Zambezia (1987), XIV, (i).TSETSE CONTROL HUNTING IN ZIMBABWE, 1919-1958G. F. T, CHILDDepartment of National Parks and Wild Life Management, ZimbabweandT. RINEYFormerly, Forest and Forest Products Division, FAO, RomeTSETSE, Glossina spp., are blood-sucking flies and the vectors of trypanosomesthat cause sleeping sickness in man and nagana in cattle. Zimbabwe is one ofseveral countries in Africa where attempts have been made to controi these fliesby eliminating the large wild vertebrate hosts on which they depend for food.Spread of tsetse (mainly G. morsitans) in Zimbabwe in the early 1900s led topublic concern as it threatened several cattle-ranching areas. A Chief Entomologistwas appointed in 1909 to define the extent of the country-wide problem(Cockbill, 1964) and hunting commenced in the area of the Gwaai and Shanganiriver confluence in 1919.The declared aim of this and all subsequent government hunting operations,before 1960, was to eliminate large wild vertebrates (Jack, 1914). This action wastaken in the belief that a recent recession in the range of the tsetse had resultedfrom the rinderpest pandemic that swept through this part of Africa in 1896decimating wild and domestic ungulates. Summers (1967), supported by Child(1968a) and Campbell and Child (1971), using historical evidence, cast doubt onthis assumption, as the continuous ranges of the fly in the southern belt(Zimbabwe, the Transvaal and Botswana) and in the northern belt (Zimbabweand Botswana) were receding before the rinderpest. Summers goes further andconcludes, using early human settlement patterns, that this range had waxed andwaned previously.Tsetse control hunting operations in Zimbabwe have been well documentedand provide one of the few early indices of wildlife populations in the country.This article examines these returns between 1919 and 1957/8, especially themore detailed records from 1933, with a view to determining whether the huntingwas effective in eliminating large mammals and highlighting points of zoo-geographical interest relating to the species that were hunted. A secondary aim isto make this data, with its biological and management implications, moreaccessible.The data were extracted from the records of the Branch of the Tsetse andTrypanosomiasis Control of the Department of Veterinary Services. They are1112TSETSE CONTROL HUNTINGaugmented by field studies in the Sebungwe Area* between 1958 and 1964, byinterviews with past and present staff of the Branch and by comparisons withsimilar tsetse control hunting in Botswana (Child et al., 1970).TSETSE CONTROL HUNTING 1919-1933Table I summarizes the available returns and shows that early hunting wasdirected at the full spectrum of large mammals occurring in the hunting areas(Fig. I). Over 57,000 animals from more than 36 species were shot. Theyincluded hartebeest, presumably Alcelaphus huselaphus, gemsbok, Oryxgazellaand wildebeest, Connochaetes taurinus, taken in the Sebungwe or ShanganiReserve hunting areas, beyond the species' present range.Red hartebeest and gemsbok are typical of the Kalahari and do not generallyoccur in the Zambezi drainage (Child and Savory, 1964; Wilson, 1975; Smithers,1983). Wildebeest no longer occur in this part of Zimbabwe; the last one was shoton tsetse control in 1936. Hemans (1935) also recorded one on the headwaters ofthe Mzola river in this general region.Early hunting returns contain a relatively high proportion of animals that areeasy to hunt. For example, black rhinoceros, Diceros hicornis, made up 0.29 percent of the total kill of 57,293 before 1933, but only 0.03 per cent of the 602,041kills after that date. In the Darwin District, 61 were shot during the first nine yearsof hunting and two in the subsequent twenty-four years. This species is verysensitive to being hunted, as recruitment to the breeding segment of thepopulation is slow (Roth and Child, 1968). Black rhino are discussed further inthe next section.Other species shot in significantly higher proportions before 1933 includesable, Hippotragus niger, (14.5 per cent : 5.6 per cent; x2 = 2,386; P < 0.005),eland, Taurotragus oryx, (4.! per cent : 1.6 per cent; x2 = 786.4; P < 0.005), androan, H. equinus, (1.9 per cent : 0.8 per cent; x2 = 260.7; P < 0.005). All arelarge conspicuous antelope, while eland are highly mobile and well known toavoid disturbances such as those that would have accompanied tsetse controlhunting. Bushpig, Potamochoerusporcus, (5.4 per cent : 3.1 per cent; x2 = 311.9;P < 0.005) and kudu, Tragelaphus strepsiceros, (15.7 per cent : 13.2 per cent;X2 = 77.79; P < 0.005), were also taken in higher numbers prior to 1933,although the differences are less marked.A striking feature of these early hunting returns is the low number of elephant,Loxadonta africana, (2) and buffalo, Synceros caffer, (32) taken in periods of upto fourteen years in regions where both species are now common. The former isespecially noteworthy as elephant are generally held to be important mechanicaldispersal agents of tsetse. The scarcity of buffalo is remarked upon by Jack*Many place-names have been altered in Zimbabwe; those used in this article are those that werein use during the period under review.G. F. T. CHILD & T, RINEY13Figure 1; TSETSE CONTROL HUNTING AREAS, 1919-1961x...Pre-1933, including thehunting operations controlledby the Native Departmentduring 1929-1938So* 1:2 500 0000 25 » 75 100 12S 150 175 WfflmSource: D. Lovemore, former head of the Branch of Tsetse and Trypanosomiasis Control.Table ITHE RESULTS OF TSETSE CONTROL HUNTING BY SPECIES AND HUNTING AREAS BETWEEN 1919 AND 1933Ol_IB2 |CC xf aQ 3lOhaSooo>03QtrtCMOldQ5OuenOS55cOATOusKCSSo933S:wQOO-jtr>o>soQ/i.0I/}S3:.;.fflX"3<LUjSi §CO »-is05 2UjO<o>ITŽ6(BQUJa-MlinUl933EiephantRhinoZebraBush pigWarthogBuffaloHartebeestTsessebeWaterbuckReedbuckSableRoanElandKuduBushbuckImpala_6176158218831422749131632621,807926101195642266Š312092131,01261293435166_65175161_Š1781812111673238169._11051521,078511871157647104496131321572441,400Š_136685649331008252749132541973361__20340510392071,04926434Š8109663164__451243714111209191588136309_57541209844281281315145470194419Š1571211371___610931600710340397481455137584_873732204868825525346278Š12775673013320071990962847347564"5341101865_.1514982709413094292917Š765202562Š2441902718831703176792421321891,5793,0995,2493226113,6012,0038,5141,0872,3429,0092,0191,210lCOmmmOOzHXor~T.CZH20DuikerKiipspringerOribiGrysbokSteenbokWildebeestGemsbokBaboonMonkeyHippoGiraffeJacka!CaracalAntbearHyenaWiid dogCheetahLeopardLionCatMiscellaneous1,201107158512S107_.4I2111226359423426344Š69Š291223314211,01158139157Š189__62_8__13955 2,140 1,485 1,01536 147 44 153___ 3127 498 290 4919 4 91343634110915041133213491456276961212-321,30665127179_-22037_224855_337158_ŠŠ229562116Š3_ŠŠ1331781363231014144256102»11,8357414321,590564131982163336141611469631912386TOTAL6,087 4,754 2,897 4,179 6,949 4,973 4,937 4,951 2,375 5,170 3,849 2,655 3,51757,29316TSETSE CONTROL HUNTING(1914), who believed that this species took loeger than most to recover after therinderpest.Until 1932 tsetse hunting was often supervised by the local NativeCommissioner or detached Assistant Native Commissioner, as but one of theirmany responsibilities. This was considered unsatisfactory and from 1933 theChief Entomologist and his staff progressively assumed direct responsibility for allhunting operations.TSETSE CONTROL HUNTING 1933-1957/8This section is based on the regular monthly hunting returns submitted by the areasupervisors. While these data have limitations by virtue of having been assembledover twenty-six years by a number of individuals who were responsible for largeareas, they do provide a valuable record of the magnitude of tsetse control huntingduring this period. They also give an accurate indication of population trendsamong the species shot in substantial numbers, especially where this is supportedby complementary evidence.General hunting patterns by areasFigures 2-11 summarize the total kills from the thirteen areas (Fig. 1) which werehunted between 1933 and 1957/8, inclusive. In many cases the graphs show asimilar form over a given sequence of years. Sipolilo (Fig. 2), Mtoko (Fig. 3) andLomagundi North (Fig. 4) show the early crest followed by a trough in the firsttwo, which is characteristic of most areas, but otherwise these areas, Shangani(Fig. 5) and Gwaai (Fig. 2) are difficult to evaluate. Hunting was either of tooshort a duration or too few animals were shot for meaningful trends to emerge.Figure 2: TOTAL KILLS IN THE SIPOLILO AND GWAAI AREAS1 000-500-0Sipolilotiwaai1933194019501957/8Before about 1939 patterns were dominated by the initial crest and trough. Inthat year four of six major hunting areas showed a trough which is not evident foiGatooma (Fig. 5), while Lomagundi/Sanyati (Fig. 6) showed a small crest. Thistrough preceded a reorganization of the hunting in the Darwin (Fig. 7),Lomagundi North (Fig. 4), Sebungwe (Fig. 8), and perhaps other, areas in whichhunting had been in progress for some years.G. F. T. CHILD & T. RINEY17Figure 