¹ Parkin, T. G., Demography and Roman Society (1992)Google Scholar, offers both a good résumé and a critique of the existing studies. For a specific high-quality study that incorporates current conceptions and methods, see Bagnall, R. and Frier, B., The Demography of Roman Egypt (1994)CrossRefGoogle Scholar.

² Morris, I., Death-Ritual and Social Structure in Classical Antiquity (1992CrossRefGoogle Scholar).

³ Bodel, J., Graveyards and Groves: A Study of the Lex Lucerina = AJAH 11 (1986 [1994])Google Scholar, now offers a standard point of departure; see also K. Hopkins, in collaboration with Letts, M., ‘Death in Rome’, ch. 4 in Death and Renewal: Sociological Studies in Roman History 2 (1983), 201–56Google Scholar.

⁴ Storey, G. R., Preindustrial Urban Demography: the Ancient Roman Evidence (2 vols), PhD thesis, The Pennsylvania State University (1992)Google Scholar, recapitulates the earlier arguments, evidence, and bibliography.

⁵ Pergola, P., ‘Le catacombe romane: miti e realtà (a proposito del cimitero di Domitilla)’, ch. 7 in Giardina, A. (ed.), Società romana e impero tardoantico, 2: Roma: politico, economia, paesaggio urbano (1986), 333–50. 489–9, at 337.Google Scholar

⁶ Müller, N., ‘Koimeterien, die altchristlichen Begräbnisstätten’, Realencyclopädie für protestantische Theologie und Kirche³ 10 (1901), 794–877Google Scholar, at § 5, based on the estimates of de Rossi and Marchi — estimates as low as about 500 kilometres have been proffered by some, but seem to be based on older and incomplete evidence.

⁷ See the description in Josi, E., ‘Cimiteri Cristiani Antichi’, Enciclopedia Cattolica 3 (1950), 1617–37Google Scholar, at 1626.

⁸ Müller, op. cit. (n. 6), § 4.3; and Leclercq, H., ‘Catacombes’, DACL 2.2 (1910), 2432–5Google Scholar, noting that de Rossi places the final abandonment of the catacombs in the mid-fifth century.

⁹ Josi, op. cit. (n. 7), 1621.

¹⁰ Guyon, J., ‘La vente des tombes à travers 1'épigraphie de la Rome chrétienne (III^e–VII^e siécles): le rôle des fossores, mansionarii, praepositii et prêtres’, MEFRA 86 (1974), 549–96CrossRefGoogle Scholar; for the practices in the Jewish catacombs see Leon, H. J., ‘The Sepulchral Formulas and Epithets’, ch. 6 in The Jews of Ancient Rome (1960), 122–34Google Scholar, with important criticisms and corrections by Williams, M. H., ‘The organisation of Jewish burials in ancient Rome in the light of evidence from Palestine and the Diaspora’, ZPE 101 (1994), 165–82Google Scholar (‘a plurality of consortia, each offering a range of facilities, from unmarked loculi at one end of the scale to spacious, arcosolia-filled cubicula at the other’).

¹¹ Josi, op. cit. (n. 7), 1626, guessing at ‘una epidemia’ — perhaps, but not necessarily: see Hopkins and Bodel above on mass burials at Rome in the pre-Christian period.

¹² Nordberg, H., Biometrical Notes: The Information on Ancient Christian Inscriptions from Rome concerning the Duration of Life and the Dates of Birth and Death, Acta Instituti Romani Finlandiae 11.2 (1963)Google Scholar, who cumulated evidence from IC 1–2; ICUR 1–3; and SICV; see, more recently, Ferrua, A., ‘Saggio biometrico sulle iscrizioni cristiane delta Nomentana e della Salaria’, RAC 64 (1988), 43–63Google Scholar. Since my study was submitted for publication, Scheidel, W., ‘Libitina's bitter gains: seasonal mortality and endemic disease in the ancient City of Rome’, Anc. Soc. 25 (1994), 151–75CrossRefGoogle Scholar, appeared. This study, which highlights the evidence for what Scheidel regards as the main disease vector that caused the high autumnal mortality in the City of Rome (malaria), is based on the figures in Nordberg, with additional samples drawn from IC and ICUR, vols 3–5.

