Petrograph: a New Software to Visualize, Model, and Present Geochemical Data in Igneous Petrology

Abstract

A new software, PetroGraph, has been developed to visualize, elaborate, and model geochemical data for igneous petrology purposes. The software is able to plot data on several different diagrams, including a large number of classification and “petrotectonic” plots. PetroGraph gives the opportunity to handle large geochemical data sets in a single program without the need of passing from one software to the other as usually happens in petrologic data handling. Along with these basic functions, PetroGraph contains a wide choice of modeling possibilities, from major element mass balance calculations to the most common partial melting and magma evolution models based on trace element and isotopic data. Results and graphs can be exported as vector graphics in publication-quality form, or they can be copied and pasted within the most common graphics programs for further modifications. All these features make PetroGraph one of the most complete software presently available for igneous petrology research.

Full text

PetroGraph: A new software to visualize, model, and present
geochemical data in igneous petrology
M. Petrelli, G. Poli, D. Perugini, and A. Peccerillo
Department of Earth Sciences, University of Perugia, Piazza Universita`, 1, 06100 Perugia, Italy (maurip@unipg.it)
[1] A new software, PetroGraph, has been developed to visualize, elaborate, and model geochemical data
for igneous petrology purposes. The software is able to plot data on several different diagrams, including a
large number of classification and ‘‘petrotectonic’’ plots. PetroGraph gives the opportunity to handle large
geochemical data sets in a single program without the need of passing from one software to the other as
usually happens in petrologic data handling. Along with these basic functions, PetroGraph contains a wide
choice of modeling possibilities, from major element mass balance calculations to the most common partial
melting and magma evolution models based on trace element and isotopic data. Results and graphs can be
exported as vector graphics in publication-quality form, or they can be copied and pasted within the most
common graphics programs for further modifications. All these features make PetroGraph one of the most
complete software presently available for igneous petrology research.
Components: 4646 words, 10 figures, 4 tables.
Keywords: data management; data plotting; geochemical modeling; petrology; software.
Index Terms: 3610 Mineralogy and Petrology: Geochemical modeling (1009, 8410); 1065 Geochemistry: Major and trace
element geochemistry; 1094 Geochemistry: Instruments and techniques.
Received 3 February 2005; Revised 5 May 2005; Accepted 11 May 2005; Published 26 July 2005.
Petrelli, M., G. Poli, D. Perugini, and A. Peccerillo (2005), PetroGraph: A new software to visualize, model, and present
geochemical data in igneous petrology, Geochem. Geophys. Geosyst., 6, Q07011, doi:10.1029/2005GC000932.
1. Introduction
[2] Handling, visualization, and modeling of geo-
chemical data are of fundamental importance in
igneous petrology. Most of these operations can
be performed using a variety of software pack-
ages such as spreadsheet programs (e.g., MS
Excel
1
or Microcal Origin
1
), interpreted scien-
tific computer languages (e.g., R
1
or Matlab
1
),
or software specifically developed for igneous
petrology such as MinPet
1
(MinPet
1
Geological
Software Inc. Web site: http://www.minpet.com/),
GCDKit [Janousek et al., 2003], IgPet
1
(IgPet
1
RockWare Web site: http://www.rockware.com/)
and PetroPlot [Su et al., 2003]. Moreover, a large
number of codes have been developed [ e.g.,
Wright and Doherty, 1970; Stormer and Nicholls,
1978; Woussen and Coˆte´, 1987; Conrad, 1987;
Holm, 1988, 1990; Nielsen, 1988; Defant and
Nielsen, 1990; Harnois, 1991; Benito and Lo´pez-
Ruiz, 1992; D’Orazio, 1993; Verma et al., 1998;
Keskin, 2002; Spera and Bohrson, 2001] to help
researchers in performing specific geochemical
models. All these software allowed users to
greatly speed up modeling of geochemical data,
but some of them are designed for old MS-DOS
operating systems and cause problems when
loaded into modern MS Windows systems.
Other programs have limited possibility of
geochemical modeling (e.g., MinPet, GCDKit
and PetroPlot). Moreover, problems can arise
when trying to run algorithms available only as
source codes; for example, the mass balance
algorithm by Stormer and Nicholls [1978]
requires a FORTRAN compiler t o be loaded,
and this may cause difficulties for the user if
G
3
G
3
Geochemistry
Geophysics
Geosystems
Published by AGU and the Geochemical Society
AN ELECTRONIC JOURNAL OF THE EARTH SCIENCES
Geochemistry
Geophysics
Geosystems
Technical Brief
Volume 6, Number 7
26 July 2005
Q07011, doi:10.1029/2005GC000932
ISSN: 1525-2027
Copyright 2005 by the American Geophysical Union 1 of 15