3: TOTAL KILLS IN THE MTOKO AREA000i5 000-4 000-3000-2000-1 ooo193319401950Figure 4: TOTAL KILLS IN THE LOMAGUNDI/DOMA ANDLOMAGUNDI NORTH AREAS3 00012 000-1 000-Lomagundi/DomaLomagundi North1933194019501957/818TSETSE CONTROL HUNTINGA peak in 1945 is evident for most areas, apart from Lomagundi/Sanyati(Fig. 6) and possibly Urungwe (Fig. 9), and is followed by a trough in all areas in1947. This coincided with a policy decision to cease hunting small antelope:steenbok, Raphicerus campestris, grysbok, R. sharpei, klipspringer, Oreotragusoreotragus, suni, Nesotragus moschatus, and oribi, Ourebia ourebi. Before that,these antelope contributed around 18 per cent of the total kill.Figure 5: TOTAL KILLS IN THE GATOOMA AND SHANGANI AREAS4 000-3 000-2 000-1 000 *-r1933194019501957/8Figure 6: TOTAL KILLS IN THE LOMAGUNDI/SANYATI AREA3 000-2000-1 000J-r1933194019501957/8G. F. T. CHILD & T. R1NEY19Figure 7: TOTAL KILLS IN THE DARWIN AREA000i2 000-1 0001933194019501957/8Figure 8: TOTAL KILLS IN THE SEBUNGWE AREA1 000800600-4002001933194019501957/820TSETSE CONTROL HUNTINGFigure 9: TOTAL KILLS IN THE URUNGWE AREA1 400"1 200-1 000-800-600-400200-1933194019501957/8Figure 10: TOTAL KILLS IN THE CHIRUNDU AREA100-0L_1933194019501957/8G. F. T. CHILD & T. R1NEY21Figure 11: TOTAL KILLS IN THE SABI VALLEY AREA400300'200100'0U-T-1933194019501957/8All areas had a peak in 1948 and 1949, followed by a dip in 1950 and a crestin 1950/1. The 1950 low resulted from a change in the recording year so that thereturns from that year were from only nine months, while the subsequent crestrepresented 15 months.High kills in all but two areas from 1953/4 to 1955/6 reflected a generalintensification of the hunting effort. In the Chirandu area (Fig. 10), for example,the number of hunters was doubled. The crest is absent for the Sabi Valley(Fig. 11), the only area in the south of Zimbabwe, and from Darwin (Fig. 7),where, according to CockbiH (1964), hunting was being phased out prior to itssuspension during 1957.A rapid drop-off in kills in all areas, except the Sabi Valley, during the last fewyears followed a Commission of Inquiry (chaired by E. Thomas) into tsetse andtrypanosomiasis (Federation of Rhodesia and Nyasaland, 1955). This recom-mended smaller hunting areas and more emphasis on fencing to create game-free/cattle-free corridors. Many hunters were diverted to fence building. Huntingwas suspended in most areas in favour of attempts to destroy tsetse habitats,through cutting of riverine vegetation in particular, and the use of insecticidessuch as DDT and dieldrin, but was resumed in 1964.The sychronous behaviour of these graphs and an ability to explain many ofthe major fluctuations in the number of kills in terms of recorded decisions22TSETSE CONTROL HUNTINGTable IITHE RESULTS OF TSETSE CONTROL HUNTING BY SPECIES AND YEARSBETWEEN 1933 AND 1957/8SpeciesElephantRhinoBuffaloZebraElandKuduRoanSabfeWaterbuckHartebeestWildebeestTsessebeNyalsImpalaReedbuckBushbuckDuikerSteenbokGrysbokKlipspringerOribiSuniWarthogBushpigBaboonLionLeopardCheetahHyenaWild dogCatJackalLynxMonkeyAntbearUnclassifiedTOTAL19332852734242,1522041,725707....31533966108484,2223139903491081,5194695993312921143982137214416.87119341564505442,8962452,016668Š.112206109669445,9064311,316422931,425548889141313202241361494616321,6871935628283704842,3341231,568534.__101465851,0358627,1215151,3984171061,2474961,33761337245852447824315022,616193625293193901.9591781.458355......11313808005977,3196801,2684591101,9714451,262642217463247169225321,62519371202802331,665841,124397,_.653197596576,9923141,7636122279384291,168516....19._.2_Š18,080193811062272051,46373891286_.492925406824,503651,3854141861,0262551,234315Š211...113,90619392282211531.34284664303205034596534,152851.6125421711,0772831.3362122127....813,7241940221163382631,786112851352497513995743,981971,5427581451,3722151,59641215131115,3291941112263534752,5772139955436572.....1,0795481,3864.8411372,2241,090212....1,6663951,3785181441.._Š...20,5101942....3343673933,3452671,2666287178._1,3716612,0246,0221292,5311,26519172,1574601,549716....22Š....Š25,14319432.._3023853563,2571971.24558950551,6515921,6746,0891062,5851,317225162,0114741,92771644251111125,157194413144184234153,2811511,160572101521,6126681,8936,5962373,4561.383162472,3474351,7943201104....1-27,269194512254653813423,55115994049889559_.1,5957611,8887,5181653,177912187402,7365062,0292201176-._...28,086194615192762273103,289100854361709611,6036271,8196,2822032,275517104332.8865161,84492316S2124,351affecting the hunting patterns suggests that the overall kill was influenced largelyby prevailing policy. This is reinforced by the similarity in the form of these graphsand the patterns of kills from individual species.Hunting by speciesTable II summarizes the kills throughout Zimbabwe of individual speciesbetween 1933 and 1957/8 when some 602,041 animals from over thirty-sixspecies were shot. Before discussing the species shot in high numbers it isinteresting to examine the returns of those taken in low numbers.Elephant: Elephant kills by area and in five- or six-year periods are summarized inTable III. As noted earlier, only two had been shot prior to 1933 and kills werelow before 1943, when numbers were generally low in the Zambezi basin inZimbabwe (Cockbill, 1964). From 1943 onwards numbers increased (Fig. 13)G. F, T. CHILD & T. RINEY23Table II (continued)SpeciesElephantRhinoBuffaloZebraElandKuduRoanSableWaterbuekHartebeestWildebeestTsessebeNyalaImpataRe»dbuckBushbuckDuikerSteenbokGrysbokKllpspringerOribfSuniWarthoflBushplgBaboonLionLeopardCheetahHyenaWild dogCatJackalLynxMonkeyAntbearUnclassifiedTOTAL194781?2761502743,27716691732396Š69Š1,1617761,4753,262Š8Š25Š2,8485571,072317_1410_Ł1ŠŠ_18,802194833103433142873,8631751,062371103_92Š1,9437201,7114,935142Š1Š3,2967372,079Ł 31921181012_ŠŠ622,16019494233932262543,9532681,00547939_104Š2,4776631,8416,749ŠŠŠŠŠ3,6288131.8847181915Š1_ŠŠŠ24,8711950583352192253,14314289134045Š58Š2,1704761,2975,280Š_ŠŠŠ2,8365051,39411339Ł 8ŠŠ__ŠŠ19,4911950/1884892563354,4892021,098425133Š78Š2,3667301,9888,128ŠŠ_ŠŠ3,8718071,83872311338ŠŠ_ŠŠŠ27,3291951/22053452S92673,4251511,05630580Š57_1,9906481,6117,186Š2314Š13,7315751,77111031527Š3__Š23,5801952/32783462782243,6312401,04924536Š9112,0046251,6238,9771603003141223,8781,1242,503114_2444Š__Š_27,4811953/45583773132914,9373061,35130150Š8082,2737671,79412,7112771,207963134Š4,5039503,219419_3561Š__ŠŠ_37,0001954/59084603383044,8391901,53929571Š7812,3027282,05613,6603841,9451,385137Š4,7511,3655,01522024279Š__Š_41,8861955/612144263042664,2471791,168343282238_1,903§152,2529,239'60548316714,7041,7215,41412323fl39_____ŠŠ34,2071956/794117135831.83056427907144471561.1948,2851....144.....3,0601,2872,592_1112521._...._.17.9691957/85414175541.6416531761Š17145391141,047TOTAL6122056,4077,4817,85177,9724,33028,63710,3711,134432,1662434,32216,64436,3905,180 173,13644_71_..2,7126832,07126._.183014,9114,37731,55913,4562,65814767,89717,14050.79410842034322S011,444744152,15915526602,041and authority was given for the species to be hunted, even in areas where it hadpreviously been scarce.Most elephant were shot by the area supervisors, who were paid a bounty of£5 for each animal killed between 1952 and 1956. This encouraged huntingoutside the prescribed hunting areas and one former supervisor admitted doing soafter having had to search for many miles to find elephant where they had becomenumerous by the late 1970s.With improved communications and supervision of hunting at all levels,excursions outside the hunting areas became more difficult and the bounty systemwas abandoned. Despite this the annual off-take continued to rise, reflecting thegreater availability of elephant, either inside or in close proximity to the huntingareas. This is indicative of increasing numbers of elephants in the Zambezi Valleyfrom about 1944, This increase is supported by historical evidence (e.g.24TSETSE CONTROL HUNTINGGumming, 1981) and has been traced by Child (1968a) and Campbell and Child(1971) for adjacent parts of Botswana, where it started around 1945.Hunting commenced in the Sabi Valley in 1941 and the first elephant (12)were shot in 1944. None were taken between 1945 and 1952/3, but from then onthe annual off-take amounted to 10, 9, 25, 50 and 37. This is a region in whichelephant were not common in 1931, but where the species is now numerous andwhere several thousand head have been culled since 1971 to protect the woodyvegetation in the Gonarezhou National Park.Table IIIELEPHANT SHOT ON TSETSE CONTROL THROUGHOUT ZIMBABWE BYFIVE- AND SIX-YEAR PERIODS BETWEEN 1933 AND 1957/8Hunting AreaSebungweUrungweGatoomaLomagundi/SanyatiLomagundi/Doma/NorthChirunduSabiDarwinMtokoTOTALKill per 10,000 animals shotPeriod1933-1938361000000101.061939-194300031100050.61944-19480198015112620818.41949-1952/313187022102391022.31953/4-1957/88159041222.131110441430.1Total9710216750251439163612_Gumming (1981) shows that elephant numbers in the less-densely-peopledparts of Zimbabwe have increased dramatically from an estimated country-widepopulation of the order of 4,000 head in 1900 to between 50,000 and 55,000by 1981.Black rhinoceros: The relatively high proportion of black rhino in the total killprior to 1933 has been noted. Of the 205 rhino shot in the next twenty-six years,149 (72 per cent) were in the Urungwe, with 40 being killed between 1933 and1938, as hunting operations became established, and 82 between 1944 and 1948.In 1944, and again in 1946, the Urungwe hunting area was extended into areas ofgood rhino habitat, where the species is still well represented, having apparentlyrecolonized parts of the former hunting area. No rhino were trapped on islandsalong the eastern shores of Lake Kariba when the dam filled between 1958 and-. Q i Child (1968b) attributes this to tsetse control hunting that took place. ŁŁŁ-.:.