A 'V 03 STATE OF ILLINOIS WILLIAM G. STRATTON. Gmernor DEPARTMENT OF REGISTRATION AND EDUCATION VERA M. SINKS, Director DIVISION OF THE STATE GEOLOGICAL SURVEY M. M. LEIGHTON, Chiei URBANA REPORT OF INVESTIGATIONS— NO. 166 SILICEOUS MATERIALS OF EXTREME SOUTHERN ILLINOIS Silica, Novaculite, Canister, Calico Rock, and Chert Gravels J. E. LAMAR PRINTED BY AUTHORITY OF THE STATE OF ILLINOIS URBANA, ILLINOIS 1953 DEC 1 5 1998 LIBRARY, STATE OF ILLINOIS WILLIAM G. STRATTON. Governor DEPARTMENT OF REGISTRATION AND EDUCATION VERA M. SINKS, Director DIVISION OF THE STATE GEOLOGICAL SURVEY M. M. LEIGHTON. Chief URBANA REPORT OF INVESTIGATIONS— NO. 166 SILICEOUS MATERIALS OF EXTREME SOUTHERN ILLINOIS Silica, Novaculite, Canister, Calico Rock, and Chert Gravels J. E. LAMAR PRINTED BY AUTHORITY OF THE STATE OF ILLINOIS URBANA, ILLINOIS 1953 MANUSCRIPT COMPLETED OCTOBER 1952 nttosBv DEC 1 5 1998 ORGANIZATION STATE OF ILLINOIS HON. WILLIAM G. STRATTON. Governor DEPARTMENT OF REGISTRATION AND EDUCATION HON. VERA M. BINKS. Director BOARD OF NATURAL RESOURCES AND CONSERVATION HON. VERA M. BINKS. Chairman W. H. NEWHOUSE. Ph.D.. Geology ROGER ADAMS. Ph.D., D.Sc. Chemistry R. H. ANDERSON, B.S.. Engineering A. E. EMERSON. Ph.D.. Biology LEWIS H. TIFFANY. Ph.D.. Pd.D.. Forestry GEORGE D. STODDARD. Ph.D.. Litt.D.. LL.D.. L.H.D., President of the University of Illinois DELYTE W. MORRIS, Ph.D.. President of Southern Illinois University GEOLOGICAL SURVEY DIVISION M. M. LEIGHTON, Ph.D.. Chief (68169— 2M— 4-53) CONTENTS Page Introduction 5 Acknowledgments . 5 Stratigraphy 5 Microcrystalline silica (tripoli) 5 Occurrence 8 Character 8 Character of deposits 8 Character of commercial silica 10 Nature of grains 10 Grain shape 10 Particle-size distribution 10 Bulk specific gravity, weight per cubic foot, and porosity 12 True specific gravity 12 Oil absorption 12 Specific surface 12 Abrasiveness 12 Fusion point 14 Chemical composition 14 Mining and processing 14 Uses 15 Resources 15 Novaculite and novaculite gravel 16 Definitions 16 Novaculite 16 Novaculite gravel 17 Character 17 Occurrence and resources 19 Mining and processing 19 Uses 20 Origin of microcrystalline silica, novaculite, and novaculite gravel 21 Introduction 21 Nature of formations before leaching 21 Origin of chert and other silica materials in Devonian formations 22 General process of disintegration of the Devonian formations 24 Interpretation of the origin of the existing lithologic phases of the siliceous Devonian formations . 25 Solid chert in deposits of considerable thickness 25 Much-fractured chert 25 Friable, pulverulent microcrystalline silica with or without chert nodules 25 Firm, microcrystalline silica with or without chert nodules 26 Chert fragments, including large chunks, mixed with clay and silica 26 Clay in beds or along fault or joint planes 27 Ganister 27 Occurrence 27 Character of deposits 28 Character of ganister 28 Mining and processing 29 Uses 29 Resources 29 Origin 30 Calico rock and siliceous shale 31 Origin 33 Page Chert gravel 33 "Lafayette" gravel and sand 33 Elco gravel 36 River gravel and sand 37 Creek gravel 38 Osage formation in Hardin and Saline counties 39 ILLUSTRATIONS Figure Page 1. Geological map 6 2. Map showing areas where silica and ganister have been mined 8 3. Clay seam in silica mine 9 4. Exposure of Grassy Knob silica 10 5. Photomicrograph of silica 11 6. Abrasiveness testing machine 13 7. Interior of silica mine 15 8. Pit of Novaculite Gravel Company 19 9. Types of fragments found in novaculite gravel 20 10. Bailey limestone outcrop 21 11. Grassy Knob chert showing partial replacement by silica 22 12. Grassy Knob chert with stylolite 23 13. Photograph of ganister 27 14. Interior of ganister mine 30 15. Calico rock 32 16. Elco gravel at Elco 37 17. Interbedded Osage chert and silica 39 TABLES Table Page 1. Succession of formations 7 2. Particle size distribution of commercial silicas; cumulative percent finer 11 3. Chemical analyses of commercial silica and of clay from a silica deposit 14 4. Chemical analyses of novaculite and clay from novaculite gravel 17 5. Sieve analyses of novaculite gravel 18 6. Character of fragments in novaculite gravel, Sample LM 104 18 7. Chemical analyses of limestone, chert and silica from the Bailey formation 26 8. Chemical analyses of ganister and of materials from the Hartline "formation" 29 9. Physical properties of calico rock 32 10. Sieve tests of "Lafayette" gravel 34 11. Sieve tests on "Lafayette" gravel and sand 3S SILICEOUS MATERIALS OF EXTREME SOUTHERN ILLINOIS Silica, Novaculite, Canister, Calico Rock, and Chert Gravels J. E. LAMAR INTRODUCTION A VARIETY OF EARTH materials composed largely of silica (SiOo), most of which do not occur in other parts of the state, are found in extreme southern Illinois (Union, Johnson, Pope, Hardin, Alexander, Massac, and Pulaski counties). They include micro- crystalline silica or "tripoli," ganister, novac- ulite, novaculite gravel, Elco gravel, chert gravel in stream beds, "Lafayette" gravel, and calico rock. This report brings to- gether data from the literature and new information concerning these mineral re- sources. ACKNOWLEDGMENTS The author is indebted to many present and past Survey colleagues for assistance in the investigations covered by this report, in- cluding Carl E. Dutton, H. B. Willman, Robert M. Grogan, Margaret Copeland Godwin, Robert R. Reynolds, R. S. Shrode, and Mary Hill Weber. Chemical analyses were made under the direction of O. W. Rees, Chemist and Head of the Analytical Division of the Geochemical Section. Thanks are also due A. M. Davis and B. W. Claflin, Illinois Minerals Co., Cairo, and P. H. Fox and G. G. Jennings, Tamms In- dustries, Inc., Chicago and Tamms, for their cooperation in matters relating to the silica industry of southern Illinois. STRATIGRAPHY The geological formations which provide the resources discussed herein and their ap- proximate thicknesses are listed in table 1, together with other geologic units, a knowl- edge of which is important to an understand- ing of possible modes of formation of the various siliceous materials. MICROCRYSTALLINE SILICA (TRIPOLI) Trade and popular usage have applied various descriptive terms to southern Illi- nois silica, including "amorphous," "decom- posed," "cryptocrystalline," and "soft." The term "silex" has also been applied errone- ously to soft silica.^ Because these names are not wholly scientifically exact in their meanings or implications, the term ?nicro- crystalline silica is used here as more accu- rately describing the southern Illinois ma- terial. The term soft silica is believed to have originated in the buffing and polishing com- pound trade to distinguish the silica pro- duced in southern Illinois and Tennessee from roughly similar materials produced in other geographic areas." It is also some- times used in the ceramic trade to distinguish microcrystalline silicas from the "hard" silica, silica flour, or ground quartz pro- duced by pulverizing quartz sand.^ The term tripoli is applied commercially to the microcrystalline silica produced in the Missouri-Oklahoma district. This term is also used technically "as a general name ^ Santmyers, R. M., Quartz and silica, Part I, General sum- mary: U. S. Bur. Mines Inf. Circ. 6472, p. 4, August 1931. - Eardley-VVilmot, V. L., Industrial minerals and rocks, Chapter I, Abrasives: New York, Am. Inst. Min. Met. Eng., p. 54, 1937. ^ Heinz, C. E., Industrial minerals and rocks. Chapter XLVII, Tripoli: New York, Am. Inst. Min. Met. Eng., p. 911, 1937. [5] SILICEOUS MATERIALS JACKSON CO. R_2_ r3 [union CO. UNION CO. PULASKI CO. Fig. 1. MICROCRYSTALLINE SILICA (TRIPOLI) Table 1. — Succession of Formations Quaternary system (Pleistocene and Recent) Clayey silt, clay; sand and gravel in places Tertiary system Pliocene series "Lafayette" formation — 20-50 feet Eocene series Paleocene series Cretaceous system Pennsylvanian system Mississippian system Upper Mississippian series Lower Mississippian series Meramec group Osage group Hartline "formation" — 30-50 feet Kinderhook group Springville formation — 60 feet Devonian system Clear Creek formation — 300 feet a, b Grassy Knob "formation" — 185 feet Bailey formation— 200-300 feet Silurian system Ordovician system •■* Indicates the presence of other geologic formations which are not discussed and whose names are omitted because they are not important in connection with this paper. For detailed data regarding the full stratigraphic succession see Weller, J. M., and Ekblaw, G. E., Preliminary geologic map of parts of the Alto Pass, Jonesboro, and Thebes quadrangles, Explanation and stratigraphic summary: Illinois Geol. Survey Rept. Inv. 70, 1940. •> Locally the Clear Creek formation rests directly on the Grassy Knob "formation." Fig. \. — [Opposite page) Generalized map of distribution of siliceous formations and asso- ciated formations in parts of Union and Alexander counties. Bottomlands of Mississippi and Cache rivers in west and south parts of map are unpatterned, as are Mississippian and Pennsylvanian sandstones, limestones, and shales in east part of map. (Compiled from Weller, Stuart, Krey, F. F., and Weller, J. M., Preliminary geologic map of the Missis- sippian formations in the Dongola, Vienna, and Brownfield quadrangles: Illinois Geol. Sur- vey Rept. Inv. 60, 1939; and Weller, J. M., and Ekblaw, G. E., Preliminary geologic map of parts of the Alto Pass, Jonesboro, and Thebes quadrangles in Union, Alexander, and Jackson counties: Illinois Geol. Survey Rept. Inv. 70, 1940.) 1 — Hartline "formation" to Salem formation. Largely limestone except Hartline chert "formation" which occurs south of Jonesboro in the western part of the; mapped area and in places includes ganister and chert gravel. 2 — Springville and Mountain Glen formations. Both formations are largely shale, but south of Jonesboro the Springville formation is highly siliceous and locally contains calico rock. 3 — Dutch Creek to Alto formations. Limestone, shale, and sandstone. A — Clear Creek formation. Chert, chert gravel, cherty limestone, and locally silica. 5 — Backbone formation. Limestone and cherty limestone. 6 — Grassy Knob "formation." Chert, chert gravel, cherty limestone, and locally silica. 7 — Bailey formation. Siliceous, cherty limestone, locally weathered to chert gravel and silica. 8 — Silurian and Ordovician formations. Principally limestone, sandstone, and shale. 8 SILICEOUS MATERIALS for a number of comparatively soft, porous, friable, microcrystalline silicas of sedimenta- ry origin.""^ The microcrystalline silica of southern Illinois qualifies as tripoli in this technical sense. Occurrence Most of the silica produced in southern Illinois has come from the Clear Creek formation, whose distribution is shown in figure 1. Silica has also been produced from the Grassy Knob "formation," most recently at Olive Branch. All of the mines currently producing tripoli from the Clear Creek for- mation are west of Elco and Mill Creek. Areas in which silica mines are known or reported are shown in figure 2. Character The natural unground southern Illinois silica, excluding the unweathered chert found in some deposits, consists chiefly of minute particles of crystalline quartz, or of aggregations of such particles, more or less firmly held together. The degree of aggregation is reflected in the friability or lack of friability of the various layers or por- tions of the deposits. The quartz particles, according to x-ray examination,"' range from 0.1 to 50 microns or more in diameter but are mostly between 1 and 5 microns. They are commonly angular, and some par- ticles show a distinct, although usually in- complete, crystal form. Some samples in- clude larger solid masses of crystalline quartz and milky quartz, either singly or in combination. Data on specific gravity and other prop- erties of the natural unground silica are given below for both the firm and friable varieties as subsequently described. Crude Crude friable firm silica silica True specific gravity. Apparent specific gravity Weight per cu. ft., lbs. . Porosity, percent. 2.65 1.31 82 50 2.65 1.44 90 46 / • ^-^ \ Wolf Lake ^5^\ Anno T. 12 "^^ "^-\ • S. m Jonesboro "^ \ -o \ r\ / 1 1 Mill T. 13 S. y^ - ' \ -. Creek ^-"^ UNION CO. • /^ALEXANDER CO 1 \"-: " ^ f / \...--' / \ ] / N, :' / 1 ' \ Elco o^^ T. 14 .