Abstract: An energy efficient query processing algorithm is proposed to process a new kind of query named node number constraint query for wireless sensor networks. The algorithm is mainly composed of two parts: query dissemination algorithm and result collection algorithm. The function of the query dissemination algorithm is to select an energy efficient way to disseminate the user's query into sensor networks. First a relevant node selection algorithm is proposed to select partial number of nodes from the network and a Steiner tree based algorithm is designed to disseminate queries to these relevant nodes. Then two energy cost models are built for the Steiner tree based query dissemination algorithm and an energy efficient flooding based query dissemination algorithm respectively. By comparing two energy consumption models, a suitable dissemination algorithm can be selected to send query to sensor networks. The energy efficient result collection algorithm presents two kinds of result collection strategy, which are direct and indirect collection strategy, and then the collection algorithm gives the condition, in which each strategy is suitable to collect results. By combining all the above mentioned algorithms, an energy efficient query processing algorithm for node number constraint query is proposed and the simulation results show that the algorithm proposed in this paper can not only guarantee the precision of the result but also save much more energy than other algorithms.