“…which is a special case of Lemma 2.7. Since V α,β,γ k,n are well-defined if γ = −1, letting γ → −1 preserves the orthogonality, which is how the orthogonality in f, g α,β,−1 was established [6], but the method does not give the decomposition of the orthogonal space. Indeed, the second and the third terms in the right hand side of [f, g] α,β,γ make sense for γ = −1 as well, whereas for the first term we parametrize the integral over T 2 as…”