Ł:? !^.,f-;^-J 1.956.G. F, T. CHILD & T. R1NEY25Elsewhere in the north of Zimbabwe, rhino kills declined from 27 between1933 and 1938 to between 3 and 13 animals in subsequent five-year periods. The13 coincided with the general intensification of hunting between 1953/4 and1957/8, when at least one was shot well outside a hunting area by a recentlyrecruited supervisor intent on securing a trophy. The species was apparentlyalready locally extinct in the Sabi Valley hunting area when operationscommenced there in 1941.Child (1968b) describes how two black rhino that were in the Nagupandehunting area broke out through the perimeter game fence within a few months ofthe start of hunting in this area in October 1961. It is safe to conclude that tsetsecontrol hunting decimated local black rhino populations, although the aboveobservations suggest that a few escaped by leaving the prescribed hunting areas.Tsessebe and hartebeest All 1,134 Lichtenstein's hartebeest, A Icelaphus lichten-sieini, recorded in Table II were from the Sabi Valley on the edge of the species'range, where it extends into Zimbabwe from Mozambique. After about i 949,fluctuations in hartebeest kills showed little relationship to the total kill from thearea and may have been due to changes in local hunting policy.Hartebeest seem sensitive to both habitat changes and hunting. It is now anendangered species within* Zimbabwe, and 66 head were reintroduced into theGonarezhou National Park from Mozambique in 1972 (Rhodesia, 1973). Theseanimals were liberated to the north of the Lundi river into a portion of the formerhunting area.Although tsessebe, Damaliscus lunatus, occupy a similar niche to that of thetwo species of hartebeest (A. Uchtensteini and A. buselaphus), the three speciestend to replace each other geographically (Dowsett, 1966; Child etai, 1972). Allbut eight of the 1,516 tsessebe shot on tsetse control from 1937 onwards camefrom the Sebungwe. It is a conspicuous and easily hunted species. Figure 21indicates a similar sequential pattern of annual kills to that of other species fromthe Sebungwe with a marked peak in 1955/6, a response to intensified huntingdue to the posting of a supervisor to Chiware (Cewali) for the first time in 1955.By 1964 there were few tsessebe in the Sebungwe hunting area. This mayhave been due to the modification of their habitats by the seasonally early burningof the range that accompanied tsetse control hunting (Riney, 1963b). Child et ah(1970) have indicated that the bush encroachment that results from such burningcan be detrimental to tsessebe populations.Wildebeest: Wildebeest shot in the Sebungwe (16) and Shangani (9) prior to1936 have teen discussed. In addition 18 were taken in the Sabi Valley between1945 and 1957/8 where the species still occurs in small numbers with a patchydistribution.Nyaia and sunk The ranges of nyala, Tmgelaphus angasi, and suni extend into the26TSETSE CONTROL HUNTINGsouth-eastern lowveld of Zimbabwe from Mozambique, and 24 and 147,respectively, were shot in the Sabi Valley. In addition there are two suni skulls inthe National Museum, Bulawayo, from tsetse control operations in the GacheGatche area of Urungwe. These and the horns of two males taken in about 1939by the late A. H. Pingstone in the Dande, midway between the Zambezi riverand the southern escarpment, are the only known records of this small antelopefrom the Zambezi Valley in Zimbabwe.Steenbok, grysbok, klipspringerandoribv As already noted, these small antelopewere not shot as a matter of policy between 1947 and 1951 /2. Grysbok was shot inhighest numbers and is discussed again below.When the hunting of these species was resumed after 1951 /2, kills were highin areas with suitable habitats, indicating that the number shot was a directreflection of the hunting effort against them. Wilson and Child (1965) suggest thatsimilar tsetse control hunting over a two-year period in Zambia may havestimulated population growth leading to the early stages of over-population ofklipspringer.Most oribi were taken from Urungwe (74 per cent), Lomagundi/Doma (11per cent), and the Sabi Valley (7 per cent), with smaller numbers from Gatooma,Lomagundi/Sanyati, Darwin and Sipolilo. This pattern of kills conforms with therange of the species described by Child and Savory (1964).Large predators: The distribution by time and area of lion, Panthera ieo, leopard,P. pardus, cheetah, Acinonyxjubatus, hyena, probably all Crocuta crocuta, andwild dog, Lycayonpictus, shot on tsetse control are summarized in Table IV. Withfew exceptions, these animals were taken in all the hunting areas.Main species hunted- Fourteen species were shot in high numbers in all or most ofthe hunting areas during the twenty-six-year period. The annual kills of theseanimals and elephant, tsessebe and grysbok are graphed in Figures 13-29. Wherethe off-take of a given species was high and might have influenced the shape of thegraph of the total kills (Fig. 12), it is compared with the total kill less the off-takefrom the species under consideration.These graphs show a remarkably consistent form for all species. Many of thecommon deflections also appear in the total kills from the various hunting areasand are readily explained in terms of recorded changes in hunting intensity oralterations in the limits of hunting areas.Graphs for twelve of the sixteen species show a clear initial crest and trough,and to these may be added tsessebe (Fig. 21). This pattern is similar to that foundfor the total kills from individual hunting areas and is to be expected after thereorganization of the hunting in 1933 or after its introduction into new areas. Asimilar pattern has been reported by Child et al (1970) for both the otf-take andthe hunter success rate for most species during the first few years of hunting in aTable IVTHE NUMBER OF LARGE PREDATORS, INCLUDING HYENA, SHOT ON TSETSE CONTROLBY HUNTING AREAS BETWEEN 1933 AND 1957/8HuntingArea 1ChirunduDarwinGatoomaLomagundi/YearsLionlunted Total152621Dorna/North 26Lomagundi/SanyatiMtokoSabiSebungweShanganiReserveSipoliloUrungweOtherTOTAL26718263826Š36756184551210108No.per year0.20.20.30.20.20.10.41.71.70.10.8ŠŠLeopardTotal114824303516724542142420No.per year0.71.81.11.21.32.10.40.91.70.58.3Š_CheetahTotal07233137008034No.per yearŠ0.30.10.10.10.10.20.3ŠŠ0.3ŠŠHyenaTotal9383102159119001782322No.per year0.61.50.10.40.88.40.63.5Š0.13.0ŠŠWild dogTotal2056101030501015530011317501No.per year1.32.20.50.41.27.10.66.010.0Š4.3Š_AS!Total4315546589512739321406434__No.per year2.96.02.22.23.718.12.212.313.30.816.7PT!oXr~O«~Jr>JDŁ»rfL.m<28TSETSE CONTROL HUNTINGFigure 12: TOTAL KILLS IN ALL HUNTING AREAS40 00030 00020 00010 0001933194019501957/8Figure 13: ELEPHANT KILLS IN ALL HUNTING AREAS1501<XH500Jj «1933194019501957/8G. F, T. CHILD &. T. RINEY29Figure 14: BUFFALO KILLS IN ALL HUNTING AREAS400-200-1933T I I T Ł I Ł 119401 I '1950T I1957/8Figure 15: ZEBRA KILLS IN ALL HUNTING AREAS4002001933ŠjŠrŠiŠiŠiŠrŠiŠiŠfŠiŠiŠiŠiŠiŠ\ŠiŠiŠr~l1940 1950 1957/830TSETSE CONTROL HUNTINGFigure 16: ELAND KILLS IN ALL HUNTING AREAS6001400-200193319401950Ł 11957/8tsetse control area in Botswana. Baboon, Papio ursinus, (Fig. 26) was anexception and these deflections were poorly-developed for elephant (Fig. 13),buffalo (Fig. 14) and grysbok (Fig. 29). The first two were, however, shot in lownumbers before 1940, while baboon and grysbok would have been less-favouredprey while hunting remained relatively easy. Apart from this exception, grysbokkills showed a close correlation with those of other species before the hunting ofthese small antelope was suspended in 1946. .The characteristic crest between 1940 and 1946 was only partially due to theintroduction of hunting in the Sabi Valley and Chirundu areas, as it is evident formost of the major hunting areas where it was attributable to some reorganizationof the hunting. The crest is present for all sixteen species, but the following low in1947 is weak in kudu (Fig. 17) and warthog, Phacochoerus aethiopicus, (Fig. 27),and absent for sable (Fig. 20), roan (Fig. 18), reedbuck, Redunca arundinum,(Fig. 23), and bushpig (Fig. 28). Returns for 1941 for several species are a mean ofthe 1940 and 1942 figures and should be used with caution as it is possible thatsome data were lost.The change