Aspects of the Bole of Man in Erosion InMhodesiaM. A. StockingDepartment of Geography, University of Rhodesia, Salisbury.Of all the factors in erosion, man has beenthe least studied. Soil erosion as distinct fromgeological erosion is essentially a man-inducedprocess. The anomolous situation whereby thehuman influences have been largely neglectedcan partially be explained in terms of the veryunreliability of man. It may be a costly andlaborious task to set up and maintain erosionplots, but once in operation virtual control isexercised over the natural environment. Rain-fall can be accurately measured; soil loss inrunoff assessed. This is not so with man: for anyrigorous measurement of his activities is fraughtwith difficulties. He is subject to the manyforces of his environment, under which twoapparently similar human beings might react intotally different fashion. Ekblaw " (1936, 2)writing on the relationship between soil scienceand geography, touches on the problem:Man is the most difficult of all things torelate scientifically to his environment be-cause he possesses as wide a degree ofadaptability as plants and animals, greatermobility as a genus than either plants oranimals and, in addition, has his peculiarpower of volition, of choosing for himselfhis own course.The problem remains as to how far it ispossible to identify a measure of order withinman's activities. If a sound quantitative studyis required, not enough order is available, sincethe sweeping generalizations necessary in a widegeographical study would tend to temper thevery important role of the individual. A cursorystudy of air photographs over adjacentEuropean farms in Rhodesia, for instance,clearly shows the very different erosional situ-ation over similar tracts of land.It appears, therefore, that any discussion onman and erosion in Rhodesia must necessarilyneglect precise cause-effect relationships. Thisproblem has been circumvented in the UnitedStates where the 'human factor' has beenmeasured in terms of known conservationmeasures and the degree to which these havebeen implemented. The approach adopted byAmerican research workers was conditioned bythe framework of the Universal Soil LossEquation (Smith and Wischmeier, 1962) whereit was hoped to reduce the causes of soilerosion down to a number of readily definableand measurable factors. This may be feasiblein the United States and indeed has been usedwith success. Thus, the 'erosion control practicefactor' is an attempt at quantifying the benefitsof various conservation practices. For example,contour planting (Moldenhauer and Wisch-meier, I960), strip cropping (Hill et al., 1944),ridging or terracing along the contour (Smithand Wischmeier, 1957) have all been assessedin terms of their value in reducing erosion withreference to standardised conditions on theerosion experiment stations in the United States.In addition, a 'cropping management factor'looks at aspects of the cover given to the soilby various crops (Stocking, 1972a). Itsdefinition encompasses measures such as thenumber of years of planting continuous rowcrops and the quantity of residues left on theground after harvesting (Wischmeier, 1960).This whole, approach by the Americanconservationists is geared to the large cornbeltfarms that are highly mechanised and techno-logically based, where, at the change of aplough, minimum tillage (i.e., least disturbanceto the soil) can be practised. The human'unreliability' factor has largely been eliminated,so much so that agricultural practices are nolonger the whim of the individual. The useof such American indices in Rhodesia, however,is dependent upon knowledge of the necessarylocal control measures and the means to imple-ment them being available. While the knowledgeand means is. to a large extent, availableamongst European farmers in Rhodesia, it isnot so in the African areas. The African farmercannot yet be expected to understand fully theintricacies of mould board ploughing, let alonesuitable crop rotations, detailed soil analysesand optimum fertilizer requirements, all ofwhich are necessary pre-requisites for the suc-cessful operation of the American indices.Therefore, one cannot effectively base an assess-ment of the human factor on the degree towhich positive conservation measures are ap-plied. An index must be sought which measuresthe degree of damage done to soils by the ac-cumulated effects of past and present land usepractices. At this stage it is