334 MORPHID^. By H. Fruhstorfer. 



The larvae live in nests on different forest -trees and especially on the climbing plants, but attack 

 one another. There are about five moults. Larva cylindrical rather slender, somewhat thickened in the middle, 

 tapering posteriorly. The last segment terminates in an indistinct tail-fork. The head is comparatively 

 large and bears a pair of horizontal, conical processes, directed anteriorly, which are sometimes only rudimen- 

 tary. Colour always bright, sometimes variegated, yellowish with red-brown dorsal stripes or cross-shaped 

 figures, back with long subdorsal tufts of bristles, of which the middle and posterior are in some cases gaily 

 coloured. Pupa similar to that of the genus Ant/ithusia but more oval. Head with two tubercles, wing-cases 

 distinctly projecting, abdomen sometimes belted with yellow ; colour green or yellowish. Pupa on twigs or leaves 

 attached by the abdomen but hanging free. The pupal stage lasts 20 — 30 days. 



The Morphids, especially the entirely blue species, form on account of their brilliance a characteristic 

 of the South American landscapes. They strike every natiu-ahst and mention of them recurs in works of travel 

 with the same stereotyped regularity as the description of the parrot's flight or the abundance of the humming- 

 birds. 



The author of these lines had the pleasure of himself observing Morpliids in the woods of Sta. Catharina, 

 but especially in a lonely side valley of the C'apivary River, which at that time was still little explored by wliite 

 men. A crystal waterfall sprinkled the roots of the forest giants, providing a point of attraction for butter- 

 flies of all kinds. Just as wild animals tread down regular tracks in order to reach water, so the Moqjliids 

 assembled here daily, following up any chance clearings to get to the cascade, as if they came to seek cooling 

 refreshment among the softly rustling arches of the tall bamboos, which were shaken and bedewed by the foam- 

 ing water. They did not come in crowds but singly, floating along quietly. And how patiently one waited, 

 until after some minutes of silent expectation a second iridescent form appeared, to be captured with the almost 

 unfailing certainty of long practice as soon as it ventured within reach of the net. 



The magic impression which the Morphids make on the European traveller seems to be shared also by 

 the natives. Dr. Koch tells us (Archiv fiir Anthropologic; Brunswick, 1906, V, part 4, p. 295) that Morpho 

 menelaus is chosen as a favourite model for the dance-masks of the red-skins on the Upper Rio Negro and Ya- 

 pura Uaupes. 



The habits of the separate species differ very much. Some, such as (ichilles and deidainia, fly unsteadily 

 at the edges of woods or along the forest-paths which intersect the tropical forests and commonly settle on leaves 

 or damp places on the ground or on the river-banks, also on fallen fruit. One species (cateyiarius) prefers the 

 underwood, is even found near human habitations and can be baited in I'ather large numbers with hung-up 

 bananas. The species of the hecuba, menelaus and rhetenor groups, however, avoid the ground; they never 

 leave the timber-forests and soar round the tops of the highest trees without appreciable movement of the 

 wings. In Ecuador it has been observed that the species of the M. menelaus group, which were flying along a 

 small wooded branch-valley, stopped short on the bank of a broad river and flew back into the shade, not ven- 

 turing even to cross a short open space in order to reach the other wooded bank. 



The range of the Morphids extends over the whole of tropical and a part of subtropical America, from 

 Mexico as far as Southern Brazil and Northern Argentina. The Amazon is its headquarters, particularly the 

 streams which run down the eastern Cordilleras of Ecuador and Peru and form the sources of that river. On the 

 Antilles no Morpho has yet been found, except on Trinidad, wliich in other respects also can scarcely be separated 

 zoogeographically from Venezuela. Concerning the vertical distribution reliable information is wanting, but it 

 is certain that some species occur up to about 2000 m. The alpine regions are avoided by the Morphos as 

 by the Hestias or Amathusiicls, as the shape of their wings is adapted to the protection of the woods. The 

 period of flight of the Moqiliids is in general a very short one and it may be assumed that, except in the pure 

 tropics, they everywhere occur (as in southern Brazil, where the author has been able to observe them himself) 

 only in the height of the tropical summer, thus in the most favourable time for the development of the butter- 

 fly life. In consequence of their not being on the wing all the year round, like many Nymphalids and Pierids, 

 they are only in quite rare cases subject to the influence of climatic contrasts and seasonal variation is only 

 known with certainty in two species (perseus and portis Hbn.). On the other hand almost all the species 

 are subject to a liigh degree of geographical modification, forming separate races whose identity with the 

 collective species must often remain doubtful, especially at the extreme limit of their range, and one may assume 

 that in the Amazon region in particular a distinct local form is developed on every important tributar3^ This 

 phenomenon recurs in the different Brazilian provinces and it is astonishing that it was left to me to notice 

 and to establish this fact. 



The examples figured are from the H. Fruhstorfer collection at Greneva, which also contains the types 

 of the newly described forms, except where the collection in the Zoological Museum at Berlin is mentioned as 

 the source. Dr. K. Jordan was kind enough to send some types from the Felder collection in the Tring Mu- 

 seum for comparison, with the help of which I have been able to correct some long-standing errors, espe- 



