I've read somewhere a proof of Gromov's nonsqueezing theorem using the fact that the action is a symplectic invariant (just like the Hamiltonian). As far as I remember, the action was expressed in terms of the radius of a sphere and also in terms of the radius of a cylinder, and the proof of the statement that the sphere doesn't fit by symplectic transormation into the cylinder was based on these expressions. But I 've read this poof only very superficially and just now wanted to read it in details, but now I don't find it, and don't remember where it was. Does somebody know this proof? Coul'd somebody tell me where can I find it? Thanks in advance.