compared to modern object-oriented, user-friendly
software.
[
3] In this contribution we present a new pro-
gram, PetroGraph (Figure 1), developed with the
aim of giving a complete platform to handle,
visualize and model geochemical data in a user
friendly envi ronment. The program runs under
Windows 98/2000/XP
1
platformsasastand-
alone application. The source code is written in
MS Visual Basic 6.0
1
and is distributed together
with the program. The code is open source and
we invite all users to contribute to the develop-
ment of the software by proposing improvement
or developing new codes. To download the soft-
ware and the tutorial, please go to http://www.
unipg.it/maurip/SOFTWARE.htm (see also ancil-
lary material).
[
4] In the following sections we present the main
features and potentialities of PetroGraph. The pa-
per is divided into five main sections (Figure 2),
including (1) input of geochemical data, (2) data
visualization, (3) development of geochemical
models, (4) data management, and (5) additional
features.
2. Data Input
[5] Geochemical data can be imported into Petro-
Graph in three file formats (Figure 2a): (1) MS
Excel
1
worksheets, (2) IgPetWin
1
formatted files,
and (3) PetroGraph (.peg) files.
[
6] MS Excel
1
worksheets need little formatting
before imported to PetroGraph, as reported in
Figure 2. The arrangement of the worksheet, how-
ever, is not much different from common analytical
outputs. In particular, the first column must contain
the sa mple name, whereas the following three
columns are dedicated to symbol and color man-
agement and to the choice to show (1) or not to
show (0) the sample in the diagrams (see the
tutorial for a detailed explanation of the arrange-
ment of the MS Excel
1
file).
Figure 1. General screen shot of PetroGraph showing some types of graphs performed by the software. (a) Total
Alkali-Silica, TAS diagram [Le Bas et al., 1986], (b) TiO
2
versus SiO
2
binary plot, (c) AFM ternary plot [Kuno,
1968], and (d) Condrite normalized REE diagram [Haskin et al., 1968]. Displayed data are from Pecceril lo et al.
[2003].
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
2of15

[7] A specific code has been written to import
IgPet
1
files (.roc format) without modifications
of the original file format.
[
8] PetroGraph can save data into text files by
applying the .peg extension. These files can be
opened by the software much faster than MS Excel
formatted spreadsheets.
3. Visualization
[9] Once data have been imported, the user can
visualize them (Figure 2b) using three different
types of diagrams commonly used in igneous
petrology: binary, ternary, and spider diagrams
(Figure 1). All diagrams can be easily generated
in few steps. For example, a binary or a ternary
diagram can be created in only 3 steps as de-
scribed in Figure 3. Several options are presented
to the user for each type of diagram. For instance,
data can be plotted in binary diagrams by using
linear and logarithmic scaling; maximum and
minimum values of the axis can be rearranged to
zoom into specific areas of the graphs. In addition,
a large number of symbols and colors can be
selected to visualize data. The most common
Figure 2. PetroGraph block diagram explaining the main features of the program.
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932petrelli et al.: petrograph software 10.1029/2005GC000932
3of15

normalizations can be chosen for rare earth ele-
ments (REE) normalized diagrams (Figure 4a;
Table 1). For example, REE data can be normal-
ized by Condrite or NASC values by selecting
the relative normalization values in the REE
spiders window (arrow in the REE spiders win-
dow; Figure 4a).
[
10] Additional spider diagrams can be also per-
formed using several normalizations (Figure 4b;
Table 1). For example, data can be normalized by
primordial mantle [Wood et al., 1979a], mid-ocean
ridge basalts (MORB) [Bevins et al., 1984], and
continental crust [Tayl or and McLennan, 1981;
Weaver and Tarney, 1984] by selecting them in
Figure 3. Screen shot of the program showing the procedure to plot a binary and ternary diagram. (a) Button of
the control bar that opens the ‘‘binary plot window,’’ (b) window that allows the user to customize and plot a binary
diagram, (c) button that allows the user to open (f) the windows in which elements can be easily selected,
(d) example of binary plot, (e) note that name and the coordinate of samples can be displayed on the lower part of
the main windows by tracking on them with the pointer, (g) button of the control bar that opens the ‘‘triangular plot
window,’’ (h) window that allows the user to customize and plot a triangular diagram, and (i) example of a
triangular plot.
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
4of15

