From whatever direction the trip be undertaken, the ascent of the volcano must be made over some one of those many immense ridges sloping from the summit to the sea west, north, and east,—like buttresses eight to ten miles long,—formed by ancient lava-torrents. Down the deep gorges between them the cloud-fed rivers run,—receiving as they descend the waters of countless smaller streams gushing from either side of the ridge. There are also cold springs,—one of which furnishes St. Pierre with her Eau-de-Gouyave (guava-water), which is always sweet, clear, and cool in the very hottest weather. But the water of almost every one of the seventy-five principal rivers of Martinique is cool and clear and sweet. And these rivers are curious in their way. Their average fall has been estimated at nine inches to every six feet;—many are cataracts;—the Rivière de Case-Navire has a fall of nearly 150 feet to every fifty yards of its upper course. Naturally these streams cut for themselves channels of immense depth. Where they flow through forests and between monies, their banks vary from 1200 to 1600 feet high,—so as to render their beds inaccessible; and many enter the sea through a channel of rock with perpendicular walls from 150 to 200 feet high. Their waters are necessarily shallow in normal weather; but during rainstorms they become torrents thunderous and terrific beyond description. In order to comprehend their sudden swelling, one must know what tropical rain is. Col. Boyer Peyreleau, in 1823, estimated the annual rainfall in these colonies at 150 inches on the coast, to 350 on the mountains,—while the annual fall at Paris was only eighteen inches. The character of such rain is totally different from that of rain in the temperate zone: the drops are enormous, heavy like hailstones,—one will spatter over the circumference of a saucer!—and the shower roars so that people cannot hear each other speak without shouting. When there is a true storm, no roofing seems able to shut out the cataract; the best-built houses leak in all directions; and objects but a short distance off become invisible behind the heavy curtain of water. The ravages of such rain may be imagined! Roads are cut away in an hour; trees are overthrown as if blown down;—for there are few West Indian trees which plunge their roots even as low as two feet; they merely extend them over a large diameter; and isolated trees will actually slide under rain. The swelling of rivers is so sudden that washer-women at work in the Roxelane and other streams have been swept away and drowned without the least warning of their danger; the shower occurring seven or eight miles off.

Most of these rivers are well stocked with fish, of which the tétart, banane, loche, and dormeur are the principal varieties. The tétart (best of all) and the loche climb the torrents to the height of 2500 and even 3000 feet: they have a kind of pneumatic sucker, which enables them to cling to rocks. Under stones in the lower basins crawfish of the most extraordinary size are taken; some will measure thirty-six inches from claw to tail. And at all the river-mouths, during July and August, are caught vast numbers of titiri[35],—tiny white fish, of which a thousand might be put into one teacup. They are delicious when served in oil,—infinitely more delicate than the sardine. Some regard them as a particular species: others believe them to be only the fry of larger fish,—as their periodical appearance and disappearance would seem to indicate. They are often swept by millions into the city of St. Pierre, with the flow of mountain-water which purifies the streets: then you will see them swarming in the gutters, fountains, and bathing-basins;—and on Saturdays, when the water is temporarily shut off to allow of the pipes being cleansed, the titiri may die in the gutters in such numbers as to make the air offensive.

The mountain-crab, celebrated for its periodical migrations, is also found at considerable heights. Its numbers appear to have been diminished extraordinarily by its consumption as an article of negro diet; but in certain islands those armies of crabs described by the old writers are still occasionally to be seen. The Père Du Tertre relates that in 1640, at St. Christophe, thirty sick emigrants, temporarily left on the beach, were attacked and devoured alive during the night by a similar species of crab. "They descended from the mountains in such multitude," he tells us, "that they were heaped higher than houses over the bodies of the poor wretches... whose bones were picked so clean that not one speck of flesh could be found upon them."...

[35]The sheet-lightnings which play during the nights of July and August are termed in creole Zéclai-tiriri, or "titiri-lightnings";—it is believed these give notice that the titiri have begun to swarm in the rivers. Among the colored population there exists an idea of some queer relation between the lightning and the birth of the little fish;—it is commonly said, "Zéclai-à ka fai yo écloré" (the lightning hatches them).

