COMMENTS & REPLY A More Informative Way to Name Plutonic Rocks— Comment by Frost et al. B. Ronald Frost, Carol D. Frost, Dept. of Geology and Geophysics, University of Wyoming, Laramie, Wyoming 82071, USA; J. Lawford Anderson, Dept. of Earth and Environment, Boston University, Boston, Massachusetts 02215, USA; Calvin G. Barnes, Dept. of Geosciences, Texas Tech University, Lubbock, Texas 79409, USA; and Marjorie Wilson, School of Earth & Environment, Leeds University, Leeds LS2 9JT, UK In a recent paper, Glazner et al. (GSA ≥10% quartz (the most common group) a more precise description of the rock. This Today, Feb. 2019) proposed a major change are described by only 23 names, many of suggestion indicates a confusion of preci- in the terminology of plutonic rocks, which share the same root name. These sion and accuracy. Modes are difficult to whereby a simplified rock name is prefixed names need not be memorized because determine in the field where, as Glazner et with the mode. In this classification, a they are present in the various IUGS dia- al. (2019) observe, the distinction between granite might be named 20,20,50 granite. grams for rock names, a diagram that is alkali feldspar and plagioclase can be sub- Glazner et al. (2019) proposed this classifi- easily pasted into field notebooks. tle. Field estimation of modes is unlikely to cation system for three reasons. First, they The argument of Glazner et al. (2019) be better than ±10%. With this precision, a maintain that rock terminology is too com- that rock names are determined by rock classified as a 25,25,40 granite would plex; they note that at least 157 igneous rock “arbitrary” boundaries is not compelling. occupy a large field in the QAP diagram names exist. The second is that the bound- These boundaries are not arbitrary: (Fig. 1). For this reason, Streckeisen (1976) aries in the International Union of Geo- The IUGS commission spent many years suggested a preliminary classification in logical Sciences (IUGS) classification are developing a system that conformed, as which granitic rocks may be named with arbitrary and hence are confusing when much as possible, with existing classifica- the termination “-oid,” as in granitoid. applied to plutonic rock units that show a tion systems. Furthermore, the rock terms Point-counting a minimum of 1000 points range in composition. Third, the IUGS sys- have meaning in the sense that geologists on stained slabs or thin sections produces a tem of classification is qualitative, and the know what to expect of a rock described more accurate determination of quartz, quantitative data from which the classifica- as tonalite instead of granite. Glazner et plagioclase, and alkali feldspar abundances tion is derived are discarded once the name al. (2019) support their arguments with that are used to identify the appropriate is determined. To solve these problems the observation that two of the plutons in IUGS rock name. However, as Glazner et Glazner et al. (2019) propose that the petro- the Sierra Nevada batholith, the Cathedral al. (2019) observe, only 5% of the analyses logic community discard the IUGS classifi- Peak Granodiorite and the El Capitan archived in the NAVDAT database have cation system and substitute a system with a Granite, contain rocks that look the same associated modal data. This means that limited number of rock names that are pre- (their Fig. 1). Thus they conclude that the their quantitative classification system, in fixed by the modal abundance of major names “granodiorite” and “granite” are in addition to being of limited value in the phases (such as quartz, alkali feldspar, and error. However, the error is not in the field, is not likely to be widely applied. plagioclase [QAP] in felsic rocks). They names of the individual rocks, it is in the A further problem with modal classifi- maintain that this is a simpler classification assumption that the Cathedral Peak cation is that even when mineral propor- and that it lends itself to a more quantitative Granodiorite contains only granodiorite tions are accurately determined there classification scheme. whereas the El Capitan Granite contains remains an inherent, irreducible uncer- We take exception to Glazner et al.’s only granite. Plutons are rarely homo- tainty. First, the abundance of feldspars in (2019) proposal and instead recommend geneous over distances greater than a few a granitoid is dependent on the cooling that geologists continue to use the IUGS tens of meters: they contain rocks with a history of the rock. A rock that cooled classification system for naming plutonic range of compositions. Whereas each relatively swiftly and contains sodium- rocks. Their first justification, that there rock named by the IUGS classification is bearing orthoclase will have a different are too many obscure terms in igneous valid, the assumption that the pluton ratio of alkali feldspar to plagioclase than petrology, was a problem recognized by name (i.e., Cathedral Peak Granodiorite) one that cooled slowly and contains