Texas Bureau of Economic Geology The Big Bend of the Rio Grande: A Guide to the Rocks, Geologic History, and Settlers of the Area of Big Bend National Park

Marathon Basin.—Marathon, Texas, is a small village on the Southern Pacific Railway and U. S. Highway 90. The town is near the center of a large topographic depression, the Marathon Basin, which is crossed by U. S. Highway 90 from east to west and U. S. Highway 385, from Marathon to the Park, on the south. The Paleozoic rocks are about 21,000 feet thick in the Marathon area; most of them are highly deformed (fig. 25) and represent a part of the Ouachita System. The basin is rimmed on the east, south, and west sides by gently dipping Cretaceous rocks. There are a few small areas of flat-lying Cretaceous rocks in the basin, and Cenozoic volcanic rocks overlie the Cretaceous formations on the western rim; the north rim is composed of tilted late Paleozoic (Permian) formations and Cretaceous rocks (fig. 5).

FIG. 25. Steeply tilted Paleozoic rocks exposed along U. S. Highway 90 about 15 miles east of Marathon, Texas.

The rocks in the Marathon Basin were deposited in the Paleozoic trough that bordered the North American Paleozoic continent on the south (fig. 6). During the late Paleozoic Era, the newly formed rocks were uplifted and the earth's crust was shortened by the squeezing of some of the formations into tight overturned northeast-trending folds. The squeezing was so great in some places that the rocks were broken along the fold's crest and the upturned mass rode forward as sheets thrust over onto younger rocks (fig. 26). Along one such fault, called the Dugout Creek overthrust, a mass of rock was pushed northward about 6 miles (King, 1937, p. 28). King has shown the character of the Marathon Basin deformation by drawing sections along three lines across the basin, and these, omitting some of the detail, are used to illustrate the complex structure (fig. 27) (in pocket).

FIG. 26. Diagram showing the development of folds, overturned folds, and thrust faults in horizontal rocks. A, A block of undeformed horizontal rocks. B, Lateral pressure applied causes the rock to fold forming a washboard-like surface. C, The concentration of pressure deformed the weakest rocks into overturned (asymmetrical) folds. D, Additional pressure caused the rocks to break along the fold crest and the upper layer began to slide forward over the lower layer. E, Additional pressure caused further forward movement; in some areas the forward thrust is measured in tens of miles.

FIG. 27. Structure sections of Marathon Basin. Modified after King (1937, cross sections, Pl. 23). (click on image for an enlargement in a new window)

The floor of the basin, in most places quite flat, is underlain by deformed, fossiliferous, marine limestone and shale, but rising above the basin floor are a few low mountains and long, sharp, sinuous ridges composed of hard formations which resist erosion. One of the ridge-making formations is a cream to yellowish-white, fine-grained, siliceous rock called novaculite. It is the same age as the Arkansas novaculite, which crops out near Hot Springs, Arkansas, and is used for the manufacture of whetstones. This stone has proved extremely valuable to artisans for honing cutting edges, and the fabrication of whet-stones from the Arkansas novaculite is a small but steady industry. The honing stones have worldwide use for sharpening carpenters' and engravers' tools, surgeons knives, and precision cutting instruments (even the pointing of hypodermic and other needles). At one place seen from U. S. Highway 385, about 10 miles south of Marathon, the novaculite is folded so that it resembles a horseshoe in plan. "Caballos" is the Spanish word for horses, and this low mountain was named Caballos Mountain; hence the name Caballos Novaculite for the siliceous rocks exposed in the Marathon Basin.

Numerous fossils have been collected from some of the formations in the Marathon Basin. They include trilobites, graptolites, corals, clams, snails, brachiopods (figs. 7, 8, 9), sponge spicules, and conodonts which are probably worm jaws and teeth (fig. 28). Some of the fossils have been matched with species collected from the Arbuckle Mountains of south-central Oklahoma, the Ouachita Mountains in southeastern Oklahoma and Arkansas, and the Appalachian Mountains of the eastern United States. This indicates that the rocks in these widely separated places are the same geologic age.

FIG. 28. Conodonts (Matthews, 1960).

The Solitario.—The Solitario is about 30 miles southwest of the Marathon Basin and is crossed by the Brewster-Presidio County line north of Lajitas (fig. 1). The area is structurally a dome (fig. 29), about 9 miles in diameter, but the center of the dome is a topographic depression drained by two intermittent streams through two narrow canyons in the basin's rim called the Lefthand and Lower Shut-ups (fig. 30). The highly deformed Caballos Novaculite and some Paleozoic limestone and shale formations are exposed in northeast-trending ridges in part of the basin floor. These rocks are of the same geologic age as those in the Marathon Basin and they form ridges which trend in the same direction. The Cretaceous sea also covered this area and massive limestone ledges such as those now exposed in the rim of the dome once covered the Paleozoic formations in the center of the depression. The Solitario uplift was probably caused by a laccolith intrusion that elevated the area. Then erosion stripped the Cretaceous rocks from the crest of the dome, and later volcanic rocks were deposited over parts of the basin floor (fig. 24).

FIG. 29. The development of laccolithic domes. A, Laccolithic dome in horizontal rocks, the intrusion not exposed. B, Cross section and plan view of a laccolith. C, An eroded laccolith; the intrusion with flanking cuestas is exposed. This type of intrusion is common in the Park.

FIG. 30. Aerial photograph of the Solitario, a dome that was probably formed by a laccolith that is not exposed. (Photograph by Aero Service Corporation, San Antonio, Texas.)

Persimmon Gap.—Persimmon Gap, named for the Mexican persimmon trees that grow there, is a wind gap carved by an ancient stream across the southern Santiago Mountains. The stream may have been the ancestral Maravillas ("wonder") Creek now flowing parallel to the east side of the Santiago Mountains, or the ancestral Nine Point Draw that now crosses the mountain range through Dog Canyon (fig. 4). Why the ancient stream changed its course is not known, but during the time that the stream was cutting Persimmon Gap, it uncovered the Paleozoic Maravillas, Caballos, and Tesnus Formations that were once buried beneath Cretaceous limestone (table 1, p. 27). The Paleozoic rocks exposed at Persimmon Gap occur in thin, northeast-trending thrust slices. A second and younger period of folds and thrust faults that occurred near the end of the Mesozoic Era deformed both the Paleozoic thrust slices and the younger Cretaceous formations.

Crustal stresses in late Mesozoic time deformed both the Paleozoic and Cretaceous rocks and elevated the Santiago Mountains, a long, northwest-trending anticlinal ridge overturned toward the southwest. Continued pushing broke the rocks along the crest of the fold and the forward thrust pushed Paleozoic rock slices toward the southwest, leaving them resting on top of the younger Cretaceous formations (fig. 26). Erosion by the ancient stream removed the Cretaceous cover, exposing a cross section of the fold (fig. 31). Maxwell et al. (1967) discussed the geology of Big Bend National Park (including Persimmon Gap) in more detail, and for further information, map, and cross sections, the reader is referred to that report and to Plate II (in pocket).

FIG. 31. A, Thrust faulting and overturned fold in Cretaceous rocks near Persimmon Gap (see also fig. 26). Elevation 3,745 feet of (B). Line indicates fault, above which the rocks were thrust or pushed to the right. B, Panoramic sketch, Persimmon Gap area. (Modified from Hazzard, in Lonsdale et al., 1955, p. 55.)

A more recent chapter in the early history of the Big Bend deals with the Indians in Persimmon Gap. Raht (1919) has written much about the Indian depredations in northern Mexico and also of the raids on early settlers, wagon trains, and military posts in West Texas. One branch of the Comanche Trail passes through Persimmon Gap where the Texas State Highway Department has erected an historical marker. Rock mills or metates used by the Indians to grind mesquite beans or grain are found in the area; they can be photographed but it is unlawful to collect artifacts in a National Park. (See a Park Ranger for the regulations.)

It seems that once a trail, always a trail, for some of the early Latin-American settlers followed the Comanche Trail; later the same route was used by cowboys, freighters, Texas Rangers, the Army, and the stage line from Marathon to Boquillas. In later years, the copper-lead-zinc-silver ore mined in Coahuila, Mexico, about 5 miles southeast of Boquillas, was hauled by burro and mule trains through Persimmon Gap to Marathon, and now the Park road passes through the Gap to join U. S. Highway 385 that leads through the Marathon gateway to Big Bend National Park.

From Persimmon Gap, a paved road extends southwesterly across the Park (fig. 4 in pocket). For some 20 miles, it closely follows the old Comanche Trail, the Marathon-Boquillas stage route, the old ore road, and a route used by the U. S. Army when hauling supplies to small garrisons stationed along the border. The road divides at the Park headquarters on the northeast side of the Chisos Mountains, about 70 miles from Marathon. The left-hand road (paved) passes east of the Chisos Mountains and takes a southeasterly course to Boquillas, Boquillas Canyon, and the Rio Grande. A right-hand, unimproved spur road, off the Boquillas road, leads to Glenn Spring and Mariscal Mountain and continues up the Rio Grande to the village of Castolon. The right-hand branch of the paved road at Park headquarters passes north of the Chisos Mountains and leads westward to join the entrance roads from Alpine and Presidio. The left-hand branch of the western road ascends the grade into the Chisos Mountains where the tourist facilities are located; farther west is a new paved road that crosses the mountains to Castolon; northwest of the Chisos Mountains a gravel spur road leads to Santa Elena Canyon (fig. 4).

Dog Canyon.—After leaving Persimmon Gap, en route toward the Park headquarters, the Santiago Mountains are on the left (east). They are a high ridge of massive Lower Cretaceous limestone; the first canyon, about 5 miles south of Persimmon Gap, is Dog Canyon (fig. 4). It is a gorge cut by Nine Point and Bone Spring Draws through the mountains, and a cross section of the structure is exposed in the canyon (fig. 32). Within the canyon, the rocks are vertical (fig. 33), but the geology differs from that seen at Persimmon Gap in that Paleozoic formations are not exposed. Dog Canyon was named by an early traveler who found a dog guarding an oxen-drawn wagon in the canyon. The owner of the oxen was never found and what happened to him is one of the Big Bend mysteries. Lieutenant Echols, who led a camel train through the Big Bend in 1860 during studies made by the United States Army to determine the value of camels as beasts of burden in the Southwest, is believed to have passed through this gap. On Echols' map and in some of the late nineteenth-century reports, the canyon was named Camel Pass.

FIG. 32. Asymmetrically folded rocks in Dog Canyon (fig. 24). The nearly horizontal rocks in the foreground can be traced to where they stand vertical in the background. Dog Canyon can be reached by a road that leaves the paved Park entrance road immediately north of the Bone Spring Draw bridge.

FIG. 33. Vertical beds of Lower Cretaceous limestone in the south wall of Dog Canyon.

Beyond Dog Canyon is the Sierra del Carmen (fig. 4). The first peak is Dagger Mountain, named for the stand of giant daggers (yucca) that grow there. It is a dome-shaped mountain that has the shape of mountains raised by igneous intrusion but erosion has not yet exposed the igneous rocks (fig. 29A). Beyond Dagger Mountain, the Sierra del Carmen is a series of long, northwest-trending ridges of massive Lower Cretaceous limestone (table 1). The rocks have been broken by faults and tilted to form fault block ridges (fig. 17). The dipping slopes are gentle, but along the faults (fractures), there are steep cliffs of massive limestone. The drainage is normally parallel to the tilted blocks.

Along the west side of the Sierra del Carmen (fig. 4) is a series of "white cliffs" that extend from near Dog Canyon southward to the Rio Grande. The "white cliffs" are formed mostly of Boquillas Formation (table 1, p. 27). The black rock layer seen in some of the "white cliffs" is a sill (fig. 24C, D) that intruded the Boquillas Formation prior to faulting.

The Rosillos Mountains are on the right (west) side of the road. In Spanish, "rosillos" means "roan," and the name of the mountains is taken from the color of the rocks. The uplift was caused by a laccolith that domed the rocks. The magma cooled beneath a rock cover and the cold head of the intrusion was not exposed until it was uncovered by erosion (fig. 24G, H). A fault along the east side of the mountain exposes the flat base (Pen Formation) on which the laccolith rests. The Grapevine and McKinney Hills farther southwest (fig. 4) have a similar origin.

