The concept of marriage squeeze expects a positive association between marriage formation and the availability of preferred mates. Previous research to test the hypothesis has had mixed results owing to inconsistent marriage measures, inconsistent age focuses and insufficient attention to migration. This study derives kinetics equations of marriage formation to link cohort age-specific mate availability to migration-adjusted marriage rate/incidence, a measure in contrast to nominal marriage rate. On testing the equations with Japanese census data for 1920–1940, it is found that, in female cohorts, mate availability impacts first marriage rate at the life-course stage from 15–19 to 20–24 years, but not at later stages. Among young females, the decline in mate availability accounted for about 21% of the decline in first marriage rate over the period 1920–1940, when there was a trend towards later but not less marriage in Japan. The study suggests that the flexibility of mate/spouse choice in females varies along the marriageable life course and is more manifest at older ages. At young ages, however, the marriage squeeze hypothesis could hold, presumably because young women are evolutionarily shaped to be choosier, perhaps postponing their marriages when preferred mates are in short supply.