possible to showonly the foundations for such an index throughcontrasts illustrating the significant differencesbetween land tenures and land use systems, theimportance of factors such as standards of edu-cation or pride of ownership and the availabilityof means and resources for conservation.THE NATIONAL SITUATIONIt is at the national level that the broadestgeneralizations must be applied. There is littledata to quote and only the major patterns ofman's activities that influence erosion can bebrought out. The mafor emphasis is on theTribal Trust Lands. It is here that man,strongly backed by natural forces, has createdfor himself the greatest actual or potentialerosional situation.The pattern of African population inRhodesia has been documented by Kay (1972).The African population density map derivedfrom Kay's study (Kay and Wheeler, 1971)clearly shows a strong clustering of populationin the rural areas to the north, east and southof Salisbury. Equally high densities Š up to 60persons per square kilometre Š are shown tothe north of Mtoko and generally in a wide arccoincident with the south-eastern Middleveld,stretching in the north from the edges of theZambezi Escarpment, through Mtoko, Rusapeand the densely-settled TTLs of the Sabi-Gutu-Buhera-Bikita-Chibi area. In contrast, the areasto the north-west are sparsely-populated. Thelarge TTLs of Gokwe and Nkai have localdensities rising to 30 p.p. sq. km., but in generalthe density is of the order of 0-10 p.p. sq. km.Moreover, it should be noted that in this rela-tively undeveloped area, endowed with poten-tially good natural resources, the increase inpopulation is over ten per cent per annum.Population density, while not directly affectingerosion, must be taken in context as a majorcontributing factor in erosion. A high densityof population on land that at the present stageof technology is incapable of supporting sucha population unless large and ever-increasingareas are brought under the plough, and whichis settled in an area where the natural incidenceof erosion is potentially very great, must beconducive to erosion. This situation is perhapsbest exemplified in parts of the south-easternMiddleveld where slopes are steep (Stocking,1972bl, vegetation is poor due to the unreliablerainfall and the pressure of population on theland is accentuated by the paucity of areassuitable for cultivation and settlement.An associated problem and one that has im-portant consequences on erosion is the conceptof land carrying capacity. In a classic studyof man-land relationships in Africa, Allanf 1949, p. 1) states that:Any area of land will support in perpetuityonly a limited number of people. An ab-solute limit is imposed by soil and climaticfactors in so far as these are beyond humancontrol, and a practical limit is set by theway in which the land is used.Allan was directly concerned with criticalpopulations for particular land-use systems. Anexcess of population above a critical limit, with-out a compensating change in the system of landusage must promote what he calls 'land degrad-*Ł >-Ik! k- Łtion' (Allan 1949, p. 1; and 1965, p, 291). Theestimation of land carrying capacity in Rhodesiapresents almost insuperable practical problems.It can be stated from the obvious fact thatserious erosion has occurred, that the carryingcapacity of the tribal areas is low under presentsystems of land usage, and that the critical popu-lation density has been exceeded in most areas.Populations of cattle, goats and donkeys arealso contributory agents to the erosion process.Holleman (1951, 358) writing of the subsistenceeconomy of the Shona tribes in the Sabi area,notes that although cultivation is by far themost important means of subsistence 'a sizeablenumber of cattle and smaller livestock is found,mostly of a poor to mediocre quality'. Thoughcattle bring in more money to the tribal areasthan sales of crops, they play little part in thenational economy of Rhodesia, being mostlymaintained for marriage and draught purposes.In this latter capacity, cattle are vital to thecultivation of fairly extensive areas, especiallyin the absence of many adult men. Here atwofold problem arises for the cultivator: thatof keeping enough cattle for ploughing andenough land under crops to feed the family.Also, individual wealth is often gauged bynumbers of livestock and there is a consequenttendency to keep the maximum number possible,a figure limited by periodic drought, lack