70 



MExMoms OF THE :n^atioxal academy op sciei^ces. 



Fig. 29 represents tbe pupa of Thyridopferyx ephemercvformix, and its close resemblance to that 

 of Oncopera intricata (fig. 33) will be seen in the presence of the large piece between the base of 

 the maxillary palpi. In (Eceticm abbotii (fig. 2S) the maxillary palpi are separated by the second 

 maxillary (labial) palpi. The former (mx. p.) is subdivided 

 into an inner and an outer small lobe. In the rsychida^ the 

 paraclypeal pieces, or tubercles, as we might call them, are 

 always present. They are convex and veiy rugose. The 

 labial or second maxillary piece in the Australian Eumetopa 

 ignohilis is of the same shape and sculpturing as in I'nyclie 

 gramincUd, but the large, round, rugose pieces on each side, 

 or first maxillary palpi, are single, not divided into two 

 parts, unless the irregularly trapezoidal jiicces between the 

 maxillary palpi and the eyepiece be the Lomologue of tlie 

 outer ijortion. 



In the Australian Metura ilongafa (fig. 30) the short 



reduced labial i)alpi are much as 



in Psyche (iramincUti, but are more 



deeply divided. The two divisions 



I am inclined to consider as the 



second maxillary (labial) palpi. In 



this genus the first maxillary palpi 



are also as in Psyche gyamhwUn. 

 It will then be seen that in the 



l»upa of this family the first and 



second maxillary palpi vary very 



much in form, as they probably do 



in the imagines, being more or less 



atro])hied in the latter, where they 



neetl to be carefully exaiiuned. On 



tlie other hand, the maxilhe them- 

 selves (lor in their pujial condition 



inhaustellate Lepidoptera they have 



retained the separated condition of 



those of the laciniate Lepidoptera), though short, are quite persistent in 

 form. 



The pupa of Phitmceticus gloverii differs from that of CEceticus abbotii 

 in the undivided first maxillary palpus (eyepiece) and the elongated 

 second maxilhe, as well as the narrower clypeal region, and the lack of 

 a cocoon or case-opener. 



By an examination of the figures it will be seen that 

 the outer division of the eyepiece varies much in size. 

 This is due to the varying width of the male antenna-, 

 which, when wide, as in Pinara (Entometa), Metrua, Thy- 

 ridopterys, and Pysche overlap and nearly conceal it, 

 while it is entirely hidden in PlataM'cticus. On the other 

 hand, in male pupa^ of llepialus and Oncopera, where the 

 antennae are small, narrow, and not pectinated, these 

 pieces are large. The end of the body has no cremaster, 

 but, what is unique, a hook arising from each vestigial 

 anal leg. 



Finally, it will bereaddy seen thatfrom an examination 

 ■of the pupa- the views of Speyer, of Chapman, and of Comstock, as to the position of the I\sychidiB 

 is fully confirmed, while I should go a little further and place them still nearer the Hepialida?. 

 They are, however, still more modified than this last-named group, since the females are wingless 



ahbotii; f'h, cocoon- 



TlG.29.— Pupa, nf Thy ridopteryx ephemercvforinis, cf ; 

 A, side view of end of body, sbowiDg oue of the two 

 terminal hooks; vestiges of 3 pairs of abdominal legs- 



