This study documents the diagenetic history of the Lower Cretaceous Edwards Group in the Balcones fault zone area of south-central Texas. The Edwards Group consists of 400 to 600 feet of porous limestone and dolomite that accumulated on the Comanche shelf in shallow-water subtidal, intertidal, and supratidal marine environments. During early diagenesis, carbonate mud neomorphosed to calcitic micrite, aragonite and Mg-calcitic allochems were altered to calcite or were leached, and evaporites formed in tidal-flat sediments. Dolomite is widespread and formed in environments ranging from hypersaline to fresh-water.
Faulting along the Balcones fault zone initiated a circulating, fresh-water aquifer system to the west and north of a fairly distinct "bad-water line," which roughly parallels the Balcones fault zone. To the south of the bad-water line, interstitial fluids remained relatively stagnant and contain over 1000 mg/l dissolved solids. Because of the differences in the chemistry of the interstitial fluids, post-faulting diagenesis in the two zones has been very different. Water in the bad-water zone can be saturated with respect to calcite, dolomite, gypsum, celestite, strontianite, and fluorite, whereas water in the fresh-water zone is saturated only with respect to calcite. Due to the change in water chemistry, rocks In the fresh-water zone have been extensively recrystallized to coarse microspar and pseudospar, extensive dedolomitization has occurred, and late sparry calcite cements have precipitated. In contrast, rocks in the bad-water zone retain fabrics associated with pre-Miocene diagenesis.
The importance of diagenesis in shallow, subsurface environments is illustrated by the fact that the Edwards Group had a stable mineralogy of calcite and dolomite before the circulation of fresh water began. In spite of the "stable" mineralogy, considerable additional diagenesis occurred in the fresh-water zone, and probably continues to occur today.
Figure 1: Location map of the study area showing the locations of cores used in the study, the location of the Edwards Aquifer with approximate flow directions, the zones of fresh-water, bad-water, and the "bad-water lines."
Figure 2: Cretaceous submarine cementation of a molluscan-intraclast grainstone by polygonal calcite cement. The boundaries where the cement meets in the intergranular spaces are thin and suture-like with triple junctures. The cement is fibrous, averaging 0.15 mm. in thickness. The long dimension of photograph is 1.7 mm. Plane-polarized light. Core DX-2, sample from 304 feet.
Figure 3: A cross-plot of d16O (PCB) with d13C (PDB) composition for selected samples from the Edwards Group, Balcones fault zone area showing clustering of data for Cretaceous-related diagenetic products (open symbols) and Miocene to Recent-related diagenetic products (solid symbols)
Figure 4: Typical Cretaceous meteoric phreatic cement. Fine to medium crystalline, isopachous, calcitic cement in Edwards grainstones. The crystals are bladed to equant, evenly surrounding the framework grains. The long dimension of photograph is 4.1 mm. Cross-polarized light. Core DX-2, sample from 312 feet.
Figure 5: Photomicrograph shows two generations of Cretaceous meteoric phreatic cement in a miliolid-intraclastic grainstone. The first cement is a fine, isopachous, equant, calcitic cement. Some compaction took place after the formation of this cement as evidenced by the spalling cement near the center of the photograph. Later, a coarsely crystalline, equant, calcite filled the remaining pore-space. The long dimension of photograph is 4.1 mm. Plane-polarized light. Core TD-3, sample from 313.5 feet.
Figure 6: Limpid dolomite crystals lining a cavity formed by the leaching of a mollusc shell. Note remarkably smooth crystal faces. Scanning electron photograph is 240 microns. San Marcos core, sample from 504 feet.
Figure 7: Neomorphism of micrite to spheres of microspar and pseudospar during the Miocene. Diameter ranges from about 5 to 50 microns. Scanning electron photomicrograph. The long dimension of photograph is 90 microns. Castle Hills core, sample from 281 feet.
Figure 8: This is a similar type of microspar and pseudospar to that shown in Figure 7. In thin-section each grain of microspar and pseudospar can be seen to be a single crystal. Grain-size is fairly uniform (averaging 20 microns) in this sample, and crystals are loafish. Long dimension of photograph is 0.77 mm. Plane-polarized light. Core TD-3, sample from 107 feet.
Figure 9: Miocene to Recent microspar in terra rossa clay. The concentration of clay (darker area in center of photograph) has strongly affected the recrystallization of microspar in adjacent areas. In areas of high concentration of clay, crystal size increases greatly and pseudospar becomes more bladed. The long dimension of photograph is 2.2 mm. Plane-polarized light. Castle Hills core, sample from 367 feet.
Figure 10: A typical dedolomite that now consists of rhombic polycrystalline aggregates of calcite floating in micrite. The aggregates replaced single crystals of dolomite during fresh-water flow related to Miocene faulting. The long dimension of photograph is 2.2 mm. Cross-polarized light. Sabinal core, sample from 403 feet.
Figure 11: A thin-section photomicrograph of a typical, "luster-mottled" or pseudometamorphic dedolomite. This common type of dedolomite consists of poikilotopic crystals of calcite enclosing and replacing rhombs of dolomite. The calcite is coarse-grained with interlocking crystals. The dolomitic rhombs have corroded edges where they have been partly replaced by calcite. The long dimension of photograph is 6.25 mm. Cross-polarized light. Core TD-3, sample from 424.5 feet.
Figure 12: This is a common type of dedolomite in which the centers of dolomite rhombs have been leached and later refilled with single crystals of calcite. Some of the rhombs in this photograph are still hollow. Intercrystalline porosity has been filled with calcite at the same time that many of the hollow, dolomitic rhombs were filled. The long dimension of photograph is 0.77 mm. Plane-polarized light. Selma core, sample from 659 feet.
Figure 13: A moldic grainstone with Miocene early meteoric, phreatic cement outside of micritic rims, and rhombic, calcitic crystals that form later, both inside and outside of the micritic rims. Much moldic and interparticle porosity remains in this sample. The long dimension of photograph is 0.77 mm. Plane-polarized light. Core DX-2, sample from 304 feet.
Figure 14: Fields of occurrence of diagenetic products plotted on a graph of salinity versus Mg/Ca ratio (modified from Folk and Land, 1975). Postulated common diagenetic pathways of calcites and dolomites in Edwards Group carbonates are shown. Also shown are the present-day compositions of waters of the fresh-water zone and bad-water zones (hatchured areas).
This study approaches the problem of documenting and understanding the diagenetic history of the Edwards Group carbonates in the Balcones fault zone area by petrographic study and by the use of various geochemical techniques. A wide variety of depositional environments were represented, from high- to low-energy, with restricted to open circulation. Environments tended to be patchy in lateral extent but, in a vertical direction, environmental trends had more continuity.
Several questions were addressed during this study. What are the petrographic characteristics of various diagenetic components, and what processes were responsible for their formation? Can the timing of diagenetic events be related to various tectonic and stratigraphic events in the history of the rock?
