body size and dietary tolerance My results show that giraffe select new shoots, pods and flowers (mainly from Ac.'i 'ia species) when these are available (section higher proportion seasonal cycle) than did Pellew to fruits steenbok. This (1984a) feeders, that Such understanding that giraffe has high 4.3.1). In fact giraffe allocated a of feeding time (on average over the food and flowers of woody plants finding conforms with that of giraffe findings are because a rate, it selective the herbivore low mass-specific intake highly contradict as general large metabolic costs should feed as but a relatively unselectively on high-fibre foods (deist, 1974; Jarman, 1974; Demment and Georgiadis, 1986). van Soest, 1986; MoNaughton and However, Bell (1971) noted that the i optimal diet for all herbivores is that which is highly nutritious and departures from digestible. this It optimum increasing bodysize. Indeed, parts are low in notavailable quality, both secondary compounds is tolerance that for increases with when high quality plant the giraffe diet is relatively in terms of fibre (section 4.4.2.4: and plant see also Hall- Martin, 1974a,b). The influences that digestive system the feeding differences ecology interpreted largely fibre contents Demment and in body sizs {i.e. ruminant/non-ruminant) of large in terms of selected herbivores of have differences diets (e.g. van Soest, 1985). While this for grazers, it may is been in Janis, suitable interpretation and have on the 1976; be a becoming increasingly apparent that plant secondary chemistry is an additional factor to be considered, particularly for browsers (Cooper 1987a; and Owen-Smith, 1985; Bobbins et al., Cooper et 4 ,4 .2.2) , Whether abilities of compounds is indicates that or browsers to unknown,but body size influences withstand study, the plant secondary circumstantial evidence it might do (section 4.4.2,4) and this requires i n v e s t i g a t ! . body size 1988;this al ■, not Either way, the influence of on the feeding ecology of herbivores is more accurately interpreted in terms of the range in diet section quality tolerated, than dietary fibre content per se (see also Bell, 1984). 6.4.2 Body size and habitat specialisation My findings indicate that with increased increased dietary tolerance body size enables the larger herbivore species to feed in a more diverse range than smaller species (section the generality of general The size trend this' habitat-use / body requires further ungulates it of habitats 3.3.1), investigation, although among African appears to styles of be well supported. Among five feeding behaviour among African antelope, Jarman (1974) recognised a tendency for small species to remain in one vegetation type throughout the year, and progressively larger progressively wider applicable to the trend African ungulate provide an explanation (du press) for why population species population) average body species to range. If size species, then Toit and a it will Owen-Smith, metabolism (energy tends to among feed in is generally in use per increase with increasing these species (Owen-Smith, 1968). Simply stated, larger species are able to spread more evenly range of through ecosystems by feeding in a wider habitats, and so metabolise a larger fraction of available plant biomass than smaller species can The above hypothesis has important implications for resource partitioning. If small-bodied ungulate species specialise on then using different niche overlap negligible. Evidence among for habitats to these this in complete ecological separation of browsers: steenbok, which occur savanna; klipspringer, outcrops; thickets and species may of their bushbuck, (Pienaar, encroach on feeding is will KNP is be the the following small open occur which 1963; the in which each other species patches only occur Smithers, on in of rocky riverine 1983). Larger these habitats but the impact distributed by their use of other habitats too. 6.4.3 For Body aiee and home range area the four species relationship between (section 3.3.2). relationship is with I and found home a close range ares The body mass exponent (1,38) in this significantly higher exponent expected scale studied body mass if home body mass than 0.75, the is assumed to range area in proportion to metabolic requirements (HcNab, 1963). Other studies on home range scaling among Bunnell, 1979; body mass herbivorous Swihart et mammals al., 1988) exponents significantly (Harestad and have also found greater than 0.75, and not significantly different to that reported here. t My data set is current not extensive debate on disproportionately larger to require for their section 3.1.2). animals metabolic home range conducted in different modifying McNab’s from resolve feel area than (reviewed based on a variety regions, the over a they are expected needs Current theories, estimates of home range enough to why are in published of atudi. j concerned with (1963) original hypothesis. That is, area must vary as a function of the minimum area an individual has to cover in order to feed itself (Scnoener, 1968; Harestad and 1981b; Lindstedt et e.1., assumption is not supported field studies. I animals were found the distances that Giraffe, example, for Acacia savanna feed in evidence that my study did would in which in order walk trees were to survive. steadily in full through leaf, to Dichrostachys cinerea that were in 3.4,3), Hence I consider a more realistic approach is ground as this resource limitation) to cover they patches of pod (section Demuth, propose that by adequate evidence from no forced (by Bunnell, 1979; 1986). X to assume an animal it profitably function of benefits will cover as much can. This area will vary as a and costs. Benefits would be determined by the productivity and heterogeneity of the environment, and costs variables, including energetics of (section 3.4.2). by various metabolic locomotion, size-dependent rate, dietary speed and tolorance, etc. 6.5 Feeding height separation Almost all giraffe feeding was at a higher level in the vegetation than smaller than that used kudu the considerable {section the generality of by Lamprey's levels in the feeding height species, Giraffe vertically was questions hypothesis by feeding (see also that at different McNaughton only are