Replacement chert in the Alibates Dolomite (Permian) of the Texas Panhandle.
The Permian Alibates Dolomite crops out along the Canadian River and its tributaries at places where the Canadian River has cut a wide valley into the High Plains of the Texas Panhandle (Fig. 1). At most places along the valley, the resistant Alibates Dolomite forms caps on steep bluffs that overlook the river, and isolated outcrops of Alibates occur along the Salt Fork of the Red River (Barnes 1968).
More than 12,000 years ago, local natives started to quarry chert from a small Alibates-capped hill within the Canadian River Valley, and tribes continued to quarry the colorful chert for weapons and tools into historic time. These quarry pits, which have been identified by archeologists as some of the oldest in North America, are now protected in the Alibates Flint Quarries National Monument near Fritch, Texas.
The Alibates Dolomite was named and first described by Gould (1907) for exposures of the dolomite along Alibates Creek in Potter County, Texas. The type section is along the creek about 2 km south of Alibates National Monument. According to Hertner (1967), the term "Alibates" was a modification of the name Allen Bates, the son of a local rancher (Bowers 1975). Patton (1923) was the first to discuss the origin of the Alibates chert. He presented three hypotheses, including a replacement theory, for the genesis of the chert but did not find evidence to support any of them. Little else had been done until Eifler (Barnes 1969), and Bowers & Reaser (1974) described the chert in the Alibates Flint Quarries National Monument area and suggested a replacement origin.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
The primary objectives of this study were to determine the petrography of the chert and chert/dolomite contact zones, and interpret the probable origin of the chert. Thirty stratigraphic sections were measured and studied to obtain a representative geographic distribution over the Alibates outcrop area, and samples were taken from each sequence of beds. More than a hundred thin sections of selected samples of dolomite, chert, and red beds were examined with a petrographic microscope. Fifteen selected samples of dolomite and chert from the Alibates were analyzed chemically for major elements and certain trace elements. Ten samples of chert-bearing rocks were selected for oxygen-isotope analysis. In addition, six representative samples of dolomite were analyzed for strontium isotopes.
The 5.0 m thick Alibates Dolomite is herein subdivided into three informal members: lower gray dolomite, middle red beds, and upper gray dolomite (Fig. 2). The lower member ranges in thickness from 1.0 to 3.0 m and is composed of medium- to finely crystalline dolomite that is distinctly laminated at most places. It is resistant to weathering and forms prominent ledges throughout the study area. The upper 10 to 30 cm of the lower dolomite commonly grades into the overlying middle red-bed member, which is relatively uniform in thickness, averaging 1.5 m. The red-bed lithology is chiefly red to brown, calcareous mudstone. Bedding or lamination is not readily apparent at most localities, but the mudstone does grade into shale at a few places. The mudstone weathers easily, and the contact with the upper dolomite member is marked by a sharp break in outcrop profile. The upper dolomite averages 0.6 m thick and is locally absent. This medium- to finely crystalline dolomite member characteristically is laminated and is intensely brecciated and fractured at some places, causing it to be less resistant to weathering than the lower dolomite member.
In the study area, the Alibates is underlain by red beds of the Permian (Guadalupian) Whitehorse Formation, and at most places is unconformably overlain by the Permian (Ochoan?) Quartermaster Formation, a red-bed interval very similar to the Whitehorse. The Quartermaster is difficult to distinguish from the Whitehorse except by stratigraphic position. Lithologically, these rock bodies are nearly identical, and it is possible that, at places where the Alibates has been removed by erosion, the Quartermaster lies directly on the Whitehorse but has not been recognized. There is also little petrographic difference between the Quartermaster and Whitehorse except that, in places, the Quartermaster may contain more clay and mica than the Whitehorse.
Dixon (1967) reviewed the problem of Alibates correlation and stated that the Quartermaster and Alibates could be either Guadalupian or Ochoan. The Texas Bureau of Economic Geology (Barnes 1969) mapped the Quartermaster, Alibates, and Whitehorse simply as a single undivided Permian-age formation. A study by Rascoe & Barrs (1972) placed the Quartermaster, Alibates and Whitehorse in the Guadalupian Series, but more recent studies (McGillis & Presley 1981; Presley 1987) have placed the Alibates as a separate formation in the Ochoan Series.
Six samples of Alibates Dolomite collected during this study were selected for strontium isotope analysis. According to Hetherington (pers. comm.), strontium-isotope ratios ([.sup.87]Sr/[.sup.86]Sr) of the selected samples indicate that both dolomite members were deposited in marine waters late in the Permian Period. As shown in Table 1, there is a marked difference in isotope ratios between the upper and lower dolomite. The lower dolomite is possibly Guadalupian; the upper dolomite probably is Ochoan.
Bowers (1975) proposed a sabkha-like depositional environment for the Alibates because of the prominent algal-mat characteristics displayed by both dolomite members, and because of the local and regional Permian stratigraphy. Although no evaporite minerals were found in the dolomite or chert during this study, the Alibates does contain anhydrite and gypsum outside the study area. McGillis & Presley (1981) described the Alibates from well logs, samples, and cores, and reported that the Alibates thickens southward into the Palo Duro basin and the lithology grades from dolomite into anhydrite and gypsum. They reported laminated anhydrite and gypsum in a core from a well in Randall County about 50 km south of the Alibates National Monument. Additional study of well data by Presley (1987) has detailed extensive evaporite-basin deposition in the Texas Panhandle, and he suggests that much of the Alibates may have been deposited as gypsum.
