20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Ill 



broad, others narrow; some are widest in the midabdominal region, 

 others across the thorax. Within these limits the character is rather 

 dependable. 



Color differences usually are not reliable except between marked 

 contrasts such as pale yellow or dark brown, not patterned or dis- 

 tinctly patterned, and larva unicolorous or dorsum much darker than 

 venter. Allowance must be made for the lack of uniform expression 

 and interpretation in the use of color and for the fact that mature 

 larvae may be darker than younger individuals, recently moulted speci- 

 mens are paler, and preserved material frequently becomes darker. 



The ninth abdominal segment offers the most important diagnostic 

 characters and should always be examined carefully, giving special 

 attention to the caudal notch, the urogomphi, and the dorsal plate. 

 Erosion may dull the points of the urogomphi, tubercles, and "teeth," 

 but important specific and generic characters are more frequently 

 drawn from this segment than from any other part of the larva. 



Prominent sculpturing such as deep pits, transverse rugae, and 

 tergal impressions appear to be reliable characters. Small or callow 

 specimens have less conspicuous sculpturing than large or fully dark- 

 ened specimens of the same species. The density of punctures and 

 the length of impressions vary considerably and usually require 

 observation on a good series of specimens to establish the common 

 range of variation. 



Setae often vary markedly both in number and in arrangement. 

 However, they are usually reliable for distinguishing such contrasts 

 as "setae absent or present" and "setae paired or unpaired." In 

 specimens in good condition the setal arrangement on the abdominal 

 mediotergites is generally dependable, but occasionally an individual 

 is found with almost double the normal complement of hairs. Young 

 larvae have fewer setae than older specimens. For these reasons, setal 

 characters should be supported by other evidence where possible. The 

 number of spinelike setae on the various segments of the legs and 

 on the thoracic episterna should not be relied upon unless the dif- 

 ferences are marked and constant. 



The nasale is a very useful structure. It is subject to considerable 

 erosion, but in the great majority of larvae the general type of nasale 

 can be determined even when worn. Where there are small lateral 

 denticles which soon wear away the structure loses much of its taxo- 

 nomic value. However, workers should become familiar with all 

 common types of nasale so that they can be recognized even when 

 eroded. 



