842 
SPHINGIDAE. General Topics. By Dr. A. Seitz. 
bright lights, we are met here with a fact that is just as curious and puzzling as the almost regular non- 
appearance of lepidoptera in the moonshine. 
The larvae of the Sphingidae, nearly all of which number among the most conspicuous insects, are 
also better known than those of the other families of Heterocera. Their shape has been described at large 
already in the volumes 2, 10, and 14, and we may therefore confine ourselves here to our knowledge of the 
American forms. Numerous forms parallel to the Indian Sphingid larvae are to be found in nearly all the 
subfamilies or tribes connecting the west with the eastern continents. They often resemble each other to such 
a degree that they exhibit the close relationship to species of the eastern hemisphere more distinctly than 
the imagines developing from them. Thus the larvae of Herse cingulata differ less from some of the palaearctic 
H. convolvuli than the latter do among their own forms. Protoparce rustica and jlorestan, excepting the granu¬ 
lated caudal horn and the shape of the oblique stripes meeting on the dorsum, are exactly like the Eurasian 
larvae of Ackerontia; it is easily understood that also the larvae of Smerinthus cerisyi resemble most exactly 
those of the Eurasian Sm. ocellatus. 
The neotropical region, however, also harbours numerous Sphingidae showing hardly any relations 
to the palaearctic members of this family, and the larvae of these forms also exhibit an appearance unparalleled 
in the Old World. Details about this will be found in dealing with the various species on the following pages; 
we merely mention here a few examples being of particular interest for their biological relations. So for in¬ 
stance there are Sphingid larvae in America showing the Geometric! manner of sitting, such as Leucorhampha 
ornatus. The resting insect holds itself with its claspers, on the one hand, and sometimes also besides with 
its pectoral legs, on the other hand, whilst the abdominal legs are drawn in; only in creeping the pectoral 
legs, especially the hindermost pair, are brought into action. If the larva is alarmed in this position, the pro¬ 
tective attitude changes into a frightening attitude, the larva copying a snake’s head. 
The way the larta accomplishes this deception is so strange that it seems to make any other explanation 
of its behaviour in cases of disturbance impossible. The larva of Leuc. ornatus, during its Geometrid-like 
sitting position, exhibits an absolute protective colour; greenish or bark-grey, in the grown-up stage it represents 
a twig of a finger’s length; the entire lack of feet — these being drawn in so far that they cannot be 
seen — very effectively intensifies the disguise into a lifeless piece of wood. On being discovered the larva 
persistently remains in this condition, also when it is gently touched; but as soon as it is more energetically 
disturbed, the insect changes its policy by erecting the front part of its body and turning it so that its under¬ 
side is uppermost. In doing so it exhibits the exact image of a snake by a dark dorsal stripe (which extends 
across the ventral side of the larva), a broadly inflated occiput, the larva’s head becoming pointed like a snout. 
The smoothly appressed pectoral feet copy part of the shielded scaly skin of a snake; two large squinting eyes 
on the underside of the 3rd or 4th segment, supported by the darting and wriggling motions of the raised 
larval body may even frighten sensible beings, because the swaying to and fro of the front body resembles 
the biting motions of venemous snakes and looks rather dangerous. In addition to this some specimens — most 
distinctly so the light greyish-green individuals — display an exact image of a transverse snake’s mouth, i. e. 
a blackish transverse streak across the larva’s head and the sides of the first segment. We do not even ex¬ 
aggerate in saying that the larva is able to change this “countenance”, so that it “makes grimaces” to a 
certain degree. The Leucorhamj)ha are very common, and when for instance one is searching the Apocynaceae 
on Mt. Corcovado near Rio de Janeiro, one will soon discover full-grown specimens, and nobody should then 
miss the opportunity of forcing the larva to show its tricks by disquieting it. The larva of Leucorkampka is 
indeed “the real arum”, as the Rev. Miles Moss says in his excellent work on the Sphingidae, and “one must 
have studied the wonder in the live insect in order that one may fully appreciate it”. 
Moreover, it must be considered that the larva only develops this disguise while it is growing. The 
larva of Leucorh. ornatus in its early stage looks somewhat like a young larva of Smerinthus. Its colour is 
green; the whole insect with its long needle-shaped caudal horn resembles the stalk of a leaf. On attaining the 
approximate thickness of a quill it copies a broken off branchlet, turns brown, and only when it has grown 
to the dimensions necessary for imitating a snake it assumes the snake-marking needed for the mimicry, 
when it promptly loses the anal horn which would interfere with such a disguise. 
A most remarkable fact is that the larvae of each of the Sphingid species choose a special kind of 
disguise and are independent from each other in the development of their frightening images. The most closely 
allied to Leucorhampha mentioned above is Madoryx pluto Or., the imago of which differs but very little from 
the somewhat larger Leucorh. triptolemus. Its larva in its early stage likewise resembles the young larva of 
Leucorhampha. and also later on it exhibits a resemblance to a twig and loses its anal horn, yet it inflates its 
anterior body in a cpiite different way forming a round knob; on being alarmed it does not turn itself round 
