Petrology of Volcanic Rocks

The Nature of Volcanic Rocks:

An Educational Tour in Three Parts

J. Nicholls
 Theoretical Petrology
The Nature of Volcanic Rocks:
An Educational Tour in Three Parts

J. Nicholls
Copyright 2011 by J. Nicholls
4204 Chippewa Rd NW
Calgary, AB T2L 1A2
Canada

iv
 The Nature of Volcanic Rocks

Contents
Preface vi
Acknowledgements vi
Introduction 1
Objectives 2
How Do Volcanic Rocks Form? 3
Part 1: Volcanic Rocks and Volcanoes 7
Introduction 7
What This Tour is Not 7
Contingency and the Nature of Historical Science 9
What to Look for in Volcanic Rocks 11
Summary, Part 1 20
Part 2: Mineralogy and Volcanic Rock Types 23
Introduction 23
Mineral Assemblages and Activity of SiO2 25
Mineral Assemblages and Activity of Al2O3 33
Rock and Glass Compositions 35
Summary, Part 2 42
Part 3: Piecing it Together 45
Rocks, Melts and Volcanic Rock Names (Optional) 45
Silica Activity and Physical Properties 50
Two Volcanoes of the East African Rift Valley 52
What Rock Types Have We Left Out? 53
What Else is Part of Volcanic Petrology? 54
Exercises 57
Introduction 57
Exercise 1: Modeling Volcanic Melts 58
Exercise 2: Compare Volcanoes, Part 1 59
Exercise 3: Making Friends with Volcanic Minerals 62
Exercise 4: Compare Volcanoes, Part 2 62
Exercise 5: Write a Review of an Article 63
Exercise 6: Develop a Volcanic Hazard Assessment 63
References Cited 65

v
 The Nature of Volcanic Rocks

Preface
Professional development, life-long learning, informal learning, and free-
choice learning are labels applied to educational activities outside the
traditional classroom and curricula. See, for example, http://otec.uoregon.
edu/informal_education.htm or the paper by Falk and Dierking (2010) in
American Scientist. Many people participate in these activities, from expert
to novice. They attend lectures, go on field trips, visit museums, take short
courses, and participate in work shops of almost endless variety. Activities
are sponsored by universities, professional societies, hobby groups,
museums, and informal groups whose members share a common interest.
The information offered by these activities seems directed to either experts
or novices or to provide an overview of a fairly large topic. Little detailed
information seems focused on the basic concepts that underlie a field of
knowledge. When offered, basic concepts are often expressed in a technical
language, jargon if you will, that only experts can follow.
A couple of years ago, a colleague asked me, “What would you do if you
only had six weeks to teach igneous petrology?” My answer, “I don’t know,”
was accompanied by a profound feeling of relief that it wasn’t my problem.
Unfortunately, his question wouldn’t go away; it kept nagging; What would
I do? With time, this Educational Tour painfully evolved from what I
would teach in six weeks into something I hope non-specialists might find
informative. When next a volcano goes off, people familiar this Tour will
know what is a valid answer to the question, “What kind of stuff erupted?”

Acknowledgements
Advice, comments, and recommendations from friends and colleagues
changed the original concept of this educational tour into what you see
today. Paul Hoskins asked the original question about what I would do with
six weeks. Leslie Reid had me look at how the content should be presented
so that people could learn it. Two students, Colin Rowell and Erin Ernst,
showed me how geophysics look at volcanic rocks. A third student, Michelle
Spita, explained why one without a mineralogical grounding would not take
the time to master the background needed to learn some of the content of an
earlier version. To all of them, thank you. They are not responsible for the
errors and misconceptions that remain. Those are mine.
Title Page Photo: The Somma crater wall from the slope of Vesuvius, Italy. A lava
flow from the 1944 eruption of Vesuvius forms the gray strip of bare rocks at the foot
of the Somma Crater wall. Vesuvius is considered the most dangerous volcano in
Europe (Scarth, 2009).

vi
 The Nature of Volcanic Rocks

Introduction
Volcanoes and volcanic rocks fascinate people, scare them, and sometimes
even kill them. Ask people what volcanoes are made of and most cannot
tell you much more than that they’re made of rocks. Every once in a while a
volcano goes off, causes a sensation in peoples minds, and makes the news
of the day. The eruptions of Eyjafjallajökull, Iceland, in April and May 2010
caused the closing of European airspace for several days resulting in large
losses in revenue by commercial carriers. Several people posted entries on
the Web asking what the stuff was that caused the closure. Oppenheimer
(2010) wrote in op-ed piece for Elements, a magazine that publishes popular
mineralogical science articles:
Then on April 14, a swarm of earthquakes presaged a much
more vigorus stage of eruption, focused at the glaciated summit
crater of Eyjafjallajökull and involving trachyandesite magma.
What, you might ask yourself, is a trachyandesite magma? Technically
and precisely, it is a mixture of melted rock matter, entrained crystals, and
dissolved volatile constituents that, on cooling, would form a rock with a
composition such that it would fall in an area labeled trachyandesite on a plot
of SiO2 versus K2O + Na2O. Now imagine your kids, parents, grandparents,
or grandkids asked you, “What came out of Eyjafjallajökull?” Do you feel
better able to answer? Somehow, I very much doubt it.
We live on a cooling planet where igneous processes are but one
manifestation of planetary heat transfer and heat loss (Verhoogen, 1980).
Expressions of these processes at the Earth’s surface are volcanoes and
their products. Modern study of these processes and products began with
the eruption of a volcano where Vesuvius stands today. Pliny the Younger
described the eruption that happened in the year 79. Ash from the eruption
buried Pompeii. Many visit the site. If you’re one of them, do you wonder
whether the ash that buried the Roman town is the same kind of stuff as the
Icelandic ash from Eyjafjallajökull volcano and, by-the-way, what is volcanic
ash?
The aim of this Educational Tour is to create a framework in which you can
place the different volcanic rocks and do so without resorting to specialized
technical terms like trachyandesite. Conceptually, the task is simple. You do
not need to name the rock but you do have to find out what minerals the rock
contains. Although finding out what minerals a volcanic rock contains may
seem a formidable task, it can often be done by searching the professional

1
 The Nature of Volcanic Rocks

literature on volcanoes. Many, perhaps most, geologically young volcanic

centers have been studied and the minerals in the rocks identified. To use
these data effectively, you need to know which minerals are critical for
determining the nature of the rocks and which are not. This Tour is devised to
provide you with the information to do just that.
This Educational Tour takes you through a description of volcanic rocks
in stages. Part 1 describes the more general features of volcanic products
and provides an overview of volcanoes and volcanic settings. For many
people, this may be enough or even more about volcanoes than they care to
know. For others, it will just be a start on the information they want. Part
2 describes the critical mineral assemblages that distinguish different rock
types. These mineral assemblages can be used to create a framework; a
framework into which we can place nearly all volcanic rocks found here on
Earth. Part 3 collects the material in Parts 1 and 2 into a coherent whole.
Several examples are placed in the framework to show how it’s done. In this
framework, the differences between volcanic rock types are usually obvious.

Objectives
If you work your way through this Educational Tour, at its end you should be
able:
• To enter any classroom from grades K through 12 as part of a
geoscience outreach program and talk about volcanic products (rocks)
and phenomena (eruptions) with confidence and hopefully without
misconceptions, at least no serious ones.
• To read and review popular accounts of volcanic activity. These popular
science stories are not just those that appear in the publications of
news organizations; they also appear in popular science magazines like
National Geographic, American Scientist, and Scientific American.
• To write accounts of volcanic products and processes similar to those
published by popular science magazines. The October, 2009 issue
of Earth, a popular geoscience journal published by the American
Geological Institute, contains articles about volcanoes you might
find interesting: False Alarm Mystery at Volcano Solved, and An
Unprecedented Look Up a Supervolcano’s Skirt. At the end of this tour,
you should have the background and knowledge to write an article with
greater scientific content than either of these articles.

2
 The Nature of Volcanic Rocks

How Do Volcanic Rocks Form?

Igneous rocks, at least those found on and in the Earth, form from melts that
originate somewhere beneath the surface of the planet by partial melting of
pre-existing solid rock. Concentration of heat, compositional changes, or
pressure decrease, accomplished by a mechanism usually unknown, brings
the temperature in some relatively small part of the Earth above its melting
point. As the mechanism continues to operate in the source region, more of
the pre-existing rock melts. Because rocks are heterogeneous mixtures of
minerals, different minerals dissolve in the melt at different stages in the
process, and the composition of the melt changes with amount of partial
melting.
Igneous melts are usually less dense than the rocks from which they form.
Consequently, they are buoyant in the surrounding rocks and will tend to
move toward the surface. If there are unbalanced mechanical forces in the
source region or along the eruption path, the melts can move laterally as
well as vertically. Most igneous rocks form from melts that have moved
from the source region. We know that volcanic rocks always come from
such melts because volcanic melts don’t originate at the Earth’s surface.
On its journey, the melt may interact with the solid rock through which it
moves by dissolving some of the rocks on its perimeter. It may crystallize
minerals during its journey and leave some or all of them behind. It may stop
and crystallize in place beneath the surface. All these processes will, in all
probability, change the composition of the melt.
If the melt crystallizes deep beneath the surface, the resulting igneous rock
is labeled plutonic. If it comes close to the surface before crystallizing, the
resulting rock is a shallow intrusive rock. Such rocks are usually studied in
the same way volcanic rocks are studied. If the melt breaches the surface of
the Earth, we call it a volcanic melt. Beneath the surface, the melt plus any
entrained minerals, and fluids is labeled magma. After eruption at the surface,
the melt, entrained minerals, and fluids are called lava. After solidification,
the product of an eruption is frequently called a flow of some kind such as
lava flow or ash flow.
Near the Earth’s surface, some melts will exsolve dissolved constituents,
which are usually of low molecular weight, in a process similar to boiling but
a more complicated one. If the external pressure is low enough, these rapidly
expanding fluids can cool and fragment the melt, producing large violent
eruptions and ash flows or ash falls. So, it turns out that volcanic ash consists
of small rock and glass fragments, the glass being chilled particles of melt.

3
 The Nature of Volcanic Rocks

Figure 1: A, Mount Shasta, California, a Cascade Range volcano,

approximately 3070 m tall. B, Cinder cones in the Cima Volcanic
Field National Natural Landmark, California. The cones are
approximately 110 m tall.

Other magmas may fragment but less violently, producing larger fragments
to form cinder cones. Others may erupt without fragmentation, producing
lava flows. The large explosive eruptions leave a crater behind and the quieter
eruptions usually build topographic forms we call volcanoes.
When handed a rock, people usually just see a lump suitable for chunking at
something. They don’t imagine the history that any rock, whether volcanic
or not, must have. Between its origin in the past and its present state in thin
section, hand sample, or outcrop, every rock has a history. A lot of the fun of
being a geoscientist comes from working through that history. The historical
path followed by a volcanic melt from origin to solidification as a rock at the
Earth’s surface can be complex and torturous.
Using our imaginations to delve into the history of events in the Earth’s past
is not a well developed trait in most humans. We pass our day-to-day lives in

4
 The Nature of Volcanic Rocks

Figure 1, cont.: C, Ngorogoro Caldera, Tanzania. Ngorogoro

Caldera is approximately 20 km in diameter. D, Blue Dragon lava
flow, Craters of the Moon National Monument, Idaho. The lava
flow has about 10 m of relief at this location.

the immediate present. We act like the surface of the Earth is permanent and
seldom look at how past events affect our daily lives. This is especially true
if similar events are separated by times longer than a few years or sometimes
even a few months. When you come across volcanic rocks in your studies or
travels, please realize that they each have their unique magmatic histories.
Later processes and events can obscure or destroy features arising from the
earlier processes, preventing us from inferring the complete history. Whether
or not one can determine the complete history, we know every volcanic rock
formed in a source region, traveled through some part of the Earth, erupted
at the surface, solidified, and created or modified a landform. It matters not
whether they formed large classic volcanoes (Figure 1A) or small cones
(Figure 1B), blasted craters (Figure 1C), or flowed quietly across the
landscape (Figure 1D).
You can gain insight into volcanic processes and volcanic rocks in several
ways. One way, obviously, is to study them in the field. Another way is to

5
 The Nature of Volcanic Rocks

look at familiar substances that in some way mimic the products of volcanic
eruptions. Exercise 1: Modeling Volcanic Melts contains recipes for a set
of experiments that you can complete in your kitchen. Solutions of sugar
and water in some ways, mimic silicate melts. Some of the solutions won’t
crystallize even in your freezer. They mimic the glass that forms during the
eruption of some magmas. Solutions that are initially cooled slowly and
start to crystallize can then be cooled quickly by placing the container in an
ice-water bath to develop textural variations. I encourage you to complete
Exercise 1 in the near future. I think it will help you better imagine what
volcanic melts and rocks look like, how their properties change with
composition and cooling, and how this affects their behavior.

6
 The Nature of Volcanic Rocks

Part 1: Volcanic Rocks and Volcanoes

Introduction
Volcanic rocks form by two distinct mechanisms: melt fragmentation
and melt solidification. Which mechanism operated can be inferred from
the textures of the resulting rocks. Volcanic rocks are assemblages of
minerals and glass. Different mineral and glass assemblages, the textures
displayed by the interlocking minerals and glass, and geologic contexts
are used by geoscientists to tell different rocks apart. Petrologists have
always used mineral assemblages as an important criteria for distinguishing
different rocks. They then gave different names to different rocks with the
consequence that people had to learn a large number of rock names to talk
about volcanic rocks. We want to avoid rock names as much as possible.
We will concentrate on the features that tell us why the rocks are different.
The important early fact to understand: mineral assemblages are the critical
criteria for distinguishing different volcanic rocks. If time and desire warrant,
we can later look at applying formal names to the rocks.

What This Tour is Not

What do you need to know in order to meet the objectives of the Tour
(see page 2)? One thing you shouldn’t need is a glossary of rock names. Rock
names seldom appear in stories written for popular science journals. Yet, in
volcanic petrology courses, students consider a plethora of terms starting
with rock names: basalt, granite, granodiorite, diabase, ijolite, and on to
jacuparangite and beyond.
If you’ve taken an introductory geoscience course, you likely learned a few
volcanic rock names such as: basalt, andesite, rhyolite, and maybe obsidian.
Do you remember what these rocks actually are? Suppose someone asked
you, what’s a basalt? What would you answer? That it’s a dark, fine-grained
rock that comes out of volcanoes? The stuff that comes out of volcanoes, stuff
that has the potential for killing millions of people, deserves characterization
by more than a few rock names. The diversity of volcanic rock types here on
Earth has led to a large number of rock names, more than anyone can readily
comprehend. The IUGS classification of igneous rocks (Le Maitre, 2002)
contains 179 names equivalent in rank to the four listed above. This means
there are approximately 90 volcanic rock names. I’ve not counted them but
no one really learns this number of names, at least not to the extent they
would use them in an email to tell other people what to expect at a volcano
they visited.

7
 The Nature of Volcanic Rocks

In addition to rock names, volcanic petrology abounds in terms that

describe the textural variations of volcanic rocks: phenocrysts, ophitic,
aphanitic, porphyritic, trachytic ... Some of you may have even heard the
word diktytaxitic. It’s a textural descriptor for a three dimensional mesh of
interlocking crystals. If you have heard it, I suspect you remember it more for
the sound it makes coming off your tongue than for its meaning.
Even the names assigned to the shapes of intrusive rocks bodies boggle the
mind. The entry for cactolith in the first edition of the Glossary of Geology,
published in 1970, must be an all time classic:
A cactolith is a quasihorizontal chonolith composed of
anastomosing ductoliths whose distal ends curl like a
harpolith, thin like a sphenolith, or bulge discordantly like an
akmolith or ethmolith.
In all fairness I must confess that the entry does begin with a description in
what seems to be ordinary English:
“[A cactolith is] An irregular intrusive igneous body of
obscurely cactus-like form...”
Although “obscurely” is not an adverb I’ve seen in a definition of anything
anywhere else and what, exactly, is a ‘cactus-like’ form? The definition
remained unchanged through the 2nd, 3rd, and 4th editions of the Glossary.
Only in the 5th edition, published in 2005, is the definition of a cactolith
dropped. Although the 5th edition no longer has entries for catctolith and
chonolith, the entries for the shapes of the other rock bodies: akmolith,
ductolith, ethmolith, harpolith, and sphenolith remain – but does anyone
remember what the words represent?
The cactolith story has come down through the years as a petrology legend,
something akin to an urban legend. When I was a graduate student, many of
us thought the definition was created as a joke. It first appeared in a work by
Charles Hunt, a geologist with the US Geological Survey, who published it
in 1953 (Hunt, et al., p. 151). The only indication I can find suggesting Hunt
wrote the definition tongue-in-cheek is a quote he used from Holmes (1920):
“… Brevity of expression is by no means an unmixed blessing, and the one
word may require a whole paragraph of explanation.” I’ll leave it to you to
decide whether Hunt was serious – or not.
For decades geoscientists have complained about the jargon associated with
volcanic rocks; without stretching time too far, one could claim they have
complained for centuries. The complainers are not just general geoscientists
8
 The Nature of Volcanic Rocks

or specialists from other fields of geoscience. Volcanic petrologists

themselves complain, but mostly to each other. The jargon persists. Can this
avalanche of terms have any significance and if it does, what can it possibly
be? I think the perceived need for all these rock names, textural terms, and
names for rock bodies rose out of the diversity of volcanic processes and
products. That the volcanic rocks possess a diversity of mineral and chemical
compositions, textures, and shapes that borders on chaos is a fact of nature.
The numerous terms reflect an attempt to bring order out of chaos.
The tour will provide a different perspective on volcanic rocks than that
taught in many geoscience courses. Not necessarily a better one and definitely
not a more comprehensive one, but a different one; one that should help you
understand the nature of volcanic processes and perhaps help you assess
whether there is something to worry about when you get near a volcano. You
will not be asked to learn the technical language of petrologists. Excepting
only a half-dozen terms or so, I’ve tried to make the text devoid of rock
terms. Toward the end I’ve included a short, optional section describing how
some of the more common rocks and their names fit into the framework we
will develop.

