Published online by Cambridge University Press: 10 July 2009
Using the inverse square law, estimates can be made of the distances at which illumination from a light-trap is equal to that from background sources. From these distances an index of trap radius can be constructed which can be considered as a measure of trap potential. Between new moon and full moon trap radii vary, depending upon the times of the night at which the trap may be operating, in ratios from about 10:1 to 15:1. A comparable index of light-trap catches can be calculated which allows catches to be examined in relation to changing radius of the trap. Analysis of a series of catches in Uganda and Ghana shows that many species are more abundant than expected in periods of moonlight, particularly at and near full moon, the biggest difference between new moon and full moon being about 10:1 for Marasmia trapezalis (Gn.), whereas the Isoptera, Bostrychidae and Spodoptera triturata (Wlk.) are 3–4 times more frequent at full moon. The ratios between catch at new moon and catch at full moon suggest that the primary determinant of catch is the frequency with which insects cross the boundary of a region of influence whose size is determined by a radius of equal energy. Deductions about the pattern of insect activity through a lunation, and nightly, and the general agreement between curves describing the change in radius of the trap and those of trap catches suggest that changes in catch over a lunation can be explained by changes in the effectiveness of the trap. When corrections are made to allow for such changes, all taxa show some increase in numbers in moonlit periods and in many taxa this increase is substantial. Correction of catches should take account of flight periodicity and this periodicity should, if possible, be confirmed by methods independent of light-traps.