For the first 20 miles south and southwest of Persimmon Gap, the Park road is on soft Upper Cretaceous marine limestone and shale (Boquillas and Pen Formations, table 1) that contain fossil oysters, clams, snails, ammonites, and fish bones (figs. 7, 8, 9). During the Upper Cretaceous, rocks such as these covered the area where the Santiago-Sierra del Carmen ranges now stand (p. 33), but they have long since been eroded away. Farther ahead, on the eastern side of Tornillo Flat, are still younger Upper Cretaceous continental sandstone and shale (Aguja and Javelina Formations, table 1). They contain dinosaur bones, fossil logs and stumps, and coal beds. On the eastern side of Tornillo Flat are the oldest Cenozoic formations known in Texas (Black Peaks, Hannold Hill, and Canoe Formations, table 1). Mammal bones, skulls, and teeth have been found in some of them, and the fossil exhibit mentioned on page 25 is at the left side of the road about one-fourth mile north of the Tornillo Creek bridge (fig. 21).

Tornillo Flat.—Tornillo Flat is a plain, rimmed by badlands, that was formed by the erosive action of Tornillo Creek and its tributaries. During initial erosion stages, on a gently sloping surface that is underlain by horizontal rock, the drainage pattern is normally simple (fig. 34). As erosion progresses, especially in areas where there are alternating rock layers of unequal permeability, intricate topographic forms develop (fig. 35A). This type of surface commonly is characterized by a maze of gullies and is called badland topography (fig. 35B). Badlands border most of Tornillo Flat and also occur at many other places in the Park. They are most often developed in a semi-arid climate, when the rains are very hard and there is little vegetation, and where the rocks are impervious clay or shale alternating with layers of soft sandstone.

FIG. 34. A, Sketch showing development of simple stream erosion. B, Photograph of arroyo development in nearly horizontal rocks.

FIG. 35. A, Sketch showing the development of step-like topography on horizontal rocks with two layers of impervious clays. B, Photograph of badlands development where there are impervious clays and little vegetation.

Erosion of tilted rock layers forms a different topography from that formed on horizontal rocks, particularly if some of the rock layers are hard. The hard rocks resist erosion and as the soft rocks are eroded, gullies occur in soft rocks while the hard rocks are left standing to form buttes, mesas, or long linear ridges. Low ridges formed in this manner are cuestas (fig. 23), but if the ridges are high and the rocks dip at a steep angle, they are hogbacks. There are several cuestas in central Tornillo Flat; hogbacks are well developed around the flanks of the McKinney Hills farther south. The wide variety of topographic features that have developed on Tornillo Flat is the result of erosion across hard and soft rock layers, some of which are nearly horizontal and others dip up to 45°.

Tornillo Flat is probably the driest part of the Park. During the dry season, there is little vegetation except creosote bush, lechuguilla, and cactus, but following a shower, plants spring up as if by magic. In a few days they are in blossom and the desert is clothed in a robe of gorgeous colors. Normally, this type of vegetation lives only a few weeks; following the next shower, new and different flowers appear and this causes a succession of blossoms, depending upon the frequency of the moisture. Prior to about 1918 to 1920, Tornillo Flat was covered by tobosa grass. The grass, named for the Tobosa Indians (Raht, 1919, p. 30), was cut by early settlers for hay—parts of an old mowing machine may perhaps still be seen on the Flat. The grass was killed by drouth and overgrazing, and only a few small clumps of the original grass remain. During recent years, as part of the conservation program, the National Park Service has attempted a reseeding project. Some of the new grass has survived and is spreading; perhaps in time Tornillo Flat will again be grass-covered as it was 50 or more years ago.

When the tobosa grass covered Tornillo Flat, the area was undoubtedly frequented by game; it was an antelope range during the early twentieth century and was probably a popular hunting ground for the Indians. The Comanche Trail passed through the Flat and the less nomadic tribes who lived in the Chisos Mountains probably hunted there also. The low sandstone cuestas (ridges) that extend across the Flat appear to have been ideal blinds for stalking game. Some of the cuestas, especially those with shallow shelters, provide evidence of other Indian activities. Numerous flint chips are scattered about or near the sandstone ridges, some of the shelters are smoke blackened, and occasionally arrow points, stone scrapers, knives, and metates have been found. Many blocks of hornfels (a black, fine-grained, metamorphosed rock) used by the Indians for making stone implements, are scattered over the surface. Their hornfels quarry was at Banta Shut-In near the south end of Tornillo Flat.

From Tornillo Flat the Park road ascends Hannold Hill and climbs onto a thick gravel deposit composed mostly of rocks washed from the Chisos Mountains. Near the top of Hannold Hill is the site where the Texas Highway Department's first camp (fig. 2) was located and shortly beyond (left) is the grave of Mrs. L. C. Hannold, who died at her nearby ranch home. Hannold Hill and the Hannold Hill Formation are named for the ranch. The gravel deposit is exposed along the road up the slope toward the Park headquarters.

From the Park headquarters road junction, the east fork skirts the upper edge of the gravel where several of the intrusive igneous rock masses can he seen, flanks the eastern side of the Chisos Mountains, and turns southeasterly toward the Rio Grande, About 5-1/2 miles from the Park headquarters is an unimproved road to the right (south) leading to Glenn Springs (pp. 61-67). Farther on, about 1-1/4 miles, is a road (north) to Dugout, marked by a small clump of cottonwood trees; this is the homestead of the late W. A. Green. An early settler dug out a spring to concentrate the water seeping from the gravel deposit flanking the east side of the Chisos Mountains; hence the name "dugout." There was a sufficient water supply for a ranch headquarters, garden, fruit trees, and flowers. Some of the palms and oleanders continue to survive. Dugout was an important overnight camp site on the Marathon-Boquillas road. The second unimproved road to the east leads to the old Boquillas post office (La Noria, "the well," on some maps), and the next old road to the east leads to Hannold's store, a popular trading post during the 1930's (pp. 101-103). Both roads are impassable to modern automobiles; the post office and store are abandoned, and the buildings are in ruins.

The gravel deposit followed by the road southeast from Hannold Hill is all out-wash debris from the Chisos Mountains. It is thin in some places, and rock mounds protrude above it, but along some of the old valleys it is as much as 500 feet thick. The underlying bedrock is exposed along most of the arroyos adjacent to the Tornillo Creek valley; in that valley near the lower bridge, where the gravel has been removed, the Upper Cretaceous formations contain many fossils. At one place, about three-tenths of a mile northeast and east of bench-mark J-742, the skull of a giant crocodilian-like reptile (Phobosuchus riograndensis, fig. 17) was collected.

Hot Springs.—About three-quarters of a mile east of the lower Tornillo Creek bridge, a gravel road turns right (south) from the pavement and leads to Hot Springs. This is the abandoned site of a small resort carved from the desert wilderness by J. O. Langford (1952, pp. 1-154), who moved to this homestead in 1909 and developed limited public bathing facilities at the hot spring in 1910. Because of border troubles, the resort was abandoned in 1913 when the Langford family moved to El Paso. They returned in 1927 to build a trading post in 1928 and cabins a few years later (fig. 36). The trading post is reached by way of a one-way road down a narrow canyon to a wider canyon cut from the Boquillas Formation (fig. 37) near where Tornillo Creek enters the Rio Grande. The bathhouse, about a quarter of a mile below the store, was destroyed in 1954; the store and cabins are still standing, but all are in a sad state of repair.

FIG. 36. The Hot Springs trading post (built in 1928) as it appeared in 1936. The display of dried catfish heads attests to one type of recreation, and fishing is still a popular sport in Big Bend National Park. (See a Park Ranger for copy of the fishing regulations.) Hills of Boquillas flagstone in background.

FIG. 37. The Boquillas Formation. This flagstone unit consists of thin-bedded limestone and shale. The interbedded shale layers are soft and easily eroded and leave the harder layers in prominent relief much like the weathered appearance of an old brick wall. The formation has many fossils (figs. 7-9).

The hot spring, one of several along this section of the Rio Grande, emerges only a few feet above the normal water level in the river; in 1936 it flowed about 250,000 gallons of water per day. The temperature is about 105° F, and although the water is mineralized it does not have a highly objectionable taste or odor. The source of the heat is not known. Some believe that it is from buried igneous rocks that are still hot. Although there are several small dikes (fig. 24J) in the area, the theory that the water is heated by unexposed hot igneous rocks has not been proved. The therapeutic value of the water is loudly acclaimed by many West Texans. The Indians apparently also used the spring waters. A crude pit in which the water formerly accumulated was allegedly excavated by the Indians. Rock mills, smoke-blackened rock cliffs, and a few Indian paintings attest that the area was used by early inhabitants.

Langford sold the Hot Springs facilities and left the Big Bend in 1942, and for the next ten years, the trading post and bathing facilities were operated by the late Margaret (Maggy) Smith. Mondays were "red-letter days" at Hot Springs, as that was the time of weekly mail delivery and also the time when the Mexican people came across the Rio Grande to trade, some from as far as 100 miles inland (fig. 38). The Mexican Customs and Immigration officials received their paychecks from Mexico City, via Marathon and then Hot Springs. The traveler who could visit Hot Springs on Mondays always had a rewarding experience.

FIG. 38. Mail day—a group of Mexican nationals who came to get their mail which was delivered once a week. After the mail truck arrived and they visited with relatives and friends, they purchased a few supplies at the nearby trading post and returned to their homes, some of which were 100 miles south of the border.

Maggy Smith made her living off the Mexican people, but she was godmother to many of them and gave back most of what she made in the way of food, clothing, medicine, and services. A Mexican never left Maggy's store hungry or cold, and in time, he would return with a chicken, pig, or produce to pay for the merchandise. Maggy Smith was known far beyond the arbitrary boundaries of the Big Bend, and if one were to meet a person who had visited the Park during the late 1940's, he probably would ask, "Say, do you know Maggy Smith?"

Boquillas.—There are villages on both sides of the Rio Grande, but the Mexican village is the older and a better established settlement (fig. 39). The site of the Texas village has been moved several times, and on some maps of the early 1900's, the village is shown at a place on Tornillo Creek about 9 miles north of the Rio Grande. The first Boquillas, Texas, was on the river bank and was established in 1894 by D. E. Lindsey (Raht, 1919, p. 283), who had an irrigated field on this side of the river and who operated an ore train in Mexico. The ore came from the old Boquillas (Del Carmen) mine at the foot of the Sierra del Carmen escarpment about 5 miles southeast of the village. The ore (lead, silver, and zinc) was smelted in a small furnace on the bank of the river near Boquillas, Mexico. Lindsey saw an opportunity for a store on the Texas side and went to San Antonio, bought a stock of supplies, shipped it to Marathon, and freighted the stock to the river. There was no road and Lindsey employed men to go ahead to clear the way and pick the best trail. It took two weeks to make the trip from Marathon, and then because of impassable cliffs, they could not haul the supplies to the place that Lindsey wanted to locate the store. Finally, burros packed the commodities to the selected site.

FIG. 39. Boquillas, Mexico, in 1936. Note that some houses are built of Boquillas flagstone, others are made of adobe, and some of both flagstone and adobe. At least one house (upper left) has painted window frames, but only a few have screens. This picture was taken during siesta; all the inhabitants were in their homes, at which time the burros occupy the village square.

The first store was a rude structure with adobe frame and counters made of packing boxes. It was not long until a settlement grew up on this side of the river and the store was enlarged. The men who were operating about 250 wagons, hauling supplies to the mine and back-hauling ore, were the principal customers. The small Mexican smelter that operated in Mexico was finally closed, and beginning about 1914, the ore was brought to this side of the river by aerial tramway (fig. 40). The northern terminal was about 4 miles north of the river; from here, the ore was freighted to Marathon for shipment to a smelter. The freight wagons drawn by mules and burros followed the old ore road which passed between the McKinney Hills and the Sierra del Carmen to join the present Park road about 15 miles south of Persimmon Gap (fig. 4). Steam engines with flat-rimmed iron wheels and a rear draw bar attached to the engine frame were also tried unsuccessfully for the ore haul. The availability of fuel was a major problem. If sufficient coal was hauled by the engine, the load of ore was materially decreased, and wood was not readily available. Some of the grades were too steep for the engines to draw a load. The slits in solid bedrock near the west end of the road tunnel were made at this time to lessen the grade for the steam engines.

FIG. 40. The aerial tramway which crossed the Rio Grande. The full buckets traveled on the inside cables, were dumped and returned to the Mexico ore dump on the outside cable. This operation closed down about the end of World War I. Most of the towers and all the cable are now down, but the scar is still visible.