General Geology of the Study Area
During the Late Jurassic and Cretaceous, regional subsidence of the Gulf of Mexico basin resulted in the deposition of an arcuate prism of sediments that thickens from a few hundred feet updip to more than 10,000 feet along the ancient shelf margin 100 to 300 miles downdip (McFarlan, 1977). During most of the early Cretaceous, a ridge complex of biogenic reefs, banks, tidal bars, channel fills, and islands existed on the shelf edge that separated the shallow-water interior of Texas from the deeper-water Gulf of Mexico basin. This band is commonly called the Stuart City reef trend (Winter, 1961; Bebout and Loucks, 1974). and is composed of rudist, coral, and algal debris. To the southeast, sediments sloped into the deeper waters of the ancestral Gulf of Mexico. Shallow-water carbonates of the Edwards Group were deposited on a generally submerged plain called the Comanche shelf (Rose, 1968) in environments ranging from open marine waters to arid, hot, supratidal flats. Typical shallow-water, restricted carbonate facies were deposited on this shelf; they shifted positions gradually with time. Euxinic and evaporitic deposits occurred to the west, and shallow-marine carbonate rocks elsewhere. Over the broad axis of the San Marcos platform, shallow-marine carbonates and dolomitic tidal-flat deposits are present. Between the tidal-flat deposits and open shelfal beds occur shallow-marine calcarenitic banks and bioclastic beds. The San Marcos Platform influenced facies tracts on the Comanche shelf more than did the Stuart City reef. The Edwards is a wide belt of shallow-water sediments deposited behind marginal banks, with the lateral succession of facies reflecting increasing restriction toward the axis of the Central Texas Platform. Sedimentation of the Cretaceous strata of Texas was controlled by a combination of structural and regional stratigraphic features.
The study area follows the trend of the Balcones fault zone, a series of en echelon, high-angle, down-to-the-coast normal faults that forms an arc averaging 10 to 12 miles in width and paralleling the trend of the underlying Ouachita system as it bends around the Texas craton (Figure 1). The area is approximately 175 kilometers long and 50 kilometers wide, and is enclosed by two groundwater divides, one in Kinney County near Brackettville and the other in Hays County near Kyle, which were considered to be the end boundaries of the aquifer (Garza, 1962b). Facies patterns within Comanchean rocks show no effect of Balcones fault movement during their deposition. The first positive evidence of movement on the Balcones fault system is shown by the abundant reworked Upper Cretaceous fossils and limestone fragments in the Lower Miocene Oakville Formation (Weeks, 1945; Raqsdale, 1960). It seems that most Balcones faulting was restricted to the Miocene (Young, 1962). Well cores were available along this fault trend in Hays, Comal, Bexar, Medina, and Uvalde counties. Three cores were available from the bad-water zone, and nine from the fresh-water zone. Previous studies have shown that the lithology, porosity. and textures of carbonates in the two zones are quite different. This study was undertaken to relate the diagenesis of the Edwards to the geochemistry of the original and later pore waters as they evolved in the system. Descriptions of techniques used in this study may be found in Ellis (1985).
General Hydrology of the Edwards Aquifer
Regional Movement of Groundwater in the Edwards Aquifer
The Edwards Aquifer is approximately 175 miles in length. It varies in width from 5 to 40 miles, and in thickness from 400 to 700 feet (Figure 1). It lies along the Balcones Escarpment, formed by the Balcones fault zone, a series of normal faults along which the Gulf Coastal Plain has been dropped relative to the Edwards Plateau. Water entering the Edwards Aquifer moves generally southward across the reservoir outcrop and then eastward through the artesian zone of the aquifer toward areas of natural discharge.
The Edwards fresh-water aquifer forms two distinct reservoirs in south-central Texas, the unconfined aquifer under the Edwards Group on the Edwards Plateau and a confined aquifer (Edwards underground reservoir) associated with the Balcones fault zone. The southeastern boundary of this aquifer is a marked and mappable "bad-water line." The bad-water line marks the transition from potable fresh water of the Edwards Aquifer to the non-potable brackish water of the bad-water zone. The boundary is usually well defined and has been traced from Comal County, Texas, into northern Mexico (Back and Hanshaw, 1977). The boundary trends northeast, similar to the strike of the Balcones faults, although for much of its length it does not appear to be fault-controlled.
No springs discharge south and east of the bad-water line, and yields from wells are considerably lower there. Apparently, the only natural discharge from the region is slow upward seepage. The bad-water line formed as a bypass boundary that meteoric water moving under structural or hvdrologic controls did not transgress. What originally may have been a random hydrologic flow boundary became more ingrained with time (Abbott, 1975).
The position of the bad-water line is largely a function of the availability of water during and shortly after the uplift owing to Balcones faulting, as the present groundwater regime was being established. The active circulation, with flushing and solution on the upgradient side, developed a major fresh-water aquifer.
Water Chemistry of the Edwards Aquifer
Water analyses were published by Pearson and Rettman (1976). They categorized the waters into five groups to show differences in the capacity of the waters to dissolve and precipitate carbonates and other minerals. For simplicity, I will summarize the characteristics of only two of their water groups: the recharge and main fresh-water groups combined, and the saline or "bad-water" group.
The fresh-water zone has total dissolved solids less than 1000 mg/l, and is generally in the 250-350 mg/l range. Sulfates and sulfides are low. The water in this zone is strongly oxidizing. Water is of the calcium bicarbonate type, in which the chloride content is generally less than 25 mg/1. Within the fresh-water zone, the ratios between the different dissolved constituents are everywhere similar. Edwards Aquifer fresh water is saturated with respect only to calcite. The pH of water in the fresh-water zone ranges from 7 to 7.6. Flow in the main freshwater-bearing part of the Edwards is so rapid that the geothermal gradient is suppressed and water temperatures vary only over the narrow range of 22 to 27o C. The chemical homogeneity and lack of a normal temperature-gradient reflect the rapid movement of water through the aquifer in fractures and solution channels. Well production is characterized by high flow rate and little or no drawdown.
Downdip, the chemistry of the water chanqes abruptly. It becomes strongly reducing, has a high sulfate content. and contains considerable quantities of hydrogen sulfide. In the saline zone, water pH is commonly less than 7.0. The saline, or "bad-water" zone, is defined by total dissolved solids in excess of 1000 mg/l, and they frequently exceed 4000 mg/l. Calcium and sulfate ions are the major cation and anion. The water is strongly reducing and can have 50 mg/l or more H2S. Sulfates are as high as 2000 mg/l. Ratios and absolute concentrations of all major ions change markedly across the bad-water line. All major dissolved constituents increase in concentration in the saline zone, and Na+l and Cl-1 become major ions in solution. In the bad-water zone, the water is saturated with respect to calcite and dolomite. Some of the waters from this zone also were saturated with respect to gypsum, celestite, strontianite, and fluorite. From the fresh-water zone there is little downdip flow through the transitional zone and into the saline zone, so water temperatures there increase and reach values of 47oC, close to the expected geothermal gradient.
