Females are more likely to attend Canadian universities and their participation rate has been increasing faster than that of males, generating a 15 percentage point gap by 2005. We investigate the determinants of attending university and explore the reasons for the increasing gender gap. As in the US literature, we find that conventional decompositions, averaging over the whole sample, attribute most of the gender gap to differences in variables, notably the University Premium. The average sample approach in these decompositions abstracts from the evolution over time of the participation rates by gender and from important detail at specific points in time, features that we try to overcome. We find that the rising "level" of the university participation rate for both women and men can be explained by changing societal norms, the dynamic influence of parental education, the University Premium, tuition fees and real income. The "2005" gap between the female and male participation rates can be accounted mostly (9 percentage points) by the gender differences in the University Premium and, to a lesser extent (6 percentage points), by differences in the coefficients in the female and male participation equations. (Contains 2 figures and 2 tables.)