Following the period of ore importation, which ended about 1919, activities in Boquillas, Texas, waned. The Lindsey store closed and most of the inhabitants moved away, but Boquillas acquired fame as a place to eat. This reputation was made and maintained by Maria (Chata) Sada and her husband, Juan. The Sadas had a small store about 200 feet northwest of the present (1967) Park Ranger station; ruins of the old building are still visible. Chata, which is a nickname for "flat nose" or "cute," managed the operations on the Texas side of the Rio Grande and Juan operated the Boquillas (Del Carmen) mine in Mexico. Both the Sadas were loved, respected, and known personally by many border residents the length and breadth of the Big Bend.

There were not many travelers to an out-of-the-way place like Boquillas in the 1920's and 1930's. It was a difficult all-day drive from Marathon—more than 100 miles—and all travelers were hungry when they arrived. Those who had been there before knew what to expect and those who arrived for the first time soon learned that Boquillas was the end of the road. Chata could hear the laboring car motor long before the vehicle was in sight; she would then start a fire in the four-hole flat-topped wood-fired stove and begin to prepare a meal. Those who were making their first visit were agreeably surprised to get a complete meal in a neat, clean dining room, with all they could eat for 50 to 75 cents. There were also four clean adobe cabins nearby and a large arbor where the guests could spend the night. Most of the travelers were law-enforcement officers, International Boundary and Water Commission engineers, mining men, geologists, ranchers, and residents from Marathon or Alpine. They always carried a bedroll and a chuck box with food, but at Boquillas, only the bedroll was used.

The Sadas were one of the few Mexican couples who did not have children of their own, but they reared several foster children to maturity. These were children whose parents could not provide as good a home as could the Sadas. They were not officially adopted but more or less "borrowed," which was a Mexican custom along that part of the Rio Grande (fig. 41). Chata also served as midwife and often acted as judge or teacher. After Juan's death, she moved to Del Rio, Texas, where she lived with one of her "adopted" sons, the owner of a prosperous automobile appliance store.

FIG. 41. The Sada "family" in front of their restaurant-home at Boquillas, Texas, in 1936. At least three of the older "adopted" sons are not present.

Boquillas Canyon.—The head of Boquillas Canyon is about 3 miles down the Rio Grande from the site of the present Boquillas village (Park Ranger station). The canyon is about 25 miles long, averages about 1,500 feet deep, and is one of the Park's most interesting features (fig. 42). It shows the results of the Rio Grande's great erosive power. The rocks have been cut away by the river and its load of sand and silt. During the late Cenozoic Era, the hard Lower Cretaceous limestone now exposed in the Sierra del Carmen Mountains was buried beneath softer layers of Upper Cretaceous rocks. At that time, the ancient river channel was located on the overlying soft rocks and when it cut through them, it was trapped in its channel and there was no alternative but to cut into the hard rocks of the buried highland. During the rainy season (summer and early autumn) the river carries many tons of sand and silt daily; these are the cutting tools that permit the river to grind its gorge deeper and deeper into the Lower Cretaceous limestone ledges exposed in the Sierra del Carmen.

FIG. 42. The head of Boquillas Canyon as seen from the look-out stop along the Park road to the canyon.

The rocks are mostly massive limestone and at most places the canyon walls rise vertically from the water's edge. There are many vertical joints, and the temperature range from the heat of the blazing sun to the cool nights leads to further cracking due to expansion and contraction in the rocks. Moisture and frost action further aid in widening the cracks and dislodging blocks of the limestone. In spite of the low rainfall, the limestone ledges in the upper part of the canyon walls are commonly pitted where some of the rock has been dissolved away. There are a few low rapids in the river bed and many boulders have fallen from the canyon walls, but the canyon is navigable, without great danger, in a small boat or rubber raft.

It was near the head of Boquillas Canyon that C. L. Neville, a Texas Ranger captain, and four rangers were escorting some surveyors down the Rio Grande by boat (Madison and Stillwell, 1958, p. 58) when they saw a band of Indians who scattered, leaving their horses. Captain Neville knew that he could not permit the horses to be reclaimed by the Indians to use them in raids on nearby ranches. Since he could not take the horses along in the boats, and since most of the ammunition had been lost when a boat had previously capsized, the horses were blindfolded and killed with an ax. To this day, some people use the term Dead Horse Canyon for the gorge and Dead Horse Mountains for the Sierra del Carmen range. The late Arthur A. Stiles, however, reported that the highlands were named Dead Horse Mountains (Sierra del Caballo Muerto) because his favorite saddle horse was killed when it fell from one of the cliffs (oral communication, August 8, 1956).

The principal range in the Sierra del Carmen in Mexico, south of the Rio Grande, forms a magnificent west-facing escarpment (fig. 43). Most of the rocks are massive Lower Cretaceous limestone with smaller amounts of shale. The massive rocks break to form steep-faced ledges, and some of the debris from the upper ledges settles on lower sloping surfaces cut in the shale. This forms light-colored bands of limestone and dark bands of debris-covered shale across the face of the escarpment. The sunset on this cliff is a gorgeous sight, the like of which is seldom seen. In the afternoon when there are wispy clouds in the western sky, the sunlight strikes the rocks with an orange glow. As the sun sets, the colors brighten to a crimson hue. After the sun disappears, the colors fade for a brief period but are again brightened by the afterglow. Eventually, however, the colors change to pink, then bluish gray and purple. These colors may be seen long after the light has disappeared from the other mountains—then all colors eventually fade into the blackness of night.

FIG. 43. The Sierra del Carmen in Mexico. a block of Lower Cretaceous limestone raised along a fault. Erosion has modified the escarpment causing it to recede. The fault is in the rugged country in front of the escarpment.

Many incidents of human interest have been recorded in or near Boquillas. The village gave its name to the Boquillas (flagstone) Formation (Udden, 1907, pp. 29-33) that crops out along the west side of the Santiago-Sierra del Carmen ranges from near Persimmon Gap southward to the Rio Grande and westward up the river to beyond Mariscal Mountain (fig. 4). Boquillas was the headquarters camp of Arthur A. Stiles, topographic engineer for the United States Geological Survey, who made the topographic map of the Terlingua and Chisos Mountains quadrangles, part of which are in the Park. It was here in December 1903 that Bill Ernst, storekeeper and Justice of the Peace, called a meeting of local settlers to recommend names to be used on the Stiles map (Madison and Stillwell, 1958, pp. 5-14). Ernst's name was used for a tinaja (rock water hole), a valley, and a tank that are in the area where he ranched. The term Ernst is used by Maxwell et al. (1967) to designate the lowest member of the Boquillas Formation. Bill Ernst was fatally shot only a few miles from Boquillas, and the switch-back curve in the old road south of the road tunnel is known to many as Ernst curve or "dead man's" curve.

On the Rio Grande flood plain, a few miles west of Boquillas, is the National Park Service tourist center, the Rio Grande Village. The site was selected because of adequate construction space in an otherwise rugged terrain. abounding with scientific values in geology, botany, archaeology, and history, pleasing scenery, and a mild winter climate. The development of the tourist area has already begun. Roads, irrigation ditches, landscape planting, picnic area, service station, and camp ground have been completed. The construction of accommodations required for taking care of overnight guests, a visitor center building where information and interpretative services will be available, parking areas, water, sewer, and related utility systems, and employee housing will be forthcoming. It is the National Park Service goal to give the visitor a park system adequate for his enjoyment and inspiration; a park system with accommodations so developed, managed, and used that posterity will enjoy to the fullest the values of their great American heritage.

Although the Rio Grande Village is becoming a visitor center, it was a center of human activity centuries before the first tourist stopped there. In the Lower Cretaceous limestone ledges at the river bank about 50 feet from the irrigation pumping station, there are 30 or more deep cylindrical holes used by the Indians for grinding mesquite beans and grain (fig. 44). It is not known what tribal band made the rock mortars or whether they lived in a village on the river flat or in the nearby smoke-blackened rocky cliffs, but judging from the number and depth of the holes, this was a well-established village long before de Vaca crossed the Big Bend country. Probably some of the aborigines lived on fish, game, and grain cultivated along the river bottoms in a manner similar to that of the tribes that de Vaca described as living in fixed villages near Presidio, Texas.

FIG. 44. A metate, or rock mill, in the limestone ledge exposed in the river bank at the west end of the Rio Grande Village. There are about 30 similar mortars at this locality.

At the eastern end of the Rio Grande Village are several warm springs; the artifacts that have been found nearby attest to Indian use and occupation. In one of the warm springs there was a species of tiny mosquito fish, Gambusia gaigei (Hubbs, 1938-39, pp. 10-11), found only in the Big Bend. They no longer occur in the warm spring but are retained in a specially prepared rearing pond where they have maximum protection. The fish are an interesting biological curiosity not only because of their size and local habitat, but also because they do not lay eggs but give birth to live offspring.

The spring area was also a settlement site for the early Anglo-Americans. The Deemer store, robbed by the Mexican bandits on May 5, 1916, was located here (pp. 65-67). The river flood plain was farmed in the early years of this century. J. M. Graham, one of the early farmers, died at the age of 91 on July 4, 1963. Frank Wedin of Marathon and, later, John (deceased) and Maria Daniels, now of Presidio, irrigated their crops by pumping water from the Rio Grande, Cotton and livestock feed were the principal products. The cotton grown on this fertile bottom land was hauled to gins at either Balmorhea or Pecos, Texas, distances of about 191 and 229 miles, respectively.

After the days of the Spanish Presidio system and more especially during the middle years of the nineteenth century, the Indians were especially aggressive in the Boquillas region. The mightiest and most dangerous were the Comanches, The Comanches were a hunter folk without permanent homes; they traveled by horseback and disliked agriculture. Their fixed abode, if a nomadic people have one, was chiefly in the rocky highlands that stretched from the upper Red River on the north southward to the Rio Grande, and from the mountains of New Mexico eastward to the San Saba Valley. Being great wanderers, fighters, horsemen, and plunderers, they lived off the land, followed the buffalo, and raided into Mexico.

During the September full moon, after the summer rains had filled the water holes (fig. 45), the Comanches started their annual southward trek to kill defenseless farmers, capture slaves, rob, plunder, and drive off the livestock. They raided most of the northern Mexican states, with especially heavy raids in Coahuila, Chihuahua, and Durango. The time of the annual onslaught by the barbaric Comanche warriors, outlaws from other tribes, and even renegades from Mexico (who passed through the Big Bend) became known as the "Mexican" or "Comanche moon."

FIG. 45. A tinaja, a natural depression in the rocks that collects water after rains and is used as an occasional water supply for both man and beast.

One of the main Comanche trails entered Big Bend National Park at Persimmon Gap (pp. 30-35), continued south to Glenn Spring, and crossed the Rio Grande at what was then called Paso del Chisos, the flattish valley west of Mariscal Mountain (fig. 1); another branch passed north of the Chisos Mountains crossing the river at Lajitas. Once in Mexico, the Indians split into small groups, plundering until they were ready to return to their home domain. When the braves started north, they had slaves, livestock, and all the plundered wealth they could carry. Many prisoners as well as livestock perished. For years the trail was marked by bones. It was such a well-marked feature that during the early days of the Mexican Republic, it was designated as the boundary between the Mexican states of Coahuila and Chihuahua at Paso del Chisos.

The Mexican border states also suffered from Apache raiders who lived at their rancherias in New Mexico and in the watered valleys of the Big Bend's Davis and Chisos Mountains. Unlike the Comanches' annual carnival of murder and robbery, the Apaches, equally as devastating struck at unexpected times throughout the year. The results of raids by the two plundering tribes left many of the frontier haciendas depopulated, the villages deserted, the livestock driven off, and the fields uncultivated. In desperation, the State of Chihuahua made a separate treaty with Bajo el Sol, main chief of the Comanches. The general terms of the treaty stated that the Comanches were to receive the protection of the Chihuahua army; in turn, they were not to raid in the State of Chihuahua but were at liberty to raid in other states; they were to fight the Apaches wherever they found them. Raht (1919, pp. 66-67) gives this account of a battle. While this treaty was in force, Bajo el Sol, his wife, and younger brother were traveling in the Sierra del Carmen foothills above Boquillas when they met a band of about 30 Apache warriors.³ His wife begged him not to fight the Apaches but he replied that his treaty with Chihuahua bound him to fight Apaches any place he found them. Thus, he sent his wife and brother back to safety and prepared to attack alone.

---

³This place is probably in the hilly belt a short distance west of the Park road tunnel, about 1 mile northwest of the Rio Grande village. The Apaches may have lived at a rancheria on the present site of the Rio Grande Village but these facts have not been definitely established.