the relative spider window (arrow in the general
spider window; Figure 4b). An important feature of
PetroGraph is the possibility to generate normali-
zation files that can be used to generate custom
spider diagrams.
[
11] One of the major tasks in igneous petrology is
rock classification and this can be easily done with
PetroGraph by using a wide choice of classification
diagrams (e.g., Q’-AN OR, TAS, etc.; Table 2);
several ‘‘petrotectonic’’ diagrams (e.g., Nb-Y, Ti-
Zr-Y, etc.; Table 2) can be also generated [Rollison,
1993]. For example, the Total Alkali versus Silica
diagram (TAS) [Le Bas et al., 1986] can be created
simply selecting it from the Diagram section of the
Plot menu (Figure 5).
[
12] All diagrams can be easily modified by a
mouse double click (Figure 6b) and a cascade
menu containing several options can be opened
by a right click (Figure 6a). All these options allow
the user to generate high-quality graphs that can be
saved to file in publication-ready form in Micro-
soft
1
metafile format or copied to the clipboard to
Figure 4. Screen shot of the program showing the procedure to plot a spider diagram. (a) Procedure to plot a REE
spider diagram. (b) Procedure t o plot a general spider diagram. Arrow in the REE spiders window: possible
normalizations for REE spiders. Arrow in the general spiders window: possible normalizations for general spiders.
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
5of15

be further elaborated in the major vector-graphic
applications (e.g., MS Office
1
, Corel
1
).
4. Modeling of Geochemical Data
4.1. Major Element Modeling
[13] Several modeling approaches by using major
and trace elements, and isotopes are implemented
in PetroGraph (Figure 2c).
[
14] Major elements are modeled by the mass
balance algorithm [Stormer and Nicholls, 1978].
This approach involves a least squares solution to
a set of linear mass balance equations (one for
each oxide) and the calculations are performed on
data consisting of chemical analyses of igneous
rocks, assumed to represent the composition of
parent and daughter magmas. This computational
approach can be used to test fractional crystalliza-
tion, assimilation, fractional melting, and magma
mixing [e.g., Stormer and Nicholls, 1978]. Mass
balance computations are performed in a user
friendly env ironment (Figure 7) by selecting
‘‘magmas’’ and ‘‘phas es’’ among the samples
belonging to the geochemical data set. Figure 7
shows the procedure to develop a Mass Balance
computation using PetroGraph: first select the
oxides (Figures 7b–7e), successively select the
initial and final ‘‘magmas’’ (Figures 7c– 7f), then
select the ‘‘phases’’ and finally calculate results
(Figure 7h).
4.2. Trace Element and Isotope Modeling
[15] Trace elements models are divided into three
main sections, involving (1) magma crystallization
processes (Table 3), (2) partial melting (Table 4,
top), and (3) magma mixing (Table 4, bottom).
PetroGraph can model mixing [Langmuir et al.,
1978] and Assimilation plus Fractional Crystalli-
zation (AFC) [DePaolo, 1981] processes, using
isotopic data.
[
16] Trace element and isotope models can be
performed by choosing the model end-members
direc tly from the plotte d samples; this strongly
simplifies the modeling procedure and allows the
user to immediately evaluate if the selected model
is suitable or not for the studied data set. Once a
model has been plotted on a graph, model param-
eters can be readily modified in the appropriate
window that will pop up with a single click of the
mouse directly on the graph. Different models can
be displayed simultaneously either in a single
graph or in different graph windows.
[
17] Figure 8 reproduces two diagrams extracted
from Peccerillo et al. [2003] and shows, as an
example, how to perform trace element and iso-
tope models with PetroGraph. The models refer to
the origin of peralkaline silicic magmas at
Gedemsa volcano (Central Ethiopian Rift), which
can be generated either by batch melting of a
basaltic rock or by fractional crystallization of a
basaltic parental melt. Trace element and isotope
models reported in Figure 8 show that fractional
crystallization accompanied by little assimilation
of the Precambrian crust is more suitable than
batch-melting to explain sample variability in the
Gedemsa magmatic system (see Peccerillo et al.
[2003] for a detailed discussion of these data)
and these models can be easily developed using
PetroGraph.
5. Data Management
[18] Data management (Figure 2d) can be per-
formed by using four principal types of operation:
(1) algebraic operations, (2) determination of geo-
chemical parameters or indexes, (3) operation re-
lated to rare earth elements, and (4) operation on
isotopes (Figure 9a).
[
19] Algebraic operations are the sum, subtraction,
division, multiplication, the elevation to an expo-
Table 1. Normalizations Available in PetroGraph for
REE and General Spider Diagrams
Normalization Source
REE Spiders
Chondrite Haskin et al. [1968]
Chondrite Masuda et al. [1973]
Chondrite Nakamura [1974]
Chondrite Boynton [1984]
Chondrite Sun and McDonough [1989]
NASC Haskin and Frey [1966]
NASC Haskin and Haskin [1966]
General Spiders
Primordial mantle Wood et al. [1979a]
Primordial mantle McDonough et al. [1992]
Primordial mantle Taylor and McLennan [1985]
Condrite Wood et al. [1979b]
MORB Bevins et al. [1984]
Upper cont. crust Taylor and McLennan [1981]
Lower cont. crust Weaver and Tarney [1984]
Average cont. crust Weaver and Tarney [1984]
Average N-type MORB Saunders and Tarney [1984],
Sun [1980]
Average OIB Sun [1980]
Custom spider allows user to generate custom
normalization file
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
6of15

