414 S. H. SCUDDER ON THE STRUCTURE AND 



mediately i^recetle them and indicate much more truly the real affinities of the group. Kirby, 

 in his recent catalogue, repeats the error of Boisduval but in a different manner, placing the 

 group at the close of the Lycaenidae and immediately preceding the Pieridag. 



Bates remarks, in the Journal of Entomology, that Eumaeus differs from all Lyca?nidae 

 in the closure of the cell by perfect nervules and adds in a note that " if the lower disco- 

 cellular nervule prove to be aborted constantly in the numerous aberrant genera of Lycre- 

 nidje of Eastern Asia, the genus Eumjeus miglit form a suljfamily of Lycosnidae, founded on 

 the closure of the wing cells." ^ Later, however, in the same Journal,^ he entirely omits 

 mention of the closure of the wing cells in his characters of the family Lyca^nidos and we 

 shall see, further on, that in Eumoeus the nervures at the extremity of the cell are not all 

 "perfect." 



The whole structure of these insects, whether we notice the high and narrow front of the 

 head, the antennae inserted in distinct pits, so far separated as to infringe upon the eyes, 

 the form and relation of the parts of the thorax, the shape and neuration of the wings, or 

 the diverse development of the front legs in the two sexes, shows that they belong to the 

 great family of Rurales, including the Vestales or Erycinidae of modern authors and the 

 Ephori or Lycaenidae of recent times. Their affinity with the Oreades is wholly imaginary, no 

 common bond of union existing between them, other than that borne by all Rurales. What 

 their position within the group may be is, as we have seen above, a matter of very diverse 

 opinion ; nearly all those authors who have given proof that they have paid close attention 

 to the structure of the genus agree, however, in placing it among tlie Ephori, in close prox- 

 imity to the Vestales. This is undoubtedly its correct general position, although whether 

 it should be classed with the one subfamily or the other, or form a third subfamily, mid- 

 way between them, will admit of discussion. We will first examine in detail the structural 

 features of the genus. 



The body of the perfect insect is somewhat robust, slightly compressed, the abdomen 

 ample, a little compressed and dorsally keeled. The head is of moderate size, thickly covered 

 with erect scales, which are short and form an even nap on the front, longer and massing 

 into irregular oljlique tufts on the summit, and into a higher, straight, transverse tuft poste- 

 riorly. Eyes of moderate size, not very full, naked, infringed upon by the posteriorly ex- 

 panding occiput. Antennae twice as long as the thorax, and less than half as long as the 

 forewings, composed of about thirty-eight joints, slender beyond the basal joints, gradually 

 expanding throughout, but more noticeably a little beyond the middle, to form the club 

 which again tapers at the tip, including the last five or six joints, to a bluntly rounded apex ; 

 the scaly covering of the stalk and club is concolorous, and the iinier naked surfiice of the 

 club is provided with a few very short fine bristles on each joint. Labial palpi horizontally 

 extended, very slender, as long as the breadth of the head, clothed with closely appressed 

 scales, the middle joint nearly equal throughout, scarcely curved upward, the apical joint 

 a little more than one third ( 5 ) or nearly one-half ( ? ) the length of the middle joint, 

 slender, pointed. Tongue as long as the hind tibife and tarsi combined. 



Wings (PI. XIV, fig. 14) large, ample, well rounded, entire, alike in form in the two sexes, 

 excepting the little greater fullness of the subcostal portion of the outer margin of the hind 

 wings in the female. Fore wings : costal margin rather strongly arched next the base, he- 



' Journ. Ent., I, 220. 2 Journ. Ent., 11, 1 76. 