Contingency and the Nature of Historical Science

Stephen Jay Gould (1985) described the history of life as contingent.
Dictionaries define the word as meaning “dependent on what may happen,”
“possible but not certain,” and “chance.” However, if the history of life is
contingent, then contingent must also mean that repetition of the events that
led to a particular life form, such as you or me, is unlikely. What happened
once is unlikely to happen again. Events governed by inorganic processes
are generally not recognized as contingent. Rather, geoscientists approach
such events as if they fall into the realm of laboratory chemistry and physics.
Do the experiment again under the same conditions and the same result will
follow. The problem is, we can never get a volcano to erupt under the same
conditions as it did before. Its a tautology that if the conditions are different,
then the event is contingent. If the processes studied by geoscientists are
contingent, then geoscience events by their very nature must be diverse in
process and product.
The idea of contingency in the operation of the universe has subtle
implications in how we approach the problems of historical science,
implications that are not fully appreciated. In my opinion, most geoscience
courses ignore or gloss over contingency. For many us, our training is such
that we need to go out of our way to look for contingent features in our

9
 The Nature of Volcanic Rocks

studies. But we first need to get a clear grasp of what contingency is.
Those who play poker know that the chances of drawing the ace, king,
queen, jack, and ten of spades, an ace-high straight flush in spades, from a
well-shuffled deck are infinitesimally small. In fact the odds are 1 chance
in 311,875,200 if we require the cards to be drawn in order from ace to ten.
Now, take a well-shuffled deck and deal five cards. When I performed the
deal, I didn’t draw the straight flush, hardly a surprise. Instead I drew the
queen of spades, the four of spades, the ten of clubs, the ten of spades, and
the king of hearts (QS, 4S, 10C, 10S, and KH). If the odds are incredibly low
that we will draw an ace high straight flush in spades, then what are the odds
of drawing a particular poker hand, such as QS, 4S, 10C, 10S, and KH, from
a well-shuffled deck? I certainly didn’t draw the QS, 4S, 10C, 10S, and KH
when I drew five cards from a second, well-shuffled deck. Nor did I get the
straight flush. The odds are actually the same as the odds of drawing the ace-
high straight flush in spades: 1 chance in 311,875,200. The two odds are the
same because all poker hands from a well-shuffled deck are equally probable.
Before the deal, all we know is that we will be dealt a poker hand. We will
enter the round of poker with that hand, no matter what the odds, simply
because we must be dealt five cards. In the universe of shuffled card decks,
different decks produce different results.
Some would argue that a particular volcanic eruption is so improbable that
it couldn’t happen by chance. Some thing must have directed it happen. But
just as a poker hand must be dealt, a volcano must erupt. Like the contingent
shuffle that led to a particular poker hand, contingent conditions lead to a
particular volcanic eruption.
Like poker hands, volcanic processes and products are contingent. Few
geoscientists studying inorganic processes have recognized contingency as
an important factor. One who did was F.J. Turner. He entitled his Presidential
Address for the Mineralogical Society of America in 1969: Uniqueness
versus Conformity to Pattern in Petrogenesis (Turner, 1970). In one of his
examples, he discussed the proliferation of names for different varieties of
rocks like those covering the ocean floors. Each eruption produced a unique
product; each eruption was a contingent event in the sense we are using
contingent here and contingency contributed to the numerous names for the
different rock types. Everything that happens during one eruption at one
volcano won’t happen again when the volcano next erupts. The order of
events are especially likely to differ. Different volcanoes, even where close
together, have different histories, products, and eruptive styles. Mauna Loa

10
 The Nature of Volcanic Rocks

and Kilauea on the Island of Hawai’i are about as close together as two large
volcanoes can get. Yet, their lava flows are distinctly different. Contingency
means we can expect the unexpected when looking at volcanic rocks. I think
fascination with volcanic rocks stems from their diversity in rock type and
their diversity of process of formation. There will be something new to learn
every time we look at another volcano or another volcanic eruption.
In this Educational Tour, we will look at the features of the rocks that derive
from the diversity of rock types. We will also look at differences between the
products from different volcanoes and differences in products from different
eruptions from the same volcano. In each case, we will be looking for unique
features. By searching out the unique for every volcano and volcanic rock
you encounter, you will be less likely to miss something important and more
likely to find something new and significant.

What to Look for in Volcanic Rocks

So, how do we learn why volcanoes behave differently? Several avenues
of inquiry are available to us. We could experimentally study the physical
properties of melts and melt-crystal mixtures to find answers to such
questions. Experimental studies of melts and crystals are a time- and
result-honored method of studying igneous processes. In many ways,
igneous petrology provided the impetus for laboratory experiments in
geoscience. However, most of us lack the time, resources, and expertise for
sophisticated experimental studies. But even simple experiments, like those
described in Exercise 1, which I hope you tried, can provide some insight
into the properties that control behavior during volcanic eruptions. For
example, the viscosities of the sugar-water solutions you made change with
temperature and with different concentrations of sugar. Solutions with higher
concentrations of sugar are more viscous. Viscosity controls how fast a melt
will flow and, like the sugar solutions, the viscosity of silicate melts changes
with temperature and composition.
Although experiments can point us along the path to deciphering the history
of a rock and volcano, a combination of field work, mineralogical studies,
and chemical analyses also lead to data-constrained inferences about the
behavior of the eruption that produced the rocks and volcano.
What, then, are the important features of volcanic rocks you need to learn?
You will get different answers from different petrologists. The ones I think
most useful when characterizing volcanoes and volcanic products are:
geologic context, texture, mafic index, silica activity, alumina activity,

11
 The Nature of Volcanic Rocks

and chemical composition. Whether all agree these are the most important
features is, to some extent, immaterial; if you know something about these
particular features for any particular volcanic rock, then you will know a lot
about its nature and the processes that formed it.
1. Geologic context: Of all the things geoscientists use to characterize
geologic features, not just volcanic rocks but geologic features in
general, geologic context is one of the most important. Seeing geologic
features in context is the prime reason for field courses. All of us who
have taught field courses can tell stories about how students came alive
as geoscientists because they really saw geologic features in context for
the first time. Often volcanic petrology courses present only a general
pattern of correlation between geologic setting and rock type. The
uniqueness of the geologic context and the rocks produced by different
eruptions deserve mention. I have heard professional geologists express
concern and confusion about finding rocks rich in the alkali metals,
sodium and potassium, in geologic settings that had no features that
would suggest rift zones or crustal rifting. They had been taught that
such rocks erupt only in rift zones.

Does a relationship exist between rock type and geologic setting? In

general, yes one exists, but exceptions enough occur to make uncritical
acceptance of a specific relationship untrustworthy. Volcanic rocks form
at rift valleys, spreading ridges, and hot spots. For example, most rocks
forming from eruptions at mid-ocean ridges contain essential plagioclase
and pyroxene in sub-equal proportions yet in Iceland, which straddles
the Mid-Atlantic ridge, can be found rocks with feldspar as the most
abundant mineral as well as the plagioclase-pyroxene-bearing rocks. The
Galapagos Islands, which sit above an oceanic hotspot, are built from a
variety of rock types. A variety of rock types erupted at the Craters of the
Moon National Monument in Idaho. The most recent products, erupted
between 2500 - 2000 years BP, are distinctly different from the lava
flows that underlie the most of the surrounding Snake River Plain, Idaho
(Stout, et al., 1994). Diversity and contingency have intruded to bring
complexity where we want simplicity.

Geologic context places the volcanic rocks we study into the framework
of Earth history, Below each volcanic center at some time in the past,
melting happened deep in the planet. The melt journeyed to the surface
where it erupted. The erupted products transformed themselves into
volcanic rocks, influencing the events at the surface, thereby changing
12
 The Nature of Volcanic Rocks

what night have been.

One last note about geologic context: If handed a volcanic rock and
asked to comment, the first thing you should do is ask, “Where did the
rock come from?” In my opinion, the geologic contexts in which rocks
are found are so important that we should not be asked to even look at
them without minimal information about their context.
2. Texture: Next to their mineral compositions, textures define volcanic
rocks. Volcanic eruptions span an energy range from violent, which
produce fine-grained fragmental products to quiet, which produce lava
flows. Between these extremes, eruptions create craters, large volcanoes,
small ones, and flows. If you get caught near one of these eruptions
and live to talk about it, the differences are obvious. In the rock record,
which tells us what happened in the geologic past, when there was no
one to observe the eruptions, field relations and textures tell us about the
nature of the eruption.

Figure 2 shows what comes from the extremely violent eruptions and
from the very quiet. Differences manifest themselves at all scales from
the volcano to the microscopic. Figure 2A and Figure 2B show two
volcanic centers that produced contrasting products. Craters of the
Moon National Monument in Idaho (Figure 2A), contains lava flows
and cinder cones. Vesuvius – Somma, Italy (Figure 2B) consists of
two volcanoes. A younger cone (Vesuvius) sits inside the crater of the
older one (Somma). The older one erupted in the year 79, producing
fragmental debris that covered the city of Pompeii, annihilating everyone
within it. Vesuvius proper grew inside the older cone during the years
following the eruption of Somma and the destruction of Pompeii.

Figure 2C and Figure 2D show outcrop exposures of products of quiet

and violent eruptions. Figure 2C shows lava flows in the wall of the
Snake River Canyon near Twin Falls, Idaho. The Snake River Plain is
underlain by lava flows. They were erupted over the last 5 million years.
Figure 2D shows layers of ash fall tuffs in the white cliffs that form a
canyon, appropriately named Tuff Canyon, in Big Bend National Park,
Texas. Although both products occur as layers, the distinction between
the two is clear. The lava flows are solid rock; the ash fall deposits
readily disaggregate. With cooling history, time, burial, and alteration,
the fragmental nature of an ash fall can become less obvious.

13
 The Nature of Volcanic Rocks

Figure 2: A, Craters of the Moon Lava and Cinder Cone Field,

Idaho . B, Vesuvius sitting in an older volcano, Somma, Italy.
Somma erupted in the year 79. Ash falls and ash flows destroyed
Pompeii. Photo taken from Pompeii.

14
 The Nature of Volcanic Rocks

Figure 2, cont.: C, Lava flows Snake River Canyon, Twin Falls,

Idaho. D, Ash fall tuffs in Tuff Canyon, Big Bend National Park,
Texas.

15
 The Nature of Volcanic Rocks

Figure 2, cont.: Photomicrographs of rocks in thin section. E,

Typical textures in a lava flow, Hawai’i. F, Textures in a unit from an
ash flow tuff, California.

16
 The Nature of Volcanic Rocks

Figure 2E and Figure 2F are photomicrographs of a lava flow from

Hawai’i and an ash flow from California. Figure 2E shows features
typical of lava flows. The crystals are intergrown, partly or completely.
Some crystals are bounded by crystal faces, partly or completely.
Some crystals have irregular shapes with bounding surfaces that fit the
surrounding crystals. The mineral assemblage in this lava flow is one of
the most common on Earth: Clear olivine with light brown, rusty edges,
brown Ca-rich pyroxene in several sizes and shapes, clear plagioclase
with low relief (it appears to lie below the surface of the other minerals),
and small opaque, Fe-Ti oxide minerals. Figure 2F displays a rock made
of brown glass, rock fragments and broken crystals. Close inspection
reveals boundaries between small glass particles. All these features
indicate a fragmental texture from an explosive eruption. Deformation
while the glass fragments were still hot caused them to coalesce and
compact.

Textures store much of the information we use to interpret how melts and
magmas change into volcanic rocks. Among the inferences are cooling
rates, whether melts arrived at the Earth’s surface at temperatures above
those required for complete melting, and whether two or more minerals
crystallized simultaneously or consecutively. They can also tell us
something about the properties of the melt. For example, some lava flows
have textures consisting of an interlocking mesh of crystals covered by
a film of glass. Almost surely, some melt drained from the solid mesh,
much like water draining through a kitchen strainer, leaving a film of
melt on the mesh of crystals and a film of water on the strainer wires.
The only difference: in the lava flow, the melt chilled to a glass whereas
the water remained liquid unless you put the strainer in the freezer.

After looking at textures to distinguish fragments from intergrown

crystals, people look next to grain-size. Students look at hand specimens
in introductory classes and observe that plutonic rocks have larger
crystals than volcanic rocks. As a generalization, distinguishing plutonic
from volcanic rocks by crystal size is not bad but it is not always good.
In northern British Columbia at Llangorse Mountain, a lava flow filled a
cirque. The interior of the body is coarser grained than the margin. The
predominant grain-size of the interior is several millimetres. The lava
probably flowed from its vent much like the other lava flows associated
with several cinder cones in the area. It then collected in the cirque.
The bowl-like depression slowed the cooling rate of the central part of

17
 The Nature of Volcanic Rocks

the body. Consequently, the grain-size is larger than found in the lava
flows in the area and larger than the grain-size of the rocks that formed
by cooling quickly at the margins of the cirque-fill. Thus, we can find
volcanic rocks as coarse as some plutonic rocks. We can also find larger
rock bodies, which all would call plutonic because of their tectonic
setting, but with margins that cooled fast and developed crystal sizes as
fine as those found in some volcanic rocks. Geologic context is a much
better indicator of whether the rock is plutonic or volcanic than is crystal
size and it is the criterion that one should use to distinguish the two
volcanic from plutonic settings in the geologic record.

Because they record so much information petrologists have coined a

large number of terms for textures and their variations. However, I’ve
found that one can use ordinary English words to describe texture
without becoming too verbose. By using ordinary English words to
describe textures, you avoid confusing your reader if they don’t know
the jargon. You make your descriptions accessible to more people.
One tends to make incomplete observations, gloss over features or,
even worse, miss significant features entirely when using specialized
terms. One descriptor becomes sufficient in the eye of the beholder and
small variations and differences are over looked. Sometimes the small
variations and differences are significant. In any case, one doesn’t want
to be thought a poor observer.

Arthur Holmes’ 1937 description of an unusual rock from Uganda serves

as a model for those who use ordinary English to describe rocks:

The specimens representing the Katunga lavas are compact fine grained
porphyritic rocks [that contain large crystals set among smaller ones]
of lustrous stone-grey to deep blue-grey colour, the phenocrysts [large
crystals] being olivine (up to 3 mm. long) of various tints from nearly
colourless to yellow-green, and melilite (up to 2 mm. square) in thin
tabular, less conspicuous crystals which, megascopically, have a shining
grey to black appearance. Some of the specimens from near the sources
of the flows are internally altered and have become fawn or buff in
colour, mottled by white zeolitic specks and reddish brown pseudomorphs
after [crystal aggregates that have the shapes of unaltered] olivine and
melilite [they chemically replaced] .

even though it could be improved with at least one more edit. The

18
 The Nature of Volcanic Rocks

words in colour, for example, are redundant and the terms porphyritic,
phenocryst, and pseudomorphs after could be replaced with the words in
brackets.

Although descriptions of rock textures in ordinary language convey more

information to more people, you will encounter textural descriptors, such
as phenocryst, in the professional literature. There, authors commonly
use the jargon. A good petrography text, such as Williams, et al. (1982)
will likely define any terms you might encounter.
3. Mafic Index: Much as I would like to avoid technical terms completely,
I find writing about volcanic rocks without resorting to a few terms
impossible. Two such terms are mafic and felsic. Minerals that form
volcanic rocks fall into two broad groups: minerals that contain essential
Fe and Mg (mafic minerals) and minerals that don’t (felsic minerals).
The common minerals found in volcanic rocks that contain essential
Fe and Mg are the olivines, pyroxenes, amphiboles, biotites, and Fe-
Ti oxides. The common minerals that lack essential Fe and Mg are the
feldspars, quartz, nepheline, leucite, and kalsilite. These minerals have
compositions dominated by K, Na, Ca, and Al. Mafic rocks contain an
abundance of mafic minerals; felsic rocks contain an abundance of felsic
minerals. The textual context determines whether the terms, mafic and
felsic, apply to the minerals or to the rocks.

The mafic index is the percentage of primary mafic minerals. Glass,

which is not a mineral, is not included in the calculation of the mafic
index nor are minerals that formed by alteration. There is a continuum of
rock types between those that contain just felsic minerals and those that
contain just mafic minerals.

Much as one might rail at jargon, the mafic-felsic concept is firmly

gounded in mineralogy, the foundation we use to characterize volcanic
rocks. Perhaps in this case, the use of technical terms is justified.

Ranges of mafic index that petrologists use to distinguish different rock

types are:

Felsic rocks: Mafic minerals constitute 0 to 30% of the mineral

assemblage.
Intermediate rocks: Mafic minerals constitute 30 to 60% of the mineral
assemblage.
19
 The Nature of Volcanic Rocks

Mafic rocks: Mafic minerals constitute 60 to 90% of the mineral

assemblage.
Ultramafic rocks: Mafic minerals constitute 90 to 100% of the mineral
assemblage.

These ranges are arbitrary and I think they were selected more for
symmetry than for convenience. Estimating a value for the mafic index
by looking at a thin section will not be precise. In addition, a bias
for mafic minerals is built into many estimates of mafic index. Mafic
minerals are almost always colored in plane polarized light and often so
in hand specimen. Our eyes are drawn to the color and we tend to over
estimate their abundance. As a result, many rocks labeled mafic in older
works really have intermediate mafic indices.

Mineral assemblages in most volcanic rocks contain both felsic and

mafic minerals. You should explicitly look for and take note of the two
kinds of minerals early in your examination. Although crystallization of
felsic and mafic minerals are not independent of each other, they can be
studied separately.

A few volcanic minerals contain both mafic elements and felsic

elements. These rocks may have unusual properties or form from melts
with unusual properties. Melilite, for example, a mineral that occurs in
rocks with extremely low SiO2 contents, contains mafic elements as well
as felsic elements.

Summary, Part 1
When you read or write an account of volcanic phenomena, pay attention
to geologic and geographic context, descriptions of the textures of the
rocks, and their mafic-felsic mineral contents. The geologic context locks
the phenomena into the geologic history of the region. Textures indicate
whether the eruptions were explosive or quiet, and the mafic index supplies
information about the composition and mineralogy of the magmas,
Most accounts that appear in the popular literature carry at least some of this
information. An article about the Yellowstone supervolcano (Achenbach, et
al., 2009), for example, includes the concept of a hot spot and a continental
plate moving over it, which is part of its geologic context. They also provide
a time-line for the large, violent eruptions that preceded the eruption that
formed the Yellowstone caldera.

