The 3rd and 4th figure in Example 9 (First-order logic using positive and negative surfaces) are a bit surprising at first, because it does not appear clearly that the edges are part of the inner surface.
(of course, it would not make sense for them to be outside the inner surface, as they involve a node from that surface, but it is not obvious visually)
I would advise that the predicate name be enclosed in the surface containing the edge.
The 3rd and 4th figure in Example 9 (First-order logic using positive and negative surfaces) are a bit surprising at first, because it does not appear clearly that the edges are part of the inner surface.
(of course, it would not make sense for them to be outside the inner surface, as they involve a node from that surface, but it is not obvious visually)
I would advise that the predicate name be enclosed in the surface containing the edge.