---

As he charged, the Apaches scattered among the rocks and bushes, but Bajo del Sol shot at them with his bow and arrows, killing two Apaches and wounding several. His arrows spent, he fought only with his spear, which his enemies could easily avoid in the rocky terrain. In some manner, one Apache had come into possession of an old shotgun ("escopeta") with only one charge. The warrior owning the shot gun hid behind a certain boulder and as Bajo el Sol was lured past the hiding place, the brave stepped out and shot him in the back of the head. The great warrior fell dead from astride his horse. And thus ended the career of one of the most notorious of the Comanche chiefs.

Presidio de San Vicente.—Presidio de San Vicente, about 10 miles up the Rio Grande from Boquillas, was built by the Spaniards during their campaigns to conquer and Christianize the Indians. Madison and Stillwell (1958, p. 50) reported that the Presidio was built in 1774. It was in ruins in 1857 (Emory, 1857, pp. 74-92); all that is now visible are small mounds of rock and adobe. Presumably, the chapel was a prominent structure and there was a stockade for the prisoners, but the individual buildings cannot be identified. The site is on a gravel terrace about 50 feet above the Rio Grande, some 2 miles southwest of the old abandoned San Vicente village in the Park. This Presidio gave its name to the mountain at the west, to the village (fig. 46), and also to the upper member of the Boquillas Formation, as the flagstone of that unit crops out near by. These Upper Cretaceous rocks crop out in the structurally low area that lies between the Sierra del Carmen on the east and Sierra San Vicente (mountain) on the west.

FIG. 46. One of the last inhabited places at San Vicente, Texas (1936).

There are historic facts and several legendary speculations centered around the old Presidio. According to Nelson (1935-36), Ugalde (one of the Spanish commandants who operated against the Indians in New Spain [northeastern Mexico and West Texas]) waged several campaigns against the Indians while he and his troops were at the Presidio. One of these demolished an Indian hacienda in Juniper Canyon of the Chisos Mountains (fig. 4) (in pocket). Partially substantiated by historic data, but mostly legend, is the story of the lost mine in the Chisos Mountains. The Conquistadors used the Presidio de San Vicente as a prison and from there worked a mine in the area. The guards blindfolded the captive Indians and marched them from the Presidio to the mine, supposedly in the Chisos Mountains (Lost Mine Peak?). There is an old trail that extends from the Presidio de San Vicente up Juniper Canyon, but if the trail ascended the slopes of Lost Mine Peak, the way is now lost in the debris along the mountain slopes. It was also said that the guards were able to see the cross at the Presidio from the mine entrance. The way to find the mine, if it ever existed, is also legend.

Those who search for the lost mine, and some do, must stand in the doorway of the old Presidio on Easter morning and look toward the Chisos Mountains. The spot where the sun's rays first light the peaks is the hidden doorway to the lost mine. The sun's rays strike the mountains at a slightly different angle most years because Easter Sunday is not on the same date, but if the sky is cloudless (regardless of the date), the sun's rays first strike Emory Peak, the highest peak, and not Lost Mine Peak.

If there was ever a Spanish gold or silver mine any place in the Chisos Mountains, it has not been found. The rocks in the high Chisos Mountains peaks that can be seen from the old Presidio are lavas of volcanic origin, and gold and silver do not usually occur in such an environment. There are a few small intrusions in the Chisos Mountains and some of these have been prospected but without favorable results. Many people have searched for the lost mine and other metallic deposits in the Chisos Mountains. One of these was the late Homer Wilson, a geologist, who divided his time between ranching and prospecting during the years 1929 to 1942. Wilson helped Harris Smith to develop the Fresno quicksilver mine in the northwestern part of the Terlingua district. He also found a gold nugget, about the size of a blackeyed pea, and several pieces of cinnabar in the Chisos Mountains (oral communication, 1937), in or near old Indian camps. Cinnabar was highly prized by the Indians for paint, and it is suspected that both the gold nugget and the cinnabar were brought into the area by them.

W. T. Burnham settled in the Big Bend country about 1914. His ranch headquarters was at Government Spring (fig. 4); he used most of the land surrounding Lost Mine Peak. During his ranch rides he did not discover a lost or hidden mine, but one day he did find a small bar of silver bullion (oral communication, 1936) in Smugglers Gap west of Lost Mine and Panther Peaks (fig. 4). Several years prior to 1914, however, a small smelter operated at the Mexico village of Boquillas. The processed ore came from the old Boquillas mine about 5 miles southeast of Boquillas and contained lead and zinc; some of the lead contained a little silver. Probably the silver bullion bar came from the Boquillas smelter and was lost from a pack animal.

San Vicente Mountain.—San Vicente Mountain dominates the country adjacent to the ancient Presidio and the abandoned village (fig. 47). The mountain is a large anticlinal fold shoved eastward by the forces that buckled the earth's crust at the end of the Mesozoic Era (fig. 26C). Erosion has stripped the soft rocks from the crest of the fold leaving the hard massive Lower Cretaceous limestone which forms the crest of the range. The Rio Grande detoured north around this barrier, so that there is a northward loop in its big southward bend. The small narrow canyon around the end of the mountain (Compton place) has long been a favorite fishing place for Big Benders. About a mile north of the canyon is a field in soft Upper Cretaceous shale and marl where C. G. Compton raised beans and maize prior to 1936.

FIG. 47. Aerial view looking southeast across the north end of Mariscal Mountain, Sierra San Vicente. and toward the Sierra del Carmen-Fronteriza escarpment south of Boquillas, Mexico. A, Cow Heaven anticline, B, Mariscal Mountain. C, Sierra San Vicente. D, Rio Grande, E, Sierra del Carmen, Mexico, F, Sierra Fronteriza, Mexico.

Mariscal Mountain.—Mariscal Mountain, also a long anticlinal ridge, is immediately west of Sierra San Vicente (fig. 47). The Rio Grande has cut Mariscal Canyon across the range about 10 miles south of the northern terminus of the fold. The canyon is about 1,600 feet deep and is carved through massive Lower Cretaceous limestone ledges (fig. 48). In the vicinity of the Rio Grande, Mariscal Mountain is separated from Sierra San Vicente by a down-dropped block, the Solis graben, about 2,000 feet deep (fig. 49). Both mountains were formed at the end of the Cretaceous (table 1), both increase in altitude southward into Mexico, and both are anticlinal folds. They differ in that Mariscal Mountain is overturned toward tile west and the crest of the fold is broken by a thrust fault (fig. 26D). The structure in the Cretaceous rocks at Mariscal is similar to that seen at Persimmon Gap (fig. 31), but no Paleozoic rocks are exposed. The rocks of Sierra San Vicente also dip steeply but they are overturned toward the east (fig. 26C) and no thrust faults have been recognized.

FIG. 48. The head of Mariscal Canyon, a gorge cut by the Rio Grande across the massive Lower Cretaceous limestone exposed in Mariscal Mountain.

FIG. 49. Sketch of a graben (sunken block) that illustrates the structure between Mariscal and San Vicente Mountains.

About the beginning of the twentieth century, D. E. Lindsey operated a small quicksilver prospect on the northern end of Mariscal Mountain about 8 miles north of the Rio Grande. On old maps, the location is shown as the Lindsey mine, but it is more commonly known as the Mariscal mine. Little is known of the early mine history except that Lindsey started the development and then sold his interests. Boss (1935, p. 561) reported that the property was operated under two different managements during the period 1917 to 1921 when 860 flasks of mercury were recovered. Since 1921, there has been some intermittent exploration and some mercury was recovered mainly by retorting the bricks from the old furnace in 1942-1943. The bricks were made locally from the Pen Clay (table 1) at a pit west of the mine, where the formation crops out immediately west of the mountain; the bricks were very porous and absorbed much of the mercury during the smelting operations. The machinery is now gone and the old mine buildings are in ruins (fig. 50). A small, low-grade fluorite deposit occurs near the top of the Lower Cretaceous limestone about a mile south of the old mine.

FIG. 50. Ruins of the old Mariscal (Lindsey) mine and part of Mariscal village as they appeared in 1963. During the World War I years, this was a flourishing community with several hundred inhabitants. Folds of the Boquillas Formation are shown and some of the ledges are sills in the Boquillas.

Cow Heaven anticline.—Cow Heaven anticline (fig. 51), one of the most picturesque folds in the Park, is northwest of Mariscal Mountain (fig. 4). It is part of a belt of folds that extends northwestward from Mariscal Mountain and passes beneath the lava in the highest Chisos Mountains peaks. The crest of the anticline is formed by a sill (fig. 22C, D) that underlies the highest part of the fold (Cow Heaven Mountain). Both north and south from the central Cow Heaven Mountain the drainage has breached the sill and here the crest of the fold is along a valley cut in the soft underlying Upper Cretaceous sandstone and shale. The right-hand side of figure 52 shows how erosion attacks folded rocks to form valleys. Rocks along the crest of an anticline are commonly more highly fractured than those along the sides. The fractures provide more opportunity for the agencies of weathering and erosion to attack the rock; this, with the increased elevation along the anticline's crest, localizes and promotes erosion.

FIG. 51. Aerial photograph showing folded rock at the north end of the Cow Heaven anticline. The spine in the background is Elephant Tusk, the elevation that appears to be in the left flank of the fold is Backbone Ridge, and farther left is Dominguez Mountain; all are intrusions. The flat-lying rocks in the central background are the tuffaceous rock of the Chisos Formation and the lava units of the South Rim Formation in the Chisos Mountains. These are deposited across the unfolded and eroded edges of the Cretaceous rock in the Cow Heaven anticline.

FIG. 52. Sketch showing how erosion develops valleys along the crest of an anticline. This is an artist's presentation of the topography and structure at Cow Heaven Mountain.

From Mariscal, a gravel road leads westward up the Rio Grande valley. During the winter months this road is normally passable (with care) but during the summer months there may be heavy rains. Most of the road passes over gravel that has been washed from the Chisos Mountains; the numerous arroyos have steep banks and the crossings are nearly at right angles to the road. In spite of difficult driving and slow rate of travel, there is much geology, vegetation, scenery, and human interest to be enjoyed from a trip along this river road, which eventually reaches Castolon and Santa Elena Canyon.

West of Mariscal Mountain is a broad valley eroded from soft Upper Cretaceous sandstone and shale that on some early maps is shown as Paso del Chisos (fig. 1; pp. 30-35). From this valley the hogback ridges on the west side of Mariscal Mountain are clearly visible (fig. 47) and in the distance (south) is a dim view of the head of Mariscal Canyon (fig. 48). The road continues westward, crossing the south end of the Cow Heaven anticline, and a cross section of the structure is visible toward the north (fig. 51). Farther northwest are the Elephant Tusk (fig. 53), an igneous plug that penetrated the north end of the Cow Heaven anticline, and Backbone Ridge and Dominguez Mountain (fig. 54). They are also plugs that deformed the flanks of the Cow Heaven fold, and on the northwestern skyline, the South Rim of the Chisos Mountains is visible. Farther northwest and closer to the Rio Grande is Punta de la Sierra (Point of the Mountain, fig. 55) and Mule Ear Peaks and toward the west is Sierra de Chino (Curly Mountain).

FIG. 53. Elephant Tusk (igneous intrusion) with dipping beds of the Chisos Formation in the foreground.

FIG. 54. Dominguez Mountain as seen from the river road. Most of the ridges are dikes, several hundred of which radiate from this intrusion like spokes from the hub of a wheel.

FIG. 55. Punta de la Sierra (the point of the mountain). Most of the ridges that extend beyond the main escarpment are dikes whose source was the Dominguez Mountain intrusion.

Hundreds of dikes radiate from Dominguez Mountain like spokes from the hub of a wheel. Many of them cut the lava at Punta de la Sierra and are mainly responsible for forming the high rugged outline of that feature. The road follows the river westward and at one place, a black rock dike crosses the Rio Grande (fig. 56). This "lava"-filled fissure is one of the many dikes that radiate from Dominguez Mountain, and it can he traced intermittently southwestward from the mountain to near the high Sierra Ponce escarpment in Mexico. This dike may have been the source of some of the lava that is now exposed along the road toward Castolon. The complex assortment of lava and ash beds in that area are included in the Chisos Formation (table 1).

FIG. 56. A dike, which can be traced most of the distance from Dominguez Mountain southwestward to near the foot of the Sierra Ponce escarpment in Mexico, is exposed in the channel of the Rio Grande.