20
 The Nature of Volcanic Rocks

A narrative could also include descriptions of the textures and how these
indicate whether the products were lava flows, fragmental flows, or ash falls,
for instance. Finally, including in your narrative an account of the amount of
felsic and mafic constituents provides the reader with at least an idea of the
mineralogical nature of the rock and magma, albeit a generalized one. For
example, an estimation of the mafic index for the Eyjafjallajökull ash would
tell most of us more about the magma than does the term tracyandesite.
Popular science accounts of volcanic centers usually lack descriptions of the
mineralogical nature of the volcanic products.
Geologic context, texture, and mafic index encapsulate much of what
is exciting and important in volcanic petrology. If you can acquire a
comprehensive knowledge of only these three features for a suite of volcanic
rocks, you will have a solid foundation for talking and writing about the
magmas and eruptions that produced them.
At this point, why don’t you try your hand with Exercise 2: Compare
Volcanoes, Part 1? The two volcanoes, Nyiragongo and Somma-Vesuvius,
you are asked to compare are about as different as any two volcanoes here
on Earth but they are similar in that they both caused death and destruction.
When you read about these two volcanoes, ask yourself how the information
in the articles fits with what we’ve discussed so far. What more do you
need to know to understand what you’re reading? After making the outline
requested for Exercise 2, you should have ideas and questions about the
volcanoes that will help you anticipate what’s coming next in the tour rather
than have the concepts drop down on you from out of the blue.

21
The Nature of Volcanic Rocks

22
 The Nature of Volcanic Rocks

Part 2: Mineralogy and Volcanic Rock Types

Introduction
In Part 2 of our Educational Tour we take a more detailed look at the
differences between volcanic rocks. First though, to even talk about volcanic
rocks, we have to talk about the minerals they contain. Our first order of
business is to introduce the minerals you need to know. Not all minerals act
as guides to rock types; only certain critical minerals distinguish significant
differences between them. Our job in Part 2 is learn which minerals are
significant and to discover why they are significant.
A relatively small number of minerals distinguish different volcanic rocks:
quartz, sanidine, plagioclase, nepheline, leucite, kalsilite, olivine, Ca-poor
pyroxenes, Ca-rich pyroxenes, titanite, perovskite, melilite, monticellite,
anorthoclase, topaz, Na-rich pyroxenes, and aenigmatite. Seventeen minerals
make the cut. They are listed in Table 1. Several of these minerals may
be familiar from introductory geoscience courses: quartz, olivine, and
plagioclase, for example. You may have encountered others, such as the
pyroxenes, but don’t remember much about them. The others are seldom,
if ever, shown to students in introductory courses. Although you probably
know about topaz because it’s a gemstone and the birthstone for November,
you may not know much about it as a mineral. The thing to remember, the
unfamiliar minerals are no more special than the familiar ones on the list.
They have distinctive habits, compositions, and properties. The more you see
them, read about them, and read about the rocks that contain them, the more
familiar they will become.
It would be nice if everyone had the skills to identify the critical minerals
both in hand specimen and thin section but many do not have the time,
inclination, or background to develop such skills. All is not lost. The
literature contains mineral descriptions for many of the rocks you’ll come
across. A little library research can do wonders. If you cannot find about the
minerals in your rocks from library research, then you can ask someone with
the expertise to identify them for you. If these methods fail, about all that’s
left is to either forget it or learn how to identify them yourself.
Minerals become familiar as you learn about their properties, such as color
and crystal form. You become familiar with them as you learn what other
minerals typically accompany them in rocks. You become familiar with
them when you know which ones are prized as gem stones. The more you
encounter these minerals in the rocks, in the literature, or in museums, the

23
 The Nature of Volcanic Rocks

Table 1: Critical Minerals

Mineral, Activity SiO2 Mafic/Felsic
Quartz Felsic
Plagioclase Felsic
Nepheline Felsic
Leucite Felsic
Kalsilite Felsic
Olivine Mafic
Ca-poor Pyroxenes Mafic
Ca-rich Pyroxenes Mafic
Titanite Mafic
Perovskite Mafic
Melilite Mafic
Monticellite Mafic
Mineral, Activity Al2O3 Mafic/Felsic
Anorthoclase Felsic
Topaz Felsic
Na-rich Pyroxenes, Mafic
Aenigmatite Mafic
Notes: Titanite and perovskite need not contain essential Fe and
Mg. However, most workers include them among the mafic minerals
because they contain essential Ti, which is a transition metal as
is Fe. Some melilite minerals, notable gehlenite, Ca2Al2Si2O7, and
Na-gehlenite, CaNaAlSi2O7, don’t contain Fe and Mg. However,
the melilites we find in volcanic rocks do contain Fe and Mg.
Consequently, melilites are placed in the mafic group.

more familiar they will become. Mineralogy texts such as those written by
Deer, et al. (1992) or Dyar, et al. (2008) can provide much of the information
needed to learn about them. A study of these texts can also help with their
identification but most people become proficient in identifying them with the
experience gained by looking at hand specimens and thin sections. It’s worth
reiterating: you don’t need to identify the minerals in the rocks that come to
interest you on this Educational Tour. You do need to find out what minerals
the rocks contain.

24
 The Nature of Volcanic Rocks

Here would be a good place to do Exercise 3: Making Friends with

Volcanic Minerals. Exercise 3 is probably the most important of all the
exercises. By doing this exercise, you will, I hope, become acquainted with
the minerals that distinguish different rock types. Distinguishing the different
rock types is the big reason taking for this Educational Tour.
The following might help put Exercise 3 in perspective: Many of us have
some idea about the nature of DNA, even though we are not life scientists.
We know it’s a double helix with four different step-like atom clusters called
bases that hold the strands together. Few of us can identify the bases or know
how to identify them. Although they have formal names, we may remember
the first letters of their names: G, C, A, and T. We might have learned that G
bonds to C and A bonds to T to form the steps between the helices. How the
steps are sequenced along the double helix determines its properties and how
the life form that carries the molecule develops and lives. What you need to
do is learn enough about each minerals in Table 1 to write a short paragraph,
similar to the what I just wrote about DNA.

Mineral Assemblages and Activity of SiO2

We have looked at three features in our quest to learn about volcanic rocks:
geologic context, textures, and mafic index. We’ve three more to go. These
features will distinguish different rock types on a finer scale than the first
three. The first two of the three may seem esoteric in an Educational Tour
for the nonspecialist. But we will reduce them to something manageable:
a recognition of different mineral assemblages. To put it another way,
the concepts may seem abstract but they can be objectively related to the
minerals found in volcanic rocks.
4. Silica Activity: This feature, silica (SiO2) activity and the next one,
alumina (Al2O3) activity, act as sign posts for distinguishing significantly
different mineral assemblages in volcanic rocks. The activity of a
constituent in a chemical system is not a familiar concept from everyday
life. We can, however, relate activity to a familiar sensation, taste.
Suppose you add sugar to water (Exercise 1). The more sugar you add,
the sweeter the solution tastes. The activity of sugar in water correlates
with its taste. The solution becomes sweeter and sweeter until it saturates
and no more sugar will dissolve. Likewise, the activity of sugar in water
increases until the solution is saturated and then the activity no longer
increases. Sugar concentration and activity of sugar are obviously
related. The relationship is not linear, however. Concentration does
not equal activity and they should be thought of as distinctly different

25
 The Nature of Volcanic Rocks

things. You may have noticed that differences in taste are less distinct
in solutions with higher concentrations than they are in more dilute
solutions. In part this happens because of the nonlinear relationship
between concentration and activity; activity doesn’t increase as rapidly
as concentration in solutions close to saturation. It is also likely that your
taste buds saturate and are not as sensitive at higher concentrations.You
can use the solutions you made for Exercise 1 to do the taste tests.

In the water and sugar experiments, the presence of a single solid phase,
sugar, constrains the activity of the constituent dissolved in the solution
(melt). But notice, there are two different things here: solid sugar and
dissolved sugar. Activity changes with the amount of dissolved sugar.
The amount of solid sugar in contact with your solution after it is
saturated has no affect on the activity. In the same way, the activity of
silica (SiO2) in silicate melts increases until quartz saturates the melt and
crystallizes. After quartz saturates the activity will no longer increase as
more SiO2 is added to the melt.

Below quartz saturation, we obviously cannot taste the melt to determine

its activity. We could, however, calculate the activity of SiO2 in the melt
from its thermodynamic properties. Formal definitions of the activities
of constituents in multicomponent melts like silicate melts derive
from thermodynamics. For our purposes, it suffices to know that each
thermodynamic component has an activity. The activity of silica is only
one of many. The calculations could take the place of taste. It would
be nice to have a more direct method, however. Calculations are not to
everyone’s taste. It turns out that we can often use the minerals in the
melts to infer silica activity. If not the exact value, then limits on values
of silica activity can frequently be placed. In the more complicated
solutions of igneous melts, more than one solid can reflect the activity of
a chemical constituent. If a particular mineral assemblage is present in
a melt, it will signal a fixed activity of silica. If only part of the mineral
assemblage is present, the the activity of silica must be above or below
the fixed value. Which side of the fixed value the acticity is on depends
on which mineral is absent from the assemblage. Most of the critical
assemblages contain just two minerals.

Before we plunge into the critical mineral assemblages, it is worth

reviewing some of the similarities between silicate melts and sugar-
water solutions. The simple water-sugar experiments provide insight

26
 The Nature of Volcanic Rocks

into several processes connected with igneous rocks: Crystallization

versus glass formation or viscosity and flow rate are examples. Now, we
discover another concept that connects sugar-water solutions and silicate
melts: the chemical activity of a melt constituent. The properties of melts
change as the values of the activities change. The activities of some
constituents have greater influence than others. Just as the activity of
sugar affects the properties of sugar-water solutions (see Exercise 1), the
activity of silica has a profound affect on the properties of igneous melts,
such as whether glass forms rather than crystals or whether a lava flow
moves fast or slow.

Carmichael, et al. (1970) introduced silica activity to igneous petrology.

Prior to publication of the Carmichael, et al. (1970) paper, petrologists
had discovered features that closely correlate with silica activity. These
features became central to how petrologists thought about differences
between igneous rocks. At about the time the 19th century turned into
the 20th, four petrologists, W. Cross, J.P. Iddings, L.V. Pirrson, and H.S.
Washington (CIPW), met at Cross’s home in New England to talk about
igneous petrology. Out of their discussions came a paper (Cross, et
al., 1902) that has influenced thinking about igneous rocks to this day.
Parallels exist between Newton and his Principia and CIPW and their
paper. Newton’s ideas pervade physics and modern civilization was
created with the ideas he gave the world in the Principia, yet very few
of us have ever read it. Likewise, the ideas developed by CIPW pervade
igneous petrology but no one reads their paper.

CIPW devised sophisticated schemes for recalculating silicate rock

analyses into sets of mineral formulae. Which set comes out of the
calculation depends on the rock composition. The set of calculated
formulae is called the CIPW norm for the rock; each rock has a unique
norm. Please note that a CIPW norm is not a set of minerals. Rather, it is
a set of calculated chemical formulae.

The genius of CIPW lies in the correlation between the set of chemical
formulae, silica activity, and the actual mineral assemblages in volcanic
rocks. CIPW embodied the essence of their ideas in an ordering
of chemical formulae to match as closely as possible, the mineral
assemblages in rocks. The ordered assemblages can be listed on an
axis of silica activity. Melts with silica activities less than that at quartz
saturation will crystallize mineral assemblages we can use to estimate

27
 The Nature of Volcanic Rocks

0
Q Orthopyroxene

En
-1
Fo
Ab
Tn
Ne Feldspar
Pf

-2
SiO2
ln a

Lc
-3 Ks Melilite
Clinopyroxene

Di Feldspathoid
Mo
-4 Di
Ak + Fo Olivine

-5
800 900 1000 1100 1200
T°C

Figure 3: Curves showing the natural logarithm of the

activity of SiO2 versus temperature. Labels are the minerals
related by differences in SiO2 content. Arrows delineate
ranges of SiO2 activity for which minerals can coexist
with silicate melt. Abbreviations for minerals: Q, quartz;
En, enstatite; Fo, forsterite, Ab, albite; Ne nepheline; Tn,
titanite; Pf, perovskite; Di, diopside; Mo, monticellite, Ak,
akermanite; Lc, leucite; Ks, kalsilite.

silica activity. The CIPW authors codified several of these. They

were based on observations petrologists had noted in the early years
of igneous petrology. The order the CIPW authors assigned to these
assemblages correlates almost perfectly with values of silica activity

28
 The Nature of Volcanic Rocks

derived through thermodynamics (Carmichael, et al., 1970).

Figure 3 shows a plot of the logarithm of silica activity versus

temperature at low pressures. On the plot are curves representing
possible mineral assemblages in rocks: quartz (Q), olivine plus Ca-poor
pyroxene (En-Fo), feldspar plus feldspathoid (Ab-Ne), titanite plus
perovskite (Tn-Pf), leucite plus kalsilite (Lc-Ks), Ca-rich pyroxene plus
monticellite (Di-Mo), and Ca-rich pyroxene plus melilite and olivine
(Di-Ak + Fo).

Silicate melts saturated with silica will coexist with quartz. Rocks
containing quartz will have silica activities defined by the topmost curve
on Figure 4. Such rocks may contain other phases such as Ca-poor
pyroxene, Ca-rich pyroxene, and feldspar. But it is the presence of quartz
that tells us that the silica activity is as high as it can get.

Petrologists had noticed that quartz and Mg-rich olivine did not occur
together in the same rocks. The CIPW authors built this incompatibility
into their calculations. Ca-poor pyroxene and olivine are related by the
transformation:

Mg2SiO4 + SiO2 = Mg2Si2O6

Olivine Melt Ca-poor pyroxene
This chemical equation tells us that if more SiO2 is added to the melt
saturated with olivine and Ca-poor pyroxene at constant pressure and
temperature, it will combine with the olivine to crystallize Ca-poor
pyroxene. Because Ca-poor pyroxene extracts SiO2 from the melt,
crystallization tends to bring the SiO2 content and activity of the melt
back to their original values. In other words, the mineral pair, olivine
plus Ca-poor pyroxene acts as a silica buffer. The curve labeled En/Fo
on Figure 4 represents the coexistence of olivine, Ca-poor pyroxene, and
melt. At silica activities above the curve, Ca-poor pyroxene saturates the
melt and no olivine crystallizes; below it, olivine saturates the melt and
Ca-poor pyroxene does not crystallize. Rather, it dissolves in the melt.
Melts with Ca-poor pyroxene but without quartz and without olivine
have silica activity values between those belonging to melts with quartz
and those belonging to melts with both Ca-poor pyroxene and olivine.

Solid solution in olivine and Ca-poor pyroxene and the crystallographic

29
 The Nature of Volcanic Rocks

structure of the Ca-poor pyroxene complicate this simple picture a bit.

The En/Fo curve on Figure 3 is drawn for the magnesian end members
of olivine and orthorhombic pyroxene. The curve is slightly displaced if
the Ca-poor pyroxene is monoclinic Mg2Si2O6 instead of orthorhombic.
In extremely iron-rich systems, the Ca-poor pyroxene-olivine curve lies
above the curve representing quartz saturation (Nicholls, et al., 1971).
Consequently, quartz and Fe-olivine can coexist together in Fe-rich
melts. If completely crystalline, rocks that form from such melts are
invariably felsic, at least those that form from such melts here on Earth.
Consequently, the antipathy between quartz and olivine described by
CIPW doesn’t exist in melts rich in FeO and SiO2.

The curve labeled Ab/Ne represents the transformation:

½ NaAlSiO4 + SiO2 = ½ NaAlSi3O8

Nepheline Melt Albite

It locates silica activities in melts that coexist with feldspar and

feldspathoid. In this case the feldspar is albite and the feldspathoid is
nepheline. Both nepheline and albite are components of solid solutions.
The natures of the solid solutions are such that the Ab/Ne curve is
only an approximate upper limit on the activity of silica in melts that
crystallize feldspathoid. Feldspars can coexist with melts having lower
silica activities than the Ab/Ne curve because of compositional and
crystallographic variations in the feldspar and feldspathoid.

The Tn/Pf curve represents the transformation:

CaTiO3 + SiO2 = CaTiSiO5

Perovskite Melt Titanite

This transformation seems incongruous on a diagram delineating

different rock types. Why should a rock type depend on which of two
minor minerals is present or absent? It’s there because the Tn/Pf curve
marks the approximate lower limit of silica activity in melts that can
crystallize feldspar and feldspars are important indicators of rock type.
It’s a matter of observation that feldspars and titanite can crystallize
together from melts whereas feldspars and perovskite do not (Smith,
1970). Feldspars can crystallize from melts with silica activities between
30
 The Nature of Volcanic Rocks

quartz saturation and those approximately equal to those defined by the

titanite-perovskite curve (Figure 3).

Melilite has rarely been found with feldspars in terrestrial rocks. Velde
and Rachdi (1988) report the first occurrence of K-feldspar and melilite.
Anorthite and melilite can crystallize together in simple chemical
systems (Berman and Brown, 1984) and they have been found together
in meteorites (Deer, et al., 1992). The dashed line on Figure 3 marks
my guess as to where the values of silica activity fall for the incoming
of melilite. It may actually fall closer to the Tn/Pf curve that marks
the last of the feldspars than I’ve drawn it. But rocks that lack both
feldspar and melilite are more common in my experience than those
that contain melilite but lack feldspar. The less frequent occurrence of
melilite-bearing melts suggests to me that a substantial difference in
silica activity can exist between melts that dissolve feldspar and those
saturated with melilite.
At some silica activity below those of melilite saturation, Ca-rich
pyroxene will dissolve. Two curves are drawn on Figure 3 that
may approximate the silica activities where Ca-rich pyroxenes
disappear. They are labeled Di/Ak + Fo and Di/Mo; they represent the
transformations:

2
3
Ca 2 MgSi 2 O7 + 13 Mg 2SiO 4 + SiO 2 = 34 CaMgSi 2 O6
Melilite Olivine Melt Ca-rich pyroxene

and
CaMgSiO4 + SiO2 = CaMgSi2O6
Monticellite Melt Ca-rich pyroxene

Rocks exist that contain olivine and melilite assemblages but which lack
Ca-rich pyroxene. Melts that crystallize to form such rocks have silica
activities below the Di/Ak + Fo curve. There is at least one rock body on
Earth that contains monticellite and olivine but no Ca-rich pyroxene. It is
a shallow intrusion that forms Haystack Butte, near Geraldine, Montana.
It presumably crystallized from a melt with a silica activity below the Di/
Mo curve.

