As many of you likely know, classical Platonic solids, the first historically well-explored polyhedra, come in pairs, called 'duals', simply by exchanging the faces and vertices. The cube is the dual of the octahedron, and the dodecahedron is the dual of the icosahedron.
But the dual of the tetrahedron, the simplest polyhedron, is also a tetrahedron.
And this leads me back to the tetrahedral models of the periodic system. In these models, the individual elements are represented by close-packed spheres. And pairs of LST periods of same length sum to give square numbers- 2+2=4; 8+8=16; 18+18=36; and 32+32=64, all squares of even integers. And pairs of THESE then sum to tetrahedral numbers (out of the Pascal Triangle): 4+16=20; 36+64=100.
Now, if the initial PT model is based on close packing of spheres, what would this imply in the dual tetrahedron? Spheres have their own equations, but if we're somehow exchanging the coordinates of the dual (faces for vertices), then what would the elements be represented by in the dual here? More spheres, or perhaps hyerboloids? And how would these be packed into the structure? In the tetrahedral models I and others have proposed, spheres are in nearest-neighbor contact (i.e. 'kissing'). But there is also rhombic structure present, and where one axis of the rhombus presents the kissing spheres, its perpendicular partner axis has the spheres separated by a gap traversing the center of the figure.
Would packing of hyperboloids focus on close contact as with kissing spheres, or more on longer-distance connectivity. I use the analogy here of the way the human cerebral cortex is structured and processes information- the typical left hemisphere deals more with nearest-neighbor connectivity (and in language more with grammatical issues), while the right deals more with long-distance connectivity and associations. This is pertinent to discussions on the way iconic vocabulary (the kind I study) is processed.
Anyway, I'm just beginning to explore these issues- any first impressions or suggestions would be welcome. With regard to the PT, would the dual tetrahedral representation of the LST structure show linkage between more far-flung memberships than the simple period and group ones that flat representations are able to, such as, for example, knight's move, etc.?
Jess Tauber