SOUND RECEPTION 97 



takes the form of a series of discharges of a large spike at a steady 

 frequency of 5-20 per second. Haskell and Belton (1 956) believed these 

 to be motor discharges, but in Rocder's and Treat's preparation they 

 were clearly afferent. In addition, two species of small spikes discharge 

 randomly, in all probability, as a result of low-level random noise in the 

 laboratory. These two spikes represent the acoustic units of the 

 tympanic organ; the large spike is a non-acoustic unit whose function 

 is not clearly understood (Fig. 65). 



Both acoustic units respond to sound by a high-frequency discharge 

 which shows no change with stimulus frequency. The two differ only 

 in their threshold. For both, the greatest sensitivity lies between 

 15,000 and 60,000 c/s. In response to a click, with many transients, or 

 a short pulse (0-02-1 msec.) of pure sound at threshold, the more 

 sensitive fibre responds after a latency of 3-5 msec, with one spike 

 (Fig. 66). When the stimulus intensity is increased, the latency de- 

 creases and there is an after-discharge of two-three spikes at a fre- 

 quency of 700 per second. A further increase in stimulus intensity 

 brings in the second acoustic unit. The variety of overlap observed 

 indicates that there is no regular alternation of activity or coupling of 

 the units. Although one cannot state categorically that only two units 

 are firing, this appears to be the case. It fits well with the histological 

 facts. 



Contrary to the situation found in orthopteran tympana, these 

 organs adapt rapidly. From an initial frequency of about 1,000 spikes 

 per second the discharge drops to 50 per cent of this value in the first 

 0- 1 second and to 25 per cent by the end of the second. 



Various operations on the tympanic organ prove that the effects 

 observed electrophysiologically are truly indicative of activity in the 

 two chordotonal sensilla. Touching the sensilla or disengaging them 

 from the tympanic membrane terminates their activity. On the other 

 hand, tearing the membrane or opening the air sac behind it fails to 

 alter the response significantly. Occluding the external tympanic recess 

 narrows the band of frequency reception to its middle range and raises 

 the threshold. 



These experiments are in agreement with the idea that this organ is a 

 displacement-sensitive receiver. They show that the tympanum must 

 be free to vibrate but that uniformly distributed elastic tension is not 

 necessary. The direction of the displacement, in so far as the sensillum 

 is concerned, is apparently not critical, since in many noctuids the axis 

 of the sensillum is not normal to the plane of the tympanic membrane ; 

 instead, it forms an acute angle with it. 