^ / / W s. / ^"^ ^K / • Tarhms •Thebes Diswood y \ T. -A. / 15 r? S. •»\ • Olive Branch 1 \ \ « Heinz, C. E., op. c'lt. ' Bradley, W. P., personal communication, 1948. Fig. 2. — Map showing approximate limits of the principal areas in which silica (dashed lines) and ganister (dotted lines) are known or re- ported to have been mined. The moisture content of Illinois silica as mined varies ; moisture determinations on three samples ranged between 15 and 25 percent by weight. Character of Deposits The silica of southern Illinois occurs in deposits an inch or less to about 30 feet thick. The tops or bottoms of some of the commercial deposits are iron-stained. The thickness of the deposits is probably fairly constant if near-surface slumping and the color of the silica are disregarded. However, as only the very white silica is now marketed, the roofs and floors of the mines are con- trolled not only by the thickness of the silica stratum but also by the amount of discolora- tion of its upper and basal portions by iron oxide. The individual beds comprising the silica deposits are usually roughly horizontal or MICROCRYSTALLINE SILICA (TRIPOLI) gently inclined, but show local waviness and distortion. In places they are cut by joints or small faults. The latter cause displace- ment of the silica beds. The joints or faults usually are bordered by a zone of discolored yellow or brown silica and may also have along them a sheet of plastic clay, composed of the clay minerals kaolinite and illite or of kaolinite only, a Fig. 3. — A vertical clay seam about 10 feet high in a silica mine pillar near a mine entry. The silica beds show a displacement of a few feet along the plane of the clay seam. Some of the clay extends a short distance to the right into the silica. An iron stained zone is visible at the top left of the photograph. fraction of an inch to an inch or more thick (fig. 3 and table 3). Some of the clay sheets contain pebbles, and in places they can be traced upward without a break into red surface gravels and clay.*^ Usually the outcrop of a silica bed is partly stained cream, yellow, or brown, and only by digging back into the deposit or by test drilling can its true color be ascer- tained. The color commonly becomes less intense, going from the outcrops to the more deeply buried parts. There appear to be several zones within the Clear Creek formation at which com- merical silica deposits occur. Judging from the distribution of mines, important zones may occur near the bottom of the formation and somewhat above the middle. The evi- dence suggests, however, that the conditions requisite for the development of commercial deposits are fulfilled only locally. Thus a zone which provides commercial silica at one place may be very cherty at others; likewise, the amount of discoloration varies from place to place. Suitable source mate- rial probably occurred in relatively defi- nite stratigraphic zones but was of patchy distribution. Therefore, if the direction of dip of a silica bed can be determined in a known deposit, and if faulting is not too complex, careful study may make it possible to predict the approximate location and posi- tion of the same stratum at other places in the immediate vicinity. There is, however, no known way of predicting whether the bed will be sufficiently white and chert- free to be of commercial grade. The silica deposits are composed of three major types of materials: friable, easily powdered, finely divided silica; porous firm silica; and chert. In addition, some deposits contain beds of clay up to two inches or more thick. The chert occurs as nodules in layers of firm silica and also as layers of angular, mostly vertically fractured chert, only partly or slightly altered to silica (fig. 4). The relative proportions of the three major types of materials vary in different deposits and from place to place in the same deposit, but usually the soft material pre- * Parmelee, C. W., and Schroyer, C. R., manuscript, Illi- nois Geol. Survey. 10 SILICEOUS MATERIALS dominates In most deposits that are operated commercially and chert is absent or occurs only in relatively minor amounts. Character OF Commercial Silica Nature of grains. — Commercially ground southern Illinois silica consists principally of two types of grains : discrete particles of quartz, and "cluster grains," that is, grains which are clusters or aggregates of minute particles of quartz (fig. 5). Some of the discrete particles are needle-like, others are more nearly equidimensional. The cluster grains, as seen under the microscope, resem- ble small fragments of sugary frosting and consist of particles similar to the discrete particles. The clusters are believed to be fragments of the natural silica which were not disintegrated during grinding. In addition, some samples contain a few relatively large crystalline quartz grains, grains of milky appearing silica, and grains Fig. 4. — An exposure of Grassy Knob "formation" silica. A band of undisintegrated chert is visible at the level of the man's hat brim, one of nodules at waist level. combining either of the foregoing with cluster type silica. A roughly quantitative study of the silica grains between 20 and 74 microns in diam- eter, separated from two commercial sam- ples, revealed that one had 93 percent cluster grains, 6 percent discrete quartz grains and milky appearing silica grains, and 1 percent of grains combining cluster type material with larger quartz particles. Com- parable figures for the second sample were 79, 14, and 7 percent. Grain shape. — Rough visual determina- tions of grain shape indicate that most of the grains in the ground silica, including both discrete particles and cluster grains, are roughly equidimensional or moderately elongate in shape. The 2 by 74 micron ma- terial in the two samples mentioned above showed for the first sample about 44 percent equidimensional grains, 48 percent moder- ately elongate, 7 percent elongate, and 1 percent of miscellaneous shapes. Corre- sponding figures for the second sample are 42, 50, 7, and 1 percent. In general the elongate grains are most common in the silica finer than 10 microns due to the abundance of needle-like quartz particles. Moderately elongate grains are somewhat more common in the 5 by 10 micron size fraction than in the coarser or finer frac- tions, whereas the equidimensional particles are more common in the silica coarser than 10 microns than in the finer sizes. Particle-size distribution. — The particle- size distribution of seven samples of com- mercial silica, selected to show the range in the commercial grades, is given in table 2. The size measurements were made by the hydrometer method and by wet sieving. Bulk specific gravity, weight per cubic foot, and porosity. — Eight samples of com- mercial silica, measured both in the loose state and after being packed down by jar- ring them, gave the following ranges in spe- cific gravity, weight, and porosity values: Loose Packed down Bulk specific gravity Weight per cu. ft. . Porosity .... . 0.29-0.70 0.92-1.19 .>4-f«lbs. 58-74 lbs. ( 73-89% 55-65% MICROCRYSTALLINE SILICA (TRIPOLI) 11 Table 2.— I'article-Size Distribution of Commercial Silicas (I n cumulative percent finer than sieve listed) Silica samples Size* (microns) 1 2 3 4 5 6 7 2 5 10 15 20 25 44 14 20 173^ 32 31 441^ 60 26H 36 38 56 61 733^ 86 47 52 55 71 84 89 93 64 66 68 80 91 94H 94 76 72H 75 843^ 93 943^ 95 85 74 783^ 87 94 95 95 981^ 96 99 98 100 100 100 53 991^ 983^ 100 100 74 100 100 a 44 microns is the size of the opening in a 325 mesh sieve. 53 microns is the size of the opening in a 270 mesh sieve. 74 microns is the size of the opening in a 200 mesh sieve. ^% Fig 5.— Photomicrograph of 30 by 40 micron silica, magnified about 600 times. The general shape of the grains is well shown. The grains are actually composed of many smaller particles, although only some of them are clearly visible. The many out-of-focus particles scattered throughout the ground mass are individual particles of the same sort as those that make up the larger grams or are small cluster grains. 12 SILICEOUS MATERIALS Those samples with the finer particle size had the lower bulk specific gravity and weight per cubic foot and the higher porosity. True specific gravity. — True specific gravity of several samples of silica, as meas- ured by the standard pycnometer method," averaged 2.644 at 25 °C. Oil absorption. — Oil absorption, as deter- mined by the standard rub-out method,'^ on eight samples of commercial silica ranged from 26.6 to 33.1 pounds of oil per 100 pounds of silica. Following is the average oil absorption of six size fractions of two samples : Size fraction Oil absorption (Microns) (lbs. oil per 100 lbs. silica) Minus 5 34 5 by 10 48 10 by 20 53 20 by 30 55 30 by 44 53 44 by 74 58 The higher oil absorptions of the size frac- tions coarser than 10 microns are believed to be due to a combination of the irregular surface and the probable porosity of the cluster grains which comprise the bulk of these fractions. Specific surface. — Specific surface, also called surface area, was determined by the nitrogen adsorption procedure on two com- merical silica samples and was respectively 38,100 and 64,400 sq. cm. per gram of silica.*^ The specific surface of three size fractions of a silica sample were: Microns Sq. cm. per gram Ox 5 47,700 10x20 23,300 30x44 17,700 Measurements of specific surface with the Blaine air permeability meter gave re- sults only about a fourth as great, and for the eight silica samples ranged from 6,550 to 14,890 sq. cm. per gram. ^A. S. T. M, designation: D 153-39. * Gardner, H. A., and Levy, S. A., Pigment and color index: Am. Paint and Varnish Manuf. Assoc. Circ. 352, p. 523. June 1929. * Determinations by the Geochemical Section, Illinois Geol. Survey. Abrasiveness. — There are no published data on the abrasiveness of southern Illinois silica, nor is there a known standard method for testing the abrasiveness of "amorphous" silica or other high-quartz content powders. Abrasiveness data given below were ob- tained by means of an apparatus patterned after that employed by Smith^° for testing the abrasiveness of calcium carbonate pow- ders. Two small type-metal blocks weighing about 3.1 grams each under a load of 68 grams were subjected to abrasion by a silica slurry on a glass plate. The plate was rotated at 55 revolutions per minute (fig. 6) and the rotating mechanism was belt-driven from a separately mounted motor to avoid vibrations in the testing unit. Abrasiveness was measured as the average weight loss in milligrams of the two metal blocks after 6000 revolutions of the glass plate. A maxi- mum variation of 10 percent was considered permissible in the weight loss between the two metal blocks. The slurry used consisted of an ethylene gljTol and silica mixture of such consistency that a uniform layer emerged from beneath the spreaders and did not hide the metal blocks as the glass plate rotated. This re- quired between 9 and 14 cc. of the glycol for the standard weight of 20 grams of silica used. The glycol was used instead of water in order to keep the viscosity of the slurry constant during the test. A new glass plate was used for each test. Eight samples of commercial silica pro- duced the following weight losses in milli- grams of the type-metal blocks: 4, 4, 9, 19, 20, 24, 24, and 25. The average loss was 16 milligrams. The average weight loss in milligrams produced by various size frac- tions from two samples of commercial silica were as follows: Size Loss in microns in milligrams Ox 5 2 5 X 10 7 10x20 17 20x30 52 30 x 44 86 44 x 53 100 *^ Smith, R. W., Machine for testing dentifrice abrasion: Ind. and Eng. Chem., Anal. Ed., vol. 12, no. 7, pp. 419-433, July 15, 1940. MICROCRYSTALLINE SILICA (TRIPOLI) 13 Fig. 6. — Abrasiveness testing apparatus. A is the type-metal block which is abraded ; a similar block on the left side of the machine is obscured. B indicates the spreaders which maintain a uni- formly thick layer of silica slurry. C is a rotating table topped with a circular glass plate bearing the silica slurry. The belt to the independently mounted motor which drives the rotating table shows in the lower part of the photograph, also the counter which records the number of revolu- tions of the table. 14 SILICEOUS MATERIALS Table 3. — Chemical Analyses of Commercial Silica and of Clay from a Silica Deposit Silica- Silica'* Silica** Silica'^ Clay« Si02 99.29 99.13 0.64 0.04 0.22 01 0.36 98.84 0.76 0.05 0.27 0.03 0.37 98.15 95 07 0.05 0.14 0.03 09 07 0.68 65.88 AI2O3 Fe203 MgO 0.36 04 0.01 23.88 0.65 0.53 CaO 0.02 0.73 NazO K2O — H2O CO2 Loss on Ignition 0.40 0.32 1.02 0.87 09 7.48 Total 99.72 100.39 100.29 100.16 100.49 » From Clear Creek formation. ^ From Grassy Knob "formation." « Clay along slip in Clear Creek formation silica deposit. Experiments to relate abrasiveness to the particle-size distribution of commercial silica samples did not yield conclusive cri- teria for forecasting abrasiveness. However, the particles coarser than about 20 microns appear to be primarily responsible for the abrasiveness of the samples. Comparison of the abrasiveness of the 20 by 30 micron size fractions of silica, and of feldspar and high-calcium limestone ground in the laboratory, showed a wear of 51 milligrams for the silica, 32 milligrams for the feldspar, and 12 milligrams for the limestone. The rate at which a sample of commer- cial silica and of a 30 by 44 micron fraction from the same sample produced wear of the metal blocks is shown below. No. revolutions of glass plate 0-2000 . . 2000-4000 . 4000-6000 . 6000-8000 . 