JOVANA
Explore Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
Back to the library
Earth Science 1922

The Reaction Principle in Petrogenesis

Norman L. Bowen

Cool one magma slowly and, crystal by crystal in a fixed order, it becomes nearly every igneous rock.

Choose your version
In depth · the introduction

Melt a rock, then cool it slowly, and it freezes back not all at once but in a strict order — and that order is why the Earth has so many different rocks.

The big idea

Volcanic rock starts as magma — molten rock, mostly a runny, dark, basalt-like liquid. As magma cools, it doesn't freeze all at once. Different minerals crystallize at different temperatures, in a fixed sequence. The first crystals to appear are dark, heavy minerals like olivine; the last to appear, only when things are much cooler, is pale quartz.

Bowen's discovery was that the early crystals keep reacting with the leftover liquid as it cools, trying to convert into whatever is stable at the new temperature. And here's the twist: if those early crystals are scooped out of the melt — for instance, by sinking to the bottom — the liquid that remains is changed. Do that over and over, and a single dark basalt magma can slowly turn into the pale, silica-rich liquid that makes granite.

How it came about

For a long time geologists were puzzled: igneous rocks come in a huge range, from dark, dense basalt to pale, light granite. Did each kind need its own special source deep in the Earth? That seemed wasteful and unlikely.

Norman Bowen, working at the Carnegie Institution's Geophysical Laboratory in Washington in the early 1900s, attacked the question in the lab. He melted mixtures of the actual minerals, cooled them under controlled conditions, and watched which crystals formed first, second, third. Out of those experiments came his 1922 paper and, six years later, his book The Evolution of the Igneous Rocks — and with them the elegant idea that one magma, crystallizing in order and shedding its early crystals, could be the parent of almost the whole rock family.

Why it mattered

Bowen turned a zoo of rock types into a single story. Instead of needing a separate origin for every rock, geologists could see them as stages along one cooling, fractionating magma. The idea reaches deep: the silica-rich granites that make up so much of the continents can be understood as the late, refined leftovers of darker, deeper magmas. It is one of the organizing ideas of all igneous geology — even as later geologists showed it is not the whole story.

A way to picture it

Think of slowly freezing very salty water. Pure ice crystals form first and float — and as you skim them off, the water left behind gets saltier and saltier, freezing at a lower and lower temperature. Cooling magma does the same with minerals instead of ice: the early crystals are 'skimmed' out (they sink), and the leftover melt drifts steadily toward a different, more silica-rich recipe. Same starting liquid, very different final rock.

A reaction-series diagram with a temperature slider: as you cool the magma from about 1200°C to 700°C, olivine then pyroxene then amphibole then biotite crystallize in turn on one branch, a plagioclase bar shifts from calcium-rich to sodium-rich on the other, and quartz forms last; a side bar shows the leftover melt becoming more silica-rich, labelled basalt, then andesite, then rhyolite.

Where it sits

A century earlier, Hutton (1788) and Lyell (1830) had shown that rocks endlessly cycle and that the Earth is unimaginably old. Bowen put chemistry inside that cycle, explaining how molten rock turns into the family of igneous rocks we see. Later, plate tectonics revealed where the parent basalt magmas come from in the first place — the melting mantle beneath mid-ocean ridges (Hess 1962; Vine–Matthews 1963) — and showed that mixing, contamination and water complicate Bowen's clean picture. But his reaction series remains the first thing every geologist learns about how a rock is born.

The original document
Original source text
N. L. Bowen · The Journal of Geology 30(3) (1922): 177–198 · Geophysical Laboratory, Carnegie Institution
Bowen sets out to explain the central puzzle of igneous petrology: why a single broad parent — a basaltic magma — gives rise to such a wide variety of rocks, from dark, dense basalts and gabbros to pale granites and rhyolites. His answer rests on a principle he draws from the laboratory's phase-equilibrium experiments on silicate melts.
The reaction principle
Crystals that form early from a cooling melt do not simply accumulate unchanged. As the temperature falls, each early crystal tends to react with the surrounding liquid, converting toward the mineral that is stable at the new, lower temperature. Crystal and liquid are continually trying to keep in step — this is the reaction relation, and it governs the whole sequence.
Two kinds of reaction series
Bowen distinguishes two patterns. In the discontinuous series, a mineral reacts with the melt and is replaced by a structurally different mineral at a definite step: olivine gives way to pyroxene, pyroxene to amphibole, amphibole to biotite. In the continuous series, a single mineral — plagioclase feldspar — stays the same mineral but changes its composition smoothly as it cools, from calcium-rich toward sodium-rich.
Fractional crystallization
The principle has a powerful consequence. If the early crystals are physically removed from the melt — settling out under gravity, or filtered away — they can no longer react back, and the leftover liquid is left progressively richer in silica and the low-temperature constituents. Step by step, one basaltic liquid evolves toward intermediate and then felsic compositions. A single parent magma can therefore yield a whole descending series of rocks.
[ … ]
Scope and the larger claim
Bowen argues that this fractionation, more than any other process, accounts for the diversity of the igneous rocks, and weighs it against rival explanations such as the wholesale mixing of magmas or large-scale assimilation of older rock. The reaction series he describes here he later consolidated into the now-famous branching diagram of his 1928 book, The Evolution of the Igneous Rocks.
Geophysical Laboratory, Washington · 1922