in the recording year in 1950 led to a dip followed by a crest forG. F. T. CHILD & T. R1NEY31Rgum 17; KUDU KILLS IN ALL HUNTING AREASSpecies kili5 00014 000-30002 0001 000 JKill fromall other species36 00032 000Ł28 00024 00020 00016 00012 000all species, similar to that noted for total kills by areas. Increased hunting intensityaround 1954 produced the expected response from all species with record killsachieved in nine of the sixteen species. Of the remainder, sable kills (Fig. 20) werehighest at the beginning of the twenty-six-year period; reedbuck (Fig. 23) werehigh then, but more so in 1945 and 1947, while the highest impala, Aepyceros32TSETSE CONTROL HUNTINGFigure 18: ROAN KILLS IN ALL HUNTING AREAS4002001933!940' T19501957/8Figure 19: WATERBUCK KILLS IN ALL HUNTING AREAS800600 -400-200-1933I '194019501957/8G. F. T. CHILD & T. RINEY33Figure 20: SABLE KILLS IN ALL HUNTING AREASKill fromall other speciesf= 0.298; n = 24; n.s.r= 0.713; /»= 10 (1946-1958); P<0.01Species kill3000-2 000-1 000-0-Ir~rŁ40 000Ł36 000Ł32 00028 00024 00020 00016 000193319401950I i ! I1957/812 00034TSETSE CONTROL HUNTINGFigure 21: TSESSEBE KILLS IN ALL HUNTING AREAS300193319401950T~"l1957/8With the exception of eight tsessebe recorded from the Gatooma Area in 1952/3, theserecords came exclusively from the Sebungwe Hunting Area.melampus, (Fig. 22), and buffalo (Fig. 14) kills were between 1948 and 1950/1,and between 1945 and 1950/1, respectively. Eland (Fig. 16), zebra, Equushurchelti, (Fig. 15), and waterbuck, Kobus ellipsiprymnus, (Fig. 19), whilereflecting the immediate crests and troughs common to other species, weregenerally shot in diminishing proportions towards the end of the twenty-six-yearperiod. Implementation of some of the recommendations in the Thomas Report(Federation of Rhodesia and Nyasaland, 1955) resulted in a dramatic decline inthe kills from all sixteen species (apart from bushpig and elephant in thepenultimate year), similar to the overall kills from areas in the Zambezi basin.The high degree of synchronization in the pattern of annual kills of thesixteen main species hunted on tsetse control, and between these and the totalannual kill from the individual hunting areas, indicates that the results of thehunting were a reflection of the pattern and intensity of the hunting. It is nowappropriate to examine the kills from individual species from two selectedhunting areas to establish whether or not they conform to these patterns.G, F. T. CHILD & T. RINEY35Figure 22: IMPALA KILLS IN ALL HUNTING AREASKill fromall otter speciesSpecies kil2 50012 0C0-1 500-1 000 1r 40 000- 38 000- 32 000Ł 28 00024 000- 20 000Ł18 000L12 00019501957/836TSETSE CONTROL HUNTINGFigure 23: REEDBUCK KILLS IN ALL HUNTING AREASKill fromall other speciesSpecies kill10 0008 0006 0004 000-2 000Ł40 000-36 000(-32 00028 00024 00020 00016 0001ŠiŠiŠiŠiŠiŠ[ŠiŠr1933 1940T~l195012 0001957/8G. F. T. CHILD & T. RINEY37Figure 24: BUSHBUCK KILLS IN ALL HUNTING AREASKill fromal! other speciesr- 0.745;/? = 24; P< 0.001Species kill3 00020001 0001933Łr i » i T i i i i r i j i i ITTI i |1940 1950 1957/840 00036 00032 000Ł 28 00024 00020 00016 000^ 12 00038TSETSE CONTROL HUNTINGRgure 25: DUIKER KILLS IN ALL HUNTING AREASKill fromall other speciesSpecies kill12 00010 0008000600040002000r« 0.752; n- 24; P< 0.0011933T T~ 1 J I T T194019501967/8Ł32 000Ł28 000Ł24 000Ł20 00016 00012 0008 000G. F. T. CHILD & T. RINEY39Figure 26: BABOON KILLS IN ALL HUNTING AREASKill froma! I other speciesSpecies kill5 0G0-4 000-3 0002 0001 000r= 0.670; n = 24; P< 0.001w36 00032 00028 00024 000-20 000-16 000JTT1933M2 000194019501957/840TSETSE CONTROL HUNTINGFigure 27: WARTHOG KILLS IN ALL HUNTING AREASSpecies killSOOOi4000-300020001 000-r-0.659; n = 24; P< 0.001i ŠiŠiŠrT~TŠiŠiŠiI ! IŠIŠ1Š1Kill fromall other species36 00032 00028 000Ł 24 000Ł 20 000Ł 16 000M2 0001933194019501957/8G. F. T. CHILD &T.RINEY41Figure 28: BUSHPIG KILLS IN ALL HUNTING AREASKill fromail other speciesr= 0,678; n = 24; P< 0.001Species kill1 500-1 000-40 00036 00032 00028 000Ł24 000Ł20 000Ł16 000L12 0001957/842TSETSE CONTROL HUNTINGFigure 29: GRYSBOK KILLS IN ALL HUNTING AREASSpecies kii!3 0002 0001 000-Kiii fromaii other species24 000Ł20 000r= 0.892; n= 14 (1933-1946); P<0.011ŠiŠr16 000L12 0001933194019501957/8Note: Hunting of smal antelope was suspended in 1946.Results of tsetse control hunting in two areasThis section deals mostly with the Urungwe and Sebungwe areas, both of whichare in the Zambezi drainage. The Branch of Tsetse and Trypanosomiasis Controlconsidered the hunting returns from these two areas, particularly those fromUrungwe, to be most complete. In addition, the authors were able to collectancillary information from the Sebungwe from late 1958 until early 1960 andfrom the Nagupande hunting area between 1962 and 1964.The total country-wide annual kills from all species are compared in Figure 30with those of the species shot in substantial numbers in the Urungwe andSebungwe during the twenty-six-year period. Obviously, in this and subsequentgraphs (Figs. 30-44), a relatively high kill-rate from one area or of a particularspecies could influence the form of the graph of the total kill with which the areaor species kill is being compared. To obviate this, the area or species kills arededucted from the total kills for comparative purposes.Even had hunting intensity remained uniform in the areas, fluctuations inG. F. T. CHILD & T. R1NEY43annual kills would be expected in reponse to such obvious variables as annualrainfall. Despite these variables, the pattern of kills of most species over givenperiods of time were remarkably similar to each other. The main exceptions in theUrungwe were baboon (Fig. 44), particularly before 1945, zebra (Fig. 32) before1939, eland (Fig. 33) and reedbuck (Fig. 39).In the Sebungwe the main divergencies were again baboon (Fig. 44) before1945, and eland (Fig. 33). Although the pattern of eland kills was similar to thatin other areas it showed a poor relationship to that in the Urungwe and to the totalkill. In both areas buffalo (Fig. 31) were shot in low numbers during the first eightto ten years. Bushpig (Fig, 43) were taken in relatively uniform numbers before1951/2 and the kills of this secretive species, which is especially difficult to hunt,did not show the marked fluctuations in later years which are evident for mostspecies.Both areas showed a general upward trend in the annual kills of baboon(Fig, 44), bushpig (Fig. 43), warthog (Fig. 42), duiker, Sylvicapra grimmia,(Fig. 41) and impala (Fig. 38), culminating in the 1952/3 to 1955/6 crest, whichcoincided with the intensification of hunting before aspects of the Thomas Reportwere implemented. Similar upward trends are also evident for the Urungwereedbuck (Fig. 39), bushbuck (Fig. 40) and buffalo (Fig. 31) kills and for theSebungwe kudu (Fig. 34) and, perhaps, roan (Fig. 35) kills. This pattern cannotbe reconciled with populations being hunted to local extinction.No obvious trends emerge from the zebra (Fig. 32), reedbuck (Fig. 39),bushbuck (Fig. 40), eland (Fig. 33), tsessebe (Fig. 21), or buffalo (Fig. 31) returnsfrom the Sebungwe, or from the figures for sable (Fig. 36), roan (Fig. 35) andkudu (Fig. 34) from the Urungwe. On the other hand, zebra (Fig. 32), waterbuck(Fig. 37) and eland (Fig. 33) from the Uruogwe, and waterbuck (Fig. 37) and,perhaps, sable (Fig. 36) from the Sebungwe were killed in declining numberstowards the end of the twenty-six-year period. It is possible that the numbers ofthese species were being gradually reduced, although none had been eliminatedby the hunting.Similarities in the patterns of most of these graphs of the kills from individualspecies in the Urungwe or the Sebungwe hunting areas and those from the samespecies, and the total off-take from all other areas, are further evidence that majordeflections in hunting successes were due to changes in country-wide huntingpolicy.Hunting In the SebungweThe area: Hunting in the Sebungwe commenced in 1919 and from then until1933 it was divided into four units (Table I), before the hunting was reorganizedas a single entity (Fig. 1). Its limits were slightly modified without changing thearea significantly when a second supervisor was assigned to the area in 1949. By1956 there were three supervisors responsible for subdivisions of the area.44TSETSE CONTROL HUNTINGFigure 30: TOTAL KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM TOTAL KILLS IN ALL AREAS45 00040 00035 00030 00000020 00015 00010 0005 000Total kill andtotai kill less area kills12 00010 0008000600040002 0000J 0jTotal less U rung weTotal less SebungweSebungweU rung we19331940195011957/8G. F. T. CHILD & T. RINEY45Figure 31: BUFFALO KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM BUFFALO KILLS IN ALL AREAS40030020011001933194019501957/8Total less SebungweTotal less U rung weU rung weSebungweFigure 32: ZEBRA KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM ZEBRA KILLS IN ALL AREAS4001300200 \100}, Total less SebungweTotal less Urungwe\ \ Urungwe*" Sebungwe1933 