71] LEPIDOPTEROUS LARVAE — FRACKER 71 



Three subfamilies are recognized, of one of which specimens have 

 not been available. 



Gracilariinae. Head as high as wide, shape normal; front not 

 reaching vertical triangle. Body cylindrical, intersegmental incisions 

 shallow. Thoracic legs present ; crochets of prolegs arranged in a lateral 

 penellipse enclosing a transverse series, all iiniordinal (Fig. 103). Gra- 

 cilaria consimilella, G. violacella, and Ornix geminatella were examined; 

 and Coriscium, the only other genus, is described as being similar. 



Lithocolletinae. Head depressed, strongly narrowed cephalad; 

 front with subparallel sides, extending to vertical triangle and often 

 widest at caudal margin of head. Body depressed, moniliform. Thora- 

 cic legs present or wanting; ventral prolegs reduced, either bearing 

 transverse rows of crochets or modified into suckers, present on segments 

 3, 4, and 5. 



LitJwcolletis hamadryadella, L. ostensackenella, Marmara salictella, 

 M. fulgidella, and Cameraria sp. (?) were studied. In Marmara both 

 thoracic legs and crochets are present, while in Cameraria neither could 

 be identified. The individuals of Lithocolletis varied in this regard. It 

 is to be noted that in many cases legs are acquired at the last molt and 

 that a persistent study of all stages will be necessary before the different 

 genera can be distinguished with any assurance of accuracy. 



Family Tortricidae 



The family of leaf-rolling caterpillars is a remarkably uniform and 

 generalized one. Larval structure indicates rather close relationship 

 with both Cossidae and Gelechioidea but is not to be relied upon, for all 

 these groups are typical Microlepidoptera in most respects and may be- 

 long to wholly different but only slightly specialized stocks. .Individual 

 fluctuating variations are rather confusing in an attempt at classification 

 but all specimens seen will trace to the family except an occasional 

 aberrant one. Greater difficulties are met in working with the genera. 

 For example, about twenty percent of codling moth larvae have mu of 

 segment 9 located on the same pinaculum with kappa and eta and will 

 therefore not trace to Cydia. The writer has not found absolutely con- 

 stant characters to distinguish the genera of this family but hopes that 

 the synopsis given here will suggest other and possibly better means of 

 separation. The larvae are of sufficient economic importance to warrant 

 the expenditure of considerable time in the study of their structure. 



Head not depressed ; front extending from one-fourth to three- 

 fourths of the distance to the vertical triangle ; adf rontals usually touch- 

 ing the vertical triangle ; ocelli six, variously arranged, but the sixth 

 always rather close to the fourth and fifth. Body cylindrical or sub- 



