Consulting the commentaries of Aquinas, Barnes, and McKirahan, this paper investigates Aristotle's claim at Posterior Analytics I, 8 (75b22) that scientific knowledge of perishable things is eternal because the premises of the demonstration are universal. Most commentators believe that Aristotle is claiming some sort of eternity for the subjects of scientific conclusions. Thus, conclusions are eternally true because these conclusions predicate some property or affection of a subject that, in some sense, always exists. However, these commentators fail to realize that because APo I, 8 is part of his argument against Platonic dialectic, what Aristotle means by eternal conclusions is not that their subjects always exist. Rather, to say that a conclusion is eternal means that the connection between subject and predicate that a demonstration has proved to hold, does so hold whenever the premisses are supposed and understood, without respect to time, or the actual present existence of particular instances of the subject. The connection holds eternally because it is a per se connection. Yet in order to be seen as per se, it indeed has to have been drawn from universal premisses. But these premisses are "commensurately universal" and not merely universal in the sense of dici de omni.