quite capable of Kenya, giraffe have been feeding to (Leuthold and (1963) separates For example time finding and For savanna browsers I suggest that levels though, so even this be complete. species ungulates tend to partition vegetation Oeorgiadia, 1986). but for feeding levels 4,3.7). This differently-sized browsing their resources kudu, overlap in giraffe from other feeding at low separation might not always in Tsavo East National Park, found to browsing below Leuthold, 1972). allocate about BOX a suggestion (seealso Leuthold, 1978) that the woody layer by elephants height This gave and of 2 re rise to the reduction of fire could bring giraffe into competition with smaller browsers. Although giraffe canopy of have exclusive access to the upper mature trees, these trees grow from saplings and seedlings, herbivory on replacement of which are food to other browsers, If saplings and seedlings was exceaaive then the mature canopy, where damaged by elephants for possibly of Schoener, example, would be retarded. long-term exploitative 1983) between the Hence the competition {senau smaller browsers and giraffe must be considered. This type of competition is evident in the boreal forests of North America. There snowshoe hares (Lopus americanus) van impact so heavily on preferred woody alaxensis, that food ia severely {A.lces alcea) plants, the abundance den,, ties (Wolff, such as Salix of browse used by moose reduced during peak hare 1980, Belovsky, 1984). I consider it unlikely that such an interaction Is of much importance among browsers in African French, 1986, do not savannas though (but see for grazers), where the smaller browsers reach the high densities that snowshoe hares do in North America. 6,6 Browser - woody plant interactions Results of this aspect indicate that induces severe compensatory of the browsing growth study (section on of Acacia shoots, 5.1.3) nifrescens which bear foliage that is more nutritious and palatable than that on lightly browsed trees. This result conforms with previous findings on the nutritional quality and growth of woody plant shoots following pruning by larqe browsing mammals (Danell et dj., 1985, Danell and HussDanell, 1985; Pellew, 19836] al., in press). Teague, 1987; Bryant et and (b> Rhoades (1985) proposed exactly grounds, postulating this that herbivores on theoretical should be either stealthy or opportunistic in countering plant defences, Stealthy herbivores should minimise damage to plants so that defence responses Schultz, 1988) , are not invoked (but see while opportunists should employ maes- attack behaviour and thereby take advantage of stressed plants. As with many the validity other plant-herbivore relations, of the above hypothesis awaits further investigation. The induction of nigrescens shoots feedback loop The scope compensatory of further of the conclusion, but growth of Acacia by severe browsing would result in a browsing (section study was too limited to 6.1.4.1). reach a I suggest that a feedback loop of this kind could result in the eradication of preferred woody browse plants where herbivores near surface water. Herbivory concentrate, such as by large mammals is not always deleterious to savanna woody plants though. Some plants, particularly the Acacias, disperse! that depend of pods schanisms hy ungulates In addition other s p e d of their I suggest 3 evolved seed (reviewed by Coe and Coe, 1987). that A. nigrescens, and perhaps too, might flowers by have upon the consumption benefit from the consumption giraffe, which have the potential to be very effective pollen vectors (section 6.2). 6.7 How does resource partitioning occur? From the results of this study I conclude that resource partitioning is ruminants in the central KNP, This is by virtue of the clearly evident among browsing following: a). Habitat specialisation among the smaller species (steenbok, klipspringer, bushbuck, etc. ) results in separation between The tendency diverse some or for larger range of communities) lessens resources in habitats habitats their specifically adaptei b). Syntopic species most of these species. species use a more vegetation impact on browse ch smaller species are . on 6,4.2). feed along distinctly foraging paths, as defined associated with feeding sites giraffe feed to (i.e at higher by the different woody plants (section 6.2), levels in the and vegetation than other species do (section 6.6). c ). Syntopic species grass, f irbs, differ and woody in (section i. .1). Such dietary separate species completely, overlap vn-ng them. the browse proportions in their differences do but reduce of diets not dietary i d). Because of morphological variations among browsers and plants, foliage accessible to on certain some browsers plants is more than it is to others. For example, giraffe (being taller) can feed in the upper canopy (having a where kudu cannot reach, narrower muzzle) can pluck and impala leaves from between the thorns of \cao:’ .a tortilla, which are an effective deterrent against kudu (Cooper and Owen- Smith, 1988; see also section 4,4.2.3) e). Dietary tolerance 1971; see also browser species nutritional increases with section may consume quality body size (Bell, 6.4.1), so the larger forage types of lower than can be tolerated and e), electivities by smaller species. f). Because of factors d) woody plants tend to for differ, particularly between large and small species (section 6.3,3). The dry season is factors that this period the four the lean result in species studied, Impala and steenbok to partitioning during are those most deserving of attention. For season occurs Giraffe and season in the KNP, so the resource by spacing kudu are steenbok are niche separation in the dry along three separated by axes (Fig 6.1). feeding height. separated by the tendency for remain on the upper catena when impala are Figure 6.L Niche separation (in the dry season) between giraffe (G), kudu (K), Impels (I), and steenbok (S), Positions of each species along the three axes are relative to other species, and the distance between positions indicates the degree of separation. "Electivities for woody food plants. 