11 Sep
2009
11 Sep
'09
1:34 p.m.
Quoting Dan Asimov <dasimov@earthlink.net>: <snip>
But (as is well-known) a perfectly regular dodecahedron can be inscribed in a perfect cube with 3 pairs of opposite edge passing through the face centers of the 3 pairs of opposite faces, resp., of the cube.
Is there something similar connecting the regular icosahedron and the regular octahedron? It will have to be a little different, since 6|12 but 8~|20. Rich