nent, and the square root. In addition, it is possible
to convert parts per million (PPM) values to weight
percent (wt%) and vice versa. These operations
allow the user to generate new variables that can be
plotted and elaborated analogously to the original
ones.
[
20] Geochemical para meters and indexes tha t
can be determined by PetroGraph are Total Iron
(FeO
tot
), Larsen Index, Solidification Index (SI),
CIPW norm, Magnesium Number (Mg#), Alumin-
ium Saturation Index (ASI), and Fe/Mg Ratio.
[
21] Regarding operations related to REE, Petro-
Graph can calculate the Europium anomaly (Eu/
Eu*), three different normalized REE ratios (La
n
/
Sm
n
,La
n
/Yb
n
,Tb
n
/Yb
n
) and the total sum of REE
(SREE).
[
22] Regarding isotopes, PetroGraph offers t he
opportunity to use the epsilon notation for Nd
and Hf isotopes and to calculate the percent devi-
ation from present-day chondritic value for
187
Os/
188
Os.
6. Additional Features
[23] Among additional features (Figure 2e), Petro-
Graph offers the opportunity to filter data sets by
applying several kinds of constraints and allows
the user to consult a solid/liquid partition coeffi-
cient database, a useful option when developing
trace element geochemical models.
[
24] Filters (Figures 9b and 9c) are useful when
only samples with specific compositional charac-
teristics (e.g., only samples with a content of SiO
2
higher than 50 wt%) are to be selected for plotting
(Figure 9c).
[
25] A complete data set of partition coefficients is
stor ed in PetroGraph (Figure 10). Data for the
Table 2. Classification and Discriminating Diagrams Performed by PetroGraph
Diagram Source
General Classification Diagrams
Binary
[Q
0
(F
0
) - ANOR] – volcanic after Streckeisen and Le Maitre [1979]
[K
2
O-SiO
2
] after Peccerillo and Taylor [1976]
[K
2
O-SiO
2
] after Middlemost [1975]
[TAS Alkalis - Silica] – volcanic after Le Bas et al. [1986]
[TAS Alkalis - Silica] – volcanic after Cox et al. [1979]
[TAS Alkalis - Silica] – plutonic after Cox et al. [1979]
[SiO
2
-K
2
O Andesite Types] after Gill [1981]
[SiO
2
- F/M] after Miyashiro [1974]
Triangular
AFM after Kuno [1968]
AFM after Irvine and Baragar [1971]
Diagrams for Basalts
Binary
[Ta/Yb - Tb/Yb] after Pearce [1982]
[Y - Cr] after Pearce [1982]
[Ti - Zr] after Pearce and Cann [1973]
Triangular
[Ti - Zr - Y] after Pearce and Cann [1973]
[Ti - Zr - Sr] after Pearce and Cann [1973]
[Nb - Zr - Y] after Meschede [1986]
[Th - Hf - Ta] after Wood [1980]
Diagrams for Granites
[Nb - Y] after Pearce et al. [1984]
[Ta - Yb] after Pearce et al. [1984]
[Rb - (Y + Nb)] after Pearce et al. [1984]
[Rb - (Yb + Ta)] after Pearce et al. [1984]
Mantle End-Members
[
87
Sr/
86
Sr -
143
Nd/
144
Nd] data from Hart et al. [1992]
[
206
Pb/
204
Pb -
143
Nd/
144
Nd] data from Hart et al. [1992]
[
206
Pb/
204
Pb -
87
Sr/
86
Sr] data from Hart et al. [1992]
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
7of15