'i feeding mainly (giraffe and (impala along the riverlines. The larger pair kudu) are separated from the smaller pair and steenbok) electivitiea for by woody plants. preferences for differences in The larger their pair share certain tree species while the smaller pair share preferences for certain shrub species. 6,8 Why does resource partitioning occur? Considering the degree to which the browse resource is partitioned among that interspecific avoided. This resource browsing ruminants, competition is raises the partitioning following questions: maintained competition? (2) ia resource of through competition resource partitioning alli or would it it is by (1) is present-day partitioning the product evolutionary the product also occur apparent largely or entirely in a of (3) is competition time? at randomly assembled group of ungulate species? Firstly, if niche separation is maintained by present- day competition, then niche expansion is to be expected in the absence of potential competitors. However, among the species is limited kudu cannot studied, the potential for niche expansion by morphology feed any cannot disperse and physiology. For example, higher than they do, and impala away from riverlines in the dry season "A" 1 that if competition is important in "petition through evolui Ly important contributing .d. ' !i 'j 6.9 . Browsers end grazers compared The feeding ecology of African savanna ungulates is currently understood mainly in terms of grazer-oriented research findings (for reasons outlined in the Introduction). My study is the first to include a range of browsers that were studied together and by the same methods. Hence this pr- "ides an opportunity to consider the applicability established for to the browsing guild of principles the grazing guild, These include the following: 1). Diet quality varies as a function of fibre content, and dietary fibre content increases with body size (Gelst, 197-1). applies to (a) the This, the Jarman-Bell principle, browsers if it is adjusted to allow for fact function of that both forage quality secondary varies chemistry content (section 4.4.2,21, and tolerance varies more closely (b) as and that a fibre dietary with body size than does average diet quality per se (Bell, 1971, 1984; section 4.4.2.41. 2). Species move season down the catenary sequence in the dry (Bell, Sinclair, 1970; 1979). Jarman, Although 1972; this does Jarman apply and to browsers, particularly giraffe (Hall-Martin, 1974b; Fellow, 1984a; section 4.3.2,21, the very small t browsers such as steenbok and klipspringer do not exhibit this pattern. These species remain in their specific habitats on the upper catena throughout the seasonal cycle. 3), Smaller species depend on larger species to trample and remove coarse plant matter, so as to facilitate access to high quality Fitzgerald, 1960; 1970, 1971; Gwynne species "grazing (elephant excluded) that there grazing species species because; efficiency (as by (a) this is Bell, does not predator they actually outcome of do not physically the benefit of smaller is debate as to whether follow larger grazing improves their generally Norton-Griffiths,1982; smaller, succession" vegetation to species. Note benefit (Vesey- 1968; browsers, as the feeding actions of large modify woody smaller Bell, McNaughton, 1976; Haddock, 1979). This principle, termed apply to plant parte and assumed); protection Sinclair, displace the they (Sinclair 1986); or larger species scramble competition narrower-muzzled feeding (b) (c) as the that favours species and (Illius the nd Gordon, 1987) , 4). Oregariousness may (McNaughton, 1984). vegetation structure increase foraging efficiency This is by the modification of to increase food yield per bite to the individual "grazing lawn" applicable to in a herd (op. clt.). This principle browsers, does although indeed not seem directly. Where browsing pressure is concentrated, food yield per bite (with nutrients) is the yield being digestible indeed improved, but not for reasons of altered plant structure. Rather, severe pruning results in altered leaf levels enhanced and chemistry, with condensed nutrient tannin levels depressed (section S.1.4.1). From the above comparisons I conclude that currently accepted principles in African ungulate feeding ecology are loosely applicable to browsers. The major exception is that of grazing succession, and others (specifically the Jarman-Bell principle hypothesis) would was accepted and the grazing lawn be more applicable to browsers if it that diet quality varies not only as a function of fibre, but of secondary chemistry as well. 6.10 The value of the guild approach, and the next step Section adopting 6.9 (above) a guild demonstrates approach consumer/resource relations. feeding ecology a particular an to advantage of investigating That is, if principles of are established for representatives of taxonomic group (e.g. ungulates) within a particular guild, these principles may be tested on i representatives of the same group in another ("control") guild. The comparison might then show where modifications are value of required to such principles increase the predictive to the taxonomic group as a Another advantage of the guild approach is that because it is resource-defined, it is flexible in the taxonomic range over which consumer-resouree consumer relations demonstrated by gronivore guild 1986). The and consumer- may be studied. This flexibility is a long-term (Brown et study on al,, 1979; the Brown desert et a!., study was initially established to test for competition between ewanded when rodents and ants, and results indicated was that rodents later compete not only with ants, but with birds as well, I suggest that the that the work developed upon groups of foliage for above example embodied is illuminating, in this thecis could be by expanding the guild to include other animals food. that depend Insects regard, particularly herbivory does in in view not only are upon a of the remove dicotyledonous priority part in this fact that insect of the browse resource, but may alter the nutritional quality of that which remains (reviewed by Schultz, 1988), A more immediate step by which the results of this study may be built modelling (see upon is Starfield and to employ ecological Bleloch, 1986). A number of relationships have been identified among the feeding patterns of browsers, and incorporation into these predictive