396 MORPHOLOGY OF THE ABDOMEN IN THE ODONATA 



drawn almost within segment two, while in the eighth it lies 

 slightly farther back in the segment than usual. In size they are 

 about equal, except that the last is always much the largest. 

 Their form varies from nearly round or oval to an inverted pear- 

 shape. Each has several pairs of lateral branches. There seems 

 to be no sex variation in the' nerve chain. The connectives are so 

 closely approximated as to form a single cord throughout most 

 of its length. 



The Muscular System 

 (Plate XXI) 



No previous exhaustive work has been done upon the nmscles 

 of Zygoptcrous larvae. Thci-e are occasional references to them 

 in the literature of general or special subjects, only. Calvert has 

 given partial figures of them in his studies of Cora and Thauma- 

 toneura, but has named only the more conspicuous ones. 



The straight, tul)u]ar body of the larva is heavil}' lined on all 

 sides with relatively thick muscles. The great number of these 

 run longitudinally and are limited to the segments in which they 

 arise. Certain muscles in the end segments of the abdomen are 

 intersegmental. Abdominal nmscles are devoid of tendons, and 

 origins and insertions occur by means of direct union to the chiti- 

 nized body wall or to the in-turned folds along the sutures and 

 joints (the ''Randwulste" of Wallengren). (plate XXIII, 

 figure 12.) 



In the naming of nmscles as distinct from one another the writer 

 has decidgd wholly upon the criterion of origin and insertion: a 

 muscle body with these regions common is considered a single 

 muscle, but if there is a common origin but two or more regions of 

 insertion, two or more muscles are taken to be present. This is a 

 problem that very frequently arises, as insect muscles lack the 

 definite sheath of vertebrate muscles and very easily split up into 

 their component fibers. Fixation is sure to bring this about 

 more or less. 



The longitudinal muscles of any sclerite are often superimposed 

 upon one another in a way that makes them very difficult to dis- 

 cover. Many of them are also quite small, in fact, so small as to 

 be indistinguishable even in stained material until laid bare. For 

 these reasons most workers, considering muscles but incidentally, 

 have overlooked them. As Wallengren's (1914) work on Aeschna 