the IUGS commission (Streckeisen, 1976; classifies all rocks within the pluton as sodium-poor microcline and plagioclase Le Maitre et al., 2002). Hence, the IUGS granodiorite, as implied by Glazner et al. with sodic rims. Furthermore, crystalliza- rock names replaced a plethora of obscure (2019), is simply false. tion of any muscovite or biotite in a rock terms. The IUGS classification scheme Glazner et al. (2019)’s third point is that will deplete the orthoclase component involves only 55 names for common plu- appending numbers that reflect modal from the feldspar matrix. Similarly, horn- tonic igneous rocks. Of these, rocks with abundances to a simple name will result in blende crystallization will deplete a GSA Today, v. 29, https://doi.org/10.1130/GSATG405C.1. Copyright 2019, The Geological Society of America. CC-BY-NC Q Q no known silexite granodiorite igneous granodiorite tonalite rocks tonalite 60 60 60 60 alkali alkali feldspar quartz monzodiorite quartz monzodiorite feldspar granite quartz monzogabbro quartz monzogabbro granite granite granite LG quartz quartz diorite quartz diorite alkali feldspar quartz alkali syenite quartz gabbro quartz gabbro feldspar quartz anorthosite syenite quartz anorthosite 10 35 65 90 35 20 20 20 F F F 20 monzodiorite alkali feldspar quartz quartz monzodiorite alkali feldspar quartz quartz monzogabbro C monzogabbro syenite syenite monzonite syenite syenite monzonite 5 5 diorite 5 5 syenite monzonite syenite monzonite diorite A P gabbro A P gabbro foid-bearing anorthosite anorthosite foid-bearing monzonite alkali syenite 10 10 foid-bearing diorite 10 50 foid 90 foid-bearing gabbro Figure 2. IUGS classification for quartz-bearing foid-bearing foid monzodiorite foid-bearing anorthosite felsic rocks showing the differentiation paths syenite monzosyenite foid followed by various granitic plutons Dashed monzogabbro lines—granitoids of Cordilleran batholiths; C— foid-bearing monzodiorite Caledonian batholiths; F—ferroan granites; foid syenite foid-bearing monzogabbro LG—peraluminous leucogranites; Q—quartz; foid diorite A—alkali feldspar; P—plagioclase. From Frost foid gabbro and Frost (2014). 60 60 Figure 1. IUGS classification for quartz-bearing foidolite felsic rocks comparing where a 25,25,40 granite would plot if the modal analyses were precise to quantify rock variability than modes. less than ±1% (red dot) compared to where it would plot if the modal analysis was precise to Together, the IUGS rock names and the ±10% (yellow circle). corresponding geochemical analyses pro- F vide insights into the most important, and frankly most exciting, questions in petrol- plagioclase component. Any statistical containing quartz diorite, tonalite, and ogy, including how igneous melts form, application of a modal classification sys- granodiorite is typical of continental arc how the magmas evolve, and how their tem would involve these uncertainties. batholiths, whereas a suite consisting of compositions reflect the tectonic environ- The IUGS classification bins samples monzonite, syenite, quartz syenite, and ments where these processes take place. into a relatively small number of rock granite is typical of ferroan granitoids Let’s leave the well-established IUGS names to give geologists a general sense from extensional environments (Fig. 2). classification system in place rather than of the rock composition. As Streckeisen We conclude that it is a mistake to to try to fix something that is not broken. (1976) puts it, “The system is merely of modify the IUGS classification system to descriptive character, as it serves to make it into a quantitative naming sys- REFERENCES CITED order the rocks that occur in nature tem. The IUGS classification system was Frost, B.R., and Frost, C.D., 2014, Essentials of according to their mineral content.” developed over nearly 20 years and Igneous and Metamorphic Petrology: Cambridge, Cambridge University Press, 314 p. Moreover, the system allows for modifi- involved 456 petrologists from 52 coun- Glazner, A.F., Bartley, J.M., and Coleman, D.S., cation of the IUGS rock name to make it tries (Le Maitre et al., 2002). The idea 2019, A more informative way to name more informative: a biotite hornblende that it should be overturned by a single plutonic rocks: GSA Today, v. 29, p. 4–10, granodiorite has more hornblende than paper is inadvisable. Glazner et al.’s https://doi.org/10.1130/GSATG384A.1. biotite. A geologist knows that magnetite (2019) proposal to use modes in a quanti- Le Maitre, R.W., Streckeisen, A., Zanettin, B., LeBas, M.J., Bonin, B., Bateman, editors, granite contains Fe3O4 as the main mafic tative way eliminates rock names that are 2002, Igneous Rocks: A Classification and component of the rock. The prefixes meaningful and well-established in the Glossary of Terms: Cambridge, Cambridge leuco- and mela- may be used to indicate geological literature and is burdened by University Press, https://doi.org/10.1017/ the abundance of mafic minerals in the the inherent imprecision of modal analy- CBO9780511535581.