All Permian formations are unconformably overlain by the Miocene-Pliocene Ogallala Formation. Although the Ogallala Formation is from 46 to 91 m thick throughout the southern Great Plains, no thick sections of Ogallala are present in the study area. The Ogallala is easily distinguished from the other formations by its light brown to buff color and by its conglomeratic phases. Locally, the formation consists of carbonate-cemented sandstone and sandy conglomerate. Caliche deposits are well developed at some localities (Brown 1956) and opalization has occurred sporadically in the formation. The opal is similar in appearance to chert in the Alibates. At most places along the Canadian River Valley where the Ogallala has been eroded and Alibates caps the bluffs, Ogallala sand has been "piped" downward into fractures in the underlying rocks, and Ogallala pebbles and cobbles partly cover the Permian strata. It is at localities where the Alibates is directly overlain by Ogallala that the most extensive chertification has taken place.
The Alibates is best known for its chert because of the ancient flint quarries, and yet the chert accounts for probably no more than two or three percent of the total exposure of Alibates. Chert occurs in most outcrops of the Alibates, and although much more abundant in the upper member, it is not confined to any particular bed or stratigraphic horizon in the dolomite. Three habits of chert observed in outcrop are small spherules (< 5 mm), irregular nodular and pod-like masses, and massive tabular sheets. Small chert spherules are most common in the lower dolomite and can be found in almost every outcrop. Irregular masses also occur in both dolomite members, but these bodies are predominant in the upper dolomite. Massive sheets have been found only at two localities: Cactus Flats and the Alibates Flint Quarries National Monument. There, chertification has taken place most extensively and chert has completely replaced the upper dolomite member (Bowers 1975).
Table 2 shows the chemical composition of Alibates rocks. Samples for analysis were chosen to obtain the best geographical distribution over the outcrop area and the most typical or representative rock types from measured sections. The object of the analyses was to determine if any unusual elements or elemental relationships exist in the dolomite and chert. Two chert, one upper dolomite, and two lower dolomite samples were selected from the National Monument because of the archaeological significance. Eight additional samples of the lower dolomite were selected to give the widest possible geographic coverage. The lower dolomite was used because it is more continuous over a large geographic area, whereas the upper dolomite is absent at some places. Two additional samples of the upper dolomite were chosen at random.
[FIGURE 3 OMITTED]
No unusual results were obtained from the analyses. Calcium/magnesium ratios of the dolomite samples range from 1.341 to 2.246 and indicate an excess of calcium in the samples analyzed. Excess calcium would be expected in dolomite that contains cavity filling spar or that has been calcitized. Some rocks are chertified dolomite (10 to 50% silica) and represent both intermediate stages of chertification and transitional contact zones between chert and dolomite rock. Chemical analyses of massive chert from the Monument area show relatively pure silica with no unusual impurities. The two samples (P-9-F1 and P-9A-F1) average 98.2 % Si[O.sub.2] with small amounts (0.2% or less) of [Al.sub.2][O.sub.3], [Fe.sub.2][O.sub.3], CaO, MgO, and Ti[O.sub.2].
The grain-size classification used in this study is that of Folk (1965) as modified by McBride and Thomson (1970) for chert. The following types of authigenic quartz are recognized:
[FIGURE 4 OMITTED]
Megaquartz: equant to elongated grains larger than 35 microns that commonly occur as cavity and vein fillings.
Microcrystalline quartz: equant grains smaller than 35 microns that commonly form pinpoint-birefringent aggregates.
Chalcedony: radiating fibers, length-fast, extinction parallel with fibers.
Quartzine: fibrous, length-slow, extinction parallel with fibers.
Lutecite: pseudofibrous, length-slow, oblique extinction.
Chert spherules. -- Spherules of chert average 2.0 mm in diameter and commonly weather dark brown, giving the appearance of beads in the light gray dolomite (Fig. 3). Although occurring in both dolomite members, the spherules are more abundant in the lower dolomite where these features are generally concentrated along some bedding surfaces.
Most spherules are composed of chalcedony; some consist of megaquartz or quartzine and lutecite (Fig. 4). Chalcedony and megaquartz spherules are cavity fillings in the dolomite and are similar to birdseye structures described by Folk (1973). Partial to total stages of cavity fillings were observed in thin sections. An empty or partly filled cavity may be less than 5.0 mm away from a completely filled cavity, thus showing extreme localization or selectivity of the chertification. No evidence was observed, however, to suggest that silica precipitation was selective of certain cavities because of pre-existing chemical or mineralogical conditions.
[FIGURE 5 OMITTED]
Some quartzine spherules appear to be cavity fillings, but others may have formed by replacing evaporite minerals. Folk (1972) described characteristics of "length-slow chalcedony" replacement of evaporite minerals, and marked similarities between his descriptions and the Alibates quartzine spherules are evident. A few spherules have shapes that are possible "ghosts" of gypsum or anhydrite crystals. Sulfate mineral replacement is suggested also by fibrous textures in the chert. Although no sulfate microlites are present in the quartzine, these possible relic gypsum textures are similar to those occurring in Permian beds on Bear Island, Svalbard near Spitsbergen. Siedlecka (1972) described and illustrated scattered spherules (1-3 mm) of length-slow chalcedony that displayed the relic sulfate fibrous texture.
The random distribution of length-slow quartzine spherules in the laminated dolomite is similar to anhydrite crystals scattered throughout dolomite beds in a modern sabkha environment, as described by Wood & Wolfe (1969). They observed that poikilotopic anhydrite is common in lagoonal dolomite sediments, but they suggested that anhydrite did not form until after the dolomite was lithified. The described lithologic textures are identical to those of the dolomite and chert of the Alibates, and, although no evaporite minerals were found in Alibates outcrops in this study, much of the Alibates in the subsurface is gypsum and anhydrite (Presley 1987). Apparently only the mineralogy has changed; quartzine and lutecite have replaced the anhydrite.
