MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

My question is about (abstract) simplicial complices.
In particular, how many are they if I consider $n$ unlabelled vertices?

For example, if $n=4$, the two complices $$ \{\varnothing, \{1\}, \{2\}, \{3\}, \{4\}, \{1, 2\}, \{3, 4\}\} $$ and $$ \{\varnothing, \{1\}, \{2\}, \{3\}, \{4\}, \{2, 3\}, \{1, 4\}\} $$ are the same, but not $$ \{\varnothing, \{1\}, \{2\}, \{3\}, \{4\}, \{1, 2\}, \{1, 3\}\} $$ (since the last two sides of this one intersect in one vertex).

If $n=3$, there are 5 of them (while the Dedekind number for 3 is 20).
They are:
- dim=2 $$\{\varnothing, \{1\}, \{2\}, \{3\}, \{1, 2\}, \{1, 3\}, \{2, 3\}, \{1, 2, 3\}\}$$ - dim=1 $$\{\varnothing, \{1\}, \{2\}, \{3\}, \{1, 2\}, \{1, 3\}, \{2, 3\}\}$$ $$\{\varnothing, \{1\}, \{2\}, \{3\}, \{1, 2\}, \{1, 3\}\}$$ $$\{\varnothing, \{1\}, \{2\}, \{3\}, \{1, 2\}\}$$ - dim 0 $$\{\varnothing, \{1\}, \{2\}, \{3\}\}$$

Since this last observation, I think that the answer is not the Dedekind number, but please prove me wrong if you think it is.

Thank you in advance, Davide

PS: I am not sure whether or not this question is related to this other one. If so, please can you explain why?
PPS: I posted this question also on Math.SE, but no one answered.

share|cite|improve this question
There is most likely an exercise in one of R. Stanley books on Enumerative Combinatorics that solves this. – Per Alexandersson Aug 14 '14 at 12:52
up vote 4 down vote accepted

This is . I checked the numbers there.

The correspondence is because the maximal elements form an antichain cover.

share|cite|improve this answer
This is also A003182(n)-A003182(n-1) ( which has many more references – Philippe Nadeau Aug 14 '14 at 13:20

Since Brendan has identified the sequence and provided values for small $n$, let me point out that the asymptotic behavior of your sequence $s(n)$ will be $$s(n) \sim \frac{1}{n!}d(n)$$ where $d(n)$ is the $n$-th Dedekind number.

To see why, note that a simplicial automorphism must at least induce a graph automorphism of the 1-skeleton, but the number of graph automorphisms of a graph on $n$ vertices are almost surely zero unless the number of edges is within $\log(n)$ of the extreme values $0$ and $\binom{n}{2}$. See Chapter 9 of Bollobas' book on Random graphs for a proof.

To answer your post-script -- this is very different from my question on homotopy classes of simplicial complexes in the sense that you are seeking isomorphism classes whereas I was only seeking homotopy classes. In particular, I would expect (at least for small $n$) for your sequence to be much larger than mine since there are many homotopy-equivalent but non-isomorphic simplicial complexes on $n$ vertices.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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