When we speak of Servius' commentary on the works of Vergil, we understand that the name of Servius, which we use mainly for convenience, cloaks in apparent unity a work that is notable for its diversity and heterogeneity. This remark pertains not only to the existence of two Servian commentaries, the one written by Servius himself in the fifth century and the one compiled several centuries afterwards and eventually published by Pierre Daniel, but also to the diverse prior sources on which both these commentaries are based. It is well known that much of the material in these commentaries is tralatician. Except in a few specific cases, however, we cannot name either the proximate or the ultimate source of any given contribution, nor can we claim to understand fully the general principles that Servius followed in compiling his work. In this paper I will review some of those cases in which we can say with certainty or with reasonable probability how some specific passages in Servius took their current form, and will attempt to clarify what these instances can tell us about Servius' working methods in general. In order to keep this essay within manageable limits, I will confine my examination to passages in which the Servian commentaries show a strong affinity with the exegetical tradition of Homer.