N SCRIBNER'S MAGAZINE for November, 1891, I gave an account of an expedition which I made with eight scientists and assistants to the northern region of the Sierra Madre Mountains, the range which extends through the whole of Mexico and may be considered as the southern prolongation of the Rockies. That expedition began in the autumn of 1890, and during the best part of the following winter we pursued our studies in this district, having the sanction of President Diaz and of the governors of the States. I found, as I expected, from preliminary investigations, that the ignorance of even intelligent Mexicans concerning the curious people who inhabit the plateaus and the barrancas or cañons of the Sierra Madre region is almost incredible. According to my most careful estimate these Tarahumari Indians number about thirty thousand. They are scattered throughout a mountainous, and, to the outside world, but little known district, many thousand square miles in area, and it is very rare that more than eight or ten families may be found living in one place. Many of them live in hillside caves.

My first expedition gave me a pretty fair notion of the physical outlines of the country and of the work to be done, should it seem wise to devote the necessary time to acquiring, by personal intercourse with the Tarahumaris, an accurate and scientific knowledge of their character and customs. For the last two years I have carried on my studies chiefly without associates, and for the last year entirely alone.

The wonderful cliff-dwellings of the Southwest, brought to notice during the last fifty years, have of late become

of absorbing interest to the intelligent American public. Amateurs as well as scientists have explored, more or less thoroughly, the cañons of the southwestern part of the United States, and have brought to light fine collections of implements used by the former inhabitants of these curious dwellings. Many excellent photographs have also been obtained. It was easy to observe, at the Columbian Exhibition at Chicago, how attractive the various sections devoted to the Cliff-Dwellers' Exhibits were to the visitors. I have, however, since my recent return from Mexico, had occasion to notice how vague and confused is the idea conveyed to the public mind by the words "cliff dwellers," and how commonly it is believed that these abodes, of which such admirable reproductions exist, are still inhabited. Perhaps the reckless writings of a traveller, lately deceased, who made living cliff-dwellers to suit the imagined want of the public, may have left some strange impressions in the minds of his readers. Whatever the reason of these wrong impressions, I will now endeavor to give here the truth about the cave-dwellers, for I have spent the best part of the last three years in exploring northern Mexico, with the cave- and cliff-dwellers especially in view. Let me say at once that I did find cave dwellers; but they are fundamentally different from certain living cliff-dwellers sketched from hearsay and imagination. Before entering upon a description of the cavedwellers I met with in the Sierra Madre, I must therefore ask the reader, for the sake of a better understanding, to forget all that he may have heard or read about living cliff-dwellers.

Cave-dwellers are found among the following tribes, counting from the north: The southern Pimas, the Tarahumaris, and the allied tribe of Huaro

early part of my explorations. Since that time, on our march southward, we found several more of those ancient communal cave - dwellings, as well as other remains of early habitations, in the form of small, square, stone houses, fortresses on the top of the mountains, etc. Ancient remains are nowhere numerous in northern Mexico, and as soon as one enters the regions inhabited by Indians, they almost disappear. Thus it is a rare thing to meet with old cave-houses; those found are always very simple and wretchedly small, and the number of houses in each cave is very limited. The caves are generally merely walled in, and the houses are one or two stories, according to the height of the cave. The building material is grit. No implements used by the builders could be discovered, but a few stone axes have been found lying on the ground, not near the caves, on the highlands, most of them of a clumsy and coarse shape.

The Indians of to-day do not take much interest in these old cave-dwellings. They attribute them to a mysterious people, the Cocoyomes, who were small of stature, did not till the soil, but ate each other and the Tarahumaris, or green herbs, and had other characteristics of the brute. At the head-waters of the Rio Fuerte I photographed several old caves, with houses that seemed of more recent origin, which the Tarahumaris told me had been built by the Tubares, a tribe now nearly extinct, with which they were constantly at war. While I have found corpses buried inside of these Tubare houses, the dead are commonly found in special caves, quite numerous throughout the Sierras, and frequently disturbed by roaming Mexican treasure-seekers, who leave few caves untouched. The people who used these burial caves seem in most cases to have been different from the present inhabitants.of the country, judging from their mode of burial and their dress.

Permit me first to try to give some idea of the physical geography of the country, its vegetation, fauna, etc. The Sierra Madre of northern Mexico, the home of the Tarahumaris and the other Indians just named, is a broad, high plateau, from six to nine thousand feet

above the level of the sea, falling rapidly down toward the west, while toward the east it gradually sinks down into the extensive lowlands of eastern Chihuahua. A few summits rise to 10,000 feet, while one of them, Cerro de Muinora, near the State of Durango, I found to be 10,450 feet, thus, no doubt, the highest in Chihuahua. There are a few llaños, but they are small. The general character of the landscape is one of small hills and valleys, sparingly watered and covered with forests of pine and oak. Along the streamlets (arroyos) which may be found in these numerous small valleys, we meet with the slender ash-trees, the young shoots and leaves of which are cooked and eaten by the Indians; farther, alders, shrubs of evonymus with its brilliant red capsules, willows, etc. Very conspicuous in the landscape everywhere is the madroña (arbutus) with its blood-red stem and branches, and its pretty, strawberrylike, edible, berries.