On the sloping plain south of Punta de la Sierra, about 20 miles west of Mariscal, is a broad area that still retains evidence of having been used as a farm and abandoned landing field; this is Johnson's ranch (fig. 57). The flattish surface is on the soft clay of the Javelina Formation, but the ancient slope is now dissected by many arroyos that head in the Chisos Mountains. The inter-arroyo flats are covered by gravel deposited by sheet wash from the Chisos Mountains; the flood plain along the Rio Grande is blanketed by river silt. Except for the climate and distance to market, it is an ideal place for an agricultural development (fig. 58). A farm project was started in 1924 by two Kentuckians, Grady and Williams, who tried to raise tobacco and who started a trading post. The tobacco farm was unsuccessful, but the store soon developed into a flourishing business. The property was purchased by Mr. and Mrs. Elmo Johnson in 1928. The store continued to have a thriving trade but the Johnsons' cotton farm and goat ranch showed very little profit because it was too far to market.

FIG. 57. The Johnson ranch home and trading post in 1936.

On July 6, 1929, the United States Army dedicated the landing field (fig. 58) at Johnson's ranch. This isolated outpost served a dual purpose; its most important use was in the training of young aviators; secondarily, it was a lookout station and check point on the International Boundary. A radio station was installed which for a time had hourly contact with Fort Bliss in El Paso, daily communication with Fort Sam Houston in San Antonio, and emergency contact with Fort D. A. Russell in Marfa. Many young aviators received their mountain-flying training over the Big Bend country, and scores of now famous people spent the night or week-end as guests in the Elmo Johnson home. One of them was a young lieutenant, Nathan Twining, who later became General Twining, Chairman of the Joint Chiefs of Staff. On the opposite side of the ledger, the name of John Dillinger also appeared in the Johnson's guest book.

FIG. 58. Aerial photograph of the Johnson ranch area in 1936 showing Rio Grande, fields, and landing strip; Chisos Mountains in the background.

The Big Bend trading posts were not the supermarkets of today but they performed a necessary service, and their operators, along with the ranchers, the cavalry, and the law enforcement officers, participated in the settlement of West Texas. The first stores were at springs, water-holes, or other overnight camping places along the oxen- or mule-drawn freight, overland stage, and mail routes. The early trading posts were small and were the prime targets for the Indian raids during the 1870's and 1880's and for the outlaws who roamed both sides of the Rio Grande, robbing or killing anyone whom they believed might have money. Although some of the trading post sites were abandoned, a few of them withstood the years of terror, and when the railroads came and the highways were improved, the general trading post sites grew into the present West Texas towns.

The trading posts in the lower Big Bend did not enjoy a phenomenal growth, for they were served neither by railroads nor highways. Their customers were and still are from the local area. The small trading post was a necessity and such trading posts will probably continue to exist in isolated areas where there are no roads and where many of the inhabitants travel by burro from the Mexican side of the river. The stores varied in size, isolation, and the quantity and variety of stocked merchandise. Most of them were on the Texas side of the Rio Grande because it was easier to obtain supplies from the United States markets. These operators not only sold merchandise but purchased products from their customers who could, in turn, purchase store items as needed. Most deals were by barter without exchange of money.

The best customers were the fur traders, who came with one to several burro loads of pelts from wild animals trapped in the mountains; during the winter months the quality of the fur was good. The furs included mostly fox, bobcat, coyote, skunk, and ringtail, but there were also a few beaver, javelina, and panther pelts, and some goat and deer hides (fig. 59). The price of pelts varied, but normally it was from 40 to 50 cents up to $1.00 each. When the furs were sold, the traders would then invest their total assets in merchandise from the store, pack it on burros, and go back to Mexico where they would barter for more furs.

FIG. 59. Furs at the Johnson ranch trading post in 1936. Elmo Johnson behind the counter and Texas Ranger, the late Pete Crawford, posing as a customer. This scene could be repeated at most Big Bend trading posts during the mid-1930s.

Customers other than the fur traders were the people who lived in small settlements as far away as 100 miles south of the border. These individuals farmed small fields where they raised frijoles (pinto beans), melons, squash, corn, and occasionally fruit and pecans. They also had to obtain merchandise from the trading post and their most common items for barter were candelilla wax, wood, chino grass hay, ropes, blankets, and goat skins. They never brought as much merchandise to the store as did the fur traders, for generally their products were the fruit of one family's efforts. They traded for necessary supplies that had to last until they had something else to sell.

The only items carried in the trading post stock were those not grown locally. Mostly, these included coffee, sugar, rice, salt, stick cinnamon, stick candy, and cigarette tobacco. Canned goods were few; Karo syrup, lard, sardines, and some fruit were most common. Prunes and raisins were sold in bulk, as was American cheese. The cheese came in large blocks and was sliced and sold a nickel's worth at a time. Many customers purchased the cheese, a five-cent box of crackers, and perhaps a can of sardines, and had lunch in the trading post. Most trading posts had a few pairs of shoes, some huaraches (sandals) made from skins and later from automobile tires, work pants (jeans or khakis), men's hats, a little jewelry (mostly earrings), and a few bolts of bright colored calico.

The trading post business flourished from about 1920 to 1930. This was a time after the Mexican revolution when there was peace and the people could accumulate a few articles for trade. After about 1930, however, the ladies' dress styles changed in the United States and furs were no longer in fashion. The bottom fell out of the fur market, which was a hard blow to the Mexican people as well as to the trading post operators. Except for the candelilla wax industry, which had a heyday during the early 1940's, the Mexican had little merchandise to barter, and many of the trading posts closed down.

The Johnsons served their community in ways other than through the trading post. They befriended the border people during sickness and death and administered what first aid they could. It was in the Johnsons' home that many border children saw their first Christmas tree. Each child received a bag of homemade candy and cookies, the first present that most of them had ever received. The Johnsons' Christmas party became a tradition along the river that has been continued by the Park residents. This policy has been extended by the Park personnel in recent years and now many of the children have received polio, smallpox, and diphtheria immunization shots, as well as school books, through the courtesy of the Park employees.

Glenn Spring.—Glenn Spring. on one branch of the old Comanche Trail, is a dependable water supply—water is a very scarce commodity in the Big Bend (fig. 4)—and that is the principal reason why the Glenn Springs village was destined to become one of the early settlements. Frequent use of the spring by the Indians is indicated by the bedrock mortars, grinding stones (fig. 60), and caches of flint and spear or arrow points. The spring is adjacent to the Glenn Spring laccolith (fig. 61). Several dikes extend away from the intrusion (fig. 62). The dikes are harder than the adjacent rocks and remain standing as "rock walls."

FIG. 60. Mortar (rock mill) in a boulder. These are common near watered Indian camps in the Big Bend. Mortars in small boulders may have been carried from camp to camp.

FIG. 61. Tilted rock ledges elevated by the Glenn Springs intrusion. Upper surface of the intrusion (A) in the right center area. The old road from Glenn Spring to Mariscal descends the flank of the intrusion and crosses hogbacks in the Aguja Sandstone. In the background are (B) the Elephant Tusk and (C) Backbone Ridge, both intrusions, and (D) Punta de la Sierra, capped by a lava.

FIG. 62. Dike filling of a fissure which originated from the Glenn Springs intrusion. The molten rock cooled and hardened in the fissure. The intrusive rock, being harder than the adjacent sedimentary beds, was left like a rock wall that now rises above the adjacent rocks and can be traced for several miles.

The first settler was a Mr. Glenn who grazed a remuda of horses in the area. In order to provide a better water supply, Glenn dug out and walled the largest spring (fig. 63). He was killed by the Indians near the spring that bears his name, and from that time on, which was probably in the 1870's, the community of Glenn Springs has had a turbulent history. During the second decade of the twentieth century, there were several ranches near Glenn Springs and the village was the social center. W. K. Ellis operated a store, rendered candelilla wax, and carried on the necessary activities of a frontier trading post. Ellis, his wife, the C. G. Comptons, and their three children lived in the village. About fifty Mexican families lived nearby in small huts (jacales) (fig. 64) and worked at the candelilla wax-rendering plant, which had a daily production of about 1,000 pounds of raw wax. It was a quiet friendly community and on May 5, 1916, some of the villagers were preparing for the Cinco de Mayo (May 5) fiesta, which is an important holiday and a time of celebration for the Mexican people.

FIG. 63. Glenn Spring in 1936. This natural water source was cleaned and walled by Glenn about 1870 when he used this water for his horses. The spring was dry in the late 1940's but began flowing again in the early 1950's. Mesquite and other brush are now so thick that it is difficult to reach the spring.

FIG. 64. Some of the abandoned adobe jacales as they appeared in 1936; now not a wall is standing in the village.

During the few weeks prior to this time, a band of desperate renegade Mexican bandits rode northward toward the Rio Grande. The leader, Rodriguez Ramirez, recruited additional men along the way. In Mexico, during those unstable times, it made little difference to the followers whether the leader was a Carranza or Villa sympathizer, for political affiliations made little difference if they were provided with food. The bandits crossed the Rio Grande at Paso del Chisos west of Mariscal Mountain (fig. 4) where they were joined by relatives and friends on the Texas side. After crossing the river, the bandits divided into two groups, one party going to the Deemer store at the east end of what is now the Rio Grande Village, the other riding toward Glenn Springs. This latter group camped in the shelter of nearby badlands or stayed in the Glenn Springs village with relatives or friends. The presence of strangers in camp was no cause for alarm, for it was Cinco de Mayo (May 5); there would be a baile (dance), and people from the ranches or from across the river were welcome. About 11 o'clock that night, the bandits moved against Glenn Springs village which was then garrisoned by nine soldiers of the Fourteenth Cavalry (fig. 65).

FIG. 65. The Glenn Springs battlefield site in 1936. The southeast end of Chilicotal Mountain is seen in the background. All the corral fences are now down and the adobe buildings are represented by a pile of adobe rubble. The old fort once stood near the left central part of this picture.

Reports⁴ differ as to the number of bandits making the raid; they also differ as to whether some soldiers were on guard or all asleep and if asleep, whether the soldiers were in the small adobe fort, outside because of the hot weather, or in tents, The bandits attacked the soldiers. some yelling "Viva Villa!" and others "Viva Carranza!" The battle endured for several hours and the handful of soldiers withstood several attacks from a much larger force. The bandits were unable to drive the soldiers from the small adobe fort, but the bandits observed that a thatching of rendered candelilla wax plants had been used to build an arbor over the small fortress to protect it from the sun. They proceeded to throw blazing firebrands onto the thatched arbor. The soldiers were soon "burned out," three were killed while trying to escape, one was seriously wounded, and at least three others seriously burned.

---

⁴Additional information regarding the Glenn Springs raid have been reported by Raht (1919, pp. 350-358), Madison and Stillwell (1958, pp. 39-42), Wood (1963), and E. E. Townsend (personal communication, 1948).

---

The bandits looted the store (fig. 66), killed one of the Compton children, and ransacked everything, taking food that was easily packed and clothing. C. D. Wood (1963, p. 67), who lived about 2-1/2 miles from Glenn Springs, heard the shooting and saw the fire. He and a neighbor, Oscar de Montel, went on foot to investigate. In the darkness they passed through the Mexican part of the settlement. Near the store, however, they were challenged by one of the bandit sentries, were fired upon, and retreated to the cover of nearby rocks. At dawn, Wood saw about 25 of the raiders saddle and pack their horses, pack the horses belonging to the soldiers, and move toward the Rio Grande. Following the bandits' departure, Wood (1963, p. 69) entered the village. He found the body of one of the raiders and seven pools of blood. If more members were in the band, they left unnoticed before daylight, and if additional bandits were killed, their bodies were probably carried back to Mexico by their comrades. The number of bandits in the raid and the exact toll will never be known.

FIG. 66. The Glenn Springs store in 1936. Water from the spring was piped to the hydrant at the foot of the cottonwood tree. Leaks in the water line and thirsty visitors wasted enough water so that the tree could grow. Now the hydrant is gone, the tree is dead, and only a pile of rubble marks the site of the store.

The bandit detachment that rode toward Boquillas completed their looting without loss of life on either side. At the Deemer store (east end of the Rio Grande Village), they took food, clothing, and money, and compelled Jesse Deemer, the store owner, Pablo Alcala, a clerk, and Monroe Paine, a Seminole-Negro, to go with them into Mexico (all three were rescued a few days later at El Pena by Major Langhorne and troops of the United States Eighth Cavalry who followed the bandits into Mexico). Next day (May 6), some of the bandits went to Boquillas where they captured the supervisory personnel and stole the payroll at the Boquillas mine. Here they also captured a doctor, Homer Powers, and four other Americans. The bandits placed their five American prisoners in a truck loaded with provisions and, with nine bandits as a guard, started south.

