No.1745. AMERICAN BEES OF THE GENUS VOMIA COCKERELL. 291 



In typical Epinomia (triangulifera) the tongue is short. In Dieu- 

 nomia it is dagger-shaped. In Paranomia it is elongated. In 

 Hoplonomia, which is really a section of Paranomia, it is long and 

 slender, as I find on examining a specimen of N. elliottii Smith, from 

 India. I also find a linear tongue in some Australian species, as 

 X. semiaurea Cockered and X. gilberti Cockerell, the latter having it 

 remarkably long and slender. 



The tongue and mouth parts of Nomia diversipes are typically 

 Halictine in character, and compared with those of Agapostemon 

 melliventris Cresson, they show no important difference. The huge 

 hyaline inner lobe of the maxillary blade is quite the same, and this 

 is especially noteworthy because Andrena shows an entirely different 

 structure, wit lino hyaline lobe. Nomiacsm not go in the Andreninae, 

 but must be considered to typify a subfamily Nomiinae, closely related 

 to the Halictinae. 



The following notes compare Halictine with Nomiine bees: 



Agapostemon melliventris Cresson. A specimen compared with 

 Nomia diversipes showed a difference in the paraglossae, which hardly 

 exceeded the apex of second joint of labial palpi, while in A", diversipes 

 they reached the level of base of last joint. The Agapostemon was a 

 male, and on examining a female of the same species, it was found that 

 the paraglossae nearly reached the tip of the labial palpi. In Agapo- 

 stemon radiatus Say the paraglossae were found to reach about the 

 level of the end of the labial palpi. The tongue and labial palpi are 

 longer in this species than in A. melliventris. 



Spliecodof/asfni texana (Cresson) has the tongue as in Nomia diver- 

 sipes, and the broad paraglossae reach the middle of the last joint of 

 labial palpi. The maxillary palpi are much larger and longer than in 

 N. diver si pi *. 



Halictus sisymbrii Cockerell differs from Nomia diversipes by the 

 much broader and shorter tongue, with the linear apical part very 

 much shorter. 



Halictus armatid ps Cresson has the tongue a little more slender than 

 in Nomia diversipes, but it tapers much more gradually to the linear 

 apical process, which is also shorter. 



It isevident that the Halictines and Nomiines have independently 

 produced groups with linear tongues. There are two such groups 

 among the Halictines which 1 consider to have generic value, namely, 

 Augochlora in America, and Thrinchostoma in Africa. It may be 

 asked why, if I consider these valid genera, I do not also accept 

 Paranomia as a genus on the same grounds. It is because I have 

 never seen a species of Augochlora or Thrinchostoma, which could 

 not be readily recognized as such, while I do not know how to divide 

 Nomia into similarly recognizable groups. This, however, ma\ 