8000-10,000 Amount of wear (milligrams) Commercial 30 by 44 tni- silica cron fraction 17 24 26 29 27 The effect of the abrasiveness tests on the 30 by 44 micron fraction mentioned above was primarily the production of particles finer than 10 microns by breakage of the original coarser particles. Fusion point. — Two samples of commer- cial silica were heated to cone 32 (about 1700°C; 3092°F) and did not fuse at this temperature. Chemical composition. — Table 3 gives the results of chemical analyses of silica pro- duced from the Clear Creek and Grassy Knob formations, also an analysis of clay, probably a mixture of the clay minerals kaolinite and illite, from a six-inch vein along a "slip" in a silica mine. Mining and Processing First production of silica from southern Illinois was reported in 1906, ^'' ^^ when three processing mills were in operation. Currently two companies operate mills in the district: Ozark Minerals Co. at Elco and Tamms Industries, Inc., at Tamms. The mill of the Olive Branch Minerals Company at Olive Branch is idle. Silica is being mined from both the Clear Creek and Grassy Knob formations by room-and-pillar methods (fig. 7). In gen- eral the procedure involves one or more entries extending underground from the outcrop of the silica stratum. Approximately parallel rooms are developed with pillars roughly 20 feet square and rooms of about the same size. The height of the rooms varies between about 10 and 30 feet and is "Mineral resources of the United States, 1906: U. S. Geol. Survey, p. 1044, 1907. (The material is reported as "infusorial earth," which at that time was not differentiated in the statistics for silica.) 12 Bain, H. F., Analysis of certain silica deposits: Illinois Geol. Survey Bull. 4, Yearbook for 1906, p. 185, 1906. MICROCRYSTALLINE SILICA (TRIPOLI) 15 commonly 15 to 25 feet. The floors of some mines have a gentle "roll." The mine roofs are carefully arched ; timbering is seldom necessary except at mine entrances. Horizontal blast holes are drilled with jack hammers in the harder layers of the silica deposits and are shot with dynamite. The silica is loaded mechanically or by hand to trucks which take it to the mills. In the currently operating mills, the silica is first passed through a ring pulverizer and is then further dry-ground in ball mills. Various size-grades are produced by air classification. The mill at Olive Branch is reported to have used a wet-grinding process. Uses Southern Illinois silica is produced in various grades of fineness and is sold as an inert mineral filler for paints, as wood filler, as a constituent of cold-water paints, as an admix for concrete, as a foundry parting or facing, for use in various ceramic products for which purpose it is sometimes called "potter's flint," for use in buflfing com- pounds, as a fine abrasive in polishes and scouring compounds, and as a polishing agent. A very fine sized grade known as "white rouge" is used especially for polish- ing optical lenses. Only white material is ordinarily used for the purposes mentioned. Southern Illinois silica has been suggested for making silicate products such as sand- lime brick.^^ Methods of bleaching off-color silica to yield a white product have been described.^* Resources The Clear Creek formation has been a major source of silica, primarily from a tract west of Mill Creek and Elco (fig. 2). The formation underlies a considerable area (fig. 1 ) beyond the limits of the Mill Creek- Elco tract, so that additional deposits may be present in the vicinity of that tract and also in other areas. Some of the silica mined in the tract near Wolf Lake may have come from the Clear Creek formation. Silica has been mined from the Grassy Knob "formation" near Olive Branch. Some of the silica deposits near Wolf Lake and a few of the deposits in the Mill Creek- Elco tract may also be in the same forma- tion. In Alexander County the lower part of the formation appears more likely to con- tain silica deposits than the higher parts of the formation. On this basis, the western part of the outcrop area of the Grassy Knob formation in Alexander County seems to offer better possibilities for silica than the eastern part. The Wolf Lake, Alexander County, and Mill Creek-Elco areas all have possibilities for additional deposits. Small amounts of silica were once mined from the Bailey formation near Thebes, but little else is known about the silica possibili- ties of the formation. In the outcrop area east of McClure and Reynoldsville, there are relatively fresh outcrops of the Bailey formation whose character suggests that if they were thoroughly leached they might give rise to silica deposits. Southeast of Thebes the formation is partly leached in places but only the surficial parts of the deposits are completely leached. It is believed that Fig. 7. — Interior of silica mine near the entrance. ^^ Williams, W. S., Artificial silicates with reference to amorphous silica: Illinois Geol. Survey Bull. 14, Year- book for 1908, pp. 277-292, 1909 Parr, S. W., and Ernest, T. R., A study of sand-lime brick: Illinois Geol. Survey Bull. 18, p. 9, 1912. *' Machin, J. S., and Tooley, F. V., Decolorization of south- ern Illinois silica: Illinois Geol. Survey Rept. Inv. 47, 1938. 16 SILICEOUS MATERIALS silica developed from the Bailey formation is more apt to resemble the Grassy Knob "formation" silica than the Clear Creek silica. NOVACULITE AND NOVACULITE GRAVEL Definitions The terms "novaculite" and "novaculite gravel" have been applied to certain southern Illinois materials for many years, but no definitions covering southern Illinois usage are known. Holbrook^'"' uses the term novac- ulite synonymously for "hard chert" but apparently makes no distinction between novaculite and novaculite gravel. It is believed that the term novaculite as now commonly used in southern Illinois is applied to dense, white or almost w^hite chert in relatively solid ledges which com- prise a deposit several feet or more thick, without other interbedded materials. The name is thus applied to chert that has a defi- nite mode of occurrence and physical prop- erties and so differentiates it from the more common deposits of porous or shattered chert. Novaculite gravel is a term loosely ap- plied in southern Illinois to a generally an- gular chert gravel which usually has not been transported from its place of forma- tion. The gravel contains varying but com- monly comparatively small quantities of microcrystalline silica and clay. The derivation of the term novaculite as applied to the Illinois chert and chert gravel is not known, but possibly the name was used because Illinois chert resembles some of the Arkansas novaculite. Holbrook remarks on the resemblance of the two materials. Dis- cussions of the origin and character of the Arkansas novaculite have suggested that it is metamorphosed sandstone,^'' quartzite,^' an extremely fine-grained sandstone, ^"^ meta- I'^Holbrook, E. A., The amorphous silica of southern Illinois: Eng. and Mining Jour., vol. 103, no. 26, p. 1136, June 30, 1917. " Owen, D. D., Second report of a geological reconnais- sance of the middle and southern counties of Arkansas, made during the years 1859 and 1860: Arkansas Geol. Survey, pp. 23-25, 1860. " Weigel, W. M., op. cit., p. 54. ^■^ Ries, H., Economic geology: New York, John Wiley, p. 288, 1930. morphosed chert^^ and chert. -^ Although the Illinois material is not identical with the few samples of Arkansas novaculite examined during this investigation, their superficial resemblance could explain the original ex- tension of the term to Illinois chert. NOVACULITE A chemical analysis of a sample of novac- ulite is given in table 4. X-ray diffraction studies^^ of a sample of novaculite revealed that it consists of alpha-quartz particles, mostly smaller than 1 micron in diameter. A sample heated to 800° C. showed marked grain growth and consisted of alpha-quartz particles in the range probably from 1 to 100 microns. The same specimen when reheated to 1500°C. consisted of roughly equal amounts of large alpha-quartz grains and of alpha-cristobalite grains produced by the inversion of small quartz grains.^^ A second portion of the same original novaculite sample when heated without in- terruption to 1500°C. consisted almost en- tirely of alpha-cristobalite with only a few medium sized alpha-quartz particles.-^ A sample of white novaculite fused be- tween cones 31 and 32 or at about 1690°C. (3074°F.). The apparent specific gravity of 6 pieces of white novaculite having no visible pores varied between 2.05 and 2.39, or about 125 to 150 pounds per cubic foot and averaged 2.21 or about 138 pounds per cubic foot. The freshest outcrop of novaculite ob- served was in the lower level at the Novacu- lite Gravel Co. pit near Tamms at the center N Vi sec. 36, T. 14 S., R. 2 W. Here a seven-foot thickness of white novaculite, stained slightly yellow along some joints, was formerly visible above the level of the railroad track and low in the slope of the hill. This deposit is probably part of the Grassy Knob "formation." 1" Branner, J. C, On the origin of novaculites and related rocks: Jour. Geol., vol. 6, pp. 368-371, 1898. 20 Purdue, A. H., and Miser, H. D., Hot Springs folio: U. S. Geol. Survey Folio 215, p. 4, 1923. Purdue, A. H., and Miser, H. D., Geology of the DeQueen and Caddo Gap quadrangles, Arkansas: U. S. Geol. Survey Bull. 808, p. 57, 1929. Tarr, W. A., and Twenhofel, W. H., Chert and flint, p. 519, in Treatise on sedimentation, by W. H. Twenhofel: Baltimore, Williams and Wilkins, 1932. 2^ Bradley, W. P., personal communication, 1948. NOFJCULITE GRAVEL 17 Table -4. — Chemical Analyses of Novaculite AND Clay from Novaculite Gravel'' £ 3 6 u ^ OJ > Ox> .^§ c-g 3 ^ §2 u c^ *-> aj rt *^ :zr^ £;S u'B Si02 • 97.67 96.19 2.16 79.42 AI2O3 1.16 13.70 Fe-iOs .33 .47 46 MgO .11 .13 CaO .00 .00 H.O .09 0.10 0.93 Loss on ignition . . . .62 Total 99.89 99.88 ■1 All samples from north pit of Novaculite Gravel Co. at Tamms. *> Partial analysis. Outcrops of the middle portion of the Grassy Knob "formation" in Union and Alexander counties and the lower part of the Clear Creek formation in Alexander County may contain considerable amounts of novaculite. The whitest deposits are most likely to be found low in ridges and hills where the novaculite is below the zone of pronounced iron staining. Novaculite has not been mined exten- sively. Production has followed usual quarrying methods designed to produce a minimum of fines, followed by hand trim- ming to remove any iron oxide stained ma- terial. Novaculite is reported to have been sold for making refractory brick and for sodium silicate. The results of an investigation on the use of novaculite for silica brick manu- facture is discussed under the uses of novac- ulite gravel. NOVACULITE GRAVEL Character The size of the chert particles varies in different deposits of novaculite gravel. Some deposits contain no material or only a little material coarser than 2 inches, others have a considerable amount of coarser chert frag- ments and still others have in addition large chunks. There is a similar variation from layer to layer within many deposits. The particle-size analysis of two samples of novaculite gravel, taken at different times and from different parts of the pit of the Novaculite Gravel Co. at Tamms, in the NEi/4 NW14 sec. 36, T. 14 S., R. 2 W. (fig. 8) are given in table 5. All the par- ticles coarser than 20 mesh were chert. In the finer sizes chert grains also predomi- nate, with lesser amounts of angular quartz grains and, in the 20 by 100 mesh fraction, a very few rounded, more or less frosted, quartz grains. The material finer than 2 microns consists principally of clay minerals. The chert fragments from the gravel sam- ples were of three principal textural kinds: (1) dense, hard and angular; (2) porous, irregular, and comparatively easy to break; and (3) partly dense and partly porous (fig. 9). Results of a count to determine the approximate abundance of these three types of fragments in a gravel sample are given in table 6. It is reported that the road-making quali- ties of novaculite gravel from different deposits vary, especially the capacity of the gravel to bond into a good road and to re- sist traffic wear and rain washing. This is often attributed to variations in the amount of "clay," including in this term not only actual clay mineral material but also more or less microcrystalline silica. Al- though clay content may partly explain the differences, variations in particle size dis- tribution and in the proportions of the three textural kinds of chert fragments may also be important in "keying together" the gravel particles in the road. A chemical analysis of a sample of pit- run novaculite gravel from the Tamms pit is given in table 4, as well as a partial anal- ysis of clay lumps hand-picked from the same deposit. Undoubtedly the high silica (SiOg) content of the clay results from the presence of particles of microcrystalline silica mixed with it. The clay is a kaolinitic clay. The apparent specific gravity of the pieces of chert without visible porosity in the novaculite gravel is similar to that of the 18 SILICEOUS MATERIALS Table 5. — Sieve Analyses of Novaculite Gravel Sieve Percent retained Cumulative per- cent retained Sample Sample LM 104^ LM-Xb Sample Sample LM 104=* LM-Xb 0" hA"'. ;;;:::::::;:::::: 1.050" .742" .525" .371" .263" 4 mesh 6 mesh 8 mesh 10 mesh 14 mesh 20 mesh 28 mesh 35 mesh 48 mesh 65 mesh lOOme.sh 150 mesh 200 mesh 270 mesh Pan" Total 0.0 5.7 0.0 10.3 3.0 13.6 17.8 10.9 14.4 16.6 10.2 11.7 9.3 8.7 9.4 4.2 6.0 3.1 4.1 3.0 3.1 2.1 2.3 1.2 1.5 0.9 1.0 0.5 0.6 0.4 0.5 0.2 0.4 0.2 0.4 0.2 0.4 0.3 0.3 0.2 0.3 6.0 15.0 100.0 100.0 0.0 5.7 16 3.0 29.6 20.8 40.5 35.2 57.1 45 4 68.8 54.7 77.5 64.1 81.7 70.1 84.8 74.2 87.8 77.3 89.9 79.6 91.1 81.1 92.0 82.1 92.5 82.7 92.9 83.2 93.1 83.6 93.3 84.0 93.5 84.4 93.8 84.7 94.0 85.0 100.0 100.0 " Pit-run gravel, excluding large chunks, b Minus IJ^-inch pit-run gravel. <= The pan fraction of Sample LM 104 contained 2.1 percent of material finer than 2 microns and that of Sample LM-X 6.5 percent. Table 6. -Character of Fragments in Novaculite Gravel Sample LM 104 Sieve openings (in inches) Plus 13^ . . . . lA by 1.050 . . . 