WPC  2B" Jl  |xMACNormal      X` hp x (#%'0*,.8135@8:<     :}D4P T I. A. 1. a.(1)(a) i) a)T,0*ÍÍ,*Í ., US!!!! !     X` hp x (#%'0*,.8135@8:<     :}D4P ,0*ÍÍ,*Í ., US!!!! ! Footnote bY }\4 P Í }\4 P  }\4 P č }\4 P !!!!, 3' \MACNormal }\4 P   PETRO.CALC.PLOT, Microsoft Excel macros to aid petrologic interpretation by GARY B. SIDDER U.S. Geological Survey Denver Federal Center Box 25046, MS905 Denver, Colorado 80225 3032365607/FAX 3032365603/gbsidder@greenwood.cr.usgs.gov October, 1993 published in: Computers & Geosciences, v. 20, no. 6 Abstract"PETRO.CALC.PLOT is a package of macros that normalizes wholerock oxide data to 100 percent, calculates the cation percentages and molecular proportions used for normative mineral calculations, computes the apices for common ternary diagrams, determines sums and ratios of specific elements of petrologic interest, and plots 33 XY graphs and five ternary diagrams. PETRO.CALC.PLOT may also be used to create other diagrams as desired by the user. The macros run in Microsoft Excel 3.0 and 4.0 for Macintosh computers and in Microsoft Excel 3.0 and 4.0 for Windows. Macros provided in PETRO.CALC.PLOT minimize repetition and time required to recalculate and plot wholerock oxide data for petrologic analysis. Key Words: Macros, recalculation, petrology, XY graphs, ternary diagrams TABLE OF CONTENTS    $,$ 3 Calculationfile [calcfile.xls] ` 7 Graph macros ` 12 Graph macroHarker diagrams [mharker.xlm] ` 13 Petrology graph macro [mxygraph.xlm] ` 14 Graph macroMgO diagrams [mmgoplot.xlm] ` 17  graph macro [mternary.xlm] ` 18 Ternary plot apicesdata [apices.xls] ` 19 How to edit your own diagrams ` 20 How to plot your own diagrams ` 22 Summary ` 26 Acknowledgments ` 26 References cited ` 27 LIST OF TABLES Table 1. Equivalent File Names for PETRO.CALC.PLOT on Macintosh Computers and PCs. Table 2. Examples of the effect of iron on recalculated wholerock oxide values. LIST OF FIGURES Figure 1. Total Alkali Silica Diagram. Figure 2. R1R2 Diagram. Figure 3. Jensen Diagram. Figure 4. MnTiP Diagram.  Introduction  Recalculation of geochemical data is a fundamental step in the interpretation of the petrologic history of igneous rocks. Methods used to recalculate the data are commonly repetitive and timeconsuming. This package of software files, named PETRO.CALC.PLOT, minimizes repetition and time required to recalculate and plot wholerock oxide data. Macros provided in PETRO.CALC.PLOT normalize wholerock oxide data to 100 percent, calculate the cation percentages and molecular proportions used for normative mineral calculations, compute the apices for common ternary diagrams, determine sums and ratios of specific elements of petrologic interest, and plot 33 XY graphs and five ternary diagrams. These files were developed with the program Microsoft Excel 3.0 for Macintosh computers\MACNormalFootnoteAny use of trade, product, or firm name is for descriptive purposes only and does not imply endorsement by the U.S. Government. ; they will also run in Microsoft Excel 4.0. They have also been modified to run in Microsoft Excel 3.0 and 4.0 for Windows. File names listed in this documentation are for the Macintosh; DOS names are in brackets. Table 1 provides a list of equivalent file names for the Macintosh and Windows versions and groups the files according to which macro uses them. PETRO.CALC.PLOT includes five macros, three worksheets for data entry and calculation, five data sets to test the calculation macro, and another five data sets to test the four graph macros. All of the XY graphs and ternary diagrams are included as individual formatted files (table 1). Figure 1 is a flow chart that schematically represents what the macros do and which files the macros utilize from data input to data or graphic output. The calculation macro normalizes wholerock oxide data to 100 percent and calculates petrologic parameters commonly used to plot and interpret major and minor oxide data such as the sum and the ratio of alkalis, the total iron/magnesia ratio, the Mg number, apices for AFM, NKC, and Jensen diagrams, the agpaitic coefficient, and the indices R1 and R2 (table 2). Three of the four graph macros plot eleven XY diagrams, and the fourth plots five ternary diagrams (fig. 1). One graph macro plots the major and minor oxides versus silica on Harker diagrams, and another plots the major and minor oxides versus MgO on MgO variation diagrams. Graphs for petrologic analysis that are plotted by the third XY graph macro include total alkalisilica, iron oxide/magnesia, Peacock, R1R2, alumina saturation, and other diagrams (table 2). Ternary diagrams plotted by the ternary graph macro include AFM, NKC, Jensen, and other diagrams. This package of macros, PETRO.CALC.PLOT, is available from the author freeofcharge at the above mailing address. Send one doublesided, highdensity 3.5inch floppy disk or two doublesided, doubledensity 3.5inch floppy disks and indicate which operating system, Macintosh or Windows, is desired. A floppy disk mailer for return mail should also be included.  Data worksheet [workshet.xls] Data worksheet [workshet.xls]  is a formatted worksheet ready to receive analytical data in the order that data are utilized by data macro [mdata.xlm]. It is also formatted to receive the results of the calculations. This worksheet allows the user to run data macro [mdata.xlm] without error due to entry of data in the wrong order. The order chosen for the oxides is that provided by the U.S. Geological Survey Branch of Geochemistry for wholerock major oxide analyses. One may type data into data worksheet [workshet.xls] directly or copy data from another worksheet into this file. Data macro [mdata.xlm] will run only if all data are entered in the exact order as presented in data worksheet [workshet.xls]. Entry of data in data worksheet [workshet.xls] is columnbased (each sample is entered in one column) for easy printing of tables for publication. Rowbased data from the user's own data file may be copied, pasted, and transposed into columnbased data by selecting EDITPASTE SPECIALTRANSPOSE from the pulldown menu. The user enters values in only the first fifteen rows (wt %/sample no. to Fe2O3/FeO) of the source data worksheet (table 2). Values for all of the major and minor oxides should be entered in the worksheet before running data macro [mdata.xlm]. However, data macro [mdata.xlm] will run successfully if a value for some, or even all, of the oxides is not entered, as long as something is entered in row 1 of the worksheet. The value 0.00 or #DIV/0! will be returned by the calculations for the normalized oxides and other sums and ratios for which values were not entered. The user does not need to enter a value for TOTAL on the worksheet, or for any other row below TOTAL (table 2). Data macro [mdata.xlm] calculates the sum of the oxides (TOTAL) and values for the other rows. Calculations run by data macro [mdata.xlm] do not recognize qualified or censored data. If any censored data are reported, for example, MnO < 0.02, then the data must be changed to a real number, such as 0.01, before running data macro [mdata.xlm]. The FORMULAREPLACE command in the menu bar may be used to globally replace censored data. Sanford and others (1993) provided an objective method to determine the replacement value of censored data. The recalculation