1 The best known English language monograph is undoubtedly the comprehensive study by Weber, Adna, The Growth of Cities in the Nineteenth Century (New York, 1899)Google Scholar, which brought together data not only for a number of European countries but also for the U.S. and other European-settled overseas countries. He reviews the findings of a large number of contemporary studies on internal migration as related to urbanization and the growth of cities as well as urban-rural demographic differentials and is a rich bibliographical source.

2 Chojnaka, Helena, “Nuptiality Patterns in an Agrarian Society,” Population Studies, XXX (July 1976), 203–26CrossRefGoogle Scholar. Quote from pages 214-15.

3 Scott, Joan and Tilly, Louise, “Woman’s Work and the Family in Nineteenth Century Europe,” Comparative Studies in Society and History, XVII (January 1975), 36–64CrossRefGoogle Scholar. Quote from page 56.

4 Knodel, John, The Decline in Fertility in Germany, 1871-1937 (Princeton, 1974).Google Scholar

5 Knodel, John and Hochstadt, Steven, “Illegitimacy in Imperial Germany: A Study of Urban-Rural Differentials,” Discussion Papers in Western European Studies (Ann Arbor, 1976).Google Scholar

6 Knodel, John and Maynes, Mary Jo, “Urban and Rural Marriage Patterns in Imperial Germany,” Journal of Family History, I (Winter 1976), 129–68.CrossRefGoogle Scholar

7 For a more detailed discussion of the differences in definitions of urban and rural among German states, see Knodel, The Decline in Fertility in Germany, 89-93; Weber, The Growth of Cities; and Knodel and Hochstadt, “Illegitimacy in Imperial Germany.”

8 For a detailed discussion of the proportions single as a tool for the study of marriage patterns, see Hajnal, John, “Age at Marriage and Proportions Marrying,” Population Studies, VII (Nov. 1953), 111–36CrossRefGoogle Scholar. For a discussion of some of the specific problems involved in using this method to study urban and rural marriage patterns in nineteenth-century Germany, see the appendix in Knodel and Maynes, “Urban and Rural Marriage Patterns.”

9 Briefly where m_i is the proportion of women married at each five year interval from 15 through 49 and F_i is the fertility of married Hutterite women in each age interval. The Hutterite schedule is chosen because it represents the highest age specific marital fertility ever reliably recorded. For a fuller discussion of this index, see Coale, Ansley J., “The Decline of Fertility in Europe from the French Revolution to World War II,” in Behrman, S. J., Corsa, Leslie Jr., and Freedman, Ronald, eds., Fertility and Family Planning (Ann Arbor, 1969)Google Scholar and Knodel, The Decline in Fertility in Germany, 33-37.

10 Knodel, The Decline in Fertility in Germany, 68-70.

11 Although census data required for these calculations are not generally available for German states for even the first half of the nineteenth century, local population studies based on church records indicate a late age at marriage predominated as far back as such records extend (see e.g., Katherine Gaskin, “An Analysis of Age at First Marriage in Europe before 1850,” unpublished manuscript, 1975).

12 Hajnal, John, “European Marriage Patterns in Perspective,” in Glass, D. V. and Eversley, D. E. C., eds., Population in History (London, 1965).Google Scholar

13 An extensive description of the method of defining administrative areas can be found in Knodel, The Decline in Fertility in Germany, 9-14. The method was modified slightly in the study of urban-rural nuptiality (see Knodel and Maynes, “Urban and Rural Marriage Patterns”) and urban-rural illegitimacy (see Knodel and Hochstadt, “Illegitimacy in Imperial Germany”).

14 See also Knodel, The Decline in Fertility in Germany, 104 and 111-12.

15 See e.g., Medick, Hans, “The Proto-industrial Family Economy: The Structural Function of Household and Family during the Transition from Peasant Society to Industrial Capitalism,” Social History, I, No. 3 (1976).Google Scholar

16 Haines, Michael, Fertility and Occupation: Coal Mining in the Nineteenth and Early Twentieth Centuries in Europe and America, Western Societies Program Occasional Papers No. 3 (Ithaca, 1975)Google Scholar; Kollman, P., “Die soziale Zusammensetzung der Bevölkerung im Deutschen Reich,” Allgemeines Statistisches Archiv, I, 540–614Google Scholar; Friedrich Prinzing, “Heiratshäufigheit und Heiratsalter nach Stand und Beruf,” Zeitschrift für Sozialwissenschaft, VI, 546-59; and Fircks, A. von, “Die Berufs—und Erwerbsthätigkeit der eheschliessenden Personen in ihrem Einflussen auf deren Verheirathbarkeit…,” Zeitschrift des Königlich Preussischen Statistischen Bureaus, XXIX (1899), 165–208.Google Scholar

17 Knodel, The Decline in Fertility in Germany.

18 Briefly where B₁ is the number of legitimate births, B is the number of total births; M_i is the number of married women in each five year age interval from 15 through 49; W_i is the number of total women in each five year age group from 15 through 49; and F_i is the fertility of married Hutterite women in each age interval. For a fuller discussion, see Coale, “The Decline of Fertility in Europe” and Knodel, The Decline in Fertility in Germany, 33-37.

19 A number of Prussian administrative areas had to be excluded because of apparent inconsistencies in the definition of rural and urban boundaries used for the civil census and the vital registration system which was ecclesiastically based until 1874. The problem is apparent when infants enumerated in the census are compared with infants expected to be surviving according to vital registration. Only areas in which the extent of agreement between the census and the vital registration was above 90 percent and in which the estimated agreement within rural and urban sectors differed by less than 5 percent are included in the present analysis. An equivalent comparison between vital registration and census data could not be made for non-Prussian areas.

