The description of the edge operation
\path ... edge[<options>] <nodes> (<coordinate>) ...
opens with the following explanation (p. 246, chapter 17 'Nodes and Edges' in the TikZ part of the manual):
The effect of the
edgeoperation is that after the main path the following path is added to the picture:
\path[every edge, 〈options〉] (\tikztostart)〈path〉;
to path. Note that, unlike the path added by the
(\tikztostart)is added before the
〈path〉(which is unnecessary for the
tooperation, since this coordinate is already part of the main path).
\tikztostartis the last coordinate on the path just before the
edgeoperation, just as for the
tooperations. However, there is one exception to this rule: If the
edgeoperation is directly preceded by a
nodeoperation, then this just-declared node is the start coordinate (and not, as would normally be the case, the coordinate where this just-declared node is placed – a small, but subtle difference). In this regard,
edgediffers from both
I don't understand the "small, but subtle difference" mentioned in the last paragraph. How can a node be the coordinate of anything? A node is a node, and a coordinate is a coordinate, except when a node has the
coordinate shape, but I don't think this is what is meant here. And if we interpret this occurrence of the word "node" as "the coordinate where the node is anchored", then what is the difference between this coordinate and "the coordinate where this just-declared node is placed"?