Neuronal mechanisms underlying control of sound production in a cricket: Acheta domesticus.

Analyses of neuronal mechanisms underlying instinctive behaviour have long been required (Tinbergen, 1951) but have not been forthcoming. The reason seems to lie in part in a lack of agreement as to suitable experimental animals. There are now signs that certain large insects, which are available in many laboratories, and whose nervous systems are amenable to electrophysiological investigation, have been selected as firstchoice subjects for detailed investigations. As examples we may refer to the work of Wilson (1961, 1964) on locust flight mechanisms, Roeder, Tozian & Weiant (1961) on the praying mantis and Hoyle (1964) on general locomotory activity in locusts. The simplicity of the motor nervous system of insects renders possible an exact interpretation of electromyograms recorded from freely moving, intact animals, thereby permitting a full analysis of the motor output associated with behaviour, provided the latter can be elicited under laboratory conditions. A possible subject for analysis in the laboratory is the innate behaviour involved in singing in crickets. A good deal of descriptive work has been done by Alexander (1961) on the nature of the songs, their functional significance and probable evolution. What makes this subject particularly appealing, however, is that songs can be elicited by local stimulation of the brain (Huber, i960, 1962). This means that a form of complex behaviour can be evoked by means (electrical stimulation) which are fully controllable and so should be an aid in investigation of this complex problem. Some species of crickets give up to four structurally distinct songs, and more are recognized in some species if the context is also taken into account. It is desirable to know the ways in which these are produced, the sites in the nervous systems where the control information is stored, how afferent clues are used to call forth a particular song and so forth. A necessary preliminary for any of these aims is to establish techniques for examining the motor output to the muscles during natural sound production in the intact animal. The methods used by one of us (Hoyle, 1964) to study the control of locomotion in locusts are adaptable to the problem and have been used successfully in the present work. The species used principally was Acheta domesticus.