The free sum is a basic geometric operation among convex polytopes. This note focuses on the relationship between the normalized volume of the free sum and that of the summands. In particular, we show that the normalized volume of the free sum of full dimensional polytopes is precisely the product of the normalized volumes of the summands.
IntroductionA (convex) polytope P ⊂ R n is the convex hull of finitely many points in R n . Equivalently, P is the bounded intersection of finitely many closed half-spaces in R n . We call a polytope lattice if its vertices are elements of Z n . Polytopes are fascinating combinatorial objects, and lattice polytopes are especially interesting due to their appearance in many contexts, such as combinatorics, algebraic statistics, and physics; see, for example, [2,26,27] Given polytopes P ⊂ R m and Q ⊂ R n , set P ⊕ Q := conv{(P, 0) ∪ (0, Q)} ⊂ R m+n .If P and Q each contain the origin, then we call P ⊕ Q the free sum of P and Q. It will be convenient to use the notation P ′ , Q ′ for (P, 0), (0, Q) ∈ R m+n , respectively, so that we simply write P ⊕ Q = conv{P ′ ∪ Q ′ }. In the present contribution, we