At low activities of SiO2, melts will crystallize kalsilite (KAlSiO4) rather

than leucite (KAlSi2O6). The coexistence of leucite and kalsilite defines

31
 The Nature of Volcanic Rocks

Table 2: Critical Mineral Assemblages and Silica

Activity
aSiO Felsic assemblage Mafic assemblage
2

~0.75 Quartz Ca-poor pyroxene or

+ Feldspars Fe-olivine in felsic rocks
+ Ca-rich pyroxene
~0.45 Feldspars Ca-poor pyroxene or
Fe-olivine in felsic rocks
+ Ca-rich pyroxene
~0.37 Feldspars Olivine and Ca-poor pyroxene
+ Ca-rich pyroxene
~0.30 Feldspars Olivine
+ Ca-rich pyroxene
~0.22 Feldspathoids and Olivine
Feldspars + Ca-rich pyroxene
~0.14 Feldspathoids Olivine
+ Ca-rich pyroxene
~0.05 Feldspathoids Olivine, Melilite
+ Ca-rich pyroxene
~0.02 Feldspathoids, esp. Olivine, melilite, or
kalsilite Montecellite

The activities listed in the left column come from converting

the logarithms of the activities on Figure 4 at a temperature
of 1000°C. If the assemblage can crystallize over a range
of silica activities, a point approximately in the middle of the
range was selected.

the Lc/Ks curve on Figure 3. the curve represents the transformation:

KAlSiO4 + SiO2 = KAlSi2O6

Kalsilite Melt Leucite

Rocks that form from melts that crystallize kalsilite but not leucite, may
be among the most SiO2 under saturated melts on Earth.

Table 2 lists in decreasing order of silica activity the critical mineral

assemblages in volcanic rocks, at least critical in terms of silica activity.

32
 The Nature of Volcanic Rocks

Felsic and mafic minerals are listed separately for convenience. You
should make sure both the felsic and mafic mineral assemblages, if both
are present, are consistent with your assignment of a rock to a specific
category in the scheme of silica activity.

Just as CIPW ignored biotites and amphiboles in their normative

schemes, Table 2 has no mention of these two minerals either. Biotite
can crystallize from melts with SiO2 activities that span the range found
in melts on Earth. Biotite is found in many quartz-bearing rocks. It
also occurs in the monticellite-bearing rock that forms Haystack Butte,
Montana. Amphiboles also occur in a plethora of rock types. Amphiboles
and biotites are both important and interesting mineral groups; they just
are not sensitive indicators of silica activity.

Mineral Assemblages and Activity of Al2O3

5. Al2O3 Activity: After SiO2, Al2O3, is almost always the next most
concentrated oxide in volcanic rocks. Chemical transformations like the
ones we used to determine silica activity, transformations like:

Mg2SiO4 + SiO2 = Mg2Si2O6

Olivine Melt Ca-poor pyroxene

have not been developed to describe the effects of Al2O3 activity.

Rather, petrologists have used inequalities between Al2O3 and Na2O,
K2O, and CaO to delineate different rock types. Rocks with different
values for Al compared to values for K, Na, and Ca have very different
characters. Petrologists have coined special terms for rocks characterized
by the inequalities. We will will call them rocks that crystallized from
melts with high, medium, low and ultra-low activities of Al2O3. The
inequalities used to characterize Al2O3 activity in volcanic rocks are:

High Al2O3 > Na2O + K2O + CaO.

Medium Na2O + K2O + CaO > Al2O3 > Na2O + K2O.
Low Na2O + K2O > Al2O3 > K2O.
Ultra-low K2O > Al2O3.

CIPW devised procedures to return mineral and chemical formulae that

characterize rocks with compositions corresponding to these inequalities.
Although not all the formulae returned by CIPW correspond to minerals,

33
 The Nature of Volcanic Rocks

enough of them do to point us toward critical minerals that characterize

rocks that fit the four inequalities. We take direction from CIPW but
delineate rock types by their mineralogy. We will not use the CIPW norm
itself to delineate them.

Nature can be perverse. The frequencies with which rocks characterized

by the four inequalities occur do not increase from top to bottom or
from bottom to top. Rather, rocks with low values of Al2O3 activity are
relatively uncommon but occur in a variety of settings and display a
wide range of silica activity. Rocks with high values of Al2O3 activity are
rare among volcanic rocks. Most volcanic rocks on Earth have medium
values of Al2O3 activity. Rocks with medium values contain the familiar
phases found in the most abundant of volcanic rocks: quartz, plagioclase,
sanidine, Ca-poor pyroxene, Ca-rich pyroxene, olivine, biotite,
amphibole, etc. Rocks found in this category are found everywhere from
the ocean floors to the tops of the most prominent of volcanoes. They are
the most abundant volcanic rocks on Earth.

Felsic, quartz-saturated, or near saturated, rocks with low values of Al2O3

activity lack plagioclase; the dominant feldspar is anorthoclase. Ca-poor
pyroxene and Ca-rich pyroxene are also absent; the pyroxene is Na-
rich. They also often contain the sodic accessory mineral, aenigmatite
(Nicholls and Carmichael, 1969). Most rocks with low values of Al2O3
have high activities of silica; some are saturated with quartz.

Less frequently mentioned in geoscience courses are rocks with both

low activities of Al2O3 and low activities of SiO2. Several volcanoes
in the East African Rift Valley have erupted to form rocks with these
characteristics. The range from felsic to mafic. Rocks with low activities
of Al2O3 occur as both fragmental deposits and as lava flows.

Rocks with ultra-low activities of Al2O3 contain unique assemblages:

Fe-rich sanidine, diopside, olivine, Mg-biotite (phlogopite), the rare
minerals, priderite (KTi8O16), and wadeite (K2ZrSi3O9) (Carmichael,
1967). These rocks are uncommon; they are found at only at a few
localities on Earth: Leucite Hills, Wyoming (Carmichael, 1967), Fitzroy
Basin, Western Australia (Prider, 1959), and Jumilla, Spain (Carmichael,
1967).

Plutonic rocks with high activities of Al2O3 exist. The characteristic

34
 The Nature of Volcanic Rocks

mineral in these rocks is a white mica. They are felsic and form from
quartz-saturated or near saturated melts. Comparable volcanic rocks are
not at all common. Topaz-bearing felsic rocks erupted in western United
States and Mexico (Christiansen, et al., 1986). They thought the topaz-
bearing volcanic rocks were alteration products of otherwise common
felsic flows with medium activities of Al2O3. Congdon and Nash (1988,
1991), however, found crystals of topaz in glass at an eruptive center in
Utah. The glassy rocks have compositions similar to melts inferred to
be the progenitors of intrusive rocks characterized by extremely large
crystals (pegmatites).

Rock and Glass Compositions

6. Chemical Composition: Many, if not most petrologists think the chemistry
of an igneous rock is its most important property. Petrologists have
expended an inordinate amount of energy and effort, time and resources,
intellect and thought on the acquisition and interpretation of the
chemistry of igneous rocks. Igneous rocks have been analyzed for major
elements, minor elements, trace elements, and isotopes in almost every
conceivable combination.

Chemical data have been used to identify rock types, infer magmatic
histories, and characterize the source regions where magmas originate.
For example, Carmichael (1962, 1963, 1964, 1965) made effective use of
chemical compositions of glassy rocks by analyzing both the rocks and
the glass. He used the rock composition to represent the original melt
and the glass composition to represent the melt after initial crystallization
but prior to eruption. From these data he could infer crystallization paths
in chemical space both forward and backward in time. Nicholls (2000)
and Nicholls and Stout (2008) used the composition of glass inclusions
in large olivine crystals to infer the presence and composition of more
primitive melts beneath Kilauea Volcano in Hawai’i. These inferences
corroborated an earlier inference by Wright (1971) that melts originating
in the source regions beneath Kilauea have MgO contents near 21%.
Russell and Stanley (1990) and Nicholls and Russell (1991) used major
and minor element abundances to delineate melting episodes and magma
batches among the recent historic eruptions of Kilauea Volcano. Halleran
and Russell (1990) used trace element chemistry to document magma
mixing in the 1969-1971 Mauna Ulu eruptions of Kilauea. More
recently, oxygen isotopes have been used to show that large explosive
eruptions of silicic magmas along the track of the Snake River Plain

35
 The Nature of Volcanic Rocks

A
UTM Grid: WGS84 Zone 12

Massive
Qrm
4810000 Rhyolite
Flow Layered
Qrf
Rhyolite
Sugary
Qrs
Qb Qrs Rhyolite
4808000 Basalt Flows Older
Qb
Than Rhyolites

Qrm
Fault
Qrf

4806000

334000 336000 338000

B
Geology from Spear and King (1982)

Figure 4: A. Big Southern Butte, a dome in the Snake River Plain,

Spear, D. B. and J. S. King (1982). The geology of Big Southern Butte, Idaho.
Idaho composed, atGeology
Cenozoic least ofinIdaho.
part,B.of glass. and
Bonnichsen In the
R. M.distance are two
Breckenridge,
other domes,Idaho
East ButteSurvey
Geological and Middle Butte
Bulletin. 26: The far one, East Butte is
395-403.
another glass-bearing dome. Middle Butte, the smallest one, is
covered by mafic to intermediate lava flows of the Snake River
Plain. A shallow glass or felsic intrustion pushed up the older lava
flows but did not breach the surface. B. Geologic map of Big
Southern Butte showing older lava flows caught atop the felsic
rocks when they punched their way to the surface.

from Oregon to Yellowstone National Park originated by partial melting

of hydrothermally altered, volcanic rocks (Bindeman, et al., 2008).

Rock chemistry furnishes the basis for identifying rocks in many

classifications. These include the original CIPW classification (Cross, et
al. 1902) and the latest classification adopted by the IUGS (Le Maitre,

36
 The Nature of Volcanic Rocks

1989, 2002). Many more examples of the uses of chemical compositions

could be cited.

Becoming proficient with the interpretations of chemical data is beyond

what we can do in a short Educational Tour. Perhaps one of the reasons
rock and magma compositions rarely receive recognition in the popular
science magazines is the time required to learn how to interpret chemical
data.

Writers of fiction often write extensively about the background of the

characters in their stories. Much of this background never makes its way
into the finished short story or novel but it does provide the writer of
fiction with a deep understanding of their characters. This understanding
shows in the quality of their story telling. In the same way, the more you
know about the volcanic phenomena and products you are writing about,
the more authoritative will be your finished product. So, even though you
may not include chemistry in your article, knowing even a few simple
ideas about the chemistry of volcanic rocks can lend your writing
credence and perhaps provide you with greater insight.

Glassy rocks are special cases in volcanic petrology. Volcanic glasses

and glassy rocks come in many varieties and are found in many settings.
Past large explosive eruptions, from Somma-Vesuvius to Krakatoa,
Mount St. Helens, Mount Pelée, and Mount Merapi are among the
largest and most dangerous on the planet. Presumably, similar eruptions
will happen in the future. Their products contain significant amounts of
glass. Fragmented melt particles, rapidly chilled to a glass, figure
prominently (Figure 2F). Glassy lava flows form large domes in many
tectonic environments. The domes of Mono Craters, California, erupted
in a Basin and Range setting. They occur in Iceland on the Mid-Atlantic
Ridge. Large glass-bearing domes erupted in the middle of the Snake
River Plain, Idaho (Figure 4).

For many compositions, glassy rocks provide the best approximations we

have to natural melts. By definition, they have not completely
crystallized. Potentially, on a different cooling path, they could have
crystallized to a mixture of minerals. Because they didn’t completely
crystallize, the question arises, what would their mineral assemblage
have been had they done so?

37
 The Nature of Volcanic Rocks

100
SiO2-rich

Wt% SiO 2 80

60
Intr
SiO2-
poor

40
Craters of
SiO2-poor

the Moon
Ultra

Llangorse
20 Mountain

0
0 20 40 60 80 100
Mafic Index
Felsic Intermediate Mafic UM

Figure 5: Comparison of a general chemical and mineralogical

classification. Orange fields mark one-to-one correspondence
between the two classifications. Sand colored fields represent rocks
that could be incorrectly identified in one or the other classifications.
There should be few if any igneous rocks with compositions that
would fall in the light yellow fields. Chemical analyses are precise
and presumably accurate. Mineral abundances, however, can be
subject to systematic error. Mineral abundances estimated with
with the microscope generally over-estimate the colored, high-relief
minerals. The mineral abundances represented on the figure were
done with non-optical methods and should be free of systematic
error. Consequently, they plot to the left of the orange fields, which
were established with mineral abundances estimated with the
microscope. The translucent area delineates the expected values
of SiO2 content and mafic index (see text). UM = Ultramafic, Intr =
intermediate.

38
 The Nature of Volcanic Rocks

Petrologists have been looking for an instrument or method of

calculation that will tell them what the completely crystalline equivalent
of glassy rock would be. At some level of detail, between mafic index
and isotopic distribution, every instrument or method will fail. Like the
fountain of youth, a perfect method for determining the completely
crystalline equivalent of a glassy rock doesn’t exist. Magmas actually
follow a contingent cooling path that results in the glassy rocks beneath
our hammer. Not all cooling paths lead to the same rock. A different
cooling path produces a different rock. A multiplicity of cooling paths
produces a multiplicity of rocks. Which of the several potential
crystalline rocks should we designate as the crystalline equivalent?

In spite of the impossible nature of our quest, we still want a way to

place the glassy rocks in context with those completely crystalline. Since
the 19th century, petrologists have used SiO2 content to characterize
volcanic rocks. Like many, if not most schemes for chemically
characterizing rocks, this one has generated its share of acrimony and
controversy. Rocks with lots of silica, the early petrologists called acid;
those with little silica, they called basic. They were under the impression
that silicic acid, H4SiO4, was a major component of igneous rocks.
Although we know today that such is not the case, the names have
persisted even though hydrogen ions are hard to find in volcanic rocks.
Many geoscientists have discarded the terms acidic and basic as rock
descriptors because they are literally meaningless, although, metabasite
seems perfectly acceptable to some metamorphic petrologists. I guess
metamorphic rocks can defy convention.

Names for ranges of silica contents carried by volcanic rocks are

convenient and useful when talking about their compositions. In an
attempt to be politically correct, I’ve used the terms SiO2-rich and SiO2-
poor instead of the traditional acid and basic. These substitutions do
avoid a bit of the jargon that plagues igneous petrology. Our primary use
of SiO2 content as an indicator of rock type will be to relate glassy rocks
to similar ones that are completely crystalline.

Figure 5 shows a plot of SiO2 content versus mafic index. In general,

completely crystalline felsic rocks are SiO2-rich and mafic rocks are
SiO2-poor. In an ideal world there would be a one-to-one correlation
between SiO2 content and mafic index with the correlation expressed by
a line with a negative slope on Figure 5. The world of volcanic rocks is

39
 The Nature of Volcanic Rocks

not ideal and a one-to-one relationship doesn’t exist. The best we can
expect is a general trend with a negative slope. Data from two volcanic
rock suites are plotted on Figure 5 and they do display general trends
with a negative slopes. A mafic index can be derived from the observed
distribution of minerals in glassy rocks. Often, mafic minerals such as
olivines, pyroxenes, and Fe-Ti oxides dominate the early crystallizing
assemblage. Such rocks will have a high mafic index but may have a
SiO2-rich composition. In such cases the mafic index is a poor indicator
of rock type. Although imperfect, Figure 5 is about the simplest
instrument you can find to infer a mafic index for a glassy rock.

Often, the crystals in a glassy rocks, and most glassy rocks contain at
least a few crystals, will constrain what is possible about the nature of
such rocks. For example, if the crystals are green pyroxenes, aenigmatite,
and anorthoclase, then the rock almost certainly has a low activity of
Al2O3. Likewise, if the mineral assemblage includes feldpathoids,
nepheline, or leucite, the melt has a lower silica activity than a quartz-
bearing rock or a rock with a Ca-poor pyroxene. In other words, don’t
ignore the minerals in glassy rocks; they can sometimes provide you
with more information than rock chemistry alone can. Put the two
together and you will have a better understanding what the rock is made
of and what the volcano is likely to do next.

We can use a bit of judicious jiggery-pokery to make Figure 5 slightly

more useful. Felsic rocks have silica percentages from somewhere in the
low to middle 80%’s to perhaps 60%. A rock composed of anorthite,
CaAl2Si2O8, which would be felsic by definition and contain 42.36%
SiO2, marks an extreme lower limit. Such a volcanic rock, which doesn’t
exist here on Earth, would have the smallest percentage silica we could
possibly expect in a felsic rock. On the other side of the diagram, at the
ultramafic edge, we would expect rocks with melilite, nepheline, or
monticellite as the principal mineral constituents. These minerals all
have silica contents near 40%. Consequently, a lower limit to the space
where we would expect Earth-bound rocks to fall would be a line
between 42% on the felsic edge and 35% on the ultramafic edge. An
upper limit would be a line between 83% on the felsic side and 35% on
the ultramafic side. These lines define an area in silica percentage – mafic
index space where most terrestrial rocks will happen. If you know the
silica percentage in the glass or glassy rock, then you can fix a range for
the expected mafic index of the glass or glassy rock. I would not try to be

40
 The Nature of Volcanic Rocks

Figure 6: Rock space defined by the three variables: activity of

SiO2, activity of Al2O3, and mafic index (see text). The orange
volume is that portion of rock space occupied by the common
volcanic rocks and that can be made from the minerals listed in
Bowen’s reaction series.

more precise than felsic, intermediate, mafic, or ultramafic with my

guess. I might not even try to go that far.

