If we don't consider theory like massive vector field, then the superficial degree of divergence is $$D= 4 - E_b -\frac{3}{2} E_f -\sum_{i}v_i \delta_i$$ where $E_b$ is the number of external boson line, $E_f$ is the number of external fermi line, $i$-th kind of vertice occurs $v_i$ times and the mass dimension of the coefficient of $i$-th kind of vertice is $\delta_i$.
I know that $D<0$ do not means the diagram is convergent, because there may be subdiagram with $D>0$.
My question is:
1 For a diagram with $D= 0$ does it mean that this diagram must be divergent? Does there exist some counterexample?
2 For a diagram with $D> 0$ does it mean that this diagram must be divergent? Does there exist some counterexample?