ofgrazing and attempts at administrative control.The Ndebele have a stronger pastoral traditionand chose to settle in the drier southern andwestern parts of the country. These peopleremain largely pastoral, although any move-ment towards commercial cultivation would bethwarted by unreliable rainfall (Kuper et al.,1954; Johnson, 1964). In 1962 there were some2 090 000 head of cattle, 650 000 goats, sheepand pigs and 65 000 donkeys in the tribal areas*(Kay,'1970, p. 80). Assuming that four of thesmall livestock are equivalent to one cattle unit,there are approximately 0,13 cattle units perhectare of African land. While this does notappear to be an unduly high density, it mustbe remembered that many areas are totallyunsuited to any form of grazing and in others,notably the Middleveld again, carrying capacityof the extensive grazings is one cow to every4-12 ha., dependent on rainfall (Department ofNative Agriculture, 1962. p. 98). It should alsobe noted that the overall density of 0,13 does*These are official figures and probably understate-ments.not take into account arable land or land un-suitable for agricultural practices. Therefore,it can be deduced that high densities of livestockare apparent in some areas. Kay (1970) showshow the situation in Rhodesia has deterioratedbetween 1961 and 1965. In 1961 in the com-munally occupied African areas as a whole itwas estimated that the total stock as a percent-age of carrying capacity was 97 per cent.Certain provinces, in particular Manicalandand Matabeleland South were overstocked. By1965 the situation had worsened considerablyand the African areas as a whole were seen tobe overstocked (114 per cent). Furthermorethe problem is bound to be more serious locallyespecially in those TTLs with high humanpopulation densities. Sister Mary Aquina (1964)quotes a good local example from the southerntip of Chilimanzi TTL near Gwelo. Due toreallocation of lands in 1965, a large numberof tribesmen and their cattle were forced tomove into this area. The cattle population wassuddenly swollen to one beast per 2 ha on landthat requires at least 6 ha per unit. In contrastto Northern Chilimanzi, the southern portionis very seriously eroded. Naturally, these localareas are at the moment the foci for erosion.In the African areas the choice for such a focusis often merely an accident of history. Tn theEuropean areas, it is to the individual farmerone has to turn. Before considering the questionof erosion at this detailed scale it is worthwhile to look into some of the regional orintra-regional patterns that occur.THE REGIONAL SITUATIONThe term 'region' is used very loosely in thiscontext as adjacent areas that have roughlysimilar natural environments. The contrast thatcan be brought out is that between differentsystems of land use which, in broad terms, isreflected in the map of land tenure. In orderto compare and contrast different areas, anaerial photographic method was adopted(Stocking, 1971). Several areas over the count-ry were chosen at random from which aparameter of intensity of erosion, gully lengths,was measured directly from the air photographs.Three of these areas fall very near each otherin the Middleveld; Bikita Tribal Trust Land,Nyahunda African Purchase Area and SabiValley Intensive Conservation Area. Mean an-nual rainfall in these three areas is in the range500-700 mm. The relief, however, is not en-Bikita Tribal Trust Land100020003000 40005000Nyahunda African Purchase AreaX2000 3000 4000 5000Sabi Valley European AreaArableGrazing5000Gully lengths m. kmFig. 1.ŠComparison of the frequency of gullies from three types of land tenure.44tirely similar and this should be borne in mindin the discussion below. Bikita is more broken,having a mean average slope of 7,0 degrees incomparison to 4,8 for Nyahunda and 3,2 forSabi Valley, though the mean for Bikita is in-creased by a few very high figures (19,4 degreesin one square kilometre, for instance).Fig. 1 is computed from a twenty per centsample of each area and illustrates the strongcontrasts between the three land tenures. Bikitahas a wide spread of erosional categories whichcan be partially attributed to a greater rangein slope values and partially to mismanagementof the land. Erosion occurs with equal severityon grazing and arable land although the verysteep slopes on which cultivation is impossibleare wholly for grazing. The African PurchaseArea shows a slightly different pattern in thatmost of the