Concentric growth rings similar to those described by King & Merriam (1969) occur in some chalcedony and quartzine spherules and are commonly accentuated by iron oxide staining (Fig. 5). Multiple generations of silica precipitation are suggested by these rings and by a few cavities filled with both chalcedony and quartzine. Some cavities show successive layers of chalcedony, quartzine, and chalcedony or megaquartz. This alternation of chert types suggests changes in the chemical environment during the stages of cavity filling. Because quartzine can form in a sulfate-rich environment (Folk 1972; 1974), quartzine interlayered with chalcedony in the cavity fillings may imply chemical changes in the interstitial waters during precipitation.
Many spherules have inclusions of finely crystalline or very finely crystalline dolomite. These carbonate microlites in the chert are generally near the quartz/carbonate boundary and appear to be crystals that were trapped in the silica precipitate. Opaque minerals, mostly iron and manganese oxides, and trace quantities of mica and clay are other impurities in the chert spherules.
Irregular chert masses. -- Irregular masses of chert occur sporadically in both dolomite members but are more abundant in the upper part of the lower dolomite. These masses are not localized geographically and occur in most Alibates outcrops. There appears to be no predominant size or shape of the masses. Sizes range from a few millimeters to a few meters, and characteristic shapes include spherical nodules, elongate pods, and lenticular stingers. Nodular and podlike masses are best exposed in the Millican quarries on the Kritser Ranch approximately 32 km north of Amarillo. Locally, the chert masses average 50 cm in longest dimension and are generally elongated normal to bedding surfaces and parallel to near-vertical fractures in the rock. Spectacular exposures of interfingering bodies of dolomite and chert occur near the head of Hackberry Canyon on the Kritser Ranch (Fig. 6) where colorful stringers of chert project from fractures into the dolomite body along bedding surfaces. Lenticular embayments and stringers are most common in the Alibates Flint Quarries National Monument and vicinity.
[FIGURE 6 OMITTED]
The chert masses all are similar in composition and internal structure. In outcrop, the chert/dolomite contact appears to be sharp, and iron oxide staining in the chert readily delineates the contact on fresh surfaces. The dark, resistant chert is most obvious on weathered surfaces. Lamination of the dolomite is continuous into the chert and is more distinctive as bands in the chert because of iron oxide coloration. In addition, unreplaced islands of laminated dolomite as large as 10 cm have been found "floating" in many chert masses. At a few places, dolomite breccia has been preserved by the chert. Banks (1990:129:144:pl. 10E) also reported silicified breccia from Alibates Dolomite near Plum Creek on the Weymouth Ranch, Moore County, Texas.
[FIGURE 7 OMITTED]
Thin sections of irregular chert masses show that most of the chert is fine- to coarse-grained microcrystalline quartz with minor amounts of chalcedony and megaquartz. The chertification process is revealed under the microscope. Microcrystalline quartz commonly occurs as an anastomosing network of stringers that invades the dolomite along porous laminae and fractures. Chert fronts of microcrystalline quartz form along the porous zones and project veinlets outward to infiltrate the dolomite. As chertification proceeds, the chert veinlets surround and isolate individual dolomite crystals thus forming a "chert-permeated" dolomite. The final stage of chertification is complete replacement of the dolomite. The entire replacement by chert or the complete transition from dolomite to chert may be displayed in a distance of less than 1.0 cm.
Massive sheet chert. -- Complete replacement of a dolomite outcrop by chert occurs at two known locations: the Alibates Flint Quarries National Monument and the Devil's Canyon-Cactus Flats vicinity, about 5 km west of the Monument. At these locations, the upper dolomite has been completely replaced by chert, thereby forming large, massive, tabular sheets that range in thickness from 0.2 to 0.6 m and extend laterally for more than 1,000 m along the outcrop. Only the upper dolomite member has been completely replaced in these two areas; the lower dolomite member contains minor amounts of chert in the common forms of spherules and nodular masses as in all other Alibates outcrops examined. Although no "remnant" dolomite has been found in the massive sheets, the common dolomite features are well preserved in the chert. As in irregular masses described above, dolomite laminae are preserved in the chert as alternating bands of red, gray, brown and several other colors. The wide variation in color of the chert probably reflects the presence of minor amounts of aluminum, iron, and manganese (Bowers 1975). Chertified dolomite breccias are usually preserved as white fragments in a red to brown matrix (Fig. 7).
In thin section, massive chert is predominantly fine-grained microcrystalline quartz. Most laminae appear as bands defined by alternating color or impurity content rather than a change of grain size or other grain characteristics. Alternating bands of very fine- and fine-grained quartz occur only in a few thin sections of massive chert rocks. These bands average 0.25 mm thick and are wavy or wispy. Iron oxide staining and trace concentrations of opaque minerals change from band to band. At chert/dolomite contacts, bands in the chert are continuous with laminae in the dolomite. Minor quantities of chalcedony and megaquartz appear as patches within the microcrystalline quartz. These patches may have been the last remaining dolomite "islands" as chertification proceeded and thereby suggest later stages of silica precipitation. The patches also represent chert spherules that were in the dolomite prior to the massive replacement.
Sedimentary breccia occurs within the massive sheets. The breccia formed either by collapse during solution of subjacent evaporite, rip-up of algal-mat layers during storm events, or a combination of these two processes. The chertified breccia appears as fine-grained microcrystalline quartz in thin section. All breccia features have been replaced uniformly by microcrystalline quartz. The rock fragments, laminae within fragments, and breccia cement are differentiated only by coloration and impurity content rather than by grain size or other grain characteristics. These petrographic characteristics are identical to those described above for lamination and indicate complete replacement of the original dolomite breccia. Replacement took place after brecciation. There is no evidence to suggest that the laminated dolomite was first replaced by chert, brecciated, and then cemented by chert or other material that was subsequently replaced by chert. At a few places, however, the massive chert has been fractured and the fractures subsequently lined or filled with more chert. In thin section, these fracture fillings are megaquartz or chalcedony, or both.