The chemistry of the bad-water zone is controlled by mixing of water from the fresh-water zone updip and deep water downdip. Deep subsurface water, downdip from the bad-water zone, consists of chloride-calcium brines with a Na+1/Ca+2 ratio less than 1 and (Cl-1-Na+1 )/Mg+2 greater than 1 (Prezbindowski, 1981). Water in the fresh-water zone averages about 1/100th the salinity of sea water, and water in the bad-water zone ranges from 1/35th to 1/4th the salinity of sea water. The Mg/Ca molar ratio increases from less than 0.5 in the fresh-water zone to 1.0 in the saline zone. The transitional waters are not simple mixtures between the extreme types, but result as fresh water flows downgradient into rock of a different mineralogic makeup and, with flow, react with this rock until water-rock equilibrium is established. The reaction driving the changes in chemistry in the transition zone and producing the marked mineralogic and petrographic changes in the Edwards is the dissolution of gypsum. The effects of these reactions are seen in the mineralogy of rocks of the fresh- and bad-water Edwards carbonates. In the bad-water cores, dolomite is the dominant mineral and gypsum is present; little alteration has occurred. The original organic material, sulfides, and evaporites have been preserved. In the fresh-water cores, calcite dominates over dolomite, and gypsum is rare. Sulfides and carbonaceous material have been oxidized.
CRETACEOUS DIAGENESIS IN THE EDWARDS GROUP
Early and late stages of diagenesis have been recognized in the carbonates of the Edwards Group. Many workers have studied diagenesis in the Edwards Group, however, except for studies by Mench (1978), Longman and Mench (1978a,b), Mench and others (1980), Ellis (1985), and Pearson and others (manuscript in prep.), only Abbott (1973, 1974) has studied Edwards diagenesis in the Balcones fault zone. "Early" diagenesis occurred from the time of sediment deposition until Balcones faulting in the Miocene, and "late" diagenesis occurred from the time of Miocene faulting to the present. Early diagenesis will be briefly discussed, omitting details of evaporite formation and silicification. (See Ellis (1985) for details of pre-Miocene diagenesis.) Late diagenesis, including neomorphism, dedolomitization, and the formation of late sparry calcite cements, will be discussed in greater detail.
Diagenesis in the Marine Environment
Carbonate allochems were rounded and broken in the environment of deposition either by physical abrasion and/or biological degradation. Micrite envelopes are a common feature of fossil fragments in skeletal grainstones in the Edwards, but are relatively rare or poorly developed in the Edwards packstones, wackestones, and mudstones.
Very few examples of submarine cementation were found within the study area. These cements are fibrous to bladed and show polygonal sutures of cement as described by Shinn (1975) (Figure 2). Folk (1984, pers. comm.) suggests that marine cements in Cretaceous seas were calcites with 2-10% Mg (not 15-20 % as in modern marine cements). Isotopic analyses of Edwards Group samples (Figure 3) fall with in the expected ranges for marine cements (Hudson, 1977); electron microprobe traverses show a slight Mg +2 memory.
Diagenesis in Meteoric Phreatic and Subsurface Environments
When Edwards Group carbonates were exposed to Cretaceous fresh water in the meteoric zone, diagenesis was extensive. Aragonite needles neomorphosed to calcite and became more equant. Magnesium ions were flushed from the grains of magnesian calcite. Continued flushing led to complete recrystallization and the transformation to an interlocking mosaic of small calcite crystals. Mg-calcite shells have lost their Mg without noticeable alteration of shell microstructure. Aragonitic allochems, unstable in the presence of meteoric water, are most commonly dissolved and later refilled with sparry calcite. Dissolution of aragonite did not take place everywhere in the Edwards Group at the same position in the diagenetic sequence. Most commonly, dissolution of aragonite followed precipitation of a thin layer of equant, medium-crystalline, calcitic cement, and preceded deposition of a later, coarser, equant calcite cement that also partially or completely filled the aragonite mold.
Fine- to medium-crystalline, bladed to equant, isopachous, calcitic cement (Figure 4) averaging 0.1 mm thick is the most common type of cement in Edwards grainstones. This type of cement precipitated in a meteoric phreatic environment, from waters with a Mg/Ca ratio of less than 2:1 and probably less than 1:1 (Folk, 1973, 1974; Folk and Land, 1975). Electron microprobe analyses of these cements showed low amounts of magnesium, which was to be expected in a meteoric Phreatic environment.
The formation of more coarsely crystalline, equant calcite cement, usually from 0.2 to 0.4 mm in size, followed the deposition of isopachous rim-cement (Figure 5). The equant calcite cement is separated from the earlier cement by a distinct break in crystal size, and a slight decrease in Mg+2 content. Waters with a Mg/Ca ratio below 1:1 were probably responsible for this phase of cementation (Folk, 1973, 1974; Folk and Land, 1975). The isotopic compositions of equant, calcitic cements (Figure 3) average d 18O = -3.81 and d 13C = +1.38. Comparison of these values with those found for submarine cement shows the equant calcite to be slightly enriched with respect to both carbon and oxygen, even though they formed in fresher waters. I believe that the submarine cements are showing effects of some isotopic exchanae because their original composition would be unstable when exposed to fresh water. Also, some of the examples of submarine cement show petrographic evidence of minor recrystallization. Whether this recrystallization was due to fresh water present at the time of formation of early calcitic cements or if it is due to flow of fresh water related to Miocene faulting, or both, cannot be determined with certainty. Early calcitic rim-cement probably results from the dissolution of aragonite and the stabilization of magnesium calcite in the meteoric phreatic environment. Petrographic evidence demonstrates most, but not all, calcitic rim cements to be precompaction. Later cements may derive a proportion of their material from carbonate released during pressure-solution and carried in from nearby rocks. The presence of relatively low compaction and high primary porosity in some Edwards rocks is a result of precompaction cementation.
Petrography and Petrology of Dolomites
Originally both the fresh-water zone and the bad-water zone contained significant quantities of dolomite. However, dedolomitization has affected much of the dolomite in the fresh-water zone. Edwards Group dolomite varies from finely crystalline, dirty, anhedral to subhedral crystals finer than 0.01 mm, to medium-crystalline, clear, subhedral to euhedral crystals as larae as 0.1 mm. Several types of dolomite have been recognized in rocks of the bad-water zone (Longman and Mench, 1978a, b). The onset of dolomitization occurred after stabilization of Mg-calcite.
Much of the porosity in dolomites of the Edwards Group formed as a result of incomplete dolomitization of muddy sediments. Following partial dolomitization, any undolomitized micrite was then dissolved by fresh water. The intercrystalline porosity thus formed in the dolomites has been measured to be as high as 43%. and values are frequently in the 25-35% range (Ellis, 1985). In addition, permeabilities in these porous dolomites can be quite high.