lend models. include: the relationships between composition, with below which the of come into to examples rainfall and identification such relationships themselves Some diet thresholds effect; the Influence of phenology (flowering, fruiting, leaf flush etc.) among certain plant species on diet composition; the relationship between rainfall and movement acroiis the catenary sequence; the relationship between feeding preference and relationship leaf condensed between browsing production and between body tannin nutritional quality, size and content; pressure and the shoot the relationships dietary tolerance, aa well as diversity of habitat use; the relationship between body size and home range area; etc. I suggest that application of inclusion of system for suitability a particularly the results the above the Transvaal of private of my useful relationships lowveld, for holdings indigenous browsing ruminants. management study would for in an be the expert assessing stocking the with REFERENCES Abrams, P,A, 1980. Some comments on measuring niche overlap, Ecology 61: 44-49. Alldredge, J.R. and Ratti, some statistical J,T. 1988. techniques for Comparison of analysis of the resource selection. J , ttlldl. Manage- 50: 157-165. Allen-Rowlandson, T.S. Organisation of strepsiceros Pallas Kudu Reserve, ALtmann, J. 1980, PAe The Social Greater and Spatial (Tragelaphus Kudu 1766) in the Andriea Voslco Cape. PhD Thesis, Rhodes Eastern 1974. Observational study of behaviour: sampling methods. Behaviour 49: 227-267, Mtmann, S.A. 1687. The impact of locomotor energetics on mammalian Altmann, S.A., Nutrients and foraging. J. Post, D.Q., Zoo!. (Lond,I 211: 216- and Klein, D,P. 1987. toxins of plants in Amboseli, Kenya. Afr, J, Ecol, 25: 279-293. Anderson, D.J. nonparametric 1982. The estimation home range: technique. A Ecology new 63: s: _ ... t Belsky, k.3. 1987. The effects of grazing: confounding of ecosystem, community and organism scales. Am. Nat. 129: 777-783, Berry, P.S. 1978. Range Luangwa Valley, Black, C.A. movements of Zambia. B. 1982. Methods Chemical and giraffe in the Atr. Wildl. J. 16: 77- of Soil Analysis Microbiological Properties. Part 2: American Society of Agronomy Inc., Madison. Botkin, D.B., Hellilo, J.M., ecosystem processes are population dynamics, and T.D. and Wu, L.S.Y, 1981. How linked to Pages 373-387 Smith {eda. ). Dynamics of large mammal in C.W. Fowler Large Mammal Populations. John Wiley and Sons, New York. grower, J.B. and Zar, J,H. 197 7. Field and Laboratory Methods for General Ecology. Wm. C. Brown, Iowa. Brown, B.W, and Hollander, Biomedical Introduction. Brown, J.H. 1976. Chapter H. 1977. John Wiley Statistics. A and Sons, New 13. Geographical Ecology of Desert Rodents. Pages 315-341 in M.L, Cody and J.H. Diamond (eds.). Ecology and Evolution of Communities. Harvard University Press, Massachusetts. Brown, J.H., Davidson, D.W., An experimental and Reichman, O.J. 1979. study of competition between seed- eating desert rodents and ants. Am. Zcol. 19: 1129- Brown, J.H., Davidson, D.W., R.S. Chapter 1986. ecology: the in J. Hunger, J.C. and Inouye, 3. Experimental community desert granivore system. Pages 41-61, Diamond and T.J. Case (eds.). Community Ecology. Harper and Row, New York. Brown, J,H. and ecological Maurer, dominance, B.A, and 1986. Cope's Body rule. size, jVature (land.) 324: 248-250. Bryant, J.P. and winter forage The role Kuropat, P.S. by subarctic 1980. Selection browsing of vertebrates: of plant chemistry. Ann. Rev, ficoi. Syat. 11: 261-286. Bryant, J.P., Chapin, Carbon/nutrient relation to F.S., balance and of Klein, boreal vertebrate herbivory. D.R. plants Olkos 40: 1983. in 357- Bryant, J .P,, Wielawi, Q.B., Clausen, T . , and Kuropat, P. 1985, Interactions of snowshoe hare and feltleaf willow in Alaska. Ecology 66: 1564-1573. Bryant, J.P., and Chapin, F.S. III. 1906. Browsing - woody plant interactions during boreal forest plant auocession. Pages Chapin III, 213-226 in P.W. Flanagan, Dyrneas (eds.). K. Van Cleve, L.A. Viereck Forest Ecosystems F.S. and C,T, in the Alaskan Taiga. Springer-Verlag, New York. Bryant, J.P., Chapin, Clausen, T.P. defense in F.S,III, Alaska paper manipulation Reichardt, 1987. Response of of birch and plant P.B. winter and chemical green alder to carbon/nutrient balance. Oecologia (Berl.) 72: 510-614. Bryant, J ., and Danell, K , , Clausen, browsing plants. on In T. t:e D.W. Provenza,,F., In press. chemistry Tallamy Phytochemical Induction and Reichardt, P., Effects of of deciduous M.J. by Herbivores- Raup mammal woody (eds,) John Wiley and Sons, New York. Brynard, A,M. and Pienaar, U.de V. 1960, Annual report of the biologists. 1958/69, Koedoe 3: 1-205, Burt, W.H. 1943. Territoriality and home range concepts as applied to mammals. J. .Vammal. 24 : 346-352. Byers, R.C., Steinhorst, R.K., and Krausman, P.R. 1984, Clarification of a technique utilization-availability data. for J. analysis Wildl. of Manage, 48: 1050-1063. Calder, W,A. III. A dimensional trade-off between space and time: constants in mammalian ecology. J■ Theor. Biol. 98: 393-400. Calder, W.A. History, III. 1984. Harvard Size, Function University Press, and Life Cambridge, Massachusetts, Carpenter, F,L. 1978. Hooks for mammal pollination? Oecologia (Berl.J 35: 123-132. Clutton-Brock, T.H. and population 657-676 in Behavioural and Albon, S.D. 1986. Competition regulation in social mammals. Pages R.M. Sibly Ecology, and R.H. Ecological Smith (eds.) . Consequences of Adaptive Behaviour. Blackwell, Oxford. Coates Palgrave, K. 1981. Struik, Cape Town. Trees of Southern Africa. Connell, J.H. 1976. Some mechanisms producing structure in natural communities: a field experiments. J.M. Diamond model and evidence from Pages 460-490 (eds.). Communities, Harvard Ecology in M.L. Cody and and Evolution University Press, of Cambridge, Massachusetts. Connell, J.H. 1980. Diversity competitors, or the ghost and the of coevolution of competition past. Oikos 35: 131-138. Conybeare, A. 1976. Notes on the feeding habits of kudu in the Kalahari sand area of Wankle National Park, Rhodesia, Arnoldia (Rhod.) 7: 1-7. Cooper, S.M. 1985. factors Influencing the Utilization of Woody Plants and forks by Ungulates. PhD thesis, University of the Witwatersrand, Cooper, S.M, a m deter feeding Jwen-Smith, N. 1986. Condensed tannins by browsing ruminants in a South African savanna. Oecologla (Berl.l 67: 142-46. Cooper, S,M. and Owen-Smith, N. 1986. Effects of plant spinescence in large mammalian Oecologia (Berl.), 68: 446-466. herbivores. sed germination responses Damuth, J , 1981a. Populefcii Damuth, J. 1987. Interspecific allometry of population Independence of use. Biol, body mass and population energy Linn. Soc. 31: 193-246. Danell, K. and Huss-Danell, K. 1986, Feeding by insects and hares on birches earlier affected by moose and Bergstrom, R. 1986. browsing, Oikoa. 