Anorthosite and labradorescence

Sometimes very abundant stuff gets undeservedly little attention. One good example is sand. Everybody seems to think that they know what it is. You cannot find less intriguing topic than that. Can you?

Well, my opinion in somewhat different. Sand is actually endlessly versatile and full of puzzles. You only need to dig deeper to see it. The same is true about some the rocks and minerals that are fairly common. Think about the plagioclase feldspar minerals. More than half of the Earth’s crust is composed of this mineral group. Therefore it isn’t really surprising that there is a rock type that is almost exclusively composed of plagioclase. It is called anorthosite (rarely named plagioclasite or labradorite as well). This rock is interesting mineralogically and petrologically but there is more to it than that. There are several traps waiting for us, and I know several people who are indeed tricked to believe wrong things.

Anorthosite from Finland (known also as spectrolite) showing labradorescence (blue spot). The width of the rock sample is 17 cm.

Because of linguistic similarities some people think that anorthosite is a rock composed of anorthite (one of plagioclase group minerals). This is not true. Anorthosite is usually composed of labradorite and sometimes bytownite or andesine as well (all plagioclase group minerals). I really don’t understand why was it necessary to engineer such a trap into which even geologists often fall.

The term “labradorite” itself is a source of confusion as well. Maybe not so much in the US because this trap was set by French and Russian geologists and creates trouble where their influence for historical reasons have been greater. The problem is that labradorite for them is a rock type containing lots of mineral labradorite. So the problem is just like the famous dolomite problem. One always has to think about what are we actually talking about — a rock or a mineral?

Anorthosite is usually defined as a leucocratic rock. Leucocratic means light-colored? Well, this is complicated. I recommend to be careful here as well. Leucocratic can be defined as ‘felsic’ or ‘not mafic’. Leucocratic minerals are those that are relatively rich in silicon and aluminum but contain little iron and magnesium. Felsic minerals (quartz, feldspar, muscovite) are generally lighter than mafic minerals (pyroxene, hornblende, biotite, olivine) but not always. Some plagioclase feldspars are definitely darker than bright green olivine crystals. And some pyroxenes are colorful as well. I again have to say that nature does not care about our classification schemes and always finds a way to laugh at us. Anorthosite in many cases is not light-colored at all. This is the result of tiny Fe-Ti oxide inclusions in plagioclase crystals that give them bluish-black hue.

One fascinating aspect associated with many anorthosites is an effect called labradorescence. It is a special form of iridescence. Plagioclase crystal is composed of many exsolution lamellae — minerals have broken up into many slabs of alternating composition. These slabs act like mirrors. Some light reflects back from the crystal surface but some portion of it penetrates the surface to be reflected back from the next lamellae which is 20…50 nanometers below the surface. Reflected lightwaves combine (this is called interference) and create peculiar colors which in our case seem to be mostly blue.

If you want to see anorthosite, you don’t have to go far – just look up at the night sky. It is a common rock type on the Moon (take a look at the post of lunar anorthosite). The highlands of the Moon surrounding the dark basaltic lava fields called maria are composed of this rock type. However, that probably is not sufficient to get to know it because 380,000 kilometers that separates us from the Moon is a bit too much.

Anorthosite is not nearly as common here as it is on the Moon. Anorthosites on the Moon are extremely old, almost as old as the Moon itself. Earth anorthosites are pretty old too, mostly from the Proterozoic Eon. Anorthosite is a plutonic rock just as granite and gabbro and it is usually associated with the latter. Therefore we can conclude that despite being officially leucocratic it actually is usually associated with mafic rocks.

Anorthosite is usually considered to be a cumulate rock. These are magmatic rocks that have crystallized from the magma which is enriched in low- or high density minerals. Anorthosite should therefore represent the upper portion of certain magma chambers where light plagioclase crystals have accumulated. However, there are lots of still unsolved questions. For example: if it really is a cumulate rock then where are the opposite mafic cumulates that are composed of pyroxene and olivine? I don’t want to say that these rocks don’t exist. They do in some other areas. These interesting rocks are collectively called peridotites. Peridotite is a common constituent of the Earth’s mantle and usually not associated with anorthosites.

One more interesting aspect is that most plutonic igneous rocks have extrusive equivalents. These pairs are gabbro-basalt, granite-rhyolite, etc. But there is absolutely no volcanic equivalent of anorthosite. Why not? And why did these rocks mostly form in the middle of the Proterozoic Eon?

Anorthosite is not an uncommon rock. It is present in nearly every continental landmass (exceptions are Greenland and Australia1) but we still know so little about it. I have to conclude this strange fact again that the the more common the rock is, the less we tend to know about it.


It came as a nice surprise to me that this post inspired several other geobloggers to write about their experiences with this rock as well. Check out the blogs of Ron Schott, Garry Hayes, Dana Hunter, Ian G. Stimpson, Callan Bentley, Ryan Jackson, and Mika McKinnon who participated in this geomeme. Please let me know if I missed your post.


1. Best, Myron G. (2002). Igneous and Metamorphic Petrology, 2nd Edition. Wiley-Blackwell.

3 comments to Anorthosite and labradorescence

Leave a Reply



You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>