Dolomite interpreted to have formed penecontemporaneously in supratidal sediments generally appears as small (less than 4 microns) rhombs with abundant inclusions; originally it probably was protodolomite. Dolomite crystals with "dirty" cores and cleaner rims are common. During diagenesis, the interstitial water changed from hypersaline to relatively fresh, but dolomite rhombs apparently continued to grow in spite of the change, forming clear euhedral overgrowths (Folk and Siedlecka, 1974). Later, in lower-salinity waters, the chemical stability of the original core of protodolomite was apparently less than that of the surrounding overgrowths, so that as a result, hollow dolomite rhombs are common. Kerr (1976) also reported zoned, hollow dolomite in his study of the Edwards Group in the Belton quarry.
Limpid dolomite with perfect rhombic shape and mirror-smooth crystal faces (described by Folk and Siedlecka, 1974) in the Edwards Aquifer was a relatively late-stage dolomite. It tends to form as overgrowths on other forms of dolomite and also to line cavities and shell molds formed by leaching (Figure 6). It did not often form in micrite but instead preferred open spaces. This suggests that leaching of micrite preceded the formation of limpid dolomite.
Stable Isotope Geochemistry of Edwards Group Dolomites
A crossplot of d18O with d13C values of Edwards Group dolomites (Figure 3) shows a linear trend. Values of d 18C vary within a small range, from +1.62 to +4.10, with an average of +2.57. Values of d 18O vary over a fairly wide range, with -5.41 being the lightest value and +0.32 the heaviest (average value is -1.98) (Mench and others, 1980; Ellis, 1985). The most 18O enriched dolomites come from the more gypsiferous members of the Edwards and tend to be typical hypersaline dolomites. These dolomites are fine-grained, cloudy crystals, commonly occurring in supratidal sequences and probably originally protodolomite. The most 18O depleted samples of dolomite are primarily limpid dolomites, associated with waters of low salinity.
Model for dolomitization in the Edwards
Petrographic, petrologic, and geochemical evidence from this study supports the hypothesis that Edwards dolomites were formed by at least two episodes of dolomitization (Ellis, 1985). The first episode was very early, produced from hypersaline brines, with location being controlled by depositional facies. The second episode was later, but with the location still mostly controlled by the location of nuclei from the earlier hypersaline dolomitization. For many dolomites, the cores formed rapidly from hypersaline brines, so that magnesium was not incorporated into the lattice in stoichiometric proportions. The rims precipitated from less concentrated solutions, and less rapidly, so that magnesium was incorporated in more stoichiometric proportions.
According to the fresh-water-mixing hypothesis, chemically more perfect dolomite forms from dilute solutions. Fresh- and saline-mixed waters also were capable of dolomitizing by themselves, without having hypersaline cores on which to build. This can be seen from the presence of pure, limpid dolomite without hypersaline cores, although mixing-zone dolomite more commonly overgrows earlier dolomite.
The present Mg/Ca value of 1:1 in the bad-water zone and salinities ranging from 1/10th to 1/100th that of sea water make the formation of dolomite within the bad-water zone a possibility today. Waters of the bad-water zone are saturated with respect to dolomite whereas these of the fresh-water zone are undersaturated. Dolomite in the Edwards has formed in environments ranging from hypersaline, to schizohaline, to fresh-water.
Evaporite minerals formed both early and late in the history of the Edwards Group. Some are related to early sabkha conditions, others as a replacement of earlier minerals, and some may be forming today within the bad-water zone. Details of evaporate formation may be found in Ellis (1985).
Petrographic, petrologic, and geochemical data on the various types of silica, combined with similar data on the dolomites and other diagenetic features of the Edwards Group, suggest a local origin for the silica and penecontemporaneous initiation of most silicification. Many molluscan fragments have undergone some form of silicification. Evaporates are frequently silicified and, locally, chert nodules are common. Chert replacement of carbonate rocks must occur under conditions where diagenetic waters are simultaneously supersaturated with respect to silica and undersaturated with respect to carbonate, since much chert replacement involves Si precipitation at the same time as carbonate solution. Details of silicification of Edwards carbonates will be found in Ellis (1985).
MIOCENE AND LATER DIAGENESIS IN THE EDWARDS GROUP
Miocene faulting resulted in the establishment of a circulating fresh-water aquifer system to the north and west of a fairly distinct bad-water line that roughly parallels the Balcones fault zone. The rocks in the fresh-water zone have been extensively recrystallized as a result of the changes in the chemistry of the water to which the rocks were exposed.
Original waters in the rocks were marine (salinity was that of sea water, Mg/Ca ratio around 3:1). Early in the diagenetic history, waters varied from hypersaline (salinity from 1 to 10 times that of sea water, Mg/Ca ratio from 10:1 to 30:1) to fairly fresh (salinity around 1/10 that of sea water and Mg/Ca ratio from 1:3 to 10:1). Cretaceous fresh-water resulted in the formation of meteoric cements and limpid dolomite. Only in the present-day fresh-water zone is extensive microspar and pseudospar found. Land and Folk (1975) proposed the realm of formation of microspar in fresh-water aquifers and lakes to be at a salinity of about 1/100th that of sea water, and a Mg/Ca ratio from 1:3 to 1:10. This salinity was much less than that for the precipitation of early, fresh-water cements. Values measured from the aquifer today give a salinity of about 1/100th that of sea water, and a Mg/Ca ratio of about 1:2 (Pearson and Rettman, 1976).
Neomorphism of Micrite to Microspar and Pseudospar
Coalescive Neomorphic Spherules
Coalescive neomorphism of micrite to microspar and pseudospar is common throughout all cores of the fresh-water zone, but does not exist in cores of the bad-water zone. Crystals in neomorphosed porous micrite in the fresh-water zone typically are spherical to subspherical and polyhedral. Diameters of the spherical grains vary from 5 to 50 microns. Small crystal faces occur on parts of some spheres, but most surfaces are covered by small irregular bumps and pits (Figure 7). In thin-section, each grain of microspar and pseudospar can be seen to consist of a single calcite crystal (Figure 8). Some grains of microspar have distinct crystal faces. They occur in a wide variety of shapes from simple rhombs to complex scalenohedra. Such crystals are most common in relatively nonporous micrites of the fresh-water zone, whereas more porous micrites have subspherical grains.
Microspar spherules are restricted to the fresh-water zone. None were observed in the bad-water zone. This suggests that the formation of the spherules is controlled by fresh-water circulation and that it postdates the establishment of the fresh-water aquifer in the Miocene. The effects of outcrop weathering can be ruled out as a cause of this neomorphism, since cores were used in this study. The altered zones do not appear to be related to Cretaceous unconformities or surfaces of exposure. Instead, neomorphism must have occurred in the shallow subsurface at depths of 25 to 200 meters.