¹³ Rebillard, E., ‘Κοιμητήριον et coemeterium: tombe, tombe sainte, nécropole’, MEFRA 105 (1993), 975–1001CrossRefGoogle Scholar, has demonstrated how the term never had the significance of graveyard or ‘necropolis’ (itself a term not commonly attested in Antiquity) but rather that of individual grave and, by extension, the grave — i.e., that of a martyr — or the sacred edifices adjacent to the burial place of a martyr.

¹⁴ Consider: ‘natus in pace’: ICUR 9046 (for death); ‘natus’: 9060 (death); ‘natale suo’: 9228 (day of death); ‘quo et natus est cuius anima cum sanctus in pace’: 15634; ‘natus in pace’: 24060 (day of death); ‘qui natus’: 24180 (death).

¹⁵ Janssens, J., Vita e morte del cristiano negli epitaffi di Roma anteriori al sec. VII (1981), 1.1.5Google Scholar, ‘I neofiti’, 26–32, notes how often Christians took care to emphasize the ritual of baptism, even when performed shortly before death itself.

¹⁶ On the new ideology see Rush, A. C., Death and Burial in Christian Antiquity, The Catholic University of America Studies in Christian Antiquity 1 (1941)Google Scholar, esp. pt. 1, ch. 3, ‘Death as Birth. The Day of Death as Dies Natalis’, 72–87. See also Stuiber, A., ‘Geburtstag’, RAC 9 (1976), 217–43Google Scholar, esp. 220–33, on the ways in which Roman practices were redefined and designated by Christians in connection with the absolute value that they placed on life-after-death, especially as exemplified by the deaths of the martyrs.

¹⁷ Tertullian, De Corona 3; cf. De Exhortatione Castitatis 11; and De Monogamia 10.

¹⁸ Not that they did not care to know how long a person lived or to record that number if possible; otherwise the plaintiff cry of ‘que quod vixit annis nescio quando’ of ICUR 24564 would make no sense.

¹⁹ Nordberg, op. cit. (n. 12), 35, and his table 5.

²⁰ F. Grossi Gondi, Trattato di epigrafia cristiana latina e greca del mondo romano occidentale (1920; reprint 1968), 96–7, who rightly rejected the explanations of Le Blant.

²¹ Nordberg, op. cit. (n. 12), 52–3 already saw this (N = 14, where only two were in excess: three and four days difference respectively). In our sample, where the number of cases rose to N = 27, there were four cases of four-day delays, three of three-days, and the rest were only one- or two-day delays between death and burial.

²² Nordberg, op. cit. (n. 12), 61.

²³ Frey, J. B., Corpus of Jewish Inscriptions: Jewish Inscriptions from the Third Century B.C. to the Seventh Century A.D.: Europe (1936; reprint 1975)Google Scholar. There are a few exceptions to this rule, but they are few indeed (e.g. Nos 68, 242, 268, 271, 457, 476, 482, and 527 in Frey's collection — of which most seem to me to be dubious or even probably non-Jewish, despite being included in Frey).

²⁴ Leon, H. J., ‘The Sepulchral Formulas and Epithets’, ch. 6 in The Jews of Ancient Rome (1960), 122–34.Google Scholar

²⁵ Frey, op. cit. (n. 23), No. 476 = ILCV 4933; cf. Leon, op. cit. (n. 24), 248 f.; the rest of the sentiments in the inscription: ‘rursum victura, reditura ad lumina rursum’, that ‘hoc tibi praestiterit pietas, hoc vita pudica’, etc., are not necessarily specifically Jewish.

²⁶ Testini, P., Archeologia cristiana² (1980)Google Scholar; Carletti, C., Iscrizioni cristiane di Roma: testimonianze di vita cristiana, secoli III–VIII (1986)Google Scholar; and his ‘Epigrafia cristiana, epigrafia dei cristiani: alle origini della terza eta dell'epigrafia’, in A. Donati (ed.), La terza etá dell'epigrafia (1988), 115–36. Galvao-Sabrinho, C. R., ‘Funerary epigraphy and the spread of Christianity in the West’, Athenaeum 83 (1995), 431–66Google Scholar, argues that the Christian revival of the ‘epigraphie habit’ was provoked by new concerns with ‘self identity’ that were centred on the problems of the body, resurrection, and on new senses of time and place.

