CHEMICAL REACTIONS OF FRUIT- FLIES. 303 



of both (the tryoyii I owe to the kindness of Mr. Froggatt) reveals nothing that can 

 be called a specific difference, and I think there is little doubt that the name iryont 

 should rank as a synonym, and that methyl eugenol will be found to attract the flies. 



We have then at Pusa five species, whose life-histories show some diversity. 

 The larvae of zonatus feed on fruits, particularly on peach, mango and guava ; 

 those of ferrugineus mainly on the fruits of Solanaceae, and on mango ; those of 

 diversns on the anthers of male flowers of Cucurbitaceae ; and those of mcmhitae 

 and candatus on the flesh of Cucurbitaceous fruits. 



Of these five species, my earlier experiments gave information regarding the 

 substances attractive to diver sus, ferrughieus and zonatus, but none respecting 

 cucurbitae and caudatus. The fact that the preferences of diversus and zonafys for 

 iso- and methyl -eugenol respectively were clearly differentiated, encouraged me to 

 hope that it might be possible ultimately to unravel the chemical relations of all 

 five species. The question of the relationship between zonatus Si,n^ ferrugineus is one 

 which I do not attempt to deal with at present, but it seems to be much closer than 

 that between zonatus and diversus. The morphological and general " bionomic '' 

 evidence is so far in agreement with the close connection revealed by the chemical 

 experiments. I refer again to this point later on. 



In the hope of attracting cucurbitae and caudatus I tried a variety of substances 

 and mixtures. Among these were : — 



(1) A solution of vanillin in hot amyl alcohol, mixed with water and exposed 



for several days in a bowd ; 



(2) a solution of vanillin in amyl alcohol, warmed with hydrogen peroxide and 



exposed in water. 

 In both cases a wine-purple solution is obtained ; in (1) the colour appears after a 

 day or two, in (2) it develops in a few minutes. At the time when these solutions 

 were being exposed (March), caudatus was known to be breeding in a patch of gourds 

 in the kitchen-garden a quarter of a mile away, but it was very scarce, only one fruit 

 containing larvae being discovered. In the course of a fortnight, four male caudatus 

 were found drowned in the amyl-vanillin solution, three in No. 1 and one in No. 2. 

 One of these was observed (in No. 1) when it settled, and its behaviour was quite 

 similar to that exhibited by zonatus and diversus under like conditions. When a 

 single fly of either of these species settles in a bowl, it soon becomes curiously fearless ; 

 as ojie of my assistants correctly phrased it, " they pay no attention to threats." 

 The same thing was noticeable with the caudatus. The amyl-vanillin solutions were 

 exposed on the same table as the eugenol derivatives which were attracting other 

 species, but no caudatus were found in or on these bowls, while the amyl-vanillin 

 had no observed attraction for any species but caudatus. No caudatus were caught 

 by solutions of amyl alcohol in water, or by vanillin in dilute ethyl alcohol, but 

 the short caudatus season was very possibly over when these were exposed. 



The nature of the chemical substance here involved is unknown ; it may be an 

 oxidation-product of vanillin. The vanilla-like smell (suspected to be sexual) emitted 

 by certain male Lepidoptera might be taken to point vaguely to possible analogies 

 between the two groups in the matter of chemico-sexual relations. I have alluded 

 already to the sharp difference between the remarkable attractiveness of iso-eugenol 