Figure 5. Screen shot of the program showing the procedure to plot classification or ‘‘petrotectonic’’ diagrams.
(a) Procedure to select a diagram; (b) example of a TAS diagram.
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
8of15

Figure 6. Screen shot of the program showing options to customize a binary plot. (a) Cascade menu generated by
click of the right mouse button on the diagram; (b) window opened by a double click of the left mouse button on the
diagram to change axis properties.
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
9of15

partition coefficient database are from Earth Ref-
erence Data and Models Web site (EarthRef; http://
www.earthref.org).
7. Summary
[26] PetroGraph is a program specifically devel-
oped to visualize, elaborate, and model geochem-
ical data. It runs on Microsoft
1
Windows 98/2000/
XP platforms and it is written in Visual Basic
1
6.0.
With its user friendly design, it is able to plot data
within binary, triangular and spider diagrams with
minimum effort. A large number of classification
and discriminating diagrams can be easily plotted;
several operations can be performed on the original
variables in order to obtain new variables and
geochemical parameters. A large number of geo-
chemical models can be calculated from major and
trace elements, and isotopes. Moreover, Petro-
Figure 7. (a) Mass Balance window and results of the mass balance calculation. (b, c, and d) On the upper part of
the Mass Balance window are reported buttons that open the windows to select (e) oxides, (f) ‘‘magmas,’’ and
(g) ‘‘phases.’’ The computation can be performed by clicking (h) the ‘‘Calculate’’ button. The output is displayed in
the lower part of the window. Results can be also exported into the clipboard or saved as text files. Data and results
reported in the presented example are the same as in the original paper by Stormer and Nicholls [1978]. Detailed
information on step-by-step procedures to perform mass balance computations is reported in the software tutorial.
Table 3. Trace Element Models Involving Crystallization Processes
Model Abbreviation
Equilibrium Crystallization [Wood and Fraser, 1976] EC
Fractional Crystallization [Neuman et al., 1954] FC
Assimilation plus Fractional Crystallization [DePaolo, 1981] AFC
In Situ Crystallization [Langmuir, 1989] In Situ C
Zone Refining [Richter, 1986] ZR
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
10 of 15

Table 4. Trace Element Models Involving Melting and Mixing Processes
Model Abbreviation
Melting Models
Batch Melting [Wood and Fraser, 1976] BM
Non Modal Batch Melting [Wood and Fraser, 1976] nMBM
Fractional Melting [Wood and Fraser, 1976] FM
Mixing Model
Mixing [Langmuir et al., 1978] Mix
Figure 8. Screen shot of the program explaining the potentialities of PetroGraph in performing trace element
modeling. (a) Window that allows the user to customize trace element models. (b) V versus Zr plot in which
Fractional Crystallization (D
V
=4.0,D
Zr
= 0.1) and Batch Melting (D
V
=4.0,D
Zr
= 0.1) models are reported. It is
clear that Gedemsa rocks’ behavior can be well explained by fractional crystallization, whereas batch melting fails in
accounting for the sample variability. (c) Sr versus
87
Sr/
86
Sr plot reporting the Assimilation and Fractional
Crystallization (AFC) model. Isotopic modeling corroborates the hypothesis indicating that fractional crystallization
couples with moderate assimilation of Precambrian crust can suitably account for Sr isotopic signature of Gedemsa
rocks; model parameters are reported in the graph. Data are from Peccerillo et al. [2003]. Detailed information on
step-by-step procedures to perform geochemical models is reported in the software tutorial.
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
11 of 15