Evidence and controlling factors. -- Klein & Walker (1995) observed that silica polymorph replacements of carbonate minerals in marine and meteoric waters is a "common early diagenetic feature of sedimentary rocks." According to Maliva & Siever (1989), outlines of "carbonate precursors" in chert bodies clearly document that most nodular chert is a replacement of carbonate strata or sediments. Evidence from this study for a replacement origin for most of the Alibates chert can be summarized as follows:
1. Chert stringers and chert embayments into the dolomite.
2. Relic laminae in the chert that pass from dolomite into chert.
3. Sedimentary dolomite breccia preserved as chert.
4. Dolomite islands "floating" in a chert matrix.
5. Rhombic crystal "ghosts" and fibrous textures in chert that may represent relic evaporite minerals.
The evidence for massive replacement in the upper dolomite is basically the same as for the irregular masses and spherules. However, the controlling factors for such complete, yet localized, replacement are not obvious. Occurrences of chert spherules and sporadic, tear-shaped masses are easily related to chertification along joints, faults, and porous zones in the dolomite, but field observations made during this study coupled with the evaluation of published geologic maps and air photos of the Monument area show no obvious structural controls for localizing the massive chert.
New regional studies of the Texas and Oklahoma panhandles may lead to recognition of controlling factors. Cooley (1984) and Dolliver (1984) identified lineaments in the region using Landsat imagery. Both areas of massive replacement appear to be along a common lineament, which may reflect some structural control. In a study of the Anadarko basin in Oklahoma, Nielson & Stern (1985) used side-looking radar images to identify regional structural trends that project into the Alibates study area. They presented evidence for reactivation of Pennsylvanian faults during Permian or more recent time, or both. This reactivation of regional structural features may have been a controlling factor for the massive replacement of the upper dolomite member.
It appears from chert/dolomite spatial relationships (proximity to fractures and larger abundance in upper dolomite) described earlier that chertification probably proceeded downward from the top of the Alibates. Because all overlying rocks have been removed by erosion, any stratigraphic or lithologic controlling factors are totally speculative. However, it is possible that the upper dolomite acted as a chemically reactive zone that resulted in the precipitation of silica from descending solutions. Locally, this zone failed to remove all the silica from aqueous solutions, and the silica leaked into the lower dolomite. No evidence to suggest topographic controls for localization of the chert was observed.
Sources of silica. -- The sources of silica for Alibates chert are not known. It is possible that the silica was derived from the host rock during early diagenesis late in the Permian Period, however, this is difficult to ascertain because of the extensive dolomitization of the rock body. The only sources of silica observed within the Alibates are detrital sand, silt, and clay in the dolomite and middle red beds. Silicate detritus in the dolomite is a possible, but extremely limited, source for the chert spherules; no biogenic sources of silica in the Alibates are known. If primary biogenic sources such as spicules and radiolarians existed, the biological elements have been obliterated by dolomitization and other processes. Some chert in the lower dolomite may have been derived from the middle red-bed member, but no evidence was found to suggest leaching of silica from the mudstone. On the contrary, incipient quartz overgrowths were observed on some sand grains from the middle red-bed member.
No local source of silica within the Alibates, past or present, could volumetrically account for the total replacement of the upper dolomite member at the Alibates Flint Quarries National Monument and at Cactus Flats. It is not probable that any known natural process could concentrate a primary Alibates silica source into such a localized body, preserve the sedimentary features of the dolomite, and not affect the overlying and underlying strata. The underlying red-bed and lower dolomite members are no different at these massive-sheet chert localities than the units are at any other Alibates exposure, and only remnants of Ogallala conglomerate overlie the chert sheets. Although it is possible that slight permeability differences within the Alibates could have caused localization of silica derived from below, evidence suggests that the source of silica was stratigraphically above the Alibates. Petrographic examination of underlying red beds revealed no chemical alteration of quartz, feldspar, or mica grains. However, quartz overgrowths do occur in overlying redbeds. In addition, multiple generations of chert as seen in thin section may also suggest multiple sources of silica.
The Alibates is known to be overlain by a present maximum thickness of 76 meters of Triassic rocks in the western part of the study area (Barnes 1969), and other studies (King & Merriam 1969 and others) suggest that Jurassic and Cretaceous rocks also may have once covered the Alibates in the Texas Panhandle. The Triassic rocks are predominantly terrigenous and contain chert and petrified logs (Barnes 1969). Silica may have been derived from bentonite, tuffaceous rock, and silicate minerals. Extensive opalization at some places in the Miocene-Pliocene Ogallala Formation (Norton 1939; Frye & Leonard 1959; Harlow, pers. comm.) indicates more recent possible sources of silica.
Analyses of groundwater from the Ogallala aquifer throughout the Great Plains region show above-normal concentrations of silica in solution (Swanberg & Morgan 1979). Although reliable chemical analyses of groundwater from the panhandles of Texas and Oklahoma are few, Swanberg (pers. comm.) feels that the Ogallala aquifer is a rich source of silica throughout the region. Because heat-flow values do not suggest a large regional thermal anomaly as a cause of the high-silica concentrations and because the water comes from continental deposits, the high values probably reflect solutions saturated with amorphous silica. According to Swanberg (pers. comm.), the source of silica is most likely Pliocene to Holocene ash from western volcanic centers such as the Jemez Mountains in New Mexico and Yellowstone in Wyoming. However, the small amounts of alkali, aluminum, and titanium oxides present in Alibates dolomite and chert (Table 2) precludes a major volcanic source (Maliva & Siever 1988:423). Indigenous plant remains (seeds, roots, stems) or wind-deposited opal phytoliths could also have been a source of silica cement in the Ogallala.
