Good morning,
Thank you for such precise guidelines. They are very helpful.
Best regards,
Jorge J. Restrepo
----- Original message -----
From: "Clemens, JD, Prof <[log in to unmask]>" <[log in to unmask]>
To: [log in to unmask]
Date: Tue, 20 Sep 2011 17:08:21 +0200
Subject: distinguishing I- and S-type granites
Hi,
Mike Brown suggested that what people need are some guidelines for
recognising S- and I-types. Here are some simple quite reliable
mineralogical criteria.
If your rock has red-brown biotite with high Al and Ti contents or
cordierite or monazite or pyrrhotite as the Fe sulphide then it's almost
certainly an S-type.
If your rock has hornblende or clinopyroxene, or titanite or magmatic
epidote or the biotite is dull brown to greenish brown or the rock
contains magnetite or alanite, you've got an I-type.
Muscovite, andalusite and garnet can occur in both sorts, especially is
you have fractionated highly felsic magmas involved. My impression is
that muscovite is more common in fractionated I-types than it is in
S-types.
Isotope characteristics should be used with circumspection. There are
high delta 18O I-types and I-types with very ancient sources can aslo
have high initial Sr isotope ratios and low epsilon Nd.
Likewise there are chemical indicators – Na2O/K2O is generally higher in
I-types, as is CaO at a given SiO2 content. ASI (= A/CNK) is a tricky
one. A lot of granites are peraluminous, but very few I-types have ASI >
1.1. There is commonly a rather telling variation in ASI with maficity
(Fe + Mg). In S-types these are commonly positively correlated, due to
the fact that mafic minerals responsible for the variation in maficity
are peraluminous. In I-types, the trend is opposite, with the more mafic
rocks having lower ASI. Accumulation of Hbl or Cpx does this.
So, as I wrote earlier, it is the totality of evidence that points to
the character of the source rock. Arriving at the point of knowing
something pretty specific about the source is rather superior to knowing
that you have an alkali-calcic granite (which could be either S- or
I-type, actually). In short, beware of magic discriminant diagrams. They
can work rather poorly and totally mislead you. Part of the reason for
this, especially with tectonic discriminant diagrams, is incorrect
assignment of the character of the rocks that were used to construct the
field boundaries in the first place and incorrect assumptions regarding
the sources of felsic magmas.
I have spoken.
JC
John D. Clemens
Professor in Geology & Exec. Head
Dept of Earth Sciences, University of Stellenbosch,
Private Bag X1, 7602 Matieland, South Africa
tel: +27 (0)21 808 3159 fax: +27 (0)21 808 3129
e-mail: [log in to unmask]
web page: http://www.sun.ac.za/geo/people/clemens_e.htm
|