This work analyzes the spatial heterogeneity of consanguinity in the Basque province of Guipúzcoa (Spain), using data provided by Catholic dispensations (1862-1980). Secular trends in consanguinity rates (%M(C)) and mean inbreeding coefficient (F) were similar in the seven Guipúzcoan regions considered, with peaks between 1881-1920 and subsequently a gradual decline. Substantial differences in consanguinity characteristics emerged when the regions were classified according to level of urbanization. Principal component analysis (accounting for more than 85% of the total variance in consanguinity variables) clearly discriminated between urbanized and less urbanized regions. The latter stand out for their high consanguinity rates (3.57-6.73%), mean inbreeding coefficient (0.00112-0.00240), and M22/M33 ratio (M22, first cousins; M33, second cousins), which ranged between 0.89-1.48. Moreover, in less urbanized regions, marital consanguinity was eminently local, and mainly conditioned by regional endogamy (71.4-85.0%). By contrast, urban subpopulations showed the lowest consanguinity rates (1.60-1.96%) and mean inbreeding coefficient (around 0.0007). In these regions, the M22/M33 ratio also exhibited high values (1.07-1.56), but this time at the expense of the contribution of the immigrant group. Discussion of the factors that could have modeled this spatial variation in consanguinity centers on: 1) demographic aspects related to the chronology and intensity of industrialization, 2) the geography of the territory and the geography of peopling, and 3) linguistic differences, expressed in the uneven distribution of Basque-speakers among the different territories considered.