Let be a matroid of rank , and for let be the number of closed sets of rank .

**Conjecture**is unimodal.

**Conjecture**is log-concave.

A sequence is *log-concave* if for all .

The first of these conjectures is due to Rota [R], the second is folklore as far as I (M. DeVos) know. The special case of proving the second conjecture for amounts to showing that and has been called the points-lines-planes conjecture. Seymour [S] proved this conjecture in the special case where every line contains at most four points, but it is still open in general.

## Bibliography

*[R] Rota, Gian-Carlo, Combinatorial theory, old and new. Actes du Congrès International des Mathématiciens (Nice, 1970), Tome 3, pp. 229--233. Gauthier-Villars, Paris, 1971. MathSciNet

[S] Seymour, P. D. On the points-lines-planes conjecture, J. Combin. Theory Ser. B 33 (1982), no. 1, 17--26. MathSciNet

* indicates original appearance(s) of problem.