408 MORPHOLOGY OF THE ABDOMEN IN THE ODONATA 



of the sixth segment on either side at their extreme anterior- 

 lateral points (on the suture) . It lies between the fore end of the 

 branchial basket and the vertically directed short intestine, 

 below the pre-rectal ampulla and above a mass of Malpighian 

 Tubules. When relaxed it bows forward to near the middle of 

 the fifth segment. Matula figures it in the fifth segment but 

 Wallengren places it in the sixth, as it is, also, in Anax. It is 

 about 8 mm. long and 1.3 mm. wide. 



The Muscular Diaphragm (or Supraintestinal Muscle of Ma- 

 tula) (plate XXII, figure 6, di) has been shown by Wallengren 

 to be a true diaphragm possessing a dorsal and a ventral portion. 

 It is attached to the lateral regions of the tergum of the extreme 

 anterior end of the fifth segment. Its ventral portion is similar 

 to but weaker than the Subintestinal Muscle; its dorsal portion 

 is a broad and very thin muscular sheath which would stand 

 vertically were it not for the large lobe of the ventriculus which 

 passes beneath it and throws it into an obUque plane. It can be 

 distinctly seen in a dissection from the dorsal side, binding down 

 the mid gut. Its attachment does not run more than two thirds 

 of the distance to the mid-dorsum on either side. Thus the dorsal 

 tracheal trunks are not affected by it. 



III. Anisoptera. Libellulinae 



Very little work has been done on the Libellulidae. Sadones 

 in 1896 gave a fairly complete account of the digestive tract, 

 and something on the respiratory system of Libellula depressa. 

 His results on the alimentary canal have been used as a basis for 

 most of the subsequent investigation of this phase of the organi- 

 zation of the Anisoptera. In 1905 Scott pubHshed an account 

 of the distribution of the tracheae in Plathemis lydia and included 

 a little work on the digestive system. 



The present work was done mainly upon Tramea Carolina, with 

 occasional comparisons with LiheUula, Plathemis, and Sjjmpe- 

 trum. The Tramea larvae are the most transparent known to 

 the writer, and many structures may be examined through the 

 body walls without dissection.. The tracheae can be traced 

 everywhere with remarkable sharpness. 



The following account makes no attempt at complete descrip- 

 tion; only such points as are peculiar to the species studied will 