²⁷ See Table 1 (‘Primary Sources’). The re-edition of the Christian epigraphy of the City of Rome begun by de Rossi, Silvagni, and Ferrua, was meant to supercede and replace the earlier project begun by de Rossi himself, of which the first two volumes and a supplement were produced (1857–1915). In collating the inscriptions, one had to be careful to eliminate possible repetitions from one corpus to another. To avoid these, care was taken to use the standard crossreference guides: Moreau, J. and Marrou, H. I., ‘Concordantiarum Tabulae’, in ILCV vol. 4: Supplementum (1967), 63–165Google Scholar; Ferrua, A. and Feissel, D., ‘C. Wessel et ICVR Concordats’, RAC 67 (1991), 37–68Google Scholar; Worp, K. A., ‘Konkordanzen zu C. Wessel, Inscriptiones Graecae Christianae Veteres Occidents’, ZPE 87 (1991), 275–90Google Scholar.

²⁸ It can be compared, for example, with the approximately 300 data sets provided by the Egyptian census returns, on which see Bagnall and Frier, op. cit. (n. 1). In this study, we are able to analyse a sample of the order of ten times that scale — a considerable improvement in quantity.

²⁹ This represents a significant decline in the percentage of stones that record relationships as found in the epigraphy of the inhabitants of the city in the pre-Christian period — about 62–63 per cent of all: see Saller, R. P. and Shaw, B. D., ‘Tombstones and Roman family relations in the Principate: civilians, soldiers and slaves’, JRS 74 (1984), 124–56Google Scholar, at 147, col. 4; and Shaw, B. D., ‘Latin funerary epigraphy and family life in the later Roman Empire’, Historia 33 (1984), 457–97Google Scholar, at 467 f.

³⁰ These proportions remained relatively constant as new data were added to the sample. As a footnote to the historical study of epigraphy, however, it might be noted that the editors ot ICUR had a disconcerting tendency to resolve fragmentary inscriptions as masculine in cases where either gender would be possible (and in a set where there is no great overall male gender preference): see, e.g., 14253 and 18096.

³¹ Shaw, B. D., ‘The Cultural Meaning of Death: Age and Gender in the Roman Family’, ch. 4 in Kertzer, D. I. and Saller, R. P. (eds), The Family in Italy from Antiquity to the Present (1991), 66–90Google Scholar, at 80 f. and table 4.5, pp. 81–2.

³² See the works cited in nn. 6–9 above, and also de Rossi, G. B., La Roma sotterranea cristiana (3 vols, 1864–1877)Google Scholar; Armellini, M., Gli antichi cimitieri di Roma e d'Italia (1893)Google Scholar; Marucchi, O., Le catacombe romane secondi gli ultitni studi e le più recenti scoperte (1903; 2nd edn, 1933)Google Scholar; Leclerq, H., ‘Coemeteria’, DACL 2.2 (1910), 2376–486Google Scholar; Styger, P., Die römische Katakomben: archäologische Forschungen über den Ursprung und die Bedeutung der altchristlichen Grabstätten (1933)Google Scholar; Fasola, U., ‘Katakomben’, Lexikon für Theologie und Kirche² 6 (1961), 20–44Google Scholar; Testini, P., Le catacombe e gli antichi cimitieri cristiani in Roma (1966)Google Scholar; Fasola, U. and Testini, P., ‘I cimitieri cristiani’, in Atti del IX Congresso internazionale di archeologia cristiana 1 (1978), 103–210Google Scholar.

³³ See, e.g., ICUR 18503, 23460, 18503, 17629, 221, 752 (a nummerarius), 4855, 5043, 5748, 5798, 13487, 13491, 13696, and 13940.

³⁴ The number of instances is frustratingly small and seems to indicate a further change in Christian values and self-presentation (also somewhat limited by the physical nature of the Christian epitaph itself). About fifty instances is an exiguous proportion of the total (about 1.5 per cent), whereas for pre-Christian Rome, the ‘occupational’ inscriptions of CIL VI (N = 1,470) seem to represent a rather larger proportion of the total (about 5 per cent); see S. Joshel, Work, Identity, and Legal Status at Rome: A Study of the Occupational Inscriptions (1992), 16 f.

³⁵ See, e.g., ICUR 17525, 22635, 23064, 1213, 1477, 4928, and 11748.

