1 The research reported in the article was supported in part by grant no. ROI HD23693 from the National Institute of Child Health and Human Development, and by Project Quest, the University Research Institute and the College of Liberal Arts of the University of Texas at Austin. Ruth Berg and Amy Holmes assisted in key parts of this research. An earlier version of this article was presented at workshops at the University of Illinois-Urbana, Western Washington University, and the University of Quebec at Chicoutimi. I received valuable advice about earlier drafts from the audiences at those seminars, as well as from Kenneth Fliess, Susan Watkins, Lauren Kattner, Donald Parkerson and JoAnn Parkerson.

2 For the earliest research, see Stouffer, Samuel A., ‘Trends in the fertility of Catholics and non-Catholics’, American Journal of Sociology 41 (1935), 143–66CrossRefGoogle Scholar; and Robinson, Gilbert K., ‘The Catholic birth-rate: further facts and implications’, American Journal of Sociology 41 (1936), 757–66.CrossRefGoogle Scholar For a survey of later research, see Mosher, William D. and Hendershot, Gerry E., ‘Religious affilitation and the fertility of married couples’, Journal of Marriage and the Family 46 (1984), 671.CrossRefGoogle Scholar

3 Jones, E. and Westoff, C. F., ‘The end of “Catholic” fertility’, Demography 16 (1979), 209–18Google Scholar; Mosher, William D. and Hendershot, Gerry E., ‘Religion and fertility: a replication’, Demography 21 (1984), 185–91CrossRefGoogle ScholarPubMed; Mosher and Hendershot, ‘Religious affiliation’.

4 See the sources cited in note 2 above.

5 Freedman, R., Whelpton, P. K., and Campbell, A., Family planning, sterility and population growth (New York, 1959)Google Scholar; Whelpton, P. K., Campbell, A., and Patterson, J. E., Fertility and family planning in the United States (Princeton, 1966)CrossRefGoogle Scholar; Westoff, C. F., Potter, R. G., Sagi, P. C. and Michler, E., Family growth in metropolitan America (Princeton, 1961)CrossRefGoogle Scholar; Westoff, C. F., Potter, R. G. and Sagi, P. C., The third child (Princeton, 1963)CrossRefGoogle Scholar; Westoff, C. F., Potter, R. G. and Sagi, P. C., ‘Some selected findings of the Princeton Fertility Study: 1963’, Demography 1 (1964), 130–5CrossRefGoogle Scholar; Ryder, Norman B. and Westoff, C. F., Reproduction in the United States: 1965 (Princeton, 1971).Google Scholar

6 Jones and Westoff, ‘The end of “Catholic” fertility’.

7 Donald, and Parkerson, Jo Ann, ‘“Fewer children of greater spiritual quality”: religion and the decline of fertility in nineteenth-century America’, Social Science History 12 (1988), 49–70.Google Scholar

8 See, for example, Tolnay, Stewart E., Graham, Stephen N. and Guest, Avery, ‘Own-child estimates of US fertility, 1886–99’, Historical Methods 15 (1982), 127–38CrossRefGoogle Scholar; and Tolnay, Stewart E. and Guest, Avery, ‘American family building strategies in 1900: stopping or spacing’, Demography 21 (1984), 9–18CrossRefGoogle ScholarPubMed, among a number of other articles cited later.

9 Haines, Michael R., ‘Fertility and marriage in a nineteenth-century industrial city: Philadelphia, 1850–1880’, Journal of Economic History 40 (1980), 151–8.CrossRefGoogle Scholar

10 The data were originally collected as part of a class Kenneth H. Fliess and I taught in 1985–1986. That class is described in Gutmann, M. P. and Fliess, K. H., ‘Reconstructing a historical community’, in Proceedings of the 1986 Academic Information Systems University AEP Conference (Milford, Connecticut, 1986)Google Scholar, and in Gutmann, M. P., ‘Teaching historical rsearch skills to undergraduates: thoughts on microcomputers and the classroom’, Historical Methods 21 (1988), 112–20.CrossRefGoogle Scholar The student papers written by the members of that class are included in Gutmann, M. P. and Fliess, K. H. eds., Fredericksburg, Texas: reconstructing a historical community (Austin, 1988).Google Scholar I thank David Gadbois and Taedoo Chung for preliminary papers on fertility in Gillespie County.

