Petrography is the study of rock under the microscope, the petrographic microscope. For a sedimentary petrologist this is one of your most important and most basic skills. To be effective you must have a solid background in the properties of minerals, especially those minerals that commonly occur in sedimentary rocks. In particular you must be familiar with the optical character of those minerals. Having at hand next to your microscope a standard mineral reference text like Nesse’s Introduction to Mineralogy is imperative. Bring such a book with you when you come to lab.
Microscopes get used by a number of students each semester. As a safety precaution and to prevent the possibility of passing around infections, like eye infections you should use glasses when viewing through the microscope. For those who do not normally wear glasses or those who wear contact lenses get yourself a set of plain safety glasses and use them.
When evaluating sandstone under the microscope there are a number of things that you should search for and observe: grain types, porosity, fossils, cements, matrix, evidence of diagenesis, grain boundary relationships just to name the basic ones. Many examples of these types of features are well illustrated in books like A Color Illustrated Guide to Constituents, Textures, Cements and Porosities of Sandstones and Associated Rocks by Peter A. Scholle or Sand and Sandstone by Pettijohn, Potter and Siever. It would be worth a student’s time to page through such books and study the various photomicrographs and illustrations. Also worth one’s while is to observe thin sections where a previous petrographer has already identified key features.
Exercise: Just to get you started with the study of sandstone and other siliciclastic rocks there are 13 samples, some with hand samples, that you should look over. Their features are characterized in Appendix A. Budget your time, expect to spend 5 minutes on each sample.
Petrographers use modal analysis as the primary means to observe a rock in thin section and they use the results of the modal analysis as the means to describe and classify sandstone.
Modal analysis is a survey that the petrographer makes to determine the percentages of the various minerals, grain types, pore types, etc. that make up the volume of a rock. It is assumed that the thin section gives a representative sampling of the bulk make-up of the rock; however this assumption may not always be valid. Small thin sections of rock with relative large components provide a poor representation or the real make-up. This can be overcome by using a larger format thin section or multiple thin sections of the same material. Rocks with strong preferred orientations of components, especially those related to depositional bedding are also troublesome. This is partially overcome by using thin sections that represent a cut normal to bedding and not one that was cut parallel to bedding.
There are two steps to a modal analysis (1) survey of what is there and (2) rapid count of what is there.
Step (1) Look at the thin section while it is held up to a window or light. Do you see big grains or small ones? Keep in mind the size of the grains. Next put it under the microscope and look at it under low magnification (2X to 5X objective) by slowly sweeping around the thin section in both plain polarized light (PPL) and cross polarized light (CPL). While doing so make a list of the various components observed. Try to make the list go from what you think is the most abundant to the least abundant of what makes up the sample. Set up a Point Count Form that has five columns. List in the left hand most column the various items observed generally with the component that you perceive being most abundant at the top in the first row and the others in the rows below. Always leave a few rows at the bottom for those items you "discover" while doing the modal analysis. You may also do this directly in EXCEL if you have your laptop with you.
You should expect to have things that you can not identify right off, the unknowns. For the "unknowns" you need to provide all the information that you can so that later on you might just be able to deduce what it was. This information you record in your notes.
Do not spend a whole lot of time on the reconnaissance. This is just a quick and dirty survey of what is there. What is important is getting a solid comprehensive working list of components. Plan ahead; ask yourself what are the objectives of the study? If all you are going to do is put a Folk clan name on the rock then your list could be as short as four components (quartz grains, feldspar grains, rock fragments, everything else). However, a modal analysis will require a lot of time and working with a much more comprehensive list will not add significantly to the required time. Use as much detail in establishing your component list as your skills allow. That way you hopefully will not have to repeat modal analysis on a thin section later after you discover something else and as your study takes a new turn.
Step (2) is the actual census of what is present. The objective here
is to make an unbiased counting of what is present. The simplest method is called
a random walk (Figure 1).
Place the center of the thin section anywhere under the objective
lens without looking down the barrel of the microscope. Then
look down the barrel and observe what is under the cross hairs at their
juncture point and identify it. You have to be honest
here and identify what is exactly under the cross hair juncture, do not cheat. Look at your listing of components and put a
tick mark in second column on the row that corresponds to the component that
you just identified. Next, without looking down the barrel of the
microscope, move the thin section a short distance in any direction.