A similar problem for the calculation of urban and rural fertility rates could arise from the fact that births were registered according to the place of occurrence rather than the place of residence of mother. To the extent that women residing outside the urban areas went to clinics or hospitals inside the city limits to give birth, the number of registered births in urban areas would include births that should be allocated to rural areas for the purpose of calculating fertility rates. The result would be to inflate artificially the urban rate and deflate the rural rate. However, since only a small proportion of births actually occurred in hospitals or clinics in the nineteenth century, the amount of distortion introduced into rural and urban rates cannot be large. In 1890, for example, only about 1 percent of all births occurred in hospitals. Of course the majority of mothers giving birth in hospitals undoubtedly lived in the city or town where the hospital was located and thus the amount of distortion introduced into the rural or urban fertility rates by births being registered outside the mother’s place of residence is substantially smaller than the percent of births that occurred in hospitals or clinics. The problem may be somewhat more serious for non-marital fertility. See Weber, The Growth of Cities, 332.

20 A similar problem complicates interpretation of the urban-rural results for the city-state of Hamburg.

21 Using the same test described in Note 19, only those cities were included for which the estimated agreement between the census and vital registration was above 90 percent and differed by less than 5 percent from the estimated agreement within the rural area and/or the entire administrative area. In addition, the level of comparisons between the 1867-68 index of marital fertility was compared to the indicated level for subsequent periods to make sure it was not unrealistically out of line with these later and presumably more reliable estimates.

22 See e.g., Henry, Louis, “Some Data on Natural Fertility,” Eugenics Quarterly, VIII, 81–91Google Scholar and Knodel, John, “Family Limitation and the Fertility Transition: Evidence from Age Patterns of Fertility in Europe and Asia,” scheduled for publication in Population Studies, XXXI (July 1977).Google Scholar

23 Shorter, Edward, Knodel, John, and de Walle, Etienne van, “The Decline of Non-Marital Fertility in Europe, 1880-1940,” Population Studies, XXV (November 1971), 375–93.Google Scholar

24 More precisely where B_I is the number of illegitimate births, U_i is the number of unmarried women in each five year age group from 15 through 49, and F_i is the fertility of married Hutterite women in each age interval. The following comparison between the illegitimacy ratio and I_h for Saxony 1880-81 indicates how misleading the former can be in indicating urban-rural differentials:

25 Knodel, The Decline in Fertility in Germany, 266-69.

26 Shorter, Knodel, van de Walle, “The Decline of Non-Marital Fertility.”

27 The median values were .070 for urban and .071 for rural; the unweighted means were .068 and .072 respectively. A similar analysis for 1900 yields virtually identical average values for the urban and rural sectors.

28 Weber, The Growth of Cities, 343-67.

29 Arthur Würzburg, “Die Säuglingssterblichkeit im Deutschen Reich während der Jahre 1875 bis 1877,” Arbeiten aus dem Kaiserlichen Gesundheitsamte, II and IV (1887 and 1888).

30 Matthiessen, P. C. and McCann, James, “The Role of Mortality in the European Fertility Transition: Aggregate-level Relations,” in Preston, Samuel H., ed., The Effects of Infant and Child Mortality on Fertility (New York, 1977).Google Scholar

31 Bourgeois-Pichat has developed a biometrie technique for estimating the amount of infant mortality directly attributable to endogenous causes based on the age structure of infant mortality after the first month of life. See Pressat, R., Demographic Analysis (New York, 1972), 92–101Google Scholar. Application of this technique to the data in Würzburg’s study produced implausibly low levels of endogenous mortality for many of the urban populations and thus has not been used. For example, for Prussia, the technique yields an estimate of endogenous death rate of 17 per 1000 live births for the urban population compared to 34 per 1000 for the rural. In contrast, during the first week of life, when close to all the infant deaths are undoubtedly endogenous in nature, the recorded urban death rate is 26 per 1000 versus 27 per 1000 for rural infants. In addition, the stillbirth rate which derives from similar endogenous causes is almost identical for the urban and rural populations (4.2 percent of all births in each case).

32 Weber, The Growth of Cities, 363-64 and E. A. Wrigley, “Births and Baptism: The Use of Anglican Baptism Registers as a Source of Information about the Numbers of Births in England before the Beginning of Civil Registration,” scheduled for publication in Population Studies, 1977.

33 Prinzing, Friedrich, “Die Kindersterblichkeit in Stade und Land,” Jahrbücher fur Nationalökonomie und Statistik, III. Folge, Bd 20 (1900), 593–644Google Scholar.

34 Knodel, The Decline in Fertility, 167.

35 Analysis of legitimate and illegitimate infant mortality by age indicates that the excess of illegitimate over legitimate infant mortality is more pronounced during the first six months of life than during the second six. In Prussia, for example, illegitimate mortality during the first half year of life for the period 1875-77 was 96 percent higher than legitimate mortality. In Contrast, mortality during the second half year–calculated as (deaths during the 6-11 months over [live births – deaths in first 6 months])–was only 70 percent higher for illegitimate births than legitimate infants. In Bavaria the corresponding differences were a 39 percent excess mortality for illegitimate infants during the first six months versus a 16 percent excess during the second six months. For Saxony, the difference was a 60 percent excess versus an 18 percent excess. This general pattern held in both the urban and rural sectors.