Figure 9. Screen shot of the program showing potentialities of PetroGraph in (a) data management and (b and c) the
data filtering window.
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
12 of 15

Graph produces high-quality graphic outputs
which can be directly used for publication.
Acknowledgments
[27] We acknowledge the useful suggestions and criticisms of
Yoshiyuki Tatsumi (Associate Editor) and two anonymous
referees. The editorial handling of W. M. White is gratefully
acknowledged. This work was funded by MIUR (G.P., A.P.)
and GNV grants.
References
Benito, R., and J. Lo´pez-Ruiz (1992), ANATEX.BAS: A pro-
gram for calculating the mineralogy of the residual solid
and trace element fractionation in partial melting, Comput.
Geosci., 18(5), 603 – 615.
Bevins, R. E., B. P. Koke laar, and P. N. Dunkley (1984),
Petrology and geochemistry for lower to middle Ordovician
igneous rocks in Wales: A volcanic arc to marginal basin
transition, Proc. Geol. Assoc., 95, 337 – 347.
Boynton, W. V. (1984), Geochemistry of the rare earth ele-
ments: Meteorite studies, in Rare Earth Element Geochem-
istry, edited by P. Henderson, pp. 63 –114, Elsevier, New
York.
Conrad, W. R. (1987), A FORTRAN program for simulating
major- and trace-element variations during Rayleigh frac-
tionation with melt replenishment or assimilation, Comput.
Geosci., 13(1), 1 – 12.
Cox, K. J., J. D. Bell, and R. J. Pankhurst (1979), The Inter-
pretation of Igneous Rocks, 450 pp, Allen and Unwin, St
Leonards, NSW, Australia.
Defant, M. J., and R. L. Nielsen (1990), Interpretation of open
system petrogenetic process: Phase equilibria constrains on
magma evolution, Geochim. Cosmochim. Acta, 54(1), 87 –
102.
DePaolo, D. J. (1981), Trace element and isotopic effects of
combined wallrock assimilation and fractional crystalliza-
tion, Earth Planet. Sci. Lett., 53, 189 – 202.
D’Orazio, M. (1993), A Macintosh BASIC program for the
interactive testing of combined assimilation and fractional
crystallization, Comput. Geosci., 19(4), 483 – 492.
Gill, J. B. (1981), Orogenic Andesites and Plate Tectonics,
358 pp., Springer, New York.
Harnois, L. (1991), TEA: A computer program in BASIC to
calculate trace-element abundances in silicate rocks and
magmas during melting and crystallization processes, Com-
put. Geosci., 17(5), 641 –653.
Figure 10. Screen shot of the program showing the solid/liquid partition coefficient database. (a) Option of the
‘‘Windows’’ menu that allows the user to open the solid/liquid partition coefficient database. (b) Window with the
periodic table of elements. (c) By clicking an element on the table, the solid/liquid partition coefficients for that
element appears.
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
13 of 15

