Page 132 - Petrology of Sedimentary Rocks

the arkose. Metamorphic rock fragments, micas, and metaquartzite fragments were
linked together at the M-pole, indicative of metamorphic source, the end-member rock
type being the “graywacke” (this is essentially the composition of the graywacke of
Naumann, I850 text). Later versions added a fourth pole for subaerially-eroded, much
older carbonate rock fragments, defining a rock known as a “calclithite” with over 50
percent CRF’s (I 959 AAPG). Clays were eliminated from the compositional name (a
major departure from the Krynine system), and placed in the textural part of the name.

Several difficulties arose with this scheme. The division between “ordinary
quartz” (Q-pole) and “stretched metaquartzite” (M-pole) is subjective, difficult enough
to determine with the microscope and certainly impossible in the field. A dark-colored
sand grain might be a chert, a basalt fragment, or a piece of slate, thus assignable to Q-
F-, or M-poles; the distinction is rarely attainable in the field and is sometimes
hifficult even with the microscope. Chert itself shows a complete gradation into shale
and siltstone, fine-grained metaquartzite, and silicified volcanic rock--and it is
difficult to define these reproducibly. The word “graywacke” is encumbered by so many
radically differing definitions that it has been rendered almost useless.

Therefore, after much soul-searching, the following revised and hopefully more
workable sandstone classification has been used by the writer since spring 1966. It is
hoped that those of us seriously interested in sandstone classification can, some decade,
all get together and use some compromise system, so this classification is tentative and
subject to change should such a miracle come to pass. I have been strongly influenced
in this change of heart by prolonged discussions with Earle F. McBride, Keith A. W.
Crook, and Harvey Blatt; many of their ideas have been incorporated into the new
revised version. This new system has been now published formally [Folk Andrews and
Lewis (I 970), New Zd. J. Geol. & Geoph.] .

The main difference between this and my previous classification is the lumping of
all rock fragments (except plutonic ones, granite and gneiss) into the third or “rock-
fragment” pole of the triangle, and the switching of chert--which is in reality a rock
fragment--from the Q-pole to the RF-pole. Van Andel (I 958 AAPG) also placed chert
with rock fragments. Unfortunately, by putting all rock fragments regardless of genesis
in one pole, the main triangle loses almost all its source-area significance; but splitting
the rock fragments on the subordinate RF triangle restores much of this loss. Lumping
of all rock fragments together to produce a “rock-fragment sandstone”--however it
might be named--was essentially the approach used by Shvetsov in the early 1930’s, and
he termed the rock a “graywacke”; this system has been followed by later Russian
petrographer (Shutov, 1965). Later workers who used the same idea include McElroy
(I 954 Austral, J. Sci., petromictic or lithic sandstone): Gilbert (I 954 book); Pettijohn
(1954 JG; I957 book); Van Andel (1958 AAPG); Crook (1960 A. J. Sci.); and MC Bride
(I 963 J. S. P.).

Splitting of the rock fragments by means of a subordinate VRF/MRF/SRF triangle
is my own idea, (though Kossovskaya used a RF triangle with different poles in 1962) as
are the names “Calclithi te” and “phyllarenite.” The percentage lines of the main
triangle are almost the same as those used by Crook (I 960), and the name “Litharenite”
was coined by McBride (1963) who shortened it from Gilbert’s (1954) term “lithic
areni te.” So much for the record of how this scheme developed--like so much else, a
mass of borrowing.

To determine a rock clan under the 1966 system, follow the following steps.

126