This fallacy is similar to either the fallacy of Composition or Division but doesn't seem to be quite the same. Sometimes "things" happen with large assemblages (even of similar elements) that cannot happen with small ones. A trivial example is in some of the WWII board game simulations. The tactical use of (for example) an armored brigade vs an infantry brigade is not really similar to their stragical use. An assembly of 6000 brigades facing 6000 is a different problem from 6 vs 6.
Another example would be to Arrow's Paradox. THe paradox of the non-transitive dice is interesting too.
Likewise, non-transitive relationships may occur in groups. A & B may interbreed, B & C may interbreed, and A cannot interbreed with C.
Perhaps this should be the Failure of Scaling Paradox.
Perhaps StJacques is needed more than ever now, he has a very clear definition of science (we'll promise not to press him about his quantizing habits).