Then look down the microscope and identify that component, record, move the
slide and repeat. The more points you examine the more accurate your analysis
becomes. Typically, a petrologist will do 300 to 600 identifications per thin
section. It is very important not to look down the barrel of the microscope
while moving the slide to a new point. Doing so first of all
adds a bias to your survey because you WILL have a tendency to move to those
things that you know or to move to the unusual items present. Secondly, doing
so is a real certain way to experience vertigo (motion sickness).

Figure 1. Thin section
with a map of the track of a random walk for a modal analysis.
If you encounter the situation where the cross hair juncture falls exactly on the boundary between two different components then make a mental rule and identify the component to the right, to the left, to the top, or to the bottom and apply the same rule to every incidence where the cross hair juncture overlies a boundary.
If you have the luxury of a mechanical stage you can adapt a more systematic gird pattern to your search (Figure 2). Again do not look down the barrel of the microscope when moving the slide. If when using a grid the grid lines may parallel a fabric feature of the rock. This will cause a little directional bias in your results. If you have a real decent mechanical stage it may allow you to set the spacing between stopping points. Select a spacing that is about twice the average component width.

Figure 2. Typical
search pattern for modal analysis using a mechanical stage where each dot
represents the location of a determination.
You can use a mechanical counter or
even a spreadsheet program in place of a sheet of paper and marking down tick
marks for each identification. Mechanical counters
usually keep a running tally of item counted and then ring a bell for every 100
items counted. A spreadsheet program can be set up to do the same.
This facilitates the modal analysis process by not having to stop and count up how far along you are in the census. However mechanical counters have a limited number of
channels for counting. A good thorough modal analysis my
have you counting 25 or more components. Spreadsheet programs can handle such
but require time consuming moving up and down to enter
a new number. This can be overcome by programming each
key on the keyboard to represent an item being counted.
Once you have achieved the desired number of observation for the thin section then sum up the tick marks for each component and record in the third column. Add up those numbers to get the total number of observations. In the fourth column report the percent for each component, the percent of the volume of the rock that is represents. If there is an item on your list that you saw in thin section but did not land on report it as being less than the percentage that it would have been it you landed on it just once (report as <0.3% for a 300 count analysis).
In the fifth column quantify your accuracy-using (Figure 3), which allows for determining the "ninety-five percent confidence limits" of your modal analysis. That is the results you get will be reproducible 95% of the time within the prescribed limits. The 'n' or the vertical scale is the total number of identifications or hits for that modal analysis (ie: 300) and the 'p' scale is the percentage of the volume of the rock represented by a particular component. The percentages within the diagram represent the 95% confidence intervals (numbers are posted above their respective lines). For example, if a modal analysis consisted of 300 determinations and a particular grain was hit on 25% of the time then the amount of the volume of the rock occupied by that grain type would be 25%±5% or the 95% confidence limits would be 20% to 30%. It is the ±5% that gets posted in the fifth column.
Figure 3. Ninety-five percent confidence limits for object proportions from modal analysis. The "n" denotes the number of total identifications and the "p" denotes the calculated percentage of the total for a particular object type.
The modal analysis must be done so that in the end you will be able to differentiate between all the various types of detrital grains, all the cements, matrix, authigenetic replacements, and the various pore types.
During your initial evaluation of a thin section you should note the various grain types which from the framework of the sample and you should note what occupies the space between these grains. Grain types include detrital mineral grains (quartz, feldspars, micas, heavy minerals etc. which were derived from pre-existing rocks) and rock fragments (bits and pieces of pre-existing rock such as limestone, granite, slate etc). But you need more detail than that!
More specifically for detrital quartz there is:
· monocrystalline quartz (single crystal lattice for entire grain),
· polycrystalline quartz (two or more crystals of quartz within the grain),
· chert (cryptocrystalline quartz forming the grain).
· You could go further here but don't. If you take sandstone petrology in the future you will!
Feldspars will be identified simply as monocrystalline or polycrystalline. If it is monocrystalline then call it …….:
· feldspars if you have no evidence to further refine your identification
· plagioclase if you can see the twinning or have stained the sample.
