No. 1771. THORAX OF HYMEN0PTERA—SN0DGRA88. 61 



the Heteroptera, has a large postnotum in the mesothorax which 

 carries the middle phragma, but it is deeply fused mesially into the 

 front of the metanotum and looks like a prescutum of this seg- 

 ment. For this reason the writer made the erroneous statement 

 in a former paper (1909) that the Belostomidse have no postnotum 

 in the mesothorax. However, that the plate in question is such is 

 amply proven by its solid lateral connections with the mesepimera. 

 Cicada, on the other hand, has a much smaller postnotum in the 

 mesothorax, but the very large middle phragma is solidly attached 

 to the lateral parts of this segment. In both Cicada and Belostoma 

 the great mass of the thoracic muscles is in the mesothorax, though, 

 judging from analogy with beetles, one would suppose that in Belos- 

 toma, at least, the hind wings must do most of the flying. 



In general, however, it is evident that the attachment of a phragma 

 to the posterior part of either segment and its size are dependent upon 

 the development of the power of flight in that segment, and that the 

 postnotal plates are developed to support the phragmas. There are, 

 of course, many apparent minor exceptions to this where a compara- 

 tively large postnotum is present bearing only a small or even a 

 rudimentary phragma. But, in such cases, the dorsal muscle fibers 

 are attached posteriorly to the postnotum itself, which thus serves 

 as both postnotum and phragma.'^ In fact, many writers have made 

 no distinction between the phragmas and the surface plates to which 

 they are attached, defining the "prescutum" as the exposed part of 

 the prephragma and the "postscutellum" as the exposed part of the 

 postphragma of any segment. The present writer, however, for rea- 

 sons based on the following facts, prefers to distinguish between the 

 phragma and its surface support. 



The reader's attention has already been directed to figure 11, 

 showing the back of a stonefly nymph, in which each thoracic tergum 

 consists of a simple notal plate separated from the one behind by a 

 wide membrane. Figure 14 is a longitudinal section through the 

 back of a similar form. The depressions mark the constrictions 

 between the segments. It is, hence, evident that the membranes 

 (Mhi, Ml)2, Mh^) are not truly intersegmental, but are postnotal in 

 position, since they occur between the notal plates {N^, N^, N^), and 

 the posterior limits of the segments. If, now, this figure be compared 

 with figures 15 and 16, showing corresponding sections through the 

 mesothorax and metathorax of adult stoneflies, it will be seen that the 

 postnotal membranes are mostly replaced by postnotal chitinizations 



oThe Odonata constitute a very prominent exception to many of the above state- 

 ments. They have large postnotal plates in each wing-bearing segment but possesa 

 neither ptu-agmas nor longitudinal thoracic muscles. Their wings are moved entirely 

 by the great dorsoventral muscles which are inserted by means of stalked disks upon 

 the bases of the wings themselves and upon the adjoining parts of the nota. It may 

 be that the postnotal plates here serve the purpose of lengthening the terga so aa to 

 give more space for the play of the wings. 