³⁶ A study is required that would extend the work of Joshel, op. cit. (n. 34), into the evidence of this later period in the history of the metropolis; for these examples see ICUR 13141, 20830, 19041, 3, 529, 1403, 1761, 2223, 4283, 4328, 5027, 34, 5047, 5053, 14583, 15389, and 944.

³⁷ For example, the cemetery ad duas lauros on the Via Labicana to the south-east of the city studied by Guyon, J., Le cimetière aux Deux Lauriers: recherches sur les catacombes romaines (1987), 352–9.Google Scholar

³⁸ For the type of analysis used here, and for the comparative figures that will be used, see Saller and Shaw, op. cit. (n. 29), with tables; and Shaw, op. cit. (n. 29), with tables.

³⁹ Janssens, op. cit. (n. 15), 1.5, ‘Le relazioni fra altri membri della famiglia’, 169–70.

⁴⁰ Janssens, op. cit. (n. 15), 2.2, ‘I rapporti con i liberti’, 179–81; and 2.3, ‘I rapporti con gli alunni’, 181–90.

⁴¹ ICUR 20819.

⁴² ICUR 10953.

⁴³ Janssens, op. cit. (n. 15), 1.1, ‘Le relazioni fra marito e moglie’, 103–32, justifiably placed first in his analysis of intra-family relationships.

⁴⁴ Janssens, op. cit. (n. 15), 1.1, ‘Le relazioni fra genitorie figli’, 132–56.

⁴⁵ Janssens, op. cit. (n. 15), 1.1, ‘Le relazioni fra figli egenitori’, 156–62.

⁴⁶ Shaw, op. cit. (n. 29), 465 f., and tables 7–14, esp. 8–10.

⁴⁷ Testini, P., ‘Aspetti di vita matrimoniale in antiche iscrizioni funerarie cristiane’, Lateranutn 42 (1976), 150–64Google Scholar.

⁴⁸ Leclercq, H., ‘Biscandens, Bisomus’, DACL 2.1 (1910), 910–15Google Scholar.

⁴⁹ Guyon, op. cit. (n. 37), 571: more than 40 per cent of gravesites purchased from fossores have their type specified (most frequently a bisomus); one-third of all contracts are already made by a family group, most often by a conjugal pair.

⁵⁰ ICUR 23891.

⁵¹ Borda, M., Lares. La vita familiare romana nei documenti archeologici e letterari (1947).Google Scholar

⁵² Saller and Shaw, op. cit. (n. 29), esp. tables 1–6.

⁵³ ICUR 1439.

⁵⁴ Consider the sentiments of bereavement in ICUR 1637; cf. No. 1978, a father for his son Dalmatius (dead at age six): a long poem on his schooling and learning Greek and Latin letters.

⁵⁵ On the tendency of urban and Christian populations to accord greater public recognition to infants and children in funerary commemoration, see Shaw, op. cit. (n. 29), 473 f.; (n. 31), 75 f.

⁵⁶ Sakamoto-Momiyama, M., Seasonality in Human Mortality: A Medico-Geographical Study (1977), 20–5, 43 f., 49.Google Scholar

⁵⁷ Office of Population Censuses and Surveys, Series DH₄ No. 16: Mortality Statistics: Injury and Poisoning … England and Wales, 1990, HMSO London (1992)Google Scholar, table 2, pp. 2–4.

⁵⁸ di Comite, L., La Mortalità in Italia, Istituto di Demografia (1974)Google Scholar, Tavola 93: ‘Indici di stagionalitá dei decessi per taluni gruppi di cause di morte, 1958–67’, p. 195.

⁵⁹ Di Comite, op. cit. (n. 58), 187; see table 4 in Ferrari, G. and Bacci, M. Livi, ‘Sulle relazione tra temperatura e mortalita nell' Italia Unita, 1861–1914’, in La populazione italiana nell' ottocento: continuità e mutamenti (1985), 273–98, at 279.Google Scholar

⁶⁰ Already noted for the so-called ‘advanced’ countries by the early 1960s: Tomlinson, R., Population Dynamics: Courses and Consequences of World Demographic Change (1965), 110–13Google Scholar; on variation for precise disease groups still discernable in these advanced ‘aseasonal’ or ‘deseasonal’ populations, see, e.g., Cox, P., Demography⁵ (1976), 133–5, and table 7.10.CrossRefGoogle Scholar

⁶¹ Dyson, T. and Crook, N., ‘Seasonal Patterns in Births and Deaths’, ch. 5 in Chambers, R., Longhurst, R. and Pacey, A. (eds), Seasonal Dimensions to Rural Poverty (1981), 135–62Google Scholar.

