EVASION OF BATS BY MOTHS — ROEDER AND TREAT 459 

 THE DETECTION OF BATS 



These experiments with artificial sounds suggest how the moth 

 ear might be expected to respond to a bat cry. A few laboratory ob- 

 servations were made with captured bats. In one of these experiments, 

 in collaboration with Dr. Fred Webster, the cries of a flying bat were 

 picked up simultaneously by a moth ear and a microphone, and 

 recorded on high-speed magnetic tape (pi. 3, fig. 2). Interesting 

 though they were, these experiments served mainly to show that the 

 full potentialities of the moth ear as a bat detector could not be 

 realized within the confines of a laboratory, and efforts were made to 

 transport the necessary equipment to a spot where bats were flying 

 and feeding under natural conditions. 



Finally, about 300 pounds of equipment was uprooted from the 

 laboratory and reassembled at dusk of a July evening on a quiet hill- 

 side in the Berkshires of western INIassachusetts, Moths attracted to a 

 light provided experim.ental material. The insect subject was pinned 

 on cork so that one of its ears had an unrestricted sound field, and with 

 the help of a microscope its tympanic nerve was exposed and placed 

 on electrodes. After amplification, the action potentials were dis- 

 played on an oscilloscope. They were made audible as a series of 

 clicks by means of headphones connected to the amplifier and were 

 stored on magnetic tape for later study. 



It was dark before all was ready, but bats immediately revealed 

 their presence to the moth ear by short trains of nerve impulses that 

 recurred about 10 times a second (pi. 4, A). The approach of a 

 cruising bat from maximum range was coded as a progressive increase 

 in the number and frequency of impulses in each train, first from one 

 and then from both A fibers. It was not long before we learned to 

 read something of the movements of the bats from these neural signals. 

 Long trains, sometimes with two frequency peaks, suggested the 

 chirps of nearby bats that echoed from the wall of a neighboring 

 house (pi. 4, B). An increase in the repetition rate of the trains 

 coupled with a decrease in the number of impulses in each train sig- 

 nified a "buzz" as the bat attacked some flying insect in the darkness 



All this was inaudible and invisible to our unaided senses. With a 

 powerful floodlight near the nerve preparation we were able to see bats 

 flying within a radius of 20 feet, and some attacks on flying insects 

 could then be both seen and also "heard" through the "buzz" as coded 

 by the moth's tympanic nerve. However, most of the sounds detected 

 by the moth ear were made by bats maneuvering well out of range of 

 the light. A rough measure of the sensitivity of the moth ear to bat 

 chirps was obtained at dusk on another occasion when the bats could 

 stiU be seen. The A cells first detected an approaching bat flying at 



