358 THE CANADIAN ENTOMOLOGIST 



faced, and that is the reason for studying insect behaviour. It is 

 not enough for the imago to deposit its eggs upon the food-plant 

 of its larva. The crisis of pupation has still to be passed. And 

 all this is done, without reflection or premeditation, by hereditary 

 predestination depending ultimately upon the specific physico- 

 chemical properties which make up the "hereditary patrimony" 

 of the organism. 



The habits of the insects of to-day afford a picture, more or 

 less altered by repetition and by circumstances, of the habits of 

 their ancestors, so that their observation has an antiquarian or 

 historical interest as well as an economic bearing, and it is not 

 necessary to mix up these two aspects. It is one of the paradoxes 

 of natural history that ancestral types can persist for untold ages 

 alongside the more differentiated types. The tropical genus 

 Peripatus is to the insect world what Amphioxus is to fishes. The 

 latter has the English name of lancelet, and by a corresponding 

 verbal transposition, Peripatus might perhaps become known as 

 the "larvelet" since it has some of the properties and much of the 

 appearance of an insect larva. Its habits are to this extent amphibi- 

 ous that whilst breathing air it requires an excess of moisture; 

 hence, like the alpine salamander amongst batrachians, it is charac- 

 teristically, though not invariably, viviparous. Whether oviparous 

 or viviparous, there is no free larval period in the life-history of 

 Peripatus; its trophic and reproductive phases are not separated. 



The multudinous traits of insects may be grouped under two 

 categories corresponding approximately with their larval and 

 imaginal phases. Idiotropic tendencies comprise the feeding, rest- 

 ing and protective devices of individual life. Phylotropic ten- 

 dencies comprise habits looking to the preservation of the race, 

 e. g., nest-building, egg-laying and brood-nursing. As an example 

 of a casual observation of an idiotropic performance I may relate 

 a small incident in my own experience. In December, 1905, I 

 was descending the bund or high embankment of one of the ancient 

 irrigation tanks of Ceylon when I noticed a dark brown Mantid 

 ensconced amidst the green foliage of a low shrub. It happened 

 to be a male of Gongylus gongylodes, a floreate species with folia- 

 ceous expansions on prothorax, legs and abdomen, known to 

 Aldrovandus and figured by him in 1602. 



