SE. 35%: .uszorhhr {v u. 3.:- . p. h .2: .3... .v . .3: timhwms fin... . wail. .1: fix: my.-. {‘19 X a 5415...: :19} 5. a :u A. J. .4 6.. . . . . .7. . v (v .er .3. v i. 5... Us? 4. 1 9:... m . .. a: x 3.4 La». 3 4. I . ”Ky; .01, 4 r. .. u..: .u. .u‘ . 0...: 112‘... I A Nu. thwghv: Sula: :. 5:3 - v a... 3 a my?4, . \3 {Lift}. 31.nw, .l-‘itu n: .6 t. ,. a! n... a? 9. $.11): 3.“??- h}. . . . t )3 W H: .3. Jim? .14., -31.]..- n um. H .3535. um”... Jain... zkwfsgwfifi§gL£§m=aMm . . z: k -005 IJBRARY Michigan State University This is to certify that the dissertation entitled INDEPENDENTLY GENERATED MAGMA BATCHES IN THE COMPOSITIONALLY ZONED ASH-F LOW SHEETS FROM THE SOUTHWEST NEVADA VOLCANIC FIELD presented by KAREN SUE TEFEND has been accepted towards fulfillment of the requirements for the PhD. in Geological Sciences Major Professor’s Signatur ‘ 6 ~ 3 3’ - 05 Date MSU is an Affirmative Action/Equal Opportunity Institution PLACE IN RETURN Box to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 2/05 czmfiC/Dateomjndd-QIS INDEPENDENTLY GENERATED MAGMA BATCI-IES IN THE COMPOSITIONALLY ZONED ASH-FLOW SHEETS FROM THE SOUTHWEST NEVADA VOLCANIC FIELD By Karen Sue Tefend A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Geological Sciences 2005 ABSTRACT INDEPENDENTLY GENERATED MAGMA BATCHES IN THE COMPOSITIONALLY ZONED ASH-FLOW SHEETS FROM THE SOUTHWEST NEVADA VOLCANIC FIELD By Karen Sue Tefend Compositionally zoned ignimbrites have been inferred to represent the eruptive product of zoned magma chambers. Topopah Spring (12.8 Ma), Tiva Canyon (12.7 Ma), Rainier Mesa (1 1.6 Ma), and Ammonia Tanks (11.45 Ma) are four compositionally zoned ash-flow sheets within the southwest Nevada volcanic field, SW Nevada. These large volume ash-flow sheets have been extensively studied with the goal of understanding the formation of large volumes of high silica magmas that are, in this case, rapidly generated and erupted within short time intervals (150,000 years between the youngest of these ash flows). Previous studies have concluded, based on major and trace element geochemistry and isotopic analyses, that the lower silica magmas and high-silica rhyolite magmas within and among each ash-flow sheet cannot be related by assimilation/fractional crystallization processes occurring within a single magma chamber. The purpose of this current study is to evaluate this conclusion using Polytopic Vector Analysis (PVA). Based on these analyses we conclude that not only can unrelated magma types be identified, but that magmas related by mixing processes can also be determined. Using PVA, it can be shown that the coevally erupted lower silica ( S 73 wt% $02) and high-silica rhyolite magmas (2 74 wt% SiOz) within Topopah Spring, Tiva Canyon, Rainier Mesa, and Ammonia Tanks are unrelated, and must represent independent magma batches. An intermediate magma type identified in Tiva Canyon was found not to be the result of mixing between the lower silica and high-silica rhyolites of Tiva Canyon; however, a Similar intermediate magma type of Ammonia Tanks can be explained as the result of mixing between more evolved portions of the lower silica magma and the coevally erupted hi gh-silica rhyolite. Rainier Mesa is unique among these ash-flow sheets in that three high-silica rhyolite magmas (HSR-l , HSR-2, and HSR-3) can be identified based on trace element geochemistry (in particular Th/Nb, and La). PVA results show that the HSR-l magma type is unrelated to the coevally erupted lower silica magma and the other two high-Silica rhyolites. However, HSR-Z and HSR-3 are related and may be considered as one magma type. End members determined by PVA for each of the high-silica rhyolite magmas of Rainier Mesa overlap in composition, which may be interpreted as the result of mixing. Indeed, sanidine and melt inclusion, and glass matrix trace-element compositions support mixing among these high-silica magmas and also with a less evolved magma type. However, mixing is limited, such that standard geochemical modeling fails. PVA is more sensitive than typical major and trace element least squares linear regression models in recognizing mixing systems. In loving memory of Sylvia B. Tefend iv ACKNOWLEDGEMENT I would like to thank Tom Vogel for his support and guidance throughout this project, as well as Lina Patino. I also thank the rest of my committee members, Duncan Sibley and William Cambray, for their efforts in making this dissertation possible. A special thanks goes out to Robert Ehrlich for his patience and assistance in explaining PVA. Thanks, Tom, for suggesting I try PVA in the first place. I would also like to thank the petrology group (Chad, Ela, Dave, Carmen, Melissa, Ryan and Sara), for some great discussions and very memorable times. A special thanks goes out to all of my office mates; I will miss the coffee runs and the group effort crossword puzzles. Also, thank you, Jason Price and Sumi Koh, for helping me get through some tough times; your friendship means a lot to me. I would also like to acknowledge my former Buckeye graduates, Cristina Millan and Jessica Albrecht (now Mrs. Olneyl), for their continuing support; great fi’iends like these are invaluable. As always, my family have been very supportive and have been an inspiration to me throughout the years; I am more proud of my family, especially my father, Clifford Tefend, then words can describe. I would like to write about each and every member of my family, but as I have eight wonderful brothers and sisters, two step brothers, fantastic brother- and sister-in-laws (including Cindy Ayers and Doug Everhart), a step sister-in- law, and an awe-inspiring, absolutely fantastic group of twenty nephews and nieces, and of course my mom, step-dad Lee, aunts, uncles, cousins, and grandmother, I would run out of paper and ink toner. This work is dedicated to all of them, but most of all, this is for my stepmom. Sylvia, we will miss you forever. TABLE OF CONTENTS List of Tables ...................................................................................... viii List of Figures ...................................................................................... x CHAPTER 1 INTRODUCTION ................................................................................. l Polytopic Vector Analysis ............................................................... 4 The VSPACE module ............................................................ 7 The PVA module .................................................................. 8 Statement of the Problem ............................................................... 10 Geologic Setting ......................................................................... 13 Geochemistry ............................................................................. 14 Mineralogy ............................................................................... 17 Isotopic Studies .......................................................................... 18 Sampling and Analytical Technique .................................................. 20 CHAPTER 2 PVA RESULTS .................................................................................. 22 Topopah Spring .......................................................................... 22 Tiva Canyon .............................................................................. 27 Ammonia Tanks ......................................................................... 36 Rainier Mesa ............................................................................. 45 CHAPTER 3 SANIDINE ANALYSES ........................................................................ 65 Topopah Spring .......................................................................... 65 Tiva Canyon .............................................................................. 68 Ammonia Tanks ......................................................................... 71 Rainier Mesa ............................................................................. 71 Evidence of Mixing in Rainier Mesa Based on Sanidines and ................... 82 CHAPTER 4 MELT INCLUSION ANALYSES ............................................................ 88 Melt Inclusion Composition of All Ash Flows ...................................... 93 ' CHAPTER 5 DISCUSSION .................................................................................... 98 CHAPTER 6 CONCLUSIONS ................................................................................. 102 APPENDIX 1. Major and trace element analyses of pumice fragments in the Rainier Mesa tuff ........................................................................ 106 vi APPENDIX 2. Major and trace element geochemistry of sanidine and plagioclase grains ........................................................................... 11 1 APPENDIX 3. Major and trace element compositions of melt inclusion ............. 163 REFERENCES .................................................................................. 181 vii LIST OF TABLES Table 1. A). Average compositions of pumice fragments for the high-silica rhyolite (HSR) and low silica (LS) compositions of Topopah Spring tuff. B). End member compositions determined for all Topopah Spring pumice fragment samples from the high-silica rhyolite and low silica pumice fragments, combined (TS—ALL). C). End member compositions determined for the dataset composed of all high-silica rhyolite pumice fragments (TS-HSR) ......... 23 Table 2. A). Average compositions of pumice fragments for the high-silica rhyolite (HSR), intermediate (INT), and lower silica (LS) compositions of Tiva Canyon tuff. B). End member compositions determined for the dataset of all Tiva Canyon high-Silica rhyolite, intermediate, and low silica pumice fragments combined (TC-ALL). C). End member compositions determined for the dataset composed of all high-silica rhyolite and intermediate pumice fragments, combined (TC-HSR+INT). D). End member compositions determined for the dataset composed of all high-silica rhyolite pumice fragments (TC-HSR) ............................................................................................ 29 Table 3. A). Average compositions of pumice fragments for the high-Silica rhyolite (HSR), intermediate (INT), and low silica (LS) compositions of Ammonia Tanks tuff. B). End member compositions determined for datasets composed of all samples of high-silica rhyolite, low silica, and intermediate pumice fragments (AT-ALL). C). End member compositions for the dataset composed of all high- Silica rhyolite and intermediate pumice fragments (AT-HSR+INT). D). End member compositions determined for datasets composed of all low silica and intermediate pumice fragments (AT-LS+INT) ........................................ 37 Table 4. A). Average compositions of pumice fragments for each of the high-silica rhyolite (HSR-l, HSR-2, and HSR-3) and low silica (LS) compositions of Rainier Mesa tuff. B). End member compositions determined for the dataset composed of all samples from the HSR-l, HSR-2, HSR-3, and LS pumice fragments (RM-ALL). C). End member compositions determined for the dataset composed of all samples from the HSR-l, HSR-2, and HSR-3 pumice fragments (RM-ALL HSR). D). End member compositions determined for the dataset composed of all HSR—l pumice fragments (RM-HSRI). E). End member compositions determined for the dataset composed of all HSR-2 and HSR-3 pumice fragments, combined (RM-HSR2+HSR3). F). End member compositions determined for the dataset composed of all low silica pumice fragments (RM-LS). G). End member compositions determined for the dataset composed of all HSR-2 pumice fragments (RM-HSRZ). H). End member compositions determined for the dataset composed of all HSR-3 pumice fragments (RM-HSR3). 1). End member compositions determined for the viii dataset composed of HSR-2, HSR-3, and all LS pumice fragments, combined (RM-HSR2+3+LS) ......................................................................... 46 Table 5. Range of trace element compositions for rim and core analyses of sanidines from Rainer Mesa pumice fragments .................................................... 74 Table 6. Ba concentrations of sanidine phenocrysts within Rainier Mesa pumice fragments. Equilibrium composition = calculated (Ba) composition for sanidines in equilibrium with magma compositions represented by each end member. Calculations are based on a mineral/melt partition coefficient of 4.30 (Mahood and Hildreth, 1983) for Ba in sanidine (for HSR). HSR-l EM3 and HSR-2 EM2 have 0 ppm Ba .............................................................. 83 ix LIST OF FIGURES Figure 1. Location of the southwest Nevada volcanic field, Nevada, showing aerial extent of the Topopah Spring (TS), Tiva Canyon (TC), Rainier Mesa (RM), and Ammonia Tanks tuffs (modified from Huysken et al., 2001) .................... 2 Figure 2. A) Th/Nb versus La plot showing division of Rainier Mesa high-silica rhyolite pumice fragments into three groups (HSR-l , HSR-2, and HSR-3). B) Cumulative frequency plot shown as a normal probability diagram for Rainier Mesa pumice fragments ..................................................................................... l 1 Figure 3. Total alkali diagram of pumice fragment compositions of the major ash- flow sheets of the southwest Nevada volcanic field (LS = low silica; INT = intermediate; HSR = high-silica rhyolite) ............................................... 15 Figure 4. Example of the results of dataset TS-ALL, in which 5 end members were determined by PVA. Each pumice fragment in this dataset has its composition uniquely defined as some proportion of each of these 5 end members, so that the sum is l (a negative fraction is the result of pre-set parameters, and essentially means zero). In this example, the proportion of EM] versus EM2 in A), and EM2 versus EMS in B) are shown for each pumice fragment. Note the two different trends displayed for the low silica (LS) pumice versus the high silica (HSR) pumice fragments, indicating that these two pumice fragment groups represent unrelated magmas ...................................................... 26 Figure 5. Graph of end member compositions determined for the TS-ALL and TS-HSR datasets for Topopah Spring. Empty enclosed area defines the range of pumice fragment compositions for low silica, intermediate silica, and high silica pumice fragments of Topopah Spring (TS), Tiva Canyon, and Ammonia Tanks, combined; shaded region is the range of pumice fragment compositions for Topopah Spring (TS), only. A). Th versus SiOz. B). La versus Rb. . . . . 2.........8 Figure 6. Example of the results of dataset TC-ALL, in which 4 end members were determined by PVA. Note the intermediate (INT) pumice fragments fail to plot between the HSR and LS pumice samples in BM 1 versus EM3. Notice also the different trends for the HSR and LS magmas, indicating that these are unrelated magma batches .................................................................. 33 Figure 7. Graph of end member compositions determined for the TC-ALL and TC-HSR datasets for Tiva Canyon. Empty enclosed area defines the range of pumice fragment compositions for low silica, intermediate silica, and high silica pumice fragments of Topopah Spring, Tiva Canyon (TC), and Ammonia Tanks, combined; shaded region is the range of pumice fragment compositions for Tiva Canyon (TC), only. A). Th versus SiOz. B).La versus Rb35 Figure 8. Graph of end member compositions for the AT-ALL, AT-LS+IN T, and the AT—HSR+INT datasets for Ammonia Tanks. Empty enclosed areas define the range of pumice fragment compositions for low silica, intermediate silica, and high silica pumice fragments of Topopah Spring, Tiva Canyon, and Ammonia Tanks (AT), combined; shaded region is the range of compositions for Ammonia Tanks (AT), only. A). Th versus SiOz. B). La versus Rb. . . ........42 Figure 9. Example of the results of dataset AT-ALL, in which 6 end members were determined by PVA. Note the intermediate (INT) pumice fragments plot between the HSR and three of the LS pumice fragment samples in EMl versus EM4 and EM2 versus EM4, indicating that the intermediate magmas result from mixing between HSR and more evolved compositions of the LS magmas. Notice also the different trends for the HSR and LS magmas, indicating that these are unrelated magma batches ....................................................... 