1.050 by .742. . . .742 by .525 . . . .525 by .371 . . . .371 by .263 . . . .263 by .185 (4 mesh) Average . Percent by number of fragments Dense fragments 67 33 57 41 32 28 38 42 Partly dense, partly porous fragments 23 41 22 15 26 18 Porous fragments 33 44 43 18 46 57 36 40 ^UlACULITE GRAVEL 19 novaculite, previously mentioned. The pres- ence of visible pores decreases the specific gravity. Occurrence and Resources Slumping usually obscures details of nat- ural outcrops of novaculite gravel, but where they are freshly exposed the deposits show considerable interbedding of diiierent ma- terials. Some layers are mostly chert, others are clayey, and still others are principally microcrystalline silica, but chert is the pre- dominant material. The beds of the deposits, though roughly horizontal, are locally wavy and irregular and in places show minor dis- placement of bedding due to small faults. Novaculite gravel deposits may be almost white, yellow, brown, pink, or red, due pri- marily to iron oxide. Some deposits are roughly uniform in color, others exhibit considerable lateral and vertical variation. Physically the color of the deposits results principally from the presence of yellow, brown or red clay and silica which occurs with and as a coating on the chert frag- ments. Some of the fragments are stained yellow, brown, or red more or less through- out but many of them when thoroughly washed are seen to be only slightly spotted or streaked on the surface. Novaculite gravel is probably available from the Grassy Knob and Clear Creek formations at many places in Union and Alexander counties. It has been produced from the Grassy Knob "formation" at Olive Branch and between Olive Branch and Tamms, and from the Grassy Knob and Clear Creek formations at Tamms. Prob- ably leaching and weathering of any of the chert and cherty limestone formations of Alexander and Union counties would give rise to some variety of novaculite gravel. The deposits may be as much as 100 feet thick or more although they are likely to be thinner. Mining and Processing Novaculite gravel is mined from open pits and usually loaded mechanically. The gravel characteristically stands in steep faces (fig. 8) which are commonly shot to loosen and throw down the gravel though some deposits may be workable with mechanical loading equipment without blasting. For- merly, at one pit small horizontal tunnels dug into the gravel face were loaded with explosives to shoot down the gravel. The gravel from some deposits is said to be used, without processing, for road con- struction. That from other deposits is passed over a coarse screen or grizzly to remove large lumps. At one time at Tamms, gravel was sized by screening ; oversize pieces w^ere crushed and then screened. Uses Novaculite gravel has been used for roads, street paving, and railroad ballast. An in- vestigation of the use of novaculite and washed novaculite gravel for the manu- facture of silica brick indicates that it is of "sufficient purity to make a high grade silica brick, and may be crushed to yield desirable distribution of grain sizes. The novaculite gravel may easily be washed to remove associated clav, after which it has Fig. 8. -Novaculite gravel of the Grassy Knob and Clear Creek formations exposed in the pit of the Novaculite Gravel Co. at Tamms. 20 SILICEOUS MATERIALS Fig. 9. — Types of fragments found in novaculite gravel. The six fragments on the left are dense smooth-surfaced chert having the fracture characteristic of this material. The six fragments on the right are porous and have highly irregular surfaces. The fragments are one to two inches. properties similar to the massive pure rock"-'" (novaculite). Silica brick suitably made from the novaculite or washed novaculite gravel are ''similar in appearance and prop- erties to the commercial silica brick made from quartzite ganister."^'^ Novaculite gravel has been dug at a num- ber of places in southern Illinois, especially in the vicinity of Olive Branch and Tamms, and has been produced or is reported to have been produced from pits at the following 22 Parmelee, C. W. and Harman, C. G., Southern Illinois novaculite and novaculite gravel for making silica refractories: Illinois Geol. Survey Rept. Inv. 117, pp. 8 and 55, 1946. 2' Parmelee and Harman, op. cit. locations. Other deposits may also have been worked. S 1/^ NE 14 NW Va sec. 36, T. 14 S., R. 2 W., and center W j/^ of the same section. Two pits operated by the Novaculite Gravel Co., Tamms. . SW 14 SE 14 NW Va sec. 28, T. 15 S., R. 2 W. SE 14 SE 14 sec. 29, T. 15 S., R. 2 W. ad- jacent to the mill of the Olive Branch Min- erals Co., Olive Branch. SW 14 SW 14 SE 14 sec. 19, T. 14 S., R. 1 W. ORIGIN OF MICROCRYSTALLINE SILICA, NOVACULITE, AND NOVACULITE GRAVEL Introduction The mode of origin of the silica and no- vaculite gravel of southern Illinois bears di- rectly on the distribution and character of the resources and is significant to an under- standing of the details of deposits in rela- tion to their commercial use. In brief, it is believed that these materials are in a broad way the result of the leaching of siliceous limestone and chert, especially calcareous chert. The various strata of the Bailey, Clear Creek, and Grassy Knob formations have been leached in a similar manner, al- though not everywhere to the same extent. In discussing origin, therefore, the forma- tions are treated as a unit, with reference to individual formations only in the case of specific differences. Nature of Formations Before Leaching An understanding of the original nature of the formations which have given rise to the microcrystalline silica and other ma- terials may be gained from a study of de- posits which are not deeply weathered and from those which are underground and well away from the outcrop. The Bailey formation, as found in wells drilled at widely separated places in the southern quarter of the state, is character- istically made up of cherty limestone with beds and nodules of tan chert. The lime- stone itself ranges from siliceous to non- siliceous. It is generally fine-grained but locally is coarse-grained.^* In general the underground Bailey is simi- lar to the formation where it crops out ex- tensively in and adjacent to the bluffs of Mississippi River in western Union and northern Alexander counties (fig. 1). The outcropping rock also shows thin shaly part- ings between the limestone layers and chert in both continuous and discontinuous layers "Workman, L. E., personal communication, 194S. Fig. 10. — A relatively fresh outcrop of Bailey limestone exposed in Mississippi River bluffs near Aldridge. Note the thin-bedded limestone, chert nodules, and the chert band at the bottom of the camera case. (fig. 10). The basal part of the formation contains chert which is blue black where fresh. Table 7 gives analyses of Bailey limestone from the blufif of Mississippi River near Aldridge (sample NF 91-92) and at Reynoldsville (sample NF 93-94) about 15 miles to the south. The basal Bailey is represented by sample 147R, obtained about two miles southeast of Thebes. The Clear Creek and Grassy Knob for- mations in the southern half of Illinois, as known from well cuttings, are cherty lime- stone similar to the Bailey formation de- scribed above. "'^ Available data regarding 25 Workman, L. E., personal communication, 1945. [21] 22 SILICEOUS MATERIALS Fig. 11. — Grassy Knob chert. The dark areas represent limestone that has not been replaced by silica. The top of the slab is a stylolitic parting. A short distance below the top a similar parting brings chert in contact with chert in some places (as at .^), limestone with chert in others (as at B), and limestone with limestone in still others (as at C). About half natural size. the formations underground are from churn drill samples and therefore are not detailed. They suggest the presence of layers of chert as well as limestone containing thin- ner chert layers and chert nodules. In the extreme northern area of outcrop, in south- ern Jackson County, the lower two-thirds of the Grassy Knob "formation" is chert but the upper third is probably alternating layers of chert and dolomitic limestone. In Union County the exposures of the middle part of the Grassy Knob ''formation" are chert, as much as 80 feet thick. -^ In Alex- ander County the exposed lower part of the formation is much weathered, but where fresh was probably interbedded limestone and chert. The Clear Creek formation in northern Union County north of an east-west line through Jonesboro is chert which contains variable thicknesses of siliceous limestone. Weller, J. M., op. cit.. p. 12. South of this line the original character of the formation has been obscured by weather- ing. Origin of Chert and Other Silica Materials in Devonian Formations The Bailey, Grassy Knob, and Clear Creek formations all contain marine fos- sils and so are believed to have been de- posited in a sea which at one time covered southern Illinois. Whether the chert and disseminated quartz now found in the for- mations is primary or secondary was not investigated. However, data which follow suggest that if the silica (SiOg) is all pri- mary, groundwater has redistributed a con- siderable amount of it within the Grassy Knob "formation," and probably within other formations since their deposition. As previously stated, the Devonian chert formations away from their outcrops are not greatly different from the same formations ORIGIN OF SILICEOUS MATERIALS 23 Fig. 12. — Grassy Knob chert showing a stylolitic contact between siliceous limestone (below) and chert (above). About one and a half natural size. at the outcrops, except that the Grassy Knob and Clear Creek formations probably are somewhat more siliceous in most of the outcrops. This suggests that the chert and other siliceous materials were primary or that widespread silicification has occurred since deposition, or both. The middle part of the Grassy Knob "formation" in places contains a breccia composed of angular chert fragments in a microcrystalline quartz matrix. A thin-sec- tion of this breccia reveals"^ that some parts of the margins of the chert fragments are bordered by a bleached zone which blends indistinctly into the matrix, whereas other parts of the margins have an abrupt, sharp contact with the matrix. Numerous rhom- bic pseudomorphs after carbonates, now quartz-filled, straddle the contact between the chert fragments and the matrix, regard- less of whether the contacts are sharp or in- distinct. The pseudomorphs are more nu- merous in the matrix than in the breccia fragments. They have slight iron stains around their margins, suggesting that they formed after a carbonate containing a small amount of iron, probably dolomite. By transmitted light, the chert of the breccia fragments is light brown and shows a minutely spherulitic structure, whereas the chert of the matrix is gray and fine-grained and lacks such a structure. At transitional contacts, the spherulitic struc- ture decreases from the fragments to the matrix. The matrix contains partly rounded quartz grains which are secondarily en- larged by silica. The breccia fragments con- tain what appear to be spines. The general appearance of the breccia sug- gests a limestone breccia which has been silicified. The Lower Devonian rocks near Rock Island, Illinois, contain a limestone breccia which, if silicified, might yield a material like the Grassy Knob breccia. Al- though the Grassy Knob breccia is regarded as older than that at Rock Island, ^^ the existence of limestone breccia in the De- vonian rocks of Illinois lends credence to the thesis that the Grassy Knob breccia was originally limestone. If so, silicification is secondary. Some massive beds of the Grassy Knob "formation" have stylolitic partings. The " Grogan, R. M., personal communicati Workman, L. E., personal communication, 1948. 