sometimes broken, country. The rocky portions are only exposed along the valleys and watercourses, whose perpendicular walls, now thrown down, appear only as sloping banks. The lithologic nature of the rock becomes more earthy; its fracture, sharper. It, however, frequently presents a rounded and blunt surface, particularly where there is a tendency to disintegration. The limestone embraced in the section of country lying between the mouths of the rivers San Pedro and Las Moras, like that above, seems to be metamorphic; it, however, differs from the limestone above the river San Pedro, in indications of having been subjected to the action of a higher temperature; and its cretaceous character is also proved by the occurrence of fossil remains, which form in some respects a transition from the adjacent geological zone. The surface of this geological belt is completely covered with drift and alluvial soil; and the growth of trees (consisting almost exclusively of mezquite) appears more liberally distributed. Whilst there are scarcely any trees to be seen upon the prominent points, the dells, basins, or flat valleys, where rain-water washes together and deposits the more fertile portion of the soil, are usually invested with scattered groves of the leguminous trees.

HOW WATERED.

The surface of this region is usually dry; it is, however, well watered, when compared to the country adjacent.

The road from San Antonio to El Paso del Norte crosses in this belt (about 40 miles wide) six clear and bold running streams, of which Las Moras, Piedras Pintas, Zoquete, and San Felipe are the most characteristic. They are somewhat similar in general appearance, and in all probability have their origin on a more solid but in a greatly deeper situated stratum; for they pour forth at once their crystal waters either from deep funnel-shaped basins or from rocky clefts. Several of these springs indicate a higher temperature than the water in the streams below.

The water of all these little streams, as also that of the Rio San Pedro or Devil's river, is strongly impregnated with carbonate of lime. Everything hanging within its touch, or in any way exposed to its action, becomes perfectly coated over by its calcareous deposite in a remarkably short time.

In consequence of the permanence and abundance of running water in these tributaries of the Rio Bravo, their bottom-land will in time be highly valued for agricultural purposes. It would be an easy matter to irrigate it, as the fall of water almost throughout is very considerable.

INTERSPERSED STRATA.

The groups of hills mentioned before as placed upon the table-lands of the country between Devil's river and the Pecos, appear again in this belt as belonging not only to the later strata of the cretaceous system, but also apparently to a still later date. These strata are usually met with, bordering and constituting the edges of the different valleys. The fossils occurring in these localities are also of the age just mentioned. As an essential characteristic, we cite here strata and shoals consisting almost solely of entire and fragmentary pieces of Exogyra, Arietina, (Roemer.) They appear either in a state of perfect preservation, or as a real breccia; the cement of which is mostly an ochre-colored calcareous sand or clay.

The stratification of this formation shows a succession of layers of variously tinted marls, of more or less coarsely grained sand, and also of differently colored limestone; all are profusely

impregnated and sprinkled over with oxide of magnetic iron. Some of these strata contain pieces of the latter formed into every shape, but most commonly give to the matrix the real habitus of volitic texture.

This imbedded formation increases as you approach its lower edge, (Las Moras,) leading to another change in the lithological features of the country along the Rio Bravo.

Before considering this change, however, another fact of much geological importance is not to be overlooked.

THE DYKE.

About twenty miles below Las Moras is Elm creek, (Arroyo de Los Olmos,) the next tributary of the Rio Bravo on the Texan side. Its valley belongs to still another geological belt; none of the more solid metamorphic limestones before referred to are now to be seen. The whole country from Las Moras to the mouth of the Rio San Juan, and even as far down as the vicinity of the Mexican town of Reynosa, forms another link of the cretaceous system; a more soft and brittle sandstone, (partly chloritic?) varying in grain, color, and cohesion, constitutes the main part of this formation. This resembles very much, if it is not in reality, the green sandstone or chloritic chalk itself.

Its northern limit, where it joins the more recent metamorphic limestone some distance below the Las Moras, is distinctly marked by a line which shows on the surface or in the soil signs of a geological disturbance. This limit is the valley of Elm creek, two miles and a half above Eagle Pass. It is wide and flat; the ridges of hills bordering it are often overthrown and washed down, whilst the horizontal strata in many places are brought into synclinal or anticlinal positions. The creek itself, not the one of that name on the El Paso road, is sluggish, and carries only a dirty, greenish, and brackish water, which often disappears in its bed, leaving only here and there small ponds and muddy pools. Out-crop of pretty extensive beds of lignite coal occur on both sides of the mouth of this creek well worthy of examination, and may prove to be of commercial value.

These coal-layers probably gave the name "Piedras Negras" to the Mexican military colony in the vicinity. The aspect of the valley of Elm creek, and the character of the country to the right and left of the Rio Bravo at this point, justify the idea that a subterraneous volcanic dike crosses the basin of the Rio Bravo. By turning for a moment from our course along the Rio Bravo, and proceeding from the mouth of Elm creek in a direction southwest by south for about seventy miles, we reach the foot of a high and bold mountain range formed of metalliferous limestone, (zechstein,) the precious contents of which once made Santa Rosa famous as a silvermining town.

