The Gromov width of complex Grassmannians

From MaRDI portal
Publication:2571371



Abstract: We show that the Gromov width of the Grassmannian of complex k-planes in C^n is equal to one when the symplectic form is normalized so that it generates the integral cohomology in degree 2. We deduce the lower bound from more general results. For example, if a compact manifold N with an integral symplectic form omega admits a Hamiltonian circle action with a fixed point p such that all the isotropy weights at p are equal to one, then the Gromov width of (N,omega) is at least one. We use holomorphic techniques to prove the upper bound.


In this paper the authors have proved that when the symplectic form is normalized in such a way that it generates the integral cohomology (in second degree), the Gromov width of the Grassmannian of the complex \(k\)-planes in \({\mathbb C}^n\) is equal to one. The Gromov width of the symplectic manifold \((M,\omega)\) is the supremum of the set of the real numbers \(a\) such that the ball \[ B(a) := \biggl\{z\in {\mathbb C}^n ; \pi\sum_{i=1}^N| z_i| ^2 < a \biggr\} \] can be symplectically embedded in \((M,\omega)\). The main technical result in the paper, presented as Proposition 2.8, gives a criterion for the existence of symplectic embedding of open subsets of \({\mathbb C}^n\) into a symplectic manifold equipped with a Hamiltonian torus action. For the above purposes the authors make use of both equivariant and holomorphic techniques. The earlier result about the Gromov width by \textit{G. Lu} (arXiv: math.SG/0103195) is also discussed and it is stated that the results of the present paper are more general as they give lower bounds for Gromov width and work for a larger class of manifolds beyond that one of the balls.











This page was built for publication: The Gromov width of complex Grassmannians

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2571371)