Chertification models. -- Several models have been proposed for the origin of nodular chert. According to these models, chertification can result from the oxidation of organic matter, oxidation of hydrogen sulfide, crystallization pressure applied to silica-carbonate boundaries during crystal growth of quartz and opal-CT, and supersaturated quartz and opal-CT and undersaturated calcite occurring simultaneously in pore waters along part of a coastal mixing zone. These models are discussed in detail by Maliva & Siever (1989). The origin of chert spherules in the upper and lower dolomite members could be explained by the mixing-zone model proposed by Knauth (1979). As noted earlier, most of the chert spherules occur along bedding surfaces. This is in agreement with Knauth's model for the origin of chert in limestone in which the chert formed in a mixing zone of meteoric water and sea water. The resulting zone of chertification is dependent on the porosity and permeability of the sediments. Quartzine and lutecite spherules from the Alibates are probable replacements of sparsely disseminated evaporite minerals, and the spherules displaying growth rings suggest seasonal(?) fluctuations of pH in the mixing zone of the model or local changes in water chemistry because of mineral reactions. However, Dawson (pers. comm.) stated that the Knauth model, which has never been demonstrated to work in nature, is "based on thermodynamic calculations, whereas, chertification (like most other diagenetic processes) is kinetically controlled."
Percolation/concentration model. -- The nature of the irregular chert masses suggests a different origin than that for the chert spherules. Therefore, a percolation/concentration model is proposed to explain the origin of these bodies. The masses have formed by chertification along inclined fracture zones and permeable horizons that extend laterally into the dolomite body. Concentrated in the upper parts of both dolomite members, chertification has proceeded generally from the top downward into the dolomite. At places on the Kritser Ranch, especially near the Millican quarries where the upper dolomite is absent, large blocks of the lower dolomite member have slumped and created funnel-shaped openings between blocks which, in outcrop, are as much as two m wide across the top. These funnel-shaped openings, together with many smaller fractures in the dolomite, are filled with younger grayish orange-pink, calcitic, arenaceous material (Ogallala?). Angular chert, nucleated within the fissure fills, spread outward into the surrounding dolomite. No chert spherules were found "trapped" in irregular masses, but chertification of the masses appears to be much younger than the spherules. The greasy luster and subtranslucent appearance of some chert along fractures in the Alibates suggests that chertification is still taking place today (Fig. 8). McBride (1988:865) noted that the chertification of carbonate clasts from Tertiary conglomerate in west Texas near Big Bend National Park "must have been fairly rapid, because it is unlikely that the proper hydrologic and chemical conditions prevailed for more than a few million years." Hattori et al. (1996:169) also reported that chertification can occur rapidly on surface rocks.
[FIGURE 8 OMITTED]
The hypothesis of silica entering the Alibates from overlying strata concurs with an earlier interpretation proposed for chertification in another part of the Great Plains (Banks 1984:74). Norton (1939) described similar chert occurrences in the Day Creek Dolomite in Kansas and concluded that "The preponderance of evidence favors the theory of replacement by silica from percolating ground water from overlying strata, the commonest source being the sandy conglomerate of the Tertiary Ogallala 'mortar beds.'"
Banks (1990:96) reported that multicolored chert occurs in the Smoky Hill Member of the Niobrara Formation. Although the member crops out in four western states, the chert occurs only at places where Ogallala directly overlies Smoky Hill Chalk. He remarked that chert in the unit resulted from "silica replacement of chalk" in the upper part of the rock body.
Walker (1960; 1962) proposed that chert-carbonate replacement is a reversible process. Calcite replaces silica (primarily detrital quartz), and silica is then transported in solution and reprecipitated nearby. Walker (1962) stated that the probable cause of such replacement reversals is fluctuation of pH in interstitial water. Conversely, experimental work by Klein & Walker (1995) indicated that the uptake of silica onto calcite surfaces in artificial seawater was irreversible and "strongly time and pH dependent." Brown (1956) and Reeves (1970) documented the formation of caliche in the Ogallala Formation in the study area. According to Reeves (1970), one result of the calichification process is oversaturation of solutions with respect to silica because of high pH. Knauth (1979) and several others have discussed the relationship of pH and silica precipitation. More recently, Williams & Crerar (1985) have shown that a carbonate chemical environment enhances silica (opal-CT) precipitation. Furthermore, the presence of magnesium (esp. Mg(OH)[.sub.2]) dramatically affects silica precipitation. Bowers (1975) suggested that it is possible for high-pH, silica-rich, groundwater from the overlying strata to percolate a relatively short distance downward into the Alibates Dolomite. The change in lithology from the terrigenous sandstone, red beds, and conglomerate to dolomite would cause a lowering of pH in the descending solutions and, together with the increase of magnesium in the dolomite, promote precipitation of silica in the dolomite.
McBride (1988:862) reported that Cretaceous carbonate pebbles and cobbles incorporated in Tertiary fluvial rocks from the Big Bend National Park area of west Texas have thin (1-5 mm) chert rinds that were formed by groundwater after deposition. He stated (p. 865) that these carbonate clasts "were replaced by opal-CT on a volume-for-volume basis."
In order to investigate further the silica diagenesis, representative chert samples from the Alibates and the overlying Ogallala were analyzed for oxygen isotopes. Ten hand samples were collected and each sample was broken in half to make duplicate sample sets. Sample 1 is petrified wood from the Ogallala less than 50 cm above the Alibates/Ogallala contact in an outcrop near Clarendon, Donley County. Sample 2 is chertified caliche from the Ogallala and is common throughout the study area. The remaining eight samples are Alibates chert from the Kritser Ranch, Cactus Flats, and the Monument area.
