446 



The Matntre Haiji (uul I he House- Fly 



Ireiich G, about a foot ileep and KS inches wide, was first made and 

 the outer sides Hned with a suitable structure consisting of boards 

 attached at the corners to stout posts. The trench was then fiMed up 

 to the original ground level with a mixture of loose earth and old nuinure 

 straw forming a light mass suitable for the pupation of the larvae. 

 The extreme hardness of the ground at Harpenden together with the 

 outer lining of wood made it improbable that the larvae would wander 

 beyond this trench, llesting on the outer edge of the original block 

 of ground F was a wood frame about 1 '1 inches deep and T) feet siiuare, A. 



Fig. 2. 



This served as a container for the manure E. The trench (1 was covered 

 in with unbleached cheese cloth attached to the inner and outer frames, 

 and on strips of wood having central holes, ordinary wire balloon traps 

 were fitted with a sliding arrangement enabling them easily to be 

 removed and replaced. At the four corners of the inner frame stout 

 canes were set up and joined to a small board D, which also carried a 

 fly trap. A tent-like cover of cheese cloth was then made to fit over 

 this framework and was fastened down with tape and tacks all round 

 the inner wood frame. In this way it was possible to note the numbers 

 and kinds of flies which liatched from the heap pro))er and from the 

 trench respectively. No bait was used, it being found (liat the photo- 

 tropic tendency of the flies was sufficient to cause them to enter the 

 traps, and it is certain that very few failed to do so. The general 



