This paper examines a nonparametric test for Granger-causality for a vector covariance stationary linear process under, possibly, the presence of long-range dependence. We show that the test converges to a non-distribution free multivariate Gaussian process, say vec (B(µ)) indexed by µ ? [0,1]. Because, contrary to the scalar situation, it is not possible, except in very specific cases, to find a time transformation g(µ) such that vec (B(g(µ))) is a vector with independent Brownian motion components, it implies that inferences based on vec (B(µ)) will be difficult to implement. To circumvent this problem, we propose bootstrapping the test by two alternative, although similar, algorithms showing their validity and consistency.