128 tHE entomologist's record. 



Parnassian .(grooving are interpretations of the primary tendency, much 

 more like each other than are the Parnassian and Nymphahd, and so 

 if the XyiiipJialis groove arises (as it certainly does) hy descent from 

 the same Papilin (archa^an) as J'arnassius does, then there is no 

 difficulty in Pii'ru also doing so. The Parnassian groove is an oval 

 (or irregular) depression, never (?) reaching the distal margin of the 

 segment. There is another feature in the grooving in which Picria 

 follows the Papilio-Nymphalid tendency, rather than the Erycinid, 

 and that is, that it is always well pronounced on the stalk, more so if 

 anything, than on the club, whilst in Erycinids the grooving is 

 always well developed on the club as compared with the stalk. It is 

 also the case that the grooved aspect of the Erycinids is more internal, 

 that of Pien's more ventral, when compared with each other. 



The other structure which Dr. Jordan believes to attach the Pierids 

 to the Erycinids, is the arrangement of the bristles ; he expresses his 

 conclusion to this effect strongly, but there is some weakness as to the 

 facts. " Sense bristles, specialised as in certain Erycinids, sometimes 

 obliterated. No connection with Xymphalidae and Papilionidar." It 

 does not appear, however, that the Pierid type ever does occur in 

 Erycinids. The Pierid type has two median lateral bristles and two 

 apical ventral bristles. Now, in Erycinids, the median lateral bristles 

 are a feature of the more specialised forms, but are unaccompanied by 

 apical bristles, whilst apical bristles occur ventrally in the earlier forms 

 without the median lateral ones. The earlier Erycinids and the earlier 

 Papilionids are alike in preserving the Lycaenid (or Hesperid) bristles, 

 an apical row. From this any other form may be derived — the Pierid 

 as easily as the Nymphalid, or as the irregular patterns common in the 

 higher Papilionids. The truth is, the sense bristles weigh very little either 

 way, but if anything rather in favour of connecting Pieris with Papilio. 



The primary arrangement of bristles is an apical row. There are 

 three (at least) very definite specialisations, all with a median lateral 

 pair of bristles. The Erycinid has no other, the Pierid has a ventral 

 apical pair, the Nymphalid a ventral basal pair. None of these can be 

 derived from each other, except the Erycinid from the other, but as 

 its origin is obvious enough, this suggestion does not arise ; all may 

 be derived from Hesperid, Lycaenid, lower Erycinid, or lower 

 Papilionid types. 



The basal progress of the l)ristles is an evidence of higher evolution, 

 the lower Hcsperiiilar and Pi/camidac having a row of apical bristles, 

 less fully apical in Lycaenids than in Hesperids and, therefore, com- 

 mencing a basal specialisation. The lateral bristles are nearly alike, 

 when at all specialised in Erycinids, Pierids, Papilionids, and 

 Nymph alids. 



In Papilio, the basal movement resulted in many various irregular 

 arrangements, but nowhere did a pair reach the base as inNymphalids. 

 In Erycinids the bristles became nearly median, but seemed unable to 

 get much out of a transverse row, so that the median and apical pair 

 never occur together. In Pierids and Nymphalids the ability to get 

 out of regular transverse alignment, seen in I'ajiilio, resulted in an 

 apical and median pair in /'/Vr/.s, a median and basal pair in .V//m- 

 phalis. The basal pair in Xi/iiiji/ialis is evidence of the highest attain- 

 able evolution of bristles in Rhopalocera. 



The Dismorphiine antenna deserves a word of rcnuirk. It seems 