194019501957/846TSETSE CONTROL HUNTINGFigure 33: ELAND KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM ELAND KILLS IN ALL AREAS500400-300200-1001Total less U rung we° Total less SebungweSebungweUrungwer~TŠiŠ[ŠiŠiŠtŠiŠiŠiŠiŠiŠi iŠiŠiŠJŠiŠiŠi iŠi i i i i1933 1940 1950 1957/8The game fence was surveyed and constructed in 1957, and huntingoperations, which had been sporadic during the year, were resumed between thisfence in the north and a cattle fence in the south (Fig. 45) in terms of the ThomasReport (Federation of Rhodesia and Nyasaland, 1955) Š the aim being to createa game-free/cattle-free corridor; but even this hunting was suspended in 1960.Hunting directed against warthog, bushpig, kudu, bushbuck, buffalo andelephant recommenced in the 540 km2 Nagupande area in October 1962 (Childet al.j 1965; Child et al., 1968). This rectangular area overlapped the old gamefence into the northern fringe of the Chiware (Cewali) subdivision of the formergame-free/cattle-free corridor that had not been hunted for eighteen months. Therest of the Nagupande area was within the pre-1957 Sebungwe area and had notbeen hunted for about six years. The area was visited frequently during the firstG. F. T. CHILD & T. RINEY47Figure 34: KUDU KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM KUDU KILLS IN ALL AREAS4 0003 0002 0001 000, Total less Sebungwe'DTotal less U rung we\*^ UrungweSebungwe1933194019501957/8Figure 35: ROAN KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM ROAN KILLS IN ALL AREAS3001200100Total less UrungweTotal less SebungweSebungweUrungwe *1933194019501957/848TSETSE CONTROL HUNTINGSABLE KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM SABLE KILLS IN ALL AREASTotal less UrungweTotal less SebungweUrungweSebungwe194019501957/8t\v a half years of resumed hunting and this provided a useful opportunity toas 'w residual effects of the hunting that had taken place between 1933 andIi. " techniques and their effect on game populationsfThe following accounto ganization of tsetse control hunting is based on the arrangement in theC; :Ł ts ubdi vision from 1958 to 1960, but was fairly typical of that in all tsetsecor.. areas. Information obtained on the ground was supplemented from there . J - < f the Branch of Tsetse and Trypanosomiasis Control and by interviewingst: . -nd former employees of the Branch.Ł ce of hunters was deployed in rotation in groups of two or three to fixedc\ located so as to achieve optimum coverage of the area but taking intoa. ti places likely to be attractive to animals. Hunting took place on abouttv^ :;;y-five consecutive days each month, before the hunters reported to theirSL , tsors and proceeded on time off. They were permitted to sell the meat fromth i.ials they shot and this and their poor wages were intended as anir '-cm to greater hunting effort.G. F. T. CHILD & T. RINEY49Figure 37: WATERBUCK KILLS IN THE SEBUNGWE AND URUNGWEAREAS ISOLATED FROM WATERBUCK KILLS IN ALL AREAS600-i500-400300-200-100-IVw v \/\o Total lessSebungweTotal less Urungwe;U rung weSebungwe[ŠiŠiŠiŠiŠiŠiŠ|ŠrŠTŠiŠiŠrŠiŠr~~tŠiŠ|ŠiŠiŠiŠiŠiŠr~~\Š|1933 1940 1950 1957/8VThe game fence was about two metres high and consisted of high-tensile steelwire on wooden posts with wooden droppers. White metal discs suspended alongthe wires, against which they rattled, were designed to make the fence moreconspicuous to animals, but its effectiveness was limited. This is not surprising asChild et al. (1965) and Child et al. (1970) indicate that warthog in theNagupande, and most species hunted on tsetse control in Botswana, crossedthrough such fences. Lewis and Wilson (1977) present numerical data of theeffectiveness of a similar fence in eastern Zambia. They found that such fences areof little hinderance to small species like warthog and duiker, although they mayinhibit movements of large animals such as sable, roan and eland.50TSETSE CONTROL HUNTINGFigure 38: MPALA KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM IMPALA KILLS IN ALL AREAS50(h2 000-1 500-1 000500Total less U rung weJK-° Total less Sebungwe'^* Sebungwe" Urungwe1933194019501957/8Wildebeest were apparently never common in the Sebungwe region andseem to have been eliminated from the tsetse control hunting area by 1936. Noother species, with the possible exception of black rhinoceros, was eliminatedfrom the hunting area, although waterbuck and possibly sable were shot indiminishing numbers towards the end of the twenty-six-year period. Thisevaluation is supported by observations in the Nagupande area, at least in thatportion rested from hunting for six years. Of the fourteen more numerous speciesin the 1933 to 1957/8 returns, waterbuck were scarce and zebra were seen in onlysmall numbers. Eland, sable and roan were reasonably common north of the oldgame fence, where sable and roan, at least, were associated with preferred habitats(Child and Wilson, 1964b; Joubert, 1976). As noted, there were also at least twoblack rhinoceros.G. F.T. CHILD &T.RINEY51Figure 39: REEDBUCK KILLS IN THE SEBUNGWE AND URUNGWEAREAS ISOLATED FROM REEDBUCK KILLS IN ALL AREAS1 000800600400200Total less Urungwe\» SebungweTotal less SebungweUrungwe1933194019501957/8The distribution of reedbuck was limited to the rather restricted areas ofsuitable habitat, but here the species was numerous even after a respite of onlyeighteen months from being hunted. Baboon, grysbok, impala and duiker werenoted in numbers after six years without hunting. During the first twelve monthsof the resumed hunting in the Nagupande area 85 bushpig, 1,357 warthog, 33bushbuck, 169 kudu, 45 elephant and 22 buffalo were shot. Immediately prior tothis hunting five parties of sportsmen were permitted to hunt the area and killedan additional 16 elephant, 17 buffalo and a range of lesser species. The elephantand buffalo had apparently penetrated from the north, but a fair proportion of theother species were taken in the south where previous hunting had ceased in 1960.Child and Wilson (1964a) examined the duiker population in the Chi warearea twenty-eight months after hunting had ceased and concluded that the specieswas exceptionally numerous with a high proportion of young animals. The52TSETSE CONTROL HUNTINGFigure 40: BUSHBUCK KILLS IN THE SEBUNGWE AND URUNGWEAREAS ISOLATED FROM BUSHBUCK KILLS IN ALL AREAS000i1 600'1 200'800400-\Total less SebungwebTotal less Urungwe\ Ł Urungwe*»*'* Sebungwe1933194019501957/6hunting returns suggested an expanding population and the later evidence wasconsistent with this postulation. Steenbok were also very common in the samearea, although it is on the fringe of the species' geographical range (Child andSavory, 1964) and tsessebe were not uncommon in appropriate habitats.As hunting had clearly been unsuccessful in eliminating the large mammalsagainst which it was aimed, it is useful to examine possible reasons.Most of the hunting was done by locally recruited tribesmen who lived at nogreat distance from the hunting area. Many had been employed as hunters formany years, when the effectiveness of the hunting techniques was evaluated overa fifteen-month period in 1959 and j960. They were generally good hunters andtrackers but poor marksmen, and their shooting ability was not improved by theG. F.T. CHILD &T.RINEYFigure 41: DUIKER KILLS IN THE SEBUNGWE AND URUNGWE AREISOLATED FROM DUIKER KILLS IN ALL AREASTotal kill less area killsArea kills10 000-9 0008 0007 000-6 000-5 000-4 000-854321000ooo-ooo-000-000-000-0-193319401950A = Sebungwe B = Total less Sebungwe C = Urungwe D = Total less.54TSETSE CONTROL HUNTINGFigure 42: WARTHOG KILLS IN THE SEBUNGWE AND URUNGWEAREAS ISOLATED FROM WARTHOG KILLS IN ALL AREAS400030002000t 000-Total less UrungwefTotal less SebungweŁSebungwe/ Ł Urungwe19331940_, , , ,Š-, 1 , fŠ-j , ,___—_19501957/8Figure 43: BUSHPIG KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM BUSHPIG KILLS IN ALL AREAS1 5001000500* Total less SebungweTotal less UrungweUrungwe-* Sebungwe19331940-fŠf-r1950-r-11957/8G. F. T. CHILD &T. RINEY55Figure 44: BABOON KILLS IN THE SEBUNGWE AND URUNGWE AREASISOLATED FROM BABOON KILLS IN ALL AREAS5 0004 0003 0002 0001000Total less UrungweTotal less SebungweUrungweSebungwe1933194019501957/8worn state of the ex-military 0.303 rifles with which they were issued. A 45 cmgrouping of shots from 25 metres was considered reasonable and few claimed tobe able to hit a 30 cm diameter tree at 100 paces with confidence. Most did notattempt to shoot even large antelope at ranges of over 75 to 100 metres.These hunters usually left their semi-permanent camps at dawn, returning fora meal around midday before hunting again from about 14.00 hours until dusk.Small species like grysbok, steenbok, duiker and bushbuck were located by sight,baboons by their vocalizations, while larger species were tracked, sometimes formany kilometres. This hunting pattern meant that most effort was within fivekilometres of the camps, except when the spoor of larger animals was beingfollowed, with the hunting intensity in the intervening areas being light butvariable, depending