One of the Americans was compelled to drive the truck, but the captives were determined not to go far into Mexico; a few miles south of Boquillas, the driver intentionally stalled the truck in a broad arroyo. Dr. Powers quickly gained the confidence of his captors by administering to their needs. There are various versions of the ensuing events (Raht, 1919, p. 352; Madison and Stillwell, 1958, p. 41; Wood, 1963, pp. 69-71), and the following account is one of them.

It became evident that the bandits did not understand the mechanics of operating a truck, and there was also difficulty in communication because of the language barrier. Finally, the captors were made to understand that the truck was overheated, so two of the guards were sent to search for water. Two more guards were sent to get mules to pull the truck from the arroyo or to pack the supplies if they could not free the truck. When the two parties of searchers were out of sight, the doctor convinced the captors that if all of them pushed, maybe they could dislodge the truck. Now it was five against five, an even match, except that the bandits were armed. The driver raced the motor, engaged the clutch just enough to give encouragement, and at a given signal while all the bandits were pushing, the Americans quickly wrenched the rifles from their captors and the bandits became prisoners. These men were taken to the Texas side of the Rio Grande where they were brought to trial at the Brewster County district court and given life sentences in the Texas penitentiary (Wood, 1963, p. 71).

The Glenn Springs raid, the Deemer store raid, and all incidents related thereto did not stop the violence or the bitterness. For several years afterward, any Mexican national was regarded with suspicion in the area. Latin-Americans known not to have been connected with the Glenn Springs raid continued to work north of the Rio Grande, but all of them carried identification.

The Chisos Mountains.—The Chisos Mountains with elevations up to 7,800 feet are the highest features in the Park (fig. 67). They are near the center of a structural trough, a large irregular graben about 40 miles wide (fig. 49), bordered on the east by the Santiago-Sierra del Carmen Mountains and on the west by Mesa de Anguila (fig. 4). This trough was first formed during the uplift that also folded the Santiago and Mariscal Mountains and elevated a mountainous ridge that is in part exposed beneath the eroded lava that caps part of the Chisos Mountains. In general, however, the trough floor was planed nearly flat by erosion that occurred after the Cretaceous period. In early Cenozoic time the trough's surface was much lower than the highlands on either the east or the west sides and therefore was the site for deposition and accumulation of sediments. The oldest rocks deposited on Tornillo Flat are clay and sandstone that contain mammal bones and teeth (pp. 24-25).

FIG. 67. Green Gulch, the only entrance by road to the Chisos Mountains Basin. Prior to 1934, only a horse trail ascended this pass; now it is a paved highway. Construction of this road, which was started by boys in a CCC Camp, was the initial construction prior to development of the area as a National Park. Casa Grande "stands as a guard" at the head of the pass. Subsidiary peaks of the Pulliam Mountain intrusion are on both right and left. Lava that caps some of the peaks forms the elevation on the left skyline.

The rocks in the Chisos Mountains are mainly of igneous origin and were erupted and deposited during the Cenozoic Era, but the rocks are younger than those deposited on Tornillo Flat. The oldest rocks in the mountains are volcanic ash mixtures of ash and clay, sandstone, and conglomerates. This sequence is overlain by thick massive lava (fig. 68), and both the sedimentary rocks and lava are penetrated and deformed by intrusions. Both intrusive rocks and massive lava cap form the high elevations in the mountains. In general, the intrusions eroded to dome-shaped peaks (Vernon Bailey Peak, fig. 69) and the lava caps square-faced peaks and mesas (Casa Grande, fig. 70).

FIG. 68. A massive lava flow, about 50 feet thick, that forms the South Rim of the Chisos Mountains.

FIG. 69. Vernon Bailey Peak, a dome-shaped intrusion at the northeast (right) side of the Window.

FIG. 70. Casa Grande (Big House), a lava-capped peak that stands guard over the Basin of the Chisos Mountains. Another view of this picturesque mountain is shown in figure 67.

The road from the Tornillo Creek bridge to Panther Pass in the Chisos Mountains ascends about 3,000 feet. The route is via Green Gulch, a beautiful canyon carved in the flank of the Pulliam Peak intrusion (fig. 67). The grades are steep, the roadway sinuous, and cars can become overheated. Water barrels for filling radiators have been placed at vantage points, and the visitor can enjoy the scenery while waiting for the motor to cool.

Alsate's Face.—Near the mouth of Green Gulch the road passes the eastern end of the Pulliam Peak intrusion. Here erosion has sculptured the intrusion; from certain places, in the proper light, the mountain profile takes the shape of a man's face that is looking skyward (fig. 71). The mountain profile is locally known as Alsate's Face.

FIG. 71. Alsate's Face, the profile of a man's face looking skyward. It is best seen in favorable light, about 1 mile east of the Basin road junction. This picture, however, was taken close, and the "chin" appears too long. This "profile" is the result of erosion on jointed intrusive rocks of unequal hardness.

Alsate was an Apache chief who was a brave warrior and one of the great Apache leaders of his time. One story relates that Alsate and his band were hard pressed by the Comanches and took refuge in the Chisos Mountains. The fighting continued and Alsate was finally killed. According to the legend, when Alsate fell, the earth shook, rumbled, and moaned, and the mountains rose. When all was again quiet, the Comanches saw the profile of Alsate's face on the mountain top. The enemy warriors left immediately and in great haste, for they believed that the mountain stronghold was still in the custody of Alsate's ghost, and so long as his face remained on the mountain top, the Chisos Mountains were the happy hunting grounds for the spirits of Alsate and his braves.

Another story of Alsate's last days goes back to 1880 (Madison and Stillwell, 1958, pp. 26-27) when a hand of Apaches was making its last stand in the Chisos Mountains. This tribe was "sort of hemmed up" in the Chisos Mountains by the Comanche Indians to the north and by the Mexicans, whose ranches and villages Alsate and his tribe raided at every opportunity, on the south. Finally, a spokesman for the Mexican garrison at San Carlos, Chihuahua, Mexico, came to Alsate and invited the entire tribe to a fiesta. Now a fiesta in the Big Bend country used to mean everything to eat and drink and dancing for days. Alsate and his tribe, intrigued by the promise of food (for they were hungry), went to San Carlos. On the first night of the fiesta, Alsate's band wined, dined, and got gloriously drunk. The next morning they awakened to find themselves in chains. The Mexicans rejoiced and had their own fiestas because Alsate had been captured. They planned to march the Indians into the interior of Mexico to be sold into slavery so that Alsate would trouble them no more.

Alsate, his squaw, and a few followers, however, managed to escape their Mexican captors (Raht, 1919, p. 277). The chief dropped out of sight, and for years most people believed him dead. After a time, however, sinister rumors began to creep over the countryside about Indians in the Chisos Mountains. Some persons claimed to have seen Alsate. Others, on awakening in the morning, found moccasin prints in their camps. Some were large, others small, as if made by a man and a woman. Most persons believed the prints were made by Alsate and his squaw, and the makers became known as "Big Foot" and "Little Foot." The campers never saw or heard anyone during the night and only food was taken.

Finally, the San Carlos authorities sent customs guards to search for the track makers. In searching about the mountains where the "ghosts" and tracks were reported to have been seen, the guards found a cave with signs of recent occupancy, but they did not find the inhabitants. The users had been able to carry in grass for a bed, to kill and eat birds, rabbits, and other small animals, and to build a fire for cooking. As the Indian "ghost" was often seen in the vicinity, the cave became known by the name "Cueva de Alsate" (Alsate's Cave), but as Alsate's ghost harmed no one, the official pursuit was stopped. In later years, some of the more venturesome people entered the cave. In an obscure corner lay the mummified remains of Alsate, the Apache chief. Nearby were the charcoal remains of a fire and a small bed. The squaw was not found, and it was supposed that she had previously returned to her people in San Carlos (Raht, 1919, pp. 278-279).

The Watchman's House.—A few miles farther into Green Gulch, to the left of the road and high on the slopes of Lost Mine Peak, is a small cave that looks like a little house and is called the Watchman's House (fig. 72). According to legend, this cave is occupied by the ghost of the Indian slave who was left to guard the mine worked by the Spaniards in the Chisos Mountains during the period that the Spanish garrison was headquartered at Presidio de San Vicente. The superstitious will not enter the cave but those who venture up the steep mountain slope to the cave's entrance find that it is a shallow re-entrant in volcanic rocks. Most agree that the legend and the view from below are more interesting than the cave itself.

FIG. 72. The Watchman's House, dark spot in light rocks, extreme left center. Although this is one of the features on Lost Mine Peak (high center) and most of the related stories are interesting, this little cave is the result of weathering and erosion. Arrow points to cave.

At the head of Green Gulch the road ascends Panther Pass and, beyond the pass, descends into the Basin (fig. 73). The Basin is a depression about 1,500 to 2,000 feet deep in the center of the highest Chisos Mountains peaks. It is about 3 miles in diameter. All the drainage flows through the Window (fig. 74), a narrow slit separating the Ward and Pulliam intrusions. One look at the Basin and the first question is—what caused it?

FIG. 73. This is the Basin as seen from high on Casa Grande during a mild dust storm. The buildings in the central area are the CCC Camp (1936), structures. This picture contains a suggestion of the mystery which has always shrouded the Chisos Mountains.

FIG. 74. The Window, or pour-off as it is more popularly called. This gorge is the outlet for all the Basin drainage—the vertical fall below is about 75 feet; Burro Mesa is pictured in the background.

The Basin.—Most people seeing the Basin for the first time suspect that it was formed by some catastrophic phenomenon—that it is perhaps a volcanic crater or the result of glacier erosion during the ice age (fig. 75, in pocket). However, no evidence for either of these earth-changing phenomena has been found. The Basin is the result of the slow processes of weathering and erosion during the geologic past. Cenozoic-age volcanic ash, sandstone, conglomerate, and lava were deposited in near-horizontal layers over the ridge of older Cretaceous rocks. These flat-lying rocks were elevated by intrusions into a broad dome during the Cenozoic Era. The fractured rocks forming the crest of the dome were eroded more rapidly than the lower rocks. Eventually, part of the lava cover was removed, exposing the underlying intrusive rocks; erosion continued, breaching the lava until the underlying softer rocks in the center of the Basin were exposed.

FIG. 75. Panoramic view of The Basin in the Chisos Mountains, Big Bend National Park, Texas. A, Emory Peak; B, Ward Mountain; C, Carter Peak; D, Vernon Baily Peak; E, Pulliam Peak; F, Casa Grande; G, Toll Mountain; H, Concession development; J, Campground; K, Saddle - horse corral. (click on image for an enlargement in a new window)

The rocks in the Basin include soft Upper Cretaceous limestone, shale, and sandstone like those found near San Vicente along the Rio Grande at elevations 4,000 feet lower, and they contain the same species of fossil clams, oysters, and snails.

Sapping was an important process in forming the Basin. The softer volcanic ash and sandstone beds underlying the massive jointed lava cap were more easily eroded than the lava. The lava was undermined and fell into the valley below, and as the sapping continued, the lava cliffs retreated, Weathering also aided the process. The daily range of temperatures caused the rocks to expand and contract, thus forming small cracks and widening larger ones. Frost action pried apart rock fragments and gravity caused them to fall. Rocks moved down the slopes toward the Basin floor and rock slides occurred on the higher slopes of some peaks. Masses of rock fragments and soil accumulate at their natural angle of repose. As the mass expands and contracts, the fragments inch their way down the slope, sliding occasionally when lubricated by rainfall.

The South Rim Trail.—The highest Chisos Mountains peaks may be reached from the South Rim trail. The trail is about 13.6 miles long (round trip) and although it is suitable for hiking, most people take the trip on horseback. The trail starts in the Basin, passes through the gap to the west of Emory Peak (fig. 76), and extends southward to the mountain rim. The lower part of the trail is on fragmental rock debris that has tumbled down the mountain slopes, but higher in the Basin rim the trail crosses fossiliferous Upper Cretaceous formations containing clams, oysters, and snails. High up in the mountains, only lava or associated extrusive rocks occur. They are part of the South Rim Formation that caps the Chisos Mountains (table 1, p. 27). The massive character of the lava can be seen in the face of the South Rim (fig. 68) and in Emory Peak, where the base of the lava cooled so fast that it is glassy. The jointing that caused the rocks to break into large blocks is well developed in the top of Emory Peak (fig. 77). One of these fissures extends about 30 feet below the debris-covered slope and has been the site for buried-treasure stories. The so-called "water tower" on the northwest flank of the mountain is a rock remnant left standing when the weathered materials in the adjacent joints were eroded away. Similar rock columns formed by weathering and erosion along joints occur in the lava rim between Toll and Emory Peaks (fig. 76). The Cowboy Boot southeast of Emory Peak was formed in like manner (fig. 78).

