Metaphysical indeterminacy is a kind of indeterminacy that cannot be explained away just by referring to a lack of knowledge or to semantic indecision. In an influential paper, Elizabeth Barnes (2014) has defended the following conditional claim: if there is metaphysical indeterminacy, then it cannot be only at the derivative level of reality. The underlying intuition behind this, as she has it, is that «if you’ve got determinate components and combine them in determinate ways, there’s nowhere for indeterminacy to come from». In order to argue for this claim, Barnes relies on two principles, that I shall call Maximal Completeness and Determinate Link. According to the former principle, a complete description of a world w is a maximal bivalent assignment of truth values to every sentence at w. According to Determinate Link, the determination link between more and less fundamental levels of reality is such to preserve determinacy from one level to the other. The aim of this paper is to argue against Barnes’ conditional claim. My strategy is two-fold. First, I argue that the Determinate Link can be rejected, for in the presence of indeterminacy it is a substantive issue whether or not the relation that connects different levels of reality is determinacy preserving. Second, I provide concrete examples, coming from the philosophy of physics, of how we can have, contra Barnes’ conclusions, metaphysical indeterminacy in the derivative ontology, and yet no indeterminacy at the fundamental level.