^{page 293 note 2} Kuczynski, R. R., Colonial Population, London: Royal Institute of International Affairs (1937)Google Scholar . Part has been reprinted in Rhodes-Livingstone Journal, no. 2 (Dec. 1944)Google Scholar , as ‘Population Movements: The Contribution of Demography to the Study of Social Problems’, pp. 18 ff.

^{page 293 note 3} Sonnabend draws attention to the value of such samples in Sonnabend, I., ‘Demographic Samples in the Study of Backward and Primitive Populations’, South African Journal of Economics, vol. ii (Sept. 1934)Google Scholar .

^{page 293 note 4} Some of the difficulties in conducting demographic surveys among non-literate peoples are recorded by Sami W. Dajani. In particular he recounts that after the Bedouins had agreed to give census information to the Palestine Administration they changed their minds and refused to do so when the Anglo-American Committee's report on Palestine was published. Dajani, Sami W., ‘The Enumeration of the Beersheba Bedouins in May 1946’, Population Studies, vol. i, no. 3 (Dec. 1947), p. 304Google Scholar .

^{page 294 note 1} Charles, Enid and Forde, C. Daryll, ‘Notes on Some Population Data from a Southern Nigerian Village’, The Sociological Review, vol. xxx, no. (April 1938)Google Scholar .

^{page 294 note 2} , A. T. and Culwick, G. M., ‘A Study of Population in Ulanga, Tanganyika Territory’, The Sociological Review, vol. xxx, no. 4 (1938), p. 365, and vol. xxxi (1939), no. 1, p. 25Google Scholar .

^{page 294 note 3} Fortes, M., ‘A Note on the Fertility among the Tallensi of the Gold Coast’, The Sociological Review vol. xxxv, nos. 4–5 (July-Oct. 1943)Google Scholar .

^{page 294 note 4} Schapera, I., Migrant labour and Tribal Life, London: Oxford University Press (1947), pp. 20 ffGoogle Scholar .

^{page 294 note 5} This study was conducted as part of the Rhodes-Livingstone Institute's comparative study of the social organization of some Central African tribes. I am grateful to the Trustees of the Institute and to the Director, Dr. E. Colson, who have given me permission to publish this material.

^{page 295 note 1} For details of this movement when it had penetrated to Northern Rhodesia, see Richards, A. I., ‘A Modern Movement of Witchfinders’, Africa, vol. viii (1935), pp. 448–61CrossRefGoogle Scholar .

^{page 298 note 1} Between a third and a half of the marriages contracted by women had been dissolved by divorce by the time that the child-bearing period had ended.

^{page 299 note 1} ndembo = elephant, hence ivory. The terms litiŵo and ndembo here both refer to the ivory beads traditionally worn by the woman in the pregnancy initiation ceremony.

^{page 300 note 1} A disease of this sort is believed in by many Central African tribes. Among the Nyanja-speaking peoples it is known as tsempho.

^{page 301 note 1} It is more usual to take the number of women between 15 and 44, though recently the age-group between 20 and 44 has been used. I have taken the age-limits 15 to 50 because of the early age of marriage, and because it appeared that the average age of cessation of child-bearing is about 50 (see Table VIII). For a discussion of the suitability of this measure of fertility where birth registration is not practised, see T. Smith, Lynn, Population Analysis, McGraw Hill Book Co., New York (1948), p. 198CrossRefGoogle Scholar .

^{page 302 note *} From The League of Nations Statistical Yearbook 1942/44. Only those countries which showed reproduction-rates based on direct census data and in which reliable age-groupings were used.

^{page 302 note †} I have used the age-limits 15-50 in these calculations.

^{page 303 note 1} The equation for the regression line is X = 0·013+0·0037 Y, where X is the G.R.R. and Y the fertility ratio. The standard error of the estimate is 0·13. A range of two standard errors was used in this estimate.

^{page 304 note 1} The total of 3·13 children per woman was divided by 2 in order to arrive at the number of females per woman.

^{page 304 note 2} M. Fortes, op. cit., p. 108.

^{page 304 note 3} I. Schapera, op. cit., p. 23.

^{page 308 note 1} The population of Kawinga, Liwonde, and Nyambi sections of what was formerly known as Liwonde district or Upper Shire District of Nyasaland, increased from 39,154 in 1926 to 64,840 in 1945. Extracted from the crude census returns in the D.C.'s office in Zomba.

^{page 308 note 2} Actually the forms for this registration do not tabulate all the categories of information necessary for the calculation of the gross and net reproduction rates. Such omissions are mainly of the ages of the mothers and ages at death, and they have been exeluded, no doubt, because it is assumed that this information is unobtainable. It is to be hoped, however, that when the system has become established, these categories will be introduced.