43 Figure 10. CIPW normative compositions of pumice fragments and end members for Figure Figure each dataset of Topopah Spring (TS), Tiva Canyon (TC), and Ammonia Tanks (AT). The open circled areas are the compositional ranges of TS, TC, and AT, combined, and are included here for reference. A). Notice the extreme compositions of several TS and one TC end member (see text for explanation). B). Notice that separating the Ammonia Tanks pumice fragments into two separate datasets (one without HSR samples, the other without LS samples), results in end member compositions that define almost the same compositional space as the AT-ALL dataset (which contains all pumice fragment samples); this is attributed to the extensive mixing among the representative magma batches, which generated an intermediate (INT) magma ........................................ 44 l 1. A). Graph of end member compositions determined for the RM-ALL and RM—ALL-HSR datasets. Empty enclosed areas define the range of pumice fragment compositions for low silica (LS), and all three high silica (HSR) pumice fragments of Rainier Mesa. B). Note how the end member compositions more accurately define the fields of pumice fragment compositions, as well as the tendency of some of the end members to overlap in composition. C) End members determined for each separate pumice fragment group (RM-LS, RM- HSRl, HSR2, and HSR-3) ................................................................. 55 12. Example of the results of dataset RM-ALL-HSR, in which 6 end members were determined by PVA. A) In this example, the proportion of EM] versus EM2 and EM6, and EM2 versus EM3 and EM4 are shown for each HSR pumice fragment. Note the two different trends displayed for the HSR-1 pumice fiagments, and the second, more scattered trend of the HSR-2 and HSR-3 pumice fragments; the HSR-2 and HSR-3 pumice fragments do not separate into two groups, statistically, and most likely represent the same, or a related, magma batch. However, HSR-l must represent a magma batch unrelated to HSR-2 and HSR-3. B) Same results for dataset RM-ALL-HSR, now with each HSR pumice fragment identified separately ................................................... 57 xi Figure 13. CIPW normative compositions of pumice fragments and end members for each dataset of Rainier Mesa (RM). The open enclosed areas are the compositional ranges of low silica pumice fragments and another area represents all three high-silica rhyolite pumice fragment compositions, combined. A). Notice the extreme compositions of most of the end members (see text for explanation). B). Notice that separating the low silica samples into one dataset, and separating the high-silica rhyolite samples into two different datasets (one with only HSR-1, the other dataset with HSR-2 and HSR-3 together) results in end member compositions that better define the total range of pumice fragment compositions ................................................................................ 62 Figure 14. Example of the results of dataset RM-HSR2+3+LS, in which 3 end members were determined by PVA. Notice that the HSR-2 and HSR-3 pumice fragments have the same linear trend, and that many of the HSR-3 samples cannot be explained as mixing between the LS and HSR-2 samples based on their location in these graphs .............................................................. 63 Figure 15. Ba, Sr, and Rb compositions of cores and rims of sanidines from high- silica rhyolite pumice fragments, Topopah Spring tuff. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Circled regions and arrows that connect cores and rims of grains are described in the text66 Figure 16. Ba, Sr, and Rb compositions of cores and rims of sanidines from high-silica rhyolite pumice fragments, Tiva Canyon tuff. Each symbol denotes the core and rim of individual sanidines; 15 % error indicated by bars on each symbol. Arrows connect cores and rims of grains that are described in the text ............................................................................................ 69 Figure 17. Ba, Sr, and Rb compositions of cores and rims of sanidines from high-silica rhyolite pumice fragments, Ammonia Tanks tuff. Each symbol denotes the core and rim of individual sanidines; 15 % error indicated by bars on each symbol. Circled regions and arrows that connect cores and rims of grains are described in the text ..................................................................... 72 Figure 18. Ba concentrations of sanidine cores and rims from each of the three hi gh- silica rhyolite pumice fragments of Rainier Mesa tuff. Note the different scales in the y axes of each graph. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Circled regions and arrows that connect cores and rims of grains are described in the text ............. 76 Figure 19. Sr concentrations of sanidine cores and rims from each of the three hi gh- silica rhyolite pumice fragments of Rainier Mesa tuff. Note the different scales in the y axes of each graph. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Circled regions and arrows that connect cores and rims of grains are described in the text ............. 78 xii Figure 20. Rb concentrations of sanidine cores and rims from each of the three high- silica rhyolite pumice fragments of Rainier Mesa tuff. Note the different scales in the y axes of each graph. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Circled regions and arrows that connect cores and rims of grains are described in the text .............. 80 Figure 21. A). Maximum and minimum Ba concentrations measured in cores (cmax and cm) and rims (rum,x and rmin) of sanidines in the Rainier Mesa high-silica rhyolites. Horizontal lines indicate the calculated Ba concentration of a sanidine in equilibrium with each of the end members (end member HSR-1 em3 and HSR-2 em2 both have 0 ppm Ba). Note'the increase in the range of Ba concentrations from HSR-1 to HSR-3. The large range in rim compositions indicates that eruption occurred prior to equilibration. Not shown: HSR-1 em2 (675 ppm Ba), HSR-2 em4 (705 ppm Ba), HSR-3 em] (3646 ppm Ba), and HSR-3 em3 (1544 ppm Ba). B). La-Rb plot of end members and their location within the range of pumice fragment compositions representative of each Rainier Mesa magma composition (open areas on graph) ............................. 84 Figure 22. A). La versus Rb (ppm) for melt inclusions from high-silica rhyolite pumice fragments from Topopah Spring, Tiva Canyon, and Ammonia Tanks, and melt inclusions from low silica pumice fragments (Topopah Spring) and intermediate silica pumice fragments (Ammonia Tanks). Note that one, possibly two melt inclusions have elevated La contents (arrows), similar to melt inclusions from low silica pumice fragments. Open and shaded fields indicate the range of low silica and high silica pumice fragment compositions, respectively. B). Melt inclusions from high-silica rhyolite and low silica pumice fragments of Rainier Mesa. Open and shaded regions represent the low silica and the high silica pumice fragments (HSR-1, HSR-2, and HSR-3), respectively ....................... 89 Figure 23. Graphs showing the trace element variation in melt inclusions from all high-silica rhyolite pumice fragments of Rainer Mesa, along with end member compositions for four separate datasets (HSR-1, HSR-2, HSR-3, and LS). Note the high Rb and Nb content of Group A inclusions. Glass matrix compositions of all high-silica rhyolite magmas are indicated on each graph. Concentrations are in ppm ................................................................................... 90 Figure 24. Graphs of end member proportions in melt inclusions from high-silica rhyolite pumice fragments. Only the Group A melt inclusions from Rainier Mesa is shown and has a distinctly different trend (circled region) in end member proportions than the melt inclusions from other ash flows, indicating that the melt compositions are related for all high-silica rhyolite magmas except for Rainier Mesa ............................................................................ 95 Figure 25. Triangular plots of trace element compositions of melt inclusions and end members. A). TS, TC, and AT melt inclusions and the 4 end members generated from the dataset TS-TC-AT-HSR. The two high La inclusions of TS, interpreted xiii as trapped melt from xenocrysts, were excluded from the dataset, and therefore not shown on these graphs. B). Group A, Group B, and Group C inclusions with end members generated from the RM-GPA, RM-GP2-and-GP3 dataset. All inclusions are from high-silica rhyolite pumice fragments ........................... 96 Figure 26. Model proposed by Cambray et al. (1995) showing normal dip-slip detachment fault with releasing step. Releasing steps along a normal fault can serve as magma chambers and account for the physical separation of different magma batches. Such releasing steps grow during fault movement, and can accommodate further inputs of magma. Multiple releasing steps can allow the physical separation of magma batches, which come into contact immediately preceding eruption .......................................................................... 99 xiv CHAPTER 1 INTRODUCTION Chemically and mineralogically zoned ash-flow sheets have been used to infer the existence of a compositionally zoned preemptive magma body. Major— and trace-element variations within these ash-flow sheets have been attributed by earlier researchers as in situ differentiation processes within the magma chamber prior to eruption, where hi gh- silica magma has formed by crystal fractionation processes from lower silica magma. However, recent studies have shown that compositionally zoned ash-flow sheets can result from open system processes where chemical variations are due to emplacement of different magma batches into the magma chamber prior to eruption. The Timber Mountain/Oasis Valley magmatic group of the southwest Nevada volcanic field (SWNVF) is a well-studied example where such processes have been inferred. The southwest Nevada volcanic field (SWNVF) contains four large ash-flow sheets, the Topopah Spring (TS), Tiva Canyon (TC), Rainier Mesa (RM), and Ammonia Tanks (AT) tuffs that are some of the best studied series of ash-flow sheets in the world (Figure 1) (Lipman et al., 1966; Christiansen et al., 1977; Byers et al., 1989; Flood et al., 1989a; Flood et al., 1989b; Schuraytz et al., 1989; Warren et al., 1989; Farmer et al., 1991; Cambray et al., 1995; Vogel and Aines, 1996; Mills et al., 1997; Bindeman and Valley, 2003). These ash-flow sheets are interpreted as being compositionally zoned due to the emplacement of discrete, independently generated magma batches into a high-level magma chamber below Timber Mountain prior to eruption (Cambray et al., 1995; Mills "7100' 118E00' mm” STONEWALL MOUNTAIN CALDERA COMPLEX r3r30' ”a __~ ,” ‘x - .......... SILENT CANYON CALDERA ‘. ,I BLACK MOUNTAIN : 'I'l CALDERA _____ . . . |\\ x ‘\ TC TIMBER MOUNTAIN— I“ ‘~\ OASIS VALLEY TS I .3700- /\ CALDERA COMPLEX ~ RM .< and g ........................ AT 3 E / Pun"... ’ ’ I I-ao'ao' " \. \ 2 I . . 1 5° "m \‘ \~, Pahrump 1 \P l . Figure 1. Location Of the southwest Nevada volcanic field, Nevada. showing aerial extents Of the Topopah Spring (TS), Tiva Canyon (TC), Rainier Mesa (RM) and Ammonia Tanks (AT) tuffs. (modified from Huysken et al, 2001 ). et al., 1997). Each of the four ash flows have a lower, volumetrically dominant rhyolitic portion containing high-silica rhyolite pumice fi'agments, and a smaller volume, less evolved upper portion containing high-silica rhyolite and more mafic pumice fragments that represent the entire compositional range of the ash flow. These lower and upper portions of each ash flow have distinct Sr- and Nd- isotopic compositions (Farmer et al., 1991). Furthermore, 8180 compositions show that the lower and upper portions of each ash-flow sheet cannot be related to one another by fractional crystallization, nor can the compositions from one ash-flow sheet be related to another ash-flow sheet by fractional crystallization (Bindeman and Valley, 2003). The origin of silicic magmas in Chemically zoned magma systems has long been an enigma; trace element studies and isotope analyses have been used to constrain the origin of silicic magmas and to infer the existence Of compositionally zoned pre-eruptive magma bodies (Farmer et al., 1989; Cambray et al., 1995; Mills et al., 1997; Bindeman and Valley, 2003). However, in systems that have diverse silicic magma batches, the question still arises ‘what is the relationship of these silicic magmas among the ash-flow sheets, and/or the relationship between the silicic magmas and the coeval mafic magmas within each ash-flow sheet?’. In this study, two types of analyses are used to answer these questions: 1) Polytopic Vector Analysis (PVA), a multivariate statistical method of determining end member compositions in a sample dataset, and 2) trace element compositions of melt inclusions and sanidine phenocrysts. Polytopic Vector Analysis Polytopic Vector Analysis (PVA) is explicitly designed to analyze samples that are mixtures; basically, each sample in a dataset is described in terms of some proportion of each end member generated by the program. Therefore, three parameters are needed to define a mixing system: 1) the number of end members, 2) the composition of each end member, and 3) the relative proportion of each end member in each sample within the dataset. Given knowledge of 1) and 2) the mixing proportions can be derived using many procedures. PVA is designed to estimate all three parameters from ambient data (e.g. chemical analyses of rock samples). The only assumptions are that every end member be present in low proportions in at least one sample and that the proportions of each end member within each sample sums to a constant (such as 1.00). All mixing systems require plotting the data onto a geometric figure termed a simplex. A simplex may occur as a line with end members defined at the two ends (two end member, like the plagioclase series), an equilateral triangle (3 end members), a symmetric tetrahedron (4 end members) or a higher dimensional equivalent. For instance a five end member system requires a four dimensional simplex. Unfortunately the term “simplex” has been used as a label for the Simplex Method of Linear programming. To avoid confusion, the PVA procedure was named Polytopic Vector Analysis. All simplexes are polytopes but not all polytopes are simplexes; only “equilateral” polytopes are simplexes. PVA was developed by and for geologists, and has evolved over a period of about 40 years; the procedure is now used in many other fields. The initial impetus came from John Imbrie who at the time was interested in grain size distributions and micropaleontological data sets (lrnbrie, 1963; Imbrie and Kipp, 1971). Imbrie, and his graduate student Ed Klovan, coded up a version of “Qmode” factor analysis where the data is placed in a covariance matrix that defines relationships between samples. This program was ultimately named “CABFAC” (Klovan, 1968; Klovan and Imbrie, 1971) and is now widely used to analyze faunal assemblages associated with climate analysis. The next major development was made at the suggestion of Al Miesch, an igneous petrologist / geochemist at the USGS. Miesch and Klovan converted CABFAC into EXTENDED QMODEL and added the QMODEL procedure (Miesch, 1976a; Miesch, 1976b; Klovan and Miesch, 1976). Miesch wanted an analytical tool to test hypotheses. His concept was that petrologists had developed models of end member systems for a variety Of igneous rock types and that a better way to evaluate sample data must exist in order to determine whether one or another model was feasible. Most petrologic data sets contain many more variables than the number of expected end members. If the variables (or analytes) were truly independent, then the data must be plotted using one reference axis for each analyte, in which case the data would plot as a multi-dimensional hypersphere. However, there are many correlations that exist between analytes such that fewer than k dimensions (where “k” is the number of variables) are necessary to enclose the data. Miesch and Klovan developed a superior way to determine the number of dimensions necessary for each variable: the now widely used “Coefficients of Determination” table. Once the dimensionality of the data is known, then the number of end members required is simple: one more than the true dimensionality. So, if this analysis determined that the system required 5 end members, for instance, and theory predicted three end members, then the theoretical system was incapable of defining the variability among samples. QMODEL permits the analyst to input compositions of theoretical end members to define a simplex, and if all sample data fits within