routine in data macro [mdata.xlm] utilizes either analyzed Fe2O3 and FeO, FeTO3 (total iron as Fe2O3), or FeO* (total iron as FeO). The user has four options for inputting data for iron. The first option is to utilize analyzed Fe2O3 and FeO. Enter values for these two, and "0" for FeTO3 and the ratio Fe2O3/FeO. However, because iron is most commonly analyzed as FeTO3 by Xray fluorescence, but ferrous iron is generally more abundant in most rocks, the user has three additional options for inputting data for iron: 2) enter the value of FeTO3 and a desired ferric/ferrous ratio for Fe2O3/FeO between 0.10 and 0.50 in increments of 0.05 (Middlemost, 1989), and "0" for FeO and Fe2O3; 3) enter the value of FeTO3, but "0" for the Fe2O3/FeO ratio, FeO, and Fe2O3; or 4) enter the value for FeO* as FeO and "0" for Fe2O3, Fe2O3/FeO, and FeTO3. If FeTO3 and the desired ratio for Fe2O3/FeO are entered (option 2), data macro [mdata.xlm] automatically determines the values for FeO and Fe2O3 and utilizes these values in subsequent calculations. If the ratio is not specified (options 3 and 4), data macro [mdata.xlm] will convert FeTO3 to FeO* or utilize the FeO* value as input by the user. Thus, options 3 and 4 produce identical calculated results. Option 2 yields slightly different calculated values, most notably the Mg number (cation percent Mg/(Mg + Fe2+) x 100), because of the different total value used to normalize the data and the lack of Fe2O3 in the calculations for options 3 and 4. Table 2 offers examples of the effect on the calculations with different iron values for two samples.  Data macro [mdata.xlm] Data macro [mdata.xlm] is a short macro written to facilitate calculation of petrologic parameters commonly used to interpret wholerock oxide data. Data are copied sample by sample from a worksheet (data worksheet [workshet.xls] or the user's own data file) in the data worksheet format and pasted into a routine (calculationfile [calcfile.xls]) that calculates normalized oxides, molecular and cation proportions, as well as apices for AFM, NKC, Jensen, and other ternary diagrams, and several ratios and sums of petrologic interest (table 2). The results in calculationfile [calcfile.xls] are copied and pasted back sample by sample into the original data worksheet below the raw analytical data. The number of samples that may be recalculated at one time is limited to 254, the number of columns available in the worksheet minus a blank column at the end of the worksheet. Data macro [mdata.xlm] continues to run until it identifies a blank cell in Row 1 in the source worksheet. A prompt then asks if the user would like to recalculate another set of data. Any number of data sets, limited only by the memory capacity of the user's computer, may be run sequentially through data macro [mdata.xlm]. For all responses while running the macros, Microsoft Excel offers only the choices OK or Cancel. Because many of the queries in the macros are YesNo choices, OK is used for Yes and Cancel is used for No. If one wishes to interrupt or halt a macro during operation, push the escape key. To run data macro [mdata.xlm], open data macro [mdata.xlm] and select MACRORUN from the menu bar. A RUN dialog box identifies the name of the macro (a 'data macro [mdata.xlm]'!cationnorms) and a cell reference ($A$1) to the macro sheet. As long as the cell reference is $A$1, the user may choose OK, Cancel, or Step. If the cell reference is different, select the name of the macro so that the cell reference reads: 'data macro [mdata.xlm]'!cationnorms. Click OK, and the macro proceeds normally. Click Cancel, and the macro stops. Enter Step, and another dialog box, SINGLE STEP, appears. This gives the user the choice of stepping through the macro, halting, or continuing the macro uninterrupted. The Step function allows the user to step line by line through the operations to see how the copy, calculation, and paste functions work. With Step, data macro [mdata.xlm] proceeds line by line; with Evaluate, one can see how each line of data macro [mdata.xlm] is answered, e.g., true or false, and the macro steps through all of the operations. With Continue, the macro runs normally. Halt stops the macro. If OK is selected from the RUN or STEP dialog box or Continue from the SINGLE STEP dialog box, an ALERT dialog box asks the user if the macro is running on the user's hard disk. If the macro is running from the floppy disk, a default path is defined for directory commands in the macro. If the macro is running on the hard disk, then a new path command needs to be entered. Click Cancel if running on the floppy disk; click OK if running from the hard disk. If one selects OK, then an ALERT dialog box asks if the path to run on the hard disk has already been saved in data macro [mdata.xlm]. If the hard disk path has been saved, click OK; if not, click Cancel and another ALERT dialog box explains that the user needs to identify the path from the hard disk to the folder or subdirectory that contains the macro and then an INPUT dialog box asks one to type the path. For example, if the macro has been copied onto the hard disk of a Macintosh computer into a folder named Macros in Microsoft Excel, the path for data macro will read: Hard disk name:Microsoft Excel:Macros:calculation macro and test data (no punctuation required after the name of the last folder), or C:\EXCEL\MACROS for PC users if copied into a subdirectory named MACROS in the directory EXCEL on the C drive. [For Macintosh only, after the path has been identified, an ALERT dialog box asks if calculationfile, the recalculation routine that is used by data macro and the four graph macros, is open. If it is open, but hidden, click OK; if not, Cancel.] Data macro [mdata.xlm] then opens and hides calculationfile [calcfile.xls] (or activates calculationfile [calcfile.xls] if it is already open) and hides data macro [mdata.xlm]. One is then prompted to open a data set with an OPEN DOCUMENT dialog box. One may select data from either the hard disk or a floppy disk. Five test data sets are provided in PETRO.CALC.PLOT if one wishes to run a test file before entering his/her own data. After the data file is opened, an INPUT dialog box asks for the name of the data set. Type the exact name of the data set as it appears at the top of the file now open and click OK. If the name is misspelled, the macro will give an error message: Macro error at cell: data macro [mdata.xlm]!A17, and one will need to restart the macro. Upper and lower case letters do not have to match exactly. Data macro [mdata.xlm] then runs through the calculation routine for all samples in the data set until it comes to a blank cell at the top of a column in the source data worksheet. Calculations for this data set are complete. The user is given the opportunity to recalculate another data set. One may recalculate as many data sets as desired. When finished, the macro reminds the user that all data sets, data macro [mdata.xlm], and calculationfile [calcfile.xls] are hidden and that they may be unhidden by selecting WINDOWUNHIDE [FILEUNHIDE in Windows] from the menu bar. If an error message is encountered during operation of data macro [mdata.xlm] or if the user decides to halt the macro while it is in operation by using the escape key, select WINDOWUNHIDE [FILEUNHIDE in Windows] to