Thanks to one minute of the Earth we know that beach sand mother rock, besides the quartz itself, is (or was?) made of olivine, amphibole and orthoclase. I wonder, what is the name of that rock and where do the 3 annihilated minerals go if only quartz is left?

  • $\begingroup$ Short answer: the rock name is granite (it has no olivine by the way). The other minerals turn into clay (also known as dust, or mud). $\endgroup$
    – Gimelist
    Aug 23, 2016 at 9:39

1 Answer 1


In only one minute of presentation there isn't time for detailed discussion, so one has to make broad generalizations which, on closer inspection, aren't quite true. It is true that most sand is made of quartz, and that most of this quartz arises from weathered granitic rocks, of which the main minerals are quartz, orthoclase and plagioclase, usually with some mica, quite rarely some amphibole and almost never olivine. In the magma olivine and quartz react to make either amphiboles (if it is a very hydrous melt) or pyroxenes.

It is possible to find a sedimentary rock made of weathered out quartz and feldspar grains. Depending upon the original chemistry the feldspar could be potassium-rich (orthoclase) or sodium-calcium-rich (plagioclase). The resulting sedimentary rock is known as an arkose. Arkose is common in dry areas of eroding granite, such as the dryer parts of sub Saharan Africa. However, it is much more common for the weathering to involve wet erosion - thousands of cycles of rain and sun, hot and cold, in which case the feldspars mostly decompose to clay (many varieties and sub-species).

Olivine is much more characteristic of basic volcanic rocks, especially basalts, which do not weather to yield quartz grains. Another rock type in which olivine is a major component is peridotite. This weathers to yield goethite (a kind of ironstone) and serpentine. Occasionally, if the pyroxene to olivine ratio is sufficiently high, the weathering can yield quartz as a secondary by-product.

Another feature of quartz: Over long periods and especially at high temperatures and/or high pressures, it is very slightly soluble. Not that you would notice. The sand isn't going to dissolve beneath your feet, but every litre of surface water contains a few milligrams of dissolved quartz as silicic acid, which takes part in a whole catalogue of geological processes. In addition, there are many other rocks besides granite which yield quartz grains, including quartzite, weathered pre-existing sandstone, schist, gneiss, and some intermediate rocks (complex chemistry between granite and basalt. Even limestone - rocks that are generally thought of as 'pure' calcium carbonate, generally have a percent or two of insoluble residue, of which quartz is the dominant component.

  • $\begingroup$ "In the magma olivine and quartz react" - quartz and Mg-olivine never coexist at any stage in magmas. Basaltic magmas do not crystallise quartz, and granitic magmas do not crystallise olivine. $\endgroup$
    – Gimelist
    Aug 23, 2016 at 9:41
  • $\begingroup$ Michael is right, as most olivine is about 85% Mg end-member, but if I remember correctly, fayalite, the much rarer iron end-member version of olivine, can co-exist with high temperature silica in the form of a fayalite-tridymire eutectic. $\endgroup$ Aug 23, 2016 at 12:58
  • $\begingroup$ "much rarer" is an understatement when discussing igneous rocks. It exists maybe only in charnockites. $\endgroup$
    – Gimelist
    Aug 23, 2016 at 19:11
  • $\begingroup$ Michael - correct again. OK, for 'much rarer', read 'extremely rare'. I have only ever read about it - in 40 years as a geologist I have never actually seen fayalite-tridymite in the same rock. $\endgroup$ Aug 24, 2016 at 3:51
  • $\begingroup$ I see now that orthoclase and plagioclase are just feldspars. This means that every 3 molecules of SiO2 has one molecule of Al and some other molecule attached. How can weathering weather the 'crystals' of the Al (K | Na | Ca) out? Separate atoms of metal can hardly be called 'crystals', IMO. $\endgroup$ Aug 31, 2016 at 7:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.