11 Jordan, T. G. has done the best recent work on the history of the Texas German community, including Gillespie County. See his German seed in Texas soil (Austin, 1966)Google Scholar, among other works. For older and more local history, see Biesele, Rudolph L., The history of the German settlements in Texas, 1831–1861 (Austin, 1930)Google Scholar; and Gillespie County Historical Society, Pioneers in God's hills; a history of Fredericksburg and Gillespie County: people and events, 2 vols (Austin, 1960).Google Scholar

12 Skolnick, M. et al. , ‘Mormon demographic history, I: Nuptiality and fertility of once-married couples’, Population Studies 22 (1978), 5–19.CrossRefGoogle Scholar

13 See Smith, Daniel Scott, ‘Parental power and marriage patterns: an analysis of historical trends’, Journal of Marriage and the Family 35 (1973), 419–28.CrossRefGoogle Scholar

14 For the methods used in this project, see Gutmann, M. P. et al. , ‘Keeping track of our treasures: managing historical data with relational database software’, Historical Methods 22 (1989).CrossRefGoogle Scholar For a list of sources for the data reported in this paper and a discussion of the special problems of the study of the demographic history of Texas, see Gutmann, M. P. and Fliess, K. H., ‘How to study southern demography in the nineteenth century: the early lessons of the Texas Demography Project’, Texas Population Research Paper, 1990.Google Scholar

15 Watkins, Susan Cotts and Gutmann, M. P., ‘Methodological issues in the use of population registers for fertility analysis’, Historical Methods 16 (1983), 109–20CrossRefGoogle ScholarPubMed; Alter, George, Family, fertility and the demographic transition (Madison, 1988).Google Scholar

16 For reasons of privacy we only used Gillespie County church data up to the end of 1910.

17 It is important to understand that we are classifying the fertility of couples according to the date of birth of the wife. While period-specific measures might be easier to compute, and marriage cohorts are more conventional, the patterns of denominational differentials and of change over time are clearer using cohort data. Summary measures like the Total Marital Fertility Rate are problematic under all circumstances because they mingle a variety of fertility experiences. The real changes took place among groups of women who were born or reached adulthood together. Their experience is best revealed cohort by cohort.

18 Women in the fourth cohort were born after 1869, and therefore only reach age 40 - at the greatest - by the time our data end in 1910. We can estimate a value for total marital fertility by assuming that age-specific marital fertility rates for ages 40–44 would have fallen to about 80, and for ages 45–49 would have fallen to about 10.

19 Stern, Mark J., Society and family strategy, Erie County, New York 1850–1920 (Albany, 1987)Google Scholar; Coale, A. J. and Zelnick, Melvin, New estimates of fertility and population in the United States: a study of annual white births from 1855 to 1960 and of completeness of enumeration in the censuses from 1880 to 1960 (Princeton, 1963).CrossRefGoogle Scholar

20 There is a discussion of the churches, their buildings and schisms in Pioneers in God's hills, as well as in Penninger, Robert, Fredericksburg, Texas: the first fifty years, trans. Wisseman, C. L. (Fredericksburg, 1971; originally published 1896), 59–65.Google Scholar

21 See the sources discussed in Gutmann and Fliess, ‘How to study Southern demography’.

22 Catholic settlers in New Braunfels, a larger German settlement between Austin and San Antonio, found themselves unwelcome, and either left the town or migrated further to settlements like Fredericksburg and Gillespie County. Lauren Kattner, ‘Land and marriage: German regional reflections in four Texas towns, 1845–1860’, unpublished paper presented at the 1989 German-American Historical Meeting.

23 In Yorktown, Texas, the Lutheran church kept two sets of sacramental registers, one which recorded the sacraments performed for ‘members’, the other for ‘non-members’. This division grew out of the requirement that Lutherans sign the so-called ‘Pittsburgh’ constitution of the General Synod of the Lutheran Church in return for church membership, which many were unwilling to do. For details, see Mgebroff, Johannes, Geschichte der Ersten Deutschen Evangelisti-Lutherischen Synode in Texas (Austin, 1902)Google Scholar, and Ziehe, H. C., A centennial story of the Lutheran Church in Texas (Taylor, Texas, 1951).Google Scholar

24 Gutmann and Fliess, ‘How to study Southern demography’.

25 Coale, A. J. and Trussell, T. James, ‘Model fertility schedules: variations in the age structure of childbearing in human populations’, Population Index 40 (1974), 185–258CrossRefGoogle ScholarPubMed; Coale, A. J. and Trussell, T. James, ‘A new procedure for fitting optimal values of the parameters of a model schedule of marital fertility rates’, Population Index 44 (1978), 211–13.Google Scholar

26 It is likely that marriage ages were in fact slowly rising, but that the data used here are contaminated because marriage ages for the youngest cohort are truncated at age 40. The somewhat younger marriage ages shown in our data for the fourth cohort are not conclusive.

