This paper studies external memory bidirectional search. That is, how bidirectional search algorithms can run using external memory such as hard drives or solid state drives. While external memory algorithms have been broadly studied in unidirectional search, they have not been studied in the context of bidirectional search. We show that the primary bottleneck in bidirectional search is the question of solution detection – knowing when the two search frontiers have met. We propose a method of delayed solution detection that makes external bidirectional search more efficient. Experimental results show the effectiveness of the approach.