On the line between the Rio Bravo and the Santa Rosa mountains, the face of the country shows many signs of a geological disturbance; the usually undulating region becomes more broken, whilst the flat, long-stretching ranges of hills are frequently overthrown. The slope of the Santa Rosa mountain is rocky, wildly broken, and steep, and large portions of the strata are entirely dislodged and most anomalously placed. The stratification here is not only seldom horizontal, but frequently thrown up vertically. There is some regularity, however, in this apparent disorder, particularly with respect to the parallelism-the characteristic of all the cordilleras of the American continent.

PARALLELISM AND VOLCANIC CROSS AXES OF THE MEXICAN CORDILLERAS.

By this parallelism is meant an inclination of the chief sierras to separate into collateral subsierras and side branches, which again join the main chain; or they are at least connected either by cross sierras, or even simple dykes. To this striking peculiarity the profiles of both the northern and southern portions of the western hemisphere are due. This mountain range of Santa Rosa presents another characteristic, and one, according to Alexander von Humboldt, peculiar to the mountain system of Mexico-it is, that the volcanic axes cross the direction of the Cordilleras almost always at right angles. There are many cross-valleys on the northeastern slope, proving the action of some volcanic disturbance; a valley some eight or nine miles east of Santa Rosa, called "El Potrero," is the most remarkable instance. Several mines are still worked here, where the metalliferous limestone is variously traversed by veins of feldspar and limestone spar; this latter usually accompanies the silver ore and galena. In the centre of this cul de sac, a better name for this so-called valley, can be seen an ancient crater, the inner walls of which are thickly coated over with a lava-like basalt of a dark red hue, whose composition differs apparently but little from that which covers in layers (20 or 30 feet thick) the cretaceous range of hills joining the northeastern slope of the metalliferous mountains.

It may, perhaps, be of importance to state that the argentiferous portion of the Sierra de Santa Rosa is not more than ten or eleven miles in length, commencing at a point called "El Cederal,” (the Cedars, Cedar Grove,) and terminating at a place, in a northwest direction, bearing the name of "Los Nogales," (the Walnuts, Walnut Grove.) The presence of ores, together with the trap or basaltic dykes branching out from the sierra at right angles, may prove the supposition which places here the origin of the volcanic power, that, pushing through the fissures of stratified rocks, causes the dyke before alluded to. Following this dyke from Santa Rosa up towards the Rio Bravo, it will be found to cross the valley of this river in the vicinity of Elm creek, as already stated.

BASALTIC HILLS-VOLCANIC DYKES IN TEXAS.

On the Texan side, the first marks of volcanic action are to be seen at the head of Leona river. Here a solitary hill of 60 to 70 feet in height occurs, formed entirely of a dark green basalt which is closely allied to that of the Santa Rosa mountains, and which also contains much hornblende and olivine. In the vicinity of Fort Inge, and also near the head of Las Moras, are several hills of the same nature; also the road from Leona to the first crossing of Devil's river leads over several places indicating volcanic action.

The west bank of the Rio Frio, at the crossing, is formed of a solid mass of basaltic rock, which undoubtedly belongs to the dyke alluded to as having its origin in the Santa Rosa mountain, and here crossing the cretaceous formation.

DR. ROEMER'S VIEWS.

Dr. Roemer, in his not yet translated work, entitled "Die Kreidebildungvon Texas," ("The Cretaceous Formation of Texas,") mentions (page 8) that plutonic or volcanic rocks were brought to him from between the San Saba and Cibolo; and according to him, granite, together with older stratified rock, is seen in narrow strips, surrounded with cretaceous strata, between the

San Saba and Pedernales. Again, about fifteen miles due north of Fredericksburg isolated granitic rocks have been met with, among which is the "Enchanted rock," of popular renown. Also, between the Llano and San Saba granite protrudes through cretaceous strata; and sixteen miles north of Fredericksburg occurs a coarse-grained granite, consisting of flesh-colored feldspar, gray quartz, and some little black mica.

In other places along the Llano very finely grained varieties of granite have been observed. Granite, frequently interspersed with veins and fragments of a white quartz, also appears at various points. Pieces of syenite, too, have been found along several of the tributaries of the Llano.

Besides these plutonic forms, trap-like rocks also seem to occur in many places of the country referred to by Dr. Roemer. Again, this author received from twenty miles to the northeast of San Antonio de Bexar pieces of a black basaltic rock, which protrudes in veins through the cretaceous limestone strata. In this basalt, as component parts, are many minute crystals of a white fossil, (glassy feldspar?) and also a dark, olivinish fossil.

