Mobile ad hoc networks are networks composed of wireless devices to create a network with the ability for self-organization. These networks are designed as a new generation of computer networks to satisfy some specific requirements and with features different from wired networks. These networks have no fixed communication infrastructure and for communication with other nodes the intermediate nodes are used. Despite having many benefits due to wireless channel and dependency of each node to the intermediate node, these networks are faced with many security concerns. One of the concerns is the possibility of black hole attacks occurrence. The occurrence of black hole attacks has challenged the security issue in this kind of networks. This type of attack which is applied on mobile ad hoc network routing protocols, declares its black hole node as the shortest route to the destination node and therefore, other nodes in the network select this node as the intermediate node in sending their packets to various destinations. As a result, these nodes can delete received packets instead of sending them to destination. To avoid this problem in mobile ad hoc networks and based on AODV routing protocol, we introduce a new method to immediately identify the black hole attack and prevent it from occurring. To detect and defend against black hole attacks in mobile ad hoc networks, we use Firefly algorithms. Firefly algorithm is a biologically motivated algorithm for multi-faceted optimizing applications which is recently developed and used in many applications. To demonstrate the effectiveness of FireFlyAODV method, we used NS-2 Simulator and compared our work with AODV protocol under black hole attack. The results of simulation show that the proposed method has higher performance in terms of packet delivery ratio, throughput, the number of removed packets and end-to-end delay