27 For another view of this question, see Fairchild, Amy, ‘The effects of transition, economy and culture on nuptiality in Fredericksburg, Texas, 1750–1910’, in Gutmann and Fliess, Fredericksburg, 57–78.Google Scholar

28 This work was originally reported in Knodel, John, ‘Natural fertility in pre-industrial Germany’, Population Studies 32 (1978), 481–510.CrossRefGoogle ScholarPubMed These results, and others are also reported in Knodel, John, Demographic behavior in the past: a study of fourteen German village populations in the eighteenth and nineteenth centuries (Cambridge, 1988), 119–52.CrossRefGoogle Scholar

29 Fliess, Kenneth H., ‘Fertility, nuptiality and family limitation among the Wends of Serbin, Texas, 1854–1920’, Journal of Family History 13 (1988), 251–64CrossRefGoogle Scholar; Haines, ‘Fertility and marriage’.

30 See Tolnay, Graham and Guest, ‘Own-child estimates’; Byers, E., ‘Fertility transition in a New England commercial center: Nantucket, Massachusetts, 1680–1840’, Journal of Interdisciplinary History 13 (1982), 17–40CrossRefGoogle Scholar; and Haines, ‘Fertility and marriage’.

31 Skolnick et al., ‘Mormon demographic history’; Ericksen, J. et al. , ‘Fertility patterns and trends among the Old Order Amish’, Population Studies 33 (1979), 255–76.CrossRefGoogle ScholarPubMed

32 See Table 1 in Fliess, ‘Fertility, nuptiality and family limitation’, p. 255, for comparisons.Google Scholar

33 Knodel, ‘Natural Fertility’; Knodel, Demographic behavior.

34 Knodel, John, The decline of fertility in Germany, 1871–1939 (Princeton, 1974).Google Scholar

35 There is an excellent discussion of these and related concepts in Knodel, Demographic behavior in the past. See especially chapters 10, ‘Trends in marital fertility and underlying natural fertility components’, and 12, ‘Starting, stopping, spacing and the fertility transition’.

36 Henry, Louis, ‘Some data on natural fertility’, Eugenics Quarterly 8 (1961), 81–91.CrossRefGoogle ScholarPubMed

37 John Knodel is one of these. Knodel, ‘Natural fertility’; Knodel, Demographic behavior; Knodel, John, ‘Birth spacing and family limitation: a critique of the Dupaquier-Lachiver technique for detecting birth control from family reconstitution data’, Annales E.S.C. 36 (1981), 473–94Google Scholar; Knodel, John, ‘Demographic transitions in German villages’, in Coale, Ansley J. and Watkins, Susan Cotts eds., The decline of fertility in Europe (Princeton, 1986), 337–89.Google Scholar

38 Dupaquier, J. and Lachiver, M., ‘Sur les debuts de la contraception en France ou les deux Malthusianismes’, Annales E.S.C. 24 (1969), 1391–406.Google ScholarAnderton, Douglas L. and Bean, Lee L., ‘Birth spacing and fertility limitation: a behavioral analysis of a nineteenth-century frontier population’, Demography 11 (1985), 169–83.CrossRefGoogle Scholar

39 See the sources cited in note 25 above.

40 Underregistration of births at all ages would occur because the fertility of unknowns in the sample data is drawn exclusively from the censuses, which might underreport children who died or moved out of their parents' home between censuses. This would appear in our data as lower age-specific marital fertility rates. The rates for 15–19 are especially troubling in this regard. To guard against the risk of such Underregistration we have included only women for whom the total number of children born as reported in the 1900 and 1910 censuses agrees with the number of children for whom we have dates of birth. All women for whom these figures do not agree have been excluded. We are fairly confident of the result, although there continues to be a risk of a bias which puts excessive numbers of childless women in the ‘unknown’ group, instead of excluding them completely from analysis, or putting them in one of the denominational groups.

41 Anderton and Bean (1985) have devised methods of analysis based on a classification of birth-interval length by final family size, but this analysis cannot be productively done for truncated data such as ours.

42 The issue is whether one searches for absolute or proportional changes in fertility. The absolute change in fertility presented in this article leads clearly to the conclusion of convergence. The relative change in fertility is not as clear, but I believe the correct interpretation is that convergence was already under way at the beginning of the twentieth century.