44: 75-81. Danell, K ,, Huss-Danell, K. Interactions between browsing moose and two species of birch in Sweden. Ecology 66: 1867-1678, Demment, M,W. and van Digestive Soest, P.J. Capacity, and Winrock Herbivores- 1983, Feeding Body Size, Strategies International, in Morrilton, Arkansas, Demment, H.W. and van explanation for Soest, P,J, 1985, A nutritional body-size patterns of ruminant and non-ruminant herbivores. Am. Nat. 126: 641-72, Demment, H.W. portable and Greenwood, computer observations of for C ,B , 1987. real-time The use of a recording of grazing behaviour in the field, J. Range Manage, 40: 284-286, de Vos, V., Bengis, R,C., Population control National Park. and of large Pages 213-231 Coetree, H.J, 1913. mammals in the Kruger in R.N, Owen-Smith du Toit, J.T. and Owen-Smith, N . (In press). Body sii du Toit, R.Fi 1966. Soil loss, hydrological changes, ipidermal Broekhoven, L., Colgi growth factor Pollination Ecology. Pergatnon Press, Oxford, Fairall, N , 1983, Production parameters of the impala, S. Afr. J. Anlm. Sci. 13: 176- Aepyceros melempus- Fairall, N. and Kleini water turnover: sized African dorc&a) and D.R. 19B4, a comparison antelope, the impala the Protein intake and of two blesbok equivalently {Damallscus {Aepyceros melampus). Can. J. Anlm. Sci. 64: (Suppl.l, 212-214, Ferrar, A.A, and Walker, herbivore/habitat National Park, B.H. 1974, An relationships analysis in the of Kyle Rhodesia, J. Sth. Afr- Vlldl. Mgmt. Ass. it 137-47, Foster, J.B. 1866, The giraffe of Nairobi National Park: home range, sex ratios, the herd and food, E. Afr. Wildl. J. 4: 139-148. Foster, J.B, and Dagg, A.I. 1972. Notes on the biology of the giraffe. B. Afr. t/lldl. J. 10: 1-16. Freeland, W,J. herbivofy by and ,Jansen, D.H. 1974. mammals: the Strategies of role of plant secondary compounds. Am. Net. 108: 169-289, t Goodall, D.H. 1986. Chapter patterning. Pages Anderson (eds.). 3. Biotope 30-40 in J. Community structure and Kikkava Ecology. and Pattern D.J, and Procsss- Blackwell, Oxford. Greene, E. 1987. Sizing Evoi. Gwynne, M.D, in up size ratios. Trends Ecol. 79-81. 1969. The relation to nutritive value of Acacia pods Acacia seed distribution by ungulates, E. Atr. IVildl. J- 7: 176-178. Gwynne, M.D. and Bell, R.H.V. vegetation components 1968. by grazing Selection of ungulates in the Serengeti National Park. Mature (bond.) 220: 390-3. Hairston, N.G. 1981. An experimental test of a guild: salamander competition. -Ecology 62: 66-72. Hall-Martin, A.J. lowveld 1974a. Food giraffe as selection by determined by Transvaal analysis of stomach contents. J. Sth. Afr. Sfildl. Ngznt. Ass. 4: Hall-Martini A.J. utilisation of 1974 b, A different note on the vegetation giraffe. Sth. Afr. J. Sci. 70: 122-123. seasonal '•ypes by Hall-Hartin, A.J. of the 1975. Seasonal diet of the Transvaal chemical composition luwveld giraffe. J. Sth, Afr, Wildl. Mgmt. Ass. 5: 19-21, Hall-Hartin, A.J,, and chemical composition lowveld giraffe. Basson, W.D. of the 1975. Seasonal diet of the Transvaal J. Sth- Afr. Wildl. Mgnt. Ass. 5: Hall-Hartin, A.J. 1985. Ecology of the African elephant Loxodonta africana Pages 56-56 in the in Research Board of South Afri.h. Board of South Africa National Pai . Division, National Kruger Parks ;compilers). National Parks Research Report 1954/85. National Parks Board, Skukuza, South Africa. Hanley, T.A., Sv^linger, D.E., Hanley, K.A. and Schoen, J.W. 1986. Relationships between analyses for fecal and rumen deer diet assessments in southeastern Alaska. Northwest Science 59: 10-16, Hansen, R.M., Mugambi, M.H. and Bauni, S,M. 1985. Diets and trophic rankings of ungulates of the northern Serengeti. J . Wildl. Manage. 49: 823-829. Harestad, A,S. and Bunnell, F.L, 1979. Home range and body weight - a re-evaluation. Ecology 60: 389-402, e, G.M. 1982. Comparisons between e trends: problems of methodology, in King’s College Sociobiology Group (eds.). Current Problems in Sociobiology. Cambridge University Press, Cambridge. .ten, J.C. and Smart, long-term exclusion N.O.E. 1984. of large The effect of herbivores on soil stomach. E. Afr. browse: structure Hofmann, R.R, 1973. The ruminant Honogr. Biol. Vol. 2. E.. Afr. Lit. Bur. , Nairobi, Hoppe, P.P. 1984. Strategies herbivores. Pages R.I. Mackie of digestion 222-243 in (eds.). in African F.M.C. Gilchrist and Herbivore Nutrition in the Craighall, South Africa. l.D. 1987. Native .ustralia. Pages 1-22 and introduced in J.B. herbivores In Hacker and J.H. Ternouth (eds.). The Nutrition of Herbivores. Academic Press, Sydney, Huntley, B,J. 1972. A note on the food preferences of steenbok. J. Sth. Afr. Vildl. Mgmt. Ass. 2: 24-26. Huntley, B.J. 1982. Southern 101-119 in Ecology Hurlbert, B.J, Huntley of Tropical S.H. 1971. diversity: a African Savannas. and B.H. Savannas. The critique and Walker Pages (eds.}. Springer-Verlag, nonconcept of species alternative parameters. Ecology 52: 577-686. Hurlbert, S.H. 1978. The measurement of niche overlap and some relatives. Ecolcgy 59: 67-77. Hurlbert, S.H. 1984. Pseudoreplication of ecological and the design field experiments. Ecol. Honogr. 