We report some recent developments from an ongoing project in which a chess endgame domain is providing benchmark experimental tests for the study of concept learning. The King and Rook against King (KRK) endgame is simple enough in chess terms but provides concept learning tasks which can be demanding, as evidenced in previous studies by a number of authors. For learning systems these tasks have highlighted problems of representation, such as the ability to express the structural relationships to be found in learning examples, and other issues like correctness, compression and comprehensibility. Our current focus is on Inductive Logic Programming methods which are based on previously developed systems for the generalisation and specialisation of normal logic programs. In the current work we are principally concerned with improving these methods to be able to handle more examples during learning. The main contribution of this work consists of a new method of incremental learning, together with new results on a set of concept learning problems taken from the KRK endgame domain.