1
$\begingroup$

Suppose we are given $(\mathbb{R}^8,\Phi)$, where $\Phi$ is the self-dual 4-form that defines $Spin(7)\subset SO(8)$ (Cayley calibration, see Notes on the Octonians, page 23). Now some 4-subspaces $V$ of $\mathbb{R}^8$ have the property that $\Phi|_V=\pm vol(V)$ (in general it is $\leqslant$ in absolute value).

QUESTION 1: Which is the dimension of the set of calibrated subspaces as a submanifold of the Grassmannian $G(4,8)$?

Now suppose we have a calibrated 8-manifold $(X,\Phi)$.

QUESTION 2: What can be said about the Cayley submanifold? Is there a Cayley submanifold containing a given Cayley subspace at some point? May be quantified 'how many' Cayley submanifolds are there?

Any idea is welcome.

$\endgroup$

1 Answer 1

3
$\begingroup$

All your questions (and much more) are answered in the 1982 paper Calibrated Geometries by Harvey and Lawson. See Acta Mathematica July 1982, Volume 148, Issue 1, pp. 47-157.

Just so you'll know: Harvey and Lawson prove the following

The dimension of the set $Cay$ of Cayley subspaces in $\mathrm{Gr}(4,8)$ is $12$, in fact, it is a smooth, connected manifold diffeomorphic to $\mathrm{SO}(7)/\left(\mathrm{SO}(3){\times} \mathrm{SO}(4)\right)$. This answers Question 1. Moreover, every $3$-plane in $\mathbb{R}^8$ lies in a unique Cayley $4$-plane.

Every real-anaytic, embedded $3$-manifold $N^3\subset\mathbb{R}^8$ lies in an embedded Cayley $4$-fold $P^4\subset\mathbb{R}^8$. Moreover, any two such Cayley extensions $P_1$ and $P_2$ are equal in some open neighborhood of $N^3$. This answers Question 2 for flat space. However, Harvey and Lawson's proof works with essentially no changes for the general case of $(X^8,\Phi)$.

$\endgroup$
2
  • $\begingroup$ Thank you very much! Although the whole paper is worth reading, do you remember (more or less) which chapter or section deals with the questions before? $\endgroup$
    – Jjm
    May 6, 2015 at 12:14
  • 1
    $\begingroup$ The parts you want are in Chapter IV (The Exceptional Geometries), but I don't remember the exact subsections. They are clearly labeled, though, so you should have no problem finding what you want. $\endgroup$ May 6, 2015 at 12:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.