find the macro and your data file. After the user is finished and tries to close the files data worksheet [workshet.xls], the test data files, or one's own data set, as well as calculationfile [calcfile.xls] and data macro [mdata.xlm], a dialog box asks if changes should be saved. To preserve the operations in data macro [mdata.xlm] and calculationfile [calcfile.xls], "NO" should be selected. The first time data macro [mdata.xlm] is run on the hard disk, the new path for the hard disk may be saved. Move the cursor to cell A1 in data macro [mdata.xlm] before saving, and then save data macro [mdata.xlm] with FILESAVE to keep the same name. If saved with a different name, the next time the new data macro is run, the macro will give an error message: Macro error at cell: new macro name!A10. After the hard disk path has been saved, the user is given a choice to customize data macro the next time it is run by deleting lines A2 through A6. This eliminates the need to answer the questions about running on the hard disk and entering the path. Data worksheet [workshet.xls], if used as the source worksheet with the user's own data, should be saved with FILESAVE AS to a new name designated by the user. To close all files at once, both open and hidden ones, when finished with a macro, hold down the shift key and pull down FILE to CLOSE ALL. Closing all files, especially including calculationfile, before the next macro is run will allow each macro to run faster.  Calculationfile [calcfile.xls] The file calculationfile [calcfile.xls] recalculates wholerock oxide data to 100 percent, volatile free, and calculates cation and molecular proportions, sums, ratios, and ternary apices for indices of petrologic interest. Although PETRO.CALC.PLOT does not calculate normative minerals, the molecular proportions needed for a CIPW norm calculation are determined (table 2). Calculationfile [calcfile.xls] also instructs the graph macros which rows of data to utilize in the source data worksheet (data worksheet [workshet.xls] or the user's own source data file) for each of the plots. All of the indices, except for the R1 and R2 factors used for the De la Roche diagram, are calculated using the volatilefree recalculated data. The R1 and R2 factors are calculated from the raw weight percentages of oxides, and FeO and Fe2O3 are grouped as Fe (De la Roche and others, 1980). LOI (loss on ignition) at 925$C is considered to represent the total bulk concentration of the volatile elements H2O, CO2, Cl, and F. Some of the petrologic indices are utilized in studies of rocks of particular compositions and thus may not be applicable to rocks of other compositions. Nonetheless, all indices are calculated by data macro [mdata.xlm] in calculationfile [calcfile.xls]. Hence, this calculation routine allows all of the user's wholerock oxide data to be manipulated in the same manner, and some indices that may not be calculated routinely are determined automatically. The following indices, with reference to the literature, are calculated:  Na2O + K2O (weight percent Na2O + K2O): this sum is commonly plotted against SiO2 on a silicavariation, or Harker, diagram, and is also known as the total alkalisilica (TAS) diagram (Cox and others, 1979; Miyashiro, 1978; Le Bas and others, 1986; Rickwood, 1989; Le Bas and Streckeisen, 1991; Le Bas and others, 1992).  Na2O/K2O and K2O/Na2O (weight percent Na2O/K2O and weight percent K2O/Na2O): two common ratios to check for alkali metasomatism (Cox and others, 1979).  FeO*/MgO (weight percent total iron as FeO divided by MgO): this quotient is commonly plotted versus weight percent SiO2 on a FMS (FeO*/MgO vs. SiO2) diagram (Miyashiro, 1973, 1974; Rickwood, 1989).  FeO*/(FeO* + MgO) (weight percent total iron as FeO divided by (total iron as FeO + MgO)): SiO2 and Al2O3 are commonly plotted versus FeO*/(FeO* + MgO) to identify tholeiitic, calcalkalic, and komatiitic rock series (Miyashiro, 1974; Viljoen and others, 1982; Anderson, 1983; DeWitt, 1989). Al2O3 plotted versus this ratio discriminates types of tholeiitic basalt (Mgrich, normal, and Ferich) and differentiates between komatiite and tholeiite rock series (Viljoen and others, 1982). This ratio may be called the mafic index if it is multiplied by 100 (Carmichael and others, 1974).  Mg number (100 x cation percent Mg divided by (cation percent Mg + Fe2+)): a common petrologic indicator of the degree of differentiation (Cox and others, 1979; Day, 1990). The Mg number is calculated differently by some investigators, with Fe3+ converted to Fe2+, and total iron as Fe2+ or Fe*; Mn is also commonly included with iron. The ratio is then: 100 x Mg/(Mg + Fe*). Note that if options 3 or 4 are selected for the input of iron (with FeTO3 or FeO* and Fe2O3/FeO = 0) to run data macro [mdata.xlm], Fe2+ equals Fe*. However, the Mg number is greater in option 2, where Fe2O3 and FeO are both calculated, because Fe2O3 is not added to FeO.  Fe2O3/FeO (weight percent Fe2O3 divided by FeO): a measure of the oxidation ratio in rocks (Middlemost, 1989). This ratio varies from 0.5 in rhyolite, trachyte, and phonolite to 0.2 for basalt, most basanite, and some foidite (Middlemost, 1989). Fe2O3/FeO is calculated by data macro [mdata.xlm] as a check on the standard ratio defined in data worksheet [workshet.xls] or the user's own data set, if used.  AFM (weight percent (Na2O + K2O), total iron as FeO*, and MgO): apices for a common triangular diagram (Nockolds and Allen, 1953; Irvine and Baragar, 1971; Brown, 1982; Rickwood, 1989). The iron apex is calculated as FeO*, or FeO + 0.8998Fe2O3, and not FeO + Fe2O3.  NKC (weight percent Na2O, K2O, and CaO): apices for a less commonly used triangular diagram (Nockolds and Allen, 1953; Field and others, 1975).  MnO/TiO2/P2O5 (weight percent (MnO x 10), TiO2, and (P2O5 x 10): indices for a ternary diagram that discriminates among five petrotectonic environments of basaltic rocks (Mullen, 1983).  K2O/TiO2/P2O5 (weight percent K2O, TiO2, and P2O5): apices for a triangular diagram to distinguish between oceanic and nonoceanic (continental) basalts (Pearce and others, 1975).  Jensen diagram (cation percentages of Al2O3, (FeO + Fe2O3 + TiO2), and MgO as written originally by Jensen (1976), but actually calculated as cation percent Al, (Fe2+ + Fe3+ + Ti), and Mg): indices for a ternary diagram to discriminate between calcalkaline, tholeiite, and komatiite volcanic rock series (Jensen, 1976; Grunsky, 1981; Jensen and Pyke, 1982; Viljoen and others, 1982; Rickwood, 1989). The Jensen diagram may also be used to classify subalkaline volcanic rocks into rhyolite, dacite, andesite, and basalt (Jensen, 1976; Grunsky, 1981; Jensen and Pyke, 1982).  A/NK (molecular proportion of Al2O3 divided by the sum of the molecular proportions of (Na2O + K2O)): this ratio is used to define peralkaline, subaluminous, and peraluminous rocks. If the ratio is <1, the rocks are peralkaline (Carmichael and others, 1974). A ratio >1 is common for peraluminous rocks, and subaluminous rocks have about equal proportions of molecular Al2O3 and Na2O + K2O (Carmichael and others, 1974).  A/CNK (molecular proportion of Al2O3 divided by the sum of the molecular proportions of (CaO + Na2O + K2O)): A/CNK 1, but with A/NK >1, defines metaluminous rocks. A ratio >1 characterizes peraluminous rocks (Carmichael and others, 1974).  