If the glass or melt composition is available, the activity of silica can be

calculated for particular pressure-temperature conditions from a
thermodynamic model, for example the model developed by Ghiorso

41
 The Nature of Volcanic Rocks

and Sack (1995). But having done that, it could become a problem
fraught exercise, trying to place the calculated value into the
mineralogically defined framework. Even though we initially constructed
the activity of SiO2 scaffold in a quantitative fashion, the ordering of
mineral assemblages is qualitative, independent of numbers. Because the
curves shown on Figure 3 and listed in Table 2 have acquired their
numerical values for compositionally pure end member phases, the
calcaulated activity of SiO2 values derived from a glass or melt
composition can be inconsistent with the numerical values for the curves.

Summary, Part 2
If you have arrived at this point, then you should be able to discuss the
geologic context of a given volcanic rock, describe its textural features,
identify the critical minerals it contains, and describe its major chemical
features. You should have the background to read and assess articles on
volcanoes and their products in the popular science literature, take part in
geoscience outreach programs, and talk about volcanic phenomena with
confidence and without misconceptions, and you should be able to write
accounts of volcanic phenomena suitable for popular science magazines.
Texture serves to distinguish fragmental rocks from those rocks that form
from flowing melts and their included crystals. Placing volcanic rocks in a
geologic context connects them with tectonic settings.
Figure 6 is a schematic diagram of SiO2 activity, Al2O3 activity, and mafic
index. It is a graphical depiction of rock space. Terrestrial rocks occupy a
large part of the space defined by the three variables. The criteria we have
developed to describe volcanic rocks extend to most of the volcanic rocks
found on our planet.
Here would be an appropriate place to do Exercise 4: Compare Volcanoes,
Part 2. For Exercise 4, you are asked to explain your comparison of
Nyiragongo and Vesuvius - Somma in an essay. After doing Exercise 2 and
4, what can you say about the connections between rock types and eruptions
at Nyiragongo and at Vesuvius? You may have discovered several features
that distinguish the products and processes of the eruptions from the two
volcanoes. Examples of such discoveries might be:
Nyiragongo produced lava flows; Somma – Vesuvius produced high-
density, fragmental flows or ash flows and ash falls.
Nyiragongo produced small amounts of ash; Somma – Vesuvius produced
more than enough ash to collapse the roofs of the buildings in Pompeii.
42
 The Nature of Volcanic Rocks

Nyiragongo sits in a rift valley where a tectonic plate is splitting apart;

Somma – Vesuvius erupts in a plate tectonic setting that is among the
more complicated on Earth.
Expanding the scope of our questions, we can ask: Do the lava flows from
Nyiragongo behave differently from the historic flows on Hawai’i? What
makes them unique? How about differences between the rocks generated
by eruptions from Vesuvius and from other explosive eruptions, such as
the Mount St. Helens eruption in 1980? Ultimately, these are the kinds of
questions you want to answer.

43
The Nature of Volcanic Rocks

44
 The Nature of Volcanic Rocks

Part 3: Piecing it Together

Rocks, Melts and Volcanic Rock Names (Optional)

As promised, this Educational Tour contains no rock names, well almost
none. As mentioned, writing about the percentage of Fe-Mg bearing minerals
in a rock without the felsic - mafic concept is impossible.
Conversations with friends and colleagues convinced me that petrologists in
particular and geoscientists in general don’t want to give up their rock names.
They find them a source of comfort and support. This, in spite of how much
they decry the chaos and confusion of rock names. In this optional section,
rock names are applied to rocks that fall in the orange region displayed on
Figure 6. The coverage doesn’t pretend to be comprehensive. Rather, the
common rocks will be placed into the rock space (Figure 6). In addition,
some particular rock types that have been the center of controversy and
contention will be placed in rock space (Figure 6).
The characteristics of the common rocks, basalt, andesite, and rhyolite fall
in the orange region on Figure 6. It is but a small part of the total rock
space available. The common rocks span the range from felsic to mafic but
crystallize from melts that have activities of SiO2 that fall in a small part of
the range of SiO2 activity occupied by volcanic rocks. They also all fall in
the medium range of Al2O3 activities. The critical minerals for these common
rocks include those in the top four categories on Table 1. The critical felsic
minerals are quartz and feldspars. The critical mafic minerals are Ca-poor
pyroxene or Fe-olivine at high SiO2 activities and olivine plus Ca-rich
pyroxene at the low end of SiO2 activity.
Basalt is usually described as the mafic representative of the set. Many
basalts, however have an intermediate mafic index rather than a mafic
one. Andesites are among the best known of the common rocks with an
intermediate mafic index. It is only one of many rocks with an intermediate
mafic index, however. Rhyolites are the common felsic rocks. The have SiO2
activities that range from quartz-saturation to the olivine - Ca-poor pyroxene
boundary,
The common rocks come from melts with medium Al2O3 activities.
Technically, these rocks are called metaluminous. Melts that have lower
Al2O3 activities crystallize peralkaline rocks. Rocks from melts with higher
Al2O3 activities are called peraluminous.
The important point here: Many people, including geoscientists, suffer

45
 The Nature of Volcanic Rocks

under the misconception that the three rock types, basalt, andesite, rhyolite,
adequately represent the varieties of rock types on Earth. They do not. The
most frequently encountered volcanic rock rocks can, perhaps, be adequately
labeled with the three names, at least to a first approximation. At many
volcanic centers, however, the three names are not adequate labels. This begs
the question: Do the differences between those adequately labeled and those
inadequately labeled make any difference? Imagine standing in front of an
on coming lava flow from Mauna Loa on Hawai’i or standing in front of an
on coming lava flow from Nyiragongo in East Africa. The Mauna Loa flow
you might outrun. The Nyiragongo flow you might not unless you’re faster
than a herd of speeding elephants. Knowing the difference in the mineralogy
expected in the melts and rocks from the two volcanoes might make a
difference in how you would react. The differences are important and calling
both rock types basalt is misleading.
In the early part of the last century, the Geological Survey of Scotland
mapped the British-Arctic volcanic province. The authors of the memoir on
Mull (Bailey, et al., 1924) developed the concept of magma types. Chemical
compositions of the rocks were used to define them. From their analytical
data they defined four such types. Kennedy (1933) concluded that two
different magma types that crystallized to mafic rocks could be recognized
world wide. These he called the tholeiitic magma type and the olivine-basalt
magma type. The names applied to these concepts have suffered minor
variations over the years: tholeiite, tholeiite basalt, tholeiitic basalt, alkali
olivine basalt, alkali basalt. Both chemical and mineralogical criteria have
served to distinguish the rocks that crystallize from the two magma types (for
example, Le Maitre, 2002).
The CIPW norm is used by many petrologists to distinguish rocks that form
from the two magma types. Tholeiitic basalts have compositions that produce
normative hypersthene (= Ca-poor pyroxene). Alkali olivine basalts have
compositions that produce normative nepheline. By its very design, the
CIPW norm calculation cannot return a result that contains neither normative
hypersthene nor normative nepheline, subject to one miniscule caveat. The
calculated result must contain at least one or the other mineral formulae. In
the real world of melts, magmas, and mafic volcanic rocks, nepheline and
Ca-poor pyroxene do not crystallize together, which seems in concordance
with the CIPW norm. In the real world of melts, magmas, and mafic volcanic
rocks, however, mafic rocks exist that contain neither nepheline nor Ca-
poor pyroxene and they are abundant. The CIPW norm and petrologic
tradition recognizes two groups of common mafic rocks. The scheme for

46
 The Nature of Volcanic Rocks

A B
Figure 7: A, Explosive eruptions and the SiO2 activities of the
melts that produced them. The eruptions at Kilauea, Eyjafjallajökull,
Ubehebe Crater, and Lunar Crater were caused by SiO2-poor
magmas (See Figure 5). B, SiO2 activities for melts that produced
lava flows. Obsidian Dome is part of the Mono Craters center,
California

distinguishing rocks developed here distinguishes three: those with Ca-poor

pyroxene, those with nepheline, and those with neither. For the latter, Stout,
et al. (1994) suggested the label olivine basalt. The three groups would then
be labeled tholeiitic basalt, olivine basalt, and alkali olivine basalt.
Olivine basalts are found in western North America from central Yukon
Territory, through British Columbia, and in the western United States at least
to within a few kilometres from the US – Mexico border. The Snake River
Plain, in particular, is covered by such rocks with the mineral assemblage
plagioclase, Ca-rich pyroxene, olivine, and Fe-Ti oxides. In Hawai’i, where
most workers label the recent products of Kilauea tholeiites, a large number
of lava flows lack both Ca-poor pyroxene and nepheline.
The point of this discussion: learn to characterize volcanic rocks by the
minerals they contain. Select a set of critical minerals that reflect the controls
imposed by chemistry and physical chemistry on melt and magma behavior

47
 The Nature of Volcanic Rocks

and then find names appropriate for the rocks. I expect you will find rock
names less arbitrary and confusing than if you learn the names and try to
make the rocks fit.
And the miniscule caveat? It is theoretically possible, but highly unlikely,
that a rock composition could exist that contains the exact amount of SiO2
(to the nth decimal place) needed to convert all the normative hypersthene to
normative olivine and leave no deficit in SiO2. There would then be no reason
to convert some normative albite into normative nepheline. The CIPW norm
would then contain neither normative hypersthene (= Ca-poor pyroxene) nor
normative nepheline.
If you have come this far, you should have a pretty good idea about the basic
features that separate one volcanic rock from another. Geologic context,
texture, mafic index, and mineral assemblage are the data we need to
characterize a rock or the products of a volcanic eruption. But what if some
one hands you a volcanic rock or you pick up an article about a volcanic
eruption? Where do you find the data you need to characterize the rocks or
the volcanic products? If you’re reading an article, there’s a good chance
the literature can provide you the data you need to characterize the products
of an eruption. If some one hands you a rock, however, you will have to
find out where it came from, determine its texture, mafic index, and mineral
assemblage.
We can use the scaffolding of features to look for in volcanic rocks to
compare volcanoes, eruptions, and volcanic products in novel ways. Figure 7
shows the relative silica activities for magmas erupted at several volcanic
centers. The eruptions differ in their mode of eruption: explosive or quiet. We
can distinguish one from the other by examining the textures of the products,
Explosive eruptions produce fragmented products. Quiet eruptions produce
lava flows.
Figure 7A shows where the melts that produced several explosive eruptions
fall on a scale of SiO2 activity. Not all explosive eruptions come from
fragmented, quartz-saturated melts. The large eruptions, at least those that
formed the chain of calderas that culminated in the Yellowstone supervolcano
are quartz-saturated (Christiansen, 2001). So was the melt that erupted from
Mount St. Helens in 1980, although it was barely at saturation (Kuntz, et al.,
1981). The melt that created the eruption of Krakatau in 1883 was saturated
with Ca-poor pyroxene but not quartz nor olivine (Symons, 1888). The iconic
representative of explosive eruptions is the eruption of Vesuvius-Somma in
the year 79. Surprisingly, the melt that erupted to destroy Pompeii was under

48
 The Nature of Volcanic Rocks

2.4
HM-4
HM-12 1175C
1178C
HM-15 HM67
2.0 1181C
1182C
HM-2
1178C

1.6
ln 

1.2
HI-12
1277C

HI-14
0.8 1302C

0.4 HI-3
HI-2 1375C
1367C

0.0
-0.86 -0.84 -0.82 -0.80 -0.78 -0.76 -0.74
ln aSiO2

Figure 8: Plot of the natural logarithm of viscosity (Giordano, et

al., 2008) versus the natural logarithm of SiO2 for melts with the
compositions of lava flows erupted during the 1968 eruption of
Kilauea Volcano, Hawai’i. The more mafic or most SiO2 poor
melts fall at the lower left part of the diagram. Melts equivalent to
the more felsic lava flows fall at the upper right. The temperatures
listed were calculated for the first crystallization on cooling at low
pressure (Ghiorso and Sack 1995).

saturated with SiO2. Leucite crystallized early, before eruption (Scaillet, et

al., 2008).
A common misconception states that mafic melts produce quiet eruptions and
felsic ones produce explosive ones. Like many generalizations, the assertion
is not always true. The 1790 eruption of Kilauea Hawai’i produced an ash

49
 The Nature of Volcanic Rocks

cloud that covered a large part of the Island of Hawai’i. The total thickness
of the ash layer ranged from over a metre at the caldera on the summit of the
volcano to zero some kilometres away (McPhie, et al., 1990). The eruption
killed several Hawaiians who were crossing the volcano. Kilauea magmas
typically produce mafic to intermediate rocks that crystallize olivine but
most of which lack Ca-poor pyroxene and in which nepheline has not been
detected. It is not as silica under saturated as was the Vesuvius melt.
The very recent eruption by the volcano with the unpronounceable name,
Eyjafjallajökull, in southern Iceland was fueled by a magma unsaturated with
SiO2 but closer to saturation than Vesuvius.
Explosive eruptions can also be generated by interactions of ground water
and magmas. Two such eruptions between SiO2-poor magmas and ground
water occurred during the geologic Recent in the U.S. southwest: Lunar
Crater, Nevada (Scott and Trask, 1971) and Ubehebe Crater, California
(Figure 7A). The craters lie close to the great circle paths followed by
airlines between Los Angeles and Calgary, Alberta, Chicago, and Halifax,
Nova Scotia. Would repeat performances in the same region disrupt air travel
in the U.S. as Eyjafjallajökull did for Europe?
Figure 7B shows the SiO2 activities for three eruptions that produced lava
flows or lava domes. The lava flows from Nyiragongo and Kilauea have been
likened because they look much the same in outcrop. The are black, have
pahoehoe surfaces and erupted from volcanoes. The flows from Nyiragongo
moved so rapidly that they trapped automobiles on the streets of Goma, a
city 60 km from the volcano and, as legend has it, they trapped and destroyed
an elephant herd caught in the path of a flow. The flows from Kilauea move
much slower than the flows from Nyiragongo. Figure 7B shows that the
flows from Kilauea and the obsidian dome are closer in the space defined by
activity of SiO2 than are the flows from Kilauea and Nyiragongo.

Silica Activity and Physical Properties

Igneous petrologists are obsessed with silicate melts. They experiment
with them, measure their properties, and analyze them for their chemical
constituents. They then take the data and develop models relating the physical
properties they measure on the experimental melts to melt composition,
temperature and pressure. Two of the best constrained models are the
thermodynamic model for silicate melts developed by Ghiorso and Sack
(1995) and the viscosity model developed by Giordano, et al. (2008). If
we have a rock analysis, then with these two models we can calculate the

50
 The Nature of Volcanic Rocks

Figure 9: Ol Doinyo Lengai volcano, northern Tanzania. The

volcano sits just east of the west wall of the East African Rift Valley.
The cliffs on the left side of the photograph form that wall. The edge
of the wall is approximately 5 km from the top of the volcano. The
volcano has erupted both sodium carbonate and silicate magmas.

temperature at which the equivalent melt starts to crystallize, its activity of

SiO2 at that temperature, and its viscosity.
Viscosity is one of the most important properties of volcanic melts; some
would claim it is its most important property. Our method of characterizing
volcanic rocks makes it relatively easy to infer viscosity from rock character.
Figure 8 is a plot of viscosity versus activity of SiO2 for melts that have
the compositions of lava flows formed during the 1968 eruption episode of
Kilauea Volcano, Hawai’i. The compositions and mineralogy of these rocks
are consistent with their derivation from a single magma batch (Nicholls
and Stout, 1988, 2008). Figure 8 illustrates several general features about
the viscosity of volcanic melts. Viscosity decreases with temperature and
increases with SiO2 content. We infer that the dominant chemical control on
viscosity is the activity of SiO2 because of the strong correlation between the
two variables shown on the graph. Activity of SiO2 is not the sole control,
however, because there is scatter about the trend. Viscosity depends on
the concentration of the other constituents in the melt, especially H2O and
fluorine (Giordano, et al., 2008). If you stirred, poked, and tasted the sugar-
water solutions you made for Exercise 1, especially after cooling them in the
refrigerator and freezer, you found that viscosity correlates positively with
sugar content and the activity of sugar. In silicate melts, viscosity increases
with silica activity.
The activities of SiO2 were calculated by introducing the rock compositions
(see references cited by Nicholls and Stout, 2008) into the thermodynamic

51
 The Nature of Volcanic Rocks

model developed by Ghiorso and Sack (1995). The viscosities were then
calculated for pressures equivalent to the Earth’s surface with the model
constructed by Giordano, et al. (2008). Temperatures for both calculations
were chosen to correspond to those at which the melts would initially start to
crystallize on cooling.

Two Volcanoes of the East African Rift Valley

Near the border between Tanzania and Kenya, along the west side of the
East African Rift Valley, a bunch of volcanoes erupted magmas that created
strange volcanic rocks.
One of them, Mount Suswa, rises some 550 m above the Rift Valley floor
about 50 km west and north of Nairobi, Kenya. The crystalline rocks of
Mount Suswa are felsic; some contain glass with SiO2 contents near 58%.
The rocks contain nepheline, indicating low activities of SiO2, and minerals
that indicate low activities of Al2O3: anorthoclase, Na-rich pyroxene, and
aenigmatite (Nash, et al. 1969). Both fragmental and lava flows make up the
volcano (Johnson, 1969).
Ol Doinyo Lengai (Figure 9), a volcano adjacent to the west wall of the East
African Rift Valley in northern Tanzania, is, perhaps, the strangest active
volcano on the planet. It erupts both sodium carbonate magmas and silicate
magmas. Some of the neighboring volcanoes also erupted both kinds of
magmas in the past (Peterson, 1989). The rocks of Ol Doinyo Lengai that
formed from some of the silicate magmas have activities of SiO2 even lower
than Suswa. Nepheline is commonly present and is accompanied by melilite
in some rocks (Peterson, 1989). I’ve looked at rocks from Ol Doinyo Lengai
that contain titanite but no feldspar. Green-brown Na-pyroxene accompanies
nepheline, titanite, and biotite. The absence of feldspar in the titanite-
bearing rocks that I looked at suggests that in rocks with low activities of
Al2O3, the feldspar-feldspathoid boundary in activity of SiO2 space doesn’t
coincide with the titanite-peroskite boundary. Rather, feldspar may cease to
crystallize at higher activities of SiO2 in melts with low activity of Al2O3 than
in melts with intermediate activities of Al2O3; a suggestion that needs to be
corroborated or refuted.
The mafic indices for the rocks that I’ve looked at from Ol Doinyo Lengai
range from intermediate to ultramafic. In addition to lava flows, ash beds
containing nepheline and Na-pyroxene were deposited at the base of the
volcano. The samples I collected contain no glass. According to a Maassai
guide, the latest eruptions in 2007-2008 deposited approximately a metre of

52
 The Nature of Volcanic Rocks

ash on the top of the cliff forming the west wall of the Rift Valley some 5 km
from the summit crater (see Figure 9).

