Analyses of historical or modern populations indicate a strong relationship between mortality level and standard of living, measured, among other factors, by degree of urbanization. The aim of this study was to assess mortality rates in children of up to 5 years of age in two populations living under different conditions in central modern Poland at the end of the 19th century: the rural parish of Kowal, under Russian partition, and Toruń, an industrial and urbanized centre under Prussian partition. Data on births and deaths were taken from birth certificate registries and from the Prussian statistics yearbooks for 1876–1894. Death rates of children aged 0–5 years were calculated, and also for annual age ranges. The urban population had lower birth rates (37.19‰), natural increase rates (8.0‰), population dynamics rates (1.26‰), which provide information about the relation between two components of a natural increase, i.e. births and deaths, and an over-mortality of boys in relation to girls. In the rural population these values were all higher: 53.67‰, 18.11‰ and 1.59‰ respectively. No impact was found of social stratification on child mortality in the wide age group of 0–5 years. However, for subsequent one-year age groups significant relationships between mortality level and size and industrialization level of the population centres were noted. The living conditions of infants in Toruń, although being in a better position as an area annexed by Prussia, were markedly worse than those of rural Kowal Parish. In the urban centre infant mortality was slightly over 269 for 1000 live born, and in Kowal Parish it was 163 for 1000 live born. The high infant mortality was balanced in Toruń by the higher mortality levels of children aged 2–5 years compared with Kowal Parish. Natural selection in the city had the greatest impact on infants, who did not have the protective influence of breast-feeding because women had to return to work shortly after giving birth. The lower infant mortality of mothers in the countryside due to longer breast-feeding led to larger family sizes. In 1871–1890 in the villages the number of children per women was about 7.42, whereas in Toruń it ranged from 4.4 to 5.2. The probability of death among children who survived the first year of life was higher in the countryside than the town. In the rural parish, perhaps because of cultural factors such as breast-feeding or working practices making full-time baby-sitting possible, children who did not reach the age of 1 year were not subjected to such intensive natural selection. Overall, differences in child mortality in the two centres in 19th central Poland resulted from ecological and cultural conditions, rather than from social and economical reasons (living under different partitions).