No. 1774. THORAX OF HYMENOPTERA—SNODORASS. 77 



Epm^) are present in each of these segments. The same is true of the 

 mesopleurum of some of the Chalcidoidea (34, 37, 45, 47, 48), in which 

 the episternum (Eijs^) and epimenmi {Eym^) are equally developed 

 and are separated by a distinct pleural suture extending fi'om the 

 coxal articulation to the wing process ( WP^) just as in Tremex 

 (1, P/Sz). In the Chalcids and some other forms, however, the meso- 

 pleurum is complicated by the presence of a plate (Ppct) in front of 

 the episternum, but this will be described later. 



In the rest of the Hymenoptera there is a distinct tendency toward 

 the reduction of the mesepimerum and the obliteration of the meso- 

 pleural suture with the result that the mesopleurum comes to consist 

 of one large plate (50, 51, 52, Pl^) carrying the wing articulation 

 ( WP2) above and the coxal articulation below. In the Ichneumo- 

 noidea the mesepimerum (20-28, 31, 32, Epm.^) is a narrow, though 

 usually perfectly distinct, plate on the posterior edge of the large 

 mesepisternum {Eps^), being widest, amongst the species figured, in 

 Odontaulacus editus (22). The suture separating this plate from the 

 episternum is either a distinct line (20, 22, 23, 25, 26, 28) or is marked 

 by a series of quadrate pits (21, 24, 27). In any case it can be identi- 

 fied as the pleural suture by examining the interior face of the thoracic 

 wall, for its course is here marked by a distinct pleural ridge (32, PR) 

 extending from the coxa to the wing process ( WP2). 



The degeneration of the mesepimerum appears to the writer clearly 

 demonstrated by this series of forms (1 , 17-28) just described. When, 

 now, it is found that, amongst the Cynipoidea (49-52) and some of 

 the Proctotrypoidea (59), there is no trace of this kind of a subdivi- 

 sion of the ])leurum, the conclusion is inevitable that the true epister- 

 num and epimerum are indistinguishably fused. Fmally, therefore, 

 when other subdivisions of the mesopleurum are discovered the con- 

 clusion that these latter are of secondary formation seems equally cer- 

 tain. In a few cases, already pointed out, the mesopleurum consists 

 of one undivided plate (50-52, PI2), but it more frec{uently (Hexa- 

 plasta, sp., 49; Proctotrypes caudatus, 57; Tropidopria conica, 59) 

 becomes differentiated into an upper wing-bearing part {PI2) and a 

 lower leg-bearing part (ph) by a horizontal or oblique suture. This 

 suture, moreover, is usually near the middle of the pleurum and is 

 always above the articulation of the coxa. Hence, it can not be con- 

 fused with the sterno-pleural suture (10, 16, 23, 27, g) which is always 

 below the coxal articulation. Yet nearly all writers on the Hymen- 

 opteran thorax have called this lower mesopleural plate the "meso- 

 sternum." To be sure, in nearly all the higher families as well as in 

 some of the lower forms, it is entirely continuous with the sternum, 

 but those species having a sterno-pleural suture clearly demonstrate 

 where the true division between the sternum and the pleurum occurs 

 when it is present. 



