120 



horizontal lines for distinct si>ecies; the three phases of the cycle may 

 be expressed by means of a diajrram, thus: 



Phase 1 : One oiiiiiivorons specie.s 



Phase 2 : Four exclusive species 



Phase 3: Several more or less onuiivorous sjx'fit^s. 



The vertical lines express continuous descent. 



Following on Phase 3, ordinary natural selection might eliminate all 

 but one of the new omnivorous species, as they would evidently come 

 closely into competition. If such elimination occurred, w^e should start 

 again as in Phase 1 of the diagram; but, if, as is perhaps usually the 

 case, the elimination was only partial when the breaking up began 

 anew, the result would probably be an increase in the total number of 

 species. 



When classifying the fauna of any plant or group of plants, we 

 should, according to the above hypothesis, need to distinguish two 

 classes, as follows: 



(1) An Endogenetic class, consisting of species which had their origin 

 on the plant or plants under consideration. 



(2) An Exoyenetie class, consisting of species which had their origin 

 on other plants. 



I use the terms endogenetic and exogeneUc as being the best I can think 

 of at present; but I am not quite pleased with them and shall be glad 

 if some member of the Association can substitute better ones. Other 

 terms I had thought of were original and derived, but original is open 

 to the objection that it conveys the idea that the species was always to 

 be found on the plant, Avhereas, in all probability, the species of plant 

 existed long before the particular species of insect came into existence. 

 But, setting aside the question of terms, I think there can be no doubt 

 that the distinction is a proper one to make, although it must be ad- 

 mitted that in many cases it will be very difficult to tell exactly which 

 class a given insect should be referred to. The same difficulty meets 

 us when we try to learn which of two countries, both inhabited by a 

 species, gave birth to that species, but to my mind the interest of the 

 inquiry is not lessened by the difficulties it presents. 



The endogenetic class will, of course, have to be divided again into 

 two subclasses, which are of extreme importance from an economic 

 point of view : 



(1«) The univorous subclass, of species still restricted to the plant 

 on which they originated. 



(16) The multivorous subclass, of species which have become mod- 

 ified in the direction of omnivoroasness. 