64: Hutchinson, G.E. 1959, Homage to Santa Rosalia or why are there so many kinds of animals? Am. Nat. 93: 146-159, Huxley, J. 1942. Evolution. The Modern Synthesis, Kariba, Rhodesia. Oeoologia (Berl.J 6 i i P.J. 1972. Seasonal initial populations distribution of large in the unflooded middle Zambezi J. Appl. Bool. 9: 283-299. P.J. 1874. The social organisation of antelope > their Jarman, P.J. Feeding ecology. Behaviour and Sinclair, strategy partitioning in A.R.E. and the 1979. pattern 48: 215- Chapter of 6. resource ungulates. Pages 130-163 in A.R.E. Sinclair and M. Norton-Griffiths (eds.). Serengeti, Dyn&mics of an Ecosystem. University of Chicago Press, Chicago. Jennrich, R.I. and Turner, non-circular home Jewell, P.Ai 1966. F.B. 1869. range, J. The concept Measurement of Theor. Biol. 22i 227- of home range in mammals, Symp, Zool. Soo. Land. IB: 85-109. Johnson, D.H. 1980. The availability measurements comparison of for evaluating preference, Ecology 61: 66-71, usage and resource V, habits of gerenuk in Tsavo National Park, Kenya. E, Leuthold, W. 1978. Ecological separation among browsing Lewis, J.A. and Starkey, R.L, 1968, Vegetable tannins, Home range, time, and body size in mammal a, Ecology Loutit, S.D., Louw, G.N., approximation of composition of the black rhinoceros Louw, G.N. and Seely, M.K, 1987. First food preferences and the chemical diet of the desert-dwelling Siceros bicornis L . Madoqua 15: 1984. Water deprivation in herbivores under arid conditions. and R.I. Pages 106-128 in P.M. Gilchrist Mackie (eds.). Herbivore Nutrition in the Subtropics and The Tropics. Science Press, Craighall, South Africa. MacMahon, J ,A ., K.G., and Schimpf, L'.J. , Bayn, R.L. approach to 1981. Anderson, D.C., Smith, An organism-centered some community and ecosystem concepts. J , Theor. riol. 68: 287-307. MacNally, R.C. 1983. On interspecific as'sesslng the significance of competition to guild structure. Ecology 64: 1646-1661, Haddock, L. 1979. The "migration" and grazing succession. Pages 104-129 in A.R.E. Sinclair and M, Norton-Qriffiths (eds.). Serengeti, Dynamics of an Ecosystem. University of Chicago Press, Chicago, Malechfk, J.C. grazing and and Balph, D,F, 1987. Diet selection by browsing livestock. Pages 121-132 in salivary chemical and inorganic nutrition. Oecologia (Serl.J 66; 478-486. HcNaughton, S.J. 1986. On plants and herbivores. Am. Nat. 128; 765-770. HcNaughton, S.J. and Georgiadis, N.J. 1986. Ecology of African grazing and browsing mammals. Ann. Rev. Scol. Syst. 17: 39-65. HcNaughton, S.J, 1987. seasonal changes in J.B. Hacker Adaptation of herbivores to in nutrient supply. Pages 391-408 and J.H. Ternouth (eds,). TVie Nutrition of Herbivores. Academic Press, Sydney. Mehansho, H . , Butler, L.G, and Carlson, D.H. 1987. Dietary tannins and salivary proline-rich proteins: interactions, inductions, and defense mechanisms. Ann. Rev. Nutr. 7: 423-440. Moermond, T.C, 1986. A mechanistic approach to the structure of animal communities: Anoiis lizards and birds, Am. Zool. 26: 23-37. Hentii, M.T. 1986. Hypothetico-deductive and inductive tannins: a Monro, R.H. critique of Wisdom et. al. Unpublished 1980. Observations on the feeding ecology of impala. S. A/r. J. Zool. 15: 107-110, Moorby, J. and Wareing, P.F. 1963. Ageing in woody plants. Ann, Sot. (Land.) 27: 291-309. Hurray, M.G. 1982. Home range, dispersal, and the clan system of the impala. Afr. J. Ecol- 20: 263-269. Mwalyosi, R.5.B. 1987. Decline of Acacia tortilis in Lake Manyara National Park, Tanzania. Afr. J. Ecol. 26: 61-53. Neu, C.W., Byers, C,R. and Peek, J.M. 1974. A technique for analysis of utilisation-availability data. J. Wildl. Manage. 38: 641-646. Norton-Griffiths, M, browsing, and 1979. The fire on influence of the vegetation gracing, dynamics of the Serengeti. Pages 310-352 in A.R.E. Sinclair and M. Norton-Griffiths teds.). Serengeti. Dynamics of an Ecosystem. University of Chicago Press, Chicago. Novellie, P.A. strategies 1978. of Comparison blesbok and of the springbok foraging in the Transvaal highveld. Novellie, P.A. S. Afr. J. WildX. Res. 8: 137- 1983. The Feeding Ecology Tragelaphus strepsiceros (Pslles) In of The Kudu The Kruger National Park. DSc thesis, University of Pretoria. Oates, L. G. 1973. Food Transvaal lowveld preferences mopane woodland. of giraffe J. Sth. in Afr. k'ildl. Ngwt. Ass. 2: 21-31. O'Connor, T.G. 1986. A Concerning the Synthesis of Field Experiments Grass Layer Southern Africa. i,. Savanna flexions of South African National Scientific Programmes Report No. 114. Owen, D.F. and Wiegert, R.G. 1976. Do consumers maximise plant fitness? ‘Oikos 27: 488-492. Owen-Smith, R.N. 1976. The social ethology of the white rhinoceros Ceratotherium simum (Burchell 1817). Z. Tierpsychol. 38: 337-304. Owen-Smith, R.N. efficiency of 1979. Assessing the foraging a large herbivore, the kudu. S. Afr. Factors influencing the ; products by large herbivores. Huntley (eds.). Ecology of Tropical and B.H. Walkei Springer- Savannas. Verlag, Berlin. among S. Vrba African (ed .). Museum Monograph gaherbi 5.H. 1985. Comparative in-Smithi R.N. and Cooper, S.M. 1987a. J preferences of ungulates J. ffildl. Manage. 61: 372-378. Owen-Smith, R.N. of woody and Cooper, plants to S.M. 19876. Palatability browsing ruminants in a South Pagel, M.D. and Greenough, J.A. 1987. Explaining size ratios in nature. Trends Bcol. Bvol. 2: 114-115. Pellew, R.A. and fire 1983a. The impacts of elephant, giraffe upon the Acacia tortilis woodlands of the Serengeti, Afr. J. Ecol. 21: 41-74. Pellew, R.A. in the 1983b. The Serengeti. production of Pellew, R.A. browser, giraffe and its food resource I. Composition, available browse. biomass, and Afr. J. Ecol. 21: 1984a. The feeding ecology of a selective the giraffe (Qtratta Camelopardalis tippelskirchi). J- Zool ■ lLon<S.) 202: 67-81. Pellew, R.A. 1984b. Food consumption and energy budgets of the giraffo. J , Appl. Ecol. 21; 141-159. Pellew, R.A, 1984c. Giraffe and okapi, Pages 634-641 in D. MacDonald 2. (ed.). The Encyclopaedia of Mammals: George Allen and Unwin, London. Pennycuick, C.J. the concept 1979. Energy of "foraging A.R.E. Sinclair and M. costs of locomotion and radius", p p , 164-184 in Norton-Griffiths leds.J. imposed SAS Institute Inc. 1986a. BAS Users Quide: Basies, Version 5 Edition, SAS Institute Inc., Cary, NC. SAS Institute Inc. 1985b. SAS Users Guide: Statistics, Version 5 Edition. SAS Institute Inc., Cary, NC. Scholler, G.B. 1972. TVie Predator-Prey Serengeti Lion. Relations. University A Study of of Chicago Press, Chicago. Schoener, T,W, 1968, Sizes of feeding territories among birds. Ecology 49: 123-141. Schoener, T.W. 1970. Non-synchronous spatial overlap of lizards in patchy habitats. Ecology 51: 408-416. Schoener, T.W. 1974a. Resource-partitioning in ecological communities. , Science (Wash., B.C.) 186: Schoener, T.W. 19746. Some competition coefficients methods for reso u of calculating . utilization spectra. Am. Nat. 108: 332-340. Schoener, T.W, 1981. An empirically based estimate of home range. TVieor. Pop. Biol, 20: 281-325. T.W. 1983. Field Senft, R.L., Coughenour, H.B., Bailey, D.W,, Rittenhouae, L.R., Sala, O.E, and Swift, D.M. 1987, Large herbivore foraging and ecological hierarchies. BioScience 37: 789-799. Shannon, C.E. Theory of and Weaver, W. 1949. Communication. The University Mathematical of Illinois Press, Urban*. Simberloff, D. 1983. testing, and Competition theory, hypothesis other community ecological buzzwords, dm, Nat, 122: 626-636. Simpson, C.D. In 1972. An evaluation of seasonal movement greater strepsiceros kudu populations Pallas - in three - Tragelaphus localities in southern Africa, Zool. Atr- 7: 197-206. Sinclair, A.R.E. and Norton-Griffiths, M. (eds. ) 1979. Serengeti, Dynamics of an Ecosystem. University of Chicago Press, Chicago. Sinclair, A.U.E. Competition and Norton-Griffiths, or facilitation ungulate populations in the M. 1982, regulate Serengeti? A hypotheses. Oecologia (fieri,; 63: 364-369, Does migrant test of Sinclair, A.R.E. 1985, Does or predation interspecific competition shape the African ungulate community? J. Anim. Ecol. 64: 899-918. Smith, J.N.M,, and Grant, P.R., Abbott, deeding habits L.K. Grant, B.R., Abbott, I.J. 1978. Seasonal variation in of Darwin's ground finches. Ecology 69: 1137-1160. Smithers, R.H.N. African 1983. The Subregion. Mammals of University the of Southern Pretoria, Pretoria. Smuts, G.L. 1982. Lion. MacMillan, Johannesburg. Southwood, T.ft.E. 1978. Methods Ecological (2nd Edition). Chapman and Hall, New York. Starfield, A.H. and Bleloch, A.L. 1986, Building Models for Conservation and Wildlife Management. MacMillan, New York. Strongi D.R., Tests of Szyska, L.A., and Simberloff, community-wide character D . 1979. displacement against null hypotheses. Evolution 33: 897-913, Sussman, R,W. and Raven, P.M. 1978. Pollination by lemurs and marsupials: an archaic coevolutionary system, Science (Wash., D.C.) 200: 731-736. Sweet, R.J. and Mpbinyane, observations on W. 1986. the ability of goats regrowth in post-browsin/ nigrescens/ Combr&tum Preliminary to control Acacia savanna apicu.latum in Botswana. J. Grassl. Soc. Sth. Afr, 3: 79-84. Swift, M.J., Heal, O.K. and Anderson, J.M. 1979. Decomposition in Terrestrial Ecosystems. Blackwell, Oxford. Swihart, R.K., Slade, N.A,, and Bergstrom, B.J. 1988. Relating body size to the rate of home range use in mammals, Ecology 69: 393-399. Taylor, Ci.R. 1969. The eland and the oryx. Sci. Am. 220: 88-95. Teague, W;R, 1987 . Defoliation and Browse Production of Acacia karroo PhD Africa. Hayne in the thesis, Eastern University Cape, South of the Witwatersrand. Terborgh, J, and Robinson, utility in S. 1986. ecology. Pages Guilds and their 66-90 in J . Kikkawa and D.J. Anderson (eds.). Community Ecology. Pattern and Process. Blackwell, Melbourne. Turner, V. 1982. Marsupials Australia, Pages 65-66 PowelV and as in pollinators J.A. Armstrong, in J.M. (eds,). Pollination A.J. Richards and Evolution. Sydney Royal Botanic Gardens, Sydney, Unwin, D,M. and Martin, P. 1987. u-vng a van Dyne, Recording behaviour portable microcomputer. Behaviour 101: 87- G.M,, Brockington, N.R,, Szocs, Z., Duek, J . and Riblc, C.A. 1980. Che .>■ 4. Large herbivore subsystem. Pages 269-637 in A.I. Breymeyer and G.M. van Dyne (eds,), Grasslands, Systems Analysis and Man. Cambridge University Press, Cambridge. van Hoven, W. 1984, Trees’ secret warning system against browsers. Custos 13: 5-16. van Soest, P.J. 1982. Wutritionai Ecology of the Ruminant. 0 1 B Books Inc., CorvnlHs, Oregon. van Wyk, P, 1984, Field Guide to the Kruger National Park, Struik, Cape Town. Trees of the t VenturCom Inc. 1984. Prelude Reference Wanuai. Printing Version VenturCom 1.2, Inc., Cambridge, Massachusetts. Vesey-Fitageraid, D.F. 1960. Grazing successicu among East African game animals. J . Mammal 41: 161-172. Walker, B.H., Ludwig, D. , R.M. 1981. Moiling, C.S. and Peterman, Stability of semi-arid grazing systems. J . Ecol. 69: 473-498. Walker, B.H. and Noy-Meir, J. 1982. stability and resilience Page in 658-690 (eds.). Ecology B.J, of of Huntley Tropical Aspects savanna and of the ecosystems. B.H. Savannas. Walker Springer- Verlag, Berlin. Walker, B.H. and Goodman, P.S. 1983. Some implications of ecosystem pp. 79-91 properties for in R.N. Large Mammals wildlife management, Owen-Smith (ed.), Management of in African Conservation Areas. Haum, Pretoria. Walker, B.H,, reported by Lewin, R. 1986. In ecology, change brings stability. Science (Wash., D.C.) 234: 1071-1073. Walker, B.H, , Emails, R.H., Owen-Smlth, R.N. and Scholes, R.J. 1987, To cull or not to cull. Lessons from a southern African drought. J. Appl. Ecol. 241 381-401. Walther, F.R. 1972. Territorial horned ungulates, examples of behaviour in with special Thomson's and reference certain to the Grant’s gazelles. Zool, Atr. 7: 303-307. Western, D. on the 1976. Water availability and its influence structure and dynamics of a savanna large mammal community. £. Atr. WLldl. J. 13: 266-88. Westoby, M. 1974, The analysis of diet selection by large generalist herbivores. Am, Nat. 108: 290-304, White, G.C., and Biotelemetry Garrott, Data - R.A. A ms, Analysis Primer. of Unpublished manuscript. Whiten, A., and Barton, R.A. 1988, Demise of the checksheet: using off-the-shelf miniature hand-held computers for remote fieldwork applications, Trends Ecol. Evol. 3: 146-148, Whitmore, J.S, 1971. South Afr. J. Sol. 67: 166-176, Africa's water budget, 5. Wiens, J.A. 1976. population responses to patchy environments. Ann. Rev, Ecol.Syst. 7: 81-120. Wiens, D., and Rourke, J.P. 1978. Rodent pollination southern African Wiens, J.A, 1984. Protea spp. Chapter IS. Resource in 276: Nature (Lond.) systems, populations, and communities. Pages 397-436 in P,W, Pricei C.N. New Slobodchikoff, and W.S. Gaud (eds.). A Ecology. Novel Systems. John Wier, J.S. 1971. Theeffect water supplies Pages 367-385 The Approaches to Interactive Wiley anti Sons, New York. in E. Scientific of creating additional in a central African national park. Duffey and Management of A.S. Watt (eds.). Anima.l and Plant Communities for Conservation, Blackwell, Oxford, Wilson, V.J. 1865. Observations Tragelaphus strepsiceros control hunting scheme in on the Pallas greater from Northern a kudu tsetse Rhodesia. E. Afr, ttildl. J, 3: 27-37, Wilson, V.J. 1970. Tragelaphus Data from strepsiceros the culling Pallas in of the kudu Kyle National Park, Rhodesia Arnoldia (Rhod.) 4: 1-26. Wisdom, C.S., 1987, Gonzalez-Coloma, A., Ecological tannin proanthocyanidins, protein protein precipitating and assays. Rundel, P.W. Evaluation binding assays of and potential, Oecologia IBerl.) 