cp's OEIS Frontend

This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.

A383998 Number of distinct truncated Graham sequences (as in Graham's Tree Reconstruction Conjecture) of length 4 on trees of order n.

Original entry on oeis.org

1, 1, 1, 2, 3, 6, 11, 20, 37, 68, 114, 188, 300, 462, 702, 1041
Offset: 1

Views

Author

Kaylee Weatherspoon, May 22 2025

Keywords

Comments

For small trees, the sequence |G|, |L(G)|, |L(L(G))|, |L(L(L(G)))| is sufficient to determine the tree, so this sequence has the same first few terms as A000055.

Examples

			For n=8, there are 23 trees but only 20 distinct truncated Graham sequences of length 4.
There are two pairs of trees on 8 vertices which have the same length-4 sequence [|G|, |L(G)|, |L(L(G))|, |L(L(L(G)))|], namely the sequence [8,7,7,9] which comes from both the (unlabeled versions of) {{1, 2}, {2, 3}, {3, 7}, {4, 5}, {5, 6}, {6, 7}, {7, 8}} and {{1, 2}, {2, 3}, {3, 8}, {4, 7}, {5, 6}, {6, 7}, {7, 8}}.
But for sequences of length 5 there are different sequences, namely
[8, 7, 7, 9, 18] and [8, 7, 7, 9, 17]: the sequence [8,7,9,17] comes from both {{1, 3}, {2, 3}, {3, 7}, {4, 6}, {5, 6}, {6, 7}, {7, 8}} and {{1, 2}, {2, 3}, {3, 8}, {4, 7}, {5, 7}, {6, 7}, {7, 8}}
So Graham's conjecture is confirmed for trees with 8 vertices, but requires using sequences of length up to 5.
		

Crossrefs

Cf. A000055.

Extensions

Corrected by Doron Zeilberger, Aug 12 2025.