24 SILICEOUS MATERIALS stylolites in some places are bounded above and below by chert, in others by limestone above and chert below, or vice versa, and in still others by limestone only (figs. 11 and 12). The stylolite surfaces are com- monly coated with thin layers of red or brown clay. A thickness of at least 4.5 mm. of limestone has apparently been dissolved in the specimen shown in figure 12, judging from the height of the stylolitic columns. The limestone below the stylolite contains 14 percent of material insoluble in acid, chiefly clay and finely divided silica. The solution of a 4.5 mm. thickness of this lime- stone would give rise to a residue about 0.6 mm. thick, most of which might be expected to accumulate along the stylolite. However, the actual amount of residual material pres- ent is much less than this, probably not over 0.1 mm., which suggests that the limestone in which the stylolite developed probably was relatively pure and that the present silica content is at least partly sec- ondary. The same line of reasoning, applied to other stylolites of greater vertical ampli- tude and involving limestone having higher siliceous residues, leads to a similar conclu- sion. As chert does not commonly form stylolites, the presence of stylolites in it also suggests secondary silicification of a limestone. Molds of trilobites and brachiopods have been found in the Grassy Knob chert at Olive Branch, and brachiopods at Tamms. Sponge spicules also are found in the Grassy Knob chert. Fossil molds occur in the Clear Creek chert. These data are not regarded as conclusive evidence for either the pri- mary or secondary origin of the chert for- mations as a whole. General Process of Disintegration of THE Devonian Formations The disintegration of the Devonian siliceous formations to microcrystalline silica and novaculite gravel deposits is thought to have come about roughly as follows : When the formations were stripped of their cover or otherwise brought within the zone of weathering, groundwater attacked them. The relatively insoluble chert was little af- fected at first, but the more readily soluble calcareous beds were gradually leached of their carbonates. This in time resulted in a series of chert strata interspersed with the soft clayey or siliceous residues of the wholly or partly leached calcareous beds which in some places contained residual nodules of chert. This caused settling and internal re- adjustments in the weathered part of the formation, with the result that the chert beds fractured. Such fracturing was addi- tional to older fracturing which resulted from diastrophic causes. The net result was increased porosity and ease of groundwater circulation, which in turn accelerated removal of calcium car- bonate and prepared the deposits for free groundwater circulation. This free circu- lation, acting over a long period of time, caused the major chemical disintegration of the cherty layers. The foregoing general process was modi- fied according to the amount of calcium carbonate and clay present in the original formations and the amount of porosity and permeability developed by the weathering. The internal readjustments within the for- mations were probably gradual, and set- tling was not of great vertical amplitude but occurred slowly without major dis- turbances of the continuity of strata. How- ever, in places small faults were formed and also open fissures, some of which ap- parently extended to the ground surface and were filled with clay, either when they were formed or at a later date. Possibly the clay came from overlying Cretaceous or Tertiary deposits. The Backbone limestone formation lies between the Grassy Knob and Clear Creek formations at some places in southern Illi- nois. In northern Union County its thick- ness may be as much as 200 feet or more, but to the south the formation is probably thin- ner and in places it may be absent or rep- resented by only a thin stratum of chert, as near Tamms. There are unconformities at the top and bottom of the formation. ^^ These variations in thickness may be due to differences in original depositional thick- ness, to erosion of the Backbone limestone 2!) Weller, J. M., Preliminary geologic map of parts of the Alto Pass, Jonesboro, and Thebes quadrangles: Illinois Geol. Survey Kept. Inv. 70, pp. 13-14, 1940. ORIGIN OF SILICEOUS MATERIALS 25 before the Clear Creek formation was de- posited, to removal of all or part of the lime- stone by solution, or to a combination of these factors. It is difficult to see how the extensive leaching which is postulated to have affected the Clear Creek and Grassy Knob formations could have taken place without also affecting the Backbone limestone. Possible evidence of such leaching may be the closely spaced stylolites which characterize 40 feet of Back- bone limestone exposed along Hutchins Creek in the eastern part of T. 11 S., R. 3 W., east of Wolf Lake. The apparent removal of the limestone from other places may be another result of severe leaching. If solution is responsible for all or part of the variation in thickness of the Backbone limestone, and if it took place before or dur- ing the leaching of the other Devonian rocks of extreme southern Illinois, the resulting thinning of the Backbone limestone and accompanying settling and fracturing of overlying strata would have aided greatly the chemical and ph3^sical disintegration of the Clear Creek formation which overlies the Backbone. Perhaps, by favoring ground- water circulation, extensive solution of the Backbone formation might also have has- tened the same processes in the underlying Grassy Knob "formation." Thus those areas where the Backbone limestone is absent or thin may be the areas where the overlying and underlying formations were most com- pletely leached ; they may therefore delineate the more favorable tracts of the occurrence of microcrystalline silica and novaculite gravel. Further information is needed re- garding the probable original distribution and thickness of the Backbone limestone as well as field evidence of its solution. Only then can the full effect of its possible solu- tion on the Clear Creek and Grassy Knob formations be accurately evaluated. Interpretation of the Origin of the Existing Lithologic Phases of the Siliceous Devonian Formations In view of the foregoing, the origin of the various present lithologic phases of the sili- ceous Devonian formation were probably as follows : (a) Solid chert in deposits of consider- able thickness. — This lithologic phase was probably originally chert which contained no carbonates, or at most only very small amounts, and therefore was not readily weathered or leached. Further it is thought to have been underlain within the zone of weathering by noncalcareous strata, and therefore it was not subjected to strains and stresses such as might otherwise have re- sulted from settling of the underlying beds as a result of leaching. (b) Much-fractured chert. — Layers or deposits of dense, much-fractured chert are thought to have been noncalcareous or only slightly calcareous and, therefore, little af- fected by solution. Fracturing resulted from readjustments which accompanied settling when the carbonates were removed from underlying calcareous materials. Lay- ers or deposits of porous, fractured chert may owe their porosity to the original pres- ence of masses of carbonates whose leaching could have had a part in fracturing the chert or may have resulted only in the porosity. In either case fracturing was more or less due to the same factors which op- erated on the dense chert. (c) Friable, pulverulent microcrystalline silica with or without chert nodules. — This silica is believed to have been originally a highly siliceous limestone. The rock, how- ever, contained sufficient carbonates so that the quartz grains were not extensively inter- locked and as a result, after the carbonates were removed, the resulting silica had a low mechanical strength and a loose texture. Chert nodules, if present, are the remains of chert nodules In the original limestone. Usually the nodules grade into the enclos- ing silica, probably as a result of the weath- ering of their peripheral portions. The in- terior part of many nodules is still fresh and blue gray, cream, or white. A newly made road cut in the SW 14 SE 14 sec. 15, T. 15 S., R. 3 W., southeast of Thebes, provided an excellent opportunity to study the effect of leaching on siliceous lime- stone. A six-inch layer of such limestone was traced about 350 feet from its un- weathered state through a series of Inter- 26 SILICEOUS MATERIALS Table 7.- -Chemical Analyses of Limestone, Chert AND Silica from the Bailey Formation ^ ^ NF91-92— 95 ft. Bailey limestone NE of Aldridge, T 11 S, R 3 W, sec. 4, SWM NF93-94— 130 ft. Bailey limestone E of Reynoldsville T \Z S, R 2 W, sec. 20, N3^N3^ 147R-20 ft. Bailey limestone and chert, road cut SE Thebes, T 15 S, R 3 W, sec. 15, SWM, SEH L503-unweathered gray siliceous lime- stone 6" thick. Loca- tion same as 147R* L504-partly weathered siliceous limestone 6" thick. Location same as 147 L505-crcam colored silica 6" thick. Location same as 147 L506-black chert 6" thick. Location same as 147R L507-brown chert 6" thick. Location same as 147R Si02 . . . . 39.46 31.75 56.2 24.35 45.38 97.81 85.71 93.85 AI2O3 . 3.59 2.71 2.03 4.47 1.47 1.54 1.25 FesOa . 1.11 1.59 5.4 1.17 2.01 .30 .57 1.33 MgO . 2.57 2.61 1.1 2.86 1.74 .30 .94 .44 CaO . 28.50 33.52 19.7 36.87 23.85 .05 5.52 .89 NaaO . .31 .28 .10 .21 .24 K2O . .73 1.31 .06 .25 .24 H2O . .26 .17 0.45 .20 .45 .08 .18 .24 CO2 . 31.06 18.90 .14 4.44 .54 Loss on ignition . 24.42 28.27 17.0 31.81 20.97 .40 5.45 1.67 Total . 99.65 100.45 99.4 100.13 100.01 100.49 100.19 99.91 * These samples were all obtained at different places laterally from the same stratum exposed in a newly made road cut mediate stages to microcrystalline silica at the outcrop of the bed. Analysis 503 (table 7) shows the composition of the original limestone, analysis L-504 the composition of a partly weathered phase, and L-505 the composition of the resulting silica. If the CO,, CaO, and MgO data in the above analyses are calculated to calcium carbonate and magnesium carbonate, satis- fying first the carbon dioxide requirements of the calcium oxide, and then combining the remaining carbon dioxide with magnesia, the following values result: Total Samp/e CaCOs MgCOz carbonates SiOi .-503 . . 65.80 4.08 69.88 24.35 L-504 . . 42.57 0.34 42.91 45.38 L-505 . . 0.09 0.19 0.28 97.81 These figures, with the data in table 7, show the conversion, by leaching, of a sili- ceous limestone to microcrystalline silica. The analysis of the silica (L-505) shows a decrease in the amounts of AUO.^, Fe^O;^, NaoO, and K2O from that in the original limestone. Normally an increase in the percentage of these constituents might be expected as a result of the removal of car- bonates. If these compounds were distrib- uted uniformly throughout the original lime- stone stratum, the decrease suggests solution and partial removal of these four constitu- ents by groundwater in the final stages of silica production. (d) Firm, jnicrocrystallirie silica with or without chert nodules. — The firm silica is thought to be the result of leaching of a calcareous chert or extremely siliceous lime- stone in which the quartz grains interlocked or were in contact. The porosity of the firm silica is probably a rough measure of the amount of carbonates removed by leaching. The chert associated with this variety of silica is subject to the same remarks as the chert described under (c). Analysis L-506 (table 7) indicates the character of one such calcareous chert. Analysis L-507 shows the effect of partial leaching on such chert. Both samples came from the same stratum, but L-507 was near the outcrop whereas L-506 was an unweathered part of the deposit. (e) Chert fragments, including large chunks, niixed with clay and silica. — This phase is believed to have been originally a series of interbedded noncalcareous chert, calcareous chert, argillaceous limestone, and siliceous limestone strata. When the car- CANISTER 27 bonates were leached from a deposit of this sort, the resultant settling caused fractur- ing of the noncalcareous chert beds, which had remained essentially unchanged be- cause they contained no carbonates. During and after the leaching, settling, and fracturing of deposits of this type, iron compounds in the upper parts of the deposits were oxidized and iron oxide was probably introduced from overlying materials. Red clay from overlying materials may also have been carried into the deposits by down- ward-percolating groundwaters. These phe- nomena extend to a depth of over 100 feet in some deposits. The novaculite gravel appears to have been formed in the above manner, although the amount of clay and iron compounds in different deposits varies. (f) Clay in beds or along fault or joint planes. — The clay in beds was probably de- rived from argillaceous limestones which for the most part contained little free silica (SiOg). Most of the clay is highly plastic, cream-colored and only mildly gritty. The clay along joint or fault planes commonly has the same physical character. Much of it may have been washed into the joints from surface deposits of Cretaceous or other clays, but some of it may have come from the clay beds resulting from the leaching of argillaceous limestone and have been squeezed or carried by groundwater into the joints. Several samples of the clay from beds and joint fillings were composed prin- cipally of kaolinite with a lesser amount of illite. CANISTER The term ganister is applied in southern Illinois to a high-silica material, usually white, cream, light yellow, or red, which is loosely consolidated and readily disinte- grated into irregular particles