192 THE entomologist's record. 



and finally, has subdorsal rows of rounded tubercles or angular j^oints 

 on some or all of the thoracic and abdominal segments. The latter 

 (Hesperid) tyj^e does not suspend itself ; has a protuberant head ; no 

 lateral head prominences ; no wing ornamentation ; wing cuticle almost 

 transj^arent ; no medio-dorsal ridge ; no subdorsal tubercles ; the 1st 

 spiracle (thoracic) with a prominent black hood on hinder margin ; the 

 antennae not reaching to the apex of the wing ; and, most important of 

 all, has no lateral movement Avhatevcr of the abdominal segments. So 

 much are (jalatea and setnele like Hesperids, that I should have been 

 easily misled into referring some Amcincan Hesjierid jjupre (Thymele 

 tityrns) sent me l)y Dr. Chapman, to the Satj'rid group of which serneJe 

 is the tyi^e. In those I have examined, I find the following gTOups : — 



f Lasiommata cegeria. 



A IV 1 Ti ) ^ \L. meqara. 

 A. Jsymi:)hahd. < "^ ■', , . 



^ ^ ' '^ rjpmcphele jamra. 



B. Hesperid. J. J 



Hipparchia hypcrantJms. 



Melanargia galatea. 

 Satyrus scmele. 



(4). I do not yet know any character on which setnele and galatea 

 can be seimrated genericall}^ but as they always have been so seiDarated, 

 I should suggest the continuance of their separation, pending further 

 observations. (3). Hyperanthus certainly has no close connection with 

 semele and galatea, it is lacking aflinities with these in every essential, 

 except that it has no conspicuous lateral head prominences. The gTound 

 colour, too, recalls galatea, but this is entirely superficial. Just as janira 

 has less developed lateral head prominences than cegeria, so hyperantlms 

 has them less developed than janira, in fact, they are so far abbreviated, 

 that they are only sej^arated from the rest of the face by a line. The 

 peculiar flatness of the face of this pupa, by which the ordinary ventral 

 part of the head is brought into line with the anterior part, and with 

 the prothorax, appears to me rather striking. The lateral movement of 

 the pupa, jnits this with the Nymphalid section. (2). Janira comes 

 close to cegeria. It suspends itself by the cremaster ; has rather less 

 develoi^ed lateral head ^prominences ; is much more slender ; the wings 

 depressed ; the ornamentation of the Avings of a decidedly different 

 type (almost identical with the form in hyperantlms). (1). JEgeria and 

 7«e^o?ra present the most perfectly Nymphalid form of the group, and in 

 this particular, are more Nymphalid than Melito'a. Their larvae, 

 however, are distinctly of the Satyrid type. 



How semele and janira have ever been put into the same genus is a 

 complete puzzle. In fact, it is difticult to explain why semele has ever 

 been located elsewhere than with galatea. A still more wide-reaching 

 fact becomes evident, from the study of the early stages of these butter- 

 flies, viz., the connection, through the group of which galatea and semele 

 are the tyi)es, between the Satyrids and Hesperids. This group has 

 purely Satyrid larva?, and evident Hesperid pupa?. It will necessitate a 

 modification in that form of classification, chiefly as yet confined to 

 American authors, where the Satyridce are placed at the top, and the 

 Hesperidm at the bottom of what is supposed to be a natural sequence 

 in the order of butterflies. If such a natural sequence be possible, it is. 

 clear that the Satyrids fall naturally between the Nymplialidm and 

 Hesperidce. 