Agpaitic coefficient (NK/A) (molecular proportions of (Na2O + K2O) divided by the molecular proportion of Al2O3): rocks with a NK/A ratio >1 are called agpaitic, which indicates that an excess of alkalis together with insufficient alumina during crystallization prohibited the formation of aluminum silicate minerals. This is common in alkalic igneous rocks. Peralkaline rocks have an agpaitic coefficient >1 (Le Bas and others, 1986).  R1 and R2  (in millications, R1 = [4,000Si 11,000(Na + K) 2,000(Fe2+ + Fe3+ + Ti)]; R2 = [6,000Ca + 2,000Mg + 1,000Al]: these parameters are calculated from the raw wholerock oxide weight percentage, not recalculated, data (De la Roche and others, 1980).  Functions 1, 2, and 3 (discriminant functions determined by multiplying the weight percent oxide and an eigenvector for each oxide in each function). These discriminant functions may be used to identify basalts from different tectonic settings on F2 vs. F1 and F3 vs. F2 plots (Pearce, 1976). F1 = [+0.0088SiO2 + 0.0102Al2O3 + 0.0066FeO 0.0017MgO 0.0143CaO  0.0155Na2O 0.0007K20 0.0774TiO2]; F2 = [0.0130SiO2 0.0129Al2O3 0.0134FeO 0.0300Mg) 0.0204CaO 0.0481Na2O + 0.0715K2O 0.0185TiO2]; and F3 = [0.221SiO2 0.0361Al2O3 0.0016FeO 0.0310MgO 0.0237CaO 0.0614Na2O  0.0289K2O 0.0532TiO2].  MgFe*Ti/Si and AlCa/Fe*NaK [weight percent ((MgO + FeO* + TiO2)/SiO2) x 100] and [weight percent ((Al2O3 + CaO)/(FeO* + Na2O + K2O))]: a plot of AlCa/Fe*NaK vs. MgFe*Ti/Si is effective to discriminate between alkaline granite, calcalkaline granite, and strongly peraluminous granite with SiO2 greater than 68 weight percent (Sylvester, 1989).  Log (CaO/(Na2O + K2O)) [Log10 of weight percent CaO divided by (Na2O + K2O)]: CaO/(Na2O + K2O) is called the calcalkali ratio (Brown, 1982), and it is commonly plotted against SiO2. Such a plot is a variation of the Peacock alkalilime index (Carmichael and others, 1974).  Graph macros Three graph macros, graph macroHarker diagrams [mharker.xlm], Petrology graph macro [mxygraph.xlm], and graph macroMgO diagrams [mmgoplot.xlm], plot eleven XY graphs each, and  graph macro [mternary.xlm] plots five ternary diagrams. All operate similarly to data macro [mdata.xlm]. Each of the macros automatically opens calculationfile [calcfile.xls] as data macro [mdata.xlm] does, and  graph macro [mternary.xlm] also opens ternary plot apicesdata [apices.xls]. The series of prompts in all four graph macros (for example, "Are you running this macro on your hard disk?", "Is calculationfile [calcfile.xls] open, but hidden, after running another macro?", and "Select a data set to plot") and entries ("Enter the name of your data set selected to plot:") are exactly the same. The graph macros are limited to plots of only five data sets with as many as 150 samples each. The user may plot more than 150 samples by changing the value 150 in cells A125 to A150 in calculationfile [calcfile.xls] to the desired number, moving to cell E1, and then saving calculationfile [calcfile.xls] with the same name. Five plot data test sets are provided to test run the four graph macros. [For Macintosh users: In Petrology graph macro .and  graph macro, when one encounters the message: "Is calculationfile open, but hidden?", one must choose OK for yes or Cancel for no. If calculationfile is open and hidden, and the user clicks OK, when graphs with field boundaries (such as the line between subalkaline and alkaline rock series) are opened, an ALERT dialog box will ask: "Update references to unopened documents?". Select No; do not select Yes. To avoid this prompt, close calculationfile before running Petrology graph macro and  graph macro .] Graph macroHarker diagrams [mharker.xlm] Graph macroHarker diagrams [mharker.xlm] plots eleven variation diagrams of the major and minor oxides versus SiO2 as the abscissa. The oxides are plotted in order from Al2O3 to MnO, except that FeTO3 follows FeO. The silicavariation, or Harker, diagram is commonly used to evaluate the consanguinity of a magma series (Carmichael and others, 1974; Cox and others, 1979; Wilcox, 1979). These series of diagrams are useful to determine the evolution of some, but not all, rock series, especially intermediate and felsic rocks (Cox and others, 1979). Petrology graph macro [mxygraph.xlm] Petrology graph macro [mxygraph.xlm] plots eleven common diagrams used for petrologic interpretation. Boundaries between fields on each diagram are designed to discriminate between rock type, rock series, and(or) tectonic environment of deposition and(or) emplacement. However, the interpretation of each diagram may lead to ambiguous conclusions. For example, some samples may plot as tholeiitic on one diagram, but calcalkalic on another. This is in part due to the fact that most of these diagrams were constructed for fresh, unaltered samples only. Also, some diagrams may be applicable only for rocks with compositions between a specific range. Check the original references cited in this paper to resolve any discrepancies in the petrologic interpretation. Harker diagramNa2O + K2O [HNA2OK2O.XLC] is a plot of total alkalis versus SiO2. This plot may be used to discriminate between rocks of the alkaline and subalkaline or tholeiite series (Rickwood, 1989). The two lines shown on this diagram represent the high and low ranges of this boundary as defined by other authors (Rickwood, 1989). Spread between the two lines may be accounted for by interlaboratory analytical precision (Rickwood, 1989). Therefore, rocks that plot within the band between the two lines cannot be reliably assigned to either the alkaline or subalkaline rock series (Rickwood, 1989). The Total AlkaliSilica diagram [TAS.XLC] is also a plot of total alkalis Na2O + K2O versus SiO2. The fields outlined on this diagram define the chemical classification of volcanic rocks as determined by the Subcommission on the Systematics of Igneous Rocks of the International Union of Geological Sciences (fig. 1; Le Bas and Streckeisen, 1991; Le Bas and others, 1992). Middlemost (1989) has utilized these fields to define standard ratios of Fe2O3/FeO. This ratio varies from 0.5 in rhyolite, trachyte, and phonolite, to 0.2 in basalt, most basanite, and some foidite, and to 0.15 in picrobasalt (Middlemost, 1989). Na2O/K2O vs. Na2O + K2O [NA2OK2O.XLC] and K2O/Na2O vs. SiO2 [K2ONA2O.XLC] demonstrate the intensity and effects of alkali enrichment or metasomatism. They are both plotted as semilog diagrams, with the ordinate a log scale and the abscissa (Na2O + K2O and SiO2, respectively) a normal arithmetic scale. Miyashiro (1975) utilized the Na2O/K2O vs. Na2O + K2O to evaluate alkali metasomatism in intermediate to mafic volcanic rocks. A curved boundary line that trends from high Na2O/K2Olow Na2O + K2O to low Na2O/K2Ohigh Na2O + K2O represents the upper limit of Na2O/K2O for fresh volcanic rocks (Miyashiro, 1975; Alvi and Raza, 1992). The K2O/Na2O ratio equals about 1.0 for most rocks with silica concentrations of about 68 weight percent and is greater than 1 for unaltered rocks with SiO2 greater than about 68 to 70 weight percent. K2O/Na2O is about 0.3 for most basaltic rocks (Cox and others, 1979). K2O/Na2O vs. SiO2 [K2ONA2O.XLC] may also be used to evaluate the maturity and position of a volcanic arc (Brown, 1982). FeOt/MgO vs. SiO2 [FEOTMGO.XLC] may be used to distinguish the tholeiite