This case study demonstrates the application of an unsupervised topic modeling algorithm to 7,773 English-language articles published in the "Library Quarterly" from 1931 to 2015. The analysis of 85 years of the journal's output follows an exploratory data analysis framework to generate novel hypotheses about the history of LIS using topic modeling, a method for identifying clusters of co-occurring words within large collections of text. The paper closely examines two topics that suggest differences in gender representation in the journal to propose and support a new hypothesis regarding the historical inclusion of gendered objects of study in LIS literature.