the simplex, then the imposed end member system is feasible. If some samples fall outside the simplex, negative mixing proportions occur and either the model end members are incorrect, or there are problems with data accuracy. Finally, in 1982 the version that has evolved into the present version of PVA was developed in the context of a particular problem. The research group under Robert Ehrlich had developed a way to quantify grain shape, and it soon became clear that grain shape frequency distributions were polymodal: the sand samples were mixtures of grains of various provenance or transport history. With the help of Klovan, PVA was developed; the principle developer was William Full (Full, et al., 1981, 1982; Ehrlich and Full, 1987). Full’s hyper-dimensional insight resulted in the creation of the DENEG procedure (Full, et al., 1981); when a simplex based on extreme samples in the dataset proved to be insufficient or was mis-oriented, the DENEG procedure allowed an iterative systematic enlargement and rotation of the simplex such that, at convergence, a simplex is defined where the compositions of each end member (located at the vertices) had non- negative components and all of the samples could contain non negative mixing proportions. At this stage the procedure was named EXTENDED QMODEL, which was then refined over the next 20 years, and the procedure was renamed PVA. The evolution in PVA continued with major improvements by Glenn Johnson, including the idea of the CD plot as well as the art and practice of PVA implementation with different data set (Johnson, 1997; Johnson, et al., 2000; Johnson, et al., 2002). The VSPACE Module PVA consists of 2 modules; the first module (VSPACE) is a variant of Q mode factor analysis that decomposes the covariance matrix into eigenvectors. Eigenvectors represent a rotation of the reference axes that were previously defined by the analytes. As with the original axes, the eigenvectors are mutually orthogonal. The orientation of each eigenvector is controlled by the orientation of the cloud of multivariate variance. By design, the first eigenvector is oriented in the direction of highest variance; the second is oriented in the direction of the highest variance residual to the first, and so on. The amount of variance each eigenvector absorbs is measured by the eigenvalue associated with each eigenvector. Because eigenvectors represent progressively less variance, a common assumption is that at some level variance is so low that it mostly consists of random noise, so that if the higher numbered eigenvectors are disregarded, only noise rather than information is lost. This is the equivalent of projecting the data from the original number of dimensions defined by the number of analytes to a lower dimension defined by the reduced set of eigenvectors. A chronic problem has been to decide how many eigenvectors to discard. VSPACE utilizes criteria first described by Klovan and Miesch (1976) and later, extended by Johnson et al. (2002). In general these criteria are based on the agreement, variable by variable, between the values in the original data and the values of each variable obtained by back calculation using progressively fewer retained eigenvectors. If, for instance, all of the variables are well approximated by two eigenvectors, then all of the relationships between samples can be displayed on a two dimensional graph. As discussed above, the number of end members is one higher than the number of necessary dimensions. Often it is unclear whether there are k or (k+1) dimensions; commonly both solutions are run. The difference between the two solutions often hangs on the “importance” of the agreement between raw data and the back-calculated values for one or two variables. The PVA Module The second module is PVA proper. The task of the PVA module is to fit a simplex that encloses the data cloud once the number of end members is determined in the first module, VSPACE. Two criteria are needed to run PVA: 1) an initial guess for the initial simplex and 2) the DENEG procedure (Full et al.,1982). PVA is an iterative procedure; that is, it starts with an initial simplex and then enlarges and reorients it so as to leave no samples outside the final simplex such that all of the vertices (end member compositions) have elements in the negative orthant. There are several ways to define the location Of the initial simplex; the preferred procedure is to choose the most mutually distant samples. The EXRAWC procedure attempts this by initializing on the k samples (where k is the number of end members), each having maximum varimax loadings on an eigenvector; next, a simplex defined by those points is constructed, and then tested, to determine whether any sample is located outside of the simplex (e.g. has negative mixing proportions). The Achilles heel of this option is its” sensitivity to outliers that actually represent data analytical error or entry errors. However, proper use of information from VSPACE can largely mitigate this problem; hence, this is the default option in PVA. It is a simple matter to determine which samples fall outside the polytope, as well as determine the composition of any vertex (the candidate end member); the DENEG procedure is an iterative procedure aimed at defining a simplex that encloses all samples and determines the compositions of the vertices. The procedure is designed to operate incrementally by moving any “side” of the simplex outwards a given distance (the DNEG value) parallel to itself or until the DENEG distance contains no samples; at this point, the procedure signals convergence. Thus, each iteration has two parts: 1) the movement of the simplex edges outwards, thus defining new vertices, and 2) changing any negative elements in the end member compositions to zero, thus rotating the simplex. Sometimes PVA does not converge, or it converges so slowly that the number of iterations is very large. If so, this can be ameliorated by either changing the DENEG value or accepting an iteration that has low negative values in either mixing proportions or end member compositions. Sometimes a lack of convergence reflects the fact that the data cloud is hyper-spherical and thus unmixing is not applicable. PVA was used to evaluate the relationship of the co-erupted magmas within each of the major ash-flow sheets of SWNVF. Pumice fragment geochemistry and melt inclusion compositions were analyzed by PVA; furthermore, the end members recognized by PVA were used to identify disequilibrium compositions in sanidine phenocrysts. Statement of the Problem As stated earlier, the low silica and high silica magmas of each ash flow represent independently generated magmas; this hypothesis, as well as the question of the relationship between the high-silica rhyolites among the ash flows, will be tested by PVA. Furthermore, previous workers had identified the presence of more than one hi gh- silica rhyolite magma represented by the pumice fragments within the Rainier Mesa tuff; these high-silica rhyolite magmas are distinguished from one another by their trace element compositions (Figure 2). PVA will be used to address the relationship of these coevally erupted magma batches of Rainer Mesa. In addition, sanidine and trace element compositions of melt inclusions were obtained in order to further evaluate the relationship among the magmas involved in the generation of these large ash-flow sheets. Trace elements are sensitive indicators to changes in magma composition, and can record open system processes (magma mixing, assimilation), crystal fractionation, the degree of partial melting and the composition of an assimilant added during magmatic evolution. Melt inclusion compositions have provided a highly useful avenue to study magma evolution (see Taylor et al., 1997; Halter, etal., 2002), because these inclusions can record the change in melt composition during crystallization. These melt inclusions represent the melt from which the host phenocryst grew, and the melt inclusion compositions may record changes in melt composition due to crystal fractionation, magma mixing or assimilation (Roedder and Weibblen, 1970; Urusov and Dudnikova, 1998; Frezzotti, 2001). Not all melt inclusions are accurate representatives of the melt composition at the time of entrapment; recognition of melt inclusion compositions altered by post-entrapment re-equilibration is important (Qin et al., 1992; Lu et al., 1995; Nielsen 10 A) 3.5 I I I I I I r I I I I I I 3.0— - <3: eie o 2.5- 43:59., 0&00 ' c *aie 2.0— O he aIe ()0 <> - Th/Nb _ ir- a; *glsaie 6 815%??3 - 0 0 10— ’9 9 <90 — 0.5? - l l J l I l l 1 1 1 I l l J 0.0 0 50 100 150 La Q RM HSR-1 Low Th/Nb, Low La 4]: RM HSR-2 High Th/Nb, Low La ale RM HSR-3 High Th/Nb, High La 0 RM LS Figure 2 A). Tthb versus La plot showing division of Rainier Mesa high-silica rhyolite pumice fragments into three groups (HSR-1, HSR-2, and HSR-3). B). Cumulative frequency plot shown as a normal probability diagram for Rainier Mesa pumice fragments. 3) Normal Probability Plot for Th/La Percent 8 O HSR-1 pumice fragments I HSR-2 pumice fragments . HSR-3 pumice fragments A LS pumice fragments Figure 2. continued. 12 et al., 1998). Melt inclusions can have regular shaped walls defining a negative-crystal shape that may indicate post-entrapment crystallization. In this study, only those inclusions with rounded shapes were selected for analysis in order to eliminate melt inclusions with post-entrapped altered compositions. Furthermore, the compositions of sanidine-bearing melt inclusions were compared with quartz-bearing and plagioclase- bearing melt inclusions in order to identify which melt inclusions may be altered due to interaction with the host phenocryst. Bacon (1989) suggested that the melt trapped during crystallization of the host phase is not representative of the bulk melt due to the formation of a ‘boundary layer’ adjacent to the growing host crystal that is enriched in those elements that diffuse slowly through the melt. However, several researchers report that this boundary layer effect is negligible or absent on melt inclusions of 2 25 pm in size (Anderson, 1974; Lu et al., 1995; Thomas et al., 2001; Fedele et al., 2003). The melt inclusions chosen in this study all have diameters of 2 35 pm. In addition to trace element compositions in the melt, trace element compositions within the solid phases are also useful in modeling changes in magma evolution. Trace element compositions of plagioclase phenocrysts have been used as evidence of Open system processes within a magma chamber such as assimilation and magma mixing Singer et al., 1995; Ginibre etal., 2002; plagioclase phenocrysts can record changes in magma compositions during crystal growth due to the low diffusivities within plagioclase attributed to the Al-Si bond (Grove et al., 1981). Geologic Setting The Timber Mountain/Oasis Valley magmatic complex (Figure 1) formed between 13 Ma and 9.5 Ma in the Southern Great Basin, and is part of the 37 Ma to 5 Ma 13 volcanic activity of the Basin and Range, which has been related to subduction of the F arallon Plate and subsequent regional extension of continental crust (Eaton, 1984; Lipman et al., 1972). Numerous calderas and associated magmatic deposits are collectively called the Southwest Nevada Volcanic Field (SWNVF); the Timber Mountain/Oasis Valley caldera is one of the more recent major calderas within the SWNVF (Christiansen et al., 1977; Byers et al., 1989). The Oasis Valley caldera formed as a result of the eruption of the Paintbrush Group, of which the Topopah Spring (12.8 Ma, 1200 km3) and the Tiva Canyon (12.7 Ma, ~900 km3) members are associated. Eruption of the Rainier Mesa tuff (1 1.6 Ma, 1200 km3) and Ammonia Tanks tuff(11.45 Ma, 900 km3) formed the Timber Mountain caldera. Each ash-flow sheet contains high- silica rhyolite pumice fragments (HSR), and a lower silica (LS) pumice fragment composition (Figure 3). In addition, Tiva Canyon and Ammonia Tanks contain an intermediate (INT), rhyolitic pumice fragment population. Rainier Mesa is unique among these ash-flows in that three high-silica rhyolite pumice fragment compositions can be identified based on Th/Nb and La (HSR-1, HSR-2, and HSR-3). Geochemistry New pumice fragment geochemistry of Rainier Mesa samples have been added to the existing database and are reported in Appendix 1. Large compositional ranges were reported for the pumice fragments within the Ammonia Tanks (59-78 wt% $02) and Rainier Mesa (57-80 wt% SiOz) tuffs of the Timber Mountain group (Mills et al., 1997) (Figure 3). The lower silica pumice fragments range from 59.4-66.9 and 56.8-72.1 wt% SiOz for the Ammonia Tanks (AT-LS) and Rainier Mesa (RM-LS) tuffs, respectively. 14 Na 2O+K 2O Topopah Spring A TS-LS A TS-HSR Na 20+K20 Ammonia Tanks 0 AT-LS 28 AT-INT . AT-HSR 16 14 12 10 (BM-503m 16 14 12 10 Old-503m TIII‘I’TIITTIII 1—ITIIIII Rhyolite lllllL Trach dacite " y A m ‘L' . i- -i :_ Topopah Spring _' r] l l l l I l 14 l l 1 l l l I l l l - 55 6o 65 70 75 Sio2 I I I T I I I I I l I I I ITIIIIII Rhyolite 0 llLllL Ammonia Tanks .: ...ILL puq.g..1 ' 55 60 65 70 75 Sio2 Figure 3. Total alkali diagram Of pumice fragment compositions of the major ash-flow sheets of the southwest Nevada volcanic field. (LS = low silica, INT = intermediate, HSR = high-silica rhyolite. 15 Na 2O+K20 Tiva Canyon Cl TC-LS + TC—INT TC-HSR Na 20+K20 Rainier Mesa 9K- RM HSR-3 Figure 3. continued. .3 a) IIFI'TIIIIIII ITIIIIfT 14 — '- ’ Rhyolite " 12 P ‘ Trachydacite T 10 __ W +++ a“ .. h ‘ 3 8 — / fl 6 > _. 4 - .. 2 _ Tiva Canyon _ 0 P I L I I I J I I I I I I I I J I I I I I T 55 60 65 70 75 SiO2 16FITITIIITI'III TIIIT'IT 14: 12 1o 8 '_andesite $9 6 4 "' . . - RaInIer Mesa 2 f— 0-1 IIIII IIIIIIIIIIIIII 55 60 65 7o 75 Sio2 16 The high-silica rhyolite (HSR) pumice fragments within the Ammonia Tanks (AT-HSR) tuff have >75.8 wt% SiOz. There is also an intermediate pumice fragment group (AT- IN T) of 68.9-73.8 wt% SiOz. Within Rainier Mesa, three separate high-silica rhyolite compositional groups are defined, based on trace element (Th, Nb, and La) concentrations (Figure 2A): a high Th/Nb, high La group (RM-HSR-3), a high Th/Nb, low La group (RM-HSR-2), and finally the low Th/Nb group (RM-HSR-l). Pumice fragment compositions in cumulative frequency plots shown as normal probability diagrams confirms the existence of the separate RM-HSR-l , RM-HSR-2, and RM-HSR-3 compositions within the Rainier Mesa ash-flow (Figure 28). The range of compositional zoning is not as large for the Topopah Springs and Tiva Canyon tuffs (Figure 3); the Topopah Spring tuff contains pumice fragments that range in composition from 69-79 wt% Si02 (Flood et al., 1989a; Schuraytz et al., 1989), and the Tiva Canyon tuff contains pumice fragments that range in composition from 65.9-77.4 wt% SiOz, (Flood et al., 1989a). The low silica pumice fragment compositions range from 69.0-73 wt% SiOz for Topopah Springs (TS-LS) and 65.9-68 wt% SiOz for Tiva Canyon (TC-LS). The high-silica rhyolite pumice fragments within the Topopah Springs (TS-HSR) and Tiva Canyon (TS-HSR) have >75.9 wt% SiOz and >74.1 wt% SiOz, respectively. Within the Tiva Canyon tuff, there is also a pumice fragment group (TC-INT) with intermediate silica compositions between 71 .0-72.6 wt% SiOz. Mineralogy The pumice fragments from these ash flows are nearly aphyric, with sanidine, albite plagioclase and quartz dominating the RM and AT phenocryst assemblages within 17 the high-silica rhyolite samples (Mills et al. 1997), and sanidine and albite plagioclase (no quartz) dominating the TS and TC phenocryst assemblages within the high-silica rhyolite samples (Flood et al., 1989a). High-silica rhyolite pumice fragments from TS, TC and AT have allanite, Chevkinite and perrierite as the LREE-bearing accessory phases, however within the RM high-silica pumice fragments, monazite represents the major LREE-bearing accessory phase (Warren, etal., 1989; Mills etal., 1997). Sphene is also present in AT and TC high-silica rhyolite pumice fragments; it has been noted by others that there is an antipathetic relationship between monazite and both allanite and Sphene (Lyakhovich, 1967; Mackie, 1928; McAdams, 1936; Rapp and Watson, 1986). A more comprehensive report of mineralogy for each of these ash flows can be found in the above cited publications on SWNVF. New analyses of sanidine and plagioclase phenocrysts from pumice fragments of all four ash-flow sheets are listed in Appendix 2. Isotopic Studies The earliest isotopic studies of SWNVF ash flows were on 87Sr/SI’ST, variations using whole rock and feldspar separates (Noble and Hedge, 1969); these authors reported more