Table 1. Experiments for developing a flytrap for collecting C. putoria.

Trap feature	Variables tested a	Rationale for experiment
Entrance hole - diameter	0.6 cm, 1.0 cm and 1.6 cm diameter entrance holes.	To determine the ideal size of hole for collecting and holding flies in a trap.
Entrance hole - diameter and number of holes	28×0.6 cm diameter entrance holes, 10×1.0 cm diameter entrance holes and 4×diameterentrance holes.	Here the total area of hole is the same for the small, medium and large holes. We hypothesized that if hole size was unimportant, all traps would collect a similar number of flies.
Entrance hole - position	Top, bottom and side, with all holes positioned17 cm above the ground	To determine whether fly catch size was dependent on whether the entrance holes were on the top, bottom or side of the trap
Entrance hole – shape test 1	Standard 1.6 cm diameter entrance holes, (1) without a cone fitted, (2) with a 1.0 cm deep paper cone with a 0.6 cm internal diameter at its tip, (3) with a 1.0 cm cone with a 1.0 cm internal diameter at its tip.	We hypothesized that small conical collars protruding from the entrance holes inside the trap lid would reduce the number of flies leaving a trap.
Entrance hole – shape test 2	Standard 1.6 cm diameter entrance holes, fittedwith and without a 0.6 cm diameter paper cone,1.0 cm deep. Traps were left overnight.	Although the 1.6 cm holes were effective at letting flies enter traps, we hypothesised that conical entrance holes would make it less likely for flies to leave the traps at night.
Trap - height	Bottom of trap positioned 0 cm, 25 cm and 120 cm above the ground.	To determine whether the height of a trap affected fly collections.
Trap - opacity	Transparent (no paper), semi-transparent (1 layer of 90 gm2 Natural Translucent Paper (Royal SovereignLtd., London, UK)) and opaque (3 layers of 90 gm2Natural Translucent Paper). The paper waspositioned inside the traps.	We hypothesized that flies remained in the trap since they were attracted to the light coming through the transparent walls. We considered that non-transparent flies would have fewer flies.
Trap - shape	Cubic 1L box (Whitefurze, Coventry UK) and acylindrical 1L box (Whitefurze, Coventry UK).	We hypothesized that flies were more likely to leave a trap which had internal corners than one which did not.
Trap - volume	0.6 L, 3.0 L and 6.0 L boxes	To determine whether the trap volume affected fly collections.
Trap – lid colour only	The plastic lids were replaced with either glosswhite, brown or blue mount board (Antique White,Chocolate Murano, Hussar Blue, Daler-Rowney,Bracknell UK) cut to fit. Entrance holes wereadded to the lid of each trap.	To determine whether the colour of the lid affected fly collections. White was the standard lid colour, brown is the colour of faeces and blue is a common colour used for attracting tsetse flies, a day-flying insect.
Trap – slab colour	Grey 1×1 m chipboard, black 1×1 m chipboard and blue 1×1 m chipboard. A 12 cm diameter pipe,10 cm in length, in the centre of each board wasinserted into the base of a standard trap	During the ‘Trap – lid colour only’ experiment we noticed flies were least attracted to black lids. We tested whether colours under the trap would deter flies from entering the trap.
Trap - gridded vs. non-gridded	1×1 m wire grid placed underneath the trap, 0.17×0.17 m wire grid positioned over the traplid, no grid. The wire grid consisted of 1 mm steelwire spaced 2.5 cm apart.	Based on the characteristics of a Scudder Grill. A grid is recognized as a means of counting flies in a given area because they (the flies) prefer to land on edges.

^aStandard trap design consisted of 10 entrance holes, each 1.6 cm in diameter, in a white lid of a 3 L polypropylene box, with transparent sides. 50 g of raw fish was placed in a 9 cm diameter white pot, covered with cotton netting.