on species.The kill-rate was also heavily influenced by the number of hunters employedat any given time and by their meat requirements. One supervisor recorded the56TSETSE CONTROL HUNTINGFigure 45: THE SEBUNGWE TSETSE CONTROL HUNTING AREAG. F. T. CHILD &T. RINEY57day in the twenty-five-day monthly hunting period when particular animals wereshot. A regular pattern emerged in which hunters tended to shoot two or threesmall animals early in the month, so as to justify their employment and satisfytheir immediate food requirements in the field. One large animal was usuallykilled towards the end of the twenty-five-day period so as to provide a goodsupply of meat to take home to sell.The situation that arose in the Chi ware subdivision in December 1959accorded with these observations. With the approach of the festive season it wasdecided to permit the hunters to return home as soon as each camp had fulfilled itsassessed quota, based on past performance. Several camps achieved this in eightor nine, instead of the usual twenty-five, days.Table V compares the kills and the average dressed carcass weight that theyrepresented, based on Roth (1966), for the twenty-four months before and afterthe hunter strength was doubled from twelve to twenty-four men in the Chirundusubsection of the Urungwe area. According to Goodier (1956), the hunting arearemained unchanged. With the exception of waterbuck, impala and kudu, thekills increased in number by 80 per cent or more, in spite of the influx ofless-experienced hunters, indicated by the marked rise in the mean number ofshots per kill. Doubling of the number of hunters led to an increase in the totalnumber of animals shot and in the mass of meat harvested of almost 70 per cent. Adrop of 15 per cent in the mass of meat per hunter still allowed 270 kg of dressedcarcass per man per month.Night-hunting with spotlamps and shotguns was introduced within the fencedcorridor in the Sebungwe in November 1958. It led to higher kills of certainspecies which is further evidence that the regular hunters had not been effective incurbing these populations. Hunters for this exercise were selected from among thebest regular hunters. They were organized into teams which were moved aboutthe area so as to provide better coverage between camps. Hunting took place fromlast light until 23.00 hours and the results of twelve months of this hunting arecompared with those from regular hunters in Tables VI and VII.In terms of the number of animals shot per hunter and the number of shots perkill, night-hunters were more efficient, but as to be expected they shot few largeanimals or diurnal warthogs. Duiker, steenbok and grysbok contributed 88 percent of the kills compared to 61 per cent by regular hunters. The main tsetse hostsin this area were bushpig, warthog, kudu and bushbuck (Cockbill, 1964) and theymade up only 3.4 per cent of the animals shot at night, as opposed to 22.5 per centof those killed during daylight hours. This means that despite the greater efficiencyin harvesting meat (116 : 84 kg per month; from Roth, 1966) each night-huntershot an average 4.25 head of primary tsetse host animals while his regularcounterpart shot 11.5 head during the twelve-month period.Besides the small antelope mentioned above, night-hunting was alsoTable VHUNTING IN THE CHIRUNDU AREA IN THE 24 MONTHS BEFORE AND AFTERTHE HUNTING STRENGTH WAS DOUBLED24-Month periodSpeciesAveragedressedweight (kg)Oct. 1952-Sept. 1954KiiiMeat (kg)Oct. 1954-Sept. 1956KillMeat (Kg)Change inkil! (%)ZebraBushpigWarthogDuikerSableWaterbuckGrysbokImpalaBushbuckKuduElandBuffalo1903632101051055272011427329521243001132930440910019035873,9908649,6001,1303,0453,1502011,0432,00021,6609,55025,66591,7177,6435571541224552832580191241641692,2511,04410,4502,55617,3122,2405,7752,94016015,6603,82027,47417,47249,855+162+ 196+80+98+90Š7+700+42+91+27+83+94TOTALSKillMeatMain tsetse hostAverage meat per manper month (kg)1,342614155,7146,488+68+70+70-15Ammunition rounds per kill 1.96Hunters per month 123.2824+67+100G. F. T. CHILD & T. RINEY59particularly effective against reedbuck and, perhaps, impala. All adapted quicklyto this new form of hunting, however, and within a few months were difficult tolocate from a vehicle using a powerful spotlamp before the hunters went to bed.Later at night they were located in numbers.The 1958/9 kill (excluding elephant) from the Chiware subdivision of thefenced corridor of some 1030 km2 is indicated in Table VI. It represented anoff-take of some 7.1 animals per km2. There was no significant selection in favourof males or females by night-hunters, although in impala (x2 = 2.639) the highernumber of females taken approached significance at the 10 per cent level. Likewisethere was little selection in favour of either sex by regular hunters except in the case ofbaboon (178 males : 100 females) and warthog (132 males : 100 females), wheremore males were taken, and steenbok (81 males : 100 females) where a greaterproportion of females were shot.The warthog sex ratio of 132 males : 100 females compares with a ratioapproaching parity in Senegal (Bigourdan, 1948), two populations in Zimbabwe,including one sampled by the first six weeks of hunting in the Nagupande area (Child,1968b; Child et al, 1968), and a preponderance of females in another Zimbabweanpopulation sampled by Gumming (1975) over seven years. A selective huntingpressure favouring females would serve to raise the reproductive potential per unit ofsurviving population and so set the stage for a population increase, as Child (1968b)has described for a number of species.A sample of 1,552 duiker skulls collected from hunting operations in theSebungwe were sexed according to Riney and Child (1964) and aged according to thecriteria developed by Riney and Child (1962), as modified by Wilson et al. (1984).About 658 (42 per cent) of the animals were in their first year, 388 (25 per cent) wereyearlings, and 506 (33 per cent) were adults over two years old. Duiker in this areamay conceive for the first time when eight to ten months old (Child and Wilson,1964a) and calve seven months later (Wilson, pers. comm.). About 58 per cent of theyearlings were females and of these about 168 were old enough to have calved. Some55 per cent of the adults were females, giving a total of about 446 (168 + 278)females that could have calved. Duiker breed throughout the year in this area(Riney and Child, 1962) giving a maximum reproductive potential of about 160calves : 100 females per year, allowing for the six-week post-partum oestrusobserved by Wilson (pers. commS). The calf to female ratio in the sample of skullswas 148 : 100.An unexpectedly high proportion of young animals was also observed in theshot sample of other populations. There were 74 male, 64 female and 32 unsexedskulls in a sample of 170 steenbok, of which 43 per cent had immature dentition.Of the sexed adults with all permanent teeth, 37 were females, giving a ratio of195 juveniles per 100 adult females. Of the 90 grysbok skulls, 38 were male, 43female, and 9 unsexed animals with fully adult dentition, for a ratio of 380Table VITHE FIRST 12 MONTHS OF NIGHT HUNTING COMPARED WITH DAY HUNTINGS0.303 Rifies (Day Hunting) Nov. 195a-Oct. 1959SpeciesSexBaboonBushpigWarthogDuikerSteenbokGrysbokReedbuckimpalaBushbuckKuduOthersTOTALSHuntersKiltAmmunition (x10)Ammunition per animalAnimals per hunterN21143591524322355144158302101929202545671352748126612232722855571235141583.14.2D261036241337658012515717433114152926154112719257981711516223583226801266231993.24.9J214251552073591321321582901618341617331041420173734761117173213056261305791793.14.5F2082819294853491021111192305611145195279192842651015232542635171304861463.03.7M3794618143249307998119217916256511_11951433691221162472214681304411373.13.4A4816641422363420548211193136199142311241115538101323152272254521284291403.33.4M30194910182820143481851665510127193368513ŠŠŠ7512141821693511263341163.52.7J24174161622211536901142046612161228549224681410414101912033941243811373.63.1J2917466814191029911202119152422133563913481119369101982124101243901393.63.1A4122631192020204015215630821254624143866125510818266410132982895871225681833.24.7S2726531071729204914517231723285126255194138210118198311133052986031215831803.14.8O394281813213822601932144072230522935641317306410718253811113654077721237482102.86.1TOTAL3632045671501693194533427951,4761,7003,17615318934222018740776591357980159689115984871711873,2183,1996.4176,0761,924Mean125.63.174.8403m05mOOz33Or~ICzHZoShotguns (Night Hunting) Nov. 1958-Oct. 1959SpeciesSexNŠŠ_Š32511_145_Š_Š__561161121.61.8DŠŠ112121830145145314134_Š_Š_19315055071.410.0J_ŠŠ33101828Š21312322_Š___13263953961.47.8F111_1_Š131730246Š123314_Š.Š21244554571.69.0MŠ_Š142236134__123Š_Š_Š16274354361.48.6A__ŠŠ__Š132033437ŠŠ_22__Š__11Š19244354371.78.6M_22_ŠŠ151227516112_.ŠŠ_ŠŠ_21163753761.57.4J22_11242852235561111Š___13341741274152.06.2J4371112342418321313411.