Hart, S. R., E. H. Hauri, L. A. Oschmann, and J. A. Whitehead
(1992), Mantle plumes and entrainment: Isotopic evidence,
Science, 256, 515 – 519.
Haskin, L. A., M. A. Haskin, F. A. Frey, and T. R. Wildman
(1968), Relative and absolute terrestrial abundances of the rare
earths, in Origin and Distribution of the Elements, vol. 1,
edited by L. H. Ahrens, pp. 889 – 911, Elsevier, New York.
Haskin, M. A., and F. A. Frey (1966), Dispersed and not-so-
rare earths, Science, 152, 299 – 314.
Haskin, M. A., and L. A. Haskin (1966), Rare earths in Euro-
pean shales: A redetermination, Science, 154, 507 – 509.
Holm, P. E. (1988), Petrogenetic modeling with a spreadsheet
program, J. Geol. Educ., 36(3), 155 – 156.
Holm, P. E. (1990), Complex petrogenetic modelling using
spreadsheet software, Comput. Geosci., 16(8), 1117– 1122.
Irvine, T. N., and W. R. A. Baragar (1971), A guide to the
chemical classification of the common volcanic rocks, Can.
J. Earth Sci., 8, 523 – 548.
Janousek, V., C. M. Farrow, and V. Erban (2003), GCDKit:
New PC software for interpretation of whole-rock geochem-
ical data from igneous rocks, Geochim. Cosmochim. Acta,
67, A186.
Keskin, M. (2 002), FC-mo deler: A Mic rosoft
1
Excel
#
spreadsheet program for modeling Rayleigh fractionation
vectors in closed magmatic systemsComput. Geosci.,
28(8), 919 – 928.
Kuno, H. (1968), Differentiation of basalt magmas, in Basalts:
The Poldervaart Treatise on Rocks of Basaltic Composition,
vol. 2, edited by H. H. Hess and A. Poldervaart, pp. 623–
688, Wiley-Interscience, Hoboken, N. J.
Langmuir, C. H. (1989), Geochemical consequences of in situ
crystallization, Nature, 340, 199 – 205.
Langmuir, C. H., R. D. Vocke, Jr., N. H. Gilbert, and R. H.
Stanley (1978), A general mixing equation with applications
to Icelandic basalts, Earth Planet. Sci. Lett., 37, 380–392.
Le Bas, M. J., R. W. Le Maitre, A. Streckeisen, and B. Zanettin
(1986), A chemical classification of volcanic rocks based on
the total alkali-silica diagram, J. Petrol., 27, 745–750.
Masuda, A., N. Nakamura, and T. Tanaka (1973), Fine struc-
ture of mutually normalised rare-earth patterns of chondrites,
Geochim. Cosmochim. Acta, 37, 239 – 248.
McDonough, W. F., S. Sun, A. E. Ringwood, E. Jagoutz, and
A. W. Hofmann (1992), K, Rb, and Cs in the earth and moon
and the evolution of the Earth mantle, Geochim. Cosmochim.
Acta, S. R. Taylor Symposium volume, 1001–1012.
Meschede, M. (1986), A method of discriminating between
different type of mid-ocean ridge basalts and continental
tholeiites with the Nb-Zr-Y diagram, Chem. Geol., 56,
207–218.
Middlemost, E. A. K. (1975), The basalt clan, Earth Sci. Rev.,
11, 337 – 364.
Miyashiro, A. (1974), Volcanic rock series in island arcs and
active continental margins, Am. J. Sci., 274, 321 – 355.
Nakamura, N. (1974), Determination of REE, Ba, Fe, Mg, Na
and K in carbonaceous and ordinary chondrites, Geochim.
Cosmochim. Acta, 38, 757 –775.
Neuman, H., J. Me ad, and C. J. Vitaliano (1 954), Trace-
element variation during fractional crystallization as calcu-
lated from the distribution law, Geochim. Cosmochim. Acta,
6, 90 – 100.
Nielsen, R. L. (1988), TRACE.FOR: A program for the calcu-
lation of combined major and trace element liquid line of
descent for natural magmatic systemsComput. Geosci.,
14(1), 15 – 35.
Pearce, J. A. (1982), Trace element characteristics of lavas
from destructive plate boundaries, in Andesites: Orogenic
Andesites and Related Rocks, edited by R. S. Thorpe,
pp. 525 – 548, John Wiley, Hoboken, N. J.
Pearce, J. A., and J. R. Cann (1973), Tectonic setting of basic
volcanic rocks de ter mi ned u sin g tr ace element analyse s,
Earth Planet. Sci. Lett., 19, 290 – 300.
Pearce, J. A., N. B. W. Harris, and A. G. Tindle (1984), Trace
element discrimination diagrams for the tectonic interpreta-
tion of granitic rocks, J. Petrol., 25, 956 – 983.
Peccerillo, A., and S. R. Taylor (1976), Geochemistry of Eo-
cene calc-alkaline rocks from Kastamonu area, Northern Tur-
key, Contrib. Mineral. Petrol., 58, 63–81.
Peccerillo, A., M. R. Barberio, G. Yirgu, D. Ayalew, M. Barbieri,
and T. W. Wu (2003), Relationships between mafic and per-
alcaline silicic magmatism in continental rift settings: A pet-
rological, geochemical and isotopic study of the Gedemsa
volcano, central Ethiopian rift, J. Petrol., 44, 2003–2032.
Richter, F. M. (1986), Simple models for trace-element frac-
tionation during melt segregation, Earth Planet. Sci. Lett.,
77, 333 – 344.
Rollison, H. (1993) , Using Geochemical Data: Evaluation,
Presentation, Interpretation, 352 pp., Longman S ci. and
Tech., Harlow, UK.
Saunders, A. D., and J. Tarney (1984), Geochemical character-
istics of basaltic volcanism within back-arc basin, in Marginal
Basin Geology, edited by B. P. Kokelaar and M. F. Howells,
Geol. Soc. Spec. Publ., 16, 59 – 76.
Spera, F. J., and W. A. Bohrson (2001), Energy-constrained
open-system magmatic processes I: General model and en-
ergy-constrained, J. Petrol., 42, 999 – 1018.
Stormer, J. C., and J. Nicholls (1978), XLFRAC: A program
for the interactive testing of magmatic differentiation mod-
els, Comput. Geosci. , 4, 143 – 159.
Streckeisen, A., and R. W. Le Maitre (1979), A chemic al
approximation to the modal QAPF classification of igneous
rocks, Neues Jahrb. Mineral. Abh., 136, 169 –206.
Su, Y., C. H. Langmuir, and P. D. Asimow (2003), PetroPlot: A
plotting and data management tool set for Microsoft Excel,
Geochem. Geophys. Geosyst., 4(3), 1030, doi:10.1029/
2002GC000323.
Sun, S. S. (1980), Lead isotopic study of young volcanic rocks
from mid-ocean ridges, ocean islands and island arcs, Philos.
Trans. R. Soc. London, Ser. A, 297, 409 – 445.
Sun, S. S., and W. F. McDonough (1989), Chemical and iso-
topic systematics of ocean basalts: Implications for mantle
composition and processes, in Magmatism in Ocean Basins,
edited by A. D. Saunders and M. J. Norry, Geol. Soc. Spec.
Publ., 42, 313 – 345.
Taylor, S. R., and S. M. McLennan (1981), The composition
and evolution of the continental crust: Rare earth element
evidence from sedimentary rocks, Philos.Trans.R.Soc.
London, Ser. A, 301, 381–399.
Taylor, S. R., and S. M. McLennan (1985), The Continental
Crust: Its Composition and Evolution, 312 pp., Blackwell
Sci., Malden, Mass.
Verma, S. P., R. Ciriaco-Villanueva, and I. S. Torres-Alvarado
(1998), SIMOCF: Modeling fractional crystallization using a
Monte Carlo approach, Comput. Geosci., 24(10), 1021 –
1027.
Weaver, B., and J. Tarney (1984), Empirical approach to esti-
mating the composition of the continental crust, Nature, 310,
575–577.
Wood, B. J., and D. G. Fraser (1976), Elementary Thermody-
namics for Geologists, 303 pp., Oxford Univ. Press, New
York.
Wood, D. A. (1980), The application of a Th-Hf-Ta diagram to
problems of tecnomagmatic classification and to establishing
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
14 of 15

