AN ESSAY ON THE CLASSIFICATION OF INSECTS. 189 



throughout Africa, except the north coast, Madagascar, Southern Asia, 

 China, Java, Sumatra, and Borneo (Sauss.). I have ventured to 

 restore Linna^us's specific name. If Latreille adopts the specific name 

 for the generic, it is no reason tliat the law of priority should be in- 

 fringed, and Linnteus's name must stand. It is unfortunate that the 

 two names should thus be the same, but the only way in which it can 

 be avoided is to change the generic name ; but it has been in such 

 general use for so many years that the confusion that would follow 

 would be worse than the evil of having the generic and specific name 

 alike, which at least has the advantage of fixing the type of the genus. 



Bellagio, East Grinstead, May 26th, 1897. 



AN ESSAY ON THE CLASSIFICATION OF INSECTS.* 



By John B. Smith, Sc.D. 



Of late years the phylogeny of insects has attracted con- 

 siderable attention from students, and much light has been 

 thrown upon the subject by the researches made. One of the 

 most notable facts has been the breaking away from the old 

 Linnean orders, and the substitution of a number of more com- 

 pact assemblages for some of the almost indefinable aggrega- 

 tions found in the old classification. New characters have been 

 sought, not only in structures visible externally, but even in 

 internal anatomical peculiarities. The subject is a very in- 

 teresting one, which the teacher is of necessity compelled to 

 study more or less, and which I was led to examine more par- 

 ticularly when the question recently came up as to the adoption 

 of some system in a general work on * Economic Entomology,' 

 which has since been published. The conclusions reached by 

 myself, while in general they agree with the latest published 

 results, have been arrived at by a somewhat different method, 

 and my ideas concerning the development of the orders are 

 somewhat unlike those heretofore accepted. I have tried to 

 adhere logically to a scheme of easy development, and have 

 made uee of some characters not heretofore particularly noted. 

 Leaving aside for the present all questions as to the origin of the 

 class " Insecta," and as to its ancestors, I start from a developed 

 hexapod — an archetypal Thysanuran with six, jointed legs ; 

 without wings ; with or without abdominal appendages other 

 than functional legs ; with no eyes or with ocelli only ; with a 

 head not greatly differing in size or form from the body seg- 

 ments ; with the thoracic segments equally developed and not 

 greatly differing except in appendages from those of the abdo- 

 men. This creature lived in moist places, perhaps partially in 



- ' Science,' n. s., vol. v. pp. 071-C77 (April SOtli, 1897). 



