1969] 
Willey and Willey — Social Displays 
297 
( ?) (buzz), alarm signals, and perhaps a tenth category, the com- 
fort motions such as the wing buzz ? wing flick, tibiotegminal click, 
mandible snap, etc. which could have subtle communication signifi- 
cance which does not involve orientation. 
At this point we should discuss the relationship of the spectro- 
gram display relative to the probable mechanisms of sound production. 
According to Watkins (1967), analysis of sounds by the Kay audio- 
spectrographs must take into account the fact that pure tones (sine 
waves) modulated with on-off pulses whose repetition rate is more 
rapid than the analysing filter can discriminate will develop a definite 
harmonic structure. The over and under tone intervals are predict- 
able from the pulse tone and pulse rate; Fourier analysis can predict 
the sound energy at each harmonic. To a degree, the original 
characteristics of the sound being analysed can be deducted from the 
harmonic structure. Likewise, if the basic tone is a spike (a very 
brief pulse of energy) , rapid spike repetition rates fuse into harmonic 
intervals equal to the repetition rate added to the preceding harmonic. 
The greatest energy (darkest band) will be exhibited at the funda- 
mental frequency of the repetition rate, and will be the lowest band 
in the trace. The femorotegminal sounds illustrated in this paper 
approximate the appearance of a pulsed spike repetition quite closely 
and this is undoubtedly related to the impact velocity of the femoral 
ridge on each peg of the tegminal file. On the other hand, we have 
no ready hypothesis for the structural basis of the flight crepitations 
which sometimes resemble spikes in themselves and at other portions 
of the spectrogram resemble the aforementioned pulse modulation 
which is beyond the resolving capacity of the analysing filter. Study 
of this problem, using oscillography and high speed cinematography 
is in progress. 
We shall discuss the several signals of A. conpsersa under their 
presumed functional categories as listed by R. Alexander (1967) 
and Otte (1968). 
Disturbance and alarm. 
We have found the flutter-squeal (Fig. 6) commonly enough to 
consider it a basic stress pattern. Its neurological basis can be 
guessed as the outlet for an overload which brings together under 
stress several independent circuits, e.g., the rapid femoral flutter 
mechanism of the male-male interaction and the increased medial 
tension on the femur as it passes over the stridulatory pegs of the 
tegmen. Similar stridulations are produced by many acridids during 
capture. 