⁶² Wrigley, E. A. and Schofield, R. S., ‘Short-Term Variations: Some Basic Patterns’, ch. 8 in The Population History of England, 1541–1871: A Reconstruction (1981), 285–355, at 285.Google Scholar

⁶³ Sargent, F., Hippocratic Heritage: A History of Ideas about Weather and Human Health (1982).Google Scholar

⁶⁴ Hdt. 11.77.

⁶⁵ Sargent, op. cit. (n. 63), 54–5 and table 2.1.

⁶⁶ Sargent, op. cit. (n. 63), 56–8, plus Appendix 11, tables 3–6.

⁶⁷ Grmek, M. D., Les maladies à l'aube de la civilisation occidentale: recherches sur la réalité pathologique dans le monde grec préhistorique, archaïque et classique (1983), 279 f.Google Scholar

⁶⁸ Lucretius, , De rerum natura, VI. 1090–137.Google Scholar

⁶⁹ ILCV 4756 = NdS (1916), 126 f.; an image of the sarcophagus, originally found in the cemetery of S. Cyriacus on the Via Ostiense, and now in the Museo delle Terme, can be found in Wilpert, G., I sarcophagi Cristiani antichi, Volume primo, Tavole (1929), tav. LIX.4Google Scholar; with discussion in Volume primo, Testo (1929), 79–80.

⁷⁰ B. D. Shaw, ‘The seasonal birthing cycle of Roman women’ (forthcoming).

⁷¹ Herlihy, D. and Klapisch-Zuber, C., Les Toscans et leurs families. Une etude du catasto florentin de 1427 (1978), 192Google Scholar; del Panta, L., Le epidemie nella storia demografica italiana: secoli XIV–XIX (1980), 47–63.Google Scholar

⁷² Corsini, C., ‘Problemi di utilizzazione dei dati desunti dai registri di sepolture e morti’, in Problemi di utilizzazione delle fonti di demografia storica: Atti del seminario di demografia storica 1972–73, 2 (1974), 93–135Google Scholar, at 124–30, ‘La stagionalità’, with ‘Tabella 7: Indici di stagionalità dei decessi per zona, negli anni di crisi et nel resto del periodo preso in esame’, and ‘Grafico 3: Indici di stagionalità dei decessi nelle zone indicate’.

⁷³ Natale, M. and Bernassola, A., La mortalità per causa nelle regioni italiane (1973), 73 f. and fig. 8Google Scholar, ‘Curve di mortalità per sesso delle generazioni del 1870, 1910, 1950’.

⁷⁴ Delille, G., ‘Un problema di demografia: uomini e donne di fronte alla morte’, in Sori, E. (ed.), Demografia storica (1975), 257–84Google Scholar, at 272 f.: it is especially true that in crises of mortality more women and children die than do men: see p. 273, a table of mortality on the early nineteenth-century South. A month-by-month tally under ‘normal’ conditions still shows more women than men dying in the early spring and in late summer/early fall (July to October). A second case, that of Casoria, reveals a worse scenario where female mortality consistently outpaced that of males in every month of the year, being particular bad in early spring (May) and then again in the July to October period.

⁷⁵ Delille, op. cit. (n. 74), 278–9, and chart on p. 273; thus confirming on a regional basis the global observations made by Sen, Amartya, ‘More than 100 million women are missing’, NRYB (20 December, 1990), 61–6Google Scholar, and Johansson, S. R., ‘Welfare, mortality, and gender: continuity and change in explanations of male/female mortality differences over three centuries’, Continuity and Change 6 (1991), 135–77CrossRefGoogle Scholar.