and calcalkaline rock series for rocks with FeO*/MgO >2, but <5, which corresponds to SiO2 >55.6 and <74.8 weight percent (Miyashiro, 1973, 1974; Rickwood, 1989). The boundary line at FeO*/MgO values less than 2.0 is not located precisely, and the tholeiitic and calcalkaline series cannot be discriminated successfully (Miyashiro, 1974; Rickwood, 1989). The boundary line on this diagram is drawn solid only for the recommended range of SiO2, 55.6 to 74.8 weight percent. SiO2 vs. FeOt/(FeOt + MgO) [FEOFEMG.XLC] shows the boundary line between the tholeiitic rock series (or Ferich rocks) and the calcalkaline rock series (or Mgrich rocks) (Miyashiro, 1974; Anderson, 1983; DeWitt, 1989). This diagram portrays the intensity of iron or magnesium enrichment within an igneous rock series or the relative enrichment between rocks of a particular igneous body or area (Anderson, 1983; DeWitt, 1989). It is also a discriminant diagram between island arc, continental arc, and continental collision granitoids and riftrelated and continental epeirogenic uplift granitoids (Maniar and Piccoli, 1989). Mg # vs. SiO2 [MG#SI.XLC] displays the variation of Mg number versus silica. Mg number is a common petrologic indicator of the degree of differentiation of igneous rocks (Cox and others, 1979; Day, 1990). Basalts with a Mg number between 79 and 70 are thought to have been in equilibrium with a peridotitic mantle (Day, 1990), and more evolved rocks have a lower Mg number. Rhyolites commonly have a Mg number between 15 and 30. Modified Peacock diagram [PEACOCK.XLC] is a plot of log10CaO/(Na2O + K2O) against SiO2. It may be used to classify the chemical affinity of igneous rocks and to identify their tectonic environment (Brown, 1982; Sims and others, 1989). Such a plot is a variation of the Peacock alkalilime index, in which rocks with equal abundances of CaO and (Na2O + K2O) are classified as: alkalic (SiO2 <51 weight percent); alkalicalcic (SiO2 between 51 and 56 weight percent); calcalkalic (SiO2 between 56 and 61 weight percent); or calcic (SiO2 >61 weight percent) (Carmichael and others, 1974; Brown, 1982). On the modified Peacock diagram [PEACOCK.XLC] , the silica concentration where the log of the ratio is zero (equal abundances of CaO and (Na2O + K2O)) defines the field (alkali to calcic) of the Peacock alkalilime index (Brown, 1982). A/NK vs. A/CNK diagram ANKACNK.XLC] shows chemical nomenclature for igneous rocks based on alumina rather than silica saturation. Metaluminous rocks have A/CNK 1, but with A/NK >1; the A/CNK ratio is >1 for peraluminous rocks, and A/NK is commonly >1; peralkaline rocks have A/NK <1 (Carmichael and others, 1974; Brown, 1982; Maniar and Piccoli, 1989). Continental collision granitoids may be distinguished from island arc and continental arc granitoids on this diagram (Maniar and Piccoli, 1989). R1R2 diagram [R1R2.XLC] provides a chemical classification of igneous rocks based on cationic parameters. A single classification grid is applicable to both volcanic and plutonic rocks (fig. 2; De la Roche and others, 1980). When plotted as R2 vs. R1, boundaries are defined to classify volcanic and plutonic rock families (De la Roche and others, 1980), to identify rocks of calcic, calcalkalic, alkalicalcic, and alkalic affinity (DeWitt, 1989), and to distinguish the tectonomagmatic association of granitic rocks (Batchelor and Bowden, 1985). TiO2ĩMg # [TIO2MG#.XLC] indicates the degree of fractionation of igneous rocks, based on the Mg number, as well as the evolutionary path of a particular rock suite. For example, Day (1990) utilized this plot to distinguish two suites of mafic tholeiitic rocks in the Rainy Lake area, Minnesota. Graph macroMgO diagrams [mmgoplot.xlm] Graph macroMgO diagrams [mmgoplot.xlm] plots eleven MgOvariation diagrams. These diagrams are typically plotted for intermediate to mafic, especially basaltic, and ultramafic rocks (Cox and others, 1979). They are used to analyze crystallization paths in a magma because the solids crystallizing from a magma generally contain more MgO than the liquid (Cox and others, 1979). DonnellyNolan and others (1991) used MgOvariation diagrams to distinguish between lavas of the Cascade Range and Medicine Lake volcano, California.  graph macro [mternary.xlm] Five ternary diagrams are plotted by  graph macro [mternary.xlm]. The windows for these diagrams have been sized so that the triangles print as equilateral triangles. AFM diagram [AFM.XLC] (Na2O + K2OFeO*MgO) is the most commonly plotted ternary diagram. It is used to subdivide subalkaline rocks into tholeiite and calcalkaline series (Irvine and Baragar, 1971; Carmichael and others, 1974; Cox and others, 1979; Rickwood, 1989). The boundary line shown on AFM diagram [AFM.XLC] is that defined by Irvine and Baragar (1971, fig. 2A, p. 528). NKC diagram [NKC.XLC] (Na2OK2OCaO) is another common ternary diagram. Nockolds and Allen (1953) displayed the calcalkaline trend on a NKC diagram for several suites of volcanic and plutonic rocks. Field and others (1975) noted that this diagram may also be used to depict the chemical effects of hydrothermal alteration associated with porphyrytype mineralization. The calcalkaline trend line on NKC diagram [NKC.XLC] is that from Field and others (1975). Jensen diagram [JENSEN.XLC] (fig. 3) is a ternary plot of the cation percentages of Al2O3, (FeO + Fe2O3 + TiO2), and MgO. It is a modification of the original Jensen Cation Plot (Jensen, 1976). Jensen diagram [JENSEN.XLC] incorporates changes proposed by Jensen and Pyke (1982) and Viljoen and others (1982), as discussed by Rickwood (1989). It may be used to define three distinct differentiation trends (komatiitic, tholeiitic, and calcalkalic) in volcanic rocks and to name the major rock types (basalt to rhyolite) of subalkalic volcanic rocks (Jensen, 1976). The three trends pass through or near the letters K, TH, and CA, respectively, on Jensen diagram [JENSEN.XLC] (fig. 3). The curved line dividing the tholeiitic and calcalkalic fields on Jensen diagram [JENSEN.XLC] corresponds closely to that on AFM diagram [AFM.XLC]. KTiP diagram [KTIP.XLC] is a ternary diagram (K2OTiO2ĩP2O5) designed to discriminate between oceanic and nonoceanic (continental) basalts. It is effective for nonalkalic primitive, not fractionated, basalts (Pearce and others, 1975). Pearce and others (1975) recommended screening samples on an AFM diagram and plotting on KTiP diagram [KTIP.XLC] only those with <20 percent Na2O + K2O (as a proportion of Na2O + K2OFeO*MgO). MnTiP diagram [MNTIP.XLC] is a ternary plot with MnO x 10, TiO2, and P2O5 x 10 as apices to discriminate among five petrotectonic environments of basalt and basaltic andesite with SiO2 between 45 and 54 weight percent (Mullen, 1983). The five environments are (fig. 4): island arc calcalkaline basalt (CAB), island arc tholeiite (IAT), midocean ridge and marginal basin basalt (MORB), ocean island tholeiite (OIT), and ocean island alkalic basalt (OIA). Continental tholeiitic basalt overlaps portions of all five oceanic fields, and boninite plots within the island arc fields of CAB and IAT (Mullen, 1983).   