area could be classified as moderate-ly eroded, i.e., in the range 500-1 500 metresof gullies per square kilometre. Nyahunda alsoshows a fairly wide range in erosional cate-gories. It is interesting to note that here againthe severe erosion occurs on the steep grazinglands. The European area presents a verydifferent picture. Over a quarter is slightlyeroded and no more than fifteen per cent isseverely eroded. This fifteen per cent is asso-ciated with African squatter and/or farmlabourer settlements on the periphery of theI.G.A. Arable land wholly occupies the highesterosional category of 2 500-3 000 m kms"2 in anarea that is almost totally European ranching.Keech (1968) quotes figures from which asimilar comparison can be made. The area con-cerned is the Tokwe-Selukwe district, east ofGwelo, on the edge of the Highveld at about1 200 metres (4 000 ft.). Table I shows twoTribal Trust Lands and one European farmingarea:Table IDISTRIBUTION OF EROSION IN THE TOKWE-SELUKWE DISTRICTErosionalCategoriesmTokweICA7425100Percentage ofSelukweTTL02549251areaChilimanziTTL8294119301 -506507-11811182-18561857-3207numeration of the erosional categories is theresult of conversion from the unit of yards ofSource: After Keech (1968); the seemingly oddgullies per 1 600 acres.The district is clearly less severely erodedthan the Sabi area but the contrast between theland tenures is no less apparent.THE LOCAL SITUATIONIt is at the local scale that the activities ofman on his soil are most evident. At theregional scale the sum of the effect of manyindividual farmers was compared to the sumfrom another group of individuals. The sum-mation necessarily hides the role of the individ-ual, a role that has vital consequences in erosion.A few of the major factors will be consideredin the light of our present state of knowledgeof the complex interrelations between man anderosion.Dip Tanks and Watering Places for CattleAll cattle in tribal areas have to be dippedin communal dipping tanks once a week fromSeptember to April and once a fortnight in thefour dry months in order to control tick-bornediseases. The hazardous nature of the dippingtank must be seen in the light of the density oftanks and the numbers of cattle using the tanks.From air photo analysis the location of dippingtanks was noted for Mtoko TTL. The densityof tanks varied from one per 100 to one per600 sq. kms. In the northern part of MtokoTTL and neighbouring parts of Maramba,Uzumba and Pfungwe TTLs, a total area of650 sq. kms. six dip tanks were found. Thereis an estimated cattle population in this areaof 100 000, each dip tank serving some 17 000cattle. Moreover, the maximum trekking dis-tance to a tank in Mtoko is twenty kilometres.The cumulative effect of these cattle movementsŠ presumably along the same paths Š and thegathering of a large number of cattle at thesame points each week must cause a very greaterosion hazard.Fig. 2 illustrates the incidence of gullyingaround a cattle dip in Mtoko. The gullying isvery noticeable in the immediate vicinity of thedip which is located next to a stream, and onthe trail paths radiating from the dip. The dipis badly located, being cradled between verysteep and bare kopjes in a narrow valley. Thesite was obviously chosen as being the mostaccessible from the surrounding countryside.Both Wright (1970) in a newspaper articleand the Advisory Committee which reported onAfrican Agriculture (Southern Rhodesia, 1962,Major cattle trailsProminent kopjesGully linesNatural waterwaysFig. 2.ŠGully erosion around a cattle dip in Mtoko TTL.6*fr *p. 182), criticize the present policy of cattledipping. Wright questions the plunge tank'seffectiveness in controlling disease and bothsuggest the decentralization of tick control bymeans of spray races and hand spraying usingmodern tickicides.To a lesser extent cattle watering points bothon European and tribal land can be dangerousfor erosion. In the dry season these wateringpoints are often limited in number and wherethey are dams or stream beds the vicinity is oftensteep sided and very susceptible to erosion. Theair photographs did indeed show considerableconcentration of gullying near natural water-ways. Since at these points the water table isnearer the sxirface, the erosion situation isfurther aggravated by increased runoff due towater-logged soils.Basic and Agricultural EducationIn order to implement some of the morerudimentary forms of conservation, a certainbasic standard of