⁷⁶ Tabutin, D., ‘La surmortalité féminine en Europe avant 1940’, Population 33 (1978), 121–48CrossRefGoogle Scholar. The pattern has been reversed only since the Second World War in more developed countries so that it is now generally true that women have a consistently higher life expectancy than men, even in most so-called ‘developing’ countries, see: ‘Patterns of Sex Differentials in Mortality in Less Developed Countries’, in A. D. Lopez and L. T. Ruzicka (eds), Sex Differentials in Mortality: Trends, Determinants and Consequences (1983), 7–32. Differences in mortality rates are most apparent in the fifteen to forty-four age range.

⁷⁷ Evidence drawn from Ferrari and Livi Bacci, op. cit. (n. 59), 280: ‘Tabella 5: Indici di stagionalità dei decessi per classi di età, Italia 1869’.

⁷⁸ Di Comite, op. cit. (n. 58), 190 f.; table 91, p. 190 and fig. 21, p. 191.

⁷⁹ Ferrari and Livi Bacci, op. cit. (n. 59), 280: ‘Tabella 5: Indici di stagionalita dei decessi per classi di età, Italia 1869‘.

⁸⁰ Patlagean, E., Pauvreté économique et pauvreté sociale à Byzance 4e–7e siècles (1977), 92–4.CrossRefGoogle Scholar

⁸¹ Boyaval, B., ‘La mortalité saisonnière dans l'Egypte gréco-romaine’, in Vercoutter, J. (ed.), Livre de centenaire, 1880–1980, de l'Institut français d'archéologie orientate du Caire (1981), 281–6Google Scholar; the use of the ‘mummy labels’ as evidence is discussed by the same author in his, ‘Remarques à propos des indications d'âges des étiquettes des momies’, ZPE 18 (1975), 49–74Google Scholar; for their chronological distribution see his, ‘La date sur les etiquettes de momies’, BASP 18 (1981), 101–18Google Scholar, at 116 (before the second century: 3; A.D. 100–150: 4; A.D. 150–211: 15; third-century: 54).

⁸² There were forty of these known at the time Montevecchi made the first comprehensive study of them: Montevecchi, O., ‘Ricerche di sociologia nei documenti dell'Egitto greco-romano, V: Le denunce di morte’, Aegyptus 25 (1946), 111–29Google Scholar; the number has now more than doubled: Casarico, L., Il controllo delta popolazione nell'Egitto romano, 1: Le denunce di morte, Corpora Papyrorum Graecarum 2 (1985)Google Scholar; supplemented by Molyviati-Toptsi, U., ‘A death certificate from the Berkeley collection’, ZPE 77 (1989), 281–2Google Scholar, and the new finds listed by her.

⁸³ Compare Montevecchi, op. cit. (n. 82), 112–13, and Casarico, op. cit. (n. 82), 25–33, with Bagnall and Frier, op. cit. (n. 1), 6–9, and their table 1.2 and fig. 1.1.

⁸⁴ Not solely by males, as Montevecchi, op. cit. (n. 82), 121, thought; Ulrich Wilken was right (Griechische Ostraka aus Aegypten und Nubien 1 (Leipzig, 1899; reprint, New York, 1979), 454) — women were included, though perhaps only in default of other declarants. There are now fifteen female cases: see Casarico, op. cit. (n. 82), 12.

⁸⁵ The dates on which they were filed (which is sometimes different than the month of death), however, are sometimes much later in the year, but always, it seems, within six months of the month of Mecheir (late January/February) — with a tendency to ‘bunch’ in the months immediately before and including Mecheir — a pattern that prompted the hypothesis that the filing had to be made by that month for fiscal purposes: Montevecchi, op. cit. (n. 82), 117–19, and Casarico, op. cit. (n. 82), 17; an hypothesis first proposed by Wilcken: Mitteis, L. and Wilcken, U., Grundzüge und Chrestomathie für Papyruskunde (1912), 1, 196Google Scholar (though, as Montevecchi cautions, there is nothing in the form of the documents themselves that states as much).

⁸⁶ I surveyed the decades of the 1920s and 1930s in the statistical charts furnished by the Annuaire statistique de l'Egypte under the section entitled ‘Distribution mensuelle des naissances et des décès par gouvernats et provinces’; the years 1923 (Annuaire statistique de l'Egypte, 1923–1924, Cairo, 1925, tableau IV, pp. 50–1) and 1926 (Annuaire statistique de l'Egypte, 1926–1927, Cairo, 1928, tableau iv, pp. 60–1) were chosen as exemplary of the general pattern. It should be noted that the seasonal mortality for Lower Egypt (the Delta) has been rather more erratic and unpredictable than that of Upper Egypt. Since the figures were given for provinces, it was also possible to track the specific figures for the ‘Fayoum’, from whence most of the papyrological evidence derives, to see if there was a seasonal regime of death that was peculiar to its environment.