Results of the analyses are given in Table 3. The data show that the isotopic ratios for two nearly pure (oxygen yield >97.5%) Ogallala chert samples are 29.7 and 31.1. Isotope ratios of relatively pure (oxygen yield >97.0%) Alibates chert range from 28.3 to 32.2 with the highest ratio occurring at the Alibates Flint Quarries National Monument. Average ratio for Ogallala chert is 30.4 and average ratio for Alibates chert is 30.0. The high (heavy) [[delta].sup.18]O values in these Alibates and Ogallala samples suggest that the chert crystallized in relatively cold, oxygen-enriched meteoric waters.
Unfortunately, the data are not definitive, The range of ratios is larger than anticipated, and it is not known how much of this difference is attributable to natural variation within a given type of sample. If most of the discrepancy is caused by natural variation and the range of values that may be obtained from a single hand sample of chert is large, it could be concluded that such a large range of values further substantiates the hypothesis of multiple sources and episodes of silica precipitation for the Alibates chert. The value and reliability of the oxygen-isotope analyses as a tool could be increased substantially by developing a much larger data base for Permian chert.
The overlap of values for both Alibates and Ogallala chert probably indicates that the composition and temperature of solutions were similar during chert formation. A comparison of these data with a temperature scale applied to isotope ratios from the late Paleozoic Arkansas Novaculite (Jones & Knauth 1979) suggests that isotopic temperatures ranged from approximately 21.0[degrees] to 37.O[degrees]C during crystallization of the Alibates chert. Most workers today believe that chert takes a relatively short time to form (i.e. McBride 1988; Klein & Walter 1995); the large temperature range indicated by the analyses of this study suggests that it formed during an interval of time characterized by major climatic changes. This interpretation is certainly questionable in light of all other evidence presented here that suggests multiple stages of chertification throughout the history of the Alibates. Some of the Alibates samples have delta-values that are similar to ratios reported both for older (Jones & Knauth 1979) and younger chert (Land 1977). Knauth (pers. comm.) remarked that "The similarity of isotopic composition of these Permian cherts to younger cherts suggests that [.sup.18]O-enriched evaporite waters were probably not involved in the silicification of the evaporite minerals. These were probably fresh waters super-saturated with respect to silica which were dissolving the evaporities."
The physical and petrographic characteristics of the Alibates chert document its origin by replacement; the source of the silica, however, remains speculative. The nature of the chert/dolomite relationships as studied in outcrop, plus the oxygen isotope and Ogallala water chemistry data, support the following conclusions:
1. Most of the silica came from sources outside of the Alibates Dolomite.
2. The sources of silica probably were stratigraphically above the Alibates Dolomite.
3. Most of the chert was localized in the dolomite by percolation along fractures and porous/permeable zones.
Table 1. Strontium isotope ratios for selected samples of Alibates Dolomite. Upper Dolomite Lower Dolomite Unit [.sup.87]Sr/[.sup.86]Sr Unit [.sup.87]Sr/[.sup.86]Sr (top) 0.70731 (top) 0.70707 (middle) 0.70734 (middle) 0.70711 (base) 0.70729 (base) 0.70704 Limit of analytical error = +/- 0.00003. Laboratory techniques used in sample analyses are discussed by Burke (1982) and others. Table 2. Chemical analyses of selected Alibates chert and dolomite samples (Mineral Studies Laboratory, Bureau of Economic Geology, The University of Texas at Austin, 1974). Sample (A) Member [Al.sub.2][O.sub.3] Cao [Fe.sub.2][O.sub.3] P-1-6 Upper 0.150 30.200 0.060 P-2-3 Lower 0.150 30.100 0.100 P-2-F1 Lower 0.150 20.100 0.200 P-4-4 Lower 0.250 31.400 0.080 P-9-4 Lower 0.150 30.000 0.120 P-9-9 Upper 0.250 52.800 0.070 P-9-F1 Upper 0.200 0.130 0.195 P-9A-1* Lower 0.200 29.700 0.130 P-9A-F1* Upper 0.200 0.130 0.140 M-2-3 Lower 0.250 29.500 0.110 M-2-5 Upper 0.250 26.700 0.080 H-5-2 Lower 0.150 35.500 0.090 H-10-2 Lower 0.150 29.900 0.090 C-1-2 Lower 0.200 29.500 0.120 D-1-4 Lower 0.050 24.200 0.050 Sample (A) MgO Mn[O.sub.2] Si[O.sub.2] SrO Ti[O.sub.2] P-1-6 20.500 0.060 1.590 Tr 0.000 P-2-3 21.300 0.080 0.850 Tr 0.000 P-2-F1 13.900 0.055 33.500 Tr 0.000 P-4-4 19.400 0.150 1.850 Tr 0.000 P-9-4 20.900 0.060 1.180 Tr 0.000 P-9-9 1.500 0.120 2.160 Tr 0.000 P-9-F1 0.020 0.003 98.100 0.000 0.015 P-9A-1* 21.000 0.025 1.910 Tr 0.000 P-9A-F1* 0.020 0.007 98.300 0.000 0.020 M-2-3 21.200 0.060 2.280 Tr 0.000 M-2-5 19.900 0.040 12.900 Tr 0.000 H-5-2 15.800 0.060 2.260 Tr 0.000 H-10-2 21.100 0.045 1.180 Tr 0.000 C-1-2 21.100 0.060 1.460 Tr 0.000 D-1-4 17.400 0.075 17.000 Tr 0.000 Sample (A) [H.sub.2]O Ign Ls. (B) Total P-1-6 0.240 46.890 99.690 P-2-3 0.160 47.130 99.870 P-2-F1 0.230 31.440 99.575 P-4-4 0.190 46.070 99.390 P-9-4 0.180 46.780 99.370 P-9-9 0.190 42.080 99.170 P-9-F1 0.230 0.940 99.833 P-9A-1* 0.240 45.980 99.185 P-9A-F1* 0.390 1.030 100.237 M-2-3 0.180 46.250 99.830 M-2-5 0.210 39.770 99.850 H-5-2 0.160 45.420 99.440 H-10-2 0.190 46.870 99.525 C-1-2 0.180 46.250 98.870 D-1-4 0.210 38.960 97.945 (A) Sample designation indicates county (P-Potter; M-Moore; H-Hutchinson; C-Carson; D-Donley), measured section, and unit; for detailed description refer to Bowers (1975). (B) Ign. Ls. = Ignition Loss. * Chert sample from Alibates Flint Quarries National Monument. Table 3. Oxygen-isotope ratios of selected chert samples from the Alibates and Ogallala formations. Yield [delta] [.sup.18]O Yield Sample Number (A) (%) (SMOW (B)) Sample Number (A) (%) Ogallala Alibates 1. (D-1-8) 97.6 29.7 3. (D-1-6A) 84.4 2. (HS-1-1) 98.4 31.1 4. (P-2B-3) 85.2 Average 30.4 5. (P-3-13) 99.7 6. (P-4-7) 98.5 7. (P-9-C)* 97.0 8. (P-9-F1)* 99.2 9. (P-9A-F)* 99.2 10. (P-10-3) 96.2 Average [delta] [.sup.18]O Sample Number (A) (SMOW (B)) Ogallala 1. (D-1-8) 29.0 2. (HS-1-1) 27.1 Average 29.7 30.7 28.3 30.7 32.2 32.7 30.0 (A) Sample designation indicates county, (D-Donley; HS-Hutchinson; P-Potter), measured section and unit; for detailed description refer to Bowers (1975). (B) Standard mean ocean water. * Chert sample from Alibates Flint Quarries National Monument.
We thank William E. Dyer, Superintendent, and the National Park Service for allowing us to measure sections and collect samples at the Alibates Flint Quarries National Monument. Park Ranger Edwin Day provided valuable assistance in the field and shared his knowledge of the Alibates with us. We also acknowledge Tom Kritser for permission to study Alibates outcrops on his property.
Cliff Osburg and W. V. Harlow provided field data which aided this study. Thanks are also due to D. A. Schofield for providing chemical analyses, to E. Hetherington for providing strontium-isotope analyses, to L. Paul Knauth for providing oxygen-isotope analyses, and to Chandler Swanberg for providing regional groundwater data.
Early versions of the manuscript benefited from critical evaluation by D. H. Campbell, M. Kastner, L. P. Knauth, L. S. Land, E. F. McBride, N. D. Smith, and D. H. Zenger. Superb editing by Journal reviewers, W. C. Dawson and J. R. Underwood, Jr., significantly improved the manuscript. Funds for this project were provided in part by a Grant-in-Aid of Research from Sigma Xi, The Scientific Research Society of North America.
Banks, L. D. 1990. From mountain peaks to alligator stomachs: A review of lithic sources in the Trans-Mississippi South, the Southern Plains, and adjacent Southwest. Oklahoma Anthropological Soc., Memoir 4, 179 pp.
Banks, L. D. 1994. Lithic resources and quarries, in Prehistory of Oklahoma, R. E. Bell, ed. New York, Academic Press, Inc., 432 pp.
Barnes, V. E., Project Director. 1968. Plainview sheet: University of Texas at Austin, Austin, Texas, Bureau of Economic Geology, Geologic Atlas of Texas, scale 1:250,000.
Barnes, V. E., Project Director. 1969. Amarillo sheet: University of Texas at Austin, Austin, Texas, Bureau of Economic Geology, Geologic Atlas of Texas, scale 1:250,000.
Bowers, R. L. 1975. Petrography and petrogenesis of the Alibates dolomite and chert (Permian), northern Panhandle of Texas [M.S. thesis]. University of Texas at Arlington, Arlington, Texas, 155 pp.
Bowers, R. L. & D. F. Reaser. 1974. Local chert occurrence in Alibates Dolomite, Alibates National Monument and vicinity, northern Panhandle of Texas. Geological Society of America Abstracts with Programs, 1974 South-Central Meeting, 5:96.
Brown, C. N. 1956. The origin of caliche on the northeastern Llano Estacado, Texas. J. Geology, 64:1-15.
Burke, W. H., R. E. Denison, E. A. Hetherington, R. B. Koepnick, H. F. Nelson & J. B. Otto. 1982. Variation of seawater [.sup.87]Sr/[.sup.86]Sr throughout Phanerozoic time. Geology, 10:516-519.
Cooley, M. E. 1984. Linear features determined from Landsat imagery in the Texas and Oklahoma panhandles. U.S. Geological Survey Open-File Report 84-589, map, scale 1:500,000.
Dixon, G. H. 1967. Northeastern New Mexico and Texas-Oklahoma panhandles, in Paleotectonic investigations of the Permian System in the United States. U.S. Geological Survey Professional Paper 515, pp. 65-80.
Dolliver, P. N. 1984. Cenozoic evolution of the Canadian River Basin. Baylor Geological Studies Bull., 42, 96 pp.
Folk, R. L. 1965. Petrology of sedimentary rocks. Austin, Texas, Hemphill's, 159 pp.
Folk, R. L. 1972. Length-slow chalcedony: a new testament for vanished evaporites. J. Sed. Pet., 41:1045-1058.
Folk, R. L. 1973. Evidence for peritidal deposition of Devonian Caballos Novaculite, Marathon Basin, Texas. Amer. Assoc. Petroleum Geologists Bull., 57:702-725.
Folk, R. L. 1974. Petrology of sedimentary rocks. Austin, Texas, Hemphill's, 182 pp.