radiogenic (higher 87Sr/86Sr; ) values for the lower portion (more silicic) versus the upper portion (more mafic) of each ash-flow sheet. Nd and Sr isotopic variations for these ash-flow sheets are reported by Farmer et al. (1991); these authors report lower initial eNd values and higher 87Sr/“Sri for the lower portions of these ash flows as compared to the upper, less evolved portions. Oxygen isotope data was reported by Farmer et a1. (1991) and more recently by Bindeman and Valley (2003). Bindeman and Valley (2003) interpret their data to indicate that the lower silica and high-silica rhyolite 18 magmas involved in the formation of each ash flow are not related by fractional crystallization and assimilation within a single magma chamber, and that these magmas most likely represent independent magma batches. Furthermore, Bindeman and Valley (2003) report that the low silica magmas involved in the generation of each ash flow cannot be related by fractional crystallization, and that the high-silica rhyolite magmas of each ash flow likewise cannot be related by fractional crystallization. Over time, both the low silica and the high-silica rhyolite magmas of the Topopah Spring, Tiva Canyon, and Ammonia Tanks tuffs show a decrease in 8'80, which is consistent with these magmas representing an increase in the contribution of mantle- derived melts; however, Rainier Mesa magmas (both low silica and high silica) have an elevated 8'80 isotopic signature. Bindeman and Valley (2003) proposed that the Rainier Mesa magmas were derived from an l8O-enriched crustal source; this source region dominates the isotopic signature of the Rainier Mesa magmas, thereby obscuring the isotopic contribution from mantle-derived melts. Radiogenic isotopic analyses also provide a stratigraphic framework for these ash flow sheets. Sawyer et al. (1994) reported 40Ar/39Ar isotopic ages to constrain the timing of eruptive events of the SWNVF. More recently, Huysken et al. (2001) reported 40Ar/39Ar isotopic ages for the Post-Grouse Canyon tephra (Oldest age at 13.52 i 0.06 Ma) and the Pre-Rainier Mesa tephra sequences (between 12.79 and 11.84 Ma) that place age constraints on the overlying Topopah Springs and Rainier Mesa tuffs, respectively. Of particular interest is the reported geochemistry of the Pre-Rainier Mesa tephra, which has a lower portion (12.79 i 0.10 Ma) that is geochemically the equivalent of a mixture between Tiva Canyon low silica magma (TC-LS) and Rainier Mesa HSR-1, and an upper 19 portion of the tephra sequence (11.84 i 0.18 Ma) with a geochemistry similar to Rainer Mesa HSR-1, indicating that the 11.6 Ma Rainier Mesa magmas were rapidly generated (within 100,000 years) after eruption of the 12.7 Ma Tiva Canyon magmas. Sampling and Analytical Techniques Most of the pumice fragments were collected by previous workers (Flood et al., 1989a; Flood et al., 1989b; Schuraytz, et al., 1989; Mills et al., 1991; Bindeman and Valley, 2003 ), and they report bulk pumice and mineral analyses. Melt inclusion and matrix glass compositions were reported by Vogel and Aines (1996). Additional Rainier Mesa pumice fragments were collected for this study. For new bulk chemistry analyses of these additional samples, the pumice fragments were leached in a glacial acetic acid/sodium acetate solution to remove secondary carbonate precipitation, and then ground by hand using a ceramic mortar and pestle into a fine powder for fusing into glass disks for analysis. Major and minor elements of pumice fragments were measured using a Rigaku S-Max X-ray fluorescence spectrophotometer (XRF) at Michigan State University. New trace element data was collected by a laser ablation-inductively coupled mass spectrometer (LA-ICPMS) at Michigan State University (Cetac LSX 200+ and Micromass Platform ICP-MS). Earlier reported trace element compositions of pumice fragments were collected by XRF (Rb, Sr, Y, Zr, Nb, La, Ba) and instrumental neutron activation analysis (INAA) at Michigan State University (see above references). All major element data have been normalized to 100% anhydrous conditions. For melt inclusion and phenocryst studies, phenocrysts were liberated from lightly crushed portions of pumice fragments and placed in immersion oil for microscopic 20 examination to select melt inclusion-bearing grains. Only phenocrysts with melt inclusions >35 um were isolated and later mounted in resin on a microscope slide for analysis. After the resin hardened, the grains were exposed by hand grinding using abrasive paper, and later polished using diamond paste on a grinding wheel. Major and minor element analyses for melt inclusions and host sanidine and plagioclase were obtained using a Cameca SX-100 electron microprobe at University of Michigan. Microprobe operating conditions were 15 kv accelerating voltage, counting times were 30 seconds per element, with Na and analyzed first, using a beam spot size was 5 pm. Trace element analyses of melt inclusions and the phenocryst host were obtained by depth profiles on a LA-ICPMS at Michigan State University; depth profiles were 25 pm in diameter, with 10 seconds ablation, at a rate of l uni/sec. Element concentrations were calculated using Ca and Al as internal standards and a NIST 612 glass disc as an external standard. 21 CHAPTER 2 PVA RESULTS Topopah Spring An initial dataset containing all pumice fragments from the low silica (LS) and high-silica rhyolite (HSR) groups were analyzed (Figure 3); the average pumice fragment compositions of these samples are given in Table 1A. PVA for this dataset (labeled TS- ALL), generated five different end members (EM), the compositions of which are listed in Table 1B. The proportion of each of these end members within each individual pumice fragment is also generated by PVA, and an example of the variation in these proportions is shown in Figure 4. Note that the LS and HSR pumice fragments have different trends; in this figure, the proportion of EM] versus EM2 within each individual pumice fragment shows that the HSR pumice fragments have very little variation in EM2 (0-4%), and a wide range of EM] proportions (IO-47%). On the other hand, the LS pumice fragments have more variation in EM2 (13-32%) and low proportions of EM] (0- 18%). Different trends for these two pumice fragment groups are also evident in EM2 versus EMS (Figure 4). Each trend describes the evolution of each magma group, which indicates that the Topopah Spring low silica and high silica samples represent separate magma batches with their own evolutionary trends. If the two pumice fragment groups were related by mixing, they would display similar trends in these EM versus EM graphs. The compositions of these 5 end members also reveal that the LS and HSR magmas are separate magma batches, because the compositions of the end members are 22 A) Ash-flow tuff TS TS TS TS pumice group LS LS HSR HSR #samples n= 12 n=26 ave st. dev. ave st. dev. wt%: Si02 70.7 1 .46 77.3 1 .00 1102 0.4 0.07 0.11 0.027 A1203 15.4 0.81 12.7 0.963 Fe0 1.6 0.26 0.84 0.120 M90 0.4 0.13 0.17 0.104 Mn0 0.1 0.02 0.07 0.021 Ca0 1 .1 0.43 0.63 0.347 Na20 3.8 0.26 2.97 0.340 K20 6.5 0.42 5.16 0.281 P205 0.1 0.03 0.02 0.016 ppm: Th 20.4 1.18 23.8 3.19 La 169 40.6 34.4 4.33 Nb' 33.2 3.14 Rb 141 12.6 196 15.7 Sr 88.3 59.0 20.3 10.0 Zr 533 123 145 31.1 Ce 280 57.5 73.0 10.9 Ta 0.9 0.10 1.47 0.318 Sm 11.2 1.54 5.75 0.746 Eu 2.3 0.80 0.32 0.190 Tb 0.9 0.05 0.77 0.135 Yb 3.0 0.16 3.11 0.526 Lu 0.5 0.10 0.51 0.093 Sc 5.7 1.30 2.13 0.458 Cs 3.5 0.51 5.74 1.31 Hf 11.8 1.77 5.31 0.964 Ba 954 972 93.6 47.5 Table 1 A). Average compositions Of pumice fragments for the high-silica rhyolite (HSR) and lower silica (LS) compositions of Topopah Spring tuff. Also shown are the 'Nb analyzed in only 5 HSR pumice fragments. 23 3) Dataset TS-ALL: End member # EM1 EM2 EM3 EM4 EM5 wt%: Si02 74.6 42.9 75.2 78.3 81.4 1102 0.135 1.59 0.00 0.06 0.05 Al203 17.0 28.1 4.09 12.5 8.87 Fe0 1.05 5.01 0.33 0.566 0.62 M90 0.087 1.26 1.78 0.109 0 Mn0 0.227 0.231 0 0.055 0 Ca0 0 1.91 11.0 0 0.10 Na20 1 .47 6.47 4.60 2.65 4.20 K20 5.31 11.2 2.61 5.66 4.68 P205 0 0.38 0.31 0 0.06 ppm: Th 30 20 14 19 21 La 28 697 0 5 32 Nb Rb 271 0 0 205 169 Sr 0 338 351 0 13 Zr 179 2050 21 84 48 Ce 93 1 135 7 3 59 Ta 2 0 1 1 1 Sm 7 35 1 3 5 Eu 0 10 1 0 0 Th 1 2 1 0 1 Yb 4 5 4 2 3 Lu 1 1 1 0 0 Sc 1 19 2 2 2 Cs 9 0 0 4 6 Hf 7 39 2 3 4 Ba 340 4320 990 0 0 Table 1 8). End member compositions determined for all Topopah Spring pumice samples from the high-silica rhyolite and low silica pumice fragments, combined (T S-ALL). 24 C) Dataset TS-HSR: End member # EM1 EM2 EM3 wt%: Si02 78.5 79.1 74.2 Ti02 0.081 0.083 0.166 Al203 12.2 9.52 15.1 Fe0 0.80 0.674 0.893 M90 0.055 0.284 0.304 Mn0 0.066 0 0.135 Ca0 0.219 1.7 0.811 N320 2.88 3.8 2.84 K20 5.16 4.76 5.44 P205 0 0.085 0.017 ppm: Th 20.8 18.7 28.9 La 27.1 34.5 50.0 Nb Rb 197 138 229 Sr 3.69 49.4 36.2 Zr 108 113 221 Ce 54.5 58.6 114 Ta 1.22 1.18 1.77 Sm 5.16 4.76 6.88 Eu 0.167 0.323 0.431 Tb 0.659 0.651 0.878 Yb 2.80 2.65 3.58 Lu 0.484 0.499 0.537 SC 2.20 2.45 1.76 Cs 5.49 3.25 7.84 Hf 4.38 4.14 7.32 Ba 73.5 0 249 Table 1 C). End member compositions determined for the dataset composed of all high-silica rhyolite pumice fragments (T S-HSR). 25 A) 0.5 A I I I 04 A - 0.3 - i — EM1 0.2 — A - A All AA ,9 _ 0.1 T A A A A A 0.0 — A — _01 l l I 1 ' -0.1 0.0 0.1 0.2 0.3 0.4 EM2 B) 0.4 I I I I I A 0.3 h A A A A - g AA A 0.2 T A ‘ EM2 A 0.1 — _ A AA 00 __ A AM %% -O.1 1 1 l 1 l 0.0 0.1 0.2 0.3 0.4 0.5 0.6 EM5 Figure 4. Example Of the results Of dataset TS—ALL, in which 5 end members were determined by PVA. Each pumice fragment in this dataset has its composition uniquely defined as some proportion Of each Of these 5 end members, so that the sum is 1 (a negative fraction is the result Of preset parameters, and essentially means zero). In this example the proportion Of EM1 versus EM2 in A), and EM2 versus EM 5 in B) are shown for each pumice fragment. Note the two different trends displayed for the low silica (LS, A ) pumice versus the high-silica (HSR, A) pumice fragments, indicating that these two pumice fragment groups represent unrelated magmas. 26 unrealistic. End members that represent related magma batches should have compositions that closely constrain the sample populations that they represent; Figure 5 shows that one of the end members for the TS-ALL dataset has an extremely low Si02 content (Figure 5A), and a very high La content (Figure 5B), whereas the other four end members plot near or within the field of pumice fragment compositions (shaded region on figure). The results of a second dataset comprised of just the HSR samples (dataset TS-HSR) yields three end members, the compositions of which more accurately represent the range of HSR pumice fragment compositions (Table 1C, Figure 5A and B). Tiva Canyon Tiva Canyon pumice fragments represent a low silica magma (LS), a high-silica rhyolite magma (HSR), and an intermediate, rhyolitic magma composition (INT) (Figure 3, Table 2A). PVA of a dataset containing all three pumice fragment compositions (dataset TC-ALL, Table ZB) reveals that the HSR and LS are not related by mixing, as each of these compositional groups have a unique trend in EM proportions (Figure 6). This figure also shows that the INT pumice fragments cannot be the result of mixing between the LS and HSR magmas; note that the EM3 proportion within the INT samples is too low to be consistent with mixing between LS and HSR magmas. Although the compositions of these end members of the TC-ALL dataset are closer representatives of actual pumice fragment compositions (Figure 7) than the previous HSR and LS combined dataset for Topopah Spring (TS-ALL), there is still one end member with uncharacteristically low Si02 content for Tiva Canyon magmas. This is also evident in the trace element-trace element plot (Figure 7B). A second dataset composed of all HSR 27 A) 3) 60 50- 40? Th 30- 20-A 10" | J I l r l 0 4O 45 50 55 60 65 70 75 80 85 8102 70MrIIIIIIIIIIIIIII 600 " 500 — 400 "' La ' 300 " 200 — 100 0g' 50 Rb Figure 5. Graph Of end member compositions determined for the TS-ALL ( A) and TS—HSR ( O) datasets for Topopah Spring. Empty enclosed area defines the range of pumice fragment compositions for low silica, intermediate silica, and high-silica pumice fragments Of Topopah Spring (TS), Tiva Canyon (TC), and Ammonia Tanks (AT), combined; shaded region is the range of pumice fragment compositions for Topopah Spring, only. A). Th versus Si02. B). La versus Rb. 28 A Ash-flow tuff TC TC TC TC TC TC pumice group LS LS INT INT HSR HSR #samples n=4 n=5 n=37 ave st. dev. ave st. dev. ave st. dev. wt%: Si02 66.6 0.594 72.2 1.18 75.9 0.60 Ti02 0.613 0.005 0.32 0.06 0.15 0.01 AI203 17.5 0.34 14.8 0.74 13.1 0.48 Fe0 2.02 0.173 1.30 0.25 0.90 0.14 M90 0.69 0.083 0.31 0.14 0.44 0.41 Mn0 0.13 0.022 0.12 0.03 0.09 0.02 C30 1.63 0.33 0.88 0.46 0.26 0.06 N320 4.49 0.28 3.69 0.41 3.04 0.29 K20 6.3 0.31 6.53 0.14 6.12 0.64 P205 0.11 0.041 0.02 0.01 0.01 0.00 ppm: Th 13.9 1.28 20.2 0.81 24.1 2.78 La 215 12.3 75.5 19.4 30.4 4.89 Nb‘ 37.6 0.99 Rb 123 51.1 144 12.9 212 37.0 Sr“ 222 62.5 69.3 10.3 14.1 4.33 Zr 844 35.4 533 1 16 232 26.4 Ce 400 22.4 165 30.2 67.4 11.3 Ta 0.90 0.20 1.32 0.41 1.67 0.22 Sm 13.5 0.82 9.80 1.52 6.01 0.99 Eu 4.77 0.54 0.98 0.19 0.28 0.11 Tb 0.9 0.0 1.10 0.20 0.95 0.13 Yb 3.03 0.33 4.02 0.45 3.92 0.39 Lu 0.53 0.21 0.76 0.15 0.68 0.14 SC 7.30 0.5 3.48 0.86 1.63 0.16 Cs 2.88 1.31 3.74 0.49 5.34 0.68 Hf 15.1 1.03 12.9 2.56 7.94 0.53 Ba 2621 585 352 129 89.8 55.6 Table 2 A). Average compositions of pumice fragments for the high-silica rhyolite (HSR), intennediate silica (INT), and lower silica (LS) compositions of Tiva Canyon tuff. *Nb analyzed in only 5 HSR pumice fragments. ”Sr analyzed in only 7 HSR pumice fragments. 29 3) Dataset TC-ALL: End member # EM1 EM2 EM3 EM4 wt%: Si02 77.2 60.2 75.1 76.2 Ti02 0.105 0.895 0.134 0.135 Al203 11.9 19.9 13.9 13 Fe0 0.746 2.79 0.992 0.701 M90 0 0.49 1.7 0.01 Mn0 0.038 0.137 0.128 0.127 CaO 0.106 2.49 0.151 0.416 Na20 2.17 4.89 3.45 3.79 K20 7.67 7.45 4.39 5.58 P205 0.01 0.15 0.01 0 ppm: Th 22.6 7.14 22.6 26.6 La 22.1 308 25.8 11.1 Nb Rb 207 0 183 272 Sr Zr 116 1230 191 368 Ce 35.8 570 51 .5 69.4 Ta 1.47 0.18 1.62 2 Sm 2.93 17.3 5.53 10.1 Eu 0.172 6.7 0.276 0 Tb 0.584 0.771 0.877 1.57 Yb 3.09 2.11 3.57 5.81 Lu 0.416 0.274 0.665 1.24 Sc 1.27 10.6 1.49 1.22 Cs 5.07 0 5.33 6.21 Hf 5.29 20 6.96 12.4 Ba 120 3620 140 0 Table 2 8). End member compositions determined for the dataset Of all Tiva Canyon high-silica rhyolite, intermediate. and low silica pumice fragments combined (T C-ALL). 30 C) Dataset TC-HSR+INT: End member 1! EM 1 EM 2 EM 3 EM 4 wt%: Si02 77.5 68.7 75.6 76.5 1102 0.079 0.471 0.108 0.162 AI203 11.7 16 13.6 13 Fe0 0.747 1.74 0.914 0.632 M90 0 0 1.68 0.051 Mn0 0.046 0.143 0.125 0.089 Ca0 0.029 1 .47 0.076 0.383 Na20 2.02 4.01 3.42 3.79 K20 7.83 7.25 4.35 5.27 P205 0.00 0.04 0.01 0.01 ppm: Th 21.2 16.9 23.1 29.2 La 10.2 123 16.2 41.0 Nb Rb 177 47.8 196 364 Sr Zr 70.5 792 153 359 Ce 8.60 263 33.1 119 Ta 1.4 1.01 1.66 2.15 Sm 1.98 14.2 4.97 9.88 Eu 0 1.72 0.101 0.556 Tb 0.508 1.27 0.872 1.49 Yb 2.73 4.17 3.59 5.98 Lu 0.368 0.748 0.673 1.19 Sc 0.723 5.15 1.14 2.22 Cs 4.65 1.56 5.6 7.73 Hf 4.43 17.2 6.45 11.6 Ba 19.8 601 80.3 29.6 Table 2 C). End member compositions determined for the dataset composed Of all high—silica rhyolite and intermediate pumice fragments, combined (1' C-HSR+INT). 31 0) Dataset TC-HSR: End member # EM1 EM2 EM3 EM4 wt%: Si02 77.1 74.8 76.6 74.6 Ti02 0.147 0.148 0.113 0.196 A1203 12.8 14 12.1 13.3 Fe0 0.478 0.986 0.721 1.3 M90 0.19 1.45 0 0.601 Mn0 0.080 0.132 0.045 0.109 CaO 0.344 0.191 0.079 0.611 Na20 3.76 3.63 2.44 2.53 K20 5 4.58 7.77 6.68 P205 0.011 0.014 0.005 0.015 ppm: Th 29.8 21.9 20.0 23.8 La 45.3 24.4 5.91 66.7 Nb Rb 365 184 223 88.7 Sr Zr 365 212 129 313 Ce 125 52.0 10.4 128 Ta 2.27 1.5 1.2 1.79 Sm 10.1 5.46 1.62 9.63 Eu 0.675 0.214 0 0.935 Tb 1 .45 0.864 0.540 1.09 Yb 5.81 3.47 2.65 4.55 Lu 1.18 0.683 0.559 0.416 Sc 2.45 1 .48 0.722 2.64 Cs 7.75 5.24 5.36 3.03 Hf 11.1 7.37 5.59 9.59 Ba 0 140 35.3 258 Table 2 D). End member compositions determined for the dataset composed Of all high-silica rhyolite pumice fragments (T C-I-ISR). 32 0.7 - ._ _ E] - 0.5 ~— .__ g; 1 Cl EM1 03 — a a .. + + _ 0.1 - 1% E] D __ + + D 1:1 .. _01 l 1 I 1 l l I —01 0.1 03 0.5 0.7 EM2 0.7 I I B I I I I r I r f—- D u—i 0.5 — '31 - _ + + - EM2 0.3 + D _ I 0.9 EM3 Figure 6. Example Of the results Of dataset TC-ALL, in which 4 end members were determined by PVA. Note the intermediate (INT,+) pumice fragments fail to plot between the HSR (CI) and LS (El) pumice samples in EM1 versus EM3. two different trends displayed for the low silica (LS) pumice versus the high-silica (HSR) pumice fragments, indicating that the intermediate magmas result from mixing between HSR and LS magmas. Notice also the different trends for the HSR and LS magmas, indicating that these are unrelated magma batches. 