Abstract

A descriptive classification of detrital, siliciclastic sandstones is proposed; origin and genesis are not specifically implied. It is similar in style and format to those used in igneous and metamorphic petrology. The nomenclature, which is binomial, is based on (1) overall texture and (2) composition of the volumetrically predominant sand-size clast, which is specifically named. General to specific names may be used depending on precision and desirability. There are three textural categories: (1) Sandstone, a sedimentary rock in which the volume of sand-size clasts is greater than the volume of either clasts smaller than or clasts larger than sand-size; (2) Arenite, a subdivision of sandstone, which contains less than 15 percent matrix (interstitial detritus, recrystallized or not, smaller than 30 μ); and (3) wacke, a subdivision of sandstone, which contains 15 percent or more matrix. The compositional part of the name is based on the sand composition as plotted on the commonly used Quartz-Feldspar-Lithic fragment triangular diagram. Each end-member of the triangle is an identifiable detrital component; the appellation specifically names the predominant, recognizable, sand-size detrital clast; and the triangle is divided symmetrically so there is no predisposition favoring one end-member over another. Three levels of compositional specificity are provided: (1) A congruent, 3-fold division, quartzose, feldspathic and lithic, depending on the predominant sand-size clasts; (2) A 6-fold division which separates the above three along the 75 percent line, giving supraquartzose (75%), subquartzose (< 75%), suprafeldspathic, subfeldspathic, supralithic, and sublithic; and (3) A 15-fold division accommodating nearly pure end-members (≥ 90%) as quartz-rich, feldspar-rich, and lithic fragment-rich and identifying the second-most predominant component, such as lithiquartzose (Q ≥ 75%> L > F), lithisubquartzose (75% > Q > L > F), etc. Thus a sandstone containing 20% matrix, 50% feldspar, 20% quartz, and 10% lithic fragments would be a “quartzosubfeldspathic wacke,” whereas one containing 10% matrix, 85% lithic fragments, and 5% feldspar would be a “lithic fragment-rich arenite.” The proposed classification overcomes problems of previous ones in that it (1) allows for degrees of specificity, (2) identifies both the majority components and texture and (3) avoids genetic, ambiguous, or widely redefined terms such as arkose and greywacke.

Sandstone petrography entered the realm of quantitative analysis and classification after Sorby perfected thin section petrography in the 1850’s (Dott, 1964). No longer need a sandstone be described in qualitative terms. Rather it could be described and, consequently, classified quantitatively in terms of detrital composition, measured textural parameters, and type and degree of lithification.

With the new wealth of quantitative petrographic data, sandstone classification became an ever more pressing problem. Which criteria would be used? What constituted similar or different? Were the pre-petrographic microscope systems still viable? Must old names be discarded and replaced by new? Need old names be redefined? What should a name denote or imply? What was the purpose of classification?

Indeed, what is the purpose of sandstone classification? Geologists have grouped themselves, more or less, into two separate schools. There are those whose intent is to name collectively rocks based on similarity of selected physical and/or chemical attributes, thus their classification systems are essentially, if not purely, descriptive. On the other hand, there are those whose intent is to determine the origin and genesis of the rock via the classification itself, thus their classification systems are essentially genetic. Genesis may denote one or more of the following: Geologic, topographic, and/or climatic provenance; the transportation system and its characteristics; environment of deposition; diagenesis; and post-diagenetic processes and products. Although members of the descriptive (presumably more basic) school, we remain frustrated by the many available systems. Some are too general; some are too specific; many emphasize the detrital quartz component and de-emphasize the lithic-fragment and feldspar components; some supposedly descriptive nomenclatures retain sandstone names which themselves have strong genetic implications. In the case of some names, for example, greywacke, there are many diverse definitions available. Conversely, for a sandstone of a given composition and texture, there may be many possible appellations.

Our purpose here is to present a descriptive classification system for detrital sandstones which overcomes some of the problems of previous systems. Our system is predominantly descriptive; it is based on both sandstone texture and detrital composition; it gives equal emphasis to the major sand-size detrital components, thus discriminating equally the end-members of the popularly used triangular diagram. The system is equally applicable to sandstones in modern or ancient mobile belts, cratons, and plate boundaries, be they convergent, divergent, or strike-slip. The nomenclature specifically identifies the volumetrically predominant sand-size detrital component, and a three-fold hierarchy of names, from general to specific, is used. In summary, our proposed system is based on the premise that a name is simply a label that allows two or more persons to communicate about the same object, quantity, or concept; a name should convey a maximum of information and a minimum of ambiguity.