The geographical distribution of the rocks of which Dr. Roemer speaks permits only the conclusion that all the marks of plutonic or volcanic formation must belong to the same system, which, traversing the upper limit of the more recent cretaceous strata in the valley of the Rio Bravo, shows itself in the shape of the low basaltic hills mentioned as occurring at the crossing of the Rio Frio, and at the heads of the rivers Leona and Las Moras.

There is no doubt that this dyke continues its northeastern direction, accompanying as an outlayer of the higher regions of the Guadalupe and Ozark mountains, and thus probably crosses the whole of Texas, and possibly Arkansas.

METEORIC IRON.

With regard to meteoric iron, to which Dr. Roemer refers in connexion with the plutonic rocks, and of which he mentions a large specimen now preserved in Yale College, we have to state that, besides magnetic iron ore, which is scattered in loose innumerable pieces of every shape and size over the whole surface of the cretaceous basin of the Rio Bravo, meteoric iron is known to exist about ninety miles northwest of Santa Rosa. An American resident of this town, Dr. John Long, called my attention to a piece weighing some twenty-five pounds, which was then in the possession of a Mexican; small pieces had been cut from it, and hammered out without the aid of fire into some trifling articles. It is said that the whole surface of the area (embracing about thirty acres of land) where the deposition of this valuable mineral occurs is covered with blocks of it, of greater or less extent, some containing as much, and even more, than thirty-six cubic feet.

GREEN SAND WITH LIGNITE

The upper limits of that portion of the cretaceous basin, which consists chiefly of strata of green sand, and the course of the volcanic dyke discussed above, seem rather to run parallel than approach each other.

So far as our observations extended, the main portion of the cretaceous basin, from Las Moras to the vicinity of Reynosa, forms a belt of 380 to 400 miles in width.

The upper part of this belt commences in the vicinity of Las Moras, and terminates some few miles above Laredo, a distance of about 200 miles, whilst the lower part begins where the former ends, and reaches as far as the vicinity of Reynosa, showing a width of about 340 miles. Both of these parts are distinctly characterized by strata of green sand, (chloritic chalk,) which change, according to the amount of oxyde of iron they contain, into variously tinted sandstone shoals. The solidity of the strata varies very much. They are sometimes formed into very solid rocks, well suited for mechanical or architectural operations; again, they consist of loose and coarsely-grained sandstone slate, which rapidly crumbles on exposure to the air. All these green sand strata are frequently intersected with layers of debris of analogous character.

In several places where these green sand strata were disintegrating, and being carried off by the action of the waters, there was observed a white, salty efflorescence, which may possibly be "ammonia." The "Rocky walls" near the mouth of the Arroyo Castaño, which is about 40 miles below Eagle Pass, and near the Presidio de San Juan el Bautista, are remarkable for this efflorescence, as also some terraces below this point. The frequent occurrence of a certain chenopodium, containing a large amount of this salt, and often covering exclusively wide tracts of sandy bottom-land along the Rio Bravo, may prove more conclusively the peculiar elements of the green sandstone.

The green sand, particularly in the upper belt, is often and variously intersected by strata of different nature, though certainly closely allied with the same system.

Strata of sandy or argillaceous marls, or blue or grayish clay, all profusely impregnated with oxide of iron, and even layers of corresponding debris, often intersect the green sand strata.

The general characteristic of this belt and its subdivisions is the strict horizontality of its strata throughout. It is only here and there that some slight local disturbance has taken place, as, for instance, near Laredo, and again some 40 or 50 miles above, where a dip of about 8° W.S.E. and E. is exposed.

The following peculiarities may serve to characterize the two subdivisions of the green sand belt:

LIGNITE COAL.

From Las Moras to the vicinity of Arroyo Sombreretillo, which is about 10 miles above Laredo, lignite coal occurs quite frequently. None came under our observation below this point; outcrops of it, however, are said to be found in the neighborhood of Roma, some 10 miles above the mouth of the Rio San Juan.

Though there is not much doubt of the existence of lignite below the Arroyo Sombreretillo, our observations have led us to the conclusion that it is more sparsely distributed.

These lignites vary both in appearance and quality; sometimes they are found to be scaly or slaty, and of a dull earthy fracture, sometimes resinous and sharply edged. Prints, and even remains, of plants, preserved in these coals, indicate vegetable forms of the higher orders, as gramineæ, (perhaps reed and cane,) and even parts of dicotyledonous trees, such as willow or ash. Other specimens of coal from below appear more amorphous; but it contains so much bitumen as to be of no use in the blacksmith's forge, where it runs together and becomes baked into a solid mass.

The localities remarkable for the most considerable deposits of lignite coal are the following: On both sides of the mouth of Elm creek, near Eagle Pass, particularly on the north bank of this water-course, where layers from 3 to 4 feet thick are exposed. On the south bank of this