109 



mained distinct. These wings illustrate, therefore, what has 

 been termed specialization by reduction. 



' There are, however, several orders of insects in which, as 

 a rule, some of the wing-veins have a greater number of 

 branches than in the hypothetical type. These illustrate 

 what has been termed specialization by addition. 



Among the orders in which 'specialization of wing-veins 

 by addition has taken place is the Neuroptera ; and several 

 genera of this order will be studied as examples of this 

 method of specialization. We will, however, first point out 

 the more important features of specialization of wing-veins 

 by addition. 



In speaking of an increase in the number of veins, refer- 

 ence is made only to a multiplication of the branches of the 

 principal veins. In no case is there an increase in the num- 

 ber of principal veins. And this increase in the number of 

 branches may be confined to one or two of the principal 

 veins in the preanal area, while the number of the branches 

 of some of the other veins may be reduced, the expanding 

 of some parts of the preanal area resulting in a crowding of 

 other parts. In some cases we will find that the multiplica- 

 tion of wing-veins extends to the anal area also ; in others 

 we will find the anal area greatly reduced. But even in 

 those cases where the anal area is reduced, the total result 

 has been the production of a n:iany-veined wing. 



In the many-veined wings both the longitudinal veins and 

 the cross-veins are increased in number. In most cases 

 where there are many cross-veins it is impracticable to dis- 

 tinguish from others those particular cross-veins to which we 

 applied special names in describing the few-veined wings. 

 (See page 91.) But in the case of the longitudinal veins it 

 is necessary to distinguish the primitive veins, that is, those 

 of our hypothetical type, from the veins that have been de- 

 veloped in addition to these. For if this is not done it will 



