( 63 ) 



lefehuiri). Tlu' ciiilitli abiloiuiiial tergite is feuLly sinuate, or roniuled-triincate ; 

 the sternite strongly cliitinoiis, forming a glabrous postvaginal plate. The vaginal 

 aperture is covered in front by a broad chitiuous lobe of variable lengtli ; tliere 

 is a transverse ridge at the base of the lolje where the intersegmental membrane 

 joins the vaginal armature ; cliitin-iilate of seventh sternite rather sliort, rounded 

 at end, or sinuate, the distal part of the sternite memliranous or semimembranous. 



Larva: cylindrical, head rounded, not obviously narrowed above; horn simjily 

 curving downward, granulose ; segments 4 to 11 with oblique side-bands (except 

 in alliijilrKjci). Not known of many species. — Food: Solanaeeae. 



Pupa : tongue-case free, curved, the tip nearly touching the breast. 

 Ilab. America, from tlie North to the 8ou!]i, most plentiful in the tropics. 

 Thirty species. 



Attempts have been made by several authors to divide the assemblage of 

 species here united under Protoparce into a number of genera. We Jiave vainly 

 endeavoured to find relialile characters to distinguish generically sucli different- 

 looking insects as nistica, ochus, lej'eburei, hro)ifcf<,/lorestan, sexta, etc., etc. It 

 is true the species do not agree in all respects. Yes, there are even groups of 

 closer allied species in Protoparce as conceived by us ; but the differences between 

 these groups disapjjcar, the gaps are filled up by intergraduate forms, when all 

 the species are compared. Some species have a larger and more prominent head 

 than others, some have longer antennae, larger or more obviously lashed eyes, or 

 long and strong spines on the foretarsus, or possess very long sjjurs, or mesial 

 tufts of half-erect scales on the abdomen, or similar tutts at the apical edges of 

 the tergites; many have yellow abdominal patches, and many have none; there may 

 be a large pulvillns in one species, and a near relative may be without such a pad. 

 Not one of these characters can be relied upon in splitting up the genus Protoparce 

 into several definable genera, which contain at the same time the sj)ecies which 

 are nearly related to one another. It is only possible to define such " genera " 

 as Diliulia and Si/Z)/gki, if one disregards the intergraduate forms altogether. 



\Ve emphasize that the presence or absence of a pulvillns, though a very 

 convenient character for the separation of the species in a key, cannot serve as a 

 basis for the division of the genus into two natural groups. 



The genus is essentially tropical ; two species extend far into the southern 

 temperate zone {sexta, petuniae), two into the northern temperature zone {sexta, 

 (juinqmmacidatus). It represents the stock from which several genera of the 

 Nearctic region have originated. There are two very interesting jioints to be 

 observed in this respect : 



1. The similarity between the genera Atreus, C/ilaenogramma, Daremma, and 

 some species of Protoparce {r>istka,Jforcstan), in pattern and structure is so close 

 that one can scarcely fail to recognise the near relationship of these insects. 



2. The distinguishing characters of those offshoots from Protoparce — i.e. from 

 ancestral forms which would fit into Protoparce as conceived by us— are essentially 

 negative, consisting of differences which are the outcome of a progressive reduction 

 of organs. It is very interesting to follow this reduction of organs as it goes on 

 and on in the assemblage of genera placed here between Protoparce and Ellema. 

 The several series into which these genera arrange themselves demonstrate clearly, 

 by the close connection still existing between the genera of each series, that the 

 direction of development is from the forms with developed organs to those with 

 reduced organs ; that Protoparce is not the heterogenous product of a convergent 



