Fundamental Concepts 2: The Rock Types

To a geologist there are three broad classes of rocks, each of which has many subclasses. The confusing thing is that there are different ways these subclasses are organized, and these reflect the different ways geologists understand substances. These are:

a) genetically. How did this rock come to be how it is?

b) chemically. What is the bulk or trace chemistry of this rock?

c) mineralogically. What are the minerals in this rock?

In terms of ultimate origins, all rocks are either the result of Earth processes or else are meteorites. In terms of Earth processes, we can think of these as 'stuff still happening' which applies in the vast majority of cases, and a few odd cases from the early history of the Earth - cases that predate the current regimes.

What do I mean by regime? Well, really, rocks are chemicals and the minerals you get, the shape of those minerals, the condition of those minerals and so on is all a function of chemistry, pressure, temperature, and deformation (being squeezed or faulted or...). So the names geologists use for rocks are a mix of descriptors that help us relate to the pressure, temperature, deformation, and chemical conditions when that rock - or mineral - formed either directly (name the collection of minerals, name the genesis itself) or indirectly.

A mineral is a distinct chemical compound with a regular crystal lattice, and usually with fixed or at least highly constrained chemical constituents. A rock is a collection of minerals plus voids and perhaps some captured liquids and so on.

So we want our rock names to capture how those minerals in that combination were formed, and perhaps how much they got deformed during or subsequent to formation. And the chemistry. And the observable field characteristics (things you'd see if you were out looking at an outcrop). All of which is why rock classification is kind of a mish-mash.

The final, kind of fuzzy idea that is hugely important is that of stability. A mineral is generally metastable, which means that it may not 'like' being at the surface of the earth very much but it doesn't spontaneously degrade to forms that are stable at the surface. For example, feldspars are formed at moderate temperatures and pressures and will degrade to clay at the Earth's surface but this takes time, chemical attack, energy, and so on because the feldspar grains are metastable. Metastability also applies to subsequent geological events. If we squish and heat and lubricate (with water, salty water, bubbly salty water,...) a rock some minerals may stay as is but some may transform to new minerals (metamorphism) or incorporate atoms from the fluids and give some atoms up while making new minerals (metasomatism). However, the pre-existing state will often be discernable directly or indirectly - it isn't a 'whole new rock.' If it weren't for two factors - metastability and the fact that most attack is via fluid access which requires fractures - there would be no rocks at the surface, only lithogenic soil.

Okay, so we can now start on rock types!

Igneous rocks are rocks that formed from melted rock and have remained pretty much intact since. The minerals that form are a complex mix of the temperature and pressure during melt cooling, amount of water, whether minerals can settle out of the melt or not, and what the starting composition was. Equally important, the composition of a melt you get when you heat a rock from cold to molten is a complex mix of temperature and pressure, amount of water, whether you separate the melt from the source or not, and what the starting composition was.

The one crucial detail is, and this is confusing, that the composition of the melt is not the same as the composition of the rocks that are melting - the source. For example, if you take a bunch of common rock types found on the continents and mix them thoroughly and heat them, the first melt you get is more or less the composition of a granite, and as you melt more and more of the source the composition will shift from granite towards the composition of the source. When it is all melted, the composition of the melt will be the same as the source rocks. That's right, here it is again: the original composition of your 'mix' can have a fairly wide range and the initial melt will be virtually identical.

This means that you can get a repeatable set of basic rock types from a huge range of meltable sources, i.e., the vast majority of igneous rocks are from a very limited collection. Of course, experts like to concentrate on the exotic rock types, but more than 90% of all igneous rocks are roughly granites or roughly basalts. That'll be important when we make continental and oceanic crust in the next installment, since these are... more or less granite and basalt, respectively.

Our next rock family is sediments. Sedimentary rocks have two huge divisions: chemical sediments and clastic sediments. Chemical sediments are created by precipitation of minerals directly from a fluid at surface or near-surface conditions. Limestone is a great example, in that case with life playing a huge role in the precipitation. Clastic sedimentary rocks are where fragments eroded from pre-existing rocks are assembled and cemented together to form a new rock without melting or other high temperature or pressure process such as explosive fracturing. Sandstone is a good example. Importantly, weathering attacks some minerals more than others. Quartz tends to be well preserved in temperate climates, but feldspar and minor minerals tend to degrade to streams, so in a temperate or cold climate the rivers will contain sand grains, rock fragments, and fine silt composed of a mix of tiny fragments and clay grains. In a tropical climate plant-mediated attack on quartz is prevalent and you tend to get mostly the finer clay constituents. Since sandstones tend to get deposited different places than siltstones, this actually acts as a bit of a chemical separator - aluminous clay minerals end up in shales and silica ends up in sandstones, broadly speaking.

Clastic sedimentary rocks have to be cemented, and the most common cements are silica and calcium carbonate. Without cement, the sediment isn't rock!

Chemical sediments tell us a lot about their environment if we can deduce the living conditions of the critters that made them - paleogeography and paleoecology. Clastic sediments tell us a lot about their environment if we can deduce the physical conditions by which those particular size particles would end up together (and preserved). A sandstone might have been a beach or a river deposit, a shale may have been a part of a deep lake bottom or an ocean floor deposit, and so on.

Of course, both types of sediments have one thing in common: when they were initially deposited they were on the Earth's surface (i.e. at the water-rock or water-air interface). They can be buried, faulted, intruded, and so on but ultimately they were once at the surface! One is reminded of Leonardo recognizing fossil shells in sedimentary rocks on a mountain top and realizing that they were similar to those forming along the nearby seashore.

Of course, clastic sediments ultimately have to have been eroded from somewhere, and the ions in solution that get turned into chemical sediments have to originate somewhere too!

Metamorphic rocks are the final type and a little fuzzier to understand. A metamorphic rock is a pre-existing rock that is affected by temperature, pressure, chemical attack, and deformation to a degree that significant mineral and textural changes happen without obliterating the original rock type (by making a new igneous one!). Texture means the arrangement of the individual mineral grains (aligned or random, for example). The problem with these is 'what does significant mean?" As a rule of thumb, if the whole rock is affected thoroughly, then that is a metamorphic rock, if not, we say something like 'partially metamorphosed' or 'cut by veins showing alteration' or... sigh. It just isn't simple.

Every metamorphic rock has a protolith, i.e. the rock that was 'before.' The protolith can be metamorphic, igneous, or sedimentary.

Okay, that's a lot, so a few examples and we'll stop for now.

Granite: Igneous, intrusive (cooled while buried).

Rhyolite: Igneous, extrusive (cooled after being extruded or blown onto the earth surface; chemically equivalent to granite).

Sandstone: Sedimentary, clastic (cemented grains of sand-sized grains from erosion of a generally quartz rich source like a granite or a, uh, sandstone)

Limestone: Sedimentary, chemical (fragments of shells, calcium carbonate in composition, glued together by more calcium carbonate).

Schist: a metamorphic rock that has been heated and deformed enough that abundant mineral growth has occured, and with a chemical composition such that platy minerals like micas can form.

Gneiss: a metamorphic rock that has started to melt, and so shows bands of igneous material between remaining bands of source rock in a complex mess (often folded and contorted and... )

You can take a few dozen geology courses to get the details of the rock types, but that'll get you started, once we start describing environments and ultimate causes.