At the beginning of the twentieth century, life expectancy at birth was much lower in Flanders, the northern part of Belgium, than in Wallonia, the southern part of the country. In the literature, this excess mortality is mainly attributed to high levels of infant mortality caused by bad feeding practices and low-quality drinking water. The regional variability of mortality risks at other ages during this period has received less attention. In this article, we reconstruct age-, sex- and disease-specific death rates for the 41 districts of Belgium around the year 1910. To show the mortality variations, we construct maps according to indirect standardised mortality rates that reflect the deviation from the national average. Our spatial analysis shows that there was a clear-cut Flemish-Walloon divide in general mortality only for infants and children under the age of 7. For older children, adolescents, and young and elderly adults, low and high mortality were observed in both regions. For disease-specific mortality, moreover, a geographical pattern was only visible for infants, who consistently had the highest death rates in Flanders. Hence, the spatial disparities in general and disease-specific mortality cannot be simplified according to a Flemish-Walloon divide. Furthermore, we noted large differences among districts belonging to the same province, and in the ranking of the districts by age. In other words, high mortality levels of infants, children, adolescents and adults did not per se appear in the same districts. From adolescent ages onwards, there were also large differences in the ranking of districts by sex- specific mortality. This strongly suggests the importance of sex-specific determinants of health and mortality at these ages.