an inch or less in size (fig. 13). The southern Illinois material is reported to have been used to make silica brick wherein it served the same purpose as a variety of crushed quartzite called ganister. Presumably this similarity of use gave the Fig. 13. — Photograph of ganister showing the diversity of particle size and shape, the type of disintegration which occurs when it is broken free from the natural deposit. Natural size. Photo by L. D. Vaughan. name ganister product. to the southern Illinois Occurrence Ganister of light enough color to make it currently of commercial grade appears to be largely restricted to areas in the vicinity of Mill Creek and Elco, and present and past commercial production has come mostly from these tracts (fig. 2). The commercial de- posits are probably of spotty distribution. North of Mill Creek to within a few miles of Jonesboro a somewhat similar material occurs in some places, but it is usually quite red, more coarsely granular, and is believed to be more cherty than the ganister deposits which have been worked. It has been used locally for surfacing roads. It is thought that the ganister deposits oc- cur in the Hartline "formation" (table 1, 28 SILICEOUS MATERIALS fig. 1 ) , although some ganister may also have been formed from calcareous siliceous beds of the Springville formation. Character of Deposits The ganister deposits vary in thickness, but a maximum is thought to be about 25 feet; most of the deposits worked commer- cially probably averaged between 10 and 15 feet thick. They are stratified, but the bedding is often irregular and in places contorted. Some deposits contain a few chert bands. Local vertical clay seams are present and may represent open fractures which were filled by clay from overlying deposits. Some fissures, ranging from 6 inches to 2 feet in width, contain gravel. According to Savage, "^^ such fissures are found in most of the ganister mines, and some of them can be traced upward to their connection with an overlying gravel deposit. Faults, probably of small displacement, are visible in some mines, and the ganister on both sides of the fault is dissimilar, indicat- ing a difference in the original rock mate- rial from which the ganister was formed. The lateral extent of individual commer- cial deposits is difficult to forecast. Although the ganister is believed to be a relatively con- sistent stratigraphic unit over limited areas, it varies in thickness and is probably dislo- cated in places by faulting. Both lateral changes in color and variations in the nature of the deposits, such as the presence of chert beds, may alter the commercial usabil- ity of the deposits. Character of Ganister As it comes from the mines, ganister consists of material ranging from lumps to powder. The lumps are readily broken down to pieces an inch or less in size. Re- sults of a particle size analysis of a sample of mine-run ganister (NF 415) is given below. When soaked in water, large lumps of ganister and some pieces of lesser size dis- integrate. A sample of mine-run ganister (LM 105) from a different deposit than ^" Savage, T. E., Geology and mineral resources of the Jonesboro quadrangle: Illinois Geol. Survey, manu- script. NF 415, soaked in water, and screened, ap- pears below. The particle size distribution represents the natural particle size of the source deposit. Sieve Inches 1 . %. Mesh 4 . 6 . 8 . 10 . 14 . 20 . 28 . 35 . 48 . 65 . 100 . 150 . 200 . 270 . 325 . Pan . NF415 LM105 Mine-run Mine-run, Percent water- by weight disintegrated Percent by weight — 9.9 1.2 — 4.3 — 3.1 4.5 3.7 — 3.9 1.1 5.5 5.8 7.5 6.6 7.0 6.4 8.1 7.4 9.0 5.8 6.1 6.0 5.3 4.8 3.1 6.1 2.9 4.0 2.8 5.1 5.3 3.4 1.5 — 0.9 16.2 18. Both samples contain a small amount of clay, probably not more than 2 percent, which is believed to be kaolinite.^^ The weight per cubic foot of a sample of mine-run ganister was 77 pounds; a simi- lar figure for minus |4-inch ganister was 70 pounds. Two large lumps of ganister aver- aged 99 pounds per cubic foot and indicate roughly the weight per cubic foot of the ma- terial in the natural deposits. Microscopic examination at low magnifi- cation showed a sample of Mill Creek ganis- ter to consist of grains and granular aggre- gates of irregular particles of opaque to translucent white, cream, or yellow material. A few rounded grains of clear quartz are present and also a few euhedral quartz grains. Examination of the smaller frag- ments of the sample by x-ray diffraction re- vealed that most of them are aggregates of quartz particles ranging largely between 1 and 5 microns in size."- Many of the larger ^^ Bradley, W. F., personal communication, 1941. ^^ Bradley, W. F., personal communication, 1941. CANISTER 29 Table -Chemical Analyses of Ganister and of Materials FROM THE HaRTLINE "FoRMATIOn" SiOa . . . AI2O3 . . . FeaOa . . . MgO . . . CaO . . . Na-20 . . . K2O . . . MnO . . . H2O- . . . CO2 ■ . ■ . . Loss on ignition Total Ganister 98.15 1.26 .15 .10 .15 .03 .13 .003 .15 .46 100.43 Hartline "formation" L-501'' 49.40 .60 .54 3.30 23.79 .09 .20 .05 21.55 22.19 100.11 L-502' 95.04 1.18 .78 .69 .50 .26 .25 .11 .31 1.21 99.91 NF-72d 86.08 7.21 2.68 .36 .50 1.08« 2.93 99.76 a Mine run from 12 ft. face in mine in SW corner SE ^ NE H sec. 1, T. 14 S., R. 2 W., near Mill Creek. ^ L-501, Hartline "formation," unweathered calcareous, siliceous rock from lower 6 ft. of an outcrop in road cut, NW ^ NW ^ SE 1.4 sec. 30, T. 13 S., R. 1 W., near Springville. •^ L-502, Hartline "formation," weathered .siliceous rock, upper 6 ft. of same outcrop as L-501. *! NF-72, Hartline "formation," weathered material from 12 feet of "gravel" in pit along road, NW ^ SE 14 sec. 36, T. 12 S., R. 2 W., near Jonesboro. « +H2O. lumps are probably similar but some have a hard chert-like core or are comprised large- ly of chert. The chemical analysis of a sample of ganister is given in table 8. The sample contained a small amount of clay which x-ray diffraction studies'^"' indicate is prob- ably kaolinite. It is likely that the alumina shown in the analysis of the ganister is com- bined with silica as this clay mineral. Prob- ably the clay content and the silica content vary in different deposits. The fusion point of two samples of ganister was above cone 32 (about 1700°C. [3092°F.]). Mining and Processing The Western Firebrick Co. of Granite City, Illinois, for a number of years has intermittently operated a ganister mine in the E 1/7 sec. 1, T. 14 S., R. 2 W. An- other mine in the same area was worked by the Southern Illinois Mining Corpora- tion of Murphysboro some years ago but is not known to have been operated recently. Usual production procedures have involved mining by the room-and-pillar method (fig. 14) hand loading, and trucking to railroad loading points. Locally red, coarse ganister ^^ Bradley, W. F., personal communication, 1941. or ganister-like material has been dug from open pits for road surfacing. Uses At present southern Illinois ganister is used mainly in making refractories. In the past it is said to have been used also for open hearth roof linings, gas retorts, silica brick, and to have been blended with flint clay, in amounts ranging from 10 to 50 per- cent, for the manufacture of refractory brick. ^* Some red ganister or ganister-like material is used for road surfacing. Resources Apparently there is a considerable ton- nage of ganister available in the Mill Creek- Elco district of southern Illinois. For rea- sons previously stated, some deposits may prove to be of variable character, and the extent of individual deposits can probably not be determined from outcrops. Test drilling appears to be the most logical method for the exploration of any sizable area, but small-scale test pitting and tun- neling will serve to prove up smaller ton- nages and deposits. ^■^ Parmelee, C. W., and Shroyer, C. R., manuscript, Illi- nois Geol. Survey, and personal communication. 30 SILICEOUS MATERIALS Fig. 14. — Interior of idle ganister mine near the center E ^/^ sec. 1, T. 14 S., R. 2 W., as seen from its entrance. The ganister beds dip toward the left side of the photograph. The room-and- pillar method of mining is shown. The rooms are ten to fifteen feet high. Origin The origin of the ganister deposits is not entirely clear. Springville shale crops out a short distance away from the mine entrances in some deposits and at the same level, sug- gesting that the ganister is derived from this shale. However, the presence of bands of chert nodules in some ganister deposits, even though uncommon, points to lime- stone or possibly calcareous chert as the source material rather than shale. The de- velopment of a siliceous material like gan- ister from the Springville shale would in- volve the removal of alumina from it by groundwater. Although this is not im- possible, a much more common phenomenon in the general area wherein the ganister de- posits occur is the leaching of carbonates from siliceous limestone or calcareous chert. The Hartline "formation" which overlies the Springville shale is chert or in places a siliceous limestone and is therefore a logi- cal source of the ganister. The top of the Springville shale is an erosional unconformity. This is well shown in a cut along the road near the center of the NE 14 NW 14 sec. 18, T. 14 S., R. 1 W. at the northeast edge of Elco where a "pocket" of red ganister about 15 feet thick with a maximum width of roughly 35 feet is bounded on each side and below by highly siliceous shale of the Springville formation. This outcrop suggests that the apparent con- tinuity of other more poorly exposed ganis- ter deposits with the Springville shale may actually indicate only that they occur in depressions, some of them probably of con- siderable size, in the top of the shale forma- tion. Although the Hartline "formation" is largely chert, it is thought that the basal part, which gave rise to the ganister, was a siliceous limestone, highly calcareous chert, or both. The original silica in these rocks apparently consisted of aggregates or irregu- lar masses ranging from a fraction of a milli- CALICO ROCK AND SILICEOUS SHALE 31 meter to an inch or more in size. Leaching the carbonates from the parent rock gave rise to the ganister. Probably the leaching of siliceous limestone produced the ganister, whereas calcareous chert or interbedded cal- careous chert and siliceous limestone may have produced the red, more coarsely granu- lar, ganister-like material used for road sur- facing. It is not known whether the leach- ing of a parent material occurring in de- pressions in the top of the Springville shale is requisite to the formation of commercial grade ganister but there may be a relation- ship. The nature of the parent material from which the ganister was formed is thought to be indicated or approximated by the rock which was exposed on a ridge about a mile north of Mill Creek in the NW ]4 NW 14 SE 14 sec. 30, T. 13 S., R. 1 W., during grading for a road. Here 6 feet of thin-bedded reddish gravelly material, hav- ing an uneven upper surface and composed of irregular porous fragments of iron-stained siliceous rock (Sample L-502, table 8), over- lies 6 feet or more of gray calcareous sili- ceous rock resembling a gray limestone (Sam- ple L-501) in layers from about Vz inch to about 4 inches thick. The lower material grades into the upper. The upper red material is regarded as the residue from the leaching of a rock which was originally like the underlying gray rock. When leached of its carbonates by acid, the gray rock gives a residuum very similar to the red material except in color. If the data for sample L-501 of the gray rock are recalculated on a carbonate-free basis, the resulting analysis closely approxi- mates that of the overlying red material. Sample NF-72 was obtained from a 12- foot exposure in a "gravel" pit on a ridge further north near Jonesboro. It was rough- ly similar to Sample L-502 but contained several rotted chert layers. The material in this pit is quite red and is a part of the Hart- line ''formation." It is higher in alumina and so is more clayey than the other samples mentioned above. A comparison of the analyses of samples L-502 and NF-72 with the analysis of gan- ister also given in table 8 shows much similarity between the two types of mate- rials. This fact suggests that both were derived from similar source material by a common method of origin. The formation of ganister from the cal- careous siliceous parent rock of the Hart- line "formation" by leaching is believed to have been accompanied by most of the phe- nomena which affected the Devonian cherty formations during leaching, including differ- ential settling, minor faulting, and the development of open joints and fissures due to either or both of these processes. The clay and gravel found in some of the joints at the present time probably washed or slumped into them from above. The known sources of white, cream, and light yellow ganister are restricted to the Elco and Mill Creek areas though this is not the limit of the area underlain by the parent Hartline "formation." The reasons for this are not evident but may be one or more of the following: ( 1 ) The parent rock