Ternary plot apicesdata [apices.xls] Ternary plot apicesdata [apices.xls] is a worksheet that is used by  graph macro [mternary.xlm] to plot the outline of the ternary diagrams as well as the boundary lines and data points on each plot. As with calculationfile [calcfile.xls], it should never be overwritten and saved with new data or with a new name.  graph macro [mternary.xlm] automatically opens ternary plot apicesdata [apices.xls]. The macro then copies the coordinates for the ternary apices of the five ternary plots from the source data worksheet and pastes the values of the apices into ternary plot apicesdata [apices.xls]. Formulae in ternary plot apicesdata [apices.xls] convert the ternary coordinates into axis values necessary to plot the ternary diagram as an XY graph.  graph macro [mternary.xlm] copies the X and Y coordinates from ternary plot apicesdata [apices.xls] and pastes the data onto each of the ternary diagrams. Ternary plot apicesdata [apices.xls] is formatted to plot as many as five data sets with as many as 150 samples each. To plot more than 150 samples, formulae in cells H5H6, H126H127, H246H247, H366H367, and H486487 must be copied and pasted into columns FB to IV in the respective rows. The number of samples is limited to 249 by the size and arrangement of ternary plot apicesdata [apices.xls].  How to edit your own diagrams The plots generated by the graph macros may all be edited and reformatted, if necessary, by using standard operations in Microsoft Excel 3.0 and 4.0. For example, double clicking on the X axis of any of the XY plots opens a dialog box in which one may change settings for the axis style, color, and weight, the tick mark type, and the tick labels. The text, font, or scale of the axis in this dialog box can also be changed. To change the scale, select Scale and enter the desired range of and interval between units in the CATEGORY (X) AXIS SCALE dialog box. Those plots with defined field boundaries will retain their position with new scale designations because all of the boundary lines on all diagrams are plotted with X and Y coordinates. The scale on ternary diagrams cannot be displayed and should not be changed. All text and field labels on the diagrams may be edited. For example, on FeOt/MgO vs. SiO2, one may click on Tholeiite Series. A box with eight filled squares appears around Tholeiite Series, and Tholeiite Series appears in the menu bar as Text 1. One may edit, delete, or move the text: to edit, click in the menu bar, type the change, and enter return; to delete, enter delete and return; to move, hold down the mouse button and move the text. If one selects the title of the graph (FeO*/MgO vs. SiO2) or the label of the X axis (SiO2 (weight %)), a box with eight unfilled squares appears around the text. The text appears in the menu bar as Title or Text Axis 2, respectively. The text may be edited or deleted, but not moved. If the title in the ternary diagrams is deleted or if the font size is changed, the size of the triangle will change so that it will not be equilateral. The user will have to resize the window to produce an equilateral triangle. To add text to a diagram, simply click in the menu bar, type the text, and enter return. Text may be moved anywhere on a diagram, and its appearance may be changed by double clicking on the text or selecting it and choosing FORMAT in the menu bar. One may change the type, size, style, color, and background of font, the alignment and orientation of text, and the border and area (patterns) around the text. The number format for scales on the X and Y axes is related to the source data worksheet and cannot be modified by selecting Scale. Data macro [mdata.xlm] copies calculated data from calculationfile [calcfile.xls] and pastes values to two decimals for most calculated data into the source data worksheet. All of the graph macros then reformat some of the data in the source data worksheet so that most of the diagrams show axis values with one decimal or none. Thus, if one wishes to change the silica values for the Harker diagrams to one decimal (64.0), one needs to change the values in the source data worksheet. This may be done after the diagrams are plotted. After the chart has been saved with the desired number format, the changed number format in the source data worksheet does not have to be saved. The user may wish to add a legend to the diagrams. For diagrams without any boundary lines, such as the Harker and MgOvariation diagrams, select CHARTADD LEGEND. The legend will contain the symbol for each data set and the name of the oxide plotted because the text is taken directly from the worksheet cells that contain the data series names. To change the legend text, one may select the data set on the chart, choose CHARTEDIT SERIES in the menu bar, and type the name in the EDIT SERIES dialog box. One may also edit the worksheet cells that contain the data series name, for example, change "Al2O3" to "Carma volcanic rocks" in data worksheet [workshet.xls] or the user's own source data worksheet. After the diagram has been saved, the name changes in the source data worksheet do not have to be saved.  Adding a suitable legend to diagrams with boundary lines is difficult in Microsoft Excel 3.0. The legend contains all of the symbols and names for each boundary line on a diagram in addition to the data points. The legend also causes the ternary diagrams to be resized so that they are not equilateral triangles. The user may need to copy the ternary diagrams and XY diagrams with boundary lines, paste them into a draw or paint program, and then add a legend. To copy the diagram, hold down Shift on the keyboard and select EDIT from the menu bar and choose COPY PICTURE. Open the draw program and select EDITPASTE from the menu bar.  How to plot your own diagrams  Users of PETRO.CALC.PLOT may wish to plot their own diagrams. XY graphs may be plotted normally as charts. Note that in Microsoft Excel 3.0 and 4.0, a larger row number is always plotted as the Y axis versus a smaller row number as the X axis (the graph macros in PETRO.CALC.PLOT have been modified to avoid this restriction). To plot a diagram, for example, AlCa/Fe*NaK vs. MgFe*Ti/Si, select the two rows of data from column A to the last column with data. The selection of the first row may be accomplished most easily by using the Shift }d@m   }\4 P ѩRight Arrow [End, then ShiftRight Arrow in Windows] keys. Then select the last column of data of the second row that the user wishes to plot while holding down the  }d@m   }\4 P  [Ctrl] key. Then hold down the Shift }d@m   }\4 P ѩArrow Left [End, then ShiftLeft Arrow] keys simultaneously. Both rows of data should be highlighted from column A to the last row of data. Select FILENEW from the menu bar, and CHART from the NEW dialog box. Select "Xvalues for XY chart" from the FIRST ROW CONTAINS dialog box. The chart may be reformatted as discussed above. For example, select CHARTATTACH TEXT, and then select Category (X) Axis, Value (Y) Axis, and Chart Title sequentially to label the X and Y axes and entitle the chart, respectively. One may also use a formatted chart in PETRO.CALC.PLOT to plot additional data. For example, to construct a traditional Peacock diagram of CaO and Na2O + K2O versus SiO2, open Harker diagramNa2O + K2O [HNA2OK2O.XLC] and a source data worksheet. Select the rows of data for SiO2, CaO, and Na2O + K2O from column A to the last column of data. Again, the easiest way to select the data is to move to cell A20 (SiO2), use Shift }d@m   }\4 P ѩRight Arrow [End, then ShiftRight Arrow] to move to the end of the row of data for SiO2, hold down the  }d@m   }\4 P  [Ctrl] key and select the last row of data in CaO, hold down the  }d@m   }\4 P  [Ctrl] key and select Na2O + K2O in column A, and use Shift }d@m   }\4 P ѩRight Arrow [End, then ShiftRight Arrow] to move to the end of the row of data for Na2O + K2O. Choose EDITCOPY from the menu bar, and then activate Harker diagramNa2O + K2O [HNA2OK2O.XLC] from WINDOW. Select EDITPASTE SPECIAL, and then choose Values (Y) in Rows, Series Names in First Column, and Categories (X Values) in First Row from the PASTE SPECIAL dialog box (the first two will already be selected). Remove the text Alkaline Series and Subalkaline or Tholeiite Series and the boundary lines by selecting them, and entering delete and then return on the keyboard. One may edit and reformat the title, scales, and axes as described above. The data symbols may be changed by selecting a data point from each data set, choose FORMATPATTERNS, and change the symbol style and color (foreground and background) in the Marker box. Ternary diagrams are more complex to plot. The following steps will allow one to plot a ternary diagram using files in PETRO.CALC.PLOT. Calculate the ternary apices for the desired elements by the following: 1) sum the three (or more) components that will be plotted; 2) divide each constituent (or the sum of constituents) for each apex by the total of all of the components, as calculated in step 1, and multiply by 100; and 3) sum the calculated values for the three coordinates (the sum should equal 100, not 1.0). Copy the data in the two rows that will plot as the B and C apices from column A to the last column of data. For example, for an AFM diagram, F (top) and M (lower right) are the B and C apices. Open ternary plot apicesdata [apices.xls] and move to cell G1. Select EDITPASTE from the menu bar. Cells H1 and H2 and the rest of rows 1 and 2 should contain the coordinates for the B and C apices, respectively; cells G1 and G2 should contain the names of the components in rows 1 and 2. The values from cells H5 and H6 to the end of the user's data in rows 5 and 6 (not including the values 0.00 in the two rows) are the calculated data that will be plotted on the ternary diagram. To construct a new ternary diagram, select cells D4 to E7 in ternary plot apicesdata [apices.xls]. Choose FILENEW from the menu bar, and CHART from the NEW dialog box. Pick "XValues for XY Chart" from the FIRST COLUMN CONTAINS dialog box. Select the data point in the top center of the diagram. Choose FORMATPATTERNS from the menu bar. Select Custom from the Line box, and choose the second line type from Weight. Pick None from the Marker box, and then OK. Choose CHARTAXES from the menu bar, and deselect both Category (X) Axis and Value (Y) Axis from the AXESMAIN CHART dialog box. A triangle is now outlined in the chart window. Activate ternary plot apicesdata [apices.xls] and select the data to be plotted from rows 5 and 6 in column H to the last column that contains the user's data (rows 5 and 6 contain 150 columns formatted with formula; those columns that do not contain a value in rows 1 and 2 will have 0.00 in rows 5 and 6). Select EDITCOPY from the menu bar. Activate the new chart from WINDOW in the menu bar. Choose EDITPASTE SPECIAL from the menu bar, and select Values (Y) in Rows (already selected) and Categories (X Values) in First Row from the PASTE SPECIAL dialog box. Select a data point in the ternary diagram, and choose FORMATPATTERNS from the menu bar. Select None from the Line box, and choose a desired symbol from the Marker box. Labels for the apices may be added as unattached text by clicking in the menu bar and typing the text. The text may then be formatted in the font and size desired, and moved to the appropriate place on the diagram. A title may be added as unattached text, or by choosing CHARTATTACH TEXT from the menu bar and selecting Chart Title from the Attach Text To box. The window will have to be resized to create an equilateral triangle. The user may also use a formatted ternary diagram in PETRO.CALC.PLOT to create a new ternary diagram. For example, copy the data to be plotted in ternary plot apicesdata [apices.xls] from rows 5 and 6 in column H to the last column that contains the user's data, as discussed above. Open AFM diagram or another ternary diagram. Choose EDITPASTE SPECIAL from the menu bar, and select Values (Y) in Rows (already selected) and Categories (X Values) in First Row from the PASTE SPECIAL dialog box. Reformat the data points under FORMATPATTERNS, delete undesired text and boundary lines, and edit the text for the apices and title. The window should be sized properly for an equilateral triangle as long as the same size and style of font are used for the apices and title. Save the new ternary diagram with a new name under FILESAVE AS.  Summary The software package PETRO.CALC.PLOT prepares and plots geochemical data for petrologic analysis. It provides a routine, rapid method, in which wholerock oxide data are entered into a worksheet just once, normalized, cation and molecular percentages, apices for common ternary plots, and sums and ratios of petrologic interest are calculated, and the results are pasted into the original worksheet. It also plots 33 XY graphs and five ternary diagrams, and it may be used to create other diagrams as desired by the user. Together, these files facilitate the interpretation of the petrologic history of igneous rocks.  Acknowledgments Ellen Sanchez from the Technology Information Center at the U.S. Geological Survey in Denver, Matthew Triplett, William Metters, and others of Microsoft Corporation in Redmond, Washington, provided important tips on customizing the macros. Richard Wanty, U.S. Geological Survey, supplied the formulae for plotting the ternary diagrams. Julie DonnellyNolan, Dana Bove, Jim Calzia, and Lance Forsythe reviewed an earlier version of this manuscript and significantly improved both the text and the macros.  References cited Alvi, S.H., and Raza, M., 1992, Geochemical evidence for the volcanic arc tectonic setting of the Dhanjori volcanics, Singhbhum craton, eastern India: Geological Magazine, v. 129, no. 3, p. 337348. Anderson, J.L., 1983, Proterozoic anorogenic granite plutonism of North America, in Medaris, L.G., Jr., Byers, C.W., Mickelson, D.M., and Shanks, W.C., eds., Proterozoic geology; selected papers from an international Proterozoic symposium: Geological Society of America Memoir 161, p. 133154. Batchelor, R.A., and Bowden, P., 1985, Petrogenetic interpretation of granitoid rock series using multicationic parameters: Chemical Geology, v. 48, no. 1, p. 4355. Brown, G.C., 1982, Calcalkaline intrusive rocks: their diversity, evolution, and relation to volcanic arcs, in Thorpe, R.S., ed., Andesites: John Wiley and Sons, New York, p. 437461. Carmichael, I.S.E., Turner, F.J., and Verhoogen, J., 1974, Igneous petrology: McGrawHill Book Company, New York, 739 p. Cox, K.G., Bell, J.D., and Pankhurst, R.J., 1979, The interpretation of igneous rocks: George Allen & Unwin, London, 450 p. Day, W.C., 1990, Petrology of the Rainy Lake area, Minnesota, USAimplications for petrotectonic setting of the Archean southern Wabigoon subprovince of the Canadian Shield: Contributions