454 SYNOPTIC COLLECTION. 



In the family of Limnophilidae, the members of which 

 have slightly curved horn-like cases, as seen in PL 1153, 

 fig. 2, there is one genus, Enoicyla pus ill a Burm. (PI. 

 1153, figs. 1-3), that lives on the land. This is the only 

 Trichopterous insect so far known that is terrestrial. 

 The figure of the larva (fig. i ; fig. 2, larva in its case) 

 does not exhibit any vestiges of respiratory filaments 

 which we should expect to find if the ancestors were 

 aquatic. It may be, however, that a sufficient period 

 of time has elapsed for the terrestrial habitat to cause the 

 complete disappearance of the branchial filaments. This 

 seems the more probable since the female (fig. 3) is spe- 

 cialized by reduction, having lost both pairs of wings. If 

 this is not the case, then the insect is primary and should 

 be placed as the most primitive of the Trichoptera. 



Order 14. LEPIDOPTERA. 



The caterpillar-type of larva has now become so fixed 

 in the insect organization that it is found with scarcely 

 an exception in the members of this immense order of 

 Lepidoptera. The variations that occur are mainly in the 

 minor details of structure, while the fundamental form and 

 characters remain essentially the same. Although this 

 is true speaking broadly, we shall see farther on that there 

 is one moth, Melittia satyriniformis Hbn , which throws 

 light on the genealogy of the order, since it passes 

 through a stage when first hatched that is comparable 

 with the Thysanuriform stage of the more generalized 

 insects. This stage, however, is brief, and the larva has 

 attained the caterpillar form when half grown. As we- 

 have stated elsewhere the common custom of calling cat- 

 erpillars "worms" is misleading and therefore lias not 

 been followed in this work. The word was given because 

 the caterpillar, like the worm, is cylindrical and seg- 

 mented, but if all cylindrical, segmented animals were to 