Microspar and Pseudospar Associated with Red Clay
Clay minerals also play an important role in the formation of microspar and pseudospar, apparently because of their tendency to attract magnesium ions (Folk, 1974; Longman, 1977). By acting as a Mg-ion "sump", the clay "liberates" the calcite micrite from a "cage" of Mg ions and allows it to recrystallize to coarser size. Although clay is sparse in the Edwards Aquifer, patches of Pleistocene terra rossa occur in several cores from the fresh-water zone. These patches are associated with much microspar and pseudospar in grains up to 0.5 mm. long. Microspar grains are tightly packed in these terre rosse and crystal boundaries are difficult to distinguish with the SEM. However, thin sections show the microspar-pseudospar to be loafish or bladed. In areas of higher concentration of clay, crystal size greatly increases and the pseudospar becomes more bladed (Figure 9). Clayey impurities have been segregated during neomorphism. In hand specimen, the orange patches appear to be largely clay, but in thin section they appear to consist primarily of microspar and pseudospar grains that have displaced small amounts of clay. During the growth of microspar and pseudospar, there is a marginal expulsion of undigested impurities such as clay.
Microspar related to red clay is most abundant in the more porous zones of the fresh-water zone cores located nearest the axis of the San Marcos Arch. The area of the San Marcos arch began to rise in early Washita time causing erosion on the surface of the Edwards Group in this area (Rose, 1972). Porosity enhancement in the Edwards was greatest in the exposed area of the Edwards during late Early Cretaceous time. This area was later subjected to greatest fault displacement during the Miocene (Woodruff and Abbott, 1979). The faults and joints superimposed additional avenues for the development of porosity and permeability into an area that already had considerable secondary porosity. Development of porosity continued during the later Tertiary and Quaternary. Terra rossa could have entered the openings thus created after Miocene faulting, but evidence of terra rossa prior to the Kansan is lacking (K. Young, pers. comm., see article in this volume.)
Microspar associated with the clay minerals in the terra rossa can be more accurately dated as to the time of its formation than the rounded microspar spherules formed by the action of interstitial fluids in the porous micrite. Caves along the Balcones fault zone contain many vertebrate fossils of Mid-Pleistocene and younger age, but no vertebrate remains older than Pleistocene have yet been found (Lundelius and Slaughter, 1971). This suggests that the terra rossa in the caves must be younger than Pliocene and that the microspar in the terra rossa probably formed in the Pleistocene or Holocene.
Geochemistry of Microspar Formation
Folk and Land (1975) proposed the realm of formation of microspar in fresh-water aquifers and lakes to be at a salinity of about 1/100th that of sea water and a Mg/Ca ratio from 1:3 to 1:10, with a salinity much lower than for the precipitation of early, fresh-water cements. Salinities measured from the aquifer today are about 1/100th that of sea water. Mg/Ca ratios are about 1:2 (Pearson and Rettman, 1976).
Rapid flushing throughout the fresh-water zone has resulted in the formation of large amounts of microspar. The flushing could have removed much Mg+2 from the micritic grains, resulting in their release from their micritic "cage," following which porphyroid neomorphism transformed micrite into microspar and pseudospar (Folk, 1974). Some Mg+2 removal occurred early in the meteoric zone (Ellis, 1985). Further Mg+2 could have been removed by clay-mineral adsorption as water percolated through the rocks as a result of Miocene faulting. Also, in late diagenesis, flushing of Mg+2 by fresh water after establishment of the aquifer resulted in the formation of much microspar. Much of this water could have been high in Ca+2 as a result of solution of evaporates within the formation. This would have further reduced the Mg/Ca ratio. Rocks of the bad-water zone, which were not exposed to late, fresh-water flushing. show no evidence of microspar or pseudospar.
Microspar and pseudospar have among the lowest values of both 13C and 180in the Edwards (Figure 3), and their values vary inversely, whereas those of the dolomites and early calcites vary together.
It has not been possible to collect samples of the formation water associated directly with the formation of microspar and pseudospar. However, it is possible to compare the isotopic composition of these calcites with that of Edwards water on a regional scale (F.J. Pearson, Jr., pers. comm., 1978) to show that they are in isotopic equilibrium, and so support the contention that the calcites are forming from present-day formation-water.
Dedolomites are found in all cores of the fresh-water zone. Luster-mottled dedolomites are the most common variety found. Polycrystalline, calcitic rhombs are the rarest type of dedolomite observed. Dedolomites with calcitic centers within dolomitic rims are common in the fresh-water zone, and are found in the bad-water zone only in Randolph core, which is very near the bad-water line. Dolomitic rims without calcitic centers are very common in the bad-water zone, which shows the relative instability of the schizohaline cores, or shows that solution of cores may have occurred in the Cretaceous, before Miocene faulting. Calcite may have grown within these rhombs within the fresh-water, phreatic zone in the shallow subsurface early in the history of diagenesis.
Moore and others (1968), Abbott (1974), Mueller (1975a), Jacka (1977, 1984), and Prezbindowski (1981) have studied dedolomites within the Edwards Group.
Types of Dedolomites in the Edwards Group
Polycrystalline rhombic calcite is a kind of dedolomite in which individual dolomitic rhombs are replaced by a mosaic of finer grained, calcitic crystals (Figure 10). These polycrystalline rhombs of calcite have sharply defined boundaries. This type of dedolomitization occurs by centrifugal replacement. Centers of crystals may have been subjected to calcitization first because they were originally a less stable form of dolomite (Ellis, 1985). Textures indicate that no large void space existed during this replacement. Dedolomites of this type generally occur in porous micrite in the Edwards Aquifer.
Luster-mottled dedolomites are mosaics of calcitic crystals coarser than the original dolomite crystals. With time, the edges of the dolomite rhombs are attacked and replaced by the calcite. Eventually, the whole rhomb may be replaced. The resulting dedolomite is luster-mottled with relict, 10- to 100-micron dolomitic crystals poikilotopically enclosed in large crystals of calcite . Centers of crystals may be subjected to calcitization first, because they were originally a less stable form of dolomite (Ellis, 1985). In some dedolomites of this type, allochems are well preserved, but in others allochem outlines are completely obliterated. The luster-mottled or pseudometamorphic type of dedolomite is the most common type in the Edwards.
Figure 12 shows leached dolomite rhombs whose centers have been filled with calcite (although some rhornbs in this photoqraph are still hollow). The calcite centers in these rhombs are single crystals in near crystallographic continuity with the dolomite. Intercrystalline porosity has been filled with calcite at the same time. Calcite may be replacing dolomite with more calcian or inclusion-rich centers, interpreted as having formed in a schizohaline environment. In these dedolomites, a distinct void space formed before precipitation of calcite.
Geochemistry of Dedolomites
Dedolomitization is a result of reactions between the Edwards dolomitic matrix and groundwater in the aquifer system formed as a result of Miocene faulting. Isotopic compositions of the dedolomites are set by these reactions. Concentrations of 18O and 13C are both low, and their values vary roughly inversely, whereas those of the dolomites and early calcites vary together (Figure 3). Isotopic values for the dedolomites follow a similar trend to those of the microspars and pseudospars. As with the microspars and pseudospars, it can be shown that dedolomites are in isotopic equilibrium with Edwards water on a regional scale, which supports the contention that the dedolomites are forming from present-day formation-water (Pearson and others, manuscript in prep.).