· orthoclase if you have stained the sample
· microcline if you can see the tartan twin pattern
Do your best to differentiate between each type of feldspar but remember sometimes you must just say it is feldspar. Expect to find detrital grains of the various micas (muscovite, biotite, and chlorite) as well as grains of glauconite, phosphorite, zircon, and any of the other possible heavy minerals.
You may also encounter as detrital grains that actually look like fragments of other rock types. These are called rock fragments. If you can identify the fragment as coming from a sedimentary rock then it is a sedimentary rock fragment (SRF). Which may further be identified as a:
· sandstone rock fragment,
· limestone rock fragment,
· shale rock fragment,
· coal fragment, etc.
If the rock fragments were derived from a volcanic source then they are volcanic rock fragments (VRF); which would include basalt rock fragments. If the source were metamorphic then you would apply the term metamorphic rock fragment (MRF). These could be further subdivided into slate rock fragment, schist rock fragments etc.. Fragments derived from phaneritic igneous rock sources and some of the coarser metamorphic rock types like gneiss generally are identified as feldspar grains unless you are working with conglomerates then you could have igneous rock fragments (IRF) or more specifically granite rock fragments, diorite rock fragments, and also then the gneiss rock fragments. The whole trick is to identify the rock fragments down to the most detailed type possible. However, you will at times be forced to use the term rock fragment if identification is not possible. The closer that you can get to identifying the rock fragment the closer you will be to knowing the source area of the sediment.
Keep track of any fossils present; these are called bioclasts. They may be calcite, calcitized aragonite, silicified aragonite, silicified calcite, dolomitized calcite or aragonite, or carbon. Keep track of the different mineral types, don’t just call then bioclasts. If you can identify the type of fossil do so.
What is between the grains? There could be
· open pore space,
· cement, or
· matrix.
Pore spaces should be described and tallied in any modal analysis. Pore types can be categorized into
· intergranular porosity (between grains),
· intragranular porosity (pores within a grain),
· moldic porosity (a pore formed by the dissolution of something which you can recognize,
· fracture porosity,
· or any other type that may jump out at you.
If you do not know what to call it then apply the term pore. Most thin sections will have little porosity and what there is may be difficult to see. Some thin sections may have been impregnated with blue epoxy in which case the pore spaces will be readily visible as blue areas.
Clay and silt size silicate mineral material, when it occurs between grains, is referred to as matrix. It may have been deposited as clay size fraction material during the depositional event or it may have formed as the result of a diagenetic process such as devitrification of volcanic glass; hard to tell in thin section how it formed. Note all the cements as pore filling material. The quartz outside of the dust rim on a detrital quartz grain is authigenic quartz cement. Calcite, gypsum, anhydrite, dolomite, and barite are all common cement minerals.
Don’t forget the unknowns and don’t forget to describe them (rule: length and detail of description is proportional to abundance). Hey, we all have unknowns; do your very best to figure out something but when everything fails call it an unknown.
Well that takes care of space but also spend some time looking at the boundaries between the various items in the thin section. Look for and make note of concavo-convex grain to grain contacts (figure 2). microstyolite contacts (figure 2), etched surfaces, euhedral surfaces, all clues to diagenesis.

Figure 2: Types of grain-to-grain contacts that can result from pressure
solution.
Lots of stuff to seek out!!!! The big stuff pops out during your initial thin section survey while the rest will materialize during the modal analysis...keep a good set of notes along the way.
Specifically for sandstone and this lab exercise the first item of business is to survey the thin section and build your list of observed items for counting. List the framework grains first in your list since they had better be the most abundant items present. Then list the various materials that fill the space between the framework grains. Most abundant to least abundant. Below is a possible brief list.
Part of the craft of doing this is planning ahead. You do not want to do a 300 count modal analysis and then at some other time do it again counting different items and you want your list organized to make it easy to use for several different purposes. Look at how the above list is organized. Items 1-4 are detrital grains listed such that the most abundant type monocrystalline quartz is first. Next follow all other types of detrital quartz grains. Then generic feldspar followed by specific feldspar types. These are the types of material that are used in many classifications of sandstone. Items 5-10 are accessory grains. Items 10-12 are what fill the space between the grains. Always leave open pore types last in the list. Now if I was to ask you what is the porosity of the sample? All you would have to do is look to the end of the list and add up the appropriate items (here just #11). If I were to ask what was the primary porosity of the sample then you quickly add up items 11, 12, and 13. Note that, with the exception of the unknown, each has a mineral specified, that way you could quickly work out a bulk chemical composition. Or quickly determine if this were limestone or sandstone!