⁸⁷ See n. 85 above; the problem is that such an interpretation must hypothesize an indirect linkage between the ‘filing dates’ of the death notices and the affixing of labels to mummy caskets and a propensity to record certain seasonal death patterns.

⁸⁸ See Table 1 ‘Sources’, for the corpora of inscriptions used in the collection of the primary data.

⁸⁹ Part of the reason for keeping the ‘Sicily’ sample separate from that of Southern Italy was the fact that most of the epigraphy was in Greek; these data were therefore initially kept separate to see if there might be any cultural differences between them and the Southern Italian set. There were, however, no significant differences that were relevant to seasonal mortality.

⁹⁰ Cardamone, A. F., ‘II ciclo stagionale dei matrimoni, delle nascite e dei decessi a Bitonto dal 1661 al 1800’, in Sori, E. (ed.), Demografia storica (1975), 227–36Google Scholar, esp. ‘Tab. 7: Indici di stagionalità dei decessi, Bitonto, 1661–1800’, p. 235.

⁹¹ Assante, F., Città e campagne nella Puglia del secolo XIX: l'evoluzione demografica (1974), 124–8Google Scholar; the figures are replicated from ‘Grafico 10: Indici di stagionalità in provincia di Bari (1816–1820)’, and ‘Grafico 11: Indici di stagionalità di Foggia (1816–1820)’, p. 126.

⁹² Ligresti, D., Sicilia moderna: le città e gli uomini (1984), 155 and ‘Tabella 24: Indici di stagionalità delle sepolture’.Google Scholar

⁹³ Bellettini, A., La popolazione del Dipartimento del Reno (1965), 188 fGoogle Scholar. and ‘Figura 22: Ciclo stagionale delle morte (1811–12)’, facing p. 190; cf. Marcuzzi, G. and Tasso, M., ‘Seasonality of death in the period 1889–1988 in the Val di Scalve (Bergamo, Pre-Alps, Lombardia, Italy)’, Human Biology 64 (1992), 215–22Google Scholar.

⁹⁴ Lassère, J. M., Ubique Populus: peuplement et mouvements depopulation dans I'Afrique romaine de la chute de Carthage à la fin de la dynastie des Sévères: 146 a.C.–235p.C. (1977), 553–4.Google Scholar

⁹⁵ Logeart, F., ‘Les inscriptions funéraires chrétiennes du Djebel Nif en-Nser (commune mixte d'Aïn M'lila)’, Revue africaine 84 (1940), 5–29Google Scholar; Lassère, op. cit. (n. 94), 554–5; location of the cemetery: S. Gsell, Atlas archeologique, f. 17 (Constantine), no. 441.

⁹⁶ Marcillet-Jaubert, J., Les inscriptions d'Altava, Publications des Annales de la Faculté des Lettres, Aix-en-Provence, 65 (1968)Google Scholar; Lassère, op. cit. (n. 94), 555; fig. 56 (N = 169, dated between A.D. 301 and 599).

⁹⁷ See the cautions in Shaw, op. cit. (n. 29), 476 f., and (n. 31), 78 f. with table 4.4.

⁹⁸ Augustine, Ep. 126.4.

⁹⁹ Cyprian, Ep. XVIII.1.2; see Clarke, G. W., The Letters of St. Cyprian of Carthage, Letters 1–27 (1984), 297–8Google Scholar, who notes that there is no reason to see in these words reference to early stages of the ‘plague of Gallus'or to specific epidemics of malaria: ‘Cyprian is merely making a deduction from general observation about a notoriously unhealthy season’. On the plague, see the description in Cypr., de Mort. and Pontius, Vita Cypr. 9.