33 I LJ r I I I I I F 0.7 - 4 — -I 0.5 -— _ EM1 _ [3 El D ‘ ... .J + + E CI 0.1 _ + Li" D D "' _._ + [:1 Cl .— _0 1 l I I I I l I I l -0 1 0 1 0.3 0 5 0.7 0.9 EM3 0.7 I EI I I m I I I D —l 0.5 _ D El __ + + EM2 0.3 ‘ _I_ T 0.1 - _ -— EL -0 1 l I I I I l I I -0 1 01 0.3 0 5 0.7 Figure 6. continued. 34 60 I I I I I A) * ‘ 50 — — 40 — Th 30 +— 20— 10— 40 45 50 55 85 $10 2 B) 700 600 500 400 La 300 200 100 Figure 7. Graph Of end member compositions determined for the TC-ALL ( a) and TC-HSR ( Q) datasets for Tiva Canyon. Empty enclosed area defines the range of pumice fragment compositions for low silica, intermediate silica, and high-silica pumice fragments Of Topopah Spring (TS), Tiva Canyon (TC), and Ammonia Tanks (AT), combined; shaded region is the range Of pumice fragment compositions for Tiva Canyon, only. A). Th versus Si02. B). La versus Rb. 35 and INT pumice fragments was next analyzed (TC-HSR+INT, Table 2C); end member- end member plots also reveal different trends in end member proportions for these pumice fragment groups, again indicating that these magma types represent separate magma batches (not shown). PVA results on a dataset with only the LS and INT samples are inconclusive due to the small number of pumice fragments with these compositions (not shown), therefore, the relationship of LS with INT cannot be established, although there seems to some relationship, based on the similar trends in BM proportions between these two pumice fragment groups (Figure 6). However, it is Clear that the INT and HSR are unrelated. PVA of the HSR dataset (TC-HSR, Table 2D) yields four end member compositions that have compositions that closely constrain the range Of Tiva Canyon HSR pumice fragment compositions (Figure 7). Note that at least one end member composition lies close to the intermediate field; sanidine phenocrysts in the HSR pumice fragments have trace element compositions that indicate mixing with a less evolved magma occurred (see following section). Limited mixing between the HSR and less evolved magma is indicated based on sanidine compositions, and the end member compositions support this. Ammonia Tanks In addition to a low silica pumice fragment (LS) and high-silica rhyolite pumice fragment (HSR) group, Ammonia Tanks also has an intermediate pumice fragment group (INT) of rhyolitic composition (Figure 3, Table 3A). PVA of a dataset comprising all three groups (dataset AT-ALL, Table 38) results in the generation of three end member compositions that better constrain the Ammonia Tanks pumice fragment compositional 36 A) Ash-flow tuff AT AT AT AT AT AT pumice group LS LS INT INT HSR HSR #samples n=9 n=10 n=34 ave st. dev. ave st. dev. ave st. dev. wt%: Si02 63.47 2.32 70.67 1 .73 77.26 0.58 Ti02 0.81 0.18 0.37 0.08 0.15 0.02 AI203 17.82 0.37 15.44 0.81 12.42 0.29 Fe0 3.41 0.97 1.50 0.29 0.73 0.14 M90 1.35 0.54 0.41 0.26 0.09 0.14 Mn0 0.12 0.02 0.10 0.01 0.08 0.01 Ca0 2.70 0.88 1.11 0.36 0.44 0.12 N320 4.63 0.19 4.05 0.30 3.32 0.28 K20 5.40 0.79 6.29 0.31 5.50 0.39 P205 0.29 0.12 0.06 0.02 0.01 0.01 ppm: Th 14.27 2.91 21.59 1.81 29.2 5.96 La 165.7 41.72 102.6 33.52 33.5 7.60 Nb 16.98 3.29 25.56 2.17 36.7 5.97 Rb 86.50 17.74 140.4 15.68 220.0 16.93 Sr 493.0 207 115.9 42.96 11.11 12.96 Zr 815.0 140.9 344.5 89.03 137.8 20.64 Ce 272.0 69.40 176.0 59.31 65.84 16.51 Sm 10.56 1.36 9.09 1.40 5.63 1.49 Eu 3.23 0.28 1.14 0.41 0.21 0.10 Tb 0.68 0.07 0.70 0.06 0.77 0.26 Yb 2.43 0.08 2.63 0.48 3.05 0.95 Lu 0.41 0.06 0.44 0.09 0.48 0.15 SC 6.79 0.62 3.19 1.01 1.33 0.35 Hf 17.18 2.31 9.43 2.35 5.30 1.28 Ba 3282 1122 569.6 171.4 115.6 71.37 Y 31.14 2.80 32.09 1.71 36.51 6.55 Table 3 A).Average compositions of pumice fragments for the high-silica rhyolite (HSR), intermediate silica (INT), and lower silica (LS) compositions Of Ammonia Tanks tuff. 37 3) Dataset AT-ALL: End member # EM1 EM2 EM3 EM4 EM5 EM6 wt%: Si02 78.8 36.3 80.8 52.1 77.8 80.1 Ti02 0 2.27 0.013 1.02 0.177 0.043 AI203 11.4 25.8 10.9 24.1 11.9 11.3 Fe0 0.114 10.2 0.315 3.32 0.786 0.489 M90 0.222 5.19 0 0.912 0 0 Mn0 0.057 0.196 0.075 0.145 0.071 0.085 CaO 0.323 8.92 0.051 2.41 0.331 0.014 Na20 1.59 6.17 3.07 6.26 3.87 4.09 K20 7.42 2.15 4.76 9.17 4.93 3.76 P205 0 1.09 0 0.163 0 0 Ippm: Th 38.1 0 19.6 3.51 24.7 39.3 La 0 260 0 377 36.1 8.93 Nb 41.8 0.3 34.8 0 54.1 34.3 Rb 223 0 258 0 221 274 Sr 0 1870 0 341 0 0 Zr 0 2270 0 1020 1810 36.2 Ce 17.4 400 0 629 95 24.7 Sm 3.74 11.9 3.28 21.6 10.4 2.91 Eu 0 8.93 0 4.69 0.214 0 Tb 0.825 0.64 0.030 0.771 1.84 0.689 Yb 4.49 2.13 0 2.45 5.65 2.80 Lu 0.647 0.24 0 0.542 0.53 0.733 SC 0 16.6 0.595 9.69 0.182 1.20 Hf 3.24 41.8 0 23.3 8.13 4.36 Ba 0 12000 0 1930 0 0 Y 25.1 29.3 29.6 22.1 88.1 27.1 Table 3 8). End member compositions determined for datasets composed Of all samples Of high-silica rhyolite, low silica. and intermediate pumice fragments (AT-ALL). 38 C) Dataset AT-HSR+|NT: End member if EM1 EM2 EM3 wt%: Si02 77.7 65.9 78.1 1102 0.129 0.543 0.116 A1203 12.2 17.8 12.2 F30 0.66 2.08 0.65 M90 Mn0 C30 0.404 1 .59 0.332 N320 3.21 4.65 3.36 K20 5.61 7.07 5.13 P205 0.009 0.094 0.006 Ippm: Th 34.7 15.5 21.9 La 29.8 156 25.4 Nb 40.1 18.6 29.5 Rb 224 72.3 229 Sr 3.4 209 0.0 Zr 125 518 103 Ce 69.7 266 30.8 Sm 5.75 12.5 4.66 Eu 0.18 1.89 0.025 Tb 1.06 0.87 0.14 Yb 4.33 2.94 0.37 Lu 0.69 0.46 0.06 Sc 1.00 4.44 1.70 Hf 6.3 13.2 2.4 83 37.1 944 147 Y 41.2 33.4 24.4 Table 3 C). End member compositions determined for datasets composed Of all high-silica rhyolite and intermediate pumice fragments (AT-HSR+INT). 39 0) Dataset AT-LS+INT: End member If EM1 EM2 EM3 wt%: Si02 59 73.1 66.2 Ti02 1 .07 0.238 0.575 Al203 18.6 14.4 17.7 Fe0 4.81 1 .04 1 .85 M90 2.15 0.223 0.426 Mn0 0.128 0.084 0.121 C30 4.03 0.815 1.10 N320 4.74 3.79 4.75 K20 4.28 6.26 6.96 P205 0.467 0.018 0.080 ppm: Th 8.91 22.6 20.3 L3 134 58.7 248 Nb 15.1 27.9 18.2 Rb 65.7 160 93 Sr 806 70 1 10 Zr 1070 214 601 Ce 212 101 420 Sm 9.25 7.68 13.6 Eu 3.96 0.446 2.89 Tb 0.67 0.698 0.697 Yb 2.29 2.47 2.85 Lu 0.35 0.415 0.502 Sc 7.95 1 .84 6.53 Hf 21 6.7 14.9 Ba 5210 151 1040 Y 32.3 32.6 29 Table 3 0). End member compositons for the dataset composed Of all low silica and intermediate pumice fragments (AT-HSR+INT), respectively. 40 fields than the TS-ALL or TC-ALL datasets did for their respective pumice fragment populations (Table 3 and Figure 8). Furthermore, the EM plots in Figure 9 confirm that the INT magmas can be described as a mix between HSR and more evolved portions of the LS magma (as first proposed by Mills et al., 1991). In this figure, the circled region indicates LS and HSR pumice fragments with compositions that can generate intermediate (INT) magmas by mixing. The more extensive mixing between the LS and HSR magmas explain why AT-ALL end members are more reasonable in terms of representing all three pumice fragment compositional fields (Figure 8). A breakdown of the dataset into HSR and INT pumice fragments (AT-HSR+INT, Table 3C), and LS and INT pumice fragments (AT-LS+IN T, Table 3D) reveals: 1) the most evolved end member in the LS+INT dataset has a HSR-like magma composition, and 2) the least evolved end member in the HSR+INT has 3 LS magma composition (notice that it falls in the more evolved end of the LS composition field). In other words, PVA recognizes a HSR end member in the LS+INT dataset, and an evolved LS end member for the HSR+INT dataset, which reflects the mixed origin of the INT pumice fragments, confirming earlier major-element multi-linear regression results on specific Ammonia Tanks pumice fragment samples by Mills, et al. (1997). The CIPW normative compositions of all end members can be used to summarize the PVA results for Topopah Spring, Tiva Canyon, and Ammonia Tanks (Figures 10A and B). Note the extreme compositions of several end members in Figure 10A which result from PVA of datasets containing unrelated pumice fragments (or, unrelated magma batches). For comparison, Figure 10B shows only those end member compositions that resulted from analysis of either a single (in the case of Topopah Springs and Tiva 41 50'1'1'1'1 A) " -i 50 — _. )— —r 40 -- — Th 30 — '— 20 — '— 10 - -J 40 45 50 55 60 65 70 75 80 85 B) 700 I I T I I r I T I I I I I T I 600 -- m 500 — _. 400 —' — L3 300 200 100 Figure 8. Graph Of end member compositions determined for the AT-ALL ( 4}), AT-LS and INT ( O), and AT-HSR and lNT ( O) datasets for Ammonia Tanks. Empty enclosed area defines the range Of pumice fragment compositions for low silica, intermediate silica, and high-silica pumice fragments Of Topopah Spring (TS), Tiva Canyon (TC), and Ammonia Tanks (AT), combined; shaded region is the range of pumice fragment compositions for Ammonia Tanks, only. A). Th versus Si02. B). La versus Rb. 42 EM1 EM2 0.6 0.5 0.4 0.3 0.2 @5223 § as 33 EM4 Figure 9. Example of the results of dataset AT-ALL, in which 6 end members were determined by PVA. Note the intermediate (INT,3@) pumice fragments plot between the HSR (Q) and three of the LS (O) pumice samples in EM1 versus EM4 and EM2 versus EM4, indicating that the intermediate magmas result from mixing between HSR and more evolved compositions of the LS magmas. Notice also the different trends for the HSR and LS magmas, indicating that these are unrelated magma batches. 43 Ano hite A A) A TS ALL EM 4— TC ALL EM 0 AT ALL EM Anorthite A A A IA ‘3, v " Quartz Albite 3) HSR ‘. I v u I; ‘3'; I. )1 Quartz T Albite A TS HSR EM [3 TC HSR EM 0 AT LS and INT EM 0 AT HSR and INT EM Figure 10. CIPW normative compositions of pumice fragments and end members for each dataset of Topopah Spring (TS), Tiva Canyon (TC), and Ammonia Tanks (AT). The open circled areas are the compositional ranges of TS, TC, and AT, combined, and are included here for reference. A). Notice the extreme compositions of several TS and and one TC end member (see text for explanation). B). Notice that separating the Ammonia Tanks pumice fragments into two separate datasets (one without HSR samples, the other without LS samples), results in end member compositions that define almost the same compositional space as the AT-ALL dataset (which contains all pumice fragment samples); this is attributed to the extensive mixing among the representative magma batches, which generated an intermediate (INT) magma. 44 Canyon) magma batch, or more than one related (in the case of Ammonia Tanks) magma batches. Rainier Mesa Rainier Mesa proved to be more complex than the other three ash-flow sheets; this ash flow contains three high-silica rhyolite pumice fragment compositions (HSR-1, HSR-2, and HSR-3) in addition to a lower silica (LS) pumice fragment group (Table 4A). Combining all pumice fragment samples into one dataset (RM-ALL) yielded five end member compositions (Table 4B). The compositions of these end members poorly constrain the actual range of pumice fragment compositions, as demonstrated in Figure 11A. A second dataset (RM-ALL HSR) excluding the LS pumice fragments was analyzed by PVA, and this time six end members were recognized (Table 4C, and Figure 1 1A). Most of these end member compositions are also poor representatives of the HSR pumice fragment compositions and have non-magma-like compositions. However, the end member-end member plots showed interesting results: 1) the HSR-1 pumice fragments are clearly unrelated to HSR-2 and HSR-3 based on the observed trends in EM proportions, and 2) the HSR-2 and HSR-3 pumice fragments are related, based on their similar trend (Figure 12A and B). The failure of dataset RM-HSR-ALL to closely define the compositional fields shown in Figure 11A is the direct result of analyzing a dataset that contains unrelated pumice fragments. Breaking down the sample population into a dataset comprised of just the HSR-1 pumice fragments (dataset RM-HSR] , Table 4D), and PVA of another dataset for the HSR-2, HSR-3 pumice fragments combined (dataset RM-HSR2+HSR3, Table 4B), and a third dataset of just LS samples (RM-LS, Table 4F) 45 A) Ash-flow tuff RM RM RM RM RM RM RM RM pumice group LS LS HSR-1 HSR-1 HSR-2 HSR-2 HSR-3 HSR-3 #samples n=29 n=45 n=25 n=22 ave st. dev. ave st. dev. ave 1 st. dev. ave st. dev. wt%: Si02 65.9 4.22 77.3 0.539 76.9 1.32 75.9 1.01 Ti02 0.663 0.34 0.115 0.018 0.174 0.026 0.240 0.032 A1203 16.9 0.85 12.8 0.355 13.3 1.28 13.1 0.511 F60 3.18 1.71 0.569 0.111 0.803 0.180 0.989 0.154 M90 1.21 0.83 0.081 0.150 0.270 0.250 0.195 0.141 MnO 0.092 0.024 0.068 0.011 0.042 0.010 0.051 0.008 CaO 2.69 1.43 0.431 0.156 0.373 0.192 0.691 0.167 Na20 4.05 0.38 3.18 0.417 2.79 0.335 2.89 0.296 K20 5.04 0.85 5.45 0.427 5.28 0.448 5.89 0.425 P205 0.255 0.209 0.018 0.019 0.045 0.042 0.033 0.013 ppm: Th 26.3 8.04 23.1 3.91 33.5 4.44 31.8 3.38 La 111 19.1 23.6 5.61 43.2 7.79 74.3 10.8 Nb 14.6 4.27 30.8 8.31 24.6 8.37 15.3 2.76 Rb 104 25.7 250 26.2 154 24.5 134 34.7 Sr 535 280 5.45 8.97 23.8 10.6 64.0 28.5 Zr 432 77.4 83.0 17.9 129 22.2 183 29.1 Ce 180 27.7 51.9 8.66 87.5 18.0 121 19.3 Ta 1.37 0.662 3.29 0.411 1.96 0.484 Sm 7.70 1.03 5.14 0.535 5.39 0.708 5.88 0.500 Eu 1.85 0.517 0.147 0.063 0.320 0.125 0.602 0.141 Tb 0.77 0.298 0.624 0.188 0.607 0.200 0.545 0.222 Yb 2.12 0.565 2.88 0.803 2.22 0.410 1.53 0.382 Lu 0.31 0.123 0.423 0.119 0.334 0.076 0.199 0.069 Sc 4.91 2.87 3.33 0.960 2.40 0.702 1.80 0.376 Hf 9.59 1.86 3.47 0.878 4.44 0.583 5.01 1.08 Ba 1560 490 63.4 73.1 70.6 51.8 261 138 Y 21.0 5.43 31.1 4.93 22.0 3.76 17.6 3.71 Table 4 A). Average compositions of pumice fragments for each of the high-silica rhyolite (HSR-1, . HSR-2, and HSR-3) and lower silica (LS) compositions of Rainier Mesa tuff. 46 3) Dataset RM—ALL: End member a EM1 EM2 EM4 EM5 ms: Si02 78.3 35.1 82.9 5.54 1102 0 1.5 0 3.84 A1203 11.9 29 11.6 35.4 FeO 0 7.04 0 19.5 M90 0 1.69 0.12 9.47 MnO 0.103 0.235 0 0.342 CaO 0 7.39 0 17 Na20 5.34 13.3 1.28 2.97 K20 4.23 3.24 4.05 2.53 9205 0 0.409 0 2.05 ppm: Th 4.79 8.88 88.9 0 La 0.0 378 11.8 240 Nb 51.9 0 50.6 5.23 Rb 477 0 44.7 0 Sr 0 1770 0 3710 Zr 0 1590 0 1420 Ce 0 525 77 389 Sm 2.93 11.8 4.18 22.8 Eu 0 5.86 0 9.84 Tb 0.437 0 0 1.58 3.46 Yb 4.82 0 0 4.88 5.87 Lu 0.792 0.087 0 0.784 0.767 Hf 1.14 28.4 0 4.3 28.3 Ba 0 6760 0 0 7690 Y 48.9 0 . 2 26.7 63.6 Table 4 8). End member compositions determined for the dataset composed of all samples from the HSR-1, HSR-2, HSR-3, and low silica pumice fragments (RM-ALL). 47 C) Dataset RM-ALL HSR: End member if EM1 EM2 EM3 EM4 EM5 EM6 wt%: Si02 75.9 65.7 69.1 82.4 89 71.8 T102 0.024 0.57 0.411 0.242 0.027 0 Al203 14.4 17.7 20.5 7.92 3.99 15.6 FeO 0.359 2.48 1.98 0.671 0 0.301 M90 0 0 2.39 0 0 0.529 MnO 0.108 0.077 0 0.012 0 0.126 CaO 0.2 2.09 0 0 0 1.39 N820 4.74 3.46 0 8.67 0 0 K20 4.25 7.65 5.36 0 6.99 10.2 P205 0.002 0.042 0.220 0.020 0 0 ppm: Th 18.7 48.9 82.6 38.3 12.9 0.0 La 0.0 196 86.3 78.1 33.8 0.0 Nb 60.5 0.0 39.9 0.0 0.0 32.2 Rb 387 0.0 0.0 0.0 132 515 Sr 0.0 263 52.0 29.1 0.0 0.0 Zr 0.0 451 311 162 31.4 0.0 06 0.0 304 197 92.7 41.3 4.4 Sm 4.50 9.02 6.94 2.50 3.53 7.54 Eu 0 2.01 0.86 0 0.31 0 Tb 1.15 0.59 1.61 0 0.33 0 Yb 6.13 0 2.89 1.78 0 0.65 Lu 0.96 0 0.48 0.68 0 0 Hf 3.4 9.5 11.1 5.6 0.5 0 Ba 12.0 ‘ 1210 0 0 20.5 96.1 Y 53.2 0 25.3 0 2.8 51.1 Table 4 C). End member compositions determined for the dataset composed of all samples from the HSR-1, HSR-2, and HSR-3 pumice fragments (RM-ALL HSR). 48 0) Dataset RM-HSR1: End memberri EM1 EM2 EM3 EM4 wt%: Si02 76.9 78.6 78.2 74.2 T102 0.143 0.0723 0.128 0.135 A1203 13.4 11.9 11.6 14.5 Fe0 0.616 0.314 0.741 0.786 M90 0 0 0 0.931 MnO 0.076 0.052 0.060 0.089 CaO 0.127 0.503 0.51 0.688 Na20 5.01 2.14 3.22 2.2 K20 3.63 6.44 5.48 6.37 9205 0.038 0 0.002 0.041 ppm: Th 30.8 13.3 24.8 26.5 La 22.8 17.4 50.2 10.7 Nb 47.3 19.8 6.1 46.8 Rb 267 282 152 253 Sr 0 2.92 7.36 18.3 Zr 67.8 56.6 132 104 C6 47.1 40.3 68.3 62.9 Sm 3.88 5.95 7.14 3.57 Eu 0.090 0.148 0.049 0.299 Tb 1.25 0.234 0 1.14 Yb 5.79 1.73 0 3.46 Lu 0.932 0.099 0 0.647 Hf 5.58 1.18 3.29 4.27 Be 21.4 157 0 10.0 Y 40.2 25.2 15.1 41.4 Table 4 0). End member compositions determined for the dataset composed of all HSR-1 pumice fragments (RM-HSR-1 ). 49 E) Dataset RM-HSR2+HSR3: End member # EM1 EM2 EM3 EM4 wt%: Si02 71.0 75.3 77.1 80.2 Ti02 0.431 0.232 0.109 0.120 AI203 14.5 15.5 12.6 10.7 Fe0 1.65 0.944 0.560 0.593 M90 0 1.06 0 0.093 MnO 0.071 0.026 0.057 0.038 CaO 1.57 0 0.730 0.254 N320 3.59 2.86 3.65 1.84 K20 6.97 3.89 5.18 6.15 P205 0.034 0.108 0.018 0.003 ppm: Th 32.3 43.4 22.2 29.7 La 146 42.3 14.5 45.7 Nb 3.49 35.6 10.4 19.1 Rb 80.6 155 218 125 Sr 184 20.6 5.65 0 Zr 37 175 63.9 90.1 Ce 218 87 56.9 72.9 Sm 7.54 5.93 4.47 4.73 Eu 1.32 0.259 0 0.412 Tb 0.303 0.903 0 0.888 Yb 0 3.3 1 .47 2.04 Lu 0 0.521 0.259 0.285 Hf 6.38 6.31 1 .55 4.49 Ba 786 44.9 0 32.1 Y 7.17 28.4 15.9 22.1 Table 4 E). End member compositions determined for the dataset composed of all HSR-2 and HSR-3 pumice fragments, combined (RM-HSR-2+HSR-3). 