445] LARVAE OF THE TENTH REDINOIDEA—YU ASA 127 



The antennae are present in all larvae of the Tenthredinoidea. Judg- 

 ing from the condition obtaining in generalized insects, it is reasonable to 

 consider the antennae of the Pamphiliidae as representing the primitive 

 ancestral type. They are long and setiform in this family and consist of 

 seven cyhndrical well-chitinized segments. The Xyelidae is closely 

 related to the preceding family in antennal characters altho a shortening 

 of the length has taken place. In the Tenthredinidae the antennae undergo 

 much modification both in the number and form of the segments. They 

 may be conical, limpet-shaped, or flattened, and the number of segments 

 varies from five to one. The antennae of the Cephidae resemble those of 

 the Xyelidae and some of the Tenthredinidae in shape and number of 

 segments. The antennae undergo steady reduction in size and number of 

 segments in the three remaining families, reaching the extreme of reduc- 

 tion in the Oryssidae, where each is represented by a button-like swelling. 

 The trend of specialization in the antennae is orthogenetic so far as the 

 families are concerned but quite diverse in the subfamilies of the Ten- 

 thredinidae. 



The mouth-parts, which include the mandibles, maxillae and labium, 

 afford a fertile field for characters which are of interest from a systematic 

 point of view. The mandibles, like the antennae, are the most persistent 

 and ever-present structures in the head of all larvae of the Tenthredinoidea. 

 The maxillary and labial palpi are typically four-and three-segmented 

 respectively. The change is in the reduction in number and size of the 

 segments. The Cephidae is normal in this respect but gradual change 

 takes place in the Xiphydriidae and Siricidae, while in the Oryssidae the 

 change has proceeded so far as to completely obliterate the maxillary and 

 labial palpi. The palpi of Phlebatrophia resemble those of the specialized 

 famihes. The families represent different stages of specialization, and 

 their relative systematic position can be indicated by the degree of changes 

 in the mouth-parts. 



The suranal process which is located on the meson of the suranal 

 lobe or the tenth urotergum is characteristic of the larvae of the Cephidae, 

 Xiphydriidae, and Siricidae. It should not be confused with the caudal 

 protuberances of certain Tenthredinidae, as these two structures are 

 of an entirely different nature. There is a minute hook-like process on the 

 caudo-meson of the tenth abdominal tergum of the Pamphiliidae. It 

 should be noted that in certain larvae of Pontania and Caulocampus the 

 caudal portion of the ultimate tergum is produced caudad as a blunt more 

 or less strongly chitinized protuberance which undoubtedly serves the 

 same function as the suranal process of the specialized families. These 

 two structural modifications of the caudal end of the body, however, are 

 not homologous with each other. The suranal process is undoubtedly 



