Changes in the Arthropod Fauna of an Experimental Bulk of Stored Wheat

In a previous paper (Coombs & Woodroffe 1963a) we described an experiment which demonstrated ecological succession in an insect population breeding in stored wheat. The experiment was on a small scale, involving two bulks of 6 kg of English wheat, stored in glass tanks in a layer 8 cm deep. Conditions were relatively dry and the bulk was too shallow to permit much heating due to insect activity. The tanks were closed by lids, thus preventing escape of insects and minimizing the chance introduction of additional species. This isolation of the insect community had two important consequences so far as the progress of the succession was concerned. The prevention of emigration maximized the denisty effects within the population, while lack of opportunity for immigration meant that the secondary species had to survive the earlier, less favourable stages in the succession; under field conditions, without restriction of access, they could have gained entry at a later, more favourable, stage. The present paper describes the progress of a similar experiment in bulk wheat under rather different conditions. The bulks were much larger, large enough to permit more heating. Conditions were damp, and fungal and bacterial decay of the grain took place. Both immigration and emigration of insects were possible and the fauna was consequently more varied and more variable in composition and numbers, than in the earlier experiment. The grain consisted of approximately 1000 kg of English Atle wheat stored in two equal bulks in a brick store of internal dimensions approximately 3-4 x 2 1 x 2-4 m. The store had a door in the middle of the eastern end and a window in the middle of the northfacing side. The floor was of concrete and the building was without any damp-proofing. The two bulks of wheat were stored along the side walls and were kept in position by wooden bulkheads, each bunker measuring 3-43 x 061 x 038 m. The bunker on the north side, under the window, was designated Bunker 1 and that on the south side, Bunker 2. Sampling The sampling procedure was designed to give an unbiased estimate of the number of each species at 3-monthly intervals and also to provide information on the horizontal and vertical distribution of each. In the previous experiment in glass tanks it was found that the insects aggregated along the edges and in the corners. In the present experiment, in order to reduce variation between samples due to this behaviour, the corners were eliminated by not sampling a block 35 cm long at each end of the bunkers, while the edge effect was incorporated into the sampling plan. Each bunker was regarded as consisting of a number of subdivisions, the positions and dimensions of which are shown in Fig. 1. The first subdivision was into outer, inner and middle regions. Each region contained a sampling strip, along the wall in the outer region, along the bulkhead in the inner region and in the middle of the middle region. Each