Edwards waters causing dedolomitization were enriched, but undersaturated, in dissolved CaSO4 and have a low Mg+2/Ca+2 ratio (Pearson and Rettman, 1976). These pore fluids dissolved gypsum and dolomite. The fluids that dissolved the sulfate in the Edwards Group were of meteoric origin, and the dissolution of CaSO4 made them chemically quite active with respect to the associated carbonate minerals. The solution then became supersaturated with respect to CaCO3 and precipitated calcite. The addition of Ca+2 to the water lowered the Mg/Ca ratio sufficiently to dedolomitize (a ratio of much less than 1:1 according to Folk and Land, 1975) and may have raised the CaCO3 solubility product (as suggested by Mercado and Billings, 1975) sufficiently to help produce radical neomorphism in associated CaCO3 sediments and rocks (Pearson and others, manuscript in prep.).
Causes of Dedolomitization in the Edwards Group and Timing of Formation
Dedolomite in the Edwards Aquifer is not directly related to recent weathering on the outcrop since cores were used in this study. Neither is it related to buried unconformities. Rocks in both the fresh-water and bad-water zones were deposited in similar environments and underwent the same diagenetic history until the establishment of the fresh-water zone in the Miocene. If dedolomite are related to subaerial exposure prior to the Miocene, it would have formed in both zones since erosion surfaces are found in both zones. Furthermore, dedolomite occurs throughout the entire 100- to 200-meter thickness of the fresh-water zone in the cores studied. It is not concentrated in a few zones as would be expected if it formed during subaerial exposure.
The several types of dedolomite in the fresh-water zone are all believed to have formed by the same general mechanism. Extensive, fresh-water flushing moved relict, Mg-rich brines from the fresh-water zone while bringing in abundant Ca+2 ions from the limestones in the recharge area. This flushing combined with the dissolution of gypsum in the fresh-water zone to raise the Ca/Mg ratio of the pore waters. The high Ca/Ma ratio and relatively rapid, fresh-water flushing caused calcite to replace dolomite. The abundance of dedolomite in the freshwater zone and its paucity in the bad-water zone indicate that dedolomitization occurred after the fresh-water aquifer system was established in the Miocene.
Late, Sparry Calcite Cement
Description and Distribution
Many grainstones in the bad-water zone are very porous. Some have been partially dolomitized. Others are cemented with equant or bladed, sparry calcite; these apparently were cemented early in diagenesis in a fresh-water, phreatic environment before the bad-water zone became established, or may have been cemented in the subsurface by Mg-poor, subsurficial waters (Folk, 1974). In the fresh-water zone, equant calcitic cement is ubiquitous. This biased distribution of sparry calcite is clearly the result of differential cementation by fresh water late in diagenesis after the establishment of the fresh-water zone (Longman and Mench, 1978a).
This cement is most commonly found filling tectonic fractures or lining fracture surfaces and obviously formed as a late cement. In general, fractures in the cores examined from the bad-water zone are closed or tight, whereas fractures in the cores from the fresh-water zone are wider, less cemented, and commonly enlarged by solution. The fresh-water zone contains many vugs that have been partially filled or filled with coarse crystalline calcite. Some of these pore-lining crystals show little evidence of dissolution, whereas others have been corroded by fresh-water flow. Analyses of Pearson and Rettman (1976) show that, in general, waters of the fresh-water zone are saturated with respect to calcite, but a number of wells have waters slightly undersaturated, which could have caused the dissolution of calcite seen in the cores. Several limestone beds in the fresh-water zone are cemented by rhombic calcite (Figure 13). These rhombs resemble dolomite in thin-section but are stained by Alizarin Red S. When examined under the SEM, calcitic rhombs differ from dolomitic rhombs in appearing slightly bloated; i.e., they have more rounded edges and less regular crystal faces. Similar rhombic calcite has been described by Perkins (1968) from rocks of a wide variety of ages and locations. Bebout and others (1977) reported rhombic calcite as a late-stage cement in rocks from the Stuart City trend. Folk (1974) suggests that this type of calcite forms in fluids with a low salinity and a low Mg/Ca ratio.
Isotope Geochemistry of Late, Sparry Calcite Cements
Isotopic evidence obtained for late, sparry calcitic cements shows that these cements formed under distinctly different chemical conditions than did earlier cements. Results of isotopic analysis of late calcitic cements and travertines can be found in Figure 3. Late calcitic cements and travertines, which can be separated regionally and petrographically from earlier calcitic cements, formed after faulting and result from reactions between earlier cements and dolomites and fresh water introduced after faulting. These reactions can be written generally as:
Fresh water + Dolomite + Gypsum Brackish Water + Calcite
(Mench and others, 1980)
The calcites produced by this reaction are distinct isotopically and texturally from earlier formed calcites. Late calcites have the lowest values of both d13C and d18O, and these values vary inversely, whereas those of earlier calcitic cements and dolomites vary together. These are as light as 7.5o/oo d13C and -10.0o/oo d16O. The carbonate fraction of these is considerably lighter than for the earlier calcites because the former calcites grew in equilibrium with meteoric water containing some organically derived carbon. There is good agreement between isotopic values measured for these late calcites and values predicted based on present-day water chemistry and various temperatures of formation. Data support the hypothesis that the late calcites are forming from present-day Edwards Aquifer water (Pearson and others, manuscript in prep.).
DIAGENETIC MODEL FOR THE EDWARDS
The diagenetic history of Edwards carbonates reflects a number of different diagenetic environments. Early stages of diagenesis began in the marine environment, the site of original formation and deposition of allochems. Later, most sediments were subaerially exposed. Most Cretaceous diagenesis occurred in the meteoric and mixing-zone environments, where a head of meteoric water developed on islands or along prograding shorelines. At the end of Edwards Group deposition, a broad, regionally exposed surface developed, and a meteoric system developed on a more regional scale. Following the formation of the post-Edwards surface, as sea level rose, the Edwards Group was buried by younger sediments, until the Central Texas platform was finally submerged in late Washita time. Major faulting occurred along the Balcones fault zone in the Miocene, which raised the Edwards Limestone in the north and west relative to sea level. Because of the faulting, conditions favorable for producing a circulating fresh-water aquifer developed on the upthrown side of the fault zone. It was the formation of this circulating groundwater system that created the last major diagenetic system that produced marked changes in the petrology of the Edwards Group carbonates.
The initial stages of diagenesis took place in the marine environment with solutions of normal marine chemistry. Major diagenetic features include the formation of micritic envelopes and a minor amount of cementation in the marine environment. Early crusts of bladed cements, probably composed of Mg-calcite, are rare in the Edwards Group within the study area.