¹⁰⁰ Bellettini, A., La popolazione italiana: un profilo storico (1987), 162 f.Google Scholar; until the turn of the century, of all countries in Western Europe, Italy and Spain had the lowest life expectancy at birth, and some of the highest general and infant mortality rates; see Ferrari and Livi Bacci, op. cit. (n. 59), 276 f., for whom the half century following Unification ‘è un periodo durante il quale la mortalità è ancora fortemente caratterizzata da condizioni di ancien régime’.

¹⁰¹ Bellettini, op. cit. (n. 93), 163 f.

¹⁰² Ferrari and Livi Bacci, op. cit. (n. 59).

¹⁰³ Ferrari and Livi Bacci, op. cit. (n. 59), table 2, p. 278.

¹⁰⁴ Bellettini, op. cit. (n. 100), 166–75, with tables 7, 8, and 9, provides a convenient summary.

¹⁰⁵ Grmek, M. D., Les maladies à l'aube de la civilisation occidentale: recherches sur la réalité pathologique dans le monde grec préhistorique, archatque et classique (1983), ch. 7Google Scholar, ‘Une grande tueuse: la tuberculose’, 261–90; and ch. 10, ‘L'hyperostose poreuse du crâne, les anémies héréditaires et l'évolution du paludisme’, 355–407 (Eng. trans., Diseases in the Ancient Greek World (1989)).

¹⁰⁶ Sallares, R., The Ecology of the Ancient Greek World (1991), 237 f.Google Scholar; Scheidel, op. cit. (n. 12), 155–64, argues forcefully that the temporal convergence of the period of highest seasonal mortality in the City of Rome with the period when falciparian malaria is most virulent (both in autumn) permits one to deduce that malaria was probably the principal lethal disease vector. While this might be true, it is difficult on this basis alone to specify malaria as the main cause of the high autumnal mortality in the city, or indeed to separate its effects from those of the other great ‘killer’ diseases of the time, including typhoid, tuberculosis, and severe gastero-intestinal infections.

¹⁰⁷ Di Comite, op. cit. (n. 58), 194 f. and fig. 22, p. 196, cf. Sallares, op. cit. (n. 106), 271 f., selects this as the main disease factor that made autumn ‘the most dangerous season of the year in ancient Greece’.

¹⁰⁸ Bellettini, op. cit. (n. 100), 170 ff. and ‘Tabella 7: Mortalità per gruppi di cause’, 171; and ‘Tabella 9: Morti per gruppi di cause e di età, 1887–89’, 174: the leading causes of death in the period were pulmonary and bronchial infections, gastro-enteritis and colitis, tuberculosis, scarlet fever (and related illnesses), infections peculiar to newborn infants. Malaria accounted for only 2.2 per cent of all mortality during the years surveyed.

¹⁰⁹ Landers, J. and Mouzas, A., ‘Burial seasonality and causes of death in London, 1670–1819’, Population Studies 42 (1988), 59–83CrossRefGoogle Scholar, at 62–3.

¹¹⁰ Scobie, A., ‘Slums, sanitation, and mortality in the Roman world’, Klio 68 (1986), 399–433CrossRefGoogle Scholar. For some comparative studies, see Wrigley, E. A., ‘A simple model of London's importance in changing English society and economy’, Past & Present 37 (1967), 44–70CrossRefGoogle Scholar, and the riposte by Sharlin, A., ‘Natural decrease in early modern cities: a reconsideration’, Past & Present 79 (1978), 126–38CrossRefGoogle Scholar.

¹¹¹ Landers, J., ‘Seasonality of Mortality’, ch. 6 in Death and the Metropolis: Studies in the Demographic History of London, 1670–1830 (1993), 203–41CrossRefGoogle Scholar, at 239–41; the analysis of mortality data for the city of Bologna and three adjacent rural districts (plains, hills, mountains) in the early nineteenth century reveals much the same cycles of death in all of them; in terms of the factors affecting their demographic make-up, it seems that pre-modern cities were not as dramatically separated from their rural environments as modern ones: Bellettini, op. cit. (n. 93), fig. 22, facing p. 190.

¹¹² Landers, J., ‘Mortality and metropolis: the case of London 1670–1830’, Population Studies 41 (1987), 59–76CrossRefGoogle Scholar; Landers and Mouzas, op. cit. (n. 109); and Wrigley and Schofield, op. cit. (n. 62), give some of the problems with the English case where burial records and the London ‘Bills of Mortality’ have been used to reconstruct seasonal mortality.