What we will use here are classifications based on the results of modal analysis or in other words classifications based on the various components which make up the sandstone.
A rule that I always go by and it would be a good practice for you to do also is that whenever you are using a particular classification you state somewhere just what classification it is that you are using and then stick to it strictly.
This way one avoids some miss communication. Pick up any geology journal and look at the names in use and you will discover quickly that the same name is applied differently by different workers. In this lab exercise we will be working with the classifications of Robert Folk and Robert Dott. Each uses the term 'quartz arenite' yet that term means different things in each classification. If you were to describe a particular sandstone as being a quartz arenite there would be some question as to what you actually meant. But if you said quartz arenite in sensu Robert Dott, 1964, well someone could look up Dott 1964 and see just what you meant. If you are writing a paper or report always somewhere in the Introduction or Methods state what nomenclature is being applied.
Folks Classification of Sandstone
Robert Folk’s classification of sandstone is based on the relative percentages of the major components of sands. The scheme uses a trilinear or ternary diagram (Figure 3).

Figure 3: Folk’s classification of Sandstone. Diagram from Folk (1980).
The main triangle or field has three poles:
To figure out where a particular sandstone would plot on this diagram go to the modal analysis. Add up the number of hits on the detrital quartz grains (Q-pole), add up the number of hits of feldspar, granite and gneiss rock fragments (F-pole), add up the number of hits on all the other rock fragments (RF-pole). Ignore everything else. Add these three groups of hits together and then divide each by that sum to get the percent for each pole.
Q% = (Q-pole)/((Q-pole) + (F-pole) + (RF-pole))
F% = (F-pole)/((Q-pole) + (F-pole) + (RF-pole))
RF% = (RF-pole)/((Q-pole) + (F-pole) + (RF-pole))
Folk classification has seven divisions or clans (figure 3):
If you find the rock sample to be litharenite or a feldspathic litharenite and if you were able to identify the various rock fragments present then you could go further by subdividing these clans. Divide the rock fragments into volcanic rock fragments (VRF), metamorphic rock fragments (MRF) and sedimentary rock fragments (SRF) and determine the relative percentages of each like you did above. For example from a 200 count modal analysis 20 hits on SRF, 20 hits on MRF and 20 hits on VRF or 30% of the sample is rock fragments but of the rock fragments 33.3% are SRF, 33.3% are VRF, and 33.3% are MRF. Plot these on a trilinear diagram using VRF, SRF, and MRF as poles. This second trilinear diagram (figure 3) defines the three subclans of the litharenites: volcanic arenite, phillarenite, and sedarenite. You could go further with sub-subclan divisions again see Figure 3 for chert arenite, calcilithite, sandstone arenite and shale arenite. For the arkoses a similar game can be played using feldspar type (Figure 3). I suppose something could also be done with the various quartz types but I have never seen such.
Dott’s classification, which is also widely used, differs some what from that of Folk. The Folk classification ignored the presence of matrix whereas Dott’s makes use of such. Dott like Folk uses trilinear diagrams but the definitions of the poles differ slightly. Notably chert falls under the quartz or Q pole not under the rock fragment or RF pole. Also the divisions and names differ considerably. Look at your matrix percentage from modal analysis if it is less than 15% then the rock is one of the arenites. If it falls between 15% and 75% it is a wacke. If greater than 75% it is a mudstone. Note also from Figure 4 how the lithic arenites and lithic graywackes can be further subdivided like in Folk’s classification.

Figure 4: Dott’s Classification of Sandstone. Modified after Dott, 1964.
Exercise: For any 5 of the assigned thin sections (see attached list) you will make a 200 count modal analysis. Percentages and 95% confidence limits will be determined. For the modal analysis differentiate between grain types to whatever degree that you can. Like wise for all the cements, matrix and pore types. You also will be making comprehensive notes of the features that you observe in each thin section. This will include a characterization of each item in your modal analysis. Next determine as completely as possible both a Folk and a Dott name for the sample. Put everything into a formal report in Microscoft WORD. The report should be of professional quality. Each thin section is to be treated separately form the others. I am looking for detail in your reports. This lab exercise and the write-up for it is a great opportunity for you to use terminology. You will turn in a digital copy only of the report.