Š11__Š515210312103191.98.6A22134_-Š495310231114661211211___Š3627713912139231.611.6s12313441565581239162512820___Š11__Š928918112181331.815.1o4371123Š37084154111425991836955Š112_102123225i2225401.818.8TOTAL14102457129817330373703416010137387513203341216nV221234454536990990170Mean8.01.7212.38f iboonfs'ishpigWarthogDuikerSteenbokGrysbokReedbuckImpalaBushbuckKuduOthersTOTALSHuntersKillAmmunition (x10)Ammunition per animalAnimals per hunterO>Table VIISUMMARY OF THE SPECIES AND SEX RATIOS OF KILLS DURING THE 12 MONTHS OF NIGHT HUNTINGCOMPARED WITH DAY HUNTINGSpeciesBaboonBushpigWarthogDuikerSteenbokGrysbokReedbuckImpaiaBushbuckKuduSubtotalOthersTOTAL0.303 RiflesKill5673197953,1763424071351591591716,2301876,417Percentageoftotal kill8.85.012.449.55.36.32.12.52.52.797.12.9ŠSex ratio(males : 100females)178891338780116129997597ŠŠ100.6Animalsperhunter4.52.56.325.22.73.21.11.31.31.4ŠŠShotgunsKill241217703101753316239864990Percentageoftotal kill2.41.21.771.010.27.63.31.60.20.399.50.4ŠSex ratio(males : 100females)140711128969976533ŠŠŠŠ84.7Animalsperhunter3.01.52.087.912.69.44.12.00.30.4Š_tomCOmoozu0XczHzoG. F. T. CHILD &T, R1NEY63juveniles per 100 adult females. The 59 reedbuck skulls consisted of 27 males, 26females, and 6 unsexed animals. Of these, 32 had sub-adult dentition and 12 wereadult females in which tooth replacement was complete, giving a ratio of between213 and 267 juveniles per 100 adult females, depending upon the number ofunsexed animals that may have been adult females.The 71 impala specimens were aged according to Child (1964) and included18 calves under twelve months old, 8 yearlings, and 26 females old enough tohave calved, judging from the month in which they were shot (Child, 1968b).This gives ratios of 69 calves and 131 calves and yearlings per 100 breedingfemales; 6 calves were small but 10 were over six months old so that there was ahigh proportion of well-grown sub-adults in the sample.While such data have limitations, they were collected by some 140 huntersover a fifteen-month period of concentrated hunting in an area that had beenhunted for twenty-six years. They suggest either a bias towards shooting younganimals, including many that would have appreared adult-sized under fieldconditions, or populations with potentially high rates of recruitment into the adultclass. The latter would be consistent with phase one of the eruptive cycle forungulates described by Riney (1963a) and conforms with the high duiker andsteenbok population densities observed by Child and Wilson (1964a) in theChiware area twenty-eight months after the hunting ceased.DISCUSSIONTsetse control hunting was designed to eliminate the wild vertebrates on whichthe fly depends naturally for food. Returns of the animals shot in the hunting areasis of historical and zoo-geographical interest to the management of wildlife in alarge portion of Zimbabwe. Here discussion is limited to the effectiveness of thehunting operations in achieving their aim of game elimination. The possibilityexists that the hunting, as practised during the period under review, was notachieving its stated objectives and that it was counterproductive to them.There is good circumstantial evidence that declining densities of fly in severalareas coincided with hunting operations and yet wildebeest in the Sebungwe andblack rhinoceros generally were the only species apparently eliminated by thehunting. Sable, roan and eland were shot in relatively higher numbers throughoutthe country before 1933 and contributed a declining proportion of the kill fromthen until 1957/8 (Fig. 46). This was also true of zebra, waterbuck and reedbuckafter 1933. With the exception of reedbuck these are all large, rather gregariousspecies, which are conspicuous and relatively easy to hunt, providing thesuccessful hunter with a good return in meat for his efforts. Apart from eland theyare all essentially grazing species.The lack of a persistent trend in the proportions of duiker, impala, bushbuck,kudu and buffalo in Fig. 46 probably relates to changes in hunting policy,64TSETSE CONTROL HUNTINGFigure 46: GROSS TRENDS IN SPECIES KILLS5-3-1-10-5-5-40-10-1-10-10-TrendDown ? UpROANSABLE *-WATER BUCK-REEDBUCKKUDUBUSHBUCKIXIIKERWAffTHOGBUSHRG^ABOQN1KB- 1S39- 1M4- 1M9- 1953/4-193« 1943 \m 1B52/3 1957/8/Vofe- The total IdH of each aptdes it abown as a percentage of the total kill of all species in allhunting areas over successive five- or six-year intervals.G. F, T, CHILD &T, R1NEY65particularly as to whether or not small antelope were shot. Tie high numberstaken of such species as duiker would tend to buffer obvious trends. Nevertheless,all these species were shot in increasing numbers until the Thomas Report(Federation of Rhodesia and Nyasaland, 1955) led to the constriction of thehunting areas. This evidence, supported by that from the Urangwe and Sebuegwewith some variations, cannot be reconciled with declining populations. Ifanything, these populations were increasing, judging from the trends in annualkills and the high proportion of young well-grown juveniles in the skull collectionfrom the Sebungwe in 1958 to 1960.Buffalo and kudu are both large species5 but like bushbuck and duiker theybecome very secretive when persecuted. Duiker, bushbuck and kudu are almostexclusively browsers, while both buffalo and impala will graze when grass isavailable. On the other hand, both tend to do well when the range has beensuppressed and woody vegetation has prospered at the expense of the perennialgrass cover.Baboon, elephant, bushpig and warthog were killed in consistently higherproportions on a country-wide basis from 1933 to 1955/6 and this trend wassufficiently strong so as to mask the effects, to the total kill, of the changes inhunting policy. With the possible exception of elephant, where annual sampleswere small, this pattern was also manifest for the Sebungwe and the Urungwe,indicating that these species increased in spite of being hunted. Baboon andbushpig are somewhat specialized mixed feeders, while warthog are grazersdepending heavily on rooting. All three are highly adaptable and usually becomevery difficult to hunt after being subjected to sustained disturbances.Table VIII summarizes these trends in the dominant animals shot on tsetsecontrol and relates them to the species gross feeding and movement patterns, Theresults are presented as percentages of the overall kill by time intervals in TableIX. Possible relationships between the total kill and gross movement patterns areinconclusive, but there was a decline in the proportion of grazers shot in lateryears, which is marked if warthog is excluded. This is compensated for mainlythrough an increase in mixed feeders, as the relative proportion of browsersremained fairly stable, especially after 1944.While hunting may have been one factor affecting the species composition ofthe fauna, the possibility exists that changes in the relative proportions of kills wasa reflection of habitat changes, especially as Riney (1963b) described suchchanges in the Sebungwe. He attributed the changes to the annual early burning ofthe range "associated with tsetse control hunting. It is well known that regularburning early in the dry season in areas of low rainfall and low soil fertility willfavour bush encroachment at the expense of the perennial grass cover. This isprecisely what Riney found in the unsettled Sebungwe, where the suppression ofTable MilTHE NUMERICAL TRENDS IN ANIMAL KILLS COMPARED WITH THE GROSS FEEDING AND GROSS MOVEMENTPATTERNS OF THE VARIOUS SPECIES HUNTED IN HIGH NUMBERS BETWEEN 1933 AND 1957/8SpeciesTrend in percentage kill Species classesUpFeedingMovementDown Grazer BrowserMixedfeederLocal Semi-locai MobileBaboonElephantZebraBushpigWarthogDuikerWaterbuckReedbuckimpalaRoanSableBushbuckKuduElandBuffaloXXXXXXXXXXXXXXXma>mOOz33Or~ICZHZ0Population TrendUpDownTable IXTHE PROPORTION OF ANIMALS SHOT IN SUCCESSIVE TIME INTERVALS,BY FEEDING AND MOVEMENT PATTERNSAnimal ClassFeedingGrazers(excluding warthog)Mixed feeders(excluding baboon)BrowsersMovementsLocalizedSemi-localized or mobilePeriod1933-1938(%)28.320.712.85.959.084.915.11939-1943(%)25.415.418.49.055.287.612.41944-1948(%)25.812.319.511.153.889.710.31949-1952/3(%)24.810.222.514.753.591.28.81953/4-1957/8<%>21.87.427.213.953.792.98.1Q71_HOIr~O68TSETSE CONTROL HUNTINGthe grasses in some areas was so marked that it had resulted in accelerated soilerosion, leading to gully formation.It may also be significant that, although warthog is a grazing species, Mitchell(1963) has described how such burning practices and the resultant changes in thespecies composition of the grasses can favour the species, at least temporarily, inthe same way that Petrides (l960,pers. comnt) demonstrated with overgrazing inUganda. Child (1968a) also indicated that roan and sable are among the firstspecies to decline, even in the absence of any significant hunting, where similarbut more arid habitats to those of the Sebungwe are subjected to the type ofburning regime practised to facilitate tsetse hunting. A reduction in roan numbers,in association with such changes in the physiognomic structure of the vegetation,conforms with Joubert's (1976) analysis