“…-In Section 3 we present a generalisation, based on a theorem of Barrow [2], for optimal kernel quadrature rules [21, Chapter 5] that have both their points and weights selected so as to minimise the worst-case error. The result, Theorem 3.4, states that such rules, if unique, converge to the N -point Gaussian quadrature rule for the functional L, which is the unique quadrature rule Q(XG, wG) such that…”