234 ANATOMY. 



abdomen. Beneath this partition, namely, at the pectoral side, a lunate 

 space remains free, through which the internal organs pass from the 

 thorax into the abdomen. Besides this most important position of the 

 internal skeleton of flies, we find, in the neighbourhood of where the 

 wings are attached, other horny arches, which serve for the insertion of 

 the alary muscles. In front also of the larger partition the scutellum 

 sends into the cavity of the thorax a small ridge, which is however as 

 unimportant as the other is important. The dorsal muscles ascend 

 obliquely through the thorax from the great partition to the meso- 

 notum, and thus hold the whole structure together. 



In the Lepidoptera, which in the structure of their thorax have 

 most resemblance to the Diptera, the conformation is already some- 

 what more complicated. In this both agree that everywhere where 

 there are exterior furrows we find corresponding interior ridges which 

 separate the points of insertion of the muscles, and thus increase their 

 firm adhesion. Such a ridge rises from the centre of the rnesonotum, 

 which passes to the scutellum, and there unites with the ridge that 

 separates the scutellum from the mesonotum. From the posterior 

 margin of the scutellum a broad partition (the mesophragma of Kirby 

 and Spence) descends, it bends first backwards and then forwards, and 

 thus forms a hook, to which the large dorsal muscles are attached. This 

 partition is analogous to the ridge of the scutellum in the Diptera. 

 The third very narrow thoracic segment leans against it, forming also 

 a posterior partition, which, however, is much more delicate and fine 

 than the first ; consequently the relations of both the partitions, in 

 comparison with those described in the Diptera, are changed, here the 

 first is the largest, and there the second. The pectoral side of the 

 thorax exhibits a central projecting ridge as the line of separation 

 between the coxae and other smaller ones corresponding with the exterior 

 furrows. 



The Hymenoptera make the direct passage from the forms already 

 described to those in which the prothorax is separated. The exterior 

 furrows of their thorax are true sutures, in which their parts are joined. 

 This has been already sufficiently explained above ( 7478.), and it 

 is there shown that the collare is the true prothorax of the Hymen- 

 optera ; we will therefore here proceed with the internal processes. In 

 the prothorax there are two strong pointed processes (PL XII. No. I. 

 f. 4. a, a), each of which has a double root; one exterior one comes 

 from the margin of the prosternum, and an interior one from the 