Notes:
99-Cu-8 This is fine-grained sandstone from one of
the geology field camp project areas near
99-Jc-6 This conglomerate is from the Jack's Cabin geology field camp project area near
00-MO-4 This sandstone sample is from the Saint Peter
Sandstone of
03-SP-01 This is a pebble that I picked up near Crested Butte, Colorado. Technically this is a rock fragment. You can spend some time looking at it with your hand lens but don't as there is nothing to see. Garb the thin section and lay it down on a piece of white paper. Note the textural nature of this; sort of a non-descript-swirlly pattern. This is the result of bioturbation. Not exactly a true trace fossil but worthy of note any way because it says the sediment experienced some sort animal activity. It is always a good practice to lay your thin section on a white paper and see what you can see; sometimes nothing and other times you may learn something. You can also learn a lot by making your own thin sections. Look at the thin section under cross polarized light (CPL) low magnification. How is the thickness? Go up one objective. There are quartz grains, some feldspar, micas, a few rock fragment, some brown amorphous areas which are kerogen and a clay matrix. The matrix accounts for more than 15% of the sample but less than 75% so by Dott this is some sort of wacke so you have a wacky rock fragment, sorry, it is Friday, it is late, and I am hungry.
03-SP-04 This is sandstone from
44-6145 Ward's Natural Science sold this as a 'graywacke'
from
44-6148 This was sold by Ward's Natural Science as a 'bitumenous; shale from
44-7353 This was sold by Ward's Natural Science as
'carbonaceous' shale from the Llewellyn Formation (Pennsylvnian)
of
CY-30 This 'graywacke'
comes from
SP 800 This siltstone is from the
Pennsylvanian of southern
SP 803 This siltstone is from the Pennsylvanian of Kentucky. Lay the thin section on a white sheet of paper and note the sedimentary structure. Next, hold it up to a light and use your hand lens on it. Note the graded bedding. Which way is up? Now put the thin section on the microscope and using low power and PPL observe the graded bedding, neat???? Note each bed grades in the same direction. This sample is from a distal turbidite. Grains are mostly quartz, some detrital mica and detrital clay and some authigenic clay (see SP 800). The opaque areas are amorphous carbon and much of the brown material is kerogen.
SP 811 This sandstone is from the Castle Rock Formation of Douglas County, Colorado. Observe the un-cut portion of the hand sample with your hand lens and note the pink grains, they are potassium feldspar. Note the grayish glassy grains; these are quartz. Note that there are many white grains and note that they exhibit cleavage; these are feldspars. Now observe the same on the cut portion of the sample and note here the abundant porosity. In hand sample you can call this an arkose. Now go to the thin section and observe under the lowest power and under crossed polarized light (CPL). What is the thickness of the thin section? Look around the thin section and not that feldspar is the dominant grain type but there is also an abundance of quartz and several different types of rock fragments. Note next that the area between the grains is dark. Now switch to plain polarized light (PPL) and observe closely the area between the grains. Observe the brown areas here and switch between CPL and PPL. Note that this material is isotropic. Also, note the pore space. Run up the magnification a bit higher and look at these. The cement here in this sample is opal. This should not be considered matrix.
96-GC-20 This sandstone is form the Sullivan Butts Latite exposed near Chino Valley, Arizona. The Sullivan Butts Latite, as the name suggests is an volcanic sequence formed during the Oligocene. It contains sedimentary rocks including sandstone and limestone. What you will see in this sample is a variety of rock fragments including granite rock fragments, numerous mono-crystalline feldspar grains (largely plagioclase), both angular and rounded mono-crystalline quartz grains suggesting a duel source area, numerous mono-crystalline grains of interesting mineralogy (don’t get bogged down in trying to figure them out), and calcite cement.
QZ039
QZ040
QZ041
QZ042
QZ044
QZ045
QZ046 This sample has been impregnated with blue epoxy.
QZ048 This sample has been impregnated with blue epoxy.
QZ049 This sample has been impregnated with blue epoxy.
QZ052
QZ056 This sample has been impregnated with blue epoxy.
99Cu5
CP571 This sample has been impregnated with blue epoxy.
SP224