Research consistently documents that girls outperform boys in literacy achievement, yet, when considering the starting point--early literacy--we do not have consensus on the origin, meaningful nature, or persistence of such differences. In this two-part study analyzing 5,816 Norwegian students (48.1% girls, average age of 6.1 years), we first considered the presence and meaningful nature of gender differences at school entry. Presence of differences, advantaging girls, was found in letter-recognition, phonemic awareness, word reading accuracy, and spelling. We found no differences in vocabulary performance. However, wide score distributions suggest that intragroup differences are more meaningful than intergroup differences. In our second analysis, using a longitudinal design with three time points (school entry, end of Grade 1, end of Grade 2), we calculated a general latent literacy score based on an IRT model. Multilevel analysis tested the persistence of gender differences after formal instruction. Growth curve analysis of the latent literacy score verified an advantage for girls at school entry and that boys' scores grew at a significantly higher rate than girls across Grades 1 and 2. Patterns of effect sizes complement the statistical analysis. In summary, findings indicate that girls' small initial advantage in literacy skill at school entry persists into Grade 2, but it is substantially attenuated with formal instruction. Implications include that gender differences are secondary to other sources of heterogeneity in early literacy and that gender differences in early literacy at school entry do not directly lead to boys' long-term underachievement in literacy. Future directions should examine if self-selection of activities in early childhood education may contribute to early literacy differences.