The second stage of diagenesis was a local meteoric stage that involved solutions of low Mg/Ca ratios, together with a mixing-zone stage in which Mg/Ca ratios are high but salinities are variable. Evidence indicates that most meteoric diagenesis occurred in the phreatic zone. This early diagenetic stage is the result of the development of fresh-water lenses below islands, or fresh-water tables beneath tidal flats. Diagenetic events in this environment included neomorphism of micrite and grains, including stabilization of Mg-calcite, limited inversion of aragonitic allochems, and extensive dissolution of aragonitic allochems. Minor amounts of syntaxial cements formed on echinodermal fragments, but the dominant form of cement consists of isopachous, bladed to equant, calcitic rims. Hypersaline dolomite formed in supratidal sabkhas. The schizohaline mixing-zone environment, between meteoric and marine environments, produced overgrowths on hypersaline nuclei of dolomite. The formation of gypsum occurred in sabkhas, penecontemporaneous with early formation of dolomite. Silicification occurred after dissolution of aragonite, since no originally aragonitic allochems appear to have been silicified, and silica is a common pore-filling in molds of aragonitic fossils. Silica nodules formed after the formation of some dolomite.
The third stage of diagenesis was the late regional phreatic stage during which a regional meteoric groundwater system developed. This stage probably occurred after exposure at the end of Edwards Group deposition. This was probably a shallow system of circulation. A deeper system of circulation could not develop until after development of the fault scarp. Coarse, equant calcite filled or partially filled pores remaining after cementation by earlier, isopachous, rim cements. Mixing-zone dolomitization as well as further silicification may still have been occurring during this stage. During this stage, Mg/Ca ratios were variable but the water was normally oxidizing.
A fourth stage of diagenesis occurred in rocks of the fresh-water zone, which developed after major Miocene faulting along the Balcones fault zone raised the Edwards Group on the north and west relative to sea level. As a result of the faulting a circulating fresh-water aquifer system developed on the upthrown side of the fault zone. Extensive neomorphism of micrite to microspar and pseudospar has occurred in the fresh-water zone following development of a circulating fresh-water system. Coalescive neomorphism of micrite is common throughout all cores of the fresh-water zone, whereas it does not exist in cores of the bad-water zone. Another type of neomorphism of micrite to microspar and pseudospar in the fresh-water zone is related to red clays introduced into the formation. Red clays are most common in the fresh-water zone in cores located nearest the axis of the San Marcos Arch. The area of the San Marcos Arch began to rise in early Washita time, causing erosion on the surface of the Edwards Group in this area (Rose, 1972). The San Marcos Arch area was later subjected to greatest fault displacement during the Miocene (Woodruff and Abbott, 1979). The joints and faults superimposed additional avenues for the development of porosity and permeability into an area that already had considerable secondary porosity. Porosity continued to develop during the Tertiary and Quaternary, and terra rossa entered the openings thus created in the middle and late Pleistocene. Paleontologic evidence suggests that the clays are younger than Pliocene, which suggests that the microspar associated with these clays formed during the Pleistocene or Holocene.
Another major diagenetic change subsequent to formation of the fresh-water zone is extensive dedolomitization of rocks in the aquifer. Edwards waters causing dedolomitization had a high Ca/Mg ratio. These pore fluids dissolved gypsum and dolomite. The solution, supersaturated with respect to CaCO3, precipitated calcite. Thus, the dissolution of gypsum and dolomite and the precipitation of calcite occurred simultaneously.
Late, sparry calcite cements are another diagenetic product in the fresh-water zone, and were a result of differential cementation by fresh water. These late calcitic cements include fracture-filling calcite, travertine, and vug-cementing calcite. In addition, rhombic calcite has been found to be a late-stage cement in several cores. Late calcitic cements are distinct isotopically and texturally from earlier calcites. The data support the hypothesis that the late calcites are still forming from present-day Edwards Aquifer water. Diagenetic pathways of calcites and dolomites are shown in Figure 14, along with approximate present-day compositions of freshwater zone and bad-water zone waters.
This study of the relationships between diagenesis and interstitial fluids in the Edwards Aquifer has demonstrated the profound effects that changes in pore-fluid chemistry have on diagenesis in shallow-subsurface environments. The carbonate rocks of the Edwards Group had already reached a so-called "stable" mineralogy of calcite and dolomite when the circulating fresh-water aquifer developed. In spite of this, considerable additional diagenesis occurred in the fresh-water zone.
Abbott, P.L., 1973, The Edwards Limestone in the Balcones Fault Zone, South-Central Texas: Unpub. Ph.D. dissert., Univ. Texas, Austin, 122 p.
Abbott, P.L., 1974, Calcitization of Edwards Group dolomites in the Balcones Fault Zone Aquifer, south-central Texas: Geology, v. 2, p. 359-362.
Abbott, P.L., 1975, On the hydrology of the Edwards Limestone, South-Central Texas: Jour. Hydrology, v. 24, p. 251-269.
Back, W., and B.B. Hanshaw, 1977, Structural and stratigraphic occurrence of "bad water" in Coahuila, Mexico: Geol. Soc. America Abstracts with Programs, v. 9, no. 7, P. 885.
Bebout, D.G., and R.G. Loucks, 1974, Stuart City Trend, Lower Cretaceous, South Texas: a carbonate shelf-margin model for hydrocarbon exploration: Univ. Texas at Austin Bur. Econ. Geol. Rept. Inv. 78, 80 p.
Bebout, D.G., R.A. Schatzinger, and R.G. Loucks, 1977, Porosity distribution in the Stuart City Trend, Lower Cretaceous, South Texas, in D.G. Bebout and R.G. Loucks (eds.), Cretaceous carbonates of Texas and Mexico: Univ. Texas Bur. Econ. Geol. Rept. Invest. 89, p. 234-259.
Ellis, Patricia Mench, 1985, Diagenesis of the Lower Cretaceous Edwards Group in the Balcones Fault Zone Area, South-Central Texas: Unpub. Ph.D. dissert., Univ. Texas at Austin, 327 P.
Folk, R.L., 1973, Carbonate petrography in the post-Sorbian age; in Ginsburg, R.N., ed., Evolving concepts in Sedimentology: Johns Hopkins Univ., Stud. Geol. no. 21, p. 118-158.
Folk, R.L., 1974, The natural history of crystalline calcium carbonate: effect of magnesium content and salinity: Jour. Sed. Petrology, v. 44, P. 40-53.
Folk, R.L., and A. Siedlecka, 1974, The "schizohaline" environment: its sedimentary and diagenetic fabrics as exemplified by Late Paleozoic rocks of Bear Island, Svalbard: Sediment. Geol., v. 11, p. 1-15.
Folk, R.L., and L.S. Land, 1975, Mg/Ca ratio and salinity: two controls over crystallization of dolomite: Am. Assoc. Petroleum Geologists Bull., v. 59, p. 60-68.
