Introduction 19 



general character of the speciahzation in a few important structures. 

 As it seems to me the primitive habit of the Coleoptera must have been 

 feeding upon a variety of decaying substances, animal and vegetable in- 

 differently, and the first Coleoptera, newly derived from some even more 

 primitive insect, must have been but poorly adapted to their work. It 

 has been shown by Sandor Gorka ' that the digestive system of such 

 as still feed upon decaying substances is of the simplest form. The habit 

 of feeding principally on animal matter, which characterizes the Ade- 

 phaga, is accompanied in the larvae as well as in the adults by adaptations 

 of structure that in classification justify making of them a sub-order. 

 The adaptation extends to the digestive system, which is highly special- 

 ized. The habit of feeding largely upon living vegetable tissue which 

 characterizes the more specialized Phytophaga and nearly all the Rhyn- 

 chophora is also accompanied by adaptations of structure, though in 

 an enth-ely different direction. These adaptations extend, as in the 

 Adephaga, to the digestive system and to the larvae; and in the Rhyn- 

 chophora, whose food is largely derived from the most recent develop- 

 ments of the vegetable kingdom, have reached a degree of specialization 

 that justifies treating that group as the highest development of the 

 sub-order Polyphaga, even if its comparatively recent origin permits 

 of tracing its descent and forbids treating it as a sub-order. 



The primitive beetles from which the two sub-orders have been 

 derived are of course extinct, and their characters must be deduced 

 from the theory just suggested. Being derived from some more primi- 

 tive generalized insect form and being the progenitors of the existing 

 forms, their structures must have been those conmion to both, but in 

 degree of adaptation exactly the opposite of that found in the most 

 specialized of existing forms. I have already pointed out that the 

 tarsi of the primitive beetle must have been composed of five equal, 

 unmodified joints and that tarsi of a less number of joints, or with joints 

 adapted to swimming or digging, must be regarded as derivatives from 

 the primitive form. Since, according to Dollo's Law,- a part once 

 lost or reduced to a vestigial condition cannot be regained in progressive 

 modification, a 3-jointed tarsus must be a derivative in comparison with 

 a 4-jointed tarsus, not vice versa, and such tarsal appendages as lobes 

 and onychium must be primitive indications, for they are lacking in highly 

 specialized beetles, but present in many more primitive insects. 



' Allgemeine Zeitschrift fur Entomologie about 1913. 



- See " A History of Land Mammals in the Western Hemisphere," New York, 1913, p. 656 

 The author, Wm. Berryman Scott, discussing the so-called law of irreversibility in evolution, decides 

 that while it is perhaps not universally exemplified, deviations are certainly exceptional. 



