394 MR. NEWPORT ON THE NERVOUS SYSTEM OF THE SPHINX LIGUSTRI. 



a shady situation, where it may suspend itself perpendicularly at rest against a wall 

 or the stem of a tree, and remain unmolested during their complete expansion. This 

 occupies but a short period : in the Butterfly it seldom exceeds more than a few mi- 

 nutes ; but in the Sphinx, whose wings are larger and stronger, it is not completed, 

 and the wings fit for flight, in less than two or three hours. 



2. Of the perfect Insect. — The nervous system of the Sphinx in its perfect condition 

 offers many interesting points for consideration, although it differs but little in its 

 general arrangement from the last stage of the pupa. Thus, there is no further alter- 

 ation in the cerebral ganglia, nor in those which constitute the thoracic mass, from 

 which nerves to the organs of locomotion are distributed ; but the whole are covered 

 in by a new structure, and do not lie, as in the larva, in the open cavity of the thorax. 

 The cords and ganglia are now inclosed on each side between the muscles, from 

 which they are separated by a semi-opake membrane of a fibrous texture. This ex- 

 tends over their upper surface, and protects them from the oesophagus, which passes 

 along above them. Before the last change in the pupa, in the beginning of June, we 

 can readily trace the distribution of nerves and obliteration of ganglia, but after this 

 period there is considerable difficulty. A very remarkable change, the obliteration 

 of a ganglion in the thorax, occurs just before this period ; but it is so rapidly effected, 

 that I have never yet been able to observe it at the actual moment of its occurrence, 

 and hence am unable to state, from positive observation, whether it be the second or 

 third ganglion that disappears. A similar change occui^ in Papilio Urticce, Linn., and 

 in this insect it takes place between the forty-eighth and fifty-eighth hour after enter- 

 ing the pupa state ; but I have not been more successful in detecting it even in this 

 instance. At first we might suppose it to be the second ganglion that disappears, as 

 we have seen that a change is taking place in it. But this, I believe, is not the case, 

 because the nerves to the two pairs of wings, which originally were each formed by 

 two roots, the first pair [Plate XIII. fig. 1 . (/)] by a root from the cord between the 

 second and third ganglion, and another from the third ganglion itself, and the second 

 [fig. I. (i)] pair by one root from the fourth ganglion, and one from the cord between 

 the fourth and the third, become now united. Now were it the second ganglion that 

 becomes obliterated, the origin of the nerves to the first pair of wings would necessarily 

 be anterior to the wliole thoracic mass. But instead of there being the third ganglion 

 and cord between the origins of these nerves, as in the larva state, they are now united 

 into two roots, and arise from those portions of the thoracic mass which pass on each 

 side of the central attachment for the muscles [Plates XIII. and XIV. figg.6. and 8. {iv)] 

 of the thorax. Each root passes diagonally outwards, the first in a direction back- 

 wards [fig. 8. (a)], the second in hke manner forwards [fig. 8. (ft)], until meeting 

 each other they unite and form a small plexus [fig. 8. (c)], and then again dividing 

 are distributed, the one [fig. 6. (a)"] to the anterior, the other [fig. 6. {d)] to the pos- 

 terior pair of wings. The reason for this curious union and complexity in the distri- 

 bution of these nerves to the wings, is not at first very evident, but a little reflection 