The Tarahumaris have names for six kinds of pine. One of these, which was first met with near Tutuaca, has a very ornamental form, on account of its slender, whiplike, hanging branches and its hanging needles, from eight to ten inches long. It grows here and there in groups at high altitudes on barren ground, and is probably a new species. The big-leaved oak-trees should also be mentioned; the leaves, which may be over ten inches long, and equally broad, are sometimes used as temporary drinking-vessels by the Indians.

Nobody can fail to observe the astonishing number of parasites and epiphytes on the trees. The yellow, round clusters growing on the branches of the oak-trees sometimes make the forests appear of a yellow hue. Lower down on the slopes of the Sierra Madre I have seen some Mexican hanging parasites, their straight, limber branches, of a fresh, dark-green color, hanging in bunches over twenty feet long. Some epiphytes, which most of the year to a casual observer look like as many tufts of hay attached to the branches, produce, during the season, extremely pretty flowers.

But flowers are not abundant in the

these barrancas. The fruits of this cactus are the best to be found in that part of the country, and the Tarahumaris have for one month a veritable Christmas feast on them.

The barrel-shaped cactus and many other kinds are eaten by the cattle, whose stomachs become so filled with spines that the Mexicans cannot make their favorite dish "menudo" (tripe) from them, but throw them away. But the Indians clean them by roasting and eat them. Fig-trees, magnoliace, the silk cotton-tree-whose roots are eaten by the natives-the chilicote (coral-tree), with its scarlet flowers, are common. Many other trees and shrubs grow luxuriantly along the river-banks or cover the rocky mountain-slopes, some of them remarkable for medicinal properties. I must mention two species of agave that grow at a considerable elevation above the bottoms of the barrancas, namely, the tsja-wee and the amole. The first is a low, ordinary-looking agave, but remarkable as being the most important of several kinds used in making an intoxicating wine. According to Indian tradition it was the first plant that God made. The other agave is called by the Mexicans amole, and is used for the same purpose as soap, its leaves, when broken and rubbed together, producing a cleansing lather. It is also employed for poisoning fish to be eaten, this poison, like so many others, having no effect upon the person who eats the fish. We are familiar with the big, flower-spikes all these agaves have. I know of nothing so astonishing as the gigantic spike that shoots upward from the comparatively small plant. Last May I came across one that I measured. It was by no means the tallest to be found, but the spike itself, without the stalk, measured 15 feet 8 inches in height. It was 70 inches in circumference at its thickest part. It seemed a pity to cut so magnificent a specimen down, but as I wanted to count the flowers, I had one of my men fell it with a few powerful blows of an axe. Counting the number of flowers, each one half as big as a man's fist, and of a brilliant yellow, upon a piece of the spine 3 inches long, I estimated the total at 24,120. As this

piece was cut out, however, from the middle and thickest part of the spike, some allowance must be made for the upper end of the plant, where the flowers were not so thick, and surely twenty thousand would be within the truth. It required two men to carry it, and as they walked they were followed by humming - birds, which fearlessly remained at work among the flowers of what they evidently considered their private garden. They might have to fly miles before finding such another.

So far as animal life goes, tracks of coons are seen everywhere at the bottom of the barrancas, while peccary, a species of pig, may be met with. Otter and fish are plentiful in the rivers, while herons, fish-hawks, and ducks are the noticeable birds. Animal life is not rich either here or on the highlands, where deer, lions, bears, rats, and many kinds of squirrels are fairly common. We found also turkeys, blackbirds, crows, green parrots, goat-suckers, and now and then the brilliant trogon. There are also many species of woodpeckers, all familiar to, and named by, the Tarahumaris.

The natives rightly count only three seasons, namely, the dry, the rainy, and the winter. The first lasts from March till June, and is very warm and windy. The rains set in as soon as the winds cease, and throughout July and August one can generally count on early thunder-storms and heavy rains in the afternoon, while the mornings are very bright. The rains do not, however, extend over a large territory, being local in character, which is very annoying to the agricultural inhabitants, who often see dark clouds rolling up, apparently full of moisture, but resulting in nothing but gusts of wind. The Tarahumari himself is not often to be deceived, for he is a remarkable prognosticator of the weather and is often consulted by the Mexicans on this point. Easterly winds bring the rain. In the winter season there are constant winds from the southeast or north, somewhat trying until you get used to them. Snow-falls in winter are by no means unknown. In Guadalupe-y-Calvo, which lies about seven thousand eight hundred feet above the level of the sea, in latitude twenty-six,