72: 395-401. Wolff, J .0. 1980. during peak Hoose-snowshoe hare densities. hare competition Proceedings of the Ranger. The North American Noose Conference 16: 238-254. Wolhuter, H . Wildlife 1948. Memories Protection of Society a Game of South Africa, Johannesburg. Youngi E. 1972. The estimating wild value of waterhole animal populations. counts J. Sth, in Afr. Wlldl. Mgmt. Ass. 2: 22-23. Zar, J.H, Edition. 1984. Biostatistical Prentice-Hall, Analysis. Englewood Cliffs, Second New Appendix AL Giraffe (G920) home range and location points. Sci.lei 1 Appendix A2. Kudu home ranges and location points. Top overlay: K905 (west), K925 (east); bottom overlay: K910 (north), K900 (south). Scale: 1 grid square s 1 km2. T Appendix A3. Steenbok and impale home ranges and location points. Top overlay! ateenbok1 , S902 (west), S907 (east). Middle overlay: impale; 1912 (squares), 191 (circles). Bottom overlay: impaia; 1913 (circles), 1909 (triangles). Scale: 1 grid square a 1 km2. t h a t follows. Hut MM; pirtertigi fiejto} tl# illoMtus ' Suple s lii ie iKOidi, lidlcitles totil ftedlii} llee ricordei In taeti Cise. I J# u ih1 1 ,tiii i'rii tu ivs n m Mjst Hof till; Fercistisa f m U $ t t a illacitloi f6rls JcjcuJfjrfstH! Setirlfiji m s i lieir«l«li?s m m Crsswfl m sa tortfitee h irtm ii lucis jtmrtU Cr w e n i iwl ek t e J r c c jv rj t rr K t onrpw Irinii rimstrn ISrttn men Sclemm btmt SiitafjH NlilMrlM Hicii So rl t1 lu UPtiWl K H (tom Mrrffii tiMeeu rolHMli IntlH ilrnerw C i j« pj rii i M f J6 il e s i U l l lU 6 I « l l ( if itp Wt i K r iM t J Ciiiuliinnili l(«HmKsiiifemi JifMiieif/n Senniesiriresi Ci l r ii i i ii i if S tn f e r oi c i fi UiH f f UfCMii MMiMifem kicn lltthcptiHI Jcitd toelfiti* lim it ciflrs lias fiilu ll Jlilzll pHtrtim h ltiilm t tlriu m SfiVli f l l K S U t i I C U i l fllflllH tiem t s n l M U torn ejjlirl Slf/tteM leiieii i mieetlfled IijsiHifrltiM tiiiii Mtintm ftclH i i i t u m "" * 3I.M ,u? M.K M tH 566 li.H ».# tl.M !M3 ,U | Nc ttetili II. lenli ilet ee«pesltlM Flint species 3 loethiy pireintigi Win; tin illotition JH M T »pr Ml W »»} Sip 0» IW a.n U.M I1.5S Itiil k«(t Urtllls H.K lacitiljrtms Sttimitittjs ctmi iecMfley vim / ii IjNs igereiiti Hi/ims secejilecsfs Ml Hi m enriilfs Snrfi nntlHli 0,11 Smi thmcm Men Urtllls tom wlttUi Imhturfts ctfisst tupiris toHilosi EtorelM frtirte tenirttifl hitirmst Cmipm sdltptrl O.ti «,« Hltll Itirisli (to lit»roslicf;s cimt Frt BlMtil itotl Foi FnUWfW Ifljisl (tildes Fit Siritniirolltoll UllergU teltwto 1,1! I J } 1 ilium tuforuti"'’ omo o!ss Hi m Hlnddil lUldintlflii Ml 0.11 I tniFirs 0° 0.M : j llto ii ulfn !iS lelesntto sieeiisM Smli toulK in iimupw trmcuw Iculi sruiluiuU (it In SiriMm fl/tMdS M lltoi/lll Ml tuontii lossiilitiin SfimiitliH ifritui 1.06 Hied illitfci Stoll ilcrtHffsi kitii intSottlsei 0 V' (Matt puitrufsm Cmsunililii 5te»lisl»! Hilt ititt till mil KOI Kill Hill Mid 30)11 HIM Hill Mill Mill 1ftl, faltis; (rt, fwv, fV, ilw . 1 5«[1« sin le skos Is, lidituisj titil W i n tin records! in eich ei S|U(i AppenSii C. Kean duration of feeding events (In seconds). Sp MpwrituiC o a t i m s , ainffe Mu l,i. Sipelie itrlcm UMie stvhleiMlI lai’thacifta! sisessi Hem eerelfelii Hn/ltere Mchieie Hi/tim Mtrofkytit W ttm m tjilm is m rm tim li ittemis irtoarpst tricluarptt t i m i smUri Pelt#em ifric m t peylliMhvi reticulate; Prttaperejw ifissstfelatos Ptmctrm ratundtfolius Mu; sstiMlt scM li tnchtpstil'i s c ltm iw birrti Meerieeje tin u SulaflM paederjeferM 5»ir»itacl^s itrlem Strythnos spirrtse r«4mi« Mritie lereiMlie ierieee termhis obontt lim n ciffrt l i m it m rk m lUiphus m rm ti Ml wedy plenM Creeperi n .i * IumU neaniot NIM W«« 5.t. 1 , 1 • $.43 SM2 10,1 IW « 'Pliste lilted 6? loicias ire ill <006? pianti, •n li the fluebir cf feedlo} eviets racgriei in eieh ceie, ttt 0,42 UW 8.M 4 0,86 2474 &W * Author Du Toit Johan Truter Name of thesis Patterns Of Resource Use Within The Browsing Ruminant Guild In The Central Kruger National Park. 1988 PUBLISHER: University of the Witwatersrand, Johannesburg ©2013 LEGAL NOTICES: Copyright Notice: All materials on the U n i v e r s i t y of t he W i t w a t e r s r a n d , J oha n ne s bu r g L i b r a r y website are protected by South African copyright law and may not be distributed, transmitted, displayed, or otherwise published in any format, without the prior written permission of the copyright owner. Disclaimer and Terms of Use: Provided that you maintain all copyright and other notices contained therein, you may download material (one machine readable copy and one print copy per page) for your personal and/or educational non-commercial use only. The University of the Witwatersrand, Johannesburg, is not responsible for any errors or omissions and excludes any and all liability for any errors in or omissions from the information on the Library website. Author Du Toit Johan Truter Name of thesis Patterns Of Resource Use Within The Browsing Ruminant Guild In The Central Kruger National Park. 1988 PUBLISHER: University of the Witwatersrand, Johannesburg ©2013 LEGAL NOTICES: Copyright Notice: All materials on the U n i v e r s i t y o f t h e W i t w a t e r s r a n d , J o h a n n e s b u r g L i b r a r y website are protected by South African copyright law and may not be distributed, transmitted, displayed, or otherwise published in any format, without the prior written permission of the copyright owner. Disclaimer and Terms of Use: Provided that you maintain all copyright and other notices contained therein, you may download material (one machine readable copy and one print copy per page) for your personal and/or educational non-commercial use only. The University of the Witwatersrand, Johannesburg, is not responsible for any errors or omissions and excludes any and all liability for any errors in or omissions from the information on the Library website.