“…= (U m (t,.)) xx on (0,f(t)), and m > 1, we can apply a well-known local result (see KALASHNIKOV, 1987) showing that if U(0, x 1 ) > 0 then U(t, x 1 ) > 0 for any t > 0. In particular, we get that U(t, x) > 0, for any x [ [ 0, f(0)) and for any t > 0.…”