10 



Rev. T. A. Marshall's monograph of 



(15) 16. Abdomen sessile, almost always rugose, 

 except the apex ; terebra concealed or 

 short ; thyridia of segments 2 — 3 visible ; 

 large robust species 9. Rhogadides. 



II. Cryptogastres, Wesm. 

 (2) 1. Fore wings with 2 cubital areolets; belly 

 concave from end to end, with sharply 

 defined edges 10. Sigalphides. 



(1) 2. Fore wings with 3 cubital areolets ; belly 



concave ; edges rellexed ; body rugose 11. Clwlonides. 



III. Areolarii, Wesm. 



(2) 1. Mesothoracic sutures invisible; radial 



areolet ample, reaching the apex of the 



wing 12. Microgasterides. 



(1) 2. Mesothoracic sutures distinct ; radial 

 areolet minute, remote from the apex 

 of the wing \Z. Agathididcs. 



EnpJiorides. 

 Perilitides. 



17. 



18. 



Blacides. 

 Liophronides. 



(13) 14. 



(10) 15. 

 (15) 16. 



IV. PoLYMORPHi,* Wesm. 

 Abdomen petiolated. 



Fore wings with 2 cubital areolets . . 14. 



Fore wings with 3 cubital areolets . . 15. 



Abdomen sessile or subsessile. 

 Fore wings with 2 cubital areolets. 



Axillary areolet of the fore wings closed 16. Calyptides 

 Axillary areolet of the fore wings open. 

 Eadius straight, its first abscissa long, 



distinct ; terebra J straight . . 

 Badius curved, its first abscissa short, 



subobsolete ; terebra $ decurved 

 Fore wings with 3 cubital areolets. 

 Eadial areolet very short, the metacarpus 



not longer than the stigma . . . . 19. Ichneutides. 

 Eadial areolet elongate, the metacarpus 



longer than the stigma. 

 Hind femora incrassated, sometimes 



toothed ; head subquadrate, vertex ex- 

 cavated, the fovea containing the 



middle ocellus 20. ndmntides. 



Hind femora simple ; head transverse, 



vertex not or hardly excavated, the 



middle ocellus not placed in a fovea. 

 Abdomen linear, longer than the thorax 21. Macrocentrides. 

 Abdomen ovate, not longer than the 



thorax. 



- Wesmael's division PolymorxM, containing heterogeneous 

 groups, has rightly been objected to as unscientific. Its use liere 

 is merely to indicate the subfamilies which do not enter into any 

 of the other divisions. 



