126 ILLINOIS BIOLOGICAL MONOGRAPHS [444 



dynamous with the larvapods, and since it is natural to believe that the 

 process of reduction has taken place very slowly by gradual suppression 

 of the appendages, it is not unreasonable to assume that the apodous 

 condition found in the Pamphiliidae represents a much later stage of 

 specialization than the condition of the polypodous larvae of the Xyelidae. 

 The biology of the Pamphiliidae also indicates that this family is perhaps 

 more specialized than the Xyelidae. However, these considerations 

 counterbalance each other, and, when all is said, it is difficult to decide 

 between the two families as to their relative degrees of specialization. 

 This somewhat drawn-out discussion leads to the following conclusions: 

 (1) the progenitor of the Hymenoptera possessed a pair of larvapods on 

 each of the first ten abdominal segments and a pair of segmented subanal 

 appendages on the eleventh segment; (2) the progenitor gave rise to dis- 

 tinct stocks which resulted in the production of larvae with larvapods in 

 one case and with subanal appendages in the other; (3) the Xyelidae 

 represents the former line of evolution and the Pamphiliidae the latter; 

 and (4) the question as to whether the Pamphiliidae is more generalized 

 than the Xyelidae or vice versa is by its nature unanswerable. To the 

 above conclusions it may be added that it is only natural and reasonable to 

 consider the Tenthredinidae as representing the further evolution of the 

 primitive stock from which the Xyelidae had evolved, and the Cephidae, 

 Xiphidriidae, Siricidae, and Oryssidae, in turn, as evolving from the 

 original stock which gave rise to the Pamphiliidae. 



The subanal appendages are present only in the Pamphiliidae and 

 Cephidae. In the former they are rather long, setiform, well developed, 

 distinctly three-segmented; in the latter they are minute, vestigial, often 

 fleshy, papilla-like, and indistinctly segmented. Since the embryonic 

 history of these appendages has not been studied, their true nature is not 

 known. There is little doubt but that they are true appendages. If they 

 represent the appendages of the ultimate segment, as has been suggested 

 by certain writers, and correspond to the so-called style of generalized 

 insects, then their presence is an indication of a primitive condition. There 

 is hardly any question as to the common origin of the subanal appendages 

 in the Pamphiliidae and Cephidae, and if these structures represent what 

 they are assumed to represent these two families must have a close affinity. 



The ocellarae are present in the Pamphiliidae, Xyelidae, Tenthredin- 

 idae, and Cephidae. They are well developed, and are usually accompanied 

 by well-defined ocularia in the first three families. In the Cephidae the 

 ocellarae are vestigial and represented by localized pigmented granules, 

 and lack ocularia. It is significant that the ocellarae are unmodified in 

 the gall-makers and leaf-miners of the family Tenthredinidae, except in 

 Phlebatrophia, where they are reduced in size and the ocularia indistinct. 

 The atrophy of the ocellarae is undoubtedly correlated with the mining 

 habit of the larvae. 



