168 A^tnals Entomological Society of America [Vol. IX, 



Third Instar. — The conditions under which these larvae must 

 be reared make it very difficult to follow the ecdyses so that the 

 exact number of instars can be determined. The following 

 described instar is identified in this paper as the third since, in 

 the examination of many specimens of different ages in the 

 field and laboratory, no indication of an instar between it and 

 the second could be found. Larval changes were also followed 

 as accurately as possible in specimens reared from eggs and 

 the same result was obtained. The writer believes that the 

 results warrant the designation of this form of the caterpillar 

 (Fig. 3) as the third instar. 



Length, 4-5 mm. General color pale yellowish. Head 

 about 0.44 mm. wide, otherwise very similar to second instar. 

 Thorax with anterior margin of cervical shield dark, remainder 

 translucent; five pairs of setae on prothorax; mesothorax with 

 one pair of lateral, filamentous gills and one pair of dorsal, 

 filamentous gills; metathorax with three pair's of filamentous 

 gills, one pair laterad, one pair dorsad, and another pair ventrad. 

 First abdominal segment with three pairs of gills similar in 

 shape and position to those of metathorax; segments II and III, 

 with one pair of dorsal, two pairs of lateral, and one pair of 

 ventral gills; abdominal segments IV and V with two pairs of 

 lateral and one pair of dorsal gills; segments VI, VII, and VIII, 

 with two pairs of lateral gills; segment IX with one pair of 

 lateral gills; segment X devoid of gills; segments IX and X 

 with several pairs of setae, one of which is terminal in position 

 and much longer than others. 



It will be noticed that the above description agrees to some 

 extent with that given by Forbes for "Stage /F", the chief 

 difference being in the maximum number of gill filaments. 

 Although the writer has never found an instar in which the 

 maximum number of gill filaments was three, he is inclined to 

 believe that the third instar of this paper is the same as "Stage 

 IV" described by Forbes. 



Later Instars. — Examination of a large series of caterpillars, 

 varying from the third instar to the full-grown state, showed 

 the existence of a number of types, based on gill characteristics, 

 gradating from the former to the latter. Each successive type 

 is characterized by increase over the preceding one in body size 

 and in the number of gill filaments. Whether or not each type 



