THE CANADIAN ENTOMOLOGIST. 419 



PHYLOGENY OF THE LITHOCOLLETID GROUP. 



(Preliminary Survey.) 



BY ANNETTE F. BRAUN, UNIVERSITY OF CINCINNATI, OHIO. 



Of the descent of this group — Cremastobombycia, Braun ; Litho- 

 colletis, Hbn. ( Phyilonorycter, Hbn.) ; Cameraria, Chapman ; Porphy- 

 rosela, Braun — from Gracilariad ancestry there seems to be no question, 

 but the exact phylogenetic relationship of the constituent members 

 remains in some doubt. Especially so is this true of the group possessing 

 abnormally flattened larvte, Cameraria, Chapman, a remarkably homo- 

 geneous complex of species, both in larval and imaginal structure and in 

 type of markings. 



In order to understand the basis of some of the recently proposed 

 theories of the descent of this group, it is necessary to review briefly some 

 of the details of structure in the early larval stages upon which such classi- 

 fication rests. The principal structural larval character possessed in 

 common by their Gracilariad ancestor and by the groups assumed to be 

 descended from it, is that all have at least two so-called Gracilarian instars. 

 The first two in Gracilaria and the first three in the Lithocolletid group 

 are of this character. Dr. T. A. Chapman, in his admirable paper 

 entitled "The Classification of Gracilaria and Allied Genera" (The 

 Entomologist, 1902, pp. 81-88, pp. 138-142, pp. 159-164), has discussed 

 the significance of this character as showing the descent of these genera 

 from Gracilariad ancestry, and as a taxonomic character, where it is ex- 

 tended into the third instar, as is the case in Cremastobombycia, Lit ho - 

 colletis (typical), Porphyrosela, and as he supposed, erroneously, however, 

 into the fourth and fifth instars of the flat-larval group. It is just with 

 regard to this point that errors in the observations of previous workers, 

 notably Chambers (Jn. Cin. Soc. Nat. Hist., II, pp. 79-93, 1879), have 

 resulted in the flat larva being regarded as more widely divergent from 

 the ordinary form in the latter instars than is actually the case. The true 

 Gracilarian larva, as it exists in the first three instars, possesses unusually 

 large labrum and labium with but very rudimentary labial palpi, flattened 

 mandibles, able to cut only the substance of the leaf directly in front of 

 them, and no maxillte that can be detected, except as represented by 

 indistinct lines on the mandibles. With the third moult of the so-called 

 cylindrical larva, the structure and form change to the normal type, with 

 all the mouth-parts present. This change in structure occurs with the 

 third moult in the flat group also, but the flattened /<?;-/« persists, and the 



December, 1909 