FIG. 76. Looking south across the Basin concession area toward Emory Peak (A). This is a view of the third highest point in the State of Texas. The peak was named for Major Emory, the first director of survey for the Mexico and United States International Boundary Commission. Toll Mountain is the flat-topped mountain at the left (B).

FIG. 77. Rock jointing in the top of Emory Peak.

FIG. 78. The Cowboy Boot, a prominent feature in the highest Chisos Mountains. This is a rock column that has been sculptured by weathering and erosion. (Photograph by Hunter's of Alpine, Texas.)

The South Rim has an elevation of more than 7,000 feet and is terminated on the south by an escarpment up to 2,000 feet high (fig. 68). Its surface is approximately a mile above the Rio Grande, 16 miles to the south. From this point of vantage, on a clear day, the visitor may see for at least 100 miles into Mexico and an equal distance in the opposite direction. This panoramic view is one of the finest in the Southwest.

The vegetation also changes on the upward climb. On the lower slopes of the Basin, the predominant trees are Mexican piñon pine, alligator and drooping juniper, and Texas madrona (fig. 79); there is a wide variety of shrubs, grasses, and some cacti. High in the mountains, there are small stands of aspen, ponderosa pine, maple, Arizona cypress, and Douglas fir (fig. 80). These are growing in a small isolated area hundreds of miles from stands of similar trees in the Rocky Mountains. How did they get there? One explanation is that the cold-weather trees and plants advanced southward as the great Pleistocene glaciers moved south. As the glaciers retreated, the climate warmed, and the cold weather flora died except in the highest and coolest elevations which were, to them, the most favorable habitat. If the Big Bend climate should grow warmer or if the rainfall in the Chisos Mountains decreases, these few stands of residual cold-weather flora probably would disappear.

FIG. 79. The piñon-juniper forest on the higher slopes of the Chisos Mountains.

FIG. 80. Piñon forest in the highest Chisos Mountains.

Boulder Meadows.—It is an easy hike to Boulder Meadows in the southern part of the Basin. The trail is on gravel that overlies Upper Cretaceous sandstone and shale. The common trees are oak, piñon pine, alligator and drooping juniper, and Texas madrona (fig. 79). In a short distance most of the trees and shrubs disappear, and beyond are grass-covered meadows with only scattered trees and boulders as large as 20 X 20 X 20 feet in diameter. The boulders are blocks from the high lava ledges that have tumbled down the mountain side. They have come to rest on clay; when the clay is wet, the boulders inch their way down the slope. Some of them are in crude rows which mark the present line of forward movement. Shallow temporary pools of water are sometimes found in the flat depressions behind the boulders and here the grass grows, forming the meadows.

The Lost Mine Trail.—Some people who take the Lost Mine trail are secretly looking for the lost mine, but most of them take the climb to enjoy the scenery, the vegetation, and the wildlife. The rocks are mostly lava but a few dikes of igneous rock filling fissures are seen along the way (fig. 22J). The most prominent dike is at the Chinese Wall, the ridge between Lost Mine and Casa Grande peaks. The dike is harder than the adjacent rocks and erosion has left the dike rock standing as a ridge which forms the divide between Green Gulch and Juniper Canyon.

The view from the orientation station on Lost Mine ridge is second only to the view from the South Rim. One can observe certain peaks in the southern Chisos Mountains, the badlands beyond, and features in Mexico. There is a commanding view of Juniper Canyon. Two springs are in the canyon. The upper spring is believed to be the place where Ugalde in 1787 met and defeated the Apache warriors from a small rancheria (Nelson, 1935-36). It was also in this canyon that a meteorite weighing about 2 tons was found in 1915 (Barnes, 1910, p. 590). Unsuccessful search for the meteorite was made in 1939; presumably it has been covered by debris slumped from the steep canyon walls.

The Window Trail.—The trail to the Window crosses Upper Cretaceous shale and slabby limestone (flagstone)—the same formations that crop out along the Rio Grande near San Vicente. The rocks were elevated to their present position by the forces that raised the Chisos Mountains. In other places, the trail crosses great gravel ridges—this gravel is rock debris carried by running water from the higher mountain slopes and deposited on the lower Basin floor. Most of the ridges are being slowly dissected and eventually the intermittent streams will carry and roll most of the debris through the Window and onto the lowlands beyond.

Oak Creek flows through the Window, and there is little or no water in the gorge during most of the year, but with heavy rain in the mountains, Oak Creek becomes a raging torrent. At such times, the banging and bouncing of the rocks makes a terrific roar as the rock load grinds forward. At the Window's brink, one can see the boulders bounce like corks as they are pushed over the edge to drop 75 feet to the bottom of the falls below, where there is a great pile of bouldery debris. When Oak Creek is in flood stage, there is little doubt that stream erosion excavated the Basin and carved the narrow Window slit through the massive intrusion that walls the Basin on its north and west sides (fig. 74).

Occasionally strange lights are seen in the Big Bend area. On certain nights, a luminous pulsating light plays restlessly on the mountain peaks or along the roads. Some have tried to bump the light or to run over it in automobiles. The cowboys have tried to shoot or rope it, some have chased it with clubs and stones, and others have become frightened and run away. The light changes in tint, form, and movement, and normally disappears just before the pursuer reaches "it." Some of the more superstitious believe that the lights are warnings from the departed Indians; some believe that lost mines glow at night; others believe that the lights are spirits that are taking back the hidden treasures from the mountain side. A more likely explanation is that the light is the phosphorescent glow (foxfire) from rotten wood or the reflection of light from tiny mineral grains in the rocks.

The Chisos Mountains in any light are magnificent, but one of the finest views is from the western side at sunset. William Ferguson recorded his impressions of the Chisos Mountains in the San Antonio Express in 1896, as follows:

Nowhere have I found such a wildy weird country. The very silence is oppressive. A man grows watchful for his own safety and becomes awe-struck by nature in her lofty moods. Emotions are stirred by the grandeur and beauty of the scenery and the ever-changing play of light and shadows. Never have I beheld such a display of glory as falls at sunset on the bald head of the Chisos Mountains, as witnessed at a distance of 25 miles. No painter could mix colors to justly portray it. No words can describe its splendour. First orange, then pink, then crimson, and last of all darkening purple, threw tints on the mountain's dark background and all faded insensible into neutral twilight. An old Mexican, who could neither read nor write, stood beside me. I pointed to the Chisos. His countenance lighted up. He exclaimed, "Bonito." It was an eloquent tribute.

A trip to the Big Bend is not complete without taking a drive on the north and west sides of the Chisos Mountains. From the road, Pulliam Peak is clearly visible to the south (fig. 81). The nearly flat ledges in the slopes are mostly lava; weathering and erosion along joints in the extrusive rocks have formed precipitous cliffs and columns, and the bald head of the Pulliam Peak intrusion is exposed along the sky line. The contact between the intrusive and extrusive rock is mostly covered, but some washes are clear and in these places the contact is visible. Farther west one can obtain a view of the Window and beyond, to the right and ahead, is Burro Mesa.

FIG. 81. Looking south at the north side of Pulliam Peak. The high elevation at right is Vernon Baily Peak (an intrusion); the Window is beyond. The high peak at extreme left is the back side of Alsate's Face (an intrusion); the Window is beyond. The massive columnar rocks in the high central area and the ledges below are mostly lava. Rock of the Aguja and Javelina Formations (table 1) are exposed in the foreground; some of these contain fragments of fossil wood and dinosaur bones.

Burro Mesa.—Burro Mesa, named for the herds of wild burros that once grazed there (fig. 82), is one of the structurally low sunken fault blocks in the Park. This earth segment was lowered along the Burro Mesa fault, which closely parallels the east face of the mesa (fig. 83), and the displaced rocks in the fault zone can be seen from the road near the north end of the mesa. Rocks along the fault dip vertically and are easily distinguished from the nearly horizontal rock layers on either side of the fracture. The highest lava unit on Burro Mesa is the same lava that caps Emory Peak in the Chisos Mountains, and the displacement (about 3,500 feet) along the fault is approximately equal to the difference in elevation between the highest point on the mesa and the top of Emory Peak. In most fault blocks, the down-dropped segment is lower than the opposite side. At Burro Mesa, the reverse condition is present. Here the hard lavas in the down-faulted block are more resistant to erosion than the soft sandstone and shale on the upthrown side. In time, erosion removed the soft rocks and now they crop out at elevations lower than the bard massive lava in the downfaulted block (fig. 83).

FIG. 82. A small group of wild burros which are commonly seen in the lowland desert areas. Occasionally there are 20 or more in a group. Most of them have never submitted to rope or saddle and are as wild as deer.

FIG. 83. Looking west at the Burro Mesa fault-line scarp. This escarpment is the result of erosion. The downdropped block is of more resistant rock and now forms the high side of the fault.

The lava and ash beds exposed on the surface of Burro Mesa were folded as they moved along the fault. They form a syncline (fig. 21), so that lava units that crop out on the eastern and western rims are buried beneath younger rocks in the middle of the mesa. The extrusive volcanic rocks were further deformed by intrusion of igneous rocks (fig. 84)—one of the plug-like intrusions is shown in the artist's sketch of the mesa (fig. 85). The Burro Mesa fault also cuts diagonally across the Chisos Pen anticline (fig. 85), which was folded after the Cretaceous rocks were deposited and before the lava and ash beds that now cap the mesa were extruded. The folds are well exposed northeast of the road near the north end of the mesa. The dark hard layers penetrating the yellowish-brown sandstone beds are sills of igneous rock that intruded the Aguja Formation. The sketch (fig. 85) shows the roughness of both Burro Mesa and the Chisos Pen anticline areas.

FIG. 84. A plug-like intrusion on Burro Mesa (fig. 85, D) that tilted the lava beds to form ring-hogbacks around the intrusion.

FIG. 85.Sketches showing the development of Burro Mesa. A, Sketch of a normal fault showing upthrown and downthrown sides. B, Sketch showing erosion of rocks on upthrown side and deposition on downthrown side. C, Further erosion resulting in development of higher elevations on the downthrown side because of the presence of hard, resistant rocks. D, Artist's sketch of the topography of Burro Mesa and the Chisos Pen anticline.

A new road, completed in 1965, leads the visitor from the Maverick entrance road east of Burro Mesa through the rugged terrain southwest of the Chisos Mountains (fig. 4). For a few miles the road follows along the east side of Burro Mesa; straight ahead (south) are good exposures of the folded Chisos Formation (fig. 86). The road crosses the southern end of the mesa with lookout views into lower Blue Creek Canyon, and it continues to Castolon and Santa Elena Canyon. Interesting geological phenomena can he seen from the road at several places. In the Chisos Formation, between the Chisos Mountains and the south end of Burro Mesa, are three dikes. These appear as rock walls that stand above the surface and are traceable for miles. They are formed of rock similar to that exposed in Ward Mountain and probably came from the Ward Mountain intrusion. In the western face of Goat Mountain (fig. 4), one can see the cross section of a former canyon (fig. 87) that was cut into volcanic rocks (Chisos Formation) and later filled by younger lava (South Rim Formation). Farther on are Mule Ear Peaks (fig. 88) formed by dikes that erosion has sculptured into the shape of a mule's ears. Near Cerro Castellan (fig. 89) are several lava necks (fig. 11B) that cooled and hardened in vents from which lava probably flowed. The largest and most spectacular of these has been mistaken for a petrified tree (fig. 90). The "wood grain" is flow structure in the rock that was formed as the lava moved up the vent; the "knot hole" probably once held a rock that was carried upward by the lava column.

FIG. 86. Crumpled beds in the Chisos Formation that crop out between the Ward Mountain intrusion in the Chisos Mountains (east), and the Burro Mesa fault (west).

FIG. 87. In the western face of Goat Mountain is a former canyon that was cut into the Chisos Formation and later filled with lava and associated extrusive material of the South Rim Formation. A, Outline of the old canyon wall; B, Chisos Formation; C, South Rim Formation.

