OF THE AMAZON VALLEY. 519 



This spocies, which is the only JtliomUi Ibimd in tlic Amazon roiiion that niii^lit lie 

 consich-nnl a ITiiinenitis, is found only at Tat)atinira, on the I't'iaivian fi-onticr. It rr- 

 senil)l('s much /. Edcssa (Uewitson, Exol. Butt. Itlioiiiin, Wix. 1*2.). a native (il'S. 1]. 15ra/.il; 

 but it is a little larger, and diU'ers considerably in the iiind-wiug' ueuratiuu of the d. lu 

 /. Edcssa there is a middle and an upper disco-cellular ncrvuh% and the u])per radial is 

 only partially aborted ; indeed, the neuration is identical witli that of /. I'lrninin and 

 /. Or'uinii ; but in Kcphcic the a])proximation to Jlt/i/wi/i/is is carried a step further; for 

 the middle and U])])er diseo-cellulars, as well as tlie upper radial, ai-i' all (piite aborted, 

 altlioui;li the lower disco-cellular and radial are not attached to the subcostal. In the ? 

 the middle diseo-eellular johis the subcostal, and there is no trace of an upper radial. 



Genus Melin.ea, nov. genus. 



The species of this distinct group were placed l)y Doul)leday in section 2 of the genus 

 Mechanilis ; but he failed to mention or misstated most of tlie ])rincipal characters of the 

 section. The genus is nearest allied to 0////'«s and Tlnivldia; in fact, it ap])roximates 

 these much more closely tlian it does Mcchauitis. From Olijras it dilTers in the fore 

 legs of the c^, in the palpi, and slightly in the wing-neuration ; iVom Tliyvidin also in 

 the fore legs of the male, in the palpi, and in theantemue. 'i'he following are its principal 

 characters. 



Pali)i short, smoothly clothed with scales, and closely ajjplird to tiie forehead; tliinl 

 joint not porrect as in 3[cchanUisM\(\. the allied genera. Antennse very long and slender. 

 Fore legs of the <? with the tibiae and tarsi more or less abbreviated, but never i-educed 

 to a rounded knob; the tibia always shorter than the femur. I'ore tarsi of the V long, 

 liliform, si)ines wid(^ a])art. Jlind-wing costal widely separated from the sidjcostal in 

 both sexes, in the c? long, reaching nearly the apex of the wing, in tlie 2 very short, 

 terminatini;- on the costa : the lower disco-cellular in l)oth sexes straight, nearly in a line 

 with the nu'dian nerviu'c ; tlie median nervure, in fact, describes a gentle curve, the lower 

 radial being placed as though it were a fourth median branch : the middle disco-cellular 

 is at right angles with th(> lower, strongly angulated in its middle, and emitting- a 

 recurrent nervule : ujjper disco-celkdar rather long, transverse, joining the subcostal at 

 about one-half the lengtli of the wing. 



The great reseml)huice in colours aiul markings between the species ol' Jfc'!iii(C'( and 

 those o\' 3f{'cli(t)i// is has led to the confounding of the two n-en(>ra ; in other words, a 

 relation of analogy has been mistaken for one of allinity, just as in the case of the two 

 subfamilies Jlclicotii/tcC' aiul ])iiii<i'iii(C, There is. however, as will be seen on comparing 

 the characters of the two genera, a wide struct ui'al dillerence in the palpi, I'ore leg.> of 

 the male, aiul neuration of the hind wind's. Some species of Mt'liiuca so nearly rescmlile 

 s])ccies of Jlt'c/i(niilis, tliat they might easily be mistaken for them. 'J'lie two an.alogons 

 forms accompany eacli other; l)ut 1 think 1 found ])roof that they are not mln/i/ci/ onv 

 to the other, in the fact that the species of the two genera do not coincide in any locality 

 on the Amazons, but vary and sescrcijate races without any nuitual specillc similarity, 

 'i'hey are very frequently accompanied by a llcUconUts assimilated to them in colours 