of the characteristics of essential roanhabitats.While tsetse may feed on a wide range of vertebrates, analysis of its blood-meals generally shows a preference for the suids and such species as bushbuck andkudu. In the Nagupande area warthog and bushpig contributed 64.8 per cent ofthe blood-meals analysed, with kudu and bushbuck at 11.5 per cent, the nextmost favoured hosts (Cockbill, 1964).It is perhaps ironic that these four species were shot in greater numbers aftertwenty-four years of tsetse control hunting, with warthog and bushpig, at least,apparently increasing in the hunting areas. While Cockbill shows a strongnegative correlation between tsetse numbers and the cumulative warthog killduring the first few months of intensive hunting in the Nagupande area from 1962to 1964, this could hardly have applied in areas where there was protractedhunting prior to 1957/8. Here any relationship between the animals killed and adecline in ly populations must have related to factors other than the mere killingof animals, which is in line with what Child et al. (1970) found in Botswana. Theysuggest that hunting may have influenced the behaviour of the host species,making them less accessible to the tsetse, for, as noted, species like duiker andother small antelope soon adjusted to changes in the hunting pattern, such as theintroduction of night shooting.If this hypothesis holds, then the net result of the hunting would have been toincrease the potential availability of tsetse food animals when hunting ceased andthe animals resumed their normal behaviour patterns. Even had a host populationbeen reduced, a rise in numbers above a critical threshold could have led to anincrease in the tsetse population (Bursell, 1961). This would explain the reporteddeterioration in the tsetse situation after the reduction in the size of the huntingareas, recommended in the Thomas Report (Federation of Rhodesia andNyasaland, 1955), when hunting was suspended in 1960.It also underlines the importance of having clearly-defined objectives in anycontrol measure such as hunting, and of systematically evaluating whether or notG. F. T. CHILD & T. RINEY69these are being achieved. In the present case, protracted costly hunting operations,over twenty-six or more years, may actually have compounded the problem theysought to alleviate. For game elimination, or selective game elimination, to beeffective it is essential that it is properly monitored against the aims to be achieved(Bursell, 1961; Vale and Gumming, 1976).AcknowledgementsThe authors are most grateful to staff at the Branch of Tsetse and Trypanosomiasisand Reclamation Š from the Officer in Charge to personnel in the field Š formaking their records available, for assisting with this and its ancillary studies and,last but not least, for their cheerful co-operation and kind hospitality. It is a creditto the Branch that studies critical of its policy received so much encouragement.The authors are especially grateful to Mr D. F. Lovemore, for many years thesenior officer concerned with field operations in the Branch, for making availablehis valuable unpublished notes. He and Dr D. H. M. Gumming kindly offeredcritically constructive advice on the draft of this article. Mr Cloud Masaraureprepared the figures.ReferencesBIGOURDAN, J. 1948 'Le phacochere et les suides dans 1'ouest africain*,Bulletin de Institut Frangais d'Afrique Noire [Dakar], X, 285-360.BURSELL, E. 1961 'Tsetse Control in Relation to Wildlife Conservation*(mimeo. [no details]).CAMPBELL, A. and CHILD, G. 1971 'The impact of man on the environmentof Botswana*, Botswana Notes and Records, III, 91-110.CHILD, G. 1964 'Growth and ageing criteria of impala, Aepycerosmelampus', Occasional Papers of the National Museums of Southern Rhodesia,XXVIIB, 128-35.1968a An Ecological Survey of North-eastern Botswana (Rome,FAO, T.A. Report 2563).1968b Behaviour of Large Mammals during the Formation ofLake Kariba (Salisbury, Trustees of the National Museums of Rhodesia,Kariba Studies 3).CHILD, G., ROBBEL, H. and HEPBURN, C. P. 1972 'Observations on thebiology of tsessebe, Damaliscus lunatus lunatus, in northern Botswana',Mammalia, XXXIV, 342-88.CHILD, G., ROTH, H. H. and KERR, M. 1968 'Reproduction and recruitmentpatterns in warthog (Phacochoerus aethiopicus) populations', Mammalia,XXXII, 6-29.CHILD, G. and SAVORY, C. R. 1964 'The distribution of large mammalspecies in Southern Rhodesia', Arnoldia (Rhodesia), I, xiv.CHILD, G., SO WLS, L. and MITCHELL, G. L. 1965 'Variations in the dentition,ageing criteria and growth patterns in warthog*, Arnoldia (Rhodesia), I,xxxviii.70TSETSE CONTROL HUNTINGCHILD, G. and WILSON, V. J. 1964a 'Delayed effects of tsetse control huntingon a duiker population5, Journal of Wildlife Management, XXVIII, 866-8.1964b 'Observations on ecology and be-haviour of roan and sable in three tsetse control areas', Arnoldia (Rhodesia),I, xvi.CHILD, G., VON RICHTER, W. and SMITH, P. 1970 'Tsetse control hunting asa measure of large mammal population trends in the Okavango Delta,Botswana', Mammalia, XXXIV, 34-75.COCKBILL, G. F. 1964 4A Second Review of the Tsetse and TrypanosomiasisPosition in Southern Rhodesia' (Salisbury Ministry of Agricuture, Dep. ofVeterinary Services, Branch of Tsetse and Trypanosomiasis Control, mimeo.).GUMMING, D. H. M. 1975 'A Field Study of the Ecology and Behaviour ofWarthog (Salisbury, Trustees of the National Museums and Monuments ofRhodesia, Museum Memoir 7).1981 'The management of elephant and other largemammals in Zimbabwe*, in P. A. Jewell and S. Holt (eds.), Problems inManagement of Locally Abundant WiM Mammals (New York, AcademicPress), 91-118.DOWSETT, R. 1966 'Behaviour and population structure of the hartebeest inKafue National Park', Puku [Occasional Papers of the Department of Gameand Fisheries, Zambia], IV, 147-54.FEDERATION OF RHODESIA AND NYASALAND. 1955 Report of the Com-mission of Inquiry on Human and Animal Trypanosomiasis in SouthernRhodesia [Chairman: E. Thomas] (Sessional Papers, C.Fed. 24).GOODIER, R. 1956 * A Survey of the Tsetse Game Position in the ChiranduArea of the Zambezi Valley* (Salisbury, Ministry of Agriculture, Dep. ofVeterinary Services, Branch of Tsetse and Trypanosomiasis Control, mimeo.).HEMANS, H. N. 1935 The Log of a Native Commissioner (Bulawayo, Booksof Rhodesia, 1971 reprint).JACK,R.W. 1914 Tsetse iy and big game in Southern Rhodesia5, Bulletin ofEntomological Research, V, 97-110.JOUBERT, S. C. J. 1976 'The Population Ecology of the Roan Antelope,Hippotragus equinus equinus (Desmarest, 1804), in the Kruger National Park*(Pretoria, Univ. of Pretoria, D.Sc. thesis).LEWIS, A. R. and WILSON, V. J. 1977 *An evaluation of a fence in the controlof wild ungulates under extensive conditions in Africa9, British VeterinaryJournal, CXXXIII, 379-87.MITCHELL, B. L. 1963 SA new aspect to the effect of bush fires in connectionwith tsetse, Glossina morsitanf, Kirkia, III, 26-8.RHODESIA, i 973 Ministry of Lands. Report of the National Parks AdvisoryBoard and Director of National Parks and WiM Life Management for 1972(Salisbury, Govt Printer).RINEY, T. 1963a 'The impact of man on the tropical environment', in Pro-ceedings of the 9th Technical Meeting of IUCN, Nairobi, 17-20.1963b *A rapid field technique and its application in describingconservation status and trends in semi-arid pastoral areas9, African Soils, VIII,159-258.G. F. T. CHILD &T. R1NEY71RlNEY T. and CHILD, G. 1962 'Breeding season and the ageing criteria for thecommon duiker (Sylvicapra grimmia)\ in Proceedings of the First FederalScience Congress, Salisbury, May 18-22, I960, 291-9.1964 'Limitations of horn height as an index toageing the common duiker (Sylvicapra grimmidf, Arnoldia (Rhodesia), I, i.ROTH, H. H. 1966 'Game utilization in Rhodesia', Mammalia, XXX,397-423.ROTH, H. H. and CHILD, G. 1968 'Distribution and population structure ofblack rhino in Kariba Basin', Sonderdruck aus Z.f. Saugetierkunde, XXXIII,214-26.SMITHERS, R. H. N. 1983 The Mammals of the Southern African Subregion(Pretoria, Univ. of Pretoria).SUMMERS, R. 1967 'Archaeological distribution and a tentative history oftsetse infestation in Rhodesia and the Northern Transvaal*, Arnoldia(Rhodesia), III, xiii.VALE, G. A. and GUMMING, D. H. M. 1976 'The effects of selectiveelimination of hosts on a population of tsetse flies (Glossina tnorstiansmorsitans Westwood (Diptera, Glossinidae))5, Bulletin of EntomologicalResearch, LXVI, 713-29.WILSON, V. J. 1975 Mamrnab of the Wankie National Park (Salisbury,Trustees of the National Museums and Monuments of Rhodesia, MuseumMemoir 4).1965 'Observations on the greater kudu, Tragelaphus strepsi-ceros Pallas, from a tsetse control hunting scheme in Northern Rhodesia*,East African Wildlife Journal, III, 27Š37.WILSON, V. J. and CHILD, G. 1965 'Notes on klipspringer from tsetse lycontrol areas in eastern Zambia', AmoMia (Rhodesia), I, xiii.WILSON, V. J., SCHMIDT, J. L. and HANKS, J. 1984 'Age determination andbody of growth of the common duiker, Sylvicapra grimmia (Mammalia)5,Journal of Zoology, CCII, 283-97.