education must be assumedin the populace. The fact that this assumptioncannot always be relied upon is one of the rootcauses for what the conservationist would call'lack of co-operation' and the educationist'inability to grasp simple and logical trains ofthought'. The author was invited to view thelatter part of a programme for the intro-duction of a short duration grazing system inChikowore's Kraal in Msana TTL near Bindura.Here, the District Commissioner and his agri-cultural staff had battled for six months with asmall group of tribal farmers to convince themthat they and their future generations wouldbenefit by fencing in paddocks, culling thosepoor cattle of mixed parentage, and allowing thegrassland, to recover from excessive grazing. Inthat this was a pioneer effort, the six monthswere well spent since the best education for theolder chiefs and kraal heads who are withoutschooling is bv example. Even so it would takemany years of piecemeal effort to join the piecesand make the TTLs not only economicallyviable but also adequately conserved.Rhodesia has on the whole a good recordin Africa for primary education; for by 1962the proportion of the children of school agewho have received some education had reachedbetween 85 and 88 per cent (Minister of NativeEducation). This bodes well for the future, fora population receptive to new ideas and theeducational ability to put these ideas intopractice is a prime requisite of conservation.However, at present the practice of collateralsuccession in which all the eldest surviving maledescendants of a common forefather are eligibleto succeed to the chieftainship (Holleman,1951), tends to make the leadership of a Shonatribe conservative in the extreme in a popu-lation that is conservative anyway. The main-tenance of one of the oldest members of thetribe at the head, a man who in all probabilityhas received no education, makes changes incustoms, particularly those related to cattle, verydifficult.Agricultural education in Rhodesia has pro-gressed hand-in-hand with primary education.In lower primary schools two hours a week areset aside for gardening and the practical aspectsof growing vegetables. Higher up the school thecurriculum is extended to field crops and thebeneficial effects of manure and fertilizer. Itis a common sight in the rainy season to see atribal school surrounded by high maize of abetter standard than crops on adjacent small-holdings. Upper primary schools are supposedto embrace aspects of soil conservation but atthis level a shortage of teachers with a know-ledge of agriculture leads to a concentration onthe more academic subjects to the detrimentof practical subjects.Away from the schools agricultural extensionprogrammes have had some success in the pro-motion of new ideas in existing farmers. Therequisite government legislation was introducedas late as 1951 in the Land Husbandry Actwhich for the first time* was an attempt to bringcompulsion rather than gentle persuasion intoagricultural extension programmes. One recom-mendation that has been slowly put into effectis the registration of African farmers into variouscategories.Three categories are recognised: that ofCo-operator who is a farmer who uses manureor fertilizer and carries out some form of ro-tation; a Plotholder who makes records of hiscropping programmes and is under tuition byan agricultural demonstrator in order to achievethe higher status of Master Farmer; a MasterFarmer, a Plotholder who has reached certain*The Natural Resources Act of 1941 had set outcertain provisions to promote good husbandry andallocation of rights and security of tenure for Afri-cans but no adequate machinery then existed forthe implementation of the Act other than for de-stocking.Table IIYear19481961 91 381Source: After Johnson (1964).CO-OPERATORS, PLOTHOLDERS AND MASTER FARMERSCo-operators Plotholders Master Farmers Total Cultivators14 293 2 017 764 213 7605 41010 454379 860minimum standards of crop and animal hus-bandry laid down by the Department of Agri-culture (Johnson, 1964). Progression to thelatter status is encouraged by opportunities ofland ownership and independence from com-munal farming on a Purchase Area plot. TableII gives an idea of the increase in numbers ofall three categories of African farmer:However, well over two-thirds of all thecultivators are unwilling or unable to adoptthe necessary practices.Conservation Facilities and PracticeThe Land Husbandry Act not only