First, the number of patients interviewed issmall. Forty-one patients with nutritional diseasesand 30 with kwashiorkor were included in thefirst stage of the survey. In the examination ofthe quality of food eaten by rural and urbanShona, 150 men and women were interviewed;50 from the urban area, 50 from Shona tribaltrust areas and 50 from European farming dis-tricts.Secondly, the random sampling procedure wasnot used in selecting the informants. The peopleinterviewed were patients attending Harare Hospi-tal only, and do not perfectly represent either alldiseased persons or all the urban and ruralShona. One may ask, for example, to What extentthe small number of urban patients with nutri-tional diseases and kwashiorkor who went toHarare Hospital for treatment was due to theavailability of other medical agencies in the townUniversity of Rhodesiasuch as government and municipal clinics andprivate physicians.Thirdly, the interview technique used in deter-mining the quality of food and dietary habits ofurban and rural patients has its limitations. Inter-viewees may suffer from faulty memory andinability to articulate, or may consciously avoidfacts which they consider embarrassing. What isneeded as a counterweight to the use of ques-tionnaires is participant observation over a fairlylong period of time.In spite of these defects, however, ProfessorGelfand must be congratulated on this richlydocumented study which will be of practicalvalue not only to medical practitioners but alsoto administrators, missionaries, historians, socio-logists and educationalists; indeed anyone whois interested in African society will find the workuseful.G. L. CHAVUNDUKAWATTS, H. L., ed. 1970 Focus on Cities. Durban, University for Social Research, 497 pp. no price indicated.This work is the printed account of the pro-ceedings of a conference organised by the Institutefor Social Research of the University of Natal inDurban, and held there in July 1968. It is almostimpossible to review such a work, so great is therange of subjects: the pattern of urbanization, thehistory of individual towns, urban sociology,health, community services, politics, architectureand planning. All that a reviewer can do is toindicate the overall nature of the contents.Most of the papers concentrate on South Africabut there are interesting case studies of Faikirk,Scotland, and Dallas, U.S.A., and equally interest-ing general surveys by Professors J. Spengler ofDuke University and K. Davies of the Universityof California; Rhodesia is only poorly representedwith short descriptive surveys of the history ofSalisbury and of the social services of its muni-cipal townships for Africans.The considerable amount of factual material onSouth African cities will no doubt be useful to awide range of students in history, politics, soci-ology and geography; but the overall effect israther disappointing. Much of the work is survey-orientated to an extent that would have horrifiedC. Wright Mills and is concerned with the margin-alia of urban life rather than basic questions. It islike the nineteenth century survey work on urbanpoverty which concentrates on descriptive detailrather than analyses of the system that madepeople poor. Planning details of Soweto or jobaspirations of Africans can hardly be meaning-fully discussed except in a context that includesthe turning circle of a Saracen armoured-car orEuropean job reservation.Similarly much of the work on urbanisationseems superficial in that it takes little acccount, toput it no more strongly, of the rural modes ofbehaviour which are, supposedly, being changed.Even this supposition, however, may be a perspec-tive that is sociologically of reliSavely little value.Surveys of urban Africans' drinking or sexualhabits might be more insightful if they were con-cerned not with divergence from those of distantrural relatives, but with their close approximationto European urban ones. For it is what ProfessorBlacking calls the 'tribalization' of Europeans inSouthern African that is sociologically interesting.Urbanization (if such a term is meaningful) ofthe poor is on a world-wide scale; decadence ofthe rich and powerful is less common. And theseare the questions that urban sociologists in South-ern Africa should be asking; otherwise, as socialanthropology was once the philosophy of imperial-ism, so sociology will become one of the tech-niques of its successors.R.S.R.85