THE SENSE REACTIONS OF FLIES. 147 



on the similarly arranged controls ; and hence one was better able to judge whether 

 the flies were attracted by the colour, or simply came for the food. 



The number of flies coming to the differently coloured foods was approximately 

 equal. The same was found in the contrpls ; showing that the flies came more or 

 less haphazard to the baits, and were not attracted by any particular colour. 



The effect of colour was further tested by means of coloured light. Wratten 

 filters were attached to one side of a large glass cage, the other sides of which were 

 darkened, so that light could only enter through the filters. In some of the 

 experiments casein, and in others honey baits, reaching the whole length of the cage, 

 were so arranged that some part of them was illuminated by light from each of the 

 filters. Different species of flies, i.e., Musca, Calliphora, Lucilia, Phormia, Fannia 

 and Sarcophaga, were then introduced and kept under observation, and their 

 behaviour towards the different colours noted, as well as the numbers feeding or 

 settling upon them. The results were negative, as in the preceding experiment. 



Further experiments testing the colour-sense of flies were made with different 

 coloured materials. These were placed in different parts of the fly-room and kept 

 under observation, and the number of flies settling upon the various colours noted. 

 In order to avoid error due to differences in texture, pieces of the same material 

 were dyed different colours. The results again showed that the flies (Musca) had 

 no preference for any particular colour, and that the choice of material upon which 

 they settled was governed not by colour, but by some other factor or factors, e.g., 

 probably texture, position, temperature, or curiosity, etc. 



Temperature. 



It is known generally that house-flies are influenced by differences in temperature 

 and that they prefer warm situations, but precise experiments to determine their 

 optimum temperature, and the maximum and minimum temperature at which 

 they will feed, appear to be lacking ; and it was to remedy this, that the following 

 experiments were made. This liking of house-flies for warmth was well shown in 

 the way they congregated in large numbers round the lighted Bunsen-burner, sitting 

 round it in a very definite circle, the size of which varied according to the distribution 

 of the heat. This was easily tested by placing larger, or smaller pieces of asbestos 

 over the flame, when the flies arranged themselves in larger or smaller circles 

 respectively (Plate X, fig. 1). The temperature of these circles was found to be 

 very constant, varying between 42°-44° C. If the gas was turned off the flies 

 ceased to sit in the ring, but came closer in to the base of the burner, up the stem 

 of which they often crept (Plate X, fig. 2), until finally, when it was no longer warm, 

 they dispersed and settled anywhere in the room. As soon as the gas was 

 re-lighted, however, they formed a circle round it once more. 



After a number of preliminary experiments in testing baits at different temperatures, 

 in which however it was not possible to keep the temperature constant for any length 

 of time and thus not possible to determine it accurately, the following apparatus 

 was set up, in order to find the optimum, maximum and minimum temperatures 

 at which house-flies would feed. This apparatus (Plate XI, fig. 1) was designed 

 so that the temperatures of the baits could be regulated and kept constant for a certain 