providedfor land and farmer registration but also soughtto promote suitable conservation systems. John-son (1964) shows how up to 1956 the work wasslow but after that date the programme wasaccelerated. He estimates that by 1961 nearlytwo-thirds of the TTLs had a proper conser-vation plan. The usual method of conservationis graded ridges and grassed waterways.On the communal tribal lands, the regularsuccession of ridges helped to demarcate eachcultivator's plot of the 'open field'. This was infact one of the major objectives of the LandHusbandry Act which attempted to allocate in-dividual rights and security of tenure in arableand grazing areas.While in theory the Land Husbandry Actshould work, in practice it can be very difficultto compel African farmers to conserve the soil.This ties in with the lack of education discussedabove. The Lands Inspectorate who are re-sponsible for enforcing the law are poorly staffedand, short of resorting to the courts, they findthat contour ridges remain partially dug andthe grazing lands still overstocked. In MtokoTTL gullies on supposedly conserved land wereas numerous as those on the more remote andunconserved land. A bad contour ridge channelswater to one point and the resulting erosion inan intense storm is worse than if no ridges hadbeen constructed.Contour ridges are invariably dug by handon tribal farmland. One reason for the oftenpoor construction is that this task is given overto the women as an extra burden to routineplanting, hoeing, harvesting and, last but notleast, rearing children. Some of the PurchaseArea farmers now employ a locally made ox-drawn contour ridger that completes the workin a fraction of the time. However, these farmsare the exception and pride of ownership hereis apparent in general farm practice. Contourridges on communal land have to suffer tramp-ling by cattle and people, especially where aridge crosses an unofficial footpath.European farmers use the same basic prin-ciples of mechanical conservation. In addition,profit motives encourage the use of fertilizersand higher plant densities which in turn dis-courage the erosion of soil. Facilities for con-servation practice are very much better for theEuropean. The Department of Conservationand Extension provides technical knowledge,contour-pegging teams and general advice onfarm planning. The European farmer naturallyhas greater capital resources to provide forsuitable ridgers, fences and the construction ofstorm waterways and roads. The European isnot so constricted by time-honoured customsparticularly with regard to possession of cattle.There is a ready appreciation of optimum stock-ing rates and a willingness to look further intothe future than the start of the next rains. Inshort, the European is more fertile ground forthe conservationist. Air photograph analysishas shown that by contrast the land is in farbetter condition and it is only the rare indi-vidual who, through short term greed oroccasionally lack of knowledge, encourages thewholesale removal of soil. The inequitableconservation facilities in Rhodesia betweenAfrican and European is one immediate openingfor the betterment of the tribal farmer. Resour-ces are limited but they could well be channelledin the most needed direction with a little re-orientation and a change of emphasis to basicconservation education rather than the morespecialised knowledge imparted at present tothe European farmer.CONCLUSIONSevere erosion in Rhodesia is largely afunction of man, either directly through hisagricultural practices or indirectly throughcattle, goat and human populations. Naturalor geologic erosion can and does occur throughthe normal processes of landform evolution.While the forms of this type of erosion are simi-lar to erosion induced by man since the con-trolling natural energy factors of slope andrainfall are the same, man has succeeded incompressing the time scale so that acceleratedsoil erosion can be over one thousand times moresevere than natural erosion in unit time.Feodoroff (1965, 150) quotes rates of accele-rated erosion in the order of 560 000 kilo-grammes per hectare, which contrasts with ratesof natural erosion in an 'average' environmentof 45 kilogrammes per hectare (Young. 