Folkman, Y., 1969, Diagenetic dedolomitization in the Albian-Cenomanian Yager Dolomite on Mount Carmel (Northern Israel): Jour. Sed. Petrology, v. 39, p. 380-385.
Garza, S., 1962b, Recharge, discharge, and changes in ground-water storage in the Edwards and associated limestones, San Antonio area, Texas: a progress report on studies, 1955-1959: Texas Board Water Eng. Bull. 6201, 42 p.
Goldberg, M., 1967, Supratidal dolomitization and dedolomitization in Jurassic rocks of Hamakhtesh-Haqatan, Israel: Jour. Sed. Petrology, v. 37, p. 760-773.
Hudson, J.D., 1977, Stable isotopes and limestone lithification: Jour. Geol. Soc. London, v. 133, p. 637-660.
Jacka, A.D., 1977, Deposition and diagenesis of the Fort Terrett Formation (Edwards Group) in the vicinity of Junction, Texas, in D.G. Bebout and R.G. Loucks (eds.). Cretaceous Carbonates of Texas and Mexico: Univ. Texas at Austin Bur. Econ. Geol. Rept. Inv. 89, p. 182-200.
Jacka, A.D., 1984, Emplacement of nonevaporitic sulfates in McKnight Formation, Maverick Basin, and associated complex diagenesis (abstr.): Am. Assoc. Petroleum Geoloqists Bull., v. 68, p. 491.
Kerr, R.S., 1976, Development and diagenesis of a Lower Cretaceous bank complex, Edwards Limestone, North Central Texas: Unpub. M.A. Thesis, Univ. Texas at Austin, 203 p.
Kinsman, D.J., and R.J. Patterson, 1973, Dolomitization process in sabkha environment (abstr.): Am. Assoc. Petroleum Geologists Bull., v. 57, P. 788-789.
Longman, M.W., 1977, Factors controlling the formation of microspar in the Bromide Formation: J. Sed. Petrology, v. 47, P. 347-350.
Longman, M.W., and P.A. Mench, 1978a, Diagenesis of Cretaceous limestones in the Edwards aquifer system of south-central Texas: a scanning electron microscope study: Sediment. Geol., v. 21, p. 241-276.
Longman, M.W., and P.A. Mench, 1978b, Dolomitization in the Edwards Limestone, south-central Texas: 10th Int. Congress on Sedimentology, Jerusalem, Israel, Abstracts, v. I., p. 392-393.
Lucia, F.J., 1961, Dedolomitization in the Tansill (Permian) Formation: Geol. Soc. America Bull., v. 72, p. 1107-1109.
Lundelius, E.L., and B.H. Slaughter, 1971, Fossil vertebrate remains in Texas caves, in E.L. Lundelius and B.H. Slaughter (eds.), Natural History of Texas Caves: Gulf Natural History, Dallas, Texas, P. 15-27.
McFarlan, E., Jr., 1977. Lower Cretaceous sedimentary facies and sea level changes, U.S. Gulf Coast; in D.G. Bebout and R.G. Loucks (eds.). Lower Cretaceous Carbonates of Texas and Mexico, Univ. Texas at Austin Bur. Econ. Geol. Rept. Inv. 89, p. 5-11.
Mench, P.A., 1978, Diagenesis related to Miocene faulting in the Edwards Aquifer system, San Antonio area, Texas: Geol. Soc. America, Abstracts with Programs, v. 10, p. 22
Mench, P.A., F.J. Pearson, Jr., and R.G. Deike, 1980, Stable isotope evidence for modern freshwater diagenesis of the Cretaceous Edwards Limestone, San Antonio area, Texas (abstr.): Amer. Assoc. Petroleum Geologists Bull., v. 64, p. 749.
Mercado, A., and G.K. Billings, 1975, The kinetics of mineral dissolution in carbonate aquifers as a tool for hydrological investigations, I. Concentration-time relationships: Journal of Hydrology, v. 24, p. 303-331.
Moore, C.H., Jr., M.R. Facundus, and D.N. Blout, 1968, Dedolmitization associated with evaporite solution, Geological Society of America Abstracts, 1968 Annual Meeting, p. 205-206.
Mueller, H.W. III, 1975a, Centrifugal progradation of carbonate bank: a model for deposition and early diagenesis, Ft. Terrett Formation, Edwards Group, Lower Cretaceous, central Texas: Unpub. Ph.D. dissert., Univ. Texas at Austin, 300 p.
Pearson, Jr., F.J., and P.L. Rettman, 1976, Geochemical and isotopic analyses of waters associated with the Edwards Limestone Aquifer, central Texas: U.S. Geological Survey, Open-File Report, 35 p.
Pearson, Jr., F.J., P.M. Ellis, and R.G. Deike, in preparation, Stable isotope evidence for modern freshwater diagenesis of the Cretaceous Edwards Limestone Aquifer, San Antonio Area Texas.
Prezbindowski, D.R., 1981, Carbonate rock-water diagenesis, Lower Cretaceous, Stuart City Trend, South Texas: Unpub. Ph.D. dissert., University of Texas at Austin, 237 p.
Ragsdale, J.A., 1960, Petrology of Miocene Oakville Formation, Texas Coastal Plain: Unpub. Thesis, Univ. Texas at Austin, 195 p.
Rose, P.R., 1972, Edwards Group, surface and subsurface, Central Texas: Univ. Texas at Austin Bureau of Economic Geology Report of Investigation 74, 198 p.
Shearman, D.J., J. Khouri, and S. Taha, 1961, On the replacement of calcite by dolomite in some Mesozoic limestone from the French Jura: Proceedings of the Geological Association London, v. 72, p. 1-12
Shinn, E.A., 1975, Polygonal cement sutures from the Holocene: a clue to recognition of submarine diagenesis (abstr.): American Association of Petroleum Geologists - Society of Economic Paleontologists and Mineralogists Annual Meeting Abstracts, v. 2, p. 68
Weeks, A.W., 1945, Balcones, Luling and Mexia fault zones in Texas: American Association of Petroleum Geologists Bulletin, v.29, p. 1733-1737
Winter, J.A., 1961, Fredericksburg and Washita strata (subsurface Lower Cretaceous), southwest Texas: Univ. Texas at Austin, Ph.D. dissert. (unpub.), 135 p.
Woodruff, Jr., C.M., and P.L. Abbott, 1979, Drainage basin evolution and aquifer development in a karstic limestone terrain, south-central Texas, U.S.A.: Earth Surface Processes, v. 4, p. 319-334.
Young, K., 1962, Mesozoic history, Llano region, in Geology of the Gulf Coast and Central Texas: Houston Geological Society Guidebook, Geological Society of America Annual Meeting, p. 98-106.
Abbott, Patrick L. and Woodruff, C. M., Jr., eds., 1986, The Balcones Escarpment, Central Texas: Geological Society of America, p. 101 -114