was low in iron oxide content, and as a result its leached product was likewise low in iron oxide and therefore light in color. (2) Most of the deposits occur low in the ridges and except at the outcrops may be below the zone of oxidation or may have been protected from oxidation. Any colora- tion by iron oxide may have been produced after the ganister was formed. (3) The deposits occur in that part of Alexander and Union counties which ap- pears to have been subjected to more ex- tensive leaching than the area further north. CALICO ROCK AND SILICEOUS SHALE In the vicinity of Jonesboro the Spring- ville formation (fig. 1, table 1 ) is commonly chiefly brown, gray, or greenish shale, but in some places the formation consists largely of highly siliceous shale somewhat resem- bling slate. In the vicinity of Mill Creek, part of the formation consists of a series of chert-like strata locally called "calico rock" because of their variegated color (fig. 15). Calico rock is usually brown along joints and on long-exposed surfaces but when ffftwamr DEC 1 5 1998 1>2 SILICEOUS MATERIALS Fig. 15. — Sawed block of calico rock showing color banding. About one-half natural size. relatively fresh much of it is irregularly mot- tled with blotches and bands of brown, yel- low, red, and purple which stand out in contrast to a white, light-gray or light-green- ish groundmass. Some calico rock is light green. The rock usually occurs in beds 1/^ to 18 inches thick and many of the beds are separated by siliceous "calico" shale or clay partings ^4 to 12 inches thick. Outcrops of the calico rock are usually well jointed and the layers are considerably broken in roughly rhombohedral blocks. The calico rock is best developed in the southeastern part of T. 13 S., R. 2 W., and ranges from about 10 to 25 feet in thickness. The calico rock is a dull-appearing, rela- tively light weight, more-or-less porous rock, probably best described as an impure variety of chert. Data on some of the physical prop- erties of three different kinds of calico rock are given in table 9. A sample of green calico rock fused to a white glass between cones 27 and 28 (about 1605 to 1615° C; 2921 to 2939°F.). A brown and purple banded sample fused to a "marble cake" patterned gray and white glass at the same approximate temperature. A chemical analysis of a sample, including both "soft" and "hard" calico rock, taken from a 10-foot outcrop along a creek in the W^ NWi/i NE14 sec. 26, T. 13 S., R. 2 W., follows: Si02 . . . .88.69 AI2O3 . . . 6.08 FesOa . . . 1.50 MgO 44 CaO 22 NasO . . . 1.20 K2O . . . . 1.20 MnO 01 TiOz 36 H.2O ...... .74 Loss on ignition 1.91 Total . .100.53 This analysis, though not exactly inter- pretable in mineralogical terms, suggests a free silica content of about 75-80 percent with the balance largely clay mineral material. Calico rock has been used in a small way locally for rubble masonry and, when crushed, for road stone. Its usefulness for refractories and other purposes has not been investigated. The siliceous shale of the Springville for- mation is well exposed in an abandoned quarry along Route 146 west of Jonesboro, in the NE 1/4 NW 14 sec. 23, T. 12 S., R. 2 W. The shale is hard, brittle, and greenish gray. It is blocky where fresh but weathers to pieces Yz- to ^-inch or less thick or to thin flakes upon severe weathering. The exposed thickness is 35 feet. The lower one-third to one-half of the exposure is somewhat cal- careous, A sample from 13 feet of the lower shale showed loss on ignition of 3.93 per- cent, indicating a possible carbonate con- tent of about 7 percent. The extent of the Table 9. — Physical Properties of Calico Rock Character of calico rock "Soft," mottled, and blotched "Hard," mottled, and blotched 'Hard, green Apparent specific gravity Weight per cubic foot, pounds 48-hour water absorption, percent by weight Porosity, percent by volume (calculated from water absorption) . 1.7 106 25 43 1.9 119 8 15 2.2 138 3 7 CHERT GRAVEL 33 siliceous shale has not been accurately deter- mined, but there are probably other de- posits in the vicinity of the quarry men- tioned above. The quarry is locally reported to have been the source of shale once used for mak- ing roofing chips. The shale is said to have had poor weather resistance in the roofing and for this reason quarrying was discon- tinued. The flaky character of the highly weathered shale at the quarry would seem to bear out this report. No other use is known to have been made of the shale. Origin The origin of the calico rock and the sili- ceous shale of the Springville formation is not fully known. There is also a ''calico shale," a light colored shale or clay with mottlings like the calico rock, found in places interbedded with calico rock or as deposits devoid of calico rock, and having an observed thickness of up to 7 feet plus. Possibly the calico rock and calico shale in their present state are products of the leaching of a calcareous, argillaceous chert and a siliceous, calcareous shale respectively. CHERT GRAVEL "Lafayette" Gravel and Sand The "Lafayette" formation of southern Illinois is variable in character though it consistently includes some gravel. Many outcrops are composed solely of gravel with more or less interbedded sand, others of gravel overlain by sand. Locally, as in parts of Alexander and Massac counties, the for- mation consists of two lithologic units. The lower unit is composed principally of red, brown, yellow, or gray sand with a few layers of rounded, brown, chert gravel. In the sand in places occurs pink, yellow, or white clay in balls or irregular pancake- shaped masses up to 10 feet long and 18 inches thick. The upper unit of the forma- tion is characteristically a rounded, brown, chert gravel with which is interbedded vari- able thicknesses of red, brown, or yellow sand. In parts of Alexander County, in the gen- eral vicinity of Fayville, there are scattered exposures of a coarse, red, clayey sand con- taining a few small pebbles, which is be- lieved to be a part of the "Lafayette" for- mation. The sand contains a few thin layers of chert gravel and locally includes masses of yellow clay. The maximum observed thickness was 17 feet. The coarse sand is underlain by as much as 16 feet of fine-grained sand. The relation of these deposits to the gravel of the "Lafayette" formation is not clear. Particle size data on the coarse sand follow r^"^ Sieve Percent retained (mesh) A B 4 6 1.1 10 2.3 3.1 20 32.3 19.8 28 31.3 29.9 35 15.7 19.9 48 4.3 8.7 65 2.6 4.1 100 1.5 2.2 150 0.2 0.3 Pan 0.3 0.2 Clay 9.6 10.5 A^16 feet of sand, NE H NW ^ sec. 27, T. IS S., R. 3 W. B — 17 feet of sand, near center E J/2, sec. 34, T. 15 S., R. 3 W. When freed of clay and surface stain, the coarse sand (except the finer sieve sizes) consists of rounded, polished, chert grains, mostly white but with a few dark gray or black grains, and of polished quartz grains ranging from angular grains with rounded edges and corners to rounded grains. They are mostly colorless, white, or cloudy ; a few are pink. Most of the pebbles in the "Lafayette" gravel are less than 3 inches in diameter, but scattered cobbles and boulders of chert as much as two feet long have been observed locally. The pebbles are characteristically rounded, including fragments up to ten inches in diameter, but boulders larger than this are usually angular or subangular. Sieve tests of several samples of "La- fayette" gravel are given in table 10. Table 1 1 gives the results of tests of four samples, 35 Lamar, T. E., Unexploited or little known minerals in Illinois, in Contributions of the Fifth Annual Mineral Industries Conference: Illinois Geo!. Survey Circ. 23, p. 217, 1938. 34 SILICEOUS MATERIALS Table 10. — Sieve Tests of "Lafayette Gravel Percent retaine d Inches or Mesh 1 2 3 4 5 2" 5.6 4.5 1 20.9 16.6 23.4 33.5 7.8 H 14.9 18.3 11.8 18.9 10.1 H 12.3 16.6 13.3 18.9 10.3 li 16.4 11.5 11.0 14.6 22.0 6 mesh 3.7 3.4 3.0 0.1 5.3 10 6.9 3.0 4.1 0.2 14.1 20 2 2 3.2 5.1 0.1 3.7 28 1.8 2.0 5.1 0.1 3.2 35 1.8 3.0 5.1 0.2 4.8 48 2.1 7.1 4.0 1.9 7.1 65 2.4 8.8 1.9 2.1 2.5 100 1.5 3.1 1.2 1.4 1.1 150 0.6 0.3 0.3 0.6 0.1 200 0.4 Tr Tr 0.4 Tr 270 0.1 Tr Tr 0.3 Tr Pan 1.5 0.3 0.2 0.3 0.4 Clay 4.9b 2.8- 6.5'^ 6.3- 7.5b Total 100.0 100.0 100.5 99.9 100.0 a Minus .004 mm. ^ Minus .002 mm. 1. 10 ft. gravel, center NE 3^ sec. 34, T. 15 S., R. 3 W., at head of Miami Hollow near Fayville, Alexander County. 2. 12 ft. gravel, NW M NW }4 NW H sec. 18, T. 16 S., R. 1 W., near Unity, Pulaski County. 3. 11 ft. gravel, S J^ NE 3^ SW i^ sec. 15, T. 16 S., R. 1 W., near Mounds, Pulaski County. 4. 8 ft. gravel, SW ^ SW 3^ SW ^ sec. 1, T. 15 S., R. 4 E., near Round Knob, Massac County. 5. 23 ft. gravel, SW }4 NW M SW 34 sec. 28, T. 14 S., R. 5 E., near Renshaw, Massac County. all taken from a gravel pit at the head of Miami Hollow, said to have been w^orked at one time by the Chicago and Eastern Illinois Railroad, presumably for ballast. The exposure at this pit is one of the thick- est known in southern Illinois and is as follows : Loess overburden 4. Gravel with interbedded brown and red sand layers (Sample 9) . 3. Sand, red, coarse grained (Sample 8) 2. Gravel, brown and red sand (Sample 7) ...._....,. . 1. Gravel and sand interbedded, in one or two foot beds (Sample 6) Covered Thickness Jeet 14 2-5 11 38 Sample 1 in table 10 was taken at a road gravel pit in the same general locality as the above section. The correlation between the stratum from which Sample 1 was obtained and the strata in the above section is not clear, but it is believed Sample 1 was obtained from beds equivalent to some part of Unit 1 above. The chemical composition of the chert pebbles comprising the "Lafayette" gravel is indicated by an analysis of the plus ^-inch chert pebbles from a representative portion of Sample 1 (table 10), as follows: Percent SiOz 96.32 AI2O3 .87 Fe203 1.76 MgO .03 CaO .10 NazO .06 K2O .09 +H2O .73 — H2O .13 CO2 . ., .16 Loss on ignition .81 Total 100.04 CHERT GRAVEL 35 Table 11. -Sieve Tests on "Lafayette" Gravel and Sand from Pit at Head of Miami Hollow, Center NE 34 Sec, 34, T. 15 S., R, 3 W., Alexander County Thick- Percent retained Sample No. ness repre- sented feet 2" 1 " 3^" Va" 10- mesh 20- mesh 40- mesh 60- mesh 100- mesh Pan Clay- Total 9 . . . . 14 11.1 11.1 21.7 14.0 7.4 3.7 5.1 15.7 5.8 0.5 3.9 100.0 8 . . . . 2 5.4 2.0 0.7 28.9 51.8 2.0 1.5 7.7 100.0 7 . . . . 11 3.9 23.6 23.6 11.3 8.4 5.3 11.3 7.7 1.0 0.5 3.4 100.0 6 . . . . 38 8.4 20.4 16.5 10.8 5.1 6 15.3 11.4 1.4 4.4 99.7 Average.^ 65 3.1 11.3 20.6 14.7 9.4 4.7 7.4 15.2 8.1 1.1 4.2 99.8 Estimated average chert chert chert chert chert chert chert chert chert qtz and quartz" 100 100 90 75 60 35 5 3 100 content- qtzlO qtz25 qtz 40 qtz 65 qtz 95 qtz 97 percent * Minus .004 mm. ^ Weighted according to thickness of beds. « Occurring as pebbles or sand grains; includes a few quartzite pebbles. The pebbles of the "Lafayette" gravel in southern Illinois are largely chert, except for well rounded vein quartz pebbles which are relatively common, and fewer quartzite pebbles or cobbles.^^ The quartzites are white, purple, brown, or dark gray. The pink, purple, and dark gray varieties are un- like any bedrock native to southern Illinois. The relative proportions of quartz grains, quartz pebbles, and quartzite pebbles to chert in the material in the Miami Hollow pit are given in table 11. The chert pebbles are dominantly brown or buff, but their color is primarily a sur- face phenomenon and usually penetrates to a depth of only i/^ to |4 inch. The insides of the pebbles are commonly gray, cream, or light buff. Black pebbles are relatively common in some deposits, and their color is likewise largely surficial. The surface of most of the chert pebbles is polished, al- though the degree of polish varies consider- ably. Some of the deposits of ''Lafayette" gravel are cemented into conglomerate, chiefly by iron oxide. The conglomerate seems to be most common in Massac County. A pit, now water-filled, formerly operated for gravel north of Metropolis, was re- ' More detailed data regarding the kinds, shape, and degree of polish of the pebbles and other special characteristics of the "Lafayette" gravel are given in: Lamar, J. E., and Reynolds, R. R., Notes on the Illinois "Lafayette" gravel: Trans. 111. Acad. Sci., vol. 44, pp. 95-108, 1951. ported to have 20 to 40 feet of cemented gravel, but whether the entire deposit was cemented or only layers therein is not known. The thickness of 40 feet of gravel reported in the Metropolis gravel pit is the maximum known in Massac or Pulaski counties. An exposure of 65 feet of "Lafayette" in Alex- ander County has been mentioned. Most of the exposures of gravel are less than 25 feet thick and many are between 15 and 20 feet thick. Good exposures of the "Lafayette" gravel are to be seen at a number of places. The samples whose mechanical analyses are given in tables 10 and 11 include some of the best outcrops. Another good exposure show- ing 25 feet of gravel occurs in the SE corner sec. 20, T. 15 S., R. 6 E., bVi miles north- east of Brookport. The "Lafayette" gravel of southern Illi- nois is found chiefly in sizable deposits in Massac, Pulaski, and Alexander counties, and the southern third of Pope County. It is not known to occur in relatively large or thick deposits in southern Illinois north of these counties, except possibly in Shaw- neetown Hills, northwest of Shawneetown in Gallatin County, where deposits described as Tertiary, and probably part of the "La- fayette" formation, are reported. ^^ Small 3^ Butts, Charles, Geology and mineral resources of the Equality-Shawneetown area: Illinois Geol. Survey Bull. 47, p. 52, 1925. 