Those interested in Frank's question regarding the origin of
"pseudosections" may, or may not, find the following helpful...

As Bas Hensen said, Roger Powell is probably responsible for the current
usage of "pseudosection" (i.e. phase diagrams depicting stable multivariant
assemblage fields for specified bulk rock compositions in P-T-bulk X
space), and it's probably worth emphasising, as Frank Spear did in his
original message and Prof Banno realised when listening to the European
petrologist, that these pseudosections are NOT equivalent to the
"pseudo-binary diagrams" used by Alan Thompson in his 1976 Am J Sci paper.

The earliest use of the term pseudosection I can find by Roger is in
Chapter 9 of his book (Powell 1978, Equilibrium Themodynamics in Petrology,
Harper & Row Publishers, London).  Roger describes two types of T-XMg
diagrams on p. 219-223.

The first diagram shows stability fields of divariant/trivariant
assemblages with changing T for a fixed bulk rock value of A/FM (i.e.
single horizontal section across an AFM diagram).  Roger calls these
"pseudobinary T-X sections" and states that "this section is only
pseudobinary because the compositions of the phases in the assemblage do
not plot in the plane of the section".  Note that the XMg axis in these
sections denotes changing Fe-Mg bulk rock composition at fixed A/FM ratio.
The diagram shows which phase assemblages are stable (hence Frank's
suggestion that they could be called assemblage stability diagrams), but
give no information on phase composition.  They simply show the XMg values
at which diviarant and trivariant field boundaries intersect the chosen
A/FM section, and how these values change with temperature.  No additional
projection is involved in the construction of these diagrams: all the
information is already in the section of the diagram.

The second diagram plots changing XMg values of individual phases in each
divariant assemblage with changing temperature.  Each divariant reaction is
a loop with one boundary defined by the compositional variation of the most
Fe-rich phase, and the other by the compositional variation of the most
Mg-rich phase.  These diagrams can depict loops for all divariant
assemblages in an AFM diagram, including assemblages stable at different
A/FM values.  They are projections of mineral compositions onto an
arbitrary horizontal Fe-Mg section across an AFM diagram, and the XMg axis
represents XMg in each phase (not bulk rock XMg).  Roger calls these
diagrams "T-X projections".

Roger's pseudobinary sections are equivalent to the phase diagrams
introduced to metamorphic petrology by Bas in his 1971 paper, although this
latter paper dealt with P-X rather than T-X sections and did not use the
term pseudosection.  Bas also showed how several P-X sections constructed
for different temperatures can be combined for a fixed XMg to derive P-T
diagrams depicting the stability of univariant and multivariant assemblages
for a single bulk composition.  Although such P-T diagrams are not
discussed in Roger's book, he and his co-workers refer to them as P-T
pseudosections in numerous later papers, presumably using the "psuedo"
terminology since they are directly comparable to the pseudobinary T-X
sections, and just like the T-X sections the compositions of the individual
phases present in the assemblages are not equivalent to the composition for
which the diagram is constructed.  All sorts of pseudosection can be found
in the literature (P-T, T-XMg, T-aCO2 etc), particularly in papers written
by Roger's group.  It is now common pratice to construct such diagrams with
the Holland and Powell data set, with the result that the term
"pseudosection" is widely used in those parts of the world where THERMOCALC
is popular.

Roger's T-X projections are equivalent to the "pseudo-binary T-X diagrams"
used by Alan Thompson, which is unfortunate as far as terminology goes.
Alan dos not say in his paper why he uses the term pseudo-binary, although
Frank gives a possible explanation in his message.  A good question would
be why Roger used the term pseudosection for the "phase assemblage
diagrams" but saw nothing pseudo in his T-X projections.  I note that Roger
lists Alan's 1976 paper in the reference list at the end of Chapter 9, and
so was presumably aware of Alan's terminology.  As Bob Tracy points out we
would need to consult work in other fields to check the earliest
descirptions of "pseudo phase diagrams", but Roger has certainly been
consistent and labels phase diagrams in all his papers as either being T-X,
P-X and P-T projections, or T-X, P-X and P-T pseudosections.

I guess we really need to ask Roger if we want an answer to Frank's question!


Ian Fitzsimons
Tectonics Special Research Centre
Department of Applied Geology
Curtin University of Technology
GPO Box U1987, Perth, WA 6845
Phone (Direct line): +61 8 9266 2455
Phone (Dept Office): +61 8 9266 7968
Fax: +61 8 9266 3153
E-mail: [log in to unmask]