278 Chemotropic Responses of Insects 



their influence in different ways: the time of day, season, degree of 

 development of the gonads, age of the insects concerned, state of hunger, 

 development of the eggs, whether oviposition is in progress or not, all 

 probably have a definite bearing upon the chemotropic responses of 

 insects, but we are totally in the dark as to how they manifest themselves. 

 We have observed, and also Richardson (1917), that the same attractive 

 substance may vary in effectiveness on different days under apparently 

 similar atmospheric conditions. This may be due to one or other of the 

 above-mentioned factors, or depend upon the actual number of insects 

 present in a given locality at a particular time, or may in some way be 

 correlated w^ith sex attraction. Feytaud in his experiments in trapping 

 moths discovered that some traps lured exceptionally large numbers, 

 ■ especially at the end of the season, and in some cases in the proportion 

 of only two females to 54 males. He attributes this to sex attraction, 

 females being trapped first, males subsequently following them. In view 

 of the complexity of the factors which have to be taken into account, 

 it is necessary to repeat every individual experiment a number of times 

 during the season, and keep very full meteorological data. It will 

 consequently be evident that reliable results, concerning the chemotropic 

 properties of even a small number of substances, take several years to 

 achieve. 



All experiments to have any scientific value must be checked by a 

 definite control. When a chemical agent is exposed, and it attracts 

 certain insects, we acquire definite data. The question then to be gone 

 into is its relative attractive value as compared with other substances. 

 To prove that any substance exhibits attractive properties we are faced 

 with a number of problems. The negative results do not necessarily 

 imply that the agent has no attractive value, as the species which may 

 respond to the stimulus may be absent from the immediate locality at 

 a given time. It is, therefore, necessary to devise a double control — 

 one to prove that certain species of insects are actually present at the 

 place and time of observation, and the other to show that the substance 

 being dealt with has, or has not, chemotropic properties. As mentioned 

 above, water may serve as one control, but to devise the second control 

 substance is less easy of achievement. It is necessary to use a substance 

 known to exhibit powerful chemotropic projierties and this, in itself, 

 is liable to vitiate the experiment in attracting insects which might have 

 otherwise responded to the substance required to be tested. The senses 

 of insects are very acute, and minute traces of a reagent are known to 

 produce marked results. We are also totally ignorant as to the distance 