Frye, J. C. & A. B. Leonard. 1959. Correlation of the Ogallala Formation (Neogene) in western Texas with type localities in Nebraska. University of Texas at Austin, Austin, Texas, Bureau of Economic Geology Report of Investigations, 39, 46 pp.
Gao, Guoqiu & L. S. Land. 1991. Nodular chert from the Arbuckle Group, Slick Hills, SW Oklahoma: a combined field, petrographic, and isotopic study. Sedimentology, 38:857-870.
Gould, C. N. 1907. Geology and water resources of the western portion of the Panhandle of Texas. U.S. Geological Survey Water Supply Paper 191, pp. 1-70.
Jones, D. L. & L. P. Knauth. 1979. Oxygen isotopic and petrographic evidence relevant to the origin of the Arkansas Novaculite. J. Sed. Pet., 49:581-598.
Hattori, Isamu, Miyuki Umeda, Tomio Nakagawa & Hirofumi Yamamoto. 1996. From chalcedonic chert to quartz chert: diagenesis of chert hosted in a Miocene volcanic-sedimentary succession, central Japan. J. Sed. Research, 66:163-174.
Hertner, H. E., ed. 1967. Three questions--three answers: a story from the life of Dr. Charles Newton Gould: Program for the dedication of an official Texas State Historical Marker honoring Dr. Charles Newton Gould, Dec. 13, 1967, Amarillo, Texas, 28 pp.
King, R. J. & D. F. Merriam. 1969. Origin of the "welded chert," Morrison Formation (Jurassic), Colorado. Geol. Soc. America Bull., 80:1141-1148.
Klein, R. T. & L. M. Walter. 1995. Interactions between dissolved silica and carbonate minerals: An experimental study at 25-50[degrees]C. Chemical Geology, in press.
Knauth, L. P. 1979. A model for the origin of chert in limestone. Geology, 7:274-277.
Knauth, L. P. & S. Epstein. 1976. Hydrogen and oxygen isotope ratios in nodular and bedded cherts. Geochimica et Cosmochimica Acta, 40, pp. 1095-1108.
Land, L. S. 1977. Hydrogen and oxygen isotopic composition of chert from the Edwards Group, Lower Cretaceous, central Texas, in Cretaceous carbonates of Texas and Mexico. University of Texas at Austin, Austin, Texas, Bureau of Economic Geology Report of Investigations Number 89, pp. 202-205.
Maliva, R. G. & Raymond Siever. 1989. Nodular chert formation in carbonate rocks. Journal of Geology, 97:421-433.
McBride, E. F. & A. Thomson. 1970. The Caballos Novaculite, Marathon region, Texas. Geol. Soc. Amer. Special Paper 122, 129 pp.
McBride, E. F. 1988. Silification of carbonate pebbles in a fluvial conglomerate by ground water. J. Sed. Pet., 58:862-867.
McGillis, K. A. & M. W. Presley. 1981. Tansill, Salado, and Alibates formations, upper Permian evaporite/carbonate strata of the Texas panhandle. University of Texas at Austin, Austin, Texas, Bureau of Economic Geology Circular, 81-8, 31 pp.
Nielsen, K. C. & R. J. Stern. 1985. Post-Carboniferous tectonics in the Anadarko basin, Oklahoma: Evidence from side-looking radar imagery. Geology, 13:409-412.
Norton, G. H. 1939. Permian redbeds of Kansas. Amer. Assoc. Petroleum Geologists Bull., 23: 1751-1819.
Patton, L. T. 1923. The geology of Potter County. University of Texas at Austin, Austin, Texas, Bull. 2330, 180 pp.
Presley, M. W. 1987. Evolution of Permian evaporite basin in Texas Panhandle. Amer. Assoc. Petroleum Geologists Bull., 71:167-190.
Rascoe, B. Jr. & D. L. Barrs. 1972. Permian System, in Geologic atlas of the Rocky Mountain region. Rocky Mountain Association Geologists, Denver, Colorado, Hirschfeld Press, pp. 143-165.
Reeves, C. C. Jr. 1970. Origin, classification, and geologic history of caliche on the southern High Plains, Texas and eastern New Mexico. J. Geology, 78:352-362.
Siedlecka, A. 1972. Length-slow chalcedony and relicts of sulphates--evidences of evaporitic environments in the Upper Carboniferous and Permian beds of Bear Island, Svalbard. J. Sed. Pet., 42:812-816.
Swanberg, C. A. & P. Morgan. 1979. The linear relation between temperatures based on the silica content of groundwater and regional heat flow: A new heat flow map of the United States: Geofisica Pura e Applicata, 117:227-241.
Walker, T. R. 1960. Carbonate replacement of detrital crystalline silicate minerals as a source of authigenic silica in sedimentary rocks. Geol. Soc. America Bull., 71:145-152.
Walker. T. R. 1962. Reversible nature of chert-carbonate replacement in sedimentary rocks. Geol. Soc. America Bull., 73:237-242.
Williams, L. A. & D. A. Crerar. 1985. Silica diagenesis, II. General mechanisms. J. Sed. Pet., 55:312-321.
Wood, G. V. & M. J. Wolfe. 1969. Sabkha cycles in the Arab Darb Formation off the Trucial Coast of Arabia. Sedimentology, 12:165-191.
Roger L. Bowers and Donald F. Reaser
Department of Geology, The University of Texas at Arlington
Arlington, Texas 76019
|Printer friendly Cite/link Email Feedback|
|Author:||Bowers, Roger L.; Reaser, Donald F.|
|Publication:||The Texas Journal of Science|
|Date:||Aug 1, 1996|
|Previous Article:||A serological antibody survey for HIV-1 and hepatitis B in northern Chihuahua, Mexico.|
|Next Article:||Two mouthless cypriniform fishes from Louisiana.|