OF WASHINGTON. 103 



forming a loop, and the outer pass over the enlarged base of 

 the sheath, and continue along the upper edge of the support 

 and unite with plate 9'. The parts are shown somewhat 

 separated at g, to better indicate their relation. The open 

 spaces between the spiculse and the support are enclosed by a 

 membrane which forms with these parts a closed passage for 

 the egg from the abdomen to the ovipositor proper. 



Another peculiar feature is the great size and abnormal 

 position of what apparently corresponds w r ith the yth ventral 

 arc of the abdomen 7'. The outer portion of this arc or plate is 

 tough, and in texture and color resembles segments 7 and 8 

 and the support 9 ; the inner portion is thinner and more 

 flexible and lighter in color, becoming membranous at the 

 point of attachment. It is attached within, and is, when the 

 ovipositor is entirely redrawn, nearly concealed by the 3rd 

 to 5th ventral arcs, as shown at a. As the ovipositor is 

 thrust out this arc is considerably drawn out, as represented 

 at b, and to thrust the ovipositor quite out, as represented at 

 c, it would seem that this sheath must necessarily be rup 

 tured ; but this is not the case with the two specimens col 

 lected having the ovipositor fully extended, and probably 

 never occurs. The explanation undoubtedly is that as the 

 ovipositor is pushed out this sheath-like plate slips back and 

 the terminal segments extend and separate at the same time 

 sufficiently to free the ovipositor. 



Nothing is known of the breeding habits of the American 

 species of Metopius; the specimens collected by me were taken 

 in the fall on Solidago bloom and in company with Trypeta 

 solidaginis, and it is possible that the Metopius is parasitic on 

 this Dipteron, the powerful ovipositor of the female being 

 well suited for piercing the Trypeta galls. The breeding 

 records of Kuropean species ascertained for me by Mr. How 

 ard do not support this belief, as they show that the rear- 

 ings have been from various Bombycids, a Noctuid, and one 

 doubtful one from Lophyrus pini. 



I have altered the characterization of the genus Metopius 

 to allow of its including this anomalous species, which I take 

 pleasure in dedicating to Prof. C. V. Riley, whose interest and 

 work in the parasitic Hymenoptera are well known. 



Metopius Gravenhorst. Head transverse, short ; face scutatiform, 

 sub-concave, with elevated margins ; eyes oval. Antennae ranging from 

 short and somewhat spindle-shaped to long and of nearly uniform thick 

 ness throughout. Thorax convex ; scutelltim quadrangular, with prom 

 inent posterior angles and carinate lateral margins. Wings medium, 

 areolet large and rhomboidal. Abdomen sessile, usually compressed 