50 F) Dataset RM-LS: End member # EM1 EM2 EM3 wt%: Si02 65 56 73 T102 0.504 1 .47 0.36 Al203 17.9 18.1 14.7 Fe0 2.46 7.32 1.51 M90 0.63 3.27 0.704 MnO 0.100 0.135 0.052 C80 2.51 6.06 0.761 Na20 4.98 3.6 2.97 K20 5.37 2.89 5.87 P205 0.155 0.766 0.068 ppm: Th 20.3 9.94 42.7 La 124 67.4 1 14 Nb 6.45 15.8 25.6 Rb 104 74.2 123 Sr 604 1220 49.5 Zr 590 365 241 Ce 172 133 212 Sm 6.53 10.4 7.79 Eu 2.08 2.99 0.854 Tb 0.463 1.47 0.81 1 Yb 1.16 3.56 2.66 Lu 0.150 0.631 0.358 Hf 12.8 8.77 5.26 Ba 2520 1770 64.9 Y 13.9 32.3 24.4 Table 4 F). End member compositions determined for the dataset composed of all low silica pumice fragments (RM-LS). 51 G) Dataset for RM—HSR2: End member # EM1 EM2 EM3 EM4 wt%: $102 75.7 77.5 75.5 81.1 T102 0.218 0.107 0.231 0.144 A1203 15.2 12.5 13.4 9.91 F90 0.828 0.597 0.959 0.793 M90 0.886 0.063 0.176 0.205 Mn0 0.024 0.060 0.037 0.043 080 0 0.562 0.793 0.078 N320 2.73 2.93 2.98 2.22 K20 4.25 5.63 5.89 5.39 P205 0.073 0.020 0.047 0.034 ppm: Th 39.6 26.3 34.1 26.6 La 45.5 25.3 63.2 39.9 Nb 30.5 14.4 16.6 21.5 Rb 149 201 135 106 Sr 23.3 0.0 48.8 31.8 Zr 157 65.3 172 121 C6 75.0 58.1 133 69.3 Sm 6.13 4.88 5.71 3.61 Eu 0.297 0 0.567 0.374 Tb 0.872 0.269 0.407 0.708 Yb 2.83 2.03 1.40 1.94 Lu 0.453 0.314 0.148 0.312 Hf 5.33 2.82 5.35 3.92 Ba 60.5 0.0 92.2 164 Y 27.0 22.3 16.6 16.4 Table 4 G). End member compositions determined for the dataset composed of all HSR-2 pumice fragments (RM-HSR-Z). 52 H) Dataset for RM-HSR3: End member # EM1 EM2 EM3 EM4 wt%: $102 71.8 78.9 75.4 75.5 Ti02 0.337 0.135 0.233 0.291 A1203 14.8 11.7 13.4 13.2 FeO 1.51 0.54 0.969 1.14 M90 0.30 0 0.036 0.450 MnO 0.059 0.034 0.041 0.071 CaO 1.19 0.376 0.792 0.719 Na20 2.98 1.87 3.44 3.17 K20 6.74 6.46 5.53 5.36 P205 0.074 0 0.035 0.043 ppm: Th 34.2 35.8 21.1 36.7 La 108 62.3 54.2 88.6 Nb 17.4 19.5 10.8 14.0 Rb 132 111 123 162 Sr 161 0.0 112 36.9 Zr 272 90.3 175 221 Ce 165 104 78.8 156 Sm 7.00 6.07 5.15 5.94 Eu 1.15 0.694 0.463 0.447 Tb 0.351 1.18 0.0 0.666 Yb 1.28 2.50 0.157 2.00 Lu 0 0.333 0.116 0.218 Hf 6.29 6.04 2.10 6.06 Ba 848 29.5 359 127 Y 25.5 25.6 10.6 14.1 Table 4 H). End member compositions determined for the dataset composed of all HSR-3 pumice fragments (RM-HSR-s). 53 1) Dataset RM-HSR2+3+LS: End member if EM1 EM2 EM3 wt%: Si02 80.3 39.3 75.6 Ti02 0.124 1.86 0.124 Al203 11.7 26.5 13.2 Fe0 0.593 9.36 0.316 MgO 0.516 4.24 0 MnO 0.024 0.197 0.0547 C30 0 8.10 0.408 Na20 1.98 6.54 3.41 K20 4.72 2.20 6.85 P205 0.0637 0.907 0 ppm: Th 33 9.85 32.9 La 1.11 184 111 Nb 27.3 16.4 8.11 Rb 170 16.9 125 Sr 0 1870 0 Zr 0 993 278 Ce 26.4 295 170 Sm 4.71 13.2 5.65 Eu 0 5.2 0.656 Tb 0.742 1.78 0.202 Yb 2.76 4.29 0.519 Lu 0.446 0.720 0.00209 Hf 2.27 20.9 5.79 Ba 0 4700 538 Y 26.7 35.7 9.53 Table 4 I). End member compositions determined for the dataset composed of all HSR-2, HSR-3, and low silica (LS) pumice fragments (RM-HSR2+3+LS). 54 A) 400'1'1‘1'1'1'1'1'1'1'1' Ln 350 300 l 250 ILLI La 200 150 100 50 0 0 50 100 150 200 250 300 350 400 450 500 550 B) Rb 400 IIIIIIIIIIIIIIIIIIIIU [j 350 I 1 300 "" '- 250 I T 1 La 200 I 150 '1 100 '— 50 ,— 0 HSR-1 _. ORM-Ls 0'.1. .1L1.1.‘ 0 RM-HSR1 0 50 100 150 200 250 300 350 400 450 500 550 Rb {t7 RM-HSR2+HSR3 Figure 11. A). Graph of end member compositions determined for the RM-ALL (It) and RM- ALL-HSR (43>) datasets. Empty enclosed areas define the range of pumice fragment compositions for low silica (LS), and all three high-silica pumice fragments of Rainier Mesa (RM). B). Note how the end member compositions more accurately define the fields of pumice fragment compositions, as well as the tendency of some of the end members to overlap in composition (see text for explanation). C). End members determined for each separate pumice fragment group (RM-LS, RM-HSR1, RM-HSRZ, and RM-HSR3). 55 C) 150 100 La 50 O RM-LS 0 RM-HSR-1 <1: RM-HSR-2 ale RM-HSR-3 Figure 11. continued. 56 A) 0.7 I I. I l I 0.5 — ’ — EM1 0.3 - a); - ale ale 9* _ 7 514* 9* *6 _ ale a” 0.1 — ale *Eale 9* ale - 559169516 516 _ ale 4,, 1 a" .* 1 1 -01 00 01 0.2 03 04 05 EM2 0.5 I I I I 04 - is - 114*? * 03 — 95* — . 216* 111 * 9* EM2 0.2 — Maggie is — ‘ *fiéflfig 914 °‘ ’ it) 4.4. ‘ 0 916 0,, _ o. 4%.. _ '01 I l l l -0.1 0.0 0.1 0.2 0.3 0.4 EM3 Q HSR-1 *HSR-Z and HSR-3 Figure 12. A). Example of the results of dataset RM-ALL-HSR, in which 6 end members were determined by PVA. In this example the proportion of EM1 versus EM2 and EM6, and EM2 versus EM3 and EM4 are shown for each pumice fragment. Note the two different trends displayed for the HSR-1 pumice fragments. and the second, more scattered trend of the HSR-2 and HSR-3 pumice fragments; the HSR-2 and HSR-3 pumice fragments do not separate into two groups statistically, and most likely represent the same, or a related magma batch. However, HSR-1 must represent a magma batch unrelated to HSR-2 and HSR-3. B). Same results from dataset RM-ALL-HSR, now with each HSR pumice fragment identified separately. 57 0.7 I I I I I A) O 0.5 — ‘3‘ its Q 9 EM1 0.3 '- *.-*- _* ** w. 0.1 - 91691315 914 -01 I l l l l 0.5 I T I l 0.4 ”- 0.3 T EM2 0.2 "' 0.1 - -01 l l l l -0.1 0.0 0.1 0.2 0.3 Q HSR-1 *HSR-Z and HSR-3 Figure 12. continued. 58 B) 0.7 I l l l I _ . _ _ f”; — _ .0 “33: E33 _ EM1 0.3 - . 3 .3, n1}: *4 ‘ _ 9 if? ale _ 4} g: 9* 5* 0.1 - 4% 10%;" 9|? ale - _ ‘54}: ‘fi' *6 * alt- - ale ale * 9* -0-1 1 L l I l -0.1 0.0 0 1 0.2 0.3 0 4 0 5 EM2 0.5 I I I I 04 b 9* - 0.3 - 319 *2; - ¥ 9K EM2 0.2 - Male ~ wt%}, :3, 2.314 . Q '5} 0.0 — 0 L1,}: __ _0.1 I I I l -0.1 0.0 0.1 0.2 0.3 0.4 EM3 Q RM-HSR-1 =33 RM-HSR-2 9% RM-HSR-3 Figure 12. continued. 59 B) 0.7 I I r I 0.5 ~— 9 _ %M‘¢‘ EM1 0.3 T 4}“ 0.1 T 9|" gf {p . 2?. _ L. cg; ,4}, _ , 5... _01 1 I l J -0 1 0 0 0.1 0.2 0 3 0.4 0.5 EM6 0.5 I I I I 915 0.4 — *2); - ale 1 9* 0.3 - I; if; _ EM2 0.2 - 0* {119K E *9 {E _ 0.1 — ’ $0 $45“ :5 ., M 6145’ $95 0.0 +— 4} -5 -0.1 l 1 1 1 -0.1 0.0 0.1 0.2 0.3 0.4 EM4 O RM-HSR-1 ‘3" RM-HSR-2 3K RM-HSR-3 Figure 12. continued. 60 resulted in end member compositions that better constrain the field of represented pumice fragment compositions (Figure 12B). Note the overlap of end member compositions for each dataset; for example, one end member in the RM-HSRI dataset has a composition that resembles HSR-2, and one end member in the RM-HSR2+HSR3 dataset closely resembles 8 LS magma, and a second end member resembles a HSR-1 magma. Although PVA rules out magma mixing between any two of these high-silica rhyolite magmas to generate a third high-silica rhyolite, it does indicate that limited mixing occurs between these three high-silica rhyolite magmas; separating HSR-2 and HSR-3 into two different datasets, RM-HSR2 (Table 40) and RM-HSR3 (Table 4H), further demonstrates that mixing between the LS and HSR-3, between HSR-3 and HSR-2, and between HSR-2 and HSR-1 is indicated (Figure 11C). Further support of this interpretation is provided by the trace-element composition of sanidine phenocrysts, melt inclusions, and glass matrix from these pumice fragment groups (see following section). Finally, Figure 13 summarizes the end member compositions (CIPW norm shown), determined by PVA on unrelated magma batches (Figure 13A), and related magma batches (Figure 13B). HSR-2 and HSR-3 remain grouped as one related magma batch, even though HSR-3 can be distinguished from HSR-2 by having elevated La concentrations. A dataset of HSR-2, HSR-3, and LS pumice fragments (RM- HSR2+3+LS, Table 41) analyzed by PVA reveals that HSR-3 cannot be derived by mixing of LS magmas with HSR-2 magmas (Figure 14). However it is clear that HSR-3 was in more contact with LS magmas than HSR-2 based on sanidine and melt inclusion trace element analyses; further evidence of this mixing will be presented in later sections. 61 Ano hite A) 1* RM-ALL EM «- <3> RM-ALL-HSR EM LS + HSR A «1». +1 .. .. Quartz Anorthite Albite 3) O RM-LS EM Q RM-HSR1 EM "’ 9K RM-HSR2+HSR3 EM 4, LS ’0 ALL HSR 0‘. Quartz Albite Figure 13. CIPW normative compositions of pumice fragments and end members for each dataset of Rainier Mesa (RM). The open enclosed areas are the compositional ranges of low silica pumice fragments and another area represents all three high-silica rhyolite pumice fragment compositions, combined. A). Notice the extreme compositions of most of the end members (see text for explanation). B). Notice that separating the low silica samples into one dataset, and separating the high-silica rhyolite samples into two different datasets (one with only HSR-1, the other dataset with HSR-2 and HSR-3 together) results in end member compositions that better defines the total range of pumice fragment compositions. 62 0.8 [E 0.6 _ O O EM1 0.4 — fig. 0 0 O 9K <> <> <> 0 0.2 - 0%? <> _ l O 0 A l 00 J l 1 v1 1 l -0.1 0.1 0.3 0.5 0.7 EM2 0.8 r _ O 06 - 0 _ O i EM1 0.4 — O O 0 E* OOOO§§ 0.2 - 0 0.0 I 4) -1 0 EM3 CC: RM-HSR-2 91$ RM-HSR-3 OLs Figure 14. Example of the results of dataset RM-HSR2+3+LS, in which three end members were determined by PVA. Notice that the HSR-2 and HSR-3 pumice fragments have the same linear trend, and that many of the HSR-3 samples cannot be explained as mixing between the LS and HSR-2 samples based on their location in these graphs. 63 \"f 1 0.7 E 0 _ <> 0 0.5 +— 0 O EM2 _ O 8 0.3 — 0 0.1 ~ -0.1 i -1 0 EM3 ‘3’ RM-HSR-Z *CRM-HSR-3 <>Ls Figure 14. continued. 64 CHAPTER 3 SANIDINE ANALYSES Topopah Spring Trace element compositions (Ba, Sr, and Rb) of sanidine phenocrysts from high- silica pumice fragments are shown in Figure 15; sanidines from the low silica pumice fragments were not analyzed. Sanidine phenocrysts from TS-HSR can be grouped into three populations based on the trace element composition of the phenocryst core. Population 1 has low Ba (~200 ppm) and low Sr (~45 ppm) core compositions, whereas population 2 has elevated Ba (near 700 ppm) and Sr (~120 ppm) in the core. Both of these sanidine populations have similar Rb compositions (~70 ppm) from core to rim. Population 3 consists of sanidines that are xenocrysts derived from a more mafic magma, and these sanidines have very high Ba (> 2500 ppm) and Sr (~600 ppm); one grain from this group is unzoned with respect to Ba (12,000 ppm), the other grain is strongly zoned with a core composition of 2500 ppm Ba and a rim composition of 15,000 ppm Ba. Both of these grains are unzoned with respect to Sr ( ~6OO ppm), and have very low Rb (~40 ppm). Zoning with respect to Ba and Sr is evident in the other sanidine populations as well, as some sanidines have core compositions that fall in population 2, but have decreasing Ba and Sr rimward so that the rim compositions match population 1 compositions. Likewise, one sanidine from population 1 has increasing Ba and Sr rimward and approaches the composition of population 2. The existence of a sanidine population (within the TS-HSR pumice fragments) with more than one core composition 65 140 1 20 1 000 800 Ba 600 400 2.5 _, rim 00 re 200 150 Sr 100 rllIIIIITIIIITIIIII rim llllllllll -( -i 66 Figure 15. Ba, Sr, and Rb compositions of cores and rims of sanidines from high-silica rhyolite pumice fragments, Topopah Spring tuff. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Circled regions and arrows that connect cores and rims of grains are described in the text. LW4-10a-1 and LW4-10a-3 are not shown due to high Ba and Sr values (> 2500 ppm Ba and ~600 ppm Sr). 150, .. 2141 i .1118; 50: " LW4-10a-1 LW4-10a-3 Figure 15. continued. 67 is consistent with open system processes such as magma mixing, with sanidines from a Ba- and Sr-enriched magma zoned with decreasing in Ba and Sr rimward, and sanidines from a Ba- and Sr-poor magma zoned with increasing Ba and Sr rimward as the phenocrysts continue to grow in the now hybrid magma. Tiva Canyon Only sanidines from the high-silica pumice fragments were analyzed, and these show smaller compositional ranges than the sanidines from high-silica pumice fragments of Topopah Springs (Figure 16). Most of these sanidines have low Ba (~3OO ppm) and Sr (~100 ppm) and are unzoned with respect to Ba with the exception of two grains which have rimward increases to >550 ppm Ba. Most of the sanidines are unzoned with respect to Sr. One sanidine grain which has consistenly high Ba (~800 ppm) from core to rim, has high Sr with slight zoning from 145 ppm to 168 ppm core to rim; note that one sanidine with a core composition that matches the composition of the majority of sanidines has a rim composition that matches the trace element composition of this high Ba grain; in fact this zoned grain with the lower Ba core is the only grain that is zoned with respect to Ba, Sr, and Rb. The remainder of sanidines from this high-silica magma have either similar Sr concentrations in core and rim, or have slightly elevated Sr in the cores. Rb concentrations for most grains are low. One grain is zoned to > 150 ppm Rb at the rim; possibly a mineral inclusion or melt inclusion was inadvertently ablated during analysis of the rim, because Rb should decrease as Ba and Sr increase due to mixing with a magma of less evolved composition. 68 1 000 800 600 Ba 400 200 core rim 200 150 Sr 100 Tlllalllillliilll —B— ~ +5! 411,2 COTS «5- 44:3,. HE'- E54 IIJTfi—HMHHII 50 Figure 16. Ba. Sr, and Rb compositions of cores and rims of sanidines from high-silica rhyolite pumice fragments, Tiva Canyon tuff. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Arrows connect cores and rims of grains described in the text 69 Tc5bi-5 ( A ) has >150ppm Rb in the rim 1 00 80 60 Rb 40 20 Figure 16. continued. 70 Ammonia Tanks Only sanidines from the AT-HSR pumice fragments were analyzed, and trace element compositions are shown in Figure 17 . The majority of sanidines have low Sr (~20 ppm) cores, and are unzoned with respect to either trace element. Three sanidines have slightly higher Ba cores (between ~100 and ~14O ppm), and decrease rimward to Ba concentrations that match the rest of the sanidines. These same grains have low Sr in the cores, with the exception of ATS-37-5, which is elevated (~40 ppm Sr), but decreases rimward to match the Sr compositions of the majority of sanidines. Also, one grain (AT5-37-4) was found to have relatively high Ba (~650 ppm core, ~300 ppm rim) and has a Sr concentration in both core and rim that matches the sanidine with elevated Sr (AT5-37-5). Rb concentrations of all sanidines in Ammonia Tanks are roughly similar, and show insignificant changes from core to rim in all but ATS-37-4 (which increases in Rb rimward). None of the sanidines have the high Ba and Sr compositions seen in the other high-silica pumice fragments within the other ash flows; this may be due to the small number of phenocrysts analyzed. Rainier Mesa One sanidine grain from a low silica pumice fragment was analyzed, and shows very high Ba concentrations (6930 to 11,300 ppm Ba, rim to core), a Sr composition between 913 and 1040 ppm, and Rb below detection limits (Table 5). Sanidines from all three high-silica rhyolite groups (RM-HSR-l , RM-HSR-Z, and RM-HSR-3) are shown in Figures 18-20 and are summarized in Table 5. 71 200 rim 150 Ba 100 \ .9. .9. .9. 'B" 50 IITIIIIIIIIITIIIIII Illllllllllllllllll AT5-37-4( O ) not shown, due to high Ba (~300 ppm, rim and ~650 ppm core) 60 rim core AT5-37-4 + _ ATS-37% — / - / / / / 41M - 4,1: <1> 50" 40- + Sr30” 20- 44 + 10" pun.uuunuu\unun-bunn-nun-u Figure 17. Ba, Sr, and Rb compositions of cores and rims of sanidines from high-silica rhyolite pumice fragments, Ammonia Tanks tuff. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Circled regions and arrows that connect cores and rims of grains are described in the text. 72 120 100 80 Rb 60 40" 20— [141 I LLJ I ll Figure 17. continued. 73 .mucmEmm: 8E3 awe—2 .9561 E0: 850.com .o momzmcm 98 new E: .8 mcoEmanoo E953 mom: ,6 0951 .m 53m... .3 E8 32 E9. 8m: .3 E8 25 E388 8 E8 F8 v E8 83 9 .3 E8 83 9 3: E8 «.8 v E9. «8 9 at E8 89 9 08 9mm... E8 9: 9 8.2 E9. 58 9 EN E9. New 9 98 E3 8? 