We prove that if a non-singular planar map Λ ∈ C 2 (R 2 , R 2 ) has a convex component, then it is injective. We do not assume strict convexity.
Introduction.Let Ω be an open connected subset of R n . We say that Λ : Ω → R n is locally injective (invertible) at X ∈ Ω if there exist neighbourhoods U X ⊂ Ω of X and V Λ(X) of Λ(X) such that the restriction Λ :we denote by J(X) the Jacobian matrix of Λ at X. By the inverse function theorem, if J(X) is non-singular then Λ is locally injective at X. It is wellknown that locally injective maps need not be globally injective, even if J(X) is non-singular for all X ∈ Ω, as in the case of the exponential map Λ(x, y) = (e x cos y, e x sin y). Injectivity (invertibility) of locally injective (invertible) maps under suitable additional assumptions has been studied for a long time. In [14] it was conjectured that every polynomial map Λ : C n → C n with constant non-zero Jacobian determinant is globally invertible, with polynomial inverse. This problem, known as the Jacobian Conjecture, was widely studied and inserted in a list of relevant problems in [20]. The Jacobian Conjecture was studied in several settings, even replacing C with other fields, but still remains unsolved for n ≥ 2 [1,4,8,24]. In [17] it was proved that asking for the determinant of J(X) not to vanish is not sufficient to