



Both forms of the conjecture are false, as shown by the following counterexample.
Take sufficiently large integers and
, and let
be disjoint
-cliques, such that their union is a clique of size
. Now add
disjoint copies of
, labelled
, with no edges between them and such that
is complete to
if
and anticomplete otherwise.
This graph has vertices and maximum degree
. Every edge in
is in a maximal clique of size
. It is straightforward to show that for every
we can choose
and
sufficiently large that this graph is a counterexample.
The original discussion of the conjecture follows.
King [K] proved that any graph with
contains a stable set intersecting every maximum clique. This used the same approach as Rabern's earlier proof for graphs with
, but exploited a more general existence condition for independent systems of representatives.
One of the key lemmas is Hajnal's clique collection lemma, which states that for any collection of maximum cliques in a graph, the sum of the intersection and the union is at least . Unfortunately the same result does not hold when restricted to large maximal cliques. Thus the case of maximal cliques would require a different approach.
This conjecture is motivated by a local strengthening of Reed's omega, Delta, chi conjecture.
Bibliography
[K] Andrew D. King. Hitting all maximum cliques with a stable set using lopsided independent transversals, J. Graph Theory, n/a, doi: 10.1002/jgt.20532.
[R] Landon Rabern. On hitting all maximum cliques with an independent set, J. Graph Theory, 66 (2010), 32--37, doi: 10.1002/jgt.20487.
* indicates original appearance(s) of problem.