912 E3 98 9 3m E8 Rm. 9 NS «-mw: E55 «9 9 8.3 E8 98 9 5mm E3 95 9 8.8 E8 «2 9 «.8 E8 98 9 ea E8 93 9 8.3 7mm: am am am am 5 mm coamanoo Eco coamanoo ER 74 Most of the sanidines from the RM-HSR-l pumice fragments are unzoned with respect to Ba, Sr and Rb, with variations within detection limits, although one grain is slightly zoned in all three trace elements, and another grain is zoned with respect to Ba only (Figures 18A, 19A, and 20A). The grain that is zoned with respect to Ba, Sr, and Rb has the lowest Ba core (38 ppm) and highest Sr and Rb rim composition compared to the other HSR-1 sanidines. HSR-2 has two sanidine populations based on the trace element compositions of the rims (Figures 18B, 19B and 20B). Population 1 is relatively unzoned with respect to Ba, whereas population 2 has Ba compositions in the core similar to population 1, except for increased Ba and Sr at the rims (~525 ppm Ba and ~85 ppm Sr). Sr compositions generally increase rimward or are relatively constant from core to rim; however, one grain has decreasing Ba and Sr rimward. Still another grain with a low Ba core increases to ~240 ppm Ba, simlar to the normally zoned (with respect to Ba) grain mentioned previously. The presence of these two grains which display either normal or reverse zoning is not uncommon in phenocryst populations in magmas that have undergone mixing. The interpretation here is that there was limited mixing between HSR-1 and HSR-2 magmas; this observation is supported by the melt inclusions (see following section). In addition, the rimward increase in Ba (to ~525 ppm) and Sr (~85 ppm) seen in population 2 sanidines indicates that a second mixing event occurred, this time involving 8 Ba- and Sr~enriched magma. Fewer grains were analyzed in the RM-HSR-3 pumice fragments, and their Ba compositions are highly variable between grains, with two of these showing extreme variations in Ba from core to rim (Figure 18C). One grain has Ba, Sr, and Rb 75 ‘15c1 : A) ' HSR-1 " - rim core - 11x) - § - Ba _ E . 50 - § - 0 : 80ml B) ' HSR-2 -* ‘700 - _. — rim core - 500— Ba 400 '— 300 — 200 *— 100 - Figure 18. Ba concentrations of sanidine cores and rims from each of the three high- silica rhyolite pumice fragments of Rainier Mesa tuff. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Circled regions and arrows that connect cores and rims of grains are described in the text. A). HSR-1. B). HSR-2. C). HSR-3. 76 C ) 2000 1500 Ba 1000 500 Figure 18. continued. Rim: 4368 1655 ppm Ba Core: 4103 i 615 ppm Ba IITI'ITII‘IIIIjIIIII rim i a / s / s m" HSR-3 lllllllllllllllljll 77 50 A) HSR-1 45~ rim core 35- ESr 5“) '- 251- 20— [in 15- bIIIIIIIIIIIIIIIII-III...-IIIIIIIIIIIIIOIIIIIIII 10 B) 140 120 — rim HSR-2 core — 100 — 80 — ESr ' 60— 40— 20 — [3’ Figure 19. Sr concentrations of sanidine cores and rims, with each graph showing phenocrysts from each of the Rainier Mesa high-silica rhyolite pumice fragment groups. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Cirlced regions and arrows that connect cores and rims of grains are described in the text A). HSR-1. B). HSR-2C). HSR-3. 78 C) 500 ; 400 — rim core 300 _ Sr b 200 — 43 4, [I] E A 43 100 — * HSR-3 E 0 I R23-7-1 u.) not shown due to high Sr (core 878 ppm. rim 633 ppm) Figure 19. continued. 79 200 A) 150 F21) 100 50 200 ‘ B) 150 FUD 100 50 _ rim core _ in I E -d r 11 .. h— = - E HSR-1 - r1m ore anneau-u-unu-n-uueaI-IIT \.. ............. inlan- HSR-2 ‘ E’f. Figure 20. Rb concentrations of sanidine cores and rims, with each graph showing phenocrysts from each of the high-silica rhyolite groups of Rainier Mesa. Each symbol denotes the core and rim of individual sanidines; 15% error indicated by bars on each symbol. Circle regions and arrows that connect cores and rims of phenocrysts are described in the text A). HSR-1. B). HSR—2. C). HSR-3. 80 C) 120 100 - rim core 80'- Rb 60 '— 1:1 40— 20 - HSR-3 ionosaouoaaa-e-IIIII-Ileana IIIOIIIIIIIIIOIIIIIJ "E'— R23-7-1 (+) not shown. (Rb below detection limit). Figure 20. continued. 81 concentrations that are identical to the HSR-2 population 1 (Figures 18C, 19C, and 20C), and a second grain (with ~500 ppm Ba in the core and rim) is identical to HSR-2 population 2 in terms of Ba content. A third grain has elevated Ba (~1200 ppm) for the core and rim; another grain has a similar core composition, but is highly zoned with respect to Ba (4368 ppm Ba at the rim). A fifih grain has a similar Ba composition (4103 ppm Ba), but this grain is normally zoned, with a large decrease in Ba rimward (536 ppm); this same grain is the only sanidine zoned in Sr (325 ppm to 178 ppm core to rim). None of the sanidines in HSR-3 show significant zoning in Rb (Figure 20C). Based on the sanidine compositions of the HSR-3 pumice fragments, it appears that some mixing occurred between two different magmas, one resembling HSR-2 and one with an even more Ba— and Sr-enriched composition; this second magma type may be the RM-LS magma. Evidence of Mixing in Rainier Mesa based on Sanidines and PVA The composition of sanidines in equilibrium with each of the end members generated by PVA was calculated using the published partition coefficient for Ba between sanidine and high-silica rhyolite melt (Mahood and Hildreth, 1983). Ba was used for these calculations as sanidine is the dominant phenocryst type and is the only phase in these magmas that would contain significant amounts of Ba (see section on mineralogy). These calculated compositions were compared to measured values in core and rim and the results are shown in Table 6 and Figure 21A. In Figure 21A, the maximum and minimum Ba concentration of cores and rims are shown. The entire range of core and rim compositions of sanidines within HSR-l falls near or between the 82 HSR-1 HSR-2 HSR-3 HSR-1 HSR-2 HSR-3 measured composition min max rim 67.8 ppm Ba 99.8 ppm Ba core 33.8 ppm Ba 67.0 ppm Ba n'm 57.2 ppm Ba 527 ppm Ba core 580 ppm Ba 242 ppm Ba rim 536 ppm Ba 4370 ppm Ba core 134 ppm Ba 4100 ppm Ba equilibrium“ composition EM1 EM2 EM3 EM4 92.0 ppm Ba 675 ppm Ba - 43.0 ppm Ba 260 ppm Ba - 396 ppm Ba 705 ppm Ba 3650 ppm Ba 127 ppm Ba 1540 ppm Ba 546 ppm Ba Table 6. Ba concentrations of sanidine phenocrysts within Rainier Mesa pumice fragments. HSR-1 EM3 and HSR-2 EM2 have 0 ppm Ba. *Equilibrium composition= calculated (Ba) composition for sanidines in equilibrium with magma compositions represented by each end member. Calculations are based on a mineral/melt partition coefficient of 4.30 (Mahood and Hildreth, 1983) for Ba in sanidine in high-silica rhyolite magma. 83 A) 1 HSR-1 em2, HSR-2 em4 4103 ppm Ba 4368 ppm Ba HSR-3 em1, HSR-3 em3 fimn 600 . HSR-3 em4 500 rmm Ba ppm .5 O O 00 O O 200 100 rmm' HSR-2 HSR-1 em3, HSR-2 em2 Figure 21. A) Maximum and minimum Ba concentrations measured in cores (Cmax and Cmin) and rims (1' max and l’ min) of sanidines in the Rainier Mesa high-silica rhyolites. Horizontal lines indicate the calculated Ba concentration of a sanidine in equilibrium with each of the end members (end members HSR-1 em3 and HSR-2 em2 both have 0 ppm Ba). Note the increase in the range of Ba concentrations from HSR-1 to HSR-3. The large range in rim compositions indicates that eruption occurred prior to equilibration. Note shown: HSR-1 em2 (675 ppm Ba), HSR-2 em4 (705 ppm Ba), HSR-3 em1 (3646 ppm Ba), and HSR-3 em3 (1544 ppm Ba). B) La-Rb plot of and members and their location within the range of pumice fragment compositions representative of each Rainier Mesa magma composition (open areas on graph). 84 3) 150 I I I HSR-3 em1 10 0 HSR-3 em4 HSR-2 em3 La 50 - HSR-2 em2 HSR-1 em 0 .1.1.1.1.1.1.1 0 50 100 150 200 250 300 350 Rb O RM-LS O RM-HSR-1 4}: RM-HSR-2 ale RM—HSR-3 Figure 21. continued. 85 calculated compositions of sanidines in equilibrium with HSR-l EM1 and EM4; note that both of these end members lie within the HSR-l compositional field in La-Rb space (Figure 21 B). Sanidines from HSR-2 have a larger range of Ba concentrations within the cores and rims (Figure 21A). The lower Ba core and rim (cm;“ and rmin) compositions are not in equilibrium with any of the HSR-2 end members, but are near equilibrium compositions of HSR-1 EM4; note in Figure 213 that HSR-1 EM4 plots near HSR-2 EM2 in the HSR- ] compositional field in La—Rb space. The similarity of HSR-2 EM2 and HSR-1 EM4 is interpreted here as representing the mixing of magmas with HSR-1 and HSR-2 compositions. The HSR-2 cores with high Ba (emu) have compositions that approach the compositions of sanidines in equilibrium with HSR-2 EM1, which lies in the compositional field of HSR-2 pumice fragments (Figure 21B). However, the maximum Ba concentration in the rims (rmax) are not in equilibrium with any of the HSR-2 end members, but their compositions approach that of the calculated concentrations of sanidines in equilibrium with HSR-3 EM4; this end member lies well within the HSR-3 compositional field (Figure 218). Sanidines from HSR-3 have the largest range of Ba compositions in both cores and rims, with maximum values in the 0.1wt% range (not shown). Lower Ba cores (Cmin) have compositions that are close to the calculated composition of sanidines in equilibrium with HSR-3 EM2 (Figure 21A); this end member falls within the compositional field of HSR-3 pumice fragments in La-Rb space (Figure 21 B). Higher Ba cores (Cmax) have Ba concentrations that exceed all HSR-3 end member compositions (see Table 6 and figure caption for 21A) and are interpreted here as xenocrysts from the RM-LS magma. Rim 86 compositions are also very high in Ba; the minimum compositions (rmm) are near the HSR-3 EM4 calculated equilibrium compositions, but the higher Ba rims (rmax) have extremely high Ba concentrations (up to 4368 ppm Ba). The large range in core compositions of HSR-3 is interpreted here as representing sanidines derived from mixing events involving HSR-2 and LS magmas. Rim compositions of HSR-2 sanidines approach values for HSR-3 sanidines, so that maximum rim values in HSR-2 are similar to minimum values for HSR-3 sanidines. The range of rim compositions for HSR-2 from rm to rmax and rm," to rmax in HSR-3 sanidines is a direct consequence of the relatively short time interval between mixing due to the limited contact of HSR-2 magmas with HSR-1 and HSR-3 magmas, and mixing due to limited contact of HSR-3 with HSR-2 and RM-LS magmas, prior to eruption which resulted in the zoned sanidines with disequilibrium compositions. The trace element compositions of melt inclusions from Rainier Mesa magmas support the role of mixing between the HSR-1 and HSR-2 magmas, as well as mixing between HSR-2 and HSR-3. In the section below, melt inclusion analyses from Rainier Mesa are reported, and their compositions are compared with the melt inclusion compositions of the other ash-flow sheets. 87 CHAPTER 4 MELT INCLUSION ANALYSES Major and trace element compositions of melt inclusions analyzed in each ash flow are reported in Appendix 3. Melt inclusions from all HSR pumice fragments of Topopah Spring, Tiva Canyon, and Ammonia Tanks typically have less Th, La, Nb and Rb than their respective pumice fragments, with the exception of one melt inclusion (LW4-10a-3-1b) from a Topopah Spring high-silica pumice fragment has elevated La (Figure 22A). The host phenocryst for the Topopah Spring melt inclusion mentioned above has extremely an high Ba and Sr composition; this phenocryst must be xenocrystic, and may indicate that some mixing has occurred between the low silica and high silica magmas, since the melt inclusion has a La content similar to the melt inclusions from the LS pumice fragments. Melt inclusions in Rainier Mesa HSR pumice fragments show a larger range in trace element compositions than the melt inclusions from other ash flows, particularly in Rb and Nb (Figures 223 and 23). These melt inclusions are categorized into three groups (Groups A, B, and C) representing early-, mid-, to late-entrapped melts; Group A inclusions represent the earliest trapped melt, followed by Group B and Group C based on the occurrence of melt inclusions within phenocrysts that contain multiple melt inclusions. For example, a quartz grain (R18-16-5) and a sanidine (99TM14-3) both have a Group A inclusion located in the inner portion of each phenocryst, and both grains have a Group B inclusion located nearer the rim. Also, a quartz grain (R8-42-6) and 88 A) 250 200 1 50 La 1 00 50 00 9 Melt inclusion from HSR‘ * Melt inclusion from lNT " {:7 Melt inclusion from LS 1.. 150 200 250 300 350 Rb l ' I <9 Group A melt inclusion 0 Group B melt inclusion X Group C melt inclusion {k melt inclusion from L8 100 200 300 400 500 600 700 Rb Figure 22. A) La versus Rb (ppm) for melt inclusions from high-silica rhyolite pumice fragments from Topopah Spring, Tiva Canyon, and Ammonia Tanks, and melt inclusions from low silica pumice fragments (Topopah Spring) and intermediate silica pumice fragments (Ammonia Tanks). Note that one, possibly two melt inclusions have elevated La contents (arrows), similar to melt inclusions from low silica pumice fragments. Open and shaded fields indicate the range of low silica and high-silica pumice fragment compositions, respectively. B). Melt inclusions from high-silica rhyolite and low silica pumice fragments of Rainier Mesa. Open and shaded regions represent the low silica and the high-silica pumice fragments (HSR-1, HSR-2, and HSR-3), respectively. 89 Glass matrix (all HSR) l . l I L I L I I I I I I I l 0 50 100 1 50 200 Nb HSR-1 EM Grou A melt inclusion . <3) p \ Found mostly in HSR-1 c1} HSR-2 EM III Group B melt inclusion \ _ _ , Found 1n all HSR -*- 11312-3 EM x Group C melt 1nclus1on ‘\ Found mostly in HSR-3 0 LS EM Figure 23. Graphs showing the trace element variation in melt inclusions from all high-silica rhyolite pumice fragments of Rainier Mesa, along with end member compositions for four separate datasets (HSR-1, HSR-2, HSR-3, and LS). Note the high Rb and Nb content of Group A inclusions. Glass matrix compositions of all high-silica rhyolite magmas are indicated on each graph. Concentrations are in ppm. 90 60 I I I F I T ' I I I I r I I 50 .— 0 ‘3’ 9 j _ x 40 _ 53’ . ‘39 *3. 9K- * j Th 30 - 0 _ _ 0* T L q 20 + * Glass matrix (all 0 HSR) 1 10 - 0 l I 1 m L 1 4 l I I 1 l ' I 0 50 100 150 La . HSR-1 EM Group A melt inclusion \ cc: HSR-2 EM + Group B melt inclusion \ fie HSR-3 EM x Group C melt inclusion 0 LS EM \ Found mostly in HSR-3 Found mostly in HSR-1 Found in all HSR Figure 23. continued. 91 sanidine (R8-42-3) with a Group B inclusion in the inner portion of each grain both have Group C inclusions near the rim. Group A inclusions have compositions that range to unusually high Rb and Nb contents (up to 3X and 20X higher than the host pumice fragment, respectively) (Figure 22B, and Figure 23A and B). These melt inclusions are predominately found in HSR-1 pumice fragments. The second group (Group B) describes melt inclusions that have La and Th concentrations that match the La and Th concentrations of the glass matrix of all three high-silica pumice fragment compositions; these melt inclusions are found in all three high-silica pumice fragment groups (HSR-1, HSR-2, and HSR-3). Group C consists of those melt inclusions that have Rb and Nb concentrations that match the glass matrix composition of all high-silica pumice fragment groups, except for elevated La and Th. These Group C melt inclusions are found almost exclusively in HSR-3 pumice fragments. It should be mentioned that not all Group B inclusions necessarily formed before Group C melt inclusions; those melt inclusions that have low Th and La, as well as low Rb and Nb, have compositions that are well within the compositional range of the glass matrix of all HSR pumice fragments, and therefore represent the last entrapped melts. What was important for this study was the identification of a group of melt inclusions that are the earliest trapped melts (Group A) and those inclusions that record an input of Th and La enriched melt (Group C). As the sanidine analyses indicate mixing has occurred among these high-silica magma types, the increase in Th and La recorded in Group C inclusions, as well as their occurrence in mainly the HSR-3 pumice fragments, most likely record the change in melt composition due to mixing with a Th-, La- enriched magma (as well as Ba— and Sr—enriched, as evidenced by the sanidine compositions) such 92 as the LS magma; the presence of a few Group C inclusions in HSR-1 and HSR-2, and the presence of a group A inclusion in HSR-2 is consistent with limited mixing among the magmas of Rainier Mesa prior to eruption. However, as shown by PVA, mixing between any two high-silica magmas does not generate the third high-silica magma type (nor can mixing between HSR-2 and LS magmas generate HSR-3 magma), and that at least two independent high-silica magmas were involved in the formation of Rainier Mesa. Melt inclusion composition of all ash flows Figure 22A shows La versus Rb concentrations of all melt inclusions from Topopah Spring, Tiva Canyon, and Ammonia Tanks tuffs along with the range of compositions represented by the pumice fragments in these ash flows. Notice that: 1) all melt inclusions from the high-silica pumice fragments of each ash flow are indistinguishable from one another, and plot along the same trend; 2) melt inclusions from low-silica pumice fragments appear to have a separate trend; and 3), at least one inclusion from a high-silica pumice fragment has a composition that more closely resembles the melt composition of a low-silica pumice fiagment. Figure 22B shows the Rainier Mesa melt inclusions for comparison. All melt inclusions from the high-silica pumice fragments of all four ash flows were analyzed as one dataset by PVA, and the results show two distinct melt populations: one group is comprised of only Group A melt inclusions from Rainier Mesa, and the second group contains all the remaining melt inclusions from the high-silica pumice fragments of all four ash flows. Group B and Group C melt inclusions from Rainier Mesa were excluded, as these inclusions have 93 compositions that reflect mixing between HSR-2, HSR-3, and LS magmas, and therefore result in a lot of scatter in the EM plots in Figure 24. It is worth mentioning that PVA utilizes the entire geochemistry of the melt inclusions when determining relationships, not just a few key trace elements, and still the Group A inclusions are singled out as unique. Also, the fact that all remaining melt inclusion compositions are grouped as one related melt implies that the high silica magmas of Topopah Spring, Tiva Canyon, Ammonia Tanks, share a common end member. Finally, the compositions of end members generated by PVA of melt compositions are compared in Figure 25. The dataset composed of melt inclusions from high-silica rhyolite pumice fragments of Topopah Spring, Tiva Canyon, and Ammonia Tanks (TSTCAT-ALL) is described by four end members, all of which are similar in terms of their compositions (Figure 25A). The Rainier Mesa inclusions were first analyzed together (RM-ALLHSR), and then subdivided into two subsets, one with all the Group A inclusions (RM-GPA), and another dataset with the remainder of inclusions (RM-GPB+GPC) (Figure 25B). Notice the wide range in end member compositions; as with the pumice fragment compositions, the large range of Rainier Mesa end member compositions is most likely due to the input from LS magmas to HSR-3. It is therefore not surprising, as some end members of the RM-ALLHSR and RM-GPB+GPC datasets have either 0 ppm Rb or 0 ppm Nb; both of these trace elements are of low concentrations within the RM-LS magma, and those end members with 0 ppm Rb and 0 ppm Nb may represent the input of LS magma. 