122 INSECT PHYSIOLOGY 



under conditions of starvation, the butterfly Pyrameis, and the fly 

 Calliphora, may be able to detect with their tarsi concentrations of 

 sugar of i 2 ,8oo 7 M or less : the threshold for the human tongue lying 

 between -64 and ^2 M. The apparent sweetness of the various sugars 

 may be very different for insects and man ; and insects may differ in 

 this respect not only between themselves, but even when the organs 

 of taste on the tarsi are compared with those on the mouthparts : in 

 Calliphora, lactose is apparently tasteless to the tarsi, but evokes a 

 response from the mouthparts. By a system of rewards and penalties, 

 the water beetle Dytiscus may be trained to associate particular 

 tastes with the presence of food. 



Insects vary enormously in the extent to which their behaviour is 

 directed by the sense of smell. Many species find their food largely by 

 this means (cockroaches, crickets, Drosphila, &c); butterflies often 

 appear to be led to the food plant upon which to lay their eggs by 

 some single odorous constituent in the leaves; parasitic Hymenop- 

 tera find by smell the larva in which to lay their eggs, sometimes 

 detecting this from a distance, sometimes following the trail of scent 

 which it has left behind it ; the following of trails of scent is doubtless 

 very important also in the orientation of ants; but the highest de- 

 velopment of this sense is seen in the males of those Lepidoptera 

 (Lasiocampidae, &c.) which can detect and locate the female from 

 distances of more than a mile by means of the olfactory organs in the 

 antennae. 



The potentialities of the sense of smell have been most fully 

 studied in the case of the bee; and in this respect the bee's perceptions 

 seem to be surprisingly like our own. Flowers odourless to us are 

 odourless to them; substances of unlike chemical composition which 

 smell alike to us - such as nitrobenzol and oil of bitter almonds - 

 are confused also by the bee; and the threshold concentration for 

 odours to be detected seems to be of the same order of magnitude for 

 bee and man. Bees can readily be trained to associate the presence of 

 a supply of food with particular scents; and they can retain the 

 memory of such associations for several weeks - far longer than they 

 can visual memories. But it is interesting to note that they can be 

 trained only with aromatic odours such as might occur in flowers, 

 and not to such substances as indol or asafoetida. 