What Rock Types Have We Left Out?

A few volcanic rock types, however, do not fit comfortably into the space
shown on Figure 6. A few lava flows that erupted on the Snake River Plain
contain mineral assemblages that would fit the space on Figure 6 except they
lack Ca-rich pyroxene (Stout and Nicholls, 1977). These lava flows have
higher FeO contents than the majority of similar appearing flows in the area.
They contain approximately 14% FeO whereas the similar appearing flows
contain approximately 10% FeO. The more abundant mafic to intermediate
lava flows of the Snake River Plain are characterized by the assemblage
plagioclase, Ca-rich pyroxene, olivine, and Fe-Ti oxides. They contain
approximately 10% FeO. In north-central British Columbia, the Nisga’a
volcano, another FeO-enriched lava erupted around 1750 - 1775 (Wuorinen,
1978, Symons, 1975). This flow contains approximately 12% FeO but does
carry Ca-rich pyroxene (Nicholls and Stout, 1994). One speculates that
the pyroxene-free, high-iron lava flows of the Snake River Plain may have
crystallized at an activity of FeO where Ca-rich pyroxene doesn’t crystallize.
Sulfide magmas likely occur within the Earth. Textures suggestive of
crystallization from a sulfide melt have been found in ore deposits. These
rocks don’t fit into Figure 6. I know of no volcanic examples here on Earth
but Jupiter’s moon, Io, may be erupting sulfide-rich melts. The sodium
carbonate rocks from Ol Doinyo Lengai and a few other volcanoes in
Northern Tanzania don’t fit into the space depicted on Figure 6 either.
There is a group of rocks, usually found in small, shallow intrusions but
occasionally as lava flows (Nicholls, 1969), that contain mafic minerals
bounded by crystal faces in a matrix of felsic minerals, which lack crystal
faces. Different rock types in the group are distinguished by different mafic
mineral assemblages in which pyroxene, biotite, or amphibole dominates.
The mafic minerals are contained in a felsic matrix of smaller crystals.
Although these rocks, usually labeled as lamprophyres, can be fit into the
scheme shown on Figure 6, texture, mafic mineral assemblage, and geologic
context have traditionally been thought to be the important criteria for
delineating members of this rock group.
The felsic to intermediate rocks with high activities of SiO2 and medium to
high activities of Al2O3 have received considerable attention from petrologists
and other geoscientists over parts of the last three centuries. Consequently,

53
 The Nature of Volcanic Rocks

several different rock types are recognized in this part of rock space. The
plutonic bodies made from these kinds of rocks are important components
of many, perhaps most, mountain belts. These are the granitic rocks students
in introductory geoscience courses learn. The different rock types are
distinguished by the relative proportions of alkali feldspar (those that contain
Na and K) and plagioclase (see Le Maitre 2002, or Philpotts and Ague, 2009).
It is straight forward to add another criterion, plagioclase-alkali feldspar ratio,
to the ones that describe rock space (Figure 6).
Here would be an appropriate place to do Exercise 5: Write a Review
of an Article. The article that you’re to review is about the Yellowstone
supervolcano. It appeared in the August, 2009 issue of National Geographic.
A journalist wrote it with contributions from a photographer and an artist
(Achenbach, et al., 2009).

What Else is Part of Volcanic Petrology?

Volcanic petrologists get their kicks by inferring magmatic histories, even
when the story is incomplete. The chemistry and mineralogy of the rocks tells
them what they can say about magmatic histories. Chemical imprints come
from compositions and processes in the source region, interactions between
melts and the rocks through which the magmas travel, and by separation
and segregation of crystals and fluids that form in the melt. Thanks to the
studies originating with Bowen (1915, 1928), imprints caused by separation
and segregation of crystals frequently receive early attention in petrologic
studies. Accounts and discussions of crystal separation and segregation
can be found in petrology textbooks under the name crystal fractionation.
Among the more recent texts are Best and Christiansen (2001), Philpotts and
Ague (2009), and Winter (2001). Older texts with many insightful passages,
which petrologists should read, include Carmichael, et al. (1974) and Turner
and Verhoogen (1960). Finally, the classic text by Bowen (1928) is one all
igneous petrologists should read and study.
If your education included an introductory geoscience course, you may have
heard about Bowen’s reaction series. It is an ordered list of minerals found
in mafic to felsic igneous rocks with medium activities of Al2O3. The volume
of rock space shown in orange on Figure 6 is the fraction of rock space
occupied by rocks that contain the minerals Bowen listed in his series. This
is the same volume of rock space that contains the common volcanic rocks.
As we have seen, this small volume is just a part of the space occupied by the
diversity of volcanic rocks found on Earth. Bowen postulated that the melts
that produce rocks that fall in this small volume can be related by crystal

54
 The Nature of Volcanic Rocks

separation and segregation of the minerals in his reaction series. If your

educational background lacks knowledge of Bowen’s reaction series, you can
find what he said about it in his book (Bowen, 1928, Chapter V).
Sometimes further research fails to corrobrate the crystal separation and
segregation hypothesis and we have to turn to the other two sources of
chemical variability: compositions and processes in the source region and
interactions between melts and the rocks through which they travel.
So, you decide to look further into the origin of a set of rocks. How should
you plan your study? Simultaneously looking at all three sources of chemical
variation can be daunting. Our Educational Tour provides the information
we need to select the most promising avenue for additional study. If the
rocks are close together in rock space, they will more likely be related by
crystal separation and segregation and less likely to have inherited significant
chemical differences from different compositions and processes in the source
region or by interactions between melts and the rocks through which they
travel. The farther apart magmas fall in rock space, the less likely they can
be related by separation and segregation of minerals. Proximity in rock space
is not an unequivocal indicator of modification by crystal separation and
segregation. Proximity only suggests that modification by crystal separation
and segregation is the most promising hypothesis to examine first.
Consider two volcanoes in the East African Rift Valley where it cuts across
the Democratic Republic of the Congo: Nyiragongo and Nyamuragira. They
are only 13 km apart but the lava flows they erupted are distinctly different.
Nyiragongo lava flows are among the most fluid on Earth. They contain
nepheline, leucite, kalsilite, melilite, olivine, and ratty-looking, corroded Ca-
rich pyroxene. The lava flows erupted by Nyamuragira contain plagioclase,
leucite, olivine, and Ca-rich pyroxene (Verhoogen, 1948). We expect the
melts from Nyamuragira to have higher activities of SiO2 than melts from
Nyiragongo because of the lack of feldspar in the Nyiragongo melts and its
presence in the melts of Nyamuragira (Figure 7). I would look for evidence
of different kinds of source regions and compositions for melts from the two
volcanoes before looking for differences between the melts caused by crystal
separation and segragation.
To paraphrase Jane Austen: It is a truth universally acknowledged, that
stratovolcanoes consist of interlayered fragmental products and flows
whereas shield volcanoes consist mostly of flows produced by quiet
eruptions. The flows in stratovolcanoes are thought to form from melts more
viscous than the melts producing the flows of shield volcanoes (compare

55
 The Nature of Volcanic Rocks

the information in http://en.wikipedia.org/wiki/Stratovolcano with the

information in http://en.wikipedia.org/wiki/Shield_volcano).
Yet, Nyiragongo is considered a stratovolcano (http://en.wikipedia.org/wiki/
Mount_Nyiragongo) whereas Nyamuragira is considered a shield volcano
(http://en.wikipedia.org/wiki/Mount_Nyamuragira). Both the observed
characteristics of the lava flows and their locations in rock space (Figure 7)
are consistent with inferences that the viscosities of the Nyamuragira melts
are larger than the viscosities of the Nyiragongo melts. Perhaps controls on
volcano shape are more complex than we think.
Our unorthodox Educational look at volcanic rocks takes us from what
the rocks contain to melt behavior to speculations about the reasons for
the shapes of volcanoes. Although many geoscientists think volcanoes
and volcanology are esoteric subjects suitable only for esoteric scientific
studies, there exist practical reasons for knowing about them. Exercise 6:
Develop a Volcanic Hazard Assessment will introduce you to one such
practical reason. Like all attempts to predict natural phenonmena, such as
climate change, hurricanes, earthquakes, and tsunamis, in time to give people
warning of danger, predicting volcanic eruptions is an uncertain undertaking.
How does our Education Tour inform your ideas about volcanic hazards.
Exercise 6 gives you an opportunity to do just that, predict what a volcano
will do and tell us how and when you would warn people of danger.

56
 The Nature of Volcanic Rocks

Exercises

Introduction
The Exercises for the Tour send you off into the world of volcanic petrology.
It’s a world unknown to most people including many geoscientists. People
study, read about, and walk on volcanoes but lots of them ignore the
rocks themselves. Here, you’re asked to connect the rock types to mineral
assemblages, rock chemistry, and a few melt properties. To help embed your
hard-won knowledge into your long range memory, you are encouraged to
present your results with essays and outlines.
Essays have features in common with many short written works: magazine
articles, opinion pieces, or short stories. All the features can be present in
the same piece of writing. A magazine article often conveys information,
an opinion piece manipulates reactions, and a short story entertains. Your
essays should do at least one of these things. Some essays might be more like
opinion pieces than entertainment whereas others might be more about facts
and their interconnections.
All four types of written work: magazine articles, opinion pieces, short
stories, and essays share a common structure. They all have a beginning, a
middle, and an end. The several kinds of written work may have different
names or labels for these parts but they all come down to the same thing:
beginning, middle, and end. In an essay, for example, the parts may be
labeled introduction, body and conclusion. Further, these labels may be used
as subheadings. A short story, on the other hand, will not have subheadings.
But the three-part structure will still be there.
All four types of written work have their own special problems and
requirements. If you write an essay that follows one kind more than
another, you will find yourself constrained by these special problems and
requirements. Most scientific and technical papers are essays. Science papers
in biology and chemistry that report the results of laboratory experiments
often follow a pattern distinguished by an acronym: IMRAD. IMRAD stands
for Introduction, Methods, Results, and Discussion. The Introduction is
the beginning. Methods and Results are the middle. The Discussion is the
end. Papers about topics in the observational and historical sciences seldom
precisely fit the IMRAD model but they will still have a beginning, middle,
and end. In short, each paper tells a story.
Authors of scientific essays face seemingly contradictory problems. Each
section must be complete in itself but the collection of sections must still

57
 The Nature of Volcanic Rocks

tell a coherent story. You will come to appreciate papers that were written to
these standards. If you are seeking information about a particular concept,
you will be impatient if you have to read both the middle and the end to find
the desired information. The introduction should have all the information you
need about the state of knowledge concerning a particular topic. But if the
author hides some of this introductory material in the end portion, it wastes
your valuable time. On the other hand, those who read the entire article will
appreciate a coherent story from beginning to end.

Exercise 1: Modeling Volcanic Melts

Melts make igneous rocks igneous. If we can understand melts, we gain
insight into the processes that form igneous rocks. We can see melts at active
volcanoes and observe chilled versions of melts in glassy rocks. Sugar-water
solutions provide analogues for silicate melts. Like all analogue comparisons,
the correlation is not perfect between the analogue and the real melt but the
comparison does deliver some insight into the behavior of the melts that
come out of volcanoes.
A kitchen stove, a refrigerator with a freezing compartment, and a few table
glasses are all the equipment required.
First, make three sugar-water solutions. Make one a saturated solution by
adding 150 ml of sugar to 100 ml of water. You may have to heat the mixture
to get all the sugar into solution. On cooling, some solid sugar should
crystallize, confirming the solution is saturated. Add 80 ml of sugar to 100
ml of water for the next solution and 25 ml of sugar to 100 ml of water for
the last. Examine the three sugar solutions for their taste, viscosity, and any
other physical property you can think of. Next, put them in the refrigerator,
which should be at a temperature a few degrees above 0°C. After twenty
four hours or so, again check the viscosity. Do the solutions flow as readily
as they did at room temperature? Have any of them started to crystallize? If
so, which ones? Finally, put the solutions in the freezing compartment for
approximately twenty four hours. What has happens after this last treatment?
Almost all volcanic melts arrive at the Earth’s surface carrying crystals. In
other words, nearly all volcanic melts arrive at the Earth’s surface saturated
with some minerals. Many melts crystallize completely as they move across
the surface as lava flows. Others only partly crystallize; they have some
glass in them. Some others don’t crystallize any further. Their viscosity
increases so much that new crystals can’t form and the melts chill to a glass.
Which of the solutions you made mimic magmas? Which ones mimic rocks?

58
 The Nature of Volcanic Rocks

Include the temperature conditions in your assessment: room temperatures,

refrigerator temperatures, and freezer temperatures. Write an essay comparing
the behavior of the sugar-water solutions to volcanic melts and rocks.

Exercise 2: Compare Volcanoes, Part 1

For Exercise 2, create an outline that compares the volcanoes Nyiragongo
with Vesuvius. You can find narratives and descriptions of the eruptions on
the web and in geoscience libraries. Chirico, et al. (2009) and Favalli, et
al. (2009) modeled the lava flows for the 2002 Nyiragongo eruption and
described the hazards associated with them. Scarth (2009), in particular, has
written a comprehensive but compact history of the Vesuvian eruptions. It is
worth reading. His book includes translations of the letters written by Pliny
the Younger, a first-hand, eye-witness account of the eruption that happened
in the year 79.
The sketch maps, Figures E1 and E2, are drawn to the same scale. Both
volcanos are a few km from the urban centers they destroyed. Nyiragongo is
smaller than Somma but both are approximately the same height. The highest
point on the Somma crater wall is approximately 1050 m above the volcano’s
base. Whereas the crater wall at Nyiragongo stands approximately 1150 m
above its base. You can check these numbers, and you should, by zooming in
on the maps which you can reach at:
http://www.gtwist.ca/Petrology/readMaps.html.

Attached to the map are some photographs showing the destruction the
volcano caused and some views of the volcano. Thure Cerling from the
University of Utah furnished the photographs of Nyiragongo.
The eruptions of Nyiragongo have been compared to eruptions on Hawai’i.
People often ask whether eruptions like those from Nyiragongo occur in
Hawai’i. No historic flows in Hawai’i have been as fluid as the flows of
Nyiragongo.
You might find a comparison of the descriptions of the Nyiragongo eruptions
with descriptions of Hawaiian eruptions useful. Jackson, et al. (1975) narrates
the 1968 eruption of Kilauea. Eruptions similar to the 79 Vesuvius – Somma
eruption have happened elsewhere since 79. They have been designated
as plinian eruptions, named after Pliny the Younger. Examples of plinian
eruptions include the Mount St. Helens eruption of 1980, the Krakatoa
eruption of 1883, and the Pinatubo eruption of 1991. Although similar, these
eruptions have different characteristics both between themselves and between

59
 The Nature of Volcanic Rocks

the eruption of Vesuvius

– Somma in 79.

Nyiragongo Volcano
and Flash-flood Lava
Flows
Nyiragongo is one of
eight large volcanoes
that march across
the Great Rift Valley
of East Africa in the
Democratic Republic
of the Congo. The
eight large volcanoes
and numerous smaller
satellite volcanoes make
up the Virunga Volcanic
field.
Nyiragongo sits Figure E1
approximately 20 km
north of the city of
Goma in the Democratic
Republic of the Congo
(see the sketch map,
Figure E1). The crater
of the volcano has had
a persistent lava lake.
In 1977 it emptied
through a crack in a
crater wall. Legend has
it that the consequent
lava flow trapped,
burned, and killed a
herd of elephants as
well as hundreds or
maybe thousands of
people. The rapidly
flowing lava caught
them; they couldn’t
outrun it. In 2002, the Figure E2

60
 The Nature of Volcanic Rocks

volcano erupted again and forced the immediate evacuation of a large number
of people from the city. The lava flows traversed through the city, engulfed,
and overturned automobiles, and killed residents. The eruption left behind
a lava flow a metre or so thick, filling the streets of Goma. Pictures of the
devastation remind one of pictures of the aftermaths of flash floods in the
southwestern United States. The lava flows moved like floodwaters but at
temperatures near 1000°C or more. Each eruption lasted only a few days at
longest.

Somma, Vesuvius, and the Destruction of Pompeii

Although Vesuvius has several unique features, it has classically served as a
model for other eruptions, historic and prehistoric. In addition, the eruption
in 79 was the first to be scientifically described. The 79 eruption of Vesuvius
was very different from the 1997 – 2002 eruptions of Nyiragongo.
The volcanic edifice at Vesuvius consists of two volcanoes. The older and
larger one is named Somma, which was the volcano that erupted in 79.
Vesuvius has erupted several times since (Scarth, 2009). The later eruptions
built the photogenic cone we call Vesuvius in the crater of Somma. The last
major eruption in 1944 produced lava flows, one of which flowed from the
recent cone along the valley between the cone of Vesuvius and the crater wall
of Somma (see Scarth, 2009, p. 271 and the photograph on the cover for this
document). Vesuvius is considered the most dangerous volcano in Europe.
Historians credit Pliny the Younger with writing the first scientific report of a
volcanic eruption. Vesuvius erupted in the year 79, first producing an eruptive
cloud that rained ash (crystal fragments and small drops of melt that turned
to glass), pumice blocks, and rock fragments on Pompeii, approximately 17
km southeast of the eruptive center (see the sketch map, Figure E2). The
eruption also ejected a column of fragmented melt, crystals, fragmented
rock, and fluid (gas) to a height estimated at 20 km (Carey and Sigurdsson,
1987). What goes up must come down; the column collapsed, producing
high-density flows made of hot glass fragments, crystals, rock fragments,
and gas that over-ran the countryside. Intermittent eruptions rebuilt the
eruption column several times. The strongest eruption raised the column to a
maximum height estimated at 32 km. The column partially collapsed between
eruptions, producing more high-density flows. Several metres of the early
ash and pumice-falls fell on Pompeii, crushing roofs, suffocating people and
animals, and burying them. Thousands were killed. The later high-density
flows completely covered Pompeii, sealing it from the rest of the world for
more than a thousand years.

