THYMELICTN.E. 91 



slender, and curving over the vertex (a character never found in the Hesperiinae) ; 

 in most of the other genera the third joint is minute, only very rarely being 

 horizontally porrected, and, when this is the case, it is always stout. Forewing : 

 cell invariably less than two-thirds the length of costa ; vein 5 curves downwards 

 at its base and consequently arises considerably nearer to 4 than to 6 ; the middle 

 discocellular being considerably longer than the lower one, frequently more than 

 twice as long as it. Hindwing usually rather elongate, but never with a con- 

 spicuous tail or tooth ; vein 5 very rarely developed. The male is frequently 

 furnished with a discal stigma on the forewing and never with a costal fold. Both 

 pairs of spurs are invariably present. 



When absolutely at rest, the butterflies hold the wings erect over 

 the back, but, when sunning themselves, they have a peculiar habit of 

 elevating their forewings and depressing the hindwings, in contra- 

 distinction to the Hesperiids proper, which sun themselves with the 

 wings fully expanded. 



Subfamily : THYMELiciNiE. 

 Tribe : THyMEucrm. 



A thorough knowledge of the egg-stage of the Urbicolids will have 

 to be obtained before the limits of this group can be determined. The 

 whole of the butterfly stirps has hitherto been defined as having an 

 upright egg, i.e., an egg with a circular transverse (horizontal) section, 

 and the micropylar axis perpendicular to the plane of attachment. 

 Careful examination of the eggs of Thymelicus acteon, Adopaea fiava 

 (than mas) and A. lineola, has led to the discovery that they are, in almost 

 all their essential characters, butterfly eggs in the strict sense, but they 

 have three axes of different lengths, so that the horizontal section of 

 the egg is not circular, whilst, in agreement with the true butterfly 

 egg, one finds that the micropylar axis is at right angles to the plane 

 of attachment. This suggests most strongly that the Thymelicines 

 may be, at least, in this respect, a specialised (and not primitive)" form 

 of the Urbicolids, and it affords a sure test by which they may be 

 separated from the Urbicolines and Cyclopidines, both of which have 

 true upright butterfly eggs with smooth unribbed shells. 



Hiibner's coitus Thymelici is defined as having — 



The wings almost entirely yellowish and unspotted — Thymelicus actaeon, 

 Esp., Pap. 36, 4 ; Hiibn., Pap. 488-490 ; T. pus tula, T. vibex, T. venula, Hiibn., 

 Pap. 665-669; T. virgula, Hiibn., Pap. 660-663; T. vitellius ; T. lima, Schiff., 

 Verz., Pap. A. 5 (thaumas, Esp., Pap. 36, 2-3 ; Hiibn., Pap. 485-487) ; T.puer. 



Speyer diagnoses {Can. Ent., x., p. 151) the Thymelicids as having: 

 The antennae half as long as the forewings, with elongate, ovoid, conically- 

 tipped club. Apical joint of the palpi nearly erect, moderately long and slender, 

 subulate, hidden to beyond its middle by the long stiff hairy clothing of the middle 

 joint. Middle tibiae with a longitudinal series of short spines. Hindwings some- 

 what produced at the inner angle. Male with a discoidal stigma, without a costal 

 fold, and without a tibial tuft. 



Speyer further notes that the tribe differs from the Pamphilidi ( = 

 Urbicolidi) in the slender, subulate apical joint of the palpi, and in the 

 absence of the hooklet on the end of the antennal club. The club, 

 however, is somewhat elongate, and thus differs from the Urbicolines 

 in which the tip is acuminate and the club usually short and stout 

 and with a short terminal crook. Kuril further also points out 

 (Pal. Gross- Schmett., p. 636) that the apical joint of the palpi is not 



* Chapman observes (in litt.) in connection with this point that, whilst " an 

 egg may vary the length of an axis, it cannot move its micropyle from tne end to 

 the top, whilst retaining the primitive form." 



