It is known that a k-term DNF can have at most 2^k ? 1 prime implicants and this bound is sharp. We determine all k-term DNF having the maximal number of prime implicants. It is shown that a DNF is maximal if and only if it corresponds to a non-repeating decision tree with literals assigned to the leaves in a certain way. We also mention some related results and open problems.