voured to solve. The trunk of the elephant has neither bone nor cartilage; and this constitutes the peculiarity of the mechanism, and involves the difficulty of exactly understanding its mode of operation; for we do not find any part without a bone except this, that is spontaneously protruded or prolonged, and so kept for some time*."

66

The centre of the trunk is pierced throughout by two long canals which are the prolongations of the nostrils, and which are separated one from the other by a fatty substance. about the third of an inch in thickness. In their whole course these channels are nearer the fore part of the trunk than the hind; and they preserve the same diameter almost throughout, till they come as high as the centre of the bone (os intermaxillare) in which the tusks are planted. At this point they suddenly turn, to approach the anterior surface of this bone, making a semi-circular curve. They are so compressed at this point, that unless there be a muscular action of the animal to dilate them, they operate as valves to prevent the ascent of any liquid to a higher point. Beyond this curve the canals again widen, and are curved back to approach the bony part of the nostril. The elephant, by this construction, can use the trunk as a reservoir for water, drawing the liquid up by suction to a certain point, beyond which it cannot pass. Cuvier considers that the trunk is not in itself an organ of smell, because the passage of any liquid through the canals would be incompatible with the delicacy of the membrane with which the nostrils in the head are lined. That membrane in man is sensibly affected with pain when any liquid enters the nose; and for the same reason, the sense of smell does not exist in the nostrils of those cetaceous animals that are constantly using

* Anatomical Account of the Elephant burnt, &c. p. 32.

them as a passage for water, such as the whale, that makes them operate as a jet-d'eau. The sense of smell in the elephant, according to this great comparative anatomist, is confined to that part of the nostrils which is enclosed in the bones of the head.

The muscles of the trunk give to these two canals, which we have described, whatever inflexions the animal desires. Although these muscles are of an extraordinary number, they may be reduced to two principal classes-those which form the body or the interior part of the trunk, and those which encompass it. These latter are all, more or less, longitudinalthat is, they begin at the circumference of the base of the trunk, and are prolonged, more or less directly, towards its point. The other class of muscles are all transverse; and cut the axis of the trunk in various directions. It may assist this description, to append a copy of the sections of the trunk given by Cuvier*.

The longitudinal muscles are divided into anterior, posterior, and lateral. The first, which are affixed to the frontal bone, form an innumerable multitude of bundles, which all descend parallel to each other, and which are alternately contracted by tendinous intersections placed at short intervals. The second, originating in the intermaxillary bones, form two beds, divided one from the other into a great number of little bundles, whose direction is oblique. The lateral muscles form two pair, which are in some degree analogous to the muscle of the upper lip.

The use of these different longitudinal muscles is sufficiently evident. When they are moved altogether, the whole of the trunk is shortened. When those on one side only are moved, the trunk is bent on that side. But further, the division of these muscles, and the tendinous intersections of the anterior class, * Anat. Comp. vol. v. pl. xxix.

Sections of the trunk of the Elephant.

A. Horizontal section, in which we see the small transverse muscles cut-some (a) across; others (b) in their length.

B. Vertical section in length, which has divided the nasal canal of the left side. The small transverse muscles which are seen in their length at b, are cut across at c;-other small muscles of the same kind are seen at their length at d. We see in their length, at e, the antagonists of these transverse muscles-that is, the small longitudinal muscles.

C. Vertical section across. The small transverse muscles are seen in their length. They have various directions, not precisely radiating from the axis to the circumference, though their course is always across. They are all within the bed of the small longitudinal muscles which the section has divided across, The principal nerves and blood-vessels are also shewn in this section; as also the two canals of the trunk.

enable the animal, at his pleasure, to shorten or to bend certain portions only of his trunk, while the rest remains prolonged, or even bent in a contrary direction. In consequence there is no sort of curve, says Cuvier, which the animal, by their means, cannot give to this instrument.

The small muscles which form the interior of the trunk, are all very distinct one from the other; and are all terminated by slender tendons, of which some cross the beds of the longitudinal muscles, to be attached to the exterior membrane which covers all the trunk, and others are planted in the membrane of the canals of the trunk. Perrault considered that all these radiated from the canals to the circumference, and that they diminished the diameter of the exterior envelope, without diminishing the diameter of the canal. This, however, is not the case: two sets of the muscles have this effect, but the other set certainly lessen the diameter of the canal, although their action can never shut the nostrils. All the transverse muscles are plunged in a cellular tissue, uniformly filled with fat. They are the antagonists of the longitudinal muscles. Cuvier adds that in their contraction they compel the trunk to elongate entirely, or in part; for their separations enable the animal to exercise them exactly within the limits which he desires. In the preceding paragraph, which is also translated from Cuvier, he assigns to the longitudinal muscles the power of shortening or of bending the trunk entirely, or in part, and he says that "there is no sort of curve which the animal, by this means, cannot give to this instrument." We must bear in mind the difference between contraction and elongation. When the longitudinal muscles are at rest, the trunk remains extended to its natural length; when they are in use it is bent or contracted. But there must be another power

to elongate the instrument with force and precision, after it is wholly or partially contracted. We must discover a mechanism, which, without bone, or cartilaginous rings, enables the animal "at pleasure to shoot it out, from a foot, upon any sudden occasion, to five feet long, and that with extraordinary force." This power must be found in the transverse muscles. The first object of the transverse muscles, or rather of two sets of them, is to keep the canals open, while the trunk is curved in various ways; for it is evident, if there were no such power, the passages would be shut, as is the case when we attempt to give contrary and sudden flexures to any elastic tube. Their second object is to elongate the trunk, and to assist in the direction of its movements. Being connected with the inner and outer membrane —that is, being attached to the membrane which covers the trunk, and that which covers the canalthey can readily diminish the space between the two substances, by their contraction. At the same time it is evident that, when the length of these muscles, from membrane to membrane, is diminished by their contraction, their thickness, which is in the direction of the length of the trunk, must be proportionally augmented; while, on the other hand, the thickness of the longitudinal muscles, which is in the direction of the width of the trunk, is proportionally diminished. From this formation it results that the trunk is more or less elongated, as the transverse muscles are more, and the longitudinal muscles less, employed. The limit to this extension of the trunk is, of course, the relaxation which the longitudinal muscles admit of, before they begin to re-act; and, just in the same manner, the resistance of the transverse ones is the limit to the shortening in length produced by the action of the longitudinal ones. The two classes of muscles are, therefore, called antagonist. The simplest popular view of the matter is to say that when the