the nature of crustal contamination of basaltic lavas of the
British Tertiary volcanic province, Earth Planet. Sci. Lett.,
50, 11– 30.
Wood, D. A., J. L. Joron, M. Treuil, M. Norry, and J. Tarney
(1979a), Elemental and Sr isotope variations in basic lavas
from Icela nd and the surroun ding oce an flo or, Co ntrib.
Mineral. Petrol., 70, 3219 –3239.
Wood, D. A., J. Tarney, J. Varet, A. D. Saunders, Y. Bougault,
J. L. Joron, M. Treuil, and J. R. Cann (1979b), Geochemistry
of basalts drilled in the North Atlantic by IPOD Leg 49:
Implications for mantle heterogeneity, Earth Planet. Sci.
Lett., 42, 77 – 97.
Woussen, G., and D. Coˆte´ (1987), HYPERFUNC: BASIC
program to calculate hyperbolic magna-mixing curves for
geochemical data, Comput. Geosci., 13(4), 421–431.
Wright, T. L., and P. C. Doherty (1970), A linear programming
and least squares computer method for solving petrologic
mixing problems, Bull. Geol. Soc. Am., 81, 1995 – 2008.
Geochemistry
Geophysics
Geosystems
G
3
G
3
petrelli et al.: petrograph software 10.1029/2005GC000932
15 of 15