114 HTMENOPTEEA. 



inside of these fields is punctured. The metasternum is hollowed on either side, 

 carinated in the centre, and finely striated. The petiole is a little longer than broad ; 

 on each side is a keel close to the edge, which is itself margined ; in the centre are two 

 narrower keels, enclosing a punctured space ; the portion enclosed by the lateral central 

 keels is smooth ; at the end of these keels is a transverse furrow. On the lower side of 

 the petiole is a curved spine; the second abdominal segment is much larger than all 

 the succeeding together ; the three following are of equal size, the next larger, the 

 following smaller but larger than the third. The anterior coxae are coarsely punctured 

 on the outer side, posterior finely punctured and striated. The posterior femora are 

 finely punctured ; beneath they are very finely and closely toothed, the teeth only 

 being visible with a strong glass ; at the base is a very indistinct tooth. At the end 

 of the stigma is a round white spot, and there is another opposite it ; there is a long 

 narrow white stripe along the costa, and the base is more or less hyaline. 



I think this is a male ; but I am not quite sure, as the apical abdominal segments 

 are retracted. If a female it would differ from H. ccerulea (which I have regarded as 

 the type of the genus) in having the ovipositor retracted. 



Subfam. EUPELMINJE. 



This subfamily possesses a well-marked peculiarity in the strongly dilated middle 

 tarsi and in the large stout spurs on the same legs. In the possession of a large, stout 

 spur on the middle tibiae the Encyrtidae agree with them, but differ otherwise in the 

 disk of the mesothorax being bordered, and in the mesothorax not being contracted 

 before the scutellum. 



Comparatively few species have been described; they are principally from the 

 Nearctic Region ; but species are known from Australia, the Sandwich Islands, North 

 and South America. The European species have been referred to six or seven genera, 

 mostly separated from each other by slight differences in the form of the legs, head, 

 and the manner in which the antennae are inserted. To arrange the. Central- American 

 species in the same lines would require the creation of several new divisions. I have 

 only done this in three cases ; for with the limited material at my command (all of the 

 species being represented by one sex only, and mostly by single specimens) I experience 

 considerable difiiculty in deciding as to what characters are of generic value, a difficulty 

 increased by the fact that such characters which, on first examination, might be regarded 

 as useful in generic separation, are found to merge into mere specific differences when 

 traced through several species. A revision of the genera of Neotropical Eupelmina 

 must therefore be deferred until the accession of more complete material shall enable 

 us to come to more definite ideas as to the limits of the genera. 



