Efficacy of Electrocuting Devices to Catch Tsetse Flies (Glossinidae) and Other Diptera

Background: The behaviour of insect vectors has an important bearing on the epidemiology of the diseases they transmit, and on the opportunities for vector control. Two sorts of electrocuting device have been particularly useful for studying the behaviour of tsetse flies (Glossina spp), the vectors of the trypanosomes that cause sleeping sickness in humans and nagana in livestock. Such devices consist of grids on netting (E-net) to catch tsetse in flight, or on cloth (E-cloth) to catch alighting flies. Catches are most meaningful when the devices catch as many as possible of the flies potentially available to them, and when the proportion caught is known. There have been conflicting indications for the catching efficiency, depending on whether the assessments were made by the naked eye or assisted by video recordings. Methodology/Principal Findings: Using grids of 0.5m2 in Zimbabwe, we developed catch methods of studying the efficiency of E-nets and E-cloth for tsetse, using improved transformers to supply the grids with electrical pulses of ~40kV. At energies per pulse of 35–215mJ, the efficiency was enhanced by reducing the pulse interval from 3200 to 1ms. Efficiency was low at 35mJ per pulse, but there seemed no benefit of increasing the energy beyond 70mJ. Catches at E-nets declined when the fine netting normally used became either coarser or much finer, and increased when the grid frame was moved from 2.5cm to 27.5cm from the grid. Data for muscoids and tabanids were roughly comparable to those for tsetse. Conclusion/Significance: The catch method of studying efficiency is useful for supplementing and extending video methods. Specifications are suggested for E-nets and E-cloth that are ~95% efficient and suitable for estimating the absolute numbers of available flies. Grids that are less efficient, but more economical, are recommended for studies of relative numbers available to various baits. Author Summary: With a view to refining studies of tsetse fly behaviour, we used a catch method, in the field in Zimbabwe, to assess and improve the efficiency of 0.5m2 electrocuting grids commonly used to sample tsetse that collide with netting while in flight or which alight on cloth. We found that the efficiency of both sorts of grid was improved if the ~40kV electricity supply pulsed at 1ms intervals, as opposed to the 5–15ms commonly used in the past. Efficiency was also improved if the energy per pulse was at least 70mJ. For the netting grid, which is ideally required to be invisible to the flies, the visibility seemed reduced if the netting were much finer than that used previously, and if the distance between the grid and its supporting frame were increased from 2.5cm to 27.5cm. The indications for muscoids and tabanids were much the same as for tsetse. Using these field results and a simple model of grid performance, we suggest the specification of grids suitable for various sampling purposes.