61
 The Nature of Volcanic Rocks

A map of Vesuvius and its neighborhood can be opened from the URL:
http://www.gtwist.ca/Petrology/readMaps.html.
If you work on the Nyiragongo – Vesuvius project in partnership with
others, you might want to try the following activity. It will consolidate and
expand what you’ve learned. Get your group together around a table and
draw lots. The first person tells the group one thing that is different between
the eruptions or tells the group how a feature differs from a similar eruption
elsewhere, Hawai’i or Mount St. Helens, for example. The next person then
does the same thing, telling the group about another feature. The only rules
are: (1) The feature described must be significant and (2) The feature must
be presented for the first time; it can’t be a repeat or paraphrased version of a
feature already presented. The last person who can tell the group something
new and significant wins. You can even offer a prize to the winner, a box of
chocolates maybe or a favorite drink. If you take notes during this activity,
you can record and learn what the others learned and likely improve your
outline by making it more complete.

Exercise 3: Making Friends with Volcanic Minerals

Write a short paragraph about each of the critical minerals listed in Table 1.
Each paragraph should be no more than half a page long. Actually, the more
concise, the better. By limiting the number of words devoted to each mineral,
you will make yourself concentrate on those features and properties that
make them real in your mind. What you put in the paragraphs is up to you.
This exercise should acquaint you with the minerals in the list. Petrologists
tend to remember mineral assemblages - what minerals are found together
in the same rocks. Mineralogists tend to remember optical properties and
crystallographic features. Geophysicists often concentrate on physical
properties. Write about those features that interest you. You can pick different
kinds of features for the different minerals if you want. Consistency is not
required. Becoming acquainted with the minerals is.

Exercise 4: Compare Volcanoes, Part 2

Revise your comparison of the two volcanoes, Vesuvius and Nyiragongo, by
incorporating what you learned in Part 2 of the Educational Tour. How much
does the material you learned in Part 2 change your original comparison?
How much did the material extend your comparison? This time, present your
comparison in an essay. An essay helps one examine complex features more
completely than does an outline. One can ignore exceptions and qualifications
in an outline that one can’t in an essay.

62
 The Nature of Volcanic Rocks

Exercise 5: Write a Review of an Article

Read and review of the article on the Yellowstone supervolcano that appeared
in the August issue of National Geographic (Achenbach, et al., 2009). Which
of the first three features that we discussed for characterizing volcanic rocks
did they describe: geologic context, texture, and mafic index? Evaluate how
well they covered them. Could the article be improved if the three features
were explicitly considered?

Exercise 6: Develop a Volcanic Hazard Assessment

Pick a historically active volcano. The history may be an oral one such as the
First Nation description of the Nisga’a Volcano eruption in the 1200’s. For
your selected volcanic center, develop an assessment of the volcanic hazard
it poses if it should erupt again. Assessing the dangers volcanoes pose in
today’s world are very real problems. We began our Tour with a comment on
the effect of volcanic eruptions on aircraft. Almost any kind of volcano can
produce ash clouds that can affect aircaft. Your assessment should address
this very real danger. Guffanti, et al., 2010 lists the volcanic centers that
produced ash clouds that affected aircraft. The past several decades have seen
dozens if not hundreds of assessments of volcanic eruptions. The problem is:
How reliable are they? If an assessment is available can you improve it? Why
is your assessment better?
Some questions you might consider:
1. What should a community do?
2. What should an individual do?
3. What should private groups, businesses, or institutions do?
4. What should public entities do?
5. What affect would an eruption have on aircraft?

63
The Nature of Volcanic Rocks

64
 The Nature of Volcanic Rocks

References Cited
Achenbach, J., Thiessen, M., and Cañellas, H., 2009, When Yellowstone
explodes: National Geographic, v. 216, p. 56-69.

Bailey, E.B., Clough, C.T., Wright, W.B., Richey, J.E., and Wilson, G.V.,
1924, Tertiary and Post-Tertiary geology of Mull, Loch Aline, and Oban,
Memoir Geological Survey of Scotland.

Berman, R.G., and Brown, T.H., 1984, A thermodynamic model for

multicomponent melts, with application to the system CaO-A1203-Si02:
Geochimica et Cosmochimica Acta, v. 48, p. 661-678.

Best, M.G., and Christiansen, E.H., 2001, Igneous Petrology: Malden,

Massachusetts, Blackwell Science, Inc, 458 p.

Bindeman, I.N., Bin, F., Noriko, T.K., and Valley, J.W., 2008, Origin and
evolution of silicic magmatism at Yellowstone based on ion microprobe
analysis of isotopicall zoned zircons: Journal of Petrology, v. 49, p. 163-
193.

Bowen, N.L., 1915, The later stages of the evolution of the igneous rocks,
Journal of Geology, v. 23, p. 1 – 89.

Bowen, N.L., 1928, The Evolution of the Igneous Rocks: Princeton, NJ,
Princeton University Press, 251 p.

Carmichael, I.S.E., 1962, Pantelleritic liquids and their phenocrysts:

Mineralogical Magazine, v. 33, p. 80-113.

Carmichael, I.S.E., 1963, The crystallization of feldspar in volcanic acid

liquids: Quarterly Journal of the Geological Society of London, v. 119, p.
95-131.

Carmichael, I.S.E., 1964, Natural liquids and the phonolitic minimum:

Geological Journal, v. 4, p. 55-60.

Carmichael, I.S.E., 1965, Trachytes and their feldspar phenocrysts:

Mineralogical Magazine, v. 34, p. 107-125.

Carmichael, I.S.E., 1967, The mineralogy and petrology of the volcanic

rocks from the Leucite Hills, Wyoming: Contributions to Mineralogy and
Petrology, v. 15, p. 24-66.

65
 The Nature of Volcanic Rocks

Carmichael, I.S.E., Nicholls, J., and Smith, A.L., 1970, Silica activity in
igneous rocks: American Mineralogist, v. 55, p. 246-263.

Carmichael, I.S.E., Turner, F.J., and Verhoogen, J., 1974, Igneous Petrology:
New York, USA, McGraw-Hill Book Company, 739 p.

Chirico, G.D., Favalli, M., Papale, P., Boschi, E., Pareschi, M.T., and
Mamou-Mani, A., 2009, Lava flow hazard at Nyiragongo Volcano, DRC,
2. Hazard reduction in urban areas: Bulletin of Volcanology, v. 71, p.
375–387.

Christiansen, E.H., Sheridan, M.F., and Burt, D.M., 1986, The geology and
geochemistry of Cenozoic topaz rhyolites from the western United States:
Geological Society of America Special Paper, v. 205, p. 1-82.

Christiansen, R.L., 2001, The Quaternary and Pliocene Yellowstone Plateau

volcanic field of Wyoming, Idaho, and Montana: U.S. Geological Survey
Professional Paper, v. 729-G, 145 p.

Congdon, R.D., and Nash, W.P., 1988, High-fluorine rhyolite: An eruptive

pegmatite magma at the Honeycomb Hills, Utah: Geology, v. 16, p. 1018-
1021.

Congdon, R.D., and Nash, W.P., 1991, Eruptive pegmatite magma: rhyolite of
the Honeycomb Hills, Utah: American Mineralogist, v. 76, p. 1261-1278.

Cross, W., Iddings, J. P., Pirrson, L. V., and Washington, H. S., 1902, A
quantitative chemico-mineralogical classification and nomenclature of
igneous rocks: Journal of Geology, v. 10, p. 555-690.

Deer, W.A., Howie, R.A., and Zussman, J., 1992, An Introduction to the
Rock-Forming Minerals, 2nd Edition: Essex, England, Longman Group
UK Limited.

Dyar, M.D., Gunter, M.E., and Tasa, D., 2008, Mineralogy and Optical
Mineralogy: Chantilly, VA, USA, Mineralogical Society of America.

Favalli, M., Chirico, G.D., Papale, P., Pareschi, M.T., and Boschi, E., 2009,
Lava flow hazard at Nyiragongo volcano,D.R.C.1.Model calibration and
hazard mapping: Bulletin of Volcanology, v. 71, p. 363-374.

Falk, J. H., and Dierking, L. D., 2010, The 95 percent solution: American
Scientist, v. 98, p. 486, DOI: 10.1511/2010.87.486.

66
 The Nature of Volcanic Rocks

Ghiorso, M.S., and Sack, R.O., 1995, Chemical mass transfer in magmatic
processes IV. A revised and internally consistent thermodynamic model
for the interpolation and extrapolation of liquid-solid equilibria in
magmatic systems at elevated temperatures and pressures: Contributions
to Mineralogy and Petrology, v. 119, p. 197-212.

Giordano, D., Russell, J.K., and Dingwell, D.B., 2008, Viscosity of magmatic
liquids: A model: Earth and Planetary Science Letters, v. 271, p. 123-134.

Gould, S.J., 1989, Wonderful Life. The Burgess Shale and the Nature of
History: New York, NY, USA, W.W. Norton & Company.

Guffanti, M., Casadevall, T. J., and Budding, K., 2010, Encounters of aircraft
with volcanic ash clouds: A compilation of known incidents, 1953-2009:
U.S. Department of the Interior, Data Series, v. 245, p. 1-12.

Halleran, A.D., and Russell, J.K., 1990, Trace elements formulated as Pearce
element ratios. II. Recognitionof open system behavior, in Russell, J.K.,
and Stanley, C.R., eds., Theory and Application of Pearce Element Ratios
to Geochemical Data Analysis, Volume 8: Vancouver, British Columbia,
Geological Association of Canada, p. 243-270.

Holmes, A. 1920. The Nomenclature of Petrology. Thos. Murby and Co.

London, 284 p.

Holmes, A., 1937, The Petrology of Katungite: Geological Magazine, v. 74,

p. 200-219.

Hunt, C.B., Averitt, P., and Miller, R.L. 1953. Geology and Geography of
the Henry Mountains region, Utah: United States Geological Survey
Professional Paper 228, 234 p.

Jackson, D.B., Swanson, D.A., Koyanagi, R.Y., and Wright, T.L., 1975, The
August and October 1968 East Rift eruptions of Kilauea Volcano, Hawaii:
United States Geological Survey Professional Paper, v. 890, p. 1-33.

Johnson, R.W., 1969, Volcanic geology of Mount Suswa, Kenya:

Philosophical Transactions, Royal Society of London, Series A 265: 383-
412. DOI: 10.1098/rsta.1969.0061.

Kennedy, W.Q., 1933, Trends of differentiation in basaltic magmas: American

Journal of Science, v. 25, p. 239-256.

67
 The Nature of Volcanic Rocks

Kuntz, M.A., Rowley, P.D., MacLeod, N.S., Reynolds, R.L., McBroome,

L.A., Kaplan, A.M., and Lidke, D.J., 1981, Petrography and particle-size
distribution of pyroclastic-flow, ash-cloud, and surge deposits, in Lipman,
P.W., and Mullineaux, D.R., eds., The 1980 Eruptions of Mount St.
Helens, Washington, U.S. Geological Survey Professional Paper 1250, p.
525-539.

Le Maitre, R.W., 1989, A Classification of Igneous Rocks and Glossary of

Terms: Oxford, England, Blackwell Scientific Publications Ltd, 193 p.

Le Maitre, R.W., 2002, Igneous rocks: a classification and glossary of terms.

Recommendations of the International Union of Geological Sciences,
Subcommission on the Systematics of Igneous Rocks. Cambridge
University Press, 236 p.

McPhie, J., Walker, G.P.L., and Christiansen, R.L., 1990, Phreatomagmatic

and phreatic fall and surge deposits from the explosions af Kilauea
volcano, Hawaii, 1790 A.D.: Keanakakoi Ash Member: Bulletin of
Volcanology, v. 52, p. 334-354.

Nash, W.P., Carmichael, I.S.E., and Johnson, R.W., 1969, The mineralogy
and petrology of Mt. Suswa, Kenya: Journal of Petrology, v. 10, p. 409-
439.

Nicholls, J., 2000, “Thermodynamics of a magmatic gas phase” 50 years

later: Comments on a paper by John Verhoogen (1949): Canadian
Mineralogist, v. 38, p. 1313-1328.

Nicholls, J., and Carmichael, I.S.E., 1969, Peralkaline liquids: A petrological

study: Contributions to Mineralogy and Petrology, v. 20, p. 268-294.

Nicholls, J., Carmichael, I.S.E., and Stormer, J.C., Jr, 1971, Silica activity
and Ptotal in igneous rocks: Contributions to Mineralogy and Petrology, v.
33, p. 1-20.

Nicholls, J., and Russell, J.K., 1991, Major-element chemical discrimination

of magma batches in lavas from Kilauea volcano, Hawaii, 1954-1971
eruptions: Canadian Mineralogist, v. 29, p. 61-74.

Nicholls, J., and Stout, M.Z., 1994, Chemical, mineralogical and petrographic
variations in the Aiyansh lava flow, B.C: Geological Association of
Canada, Program with Abstracts, v. 19, p. 82.

68
 The Nature of Volcanic Rocks

Nicholls, J., and Stout, M.Z., 1997, Epitactic overgrowths and intergrowths of
clinopyroxene on orthopyroxene: Implications for paths of crystallization,
1881 lava flow, Mauna Loa Volcano, Hawaii: Canadian Mineralogist, v.
35, p. 909-922.

Nicholls, J., and Stout, M.Z., 1988, Picritic melts in Kilauea - Evidence from
the 1967-1968 Halemaumau and Hiiaka eruptions: Journal of Petrology,
v. 29, p. 1031-1057.

Nicholls, J., and Stout, M.Z., 2008, Magmatic Histories: The 1968-
1969 Eruptions of Kilauea Volcano. http://gtwist.ca/Petrology/
MagmaticHistories/Kilauea1968Text.pdf

Oppenheimer, C., 2010, Triple Point: We told you so! Reflections on the
‘Ashpocalypse,’ v. 6, No. 4, p. 205.

Peterson, T.D., 1989, Peralkaline nephelinites. I. Comparative petrology

of Shombole and Oldoinyo L’engai, East Africa. Contributions to
Mineralogy and Petrology, Contributions to Mineralogy and Petrology, v.
101, p. 458-478.

Philpotts, A.R., and Ague, J.J., 2009, Principles of Igneous and Metamorphic
Petrology: Cambridge, England, Cambridge University Press, 667 p.

Russell, J.K., and Stanley, C.R., 1990, Origins of the 1954 to 1960 lavas,
Kilauea Volcano, Hawaii: Major element constraints on shallow reservoir
magmatic processes: Journal of Geophysical Research, v. 95, p. 5021-
5047.

Scaillet, B., Pichavant, M., and Cioni, R., 2008, Upward migration of
Vesuvius magma chamber over the past 20,000 years: Nature (London), v.
455, p. 216-220.

Scarth, A., 2009, Vesuvius, A Biography: Princeton, NJ, USA, Princeton

University Press, 342 p.

Scott, D.H., and Trask, N.J., 1971, Geology of the Lunar Crater Volcanic
Field, Nye County, Nevada: United States Geological Survey Professional
Paper, v. 599, p. I1-I22.

Smith, A.L., 1970, Sphene, perovskite and co-existing Fe-Ti oxide minerals:
American Mineralogist, v. 55, p. 264-269.

69
 The Nature of Volcanic Rocks

Spear, D. B. and King, J. S. (1982). The geology of Big Southern

Butte, Idaho. Cenozoic Geology of Idaho. B. Bonnichsen and R. M.
Breckenridge, Idaho Geological Survey Bulletin. 26: 395-403.

Stout, M.Z., and Nicholls, J., 1983, Origin of hawaiites from the Itcha
Mountain Range, British Columbia: Canadian Mineralogist, v. 21, p. 575-
581.

Stout, M.Z., and Nicholls, J., 1977, Mineralogy and petrology of Quaternary
lavas from the Snake River Plain, Idaho: Canadian Journal of Earth
Sciences, v. 14, p. 2140-2156.

Stout, M.Z., Nicholls, J., and Kuntz, M.A., 1994, Petrological and
mineralogical variations in 2,500-2,000 yr B.P. lava flows, Craters of the
Moon Lava Field, Idaho: Journal of Petrology, v. 35, p. 1681-1715.

Symons, D. T. A., 1975, Age and flow direction from magnetic measurements
on the historic Aiyansh flow, British Columbia: Journal of Geophysical
Research, v. 80, p. 2622-2626.

Symons, G.J. (Editor), 1888, The Eruption of Krakatoa and Subsequent

Phenomena. Report of the Krakatoa Committee of the Royal Society:
London, England, Trubner & Co, 494 p.

Turner, F.J., 1970, Uniqueness versus conformity to pattern in petrogenesis:

American Mineralogist, v. 55, p. 339 – 348.

Turner, F.J., and Verhoogen, J., 1960, Igneous and Metamorphic Petrology:
Inc, McGraw-Hill Book Company.

Velde, D., and Rachdi, H.E., 1988, Influence of Sr on an established

petrological incompatibility: The association melilite + K-feldspar in a
nephelinite from Djebel Targou, Central Morocco: Journal of Petrology, v.
29, p. 585-597.

Verhoogen, J., 1948, Les eruptions 1938-1940 du volcan Nyamuragira:

Brussels, Belgium, Institut des Parcs Nationaux du Congo Belge, p. 186.

Verhoogen, J., 1980, Energetics of the Earth: Washington, D.C., USA,

National Academy of Science.

Williams, H., Turner, F.J., and Gilbert, C.M., 1982, Petrography, An

Introduction to the Study of Rocks in Thin Section: San Francisco, W.H.

70
 The Nature of Volcanic Rocks

Freeman and Company, 626 p.

Winter, J.D., 2001, An Introduction to Igneous and Metamorphic Petrology:

Upper Saddle River, New Jersey, USA, Prentice Hall, 697 p.
Wright, T. L., 1971, Chemistry of Kilauea and Mauna Loa in space and time:
United States Geological Survey Professional Paper, v. 735, p. 1-40.

Wuorinen, V., 1978, Age of the Aiyansh Volcano, British Columbia: Canadian
Journal of Earth Sciences, v. 15, p. 1037-1038.

