76 
MEMOIES OF THE I^ATIONAL ACADEMY OF SCIENCES. 
- mx.p 
Its relations to tlie Cossidfe, including the Zeuzerina^, remain still to be elaborated; they are 
rather close, yet the Tortricoid affinities are very apparent, and need further examination. The 
j)upa of Zeuzera 'pyrina (fig. 40) is of the same character as iu Prionoxystus, but the maxillary 
palpi are larger, tlie lateral palpi more reduced, while the cell-breaker is very long, being much 
more developed. 
Fig. 30 shows the front of the head and maxilhe of the Oossid, Prionoxystus ro&iViffc, which is 
more Tortricid than IJepialid; paraclypeal iiieces; mx, maxillary palpi; labial palpi; 
mx„ maxilhe. Fig, 40 represents the head and end of the body of Zeuzera pyrina. 
Eemarl<s on the Coehliopodkla :. — Chapman removes this group from 
the Bombyces after a study of their larval and pupal characters. We 
should, after studying the pupie of five or six genera, agree with his 
suggestion that this and the family ^Megalopygidie (Lagoidm) should be 
removed from the Bombyces and jilaced near the Tineoids, from which 
they have undoubtedly descended. That the line of descent, however, 
was directly, as Chapman suggests, from the Eriocephalidie seems to us 
a matter of doubt. The larvje of the Cochliopodids present some 
notable differences from that of Eriocephala, wiiose so-called “eight 
pairs of abdominal legs’^ appear to be merely spine-bearing tubercles. 
Although the head of Eriocephala is partially retractile, this adaptation 
may have no phylogenetic siguifteance. 
Fig. 35 represents the front of the head of Parasa chloris, showing 
the maxillary palpi, and a lateral process (p ) connected with it, wiiich 
1 have not seen in any other pupm, and 
may be internal. 1 have also observed 
it in the cast i)upal skin of Tortricidea 
testacea. The maxilhe are either shorter 
or no longer than the large labial x>alpi. 
The paraclypeal tubercles are well devel- 
oped iu this grouj). If we compare the 
head of the pupa of liirasa and of other 
genera, especially Liniacodes and Hetero- 
genea, with that of Tinea, there will be 
observed a close resemblance, esj)ecialiy 
in the maxilhe, maxillary palx)i, and labial 
X»alpi, indicating the more or less direct 
descent of the family from some tineid 
form, perhax)s an extinct ally of Xepticula, 
since Chapman speaks of “a resemblance 
that is almost identity iu the pupa” of Xepticula as compared wdth that of Limacodes. 
Remarks on the Me(jalopyijida\--Tho^ genus Megalopyge (Lagoa) is remarkable for the shape of 
the pupa, which is somew'hat as in Cochliopodidm, confirming the view that the two families are 
allied, though still presenting some notable difierences in larval characters. Fig. 3G represents the 
pupal features as seen in the front of the head of a Megalopyge from Florida (probably Jf. crispata 
ov operciilaris). The maxillm seem to be aborted; on each side of the second maxillary (labial) 
X)alj)i, under the ej^e, are the first maxillary palpi, whose structure needs further examination. 
Very different is the head of an allied Mexican species, Lagoa superha (fig. 37), iu w^hich the 
second maxilhe (labium) are well marked, though the palx^i are only represented by two short 
lobes. Here the maxilhe are present, and the maxillary x^alj)! are rex)resented by a large lateral 
irregular round piece. 
The next series of families begins with the Tortricidcc^ from which may have descended the 
Cosskhe. As will be seen by comparing fig. 38 of the pupa of Tortrix rileyana with that of the- 
Cossidic (fig. o9, head and mouth i)arts of the x^upa of Prionoxystus rohinioi)^ Dr, Chax)mau’s 
oxnniou that Cossus has “no character at any stage to distinguish it from Tortrices” is well 
sustained. The l^ux>al characters of Zeuzera pyrina (fig. 40) also show that it belongs to the same 
VIII 
Fig. 38.— Pupa of Tortrix rileyana, Si $ . enil of body, with cremaster. 
