I 



62 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.39. 



(PN2, PN3), which form conspicuous wide transverse plates on the 

 surface of the dorsum behind the wing-bearing nota. In Alloperla 

 (fig. 15) they are weakly continuous with the notal plates, but in 

 almost all insects, where they occur, they are separated from the 

 latter by narrow membranous sutures, as in Tseniopteryx (fig. 16). 1 

 Therefore, it is clear that the so-called "postscutellum" is not a 

 difi^erentiation of the true notum, as is the prescutum, scutum, or scu- 

 tellum, but is an additional plate, and, hence, the writer's ground for 

 designating it by the more generally significant term of "postnotum." 



Asain examinino; figure 14 it will be seen that the dorsal longitudinal 

 muscles (DMd) are truly segmental at this stage of development. 

 Woodworth (1909), however, thinks otherwise, for he says, "The 

 great dorsal muscle of flight for which the phragma was developed 

 is probably only a dorsal intersegmental muscle. These extend from 

 the anterior edge of one segment to the corresponding part of the 

 next." The writer can not see how the annular constrictions of any 

 nymphal form can be anything else than the intersegmental lines. 

 They certainly appear to correspond with the grooves between the 

 embryonic somites. Moreover, the muscle somites of the embryo 

 correspond with the body somites. This is true even in adults. If 

 the thoracic and abdominal terga of Machilis be removed there are 

 uncovered muscular segments exactly corresponding with the chitin- 

 ous segments. As has already been pointed out, the postnotal mem- 

 branes of the nymph (fig. 14, Mb) are not "intersegmental," but lie 

 before the true intersegmental grooves. For this reason the longi- 

 tudinal muscles of any segment may pull the succeeding segment 

 forward, by their contraction, just as if their posterior ends were 

 inserted upon the anterior edge of the latter segment. 



If the ancestral insects were wingless creatures, as is universally 

 conceded, then it must be assumed that the primitive function of 

 the longitudinal muscles was the movement of the segments, prin- 

 cipally the retraction of each into the preceding segment for purposes 

 of locomotion or respiration. It follows next, as a corollary to this, 

 that the part these muscles play in the movement of the wings in 

 modern insects has secondarily devolved upon them in the meso- 

 thorax and metathorax. Now, in order that the contraction of these 

 muscles may change the shape of these two segments instead of 

 telescoping them, it is clear that the postnotal membranes must be 

 obhterated in some way, so that the chitinous parts shall abut against 

 each other. We can imagine that this might be efi'ected in three 

 ways: (1) By a posterior extension of each notum till it should meet 

 the succeeding notum, (2) by a chitinization of the postnotal mem- 

 branes, or (3) by a shortening of these membranes. There is no 

 evidence that the first has ever happened — no insect shows a pos- 

 terior prolongation of the notum behind the scutellum, which would 

 be a true postscutellum, though the scutellum itself is often enlarged 



