36 CONNECTICUT GEOL. AND NAT. HIST. SURVEY [BuU. 



labella are greatly reduced so that the}" will not impede the piercing- 

 operations of the j)roboscis, while in certain mvcetophilids, on the 

 other hand, they are extremely broad, membranous lobes, which may 

 become united to form a single broad lobe as in Mycetophila. 



In the higher Diptera, the membranous, lobe-like labella can be 

 dilated by blood pressure, and they are also operated ])y muscles at- 

 tached to the anterior labellar sclerite, als, (also called the discal 

 sclerite), and to the posterior labellar sclerite, pis, (also called the 

 furca, although this term has been applied to the endosternal apophy- 

 ses of the thorax, to the spring of certain Apterygota, etc.) shown in 

 Fig. 4, E. The small basal sclerites labelled h in Fig. 2, A of the 

 labium of Bittacomorpha, etc., are sometimes interpreted as the ves- 

 tigial labial palpi in these foi-ms, but they belong in the same cate- 

 gory with the numerous secondary plates developed in the basal region 

 of the labella. 



Pseudotracheae, pt, or small sclerotized grooves, resembling tra- 

 cheae in appearance, are developed on the "oral," mesal, or pseudo- 

 tracheal surface of the labella; and in association with these, small 

 prestomal teeth are formed in many muscoid flies, in which these ac- 

 cessory ''teeth" are used as rasping structures which are exposed and 

 brought into play when the labella are everted or foldeci back. These 

 ''teeth" become increasingly larger and more important in the muscoids 

 leading to the blood sucking forms,* such as Stomoxys and Glossina, 

 in which the labella are reduced and the "teeth" are used in the pierc- 

 ing operations of these flies. 



The pseudotracheae, pt, are represented by two simple, unbranched 

 channels? on each labellum in a {n-imitive dipteran of the TipuJcf- type, 

 while in such a higher type as Tahanus, each of the labella bears two 

 main pseuclotracheal trunks, one anterior and one posterior, which 

 give off more or less parallel branches extending over the pseudo- 

 tracheal, or "oral", surface of each labellum; and from such a type, 

 (he pseudotracheae of the Cyclorrhapha could be derived. In the 

 housefly, according to Hewitt (1914, p. 16), twelve pseudotracheal 

 branches extend over the oral, or pseudotracheal, surface of the label- 

 lum from the anterior trunk, and about twenty branches are given off 

 by the posterior trunk, while a ''median set of three or four pseudo- 

 tracheae run direct into the oral aperture (which) lies at the 



base of the small oral pit, which is a space kept open between the oral 

 lolies by means of the discal sclerites." 



The channels of the pseudotracheae are not complete tubes, since 

 a zigzag, longitudinal cleft or Assure (/ of Fig. 4. K) extends down 

 each of tliem, and interrupts the taenidia-like, incomplete hoops,, or 

 loops, which preA^ent the channels from collapsing. The loops are 

 bifld, or forked, at one end {i of Fig. 4,1 and K) and are broadened 

 at the other end, which is not bifid ; and the loops are so arranged that 

 the bifid end of one loop alternates with the broadened end (or base) 

 of the next loop, in the series extending along the sides of the cleft 



*The prestomal teeth are so large and powerful in the fly M%isca crassirostris, that 

 this flv can use them to scrape down to the blood-containing- tissues of cattle to ob- 

 tain blood to feed upon despite the fact that the labella are not otherwise modified 

 for piercing. 



