12 [January, 



h A — ceiichri (the space between them 

 is the metanotuni) . 



I' (in fig. 2) — prosternnm. 



^— mesosternum. 



in — metasternum. 



n — mesoplcura. 



o — metaploura. 



/; p p — 00X89. 



EXPLANATION OF FIGURES 1 and 2. 

 a (in both figures) — pronotum. 

 h 6— tegulae. 



c — middle (or front) lobe of mesonotum. 

 d rf— side lobes of ditto. 

 e — scutellum (or better perhaps xcntelhim 

 mesothoracix, to distinguish it from 



f — postscutelUim {hetiev scutellum meta- 



thoracis) . 

 q — propodeum or median segment (the 



central slit in this is what Cameron 



calls the blotch). 



Note. — The unlettered areas in fig. I arc parts of the meso- and metathorax, 

 which are seldom if ever referred to in descriptions, and I therefore ignore them. 

 The shaded space indicates a very deep impression between the meso- and meta- 

 thoracic regions. 



I may add that — 



the prothorax includes the areas n and k. 



the mesothorax „ ,, c, d, and e in fig. 1, n and / in fig. 2. 



the metathorax „ ,, fin fig. 1, o and m in fig. 2. 



the propodeiim {g in both figures) is an originally abdominal segment 

 transferred to the thorax in pupation. 



The thorax of a Saw-fly can easily be broken up into its throe constituent 

 parts of pro-, meso- and meta-thorax. If the front and middle coxae arc seized in 

 two pairs of pincers and pulled apart, the pro- and mesothorax part company. 

 Similarly by tearing the middle eoxse away from the hind coxse the mesothorax can 

 be separated from the metathorax. The so-called propodeum, though theoretically 

 an abdominal segment, is so firmly attached to the metathorax, that when the 

 abdomen is bi-oken off (e. _i7.,in a dried specimen by pushing it roughly downwards) 

 the propodeum always remains with the thorax. 



In fig. 1, h, h, the " cenchri," are two singular organs with some resemblance 

 to little tegulse. They are always present in Tent hred in idee, but I cannot find that 

 their function lias as yet been discovered. They belong to the metathorax, and 

 mark its base. 



The structui'e of the abdomen in Hi/menoptera generally has of late been 

 thoroughly investigated afresh by Dr. E. Zander (Zeitschr. f. wissenschaftliche 

 Zoologie, Leipzig, 1899 and 1900). In both sexes of T. mesomela the first five seg- 

 ments are quite simple, each consisting of a dorsal and a ventral plate, the former 

 considerably overlapping the latter, and forming together with it a regularly 

 " annulus." Then comes in the female a (ith segment, slightly differing from those 

 preceding, in that the ventral plate is smaller, more overlapped by the dorsal, and 

 with its apical margin excised on each side of a central projection (sometimes called 

 the "hypopygium "). Then — I still speak of the ? — we have a 7th dorsal plate of 

 normal form, but the ventral plate is reduced to a narrow wire-like strip or pair of 

 strips, to which are attached the " saws." The 8th dorsal plate again is nearly 

 normal in form, except that its apical margin is emarginate, and here it is not, as in 