1969,852).Man is, therefore, the significant key to theprocesses of accelerated soil erosion but un-fortunately he is the least studied of the severalfactors. At the present time, observation andoccasional measurement where statistics andfacilities permit are the only available meansfor study of man's important role in erosion inRhodesia. Attempts at quantifying man-erosionstudies are frustrated by the very lack of orderin man's activities, although if and when theagricultural economy of the whole country ex-pands, a situation may be envisaged where pre-cise measurement is possible along Americanlines.At three scales, the national, regional andlocal, significant differences and patterns inerosion can be found in Rhodesia. Possibly themajor differences lie between the land tenuresof the two main racial groups. Differences with-in these two groups must be based at themoment on population densities in the caseof the tribal areas and overall farming typesfor European areas. Game parks, Nationaland Unreserved Lands form a special problem.In recent unpublished reports some game re-serves, notably Wankie, have suffered severeerosion around or near watering points. How-ever, in the main, these areas must be consideredas being low in erosion hazard, along with theextensive European ranching areas.REFERENCESALLAN, W. 1949 How much land does a man require? Studies in African Land Usage in Northern Rhodesia,London, O.U.P. for Rhodes-Livingstone Institute, Paper No. 15.1965 The African Husbandman. Edinburgh, Oliver and Boyd.AQUINA, SR. M. 1964 The Social Background of Agriculture in Chilimanzi Reserve. Rhodes-LivingstoneJournal, 36, 7-39.DEPARTMENT OF NATIVE AGRICULTURE OF SOUTHERN RHODESIA 1962 African AgricultureIn: RHODESIA AND NYASALAND An Agricultural Survey of Southern Rhodesia, Part II: Agro-Economic Survey. Salisbury, Government Printer.EKBLAW, W. E. 1936 Soil Science and Geography. Proceedings. Soil Science Society of America, i, 1-5.FEODOROFF, A. 1965 Mecanismes de l'Erosion par la Pluie. Revue Geographie Physique et de GeologieDynamique, T, 149-63.HILL, H. O., PEEVY, W. J., McCALL, A. G. and BELL, F. G. 1944 Investigations in Erosion Control andReclamation of Eroded Land at the Blackland Experiment Station, Temple, Texas, 1939-1941. Washing-ton, U.S. Department of Agriculture, Technical Bulletin, No. 859.HOLLEMAN, J. F. 1951 Some"'Shona' Tribes of Southern Rhodesia. In: COLSON, E. and GLUCKMAN,M., eds., Seven Tribes of British Central Africa.. Manchester University Press for Rhodes-Livingstone Insti-tute.JOHNSON, R. W. M. 1964 African Agricultural Development in Southern Rhodesia: 1945-1960. StanfordUniversity, Food Research Institute.KAY, G. 1970 Rhodesia: A Human Geography. London, University of London Press.1972 Distribution and Density of African Population in Rhodesia. University of Hull, Departmentof Geography, Miscellaneous Series No. 12.KAY, G. and WHEELER, R. G. 1971 Rhodesia, African Population Density 1:1 125 000 Map. Salisbury,Surveyor-General.KEECH, M. A. 1968 Soil Erosion Survey Techniques. Proceedings and Transactions of the Rhodesian Scien-tific Association, 53, 13-16.KUPER, H., HUGHES, A. J. B. and VAN VELSEN, J. 1954 The Shona and Ndebele of Southern Rhodesia{Ethnographic Survey of Africa, Southern Africa, Part IV). London, International African Institute.MINISTER OF NATIVE EDUCATION 1962 In: SOUTHERN RHODESIA, Legislative Assembly Debates,49, c.608, 7 March.MOLDENHAUER, W. C. and WISCHMEIER, W. H. 1960 Soil and Water Losses and Infiltration Rates onIda Silt Loam as Influenced by Cropping Systems, Tillage Practices and Rainfall Characteristics. Proc. SoilSci. Soc. Am., 24, 409-13.SMITH, D. D. and WISCHMEIER, W. H. 1957 Factors Affecting Sheet and Rill Erosion. Transactions,American Geophysical Union, 38, 889-96.1962 Rainfall Erosion. Advances in Agronomy, 14, 109-48.SOUTHERN RHODESIA 1962 The Development of the Economic Resources of Southern Rhodesia withParticular Reference to the Role of African Agriculture: Report of the Advisory Committee. Salisbury,Mardon for Government Printer, CSR 15.STOCKING, M. A. 1971 Soil Erosion Problems in Rhodesia. Journal of Soil and Water Conservation, 26,239-40.1972a Planting Pattern and Erosion on a Cotton Crop. Rhodesian Science News, 6,231-2, 236.1972b Relief Analysis and Soil Erosion in Rhodesia using Multi-variate Techniques.Zeitschrift fur Geomorphlogie (N.P.), 16, 423-43.WISCHMEIER, W. H. 1960 Cropping Š Management Factor Evaluations for a Universal Soil Loss Equation.Proc, Soil Sci. Soc. Am., 24, 322-6.WRIGHT, A. 1970 Communal Dipping is Menace to Tribal Land. Sunday Mail, 12 April.YOUNG, A. 1969 Present Rate of Land Erosion. Nature, 224, 851-2.10