71
 The Nature of Volcanic Rocks

Index
79 Somma-Vesuvius eruption vi, 13, 59, 61
1750-75 Nisga’a eruption 53, 63, 70, 72
1790 Kilauea eruption 49, 59
1883 Krakatoa eruption 59
1902 Mount Pelée eruption 37
1944 Vesuvius eruption 61
1969-74 Mauna Ulu eruptions of Kilauea 35
1980 Mount St. Helens eruption 43, 59
1991 Pinatubo eruption 59
2007-8 Ol Doinyo Lengai eruptions 51-53
2010 Mount Merapi eruption 37

A
Achenbach, J. 20, 54, 63, 65
activity 26-27, 29, 31, 40
activity of Al2O3 see alumina activity
activity of SiO2 see silica activity
activity of sugar 25, 27, 51
taste and 25-26, 58
aenigmatite 23, 34, 40, 52
Ague, J.J. 54, 69
Aiyansh lava flow see Nisga’a volcano
Aiyansh Volcano see Nisgaá volcano
Alberta 50
alkali feldspar 54
alumina activity 11, 33-34, 45
high 33-34, 45
low 33-34, 45
medium 33-35, 45, 52-53
ultra-low 33-34
amphibole 19, 33, 34, 53
andesite 7, 45-46
anorthoclase 23, 34, 40, 52
ash 13, 15, 42
ash fall 13, 49
ash fall tuff 13, 15
ash flow 3, 14, 17, 42
ash flows 61
Australia 34
Averitt, P. 67, see Hunt, C.B. 8

B
Bailey, E.B. 46, 65
basalt 7, 36, 45-47
alkali olivine 46, 47
tholeiite 46
Berman, R.G. 31, 65
Best, M.G. 54, 65

72
 The Nature of Volcanic Rocks

Big Bend National Park 13, 15

Big Southern Butte 36
Bin, F. 65, see Bindeman, I.N. 36
Bindeman, I.N. 36, 65
biotite 19, 33-34, 52-53
Boschi, E. 66, see Chirico, G.D. 59
Bowen, N.L. 41, 54, 65
Bowen’s reaction series 41, 54
British Columbia 53
Brown, T.H. 31, 65
Budding, K. 67, see Guffanti, M. 63
Burt, D.M. 66, see Christiansen, E.H. 35

C
Ca-poor pyroxene 23-24, 29-30, 32-34, 40, 45-48, 50
Ca-rich pyroxene 23, 29, 31-32, 34, 45, 47, 53, 55
caldera 48-49
Calgary 50
California 37, 50
Cañellas, H 65, see Achenbach, J. 20, 54, 63
Carmichael, I.S.E 34-35, 54, 65-66, 68, see Nash, W.P. 52
and silica activity 27, 29, 66, 68, see Nicholls, J. 30
Casadevall, T.J. 67, see Guffanti, M. 63
chemical composition 9, 12, 35-37, 46, 50, 65
chemical formulae 33
Chicago 50
Chirico, G.D. 59, 66
Christiansen, E.H. 35, 54, 65-66
Christiansen, R.L. 48, 66, 68, see McPhie, J. 50
Cima Volcanic Field National Natural Landmark 4
Cioni, R. 69, see Scaillet, B. 49
CIPW 27-30, 33-34
CIPW norm 27, 34, 46, 48
chemical formulae 33
mineral formulae 27, 33, 46
classifications
chemical 38
CIPW 36
IUGS 36
mineralogical 36
Clough, C.T. 65, see Bailey, E.B. 46
cones, cinder 4-5, 13, 17, 61
Congdon, R.D. 35, 66
Congo see Democratic Republic of the Congo
contingency v, 9-12
crater vi, 4-5, 12-14, 38, 50, 52, 59-60
Craters of the Moon National Monument 5, 12-14, 38
Cross, W. 27, 36, 66, see also CIPW
crystal fractionation, crystal separation and segregation
crystal separation and segregation 54-55

73
 The Nature of Volcanic Rocks

crystalline equivalent 39
crystals 61

D
Deer, W.A. 24, 31, 66
Democratic Republic of the Congo 55
diabase 7
Dierking, L.D. vi, 67
Dingwell, D.B. 67, see Giordano, D. 49-52
diopside 34
domes 37
Dyar, M.D. 24, 66

E
East African Rift Valley v, 34, 43, 51-52, 55
East Butte 36
elements
isotopes 35, 39
major 35
minor 35
trace 35
eruptions 6, 10-13, 17, 20-21, 35, 42-43, 52, 61-62
large explosive eruptions 4, 35, 37, 43, 47-48, 50
79 Somma-Vesuvius eruption vi, 13, 59
1883 Krakatoa eruption 59
1902 Mount Pelée eruption 37
1980 Mount St. Helens eruption 43, 59
1991 Pinatubo eruption 59
2010 Mount Merapi eruption 37
Yellowstone supervolcano 20, 48, 54, 63
plinian eruptions see large explosive eruptions
quiet 13, 20, 48-49, 55
essays 57, 62
experimental studies 6, 9, 11, 50
Eyjafjallajökull volcano 1, 47, 50

F
Falk, J.H. vi, 67
Favalli, M. 66, see Chirico, C.D. 59
feldspar 12, 19, 28-32, 45, 52, 55
feldspathoid 28-30, 32
felsic intrustion 36
felsic minerals 19-21, 24, 33, 40, 53
felsic rocks 19, 32, 36, 40
Fitzroy Basin 34
flash floods 61
fragmental deposits 34
fragmentation 4, 7
fragments 3-4, 17
free-choice learning see informal learning

74
 The Nature of Volcanic Rocks

G
Galapagos Islands 12
geologic context 7, 11-13, 18, 20-21, 25, 42, 48, 53, 63
Ghiorso, M.S. 41, 50-51, 67
Gilbert, C.M. 70, see Williams, H. 19
Giordano, D. 49-52, 67
glass 6-7, 17, 19, 35-37, 40-42, 52, 58, 61
glassy rocks 35, 37, 39-40, 58
crystalline equivalent 39
Goma 61
Gould, S.J. 9, 67
grain-size 17-18
granite 7
granodiorite 7
Guffanti, M. 63, 67
Gunter, M.E. 66, see Dyar, M.D. 24

H
Halifax 50
Halleran, A.D. 35, 67
Hawai’i 11, 16-17, 35, 43, 46-47, 49, 51, 59, 62
high 33-34
Holmes, A. 8, 18, 67
hornblende 34
Howie, R.A. 66, see Deer, W.A. 24, 31
Hunt, C.B. 8, 67

I
Iceland 1, 12, 37
Idaho 37
Iddings, J.P. 66, see Cross, W 27, 36
see also CIPW
igneous rocks 58
ijolite 7
important features, volcanic rocks, alumina activity 11, 33-34
important features; volcanic rocks
alumina activity 11
geologic context 7, 11-13, 18, 20-21, 25, 42, 48, 53, 63
mafic index 11, 19-21, 25, 38-42, 48, 63
silica activity v, 11, 25-27, 29, 31-32, 40, 50-51
textures 7, 9, 11, 13, 16-18, 20-21, 25, 42, 48, 53, 63
IMRAD 57
informal learning vi
intermediate mafic index 45
intermediate rocks 19, 36, 38, 40, 50, 53
Io, Jovian Moon 53
IUGS classification 36

J
Jackson, D.B. 59, 67

75
 The Nature of Volcanic Rocks

jacuparangite 7
Johnson, R.W. 52, 67, 68, see Nash, W.P. 52
Jumilla 34

K
kalsilite 19, 23-24, 28-29, 31-32, 55
Kaplan, A.M. 68, see Kuntz, M.A. 48
Kennedy, W.Q. 46, 67
Kenya 52
Kilauea volcano 11, 35, 49, 51
Mauna Ulu eruptions of 35
King, J.S. 36, 70
Koyanagi, R.Y., 67, see Jackson, D.B. 59
Krakatoa Volcano 37
Kuntz, M.A. 48, 68, 70, see Stout, M.Z. 12, 47

L
lava flows vi, 3, 5, 11-13, 16-18, 21, 27, 34, 36, 43, 46, 48, 52-53, 55-56, 58-59, 60, 61
lavas 3, 17
Le Maitre, R.W. 7, 36, 46, 54, 68-69
leucite 23, 28-29, 31-32, 40, 48, 55
Leucite Hills 34
Lidke, D.J. 68, see Kuntz, M.A. 48
life-long learning see informal learning
Lunar Crater 47

M
MacLeod, N.S. 68, see Kuntz, M.A. 48
mafic 45
mafic index 11, 19-21, 25, 38-42, 45, 48, 63
mafic minerals 19-20, 24, 33, 40, 53
mafic rocks 19-20, 39, 46
magazine articles 57
magma types
tholeiitic basalt 46
alkali olivine basalt 46
magmas 4, 6, 20-21, 35, 39, 46, 48, 50
silicate magmas 52
sodium carbonate 53
sulfide 53
Mamou-Mani, A. 66, see Chirico, G.D. 59
Mauna Loa volcano 10
Mauna Ulu eruptions 35
McBroome, L.A. 68, see Kuntz, M.A. 48
McPhie, J. 50, 68
medium see alumina activity, medium
melilite 23-24, 28-29, 31-32, 40, 52, 55
melting, partial 36
melts
fragmented 61

76
 The Nature of Volcanic Rocks

silicate
thermodynamic model for 41, 50-51
metaluminous 45, see alumina activity, medium
Mexico 35
mica, white 35
Mid-Atlantic Ridge 12, 37
Middle Butte 36
Miller, R.L. 67, see Hunt, C.B. 8
mineral assemblages 2, 7, 17, 19-20, 25-27, 29, 37, 40, 42, 47-48, 53, 57, 62
mineral formulae 27, 33, 46
mineral properties 23, 62
minerals 1-3, 7, 9, 12, 17, 19-20, 23-28, 33, 35, 37, 40-41, 45, 47, 52, 54-55, 62
critical 23-24, 34, 42, 45, 47, 62
aenigmatite 23, 34, 40, 52
anorthoclase 23, 34, 40, 52
Ca-poor pyroxene 23-24, 29-30, 32-34, 40, 45-48, 50
Ca-rich pyroxene 23, 29, 31-32, 34, 45, 47, 53, 55
kalsilite 19, 23-24, 28-29, 31-32, 55
leucite 23, 28-29, 31-32, 40, 48, 55
melilite 23-24, 28-29, 31-32, 40, 52, 55
monticellite 23-24, 28-29, 31, 33, 40
Na-rich pyroxenes 23, 52
nepheline 19, 23, 30, 40, 46-47, 50, 52, 55
olivine 17-19, 23-24, 28-35, 40, 45-48, 50, 53, 55
perovskite 23-24, 28-30
plagioclase 23, 34, 54-55
quartz 23, 28-30, 32, 34, 45, 48
sanidine 23, 34
titanite 23-24, 28-30, 52
topaz 23-24, 35
Mono Craters 47, 50
Obsidian Dome 47, 50
Montana 31, 33
monticellite 23-24, 28-29, 31, 33, 40
Mount Pelée 37
Mount St. Helens 37, 48
Mount St. Helens, eruptions of 43, 59
Mount Suswa 52

N
Na-rich pyroxenes 23, 52
Nairobi 52
Nash, W.P. 35, 52, 66, 68
nepheline 19, 23, 30, 40, 46-47, 50, 52, 55
Newton, I. 27
Ngorogoro Caldera 5
Nicholls, J. iv, 30, 34-35, 51, 53, 66, 68-69, 70
see Carmichael, I.S.E. 27, 29
see Stout, M.Z. 12, 47
Nisga’a volcano 53, 63, 70, 72

77
 The Nature of Volcanic Rocks

Noriko, T.K. 65, see Bindeman, I.N. 36

norm see CIPW norm
normative hypersthene 46, 48
normative nepheline 46, 48
Nova Scotia 50
Nyamuragira melts 55-56
Nyamuragira volcano 55-56
Nyiragongo eruptions 59, 61
Nyiragongo lava flows 55
Nyiragongo melts 55-56
Nyiragongo Volcano 42-43, 50, 55-56, 59-62
Nyiragongo Volcano and Flash-flood Lava Flows 60

O
obsidian 7 see also glass
Obsidian Dome 47, 50
Ol Doinyo Lengai 51-53
olivine 17-19, 23-24, 28-35, 40, 45-48, 50, 53, 55
Oppenheimer, C. 1, 69
Oregon 36

P
Papale, P. 66, see Chirico, G.D. 59
Pareschi, M.T. 66, see Chirico, G.D. 59
pegmatites 35
peralkaline see alumina activity, low
peraluminous see alumina activity, high
perovskite 23-24, 28-30
Peterson, T.D. 52, 69
Philpotts, A.R. 54, 69
Pichavant, M. 69, see Scaillet, B. 49
Pinatubo volcano 59
Pirrson, L.V. 66, see Cross, W 27, 36
see also CIPW
plagioclase 23, 34, 54-55
planetary heat transfer 1
Pliny the Younger 1, 27, 39, 57, 59, 61, 63
plutonic rocks 17-18, 34
poker 10
Pompeii 13-14, 42, 48, 61
priderite 34
Principia 27

Q
quartz 23, 28-30, 32, 34, 45, 48
quartz-saturated 34-35, 45, 48

R
Rachdi, H.E. see Velde, D.
Reynolds, R.L. 68, see Kuntz, M.A. 48

78
 The Nature of Volcanic Rocks

rhyolite 45-46
Richey, J.E. 65, see Bailey, E.B. 46
rock fragments 61
rock space 41-42, 45, 53-56
rock types v, 2, 10-12, 19, 23, 25, 30, 33, 35, 42, 46, 53-54, 57
andesite 7, 45-46
basalt 7, 36, 45-47
obsidian 7
pegmatite 35
rhyolite 45-46
trachyandesite 1
Rowley, P.D. 68, see Kuntz, M.A. 48
Russell, J.K. 35, 67, 68, 69, see Giordano, D. 49-52

S
Sack, R.O. 41, 50-51, 67
sanidine 23, 34
Scaillet, B. 49, 69
Scarth, A. vi, 59, 61, 69
Scott, D.H. 50, 69
Sheridan, M.F. 66, see Christiansen, E.H. 35
silica activity v, 11, 25-29, 31-33, 40-42, 45, 47-48, 50-52
SiO2 1, 25-26, 28-34, 40-41, 45, 48-53, 55
SiO2 content 28-29, 38-40, 51-52
and mafic index 38-40
Smith, A.L. 30, 66, see Carmichael, I.S.E. 27,29
Snake River Plain 12-13, 35-37, 47, 53
Somma crater wall 59
Somma-Vesuvius volcano 37
Somma volcano 13-14, 42-43, 59-61
Vesuvius, eruption of 60-61
Vesuvius volcano vi-1, 13-14, 42-43, 50, 59, 61-62
source regions 35, 55
Spain 34
Spear, D.B. 36, 70
Stanley, C.R. see Russell, J.K.
Stormer, J.C., Jr 68, see Nicholls, J. 30
Stout 12, 35, 47, 51, 53, 70-71
stratovolcano 55
sugar-water solutions 6, 11, 25-27, 51, 58-59
Swanson, D.A. 67, see Jackson, D.B. 59
Symons, D.T.A. 48, 53, 70

T
Tanzania 51-53
Tasa, D. 66, see Dyar, M.D. 24
taste and activity 25-26, 51, 58
tectonic plates 43
textures 7, 9, 11, 13, 16-18, 20-21, 25, 42, 48, 53, 63
thermodynamic properties and silica activity v, 11, 25-29, 31-33, 40-42, 45, 47-48, 50-52

79
 The Nature of Volcanic Rocks

Thiessen, M. 65, see Achenbach, J. 20, 54, 63

titanite 23-24, 28-30, 52
topaz 23-24, 35
trachyandesite 1
trachyandesite magma 1
Trask, N.J. see Scott, D.H.
Tuff Canyon 13, 15
tuffs 15
Turner, F.J. 10, 54, 66, 70
see Carmichael, I.S.E., 54
see Williams, H., 19

U
Ubehebe Crater 47, 50
ultramafic 38
ultramafic rocks 20
United States 35, 61
Utah 35

V
Valley, J.W. 65, see Bindeman, I.N. 36
Velde, D. 31, 70
Verhoogen, J. 1, 54-55, 66, 69, 70,
see Carmichael, I.S.E. 54
Vesuvian eruptions 59
Vesuvius-Somma 48
Vesuvius volcano vi-1, 13-14, 42-43, 50, 59, 61-62
viscosity 11, 27, 49-51, 58, 67
volcanic hazards 63
volcanic melts 58
volcanic phenomena 20
volcano, shield 55-56
volcanoes v-2, 4, 7, 9-11, 13, 21, 34, 40, 42, 46, 48, 50-53, 55-62
Big Southern Butte 36
Cima Volcanic Field 4
Craters of the Moon National Monument 5, 12-14, 38
Eyjafjallajökull volcano 1, 47, 50
Kilauea volcano 11, 35, 49, 51
Mauna Ulu eruptions of 35
Mauna Loa volcano 10
Mono Craters 47, 50
Obsidian Dome 47, 50
Mount Pelée 37
Mount St. Helens 37
eruptions of 43, 59
Ngorogoro Caldera 5
Nisga’a volcano 63
Nyamuragira volcano 55-56
Nyiragongo volcano 59
Ol Doinyo Lengai 51-53

80
 The Nature of Volcanic Rocks

Pinatubo volcano 59
planetary heat transfer and 1
Somma-Vesuvius volcano 13-14, 42-43, 59-61
Yellowstone supervolcano 20, 48, 54, 63

W
wadeite 34
Walker, G.P.L. 68, see McPhie, J. 50
Washington, H.S. 66, see Cross, W 27, 36
see also CIPW
well-shuffled deck 10
Williams, H. 19, 70
Wilson, G.V. 65, see Bailey, E.B. 46
Winter 54, 72
Wright, T.L. 67, 71, see Jackson, D.B. 35
Wright, W.B. 65, see Bailey, E.B. 46
Wuorinen, V. 53, 71
Wyoming 34

Y
Yellowstone caldera 20
Yellowstone National Park 36
Yellowstone supervolcano 20, 48, 54, 63

Z
Zussman, J. 66, see Deer, W.A. 24, 31

81