1975] Robinson and Pratt — Phenology of Hexacentrus 
321 
to Wau 
Wau; MHR left in September). We have graphed the monthly 
averages for each transect and the average for the combined monthly 
totals and these are shown in Figure 5. 
Discussion 
The sampling system that we used was not only simple but arbi- 
trary. We did not attempt to determine whether the results were 
influenced by the weather conditions obtaining during the census 
period. We would guess that we were sampling ,a fairly high pro- 
portion of the males within earshot of the transects but that this 
proportion did not remain the same, necessarily, from census to 
census. We got the impression that fewer insects were singing on 
cold clear moonlit nights. Dumortier (1963; 638-9) lists a number 
of climatic variables that may affect insect song (either as exciters 
or inhibitors). For these reasons we would not place too much 
reliance on the magnitude of the variations in the census results. 
The direction of the variations may reflect some general trends, 
since the lowest counts occur towards the middle of the Southern 
Hemisphere winter when rainfall and temperatures at Wau are low 
(see Robinson, Lubin & Robinson, 1974: 132-3). 