36 SILICEOUS MATERIALS deposits occur in the western and southern parts of Union County and in southern Johnson County. The gravel is believed to be reasonably extensive, though of varying thickness, un- der some of the upland areas of Pulaski and Massac counties and parts of Alexander and southern Pope counties. The elevation of the top and base of different outcrops varies, even within short distances. The highest known deposits occur in southeastern Mas- sac and southern Pope counties where the tops of some deposits are at about 565 feet above sea level. The elevations of the top of the deposit south of Renshaw are about 540 feet and of the deposit in Miami Hol- low near Fayville about 485 feet. Other de- posits and the approximate elevations of their tops are: Mounds, 350; Villa Ridge, 430; one mile northeast of Olmsted, 480; Round Knob, 440; Shawneetown Hills, 500; and the gravel pit at Metropolis, 340 feet. The Renshaw and Fayville deposits and possibly the Brookport deposit are believed to lie on hills which have a core of consoli- dated rock, though Cretaceous sediments probably intervene between the gravel and the bedrock. The other deposits mentioned above apparently occur on ridges or hills composed principally of Cretaceous or Eocene sand, silt, and clay. Large masses of slumped gravel are common because the subjacent Cretaceous sediments are un- stable when wet. The "Lafayette" forma- tion is regarded as water laid, probably by streams. Many of the gravel deposits have an overburden of dark red or brown, sticky, silty clay or clayey silt, usually sandy and often pebbly, a few inches to about 6 feet thick, which in turn is overlain by brown, silty clay (loess) having a maximum thick- ness of possibly 40 feet near the major streams, but the thickness varies considerably from one deposit to another. An average thickness of overburden is 10 to 20 feet on the deposits which have been worked as a source of gravel. The Western Indiana Gravel Company at one time operated extensively in a de- posit of "Lafayette" graveP^ located a few miles north of Metropolis. The deposit was worked by dredge from a water-filled pit. Railroad ballast, road gravel, concrete ag- gregate, engine sand, torpedo sand, and filter sand were produced. The Chicago and Eastern Illinois Railroad is said to have obtained gravel, presumably for ballast, from the Miami Hollow deposit near Fay- ville in Alexander County. Mostly the gravel has been worked recently in roadside pits as a source of road gravel, for which purpose it serves well and has been widely used in southern Illinois. Elco Gravel In the SW 14 NE 14 NE H sec. 7, T. 14 S., R. 1 W., about II/4 miles northeast of Elco, is a deposit of chert gravel, known locally by the trade name of "Elco" gravel (fig. 16), which differs in some respects from the other chert gravels of southern Illinois. The structural details of the Elco gravel deposit are complex. It consists of angular chert gravel, including in a few places slabs of chert up to 5 feet long, with which are interbedded layers of rounded chert gravel. Both materials are generally light gray or white, though locally they are discolored by iron stains. The pebbles usu- ally show no polish. Most of the gravel is composed of pebbles less than 6 inches in diameter. The deposit shows a crude strati- fication which is roughly horizontal, but there is notable local distortion of the beds. In some places, the deposit is cut by frac- tures or small faults. The thickness of the gravel ranges up to about 50 feet. In some places the gravel rests on sand composed of quartz and chert grains, which has a visible thickness of about 10 feet. The base of the bed was covered. In other places the gravel rests on a cherty clay. Knobs of porous chert bedrock, apparently originally calcareous chert, occur at both ends of the deposit and extend upward into the gravel to heights of 20 feet. The Elco gravel has been used chiefly as road gravel. A sieve analysis of a sam- 3s Lamar, J. E., Western Indiana Gravel Company dredges flint gravel at Metropolis, Illinois: Rock Products, vol. 32, no. 7, pp. 71-73, 1929. CHERT GRAVEL 37 pie of 20 feet of pit-run gravel from the west end of the deposit follows : Cumulative Sieve Percent percent retained retained Inches 2 . . 10.2 10.2 1 , . 25.6 35.8 % 6.4 42.2 y2 , . 7.7 49.9 M 14.1 64.0 Mesh 6 , . 7.3 71.3 10 , . 8.9 80.2 20 , . 5.7 85.9 28 . . 1.9 87.8 35 . . 1.9 89.7 48 . . 4.4 94.1 65 . . 2.8 96.9 100 , . 1.1 98.0 150 , . 0.3 98.3 Pan , . 0.2 98.5 Clay- . . 1.5 100 Total . . . . . 100.0 « Minus .004 mm. The origin of the gravel is obscure. Some of it is probably water-laid ; other parts of the deposit are somewhat similar to the cherty residuum developed by leaching and weathering from some of the Devonian chert and cherty limestone formations. The fact that the gravel rests irregularly on cherty clay, coarse sand, or chert bedrock further complicates the problem of its origin. The extent of the Elco gravel is problem- atical. The most logical area for prospect- ing is in the vicinity of the known deposit, especially along the Gulf, Mobile and Ohio Railroad for about a mile north and a mile south of the known deposit. Chances for finding gravel are believed to be best on the east side of the railroad in the frontal slopes of the ridges or topographic noses. River Gravel and Sand Gravel and sand are being or have been dredged from the Ohio River near Cairo, Metropolis, Joppa, Golconda, and Eliza- bethtown, and probably at other places. Gravel is not known to have been dredged from the Mississippi River in extreme south- ern Illinois. The gravel obtained from the Ohio River consists principally of light buff to brown, rounded chert pebbles similar to those of the "Lafayette" formation found in the uplands of southern Illinois. A few light gray or dark gray chert pebbles are present. A sample of commercially produced river gravel examined in detail showed also quart- zite pebbles in moderate abundance, and a few sandstone and igneous pebbles. The lat- FiG. 16. — Elco gravel deposit near Elco. Two masses of rotted chert bedrock are visible at the extreme left of the pit with an inclined zone of gravel between them. Just above the man's head, extending to both left and right, is a distorted layer of chert slabs. The attitude of this layer is in contrast to the horizontal bedding shown in the higher parts of the deposit. 38 SILICEOUS MATERIALS ter were restricted to the gravel finer than ^-inch, were well rounded, and both light and dark in color. The river sand (minus 4-mesh) contains in its coarser sizes abundant brown chert particles such as occur in the "Lafayette" formation. About half the material be- tween 4 and 10 mesh consists of chert grains. The percentage of chert decreases with the fineness of the sand until below 20 mesh it is minor. The grains of the river sand range from angular to rounded. The chert particles are usually rounded and many of them have highly polished surfaces. Some of the quartz particles also show a high polish and, although a few are well rounded, many are angular to subrounded. The Halliday Sand Company of Cairo dredges sand and gravel from bars in the Ohio River. Products are gravel, con- crete aggregate and masonry, plastering, engine, asphalt and concrete sand. The Metropolis Sand and Gravel Company, among others, has produced sand and gravel from the Ohio River but did not report production in 1952. The Federal Materials Company of Paducah has at Metropolis a materials yard which handles river gravel and sand. Creek Gravel A considerable body of chert gravel occurs in many of the creeks which flow through the area underlain by the cherty formations of the Devonian or Lower Mississippian rocks of southern Illinois, especially in Alex- ander and southwestern Union counties. Usually the gravel is visible only in the ac- tive channels of the streams, with about 10 feet the maximum thickness commonly ex- posed. Some of the gravel is undoubtedly coming from existing outcrops but much of it may be the result of reworking an older gravel fill in the valleys. Many of the valley flats may contain an appreciable filling of gravel which is covered by a deposit of alluvial silt, probably laid down during gla- cial times, to varying depths depending on the proximity of the particular portion of the stream valley in question to the Missis- sippi River and on the gradient of the stream valley. Such silt is commonly 3 to 8 feet thick. Owl Hollow is a valley three miles long extending northward from the flat of the Mississippi River in the NE }4- sec. 3, T. 12 S., R. 3 W., Union County, about a mile east of Wolf Lake. Some years ago it was rumored that this valley was to be used as a source of gravel for railroad ballast. It is roughly representative of other valleys in the chert areas of southern Illinois, and a discussion of the chert gravel visible in it gives an idea of what may be expected in other similar valleys. The active channel of the stream in Owl Hollow is covered with chert gravel which, with local exceptions, becomes coarser upstream from the point where the creek valley joins the flat of the Mississippi River. The coarsest material generally pres- ent is about 10 inches in diameter. Most of the gravel fragments are cut by incipient joints which cause them to break rather readily when struck with a hammer. The visible thickness of the gravel ranges from one to 7}A feet and averages 2 to 3 feet, but the amount of unexposed gravel may be considerably more than this. The gravel fill in Owl Hollow was stud- ied for a distance of about 5500 feet above its mouth. In this distance the active chan- nel of the creek ranges from 15 to 80 feet wide and averages about 30 feet. It is estimated on the basis of rough measure- ments of channel width, channel length, and thickness of exposed gravel that in the active channel there are about 20,000 cubic yards of gravel. A considerably larger amount may be available beneath the ex- posed gravel and under other parts of the stream's valley bottom. In comparison to Owl Hollow, which is a relatively small valley, some of the larger streams may have a much greater quantity of chert gravel in their channels and beneath their flood plains. The creek gravel has been used to a limited extent for road making. ■liwiiaiiisi lil^iiilPlIi Wmmmsmm§ ^^iiilifiliiPii mmmvm m Fig. 17. — Disc-shaped or pillow-like masses of chert interbecUled with silica, exposed in quarry in sec. 32, T. 11 S., R. 8 E. Photo by R. M. Grogan. Osage Formation in Hardin and Saline Counties Another source of cherty siliceous mate- rials is the Osage formation of Hardin and Saline counties. Where unweathered this formation is believed to consist of a gray or dark gray siliceous limestone yvhich con- tains lenses or masses of chert, but weather- ing produces a residual chert gravel. On the ridges the thickness of the gravel is prob- ably less than 10 feet but slumping may con- centrate greater thicknesses on the low^er slopes of the ridges. The gravel is usually angular and contains numerous elongate fragments roughly triangular in cross sec- tion. With it commonly is mixed residual clay and soil. Gravel of this type consti- tutes a source of road metal reported to have been used in a limited way. In Hardin County the Osage strata un- derlie an oval -shaped tract about 3]/2 miles long and two miles wide, extending roughly southeast of Hicks in T. 11 S., Rs. 7 and 8 E.^^ In the SW 1/4 NW V4 sec. 32, T. '* Weller, Stuart, The geology of Hardin County: Illinois Geol. Survey Bull. 41, pp. 92-93, 306-307, 1920. 11 S., R. 8 E., a quarry exposes a maximum of about 45 feet of highly weathered Osage, consisting of pillow-like masses of white or buff chert in a matrix of relatively coarse light gray or cream silica (fig. 17). It is estimated that about 90 percent of the de- posit is chert and 10 percent silica. The rock has been quarried and crushed for road metal.^"^ In Saline County an outcrop of Osage rocks occurs in an isolated hill about H mile long, three miles southwest of Equality in the NE 14 NW 14 sec. 36, T. 9 S., R. 7 E.*^ Here a quarry has produced crushed stone from a deposit of steeply dipping, cherty, gray, siliceous Osage limestone. Originally the operation may have begun as a gravel pit in residual chert gravel, of which there is about 10 feet exposed as overburden on the quarry. '*- *" Grogan, R. M., personal communication, 1943. " Butts, Charles, Geology and mineral resources of the Equality-Shawneetown area: Illinois Geol. Survey Bull. 47, pp. 70-71, 1925. •^ Grogan, R. M., personal communication. Illinois State Geological Survey Report of Investigations No. 166 1953