Skip to content
Top curves

Viscosity of Magmas

The nature of volcanic eruptions is highly dependent on magma viscosity and also on dissolved gas content.

Magma Composition:

Silicate-rich magmas are typically formed at destructive plate boundaries, by partial melting and/or assimilation of crustal rocks (which are richer in silica than the rock of the mantle). Such magmas erupt as andesites and rhyolites or are intruded as granite masses. The more extensive silicate chain molecules render these magmas highly viscous, so when eruption occurs it is usually explosive (e.g. Mt St Helens).

Low-silica magmas are typically formed by partial melting of mantle rocks beneath mid-ccean ridges or at “hot spots” like Hawaii. These magmas erupt as basalts or intrude as gabbro, and are far less viscous. Eruptions are generally effusive.


Magma temperatures reflect the melting points of their mineral components. Not surprisingly, magmas formed by partial melting of mantle rocks are much hotter – well over 1200oC for some Hawaiian basalts – than is the case for crustally derived melts. Rhyolites may reach the surface at temperatures of less than 900oC, and so have much higher viscosity.

Volatile Content:

Magma invariably contains small amounts of dissolved gas (water, CO2 etc) which is released as pressure is removed. Magmas formed by melting of mantle rocks have generally low volatile contents, but those formed by partial melting of crustal rocks are often volatile-rich. A high volatile content decreases viscosity (like adding water to treacle), and is probably the main factor in enabling some highly viscous (but also volatile-rich) melts to reach the surface at all. The release of gas during eruption is particularly likely to be explosive if the magma is both viscous (as gas is released, so viscosity is immediately increased) and volatile rich.

Crystal Content:

Some magmas have already begun to crystallise by the time they reach the surface. Again, this applies particularly to the cooler, more viscous magmas typical of destructive plate margins. A “crystal mush” will clearly have increased viscosity.


The "Treacle Test" experiment is designed to enable students to investigate how temperature, volatiles and crystals may affect viscosity. It is suggested that viscosity is roughly measured by how long it takes the treacle to flow from one end of a boiling tube to the other.
Bottom curves