The distinction between macro- and microinventions is at the core of recent debates on the Industrial Revolution. Yet, the empirical testing of this notion has remained elusive. We address this issue by introducing a new quality indicator for all patents granted in England in the period 1700-1850. Our findings indicate that macroinventions did not exhibit any specific time-clustering, while microinventions were correlated with the economic cycle. In addition, we also find that macroinventions were characterized by a labor-saving bias and were mostly introduced by professional engineers. These results suggest that Allen's and Mokyr's views of macroinventions, rather than conflicting, should be regarded as complementary.