FIG. 88. Looking northeast at Mule Ear Peaks. These two mule-ear-like peaks were formed by erosion. The two peaks were the high hard parts of two dikes that were left standing as the weaker adjacent rocks were eroded away. As part of the Army Air Corps training program in the 1930's, the pilots maneuvered their planes between the two peaks. Note the sheetwash gravel deposits that extend from the highlands.

FIG. 89. Cerro Castellan, a high stack of volcanic rocks (ash, lava, and tuffaceous rocks). The lower slopes are Chisos Formation and the caprock is South Rim Formation (table 1). Note the erosion surface at top of the yellowish-white beds exposed in the southwest face. This is the unconformity between the Chisos and South Rim Formations. The white foreground is volcanic ash; most of the dark rocks are lava.

FIG. 90. A small volcanic neck near Cerro Castellan. This small neck has been erroneously identified by some as a petrified tree. The "grain" was formed by flow structure in the lava as it ascended; the "knothole" once held a boulder. The white rock is hard volcanic ash and the dark rocks in the background are lava.

Castolon village.—Castolon, now a Ranger station and trading post in the Park, is on a Rio Grande terrace near the foot of Cerro Castellan (fig. 89). Madison and Stillwell (1958, p. 52) reported that the name Castolon was derived from a settler named Castulo who lived at a spring in that vicinity. This site was visited by Lt. Echols on July 26, 1860, while he was exploring the Big Bend with camels. In his diary (entry of July 27, 1860), he made the following notations (Echols, 1860, p. 11):

After marching a few miles down the river this morning, I turned back without going to Los Chiras, satisfied that there is no better place on the river for building a post of any dimension that I had found. The river has a fine valley on each side, about twenty-five miles down; more timber and wood than a post can use. I saw one or two good sites on moderately gravel mesas, easily accessible from the river bottom, elevated just sufficiently for the purpose; plenty grazing for animals; some small canebrakes in the bottom; in a word the location is well adapted for the purpose of building a post . . .

The cavalry post was not built at that time, but small settlements did "grow up" at Castolon and at Coyote at the mouth of Alamo Creek, a few miles farther west (fig. 4, in pocket). The farmer-rancher residents raised goats and a few cattle and farmed the flood-plain areas of the Rio Grande and of Blue and Alamo Creeks. During the border trouble, 1914 to 1918, the U. S. Army established and garrisoned a small military post at Castolon. The long structure in which the trading post is now housed was the cavalry barracks, and the two large houses were the officers' quarters during the early World War I days. The small structure southeast of the store was the bath house and at one time there was a corral and feed shed. Castolon was also a Texas Ranger station and much of the time the Rangers were busy chasing stock thieves, smugglers, and other outlaws back and forth across the Rio Grande. There probably was never much illegal crossing of the Rio Grande in the Big Bend area but small herds of cattle were driven across from time to time, and should a herd of goats happen to water at the crossing point and follow the cow trail to the rocky ledges away from the river, it was almost impossible to trail the thieves. One cattleman observed that in order to ranch along the Rio Grande, one had to steal more from his neighbors than they stole from him.

The cavalry post at Castolon was abandoned in 1921 (Park records) and the land and buildings were sold to H. E. Perry and Wayne Cartledge of Terlingua. Perry owned the Terlingua quicksilver mine and Cartledge was the store manager and purchasing agent. They established the trading post, ranch, post office, and farm at Castolon. The fields were irrigated by pumping water from the Rio Grande; cotton, wheat, and stock feed were the principal crops. The old water-pumping installation is still on the river bank and the long-abandoned cotton gin was an addition to the pumphouse building. The long building with the tin roof was used principally for hay and other feed storage. The long adobe structure at the foot of the mesa, about half a mile west of the trading post, housed the first general store; later the building was used for a schoolhouse. There still remain a dozen or more adobe huts once occupied by the Castolon villagers.

The present store at Castolon has a stock of cold drinks, groceries, camp and picnic supplies, and gasoline. In the future, it will have a visitor center with interpretative services, campground, picnic area, water supply, sewage disposal, and other utilities. At present the store is a popular stopping point for fishermen or campers, and many people drive to Castolon to enjoy the scenery.

Across the Rio Grande opposite Castolon is the Mexican village of Santa Helena. population about 400. It is a cooperative agricultural colony and the principal crop is cotton. These villagers, too, irrigate their fields with water pumped from the Rio Grande. Most of the fields produce two or more bales of cotton per acre, but marketing is a major problem as the cotton is trucked over poor roads around the south end of Sierra Ponce to San Carlos and thence to Chihuahua City. Mexico, In spite of the isolation and transportation difficulties, the village seems to be prospering and growing.

Santa Elena Canyon.—The view of Santa Elena Canyon is one of the most spectacular in the Park (fig. 91). The Rio Grande, like a giant rasp, has cut through the massive Lower Cretaceous limestone layers in the mesa making a gorge, which in some places is not more than 30 feet wide but which has sheer walls approximately 1,500 feet high. Mesa de Anguila and the south end of the mesa (Sierra Ponce in Mexico) are parts of a large block that was elevated along the Terlingua fault (fig. 92). Soft Upper Cretaceous limestones and shales similar to those exposed along Terlingua Creek toward the north once covered the mesa; they probably were overlain by lava, as are some of the Upper Cretaceous beds in the Chisos Mountains. Erosion removed the soft rock from the mesa top leaving only the hard limestone in the lower slopes of what was once a much higher mesa.

FIG. 91. Aerial view of the Santa Elena Canyon. The middle of the Rio Grande at the bottom of the canyon is the International Boundary between Mexico and the United States; Sierra Ponce (Mexico), on the left, Mesa de Anguila (United States), on the right. This canyon is about 1,500 feet deep. There are places in the lower canyon where a man standing in a boat with outstretched arms can touch Mexico with the fingertips of one hand and the United States with the other. The top of the canyon is wider and in most places is 500 or more feet across, (Photograph by Hunter's of Alpine, Texas.)

FIG. 92. Sketch of Santa Elena Canyon, which is a prominent physical barrier between the United States and Mexico. The canyon is cut across an uplifted block of Lower Cretaceous limestone. The formation outcrops are indicated; most of these rock units are also exposed in the Sierra del Carmen, Mariscal Mountain, Sierra San Vicente, and Christmas Mountains area. (Modified from West Texas Geological Society Guidebook, 1955, p. 95.)

The mesa front and the canyon walls show the best cross section of the Lower Cretaceous formations exposed in the Park. The lowest rocks exposed at the mouth of the canyon (fig. 92) are the top of the Glen Rose Formation (table 1). Lower parts of the Glen Rose crop out along the Marufo Vega trail in the Sierra del Carmen. The Glen Rose also forms the lower half of the Sierra del Carmen escarpment south of the Rio Grande (fig. 43); the lowest part of the formation, with conglomerate and sandstone at the base, rests on Paleozoic rock in Persimmon Gap. The low slope beginning about 100 feet above the river at the mouth of Santa Elena Canyon is made by the Telephone Canyon Formation, which is also exposed at many places in the Sierra del Carmen. This formation correlates with the Walnut Clay farther east, part of which, in the Austin area of Central Texas, is a popular building stone. The lowest vertical cliff at Santa Elena Canyon is formed by the Del Carmen Limestone, which correlates with the Edwards Limestone in Central Texas. The Del Carmen Limestone forms the base of most escarpments in the Sierra del Carmen, the base of the canyon wall in Mariscal Canyon, and the lowest cliff in the Christmas Mountains. The middle slope in Santa Elena Canyon is formed by the Sue Peaks Formation, which also forms conspicuous slopes in the Sierra del Carmen, in the middle of Mariscal Canyon, and in some places in the Christmas Mountains. The upper canyon wall is formed by the Santa Elena Limestone. Rocks belonging to this formation also crop out at the top of most ridges in the Sierra del Carmen from near Dog Canyon southward to the Rio Grande and form the top of the high Sierra del Carmen escarpment south of the river. They also form the top layers in most of Sierra San Vicente, Mariscal Mountain, and the Christmas Mountains. On top of the mesa in Mexico (fig. 92), and also about midway between the mouth of Santa Elena Canyon and Lajitas, are some hills and ridges formed by the Del Rio Clay and Buda Limestone. These two rock units form hogback ridges along much of the west side of the Sierra del Carmen and along the Sierra San Vicente-Mariscal Mountain anticlines. Along the upper reaches of Santa Elena Canyon, south of Lajitas, the Boquillas Formation (basal Upper Cretaceous), including a massive black sill, is preserved in some downfaulted blocks along the Rio Grande.

The canyon is about 18 miles long and begins near Lajitas at the northwest corner of the Park (fig. 4). Only the lower 7-mile extent of the canyon is narrow with steep walls as it is shown in the aerial photograph (fig. 91). Near the sharp bend at the top of the photograph is a "rock slide" that has partially dammed the river. This is a pile of rock about 250 feet high and nearly a quarter of a mile long, composed of boulders as large as 20 feet long that have fallen from the vertical canyon wall. The water passes through narrow openings between and around the boulders. There is no passage wide enough for even a small boat, and those who have gone through the canyon have been compelled to portage. It was here that R. T. Hill (1901) spent three days carrying boats and supplies over the "rock slide" when he was making a survey along the International border. Others have lost their boats and cargo and a few have been drowned. There are numerous whirlpools and the person who navigates Santa Elena Canyon is cautioned not to venture too close to the "rock slide" in any type of boat.

Below the "rock slide" is a cave in the soft shaly strata below the hard limestone caprock (fig. 93). Although parts of the cave's walls and ceiling are blackened by smoke, the presence of tobacco and sardine cans suggests that the cave's chief periods of use have been in modern times. There are, however, some interesting stories associated with the cave. Of course, it is one of the many legendary sites of hidden treasure in the Big Bend, but most of the tales are about outlaws who made this cave their headquarters or about the cattle rustlers who held their herds in the cave until the pressure had subsided. The chief weakness of most of these stories is that it would be necessary to swim the herd about 2-1/2 miles up the Rio Grande in order to reach the cave, and it is also too difficult to reach for just an average outlaw "hangout."

FIG. 93. Smugglers Cave in the wall of Santa Elena Canyon (Mexico side). Many stories regarding smuggling and livestock rustling activities are associated with this cave.

Origin of the Rio Grande canyons.—Uplift in the Southwest during mid-Cenozoic time formed many fault-block mountains and intermontane basins (fig. 19). The basins received debris from the mountains; in some of them there was no through drainage and probably they contained lakes (fig. 94). As the closed basins were filled with debris, the lower ranges were probably buried. Basin waters naturally drained through the lowest outlet into a lower basin. When the second basin was filled, the water flowed into a still lower depression. In time, through-flowing drainage was established across the Trans-Pecos Texas region, and most of the basin contained great thicknesses of rock debris. The depth of the original valley-fill deposit in the Park is not known, and whether it buried such features as Mesa de Anguila, Mariscal Mountain, and the Sierra del Carmen is a matter for speculation. There are small patches of valley-fill, including lake deposits, along the present Rio Grande valley and in several of the fault-block valleys of the Big Bend; much larger deposits occur along the river toward El Paso. Once drainage was established on the valley-fill deposits, the river was trapped. When it encountered hard underlying bedrock, there was no alternative but to cut a canyon. A great deal of the Rio Grande history is not completely understood; King (1935) has discussed the problem in much more detail.

FIG. 94. Sketches showing the development of drainage across a fault-block mountain. This probably is the way the Rio Grande cut its channel across West Texas.

The subsequent history of the Rio Grande valley has been one of erosion. Large quantities of the soft materials were eroded and carried away by the river; hard rock knots were left standing high above a dissected plain. Where the stream crossed hard rock ridges, narrow steep-walled canyons were formed. Weathering of the hard rock masses left castle-like peaks (fig. 70), high mesas with precipitous rims (fig. 68), and long linear ridges (fig. 47). These features have been further sculptured along joints in the massive rock to form columns (fig. 95), spires, pinnacles, arches, and buttresses that not only add to the interest and beauty of the country but also stand as monuments to an area that at one time was much higher than it its now.

FIG. 95. Willow Mountain, north of Study Butte and east of the Alpine entrance road. This is the most spectacular display of rock joints in the southern Big Bend area. The mountain mass is an intrusive plug, which cooled below the surface and was uncovered by erosion. On cooling, the rock mass contracted and was uniformly jointed. After erosion uncovered this mass, weathering and erosion removed the soft areas to emphasize the joint pattern.