to Mineralogy and Petrology, v. 105, no. 3, p. 303321. De la Roche, H., Leterrier, J., Grandclaude, P., and Marchal, M., 1980, A classification of volcanic and plutonic rocks using R1R2ĩdiagram and majorelement analyses Its relationships with current nomenclature: Chemical Geology, v. 29, no. 2, p. 183210. DeWitt, E., 1989, Geochemistry and tectonic polarity of Early Proterozoic (17001750Ma) plutonic rocks, northcentral Arizona, in Jenney, J.P., and Reynolds, S.J., eds., Geologic evolution of Arizona: Arizona Geological Society Digest, v. 17, p. 149163. DonnellyNolan, J.M., Champion, D.E., Grove, T.L., Baker, M.B., Taggart, J.E., Jr., and Bruggman, P.E., 1991, The Giant Crater lava field: Geology and geochemistry of a compositionally zoned, highalumina basalt to basaltic andesite eruption at Medicine Lake volcano, California: Journal of Geophysical Research, v. 96, no. B13, p. 21,84321,863. Field, C.W., Briskey, J.A., Henricksen, T.A., Jones, M.B., Schmuck, R.A., and Bruce, W.R., 1975, Chemical trends in Mesozoic plutons associated with porphyrytype metallization of the Pacific Northwest: Society of Mining Engineers of AIME Preprint Number 75L359, 25 p. Grunsky, E.C., 1981, No. 16. An algorithm for the classification of subalkalic volcanic rocks using the Jensen cation plot, in Wood, J., White, O.L., Barlow, R.B., and Colvine, A.C., eds., Summary of field work, 1981: Ontario Geological Survey, Miscellaneous Paper 100, p. 6165. Irvine, T.N., and Baragar, W.R.A., 1971, A guide to the chemical classification of the common volcanic rocks: Canadian Journal of Earth Sciences, v. 8, no. 5, p. 523548. Jensen, L.S., 1976, A new cation plot for classifying subalkalic volcanic rocks: Ontario Division of Mines, Miscellaneous Paper 66, 22 p. Jensen, L.S., and Pyke, D.R., 1982, Komatiites in the Ontario portion of the Abitibi belt, in Arndt, N.T., and Nisbet, E.G., eds., Komatiites: George Allen & Unwin, London, p. 147157. Le Bas, M.J., Le Maitre, R.W., Streckeisen, A.L., and Zanettin, B., 1986, A chemical classification of volcanic rocks based on the total alkalisilica diagram: Journal of Petrology, v. 27, no. 3, p. 745750. Le Bas, M.J., Le Maitre, R.W., and Woolley, A.R., 1992, The construction of the total alkalisilica chemical classification of volcanic rocks: Mineralogy and Petrology, v. 46, no. 1, p. 122. Le Bas, M.J., and Streckeisen, A.L., 1991, The IUGS systematics of igneous rocks: Journal of the Geological Society, London, v. 148, part 5, p. 825833. Maniar, P.D., and Piccoli, P.M., 1989, Tectonic discrimination of granitoids: Geological Society of America Bulletin, v. 101, no. 5, p. 635643. Middlemost, E.A.K., 1989, Iron oxidation ratios, norms and the classification of volcanic rocks: Chemical Geology, v. 77, no. 1, p. 1926. Miyashiro, A., 1973, The Troodos ophiolitic complex was probably formed as an island arc: Earth and Planetary Science Letters, v. 19, no. 2, p. 218224.   1974, Volcanic rocks series in island arcs and active continental margins: American Journal of Science, v. 274, no. 4, p. 321355.   1975, Classification, characteristics, and origin of ophiolites: Journal of Geology, v. 83, no. 2, p. 249281.   1978, Nature of alkalic volcanic rock series: Contributions to Mineralogy and Petrology, v. 66, no. 1, p. 91104. Mullen, E.D., 1983, MnO/TiO2/P2O5: a minor element discriminant for basaltic rocks of oceanic environments and its implications for petrogenesis: Earth and Planetary Science Letters, v. 62, no. 1, p. 5362. Nockolds, S.R., and Allen, R., 1953, The geochemistry of some igneous rock series: Geochimica et Cosmochimica Acta, v. 4, no. 3, p. 105142. Pearce, J.A., 1976, Statistical analysis of major element patterns in basalts: Journal of Petrology, v. 17, no. 1, p. 1543. Pearce, T.H., Gorman, B.E., and Birkett, T.C., 1975, The TiO2ĩK2OP2O5 diagram: A method of discriminating between oceanic and nonoceanic basalts: Earth and Planetary Science Letters, v. 24, no. 3, p. 419426. Rickwood, P.C., 1989, Boundary lines within petrologic diagrams which use oxides of major and minor elements: Lithos, v. 22, p. 247263. Note: no number Sanford, R.F., Pierson, C.T., and Crovelli, R.A., 1993, An objective replacement method for censored geochemical data: Mathematical Geology, v. 25, no. 1, p. 5980. Sims, P.K., Van Schmus, W.R., Schulz, K.J., and Peterman, Z.E., 1989, Tectonostratigraphic evolution of the Early Proterozoic Wisconsin magmatic terranes of the Penokean Orogen: Canadian Journal of Earth Sciences, v. 26, no. 10, p. 21452158. Sylvester, P.J., 1989, Postcollisional alkaline granites: Journal of Geology, v. 97, no. 3, p. 261280. Viljoen, M.J., Viljoen, R.P., and Pearton, T.N., 1982, The nature and distribution of Archaean komatiite volcanics in South Africa, in Arndt, N.T., and Nisbet, E.G., eds., Komatiites: George Allen & Unwin, London, p. 5379. Wilcox, R.E., 1979, The liquid line of descent and variation diagrams, in Yoder, H.S., Jr., ed., The Evolution of the Igneous Rocks: Fiftieth Anniversary Perspectives: Princeton University Press, Princeton, New Jersey, p. 205232. FIGURE CAPTIONS Figure 1. Flow chart that schematically represents how the macros, data input and calculation files, and chart templates interact. Figure 2. Total Alkali Silica diagram. Classification of the volcanic rocks from Le Bas and Streckeisen (1991, fig. 5, p. 830). Basanite has normative olivine >10%; tephrite has normative olivine <10%. Trachyte has quartz <20% of sum of felsic minerals; trachydacite has quartz >20% of sum of felsic minerals. Standard Fe2O3/FeO ratios from Middlemost (1989, fig. 7, table I, p. 24 and 25, respectively). For basanite and tephrite, the Fe2O3/FeO ratio equals 0.2 for rocks with Na2O + K2O <6 weight percent; the ratio equals 0.3 for rocks with Na2O + K2O >6 weight percent. This figure was created by copying the chart in Microsoft Excel, pasting it into Aldus Freehand 3.0, and adding field names and iron ratios. Names and iron ratios are omitted in the Microsoft Excel chart template so that plotted data may be seen better. Figure 3. R1R2 diagram. Classification of the volcanic (fig. 2A) and plutonic rocks (fig. 2B) from De la Roche and others (1980, fig. 3, p. 189). Fig. 2A. lb, latibasalt; latiand., latiandesite; ne, nepheline; Q, quartz. Fig. 2B. mg, monzogabbro; quartz monz., quartz monzonite; ne, nepheline; Q, quartz. Field names added in Aldus Freehand 3.0, as described in Figure 2. Figure 4. Jensen diagram. Classification of subalkalic volcanic rocks and differentiation trends for komatiitic (K), tholeiitic (TH), and calcalkalic (CA) rocks (from Jensen, 1976, figs. 1 and 2, p. 7 and 8). HiFe and HiMg are abbreviations for highiron tholeiitic basalt and highmagnesium tholeiitic basalt, as originally defined by Jensen (1976). Fe, Normal, and Mg are abbreviations for ironrich tholeiite, normal tholeiite, and magnesiumrich tholeiite, as modified by Viljoen and others (1982, fig. 4.7, p. 69). Field names added in Aldus Freehand 3.0, as described in Figure 2. Figure 5. MnTiP diagram. Discriminant diagram for basalts and basaltic andesites with a concentration of silica between 45 and 54 weight percent from Mullen (1983, fig. 1, p. 54). CAB, island arc calcalkaline basalt; IAT, island arc tholeiite; MORB, midocean ridge and marginal basin basalt; OIT, ocean island tholeiite; and OIA, ocean island alkalic basalt.