24 MR. A. HAMMOND ON THE 



the central portion of the thoracic cavity towards the dorsum, 

 and are chiefly conspicuous in the alary segments, and (with the 

 exception of the Libellulidse) more especially of such insects as 

 are remarkable for their power of flight, in the production of 

 which, as is shown by Chabrier, they are chiefly instrumental. 

 They fall in two divisions, one on either side of the mesial line, as 

 may be seen in Plate I. figs. 8, 10, & 11, and in Plate II. figs. 5, 

 10, & 11. They extend from the praescutum to the post- 

 scutellum of the segment to which they belong, in every insect 

 with which I am acquainted, and are the " muscles dorsaux au 

 abaisseurs des ailes " of Chabrier, the recti dorsales of the 

 larva. 



In the Coleoptera, where the thoracic segments are unmistak- 

 ably distinct and the phragmata well developed, their attachments 

 are equally clear ; here, however, they are chiefly confined to the 

 metathorax, as the faculty of flight in those insects resides in 

 that segment. In Acrida viridissima we have an example where 

 that faculty is resident in both segments ; and accordingly we 

 find that the longitudinal muscles are present in both (see PL II. 

 fig. 11); and though the two alary segments are much more inti- 

 mately united in this insect than they are in the Coleoptera, the 

 two sets of muscles are perfectly distinct, the length of each 

 being coextensive with the limits of the segment to which it 

 belongs. 



In the Lepidoptera the same rule holds. The anterior wings 

 of these insects would appear to be the chief agents of flight ; for 

 we find the dorsal longitudinal muscles confined to the meso- 

 thorax, the vertical ones only being found in the succeeding seg- 

 ment. They extend from the praescutum to the postscutellum, 

 from the pro- to the mesophragma, both of which partitions are 

 well marked. See PL II. fig. 10, which represents a section of 

 the thorax of Liparis salicis. 



In the Hymenoptera a somewhat singular conformation exists, 

 which was first, I believe, pointed out by Macleay. The meso- 

 thoracic postscutellum is detached in the mesial line from the 

 scutellum, and only remains attached by its lateral extremities, 

 so that the narrow rim of the metathorax follows immediately 

 upon the scutellum of the preceding segment. Nevertheless 

 that the detached septum thus formed is the postscutellum of the 

 mesothorax is evidenced, as Macleay says, by the fact that when 