VIII

... We enter the upper belt of woods—green twilight again. There are as many lianas as ever: but they are less massive in stem;—the trees, which are stunted, stand closer together; and the web-work of roots is finer and more thickly spun. These are called the petits-bois (little woods), in contradistinction to the grands-bois, or high woods. Multitudes of balisiers, dwarf-palms, arborescent ferns, wild guavas, mingle with the lower growths on either side of the path, which has narrowed to the breadth of a wheel-rut, and is nearly concealed by protruding grasses and fern leaves. Never does the sole of the foot press upon a surface large as itself,—always the slippery backs of roots crossing at all angles, like loop-traps, over sharp fragments of volcanic rock or pumice-stone. There are abrupt descents, sudden acclivities, mud-holes, and fissures;—one grasps at the ferns on both sides to keep from falling; and some ferns are spiked sometimes on the under surface, and tear the hands. But the barefooted guides stride on rapidly, erect as ever under their loads,—chopping off with their cutlasses any branches that hang low. There are beautiful flowers here,—various unfamiliar species of lobelia;—pretty red and yellow blossoms belonging to plants which the creole physician calls Bromeliaceœ; and a plant like the Guy Lussacia of Brazil, with violet-red petals. There is an indescribable multitude of ferns,—a very museum of ferns! The doctor, who is a great woodsman, says that he never makes a trip to the hills without finding some new kind of fern; and he had already a collection of several hundred.

The route is continually growing steeper, and makes a number of turns and windings: we reach another bit of savane, where we have to walk over black-pointed stones that resemble slag;—then more petits-bois, still more dwarfed, then another opening. The naked crest of the volcano appears like a peaked precipice, dark-red, with streaks of green, over a narrow but terrific chasm on the left: we are almost on a level with the crater, but must make a long circuit to reach it, through a wilderness of stunted timber and bush. The creoles call this undergrowth razié: it is really only a prolongation of the low jungle which carpets the high forests below, with this difference, that there are fewer creepers and much more fern.... Suddenly we reach a black gap in the path about thirty inches wide—half hidden by the tangle of leaves,—La Fente. It is a volcanic fissure which divides the whole ridge, and is said to have no bottom: for fear of a possible slip, the guides insist upon holding our hands while we cross it. Happily there are no more such clefts; but there are mud-holes, snags, roots, and loose rocks beyond counting. Least disagreeable are the boubiers, in which you sink to your knees in black or gray slime. Then the path descends into open light again;—and we find ourselves at the Étang,—in the dead Crater of the Three Palmistes.

An immense pool, completely encircled by high green walls of rock, which shut out all further view, and shoot up, here and there, into cones, or rise into queer lofty humps and knobs. One of these elevations at the opposite side has almost the shape of a blunt horn: it is the Morne de la Croix. The scenery is at once imposing and sinister: the shapes towering above the lake and reflected in its still surface have the weirdness of things seen in photographs of the moon. Clouds are circling above them and between them;—one descends to the water, haunts us a moment, blurring everything; then rises again. We have travelled too slow; the clouds have had time to gather.

I look in vain for the Three Palmistes which gave the crater a name: they were destroyed long ago. But there are numbers of young ones scattered through the dense ferny covering of the lake-slopes,—just showing their heads like bunches of great dark-green feathers.

—The estimate of Dr. Rufz, made in 1851, and the estimate of the last "Annuaire" regarding the circumference of the lake, are evidently both at fault. That of the "Annuaire," 150 metres, is a gross error: the writer must have meant the diameter,—following Rufz, who estimated the circumference at something over 300 paces. As we find it, the Étang, which is nearly circular, must measure 200 yards across;—perhaps it has been greatly swollen by the extraordinary rains of this summer. Our guides say that the little iron cross projecting from the water about two yards off was high and dry on the shore last season. At present there is only one narrow patch of grassy bank on which we can rest, between the water and the walls of the crater.