| Configuration | Vertices | Edges | Faces | Cells | Result | Remarks | |||
| {3, 4, 3} | 24 | 96 | {}@1 | 96 | {3} | 24 | {3, 4} | {3, 4, 3} | |
| 3 * 24 | {4} | 3 * 8 | {4, 3} | 3 * {4, 3, 3} | |||||
| 16 | {6} | disconnected | |||||||
| 3 * 24 | {}@2 | 3 * 32 | {3} | 3 * 16 | {3, 3} | 3 * {3, 3, 4} | total split | ||
| 3 * 4 | {3, 4} | disconnected | |||||||
| 3 * 6 | {4} | disconnected | |||||||
| 96 | {}@3 | 16 * 2 16 |
{3} {6/2} |
disconnected | |||||
This kind of polytope occurs only in 4D. It is reasonably simple. It is also called the icositetrachoron or the 24-cell. The coordinates of the vertices can be found in Coxeter, I have scaled them so that the basic edge length is 2. This is the only polytope except for the simplices, that is self-reciprocal.
The polytope has a few interesting features. In section 1 a compound can be found consisting of three {4, 3, 3}'s. They share only edges, not higher dimensional objects. In section 2 the polytope splits totally in three {3, 3, 4}'s. These {3, 3, 4}'s are the reciprocals of the {4, 3, 3}'s in section 1, the global numbering used in this section is derived from the numbering in section 1. In section 1 we can also find hexagons (but they are disconnected). The {3}'s in section 3 form pairwise {6/2}'s, induced by the {6}'s in section 1. These faces (or face pairs) can be grouped in 4 groups of 4 faces, such that the set of faces in a group cover the vertices exactly once. This grouping can be done in 2 different ways.
This polytope can tile 4D space, the symbol becomes {3, 4, 3, 3}.