94 EM1 EM2 Figure 24. Graphs of end member proportions in melt inclusions from high-silica rhyolite pumice fragments: Topopah Spring (A), Tiva Canyon (I), Rainier Mesa Group A (43:), Ammonia Tanks (0). Only the Group A melt inclusions from Rainier Mesa are shown and have a distinctly different trend (circled region) in end member proportions than the melt inclusions from other ash flows, indicating that the melt compositions are related for all high-silica rhyolite magmas except for Rainier Mesa. 95 A) La 43> TS,TC and AT HSR melt inclusions O TS-TC-AT-HSR EM La Th Th Nb Figure 25. Triangular plots of trace element composition of melt inclusions and end members. A) TS, TC, and AT melt inclusions and the 4 end members generated from dataset TS-TC-AT- HSR. The two high La inclusions of TS, interpreted as trapped melt from xenocrysts, were excluded from the dataset, and therefore not shown on the graphs. B) Group A, Group B, and Group C inclusions with end members generated from the RM-GPA dataset, RM-GPB+C dataset, and RM-ALLHSR dataset. All inclusions are from high-silica rhyolite pumice fragments. 96 3) Th Th Figure 25. continued. La Group A (RM) melt inclusions + Group B (RM) melt inclusions 3‘ Group C (RM) melt inclusions A RM-GPA EM A RM-GPB+C EM 1:] RM-ALLHSR EM 97 CHAPTER 5 DISCUSSION Based on previous whole rock and pumice fragment geochemical and isotopic studies, the co-erupted low silica and high-silica rhyolite magmas associated with each of the four major ash-flow sheets of southwest Nevada volcanic field cannot be explained as magmas related by processes occurring within a single magma chamber, and therefore represent independently generated magma batches. That such magmas can erupt from the same nested caldera has been attributed to the extensional environment of this portion of the southern Great Basin (Cambray et al., 1995). These authors describe how releasing steps associated with, in this case, normal dip-slip detachment faults, can serve as magma chambers that expand during extension and can accommodate further influxes of magma (Figure 26). Lower rates of extension can result in a greater internal pressure within the releasing step (magma chamber), and with the addition of new magma, result in eruption. Such a mechanism for eruption in this tectonic regime explains why periods of major extension and major volcanism are not contemporaneous (Sawyer et al., 1994). Releasing steps along such a fault system can allow the spatial separation of discrete magma batches in different regions of the crust, which eventually mingle during the eruption process that results in the formation of the major ash-flow sheets in this study. The hypothesis that the low silica and high-silica rhyolite magmas associated with each ash-flow sheet are unrelated is supported by PVA of the pumice fragments, based on 1) the identification of distinct trends in the variation of end member proportions in EM 98 Releasing steps may \ serve as _ chambers \ / for magma LS magma Figure 26. Model proposed by Cambray et al. (1995) showing normal dip—slip detachment fault with releasing step. As extension continues, releasing steps can accommodate more magma. Multiple releasing steps can allow the physical separation of magma batches that come into contact immediately preceding eruption. 99 plots, and 2) the generation of end members with unrealistic compositions, which fail to constrain the range of pumice fragment compositions in the dataset. Based on these same criteria, the relationship of the three high-silica rhyolite magmas of Rainier Mesa (HSR- 1, HSR-2, and HSR-3) is also determined; HSR-1 is identified as a separate magma batch unrelated to the other coerupted magmas (HSR-2, HSR-3, and LS). HSR-3 cannot be explained as the result of magma mixing between HSR-2 and LS; however, HSR-3 and HSR-2 are very similar, and together may represent a second high-silica rhyolite magma batch. That limited mixing occurred among HSR-1, HSR-2, and HSR-3 is supported by the melt inclusions and sanidine trace element compositions. Furthermore, the determination of sanidine compositions that are in equilibrium with one of the end members allowed the identification of sanidines that are xenocrysts derived from the magma involved in mixing. In other words, if a sanidine from HSR-2 has a core composition that is in equilibrium with a HSR-2 end member that plots within the HSR-3 compositional field, then that sanidine first crystallized in HSR-3 magma, and later became incorporated in HSR-2 magma during a mixing event. Not only does the use of PVA on datasets composed of pumice fragment samples confirm the existence of independently generated magma batches associated with each of the four major ash-flow sheets, but a relationship among the high-silica magmas can be determined. Both the trace element variation of melt inclusions from high—silica rhyolite pumice fragments, and PVA of these same melt inclusions, establishes that the high-silica rhyolite magmas of Topopah Spring, Tiva Canyon, and Ammonia Tanks have a common source, and followed similar evolutionary paths. Conversely, Rainier Mesa is unique among these ash-flow sheets in that l) at least two high-silica rhyolite magma 100 compositions are determined, and 2) there are melt inclusions with high Rb and Nb (Group A inclusions) that are unique among these studied ash-flow sheets, and based on PVA, represent melt compositions that are distinct from the other ash-flow sheets, and finally, 3) Rainier Mesa magmas were derived from an l8O-enriched crustal source, which dominates the isotopic signature of the Rainier Mesa magmas. Rainer Mesa magmas were generated during a relatively large time interval (1.1 My between Tiva Canyon and Rainier Mesa) at a time when extension rates in this region of the southern Great Basin were high; this explains the lack of significant eruptions during this time, as releasing steps can accommodate more magma during extension (Figure 26). What is proposed here is that, as a result of the large amount of extension, a different region of the lower to mid level crust underwent melting which produced the Rainier Mesa magmas; this region of the crust had an elevated 6'80, perhaps due to hydrothermal alteration. Following the caldera-forming eruption of the Rainier Mesa magmas, Ammonia Tanks magmas were quickly generated and erupted a short 150,000 years after the formation of the Rainier Mesa tuff; despite the short time span between these eruptions, Ammonia Tanks magmas are more alike Topopah Springs and Tiva Canyon in terms of pumice fragment and glass geochemistry, and PVA results on TS, TC, and AT pumice fragments and melt inclusion compositions. Thus it appears that whatever crustal source was involved in the generation of Rainier Mesa magmas was no longer involved in magma generation post-Rainier Mesa. 101 CHAPTER 6 CONCLUSIONS End member compositions generated by PVA give valuable information as to whether or not a dataset comprises more than one magma batch (ie. whether or not the samples are related). If end member compositions adequately constrain the field of represented pumice fragment compositions, then the pumice fragments in the dataset (and therefore the magmas they represent) are related. Variations in the proportions of each of these end members within each pumice fragment sample result in the immediate recognition of one or more pumice fragment groups (if they represent unrelated magma batches). Breaking down a dataset into smaller subgroups is the next logical step in determining the relationship of, in this case, co-erupted magmas. As shown in Rainier Mesa, the interaction of three high-silica rhyolites and a low silica magma was detected. Furthermore, mixing between two magma batches was identified as the origin of the Ammonia Tanks INT pumice fragment group, whereas the Tiva Canyon INT pumice fragment group was not found to be a hybrid magma generated by mixing between high- silica rhyolite and low silica magmas. However, mixing is ofien subtle and not recognized in the sample geochemistry, such as within the Rainier Mesa high-silica rhyolite magmas. PVA-determined end member compositions for the HSR-1 and HSR2- HSR-3 datasets reveal an overlap in end member compositions that span more than one representative HSR composition field. These HSR pumice fragments have sanidine, melt inclusion, and glass matrix trace-element compositions that indicate mixing, which 102 involved all three HSR as well as a less evolved magma type, and PVA detects limited mixing among independent magma batches where conventional major and trace element least squares linear regression modeling fails. For the ash flows of SWNVF, independent magmas were involved in each ash flow formation, and PVA can establish a common source for the high-silica magmas of Topopah Spring, Tiva Canyon, and Ammonia Tanks despite limited mixing with less evolved magmas and the overprinting of the isotopic signature from mantle-derived melts. Until now, magma mixing has been detected by a combination of bulk geochemistry, phenocryst and glass geochemistry, and also by petrographic analysis; the results of this study show that PVA along with traditional methods of evaluation provides a powerful technique of evaluating magma petrogenesis. 103 APPENDICES 104 APPENDIX 1 105 APPENDIX 1. Major and trace element analyses of pumice fragments in Rainier Mesa tuff. Sample ID 86RM6-16 02RM19—15B 99TM-13 020724-RM17 020724RM-18 02Rm19-10 Group HSR-1 HSR-1 HSR-1 HSR-1 HSR-1 HSR-1 wt%: Si02 77.4 76.8 77.2 77.0 77.2 77.3 Ti02 0.15 0.16 0.14 0.15 0.14 0.14 A1203 13.0 13.3 12.9 12.8 12.7 12.9 Fe0 0.65 0.68 0.7 0.51 0.46 0.57 M90 0 0 0 0 0 0 MnO 0.07 0.07 0.05 0.07 0.06 0.06 080 0.1 0.1 0.38 0.35 0.34 0.34 N820 3.7 3.8 3.4 3.27 3.51 3.66 K20 4.85 4.9 5.04 5.71 5.44 4.93 P205 0.05 0.05 0.05 0.05 0.05 0.05 PPM Th 24.0 24.2 28.6 25.9 24.5 25.3 La 19.0 19.9 26.3 20.2 19.0 19.9 Nb 39.6 42.8 34.9 59.6 55.8 57.1 Rb 259 260 256 284 276 267 Sr 4.37 3.94 15 7 8 6 Zr 76.7 80.0 92 80 73 81 Ce 38.8 39.4 65.9 55.9 55.7 54.2 Ta 3.19 3.40 2.51 3.52 3.66 3.45 Sm 4.33 4.67 4.68 4.83 4.77 4.49 Eu 0.13 0.16 0.16 0.33 0.23 0.24 Tb 0.85 0.94 0.78 0.93 0.94 0.92 Yb 3.36 3.50 3.23 3.63 3.73 3.71 Lu 0.51 0.54 0.45 0.60 0.59 0.60 Hf 3.46 3.63 3.62 3.59 3.54 3.57 Ba 15.2 12.4 45.4 42.0 31 34.4 Er 3.34 3.69 3.04 3.51 3.42 3.49 V 4.65 14.5 5.35 11.2 Cr 3.66 12.5 6.91 12.3 U 5.91 6.20 5.33 10.9 13.1 9.64 Pb 37.3 58.7 91.2 53.9 Nd 16.5 17.2 19.8 17.2 17.0 16.9 Pr 4.93 5.27 6.43 5.74 5.61 5.48 Dy 5.28 5.67 4.68 5.5 5.69 5.48 Ho 1.16 1.25 1 1.24 1.26 1.23 Gd 4.5 4.94 4.48 4.86 4.77 4.72 Y 37.1 40.0 32.6 38.1 37.1 37.4 106 APPENDIX 1. Continued. Sample ID 86RM6-17 86RM27-6 86RM27-7 86RM27-8 86RM-27-9 86RM27-14 Group HSR-2 HSR-2 HSR-2 HSR-2 HSR-2 HSR-2 wt%: Si02 74.3 77.3 77.2 78.4 77.2 76.2 T102 0.21 0.19 0.21 0.19 0.18 0.17 A1203 16.2 13.5 13.7 12.8 13.0 13.9 FeO 1.3 0.56 0.63 0.44 0.86 1.03 M90 0.15 0.93 0.63 0.62 0.66 0.51 MnO 0.04 0.02 0.02 0.02 0.04 0.05 080 0.14 0.14 0.12 0.11 0.13 0.15 Na20 2.52 2.84 2.82 2.64 2.77 2.94 K20 4.96 4.42 4.56 4.7 5.01 4.93 P205 0.05 0.05 0.05 0.05 0.05 0.05 PPmi Th 38.5 36.5 34.7 33.5 34.8 32.5 La 39.9 44.5 51.1 46.3 49.2 35.9 Nb 26.0 24.7 25.2 23.2 27.6 26.8 Rb 130 160 162 160 154 155 Sr 22.0 23.6 24.3 23.8 23.3 20.8 Zr 147 135 160 153 133 113 Ce 70.5 78.0 82.0 77.0 80.7 68.4 Ta 2.19 1.77 1.74 1.71 1.87 1.9 Sm 5.43 5.83 5.92 5.48 6.12 4.74 Eu 0.29 0.27 0.34 0.31 0.33 0.26 Tb 0.72 0.76 0.74 0.71 0.82 0.69 Yb 2.5 2.59 2.38 2.26 2.6 2.32 Lu 0.40 0.40 0.36 0.35 0.4 0.36 Hf 5.14 4.62 4.98 4.98 4.49 4.03 Ba 69.4 51.9 66.2 72.0 60.3 55.9 Er 2.37 2.51 2.27 2.18 2.65 2.26 V Cr U 2.8 2.28 2.21 2.28 3.09 3.48 Pb Nd 28.4 30.1 32.4 30.3 33.0 25.1 Pr 8.61 9.39 10.33 9.5 10.36 7.78 Dy 3.86 4.14 3.93 3.68 4.22 3.67 Ho 0.84 0.91 0.84 0.79 0.96 0.82 Gd 4.97 5.48 5.51 5.14 5.68 4.68 Y 22.7 26.5 23.1 22.8 27.3 23.4 107 APPENDIX 1. Continued. Sample ID 86RM27-36 99TM-14 99TM-16 020725RM-23 02RM19-2 02RM19-3 Group HSR-2 HSR-2 HSR-2 HSR-2 HSR-2 HSR-2 wt%: Si02 80.2 75.6 77.2 76.8 76.5 75.5 Ti02 0.15 0.16 0.19 0.18 0.14 0.16 A1203 10.9 14.7 12.5 12.8 13.9 14.9 FeO 0.72 0.82 0.9 0.81 0.75 0.84 M90 0.26 0.05 0.07 0.06 0.02 0 MnO 0.04 0.05 0.04 0.04 0.05 0.05 CaO 0.14 0.12 0.51 0.48 0.40 0.41 Na20 2.25 2.85 3.55 3.21 2.90 2.52 K20 5.21 5.55 4.93 5.51 5.27 5.51 P205 0.05 0.05 0.05 0.05 0.05 0.05 PPmI Th 28.3 33.0 34.2 32.9 37.7 41.1 La 37.1 29.0 53.6 51.3 37.6 43.3 Nb 19.9 29.8 22.1 30.6 42.4 41.1 Rb 124 170 124 125 171 173 Sr 22.4 12.4 50 42 17 18 Zr 122 107 162 157 113 132 Ce 67.9 60.2 119 112 95.5 104 Ta 1.4 2.4 1.28 1.59 2.68 2.88 Sm 4.15 4.53 5.11 5.03 6.00 7.34 Eu 0.32 0.12 0.47 0.6 0.31 0.34 Tb 0.54 0.70 0.60 0.61 0.84 1.00 Yb 1.96 2.58 1.96 2.00 2.81 3.06 Lu 0.3 0.39 0.26 0.33 0.43 0.46 Hf 3.98 4.22 4.55 4.54 4.5 5.06 Ba 78.5 36.4 186 187 74.7 74.2 Er 1.78 2.46 1.8 1.81 2.5 2.92 V 4.52 19.1 7.69 6.75 Cr 2.74 13.1 8.73 7.84 U 3.14 4.57 2.68 4.9 8.62 8.62 Pb 38.5 45.1 88.3 98.0 Nd 23.7 21.9 30.8 30.2 29.4 35.7 Pr 7.43 6.8 10.7 10.1 9.58 11.6 Dy 2.88 4.05 3.1 3 4.5 4.96 l-lo 0.65 0.87 0.57 0.65 0.93 1.09 Gd 4.09 4.58 4.41 4.46 5.38 6.58 Y 17.9 25.8 18.8 17.6 27.2 30.3 108 APPENDIX 1. Continued. Sample ID 02RM19-7 02RM19—11 86RM24-1 020725-RM5 020725RM-19F Group HSR-2 HSR-2 HSR-3 L8 L8 wt%: Si02 74.4 77.4 76.3 72.2 71 .2 Ti02 0.21 0.17 0.24 0.37 0.38 Al203 16.3 12.5 13.0 14.7 15.8 Fe0 1.07 0.82 1 1.66 1.94 M90 0.09 0.04 0.09 0.3 1.17 MnO 0.05 0.04 0.05 0.06 0.04 CaO 0.42 0.45 0.17 0.98 0.97 N320 2.63 2.6 2.74 3.55 3.34 K20 4.67 5.9 6.33 5.93 4.84 P205 0.05 0.05 0.05 0.07 0.09 PPM: Th 45.4 38.4 36.3 51.3 35.2 La 48.6 51.0 67.1 130.1 82.9 Nb 42.3 34.8 21.4 21.9 32.5 Rb 123 179 145 112 93 Sr 27 24 42.2 131 219 Zr 159 136 213 359 301 Ce 109 1 17 110 259 176 Ta 2.58 1.89 1.5 1.22 2.09 Sm 6.38 6.1 5.33 7.5 7.3 Eu 0.45 0.47 0.51 1.07 1.21 Tb 0.82 0.77 0.59 0.74 0.95 Yb 2.77 2.4 2.06 2.01 3.3 Lu 0.41 0.36 0.32 0.26 0.47 Hf 5.57 4.44 5.92 8.62 7.36 Ba 101 101 173 497 792 Er 2.29 2.19 1.88 1.44 2.91 V 10.1 9.28 19.6 32.5 Cr 9.35 8.37 6.74 9.03 U 7.58 7.56 2.24 3.61 5.24 Pb 78.2 80.0 63.9 129 Nd 34.8 34.0 35.8 65.1 50.1 Pr 11.4 11.6 11.5 21.9 16.4 Dy 4.21 3.82 3.01 2.78 4.65 Ho 0.83 0.8 0.66 0.55 1.02 Gd 5.71 5.41 5.01 6.79 7.19 Y 24.2 22.4 19.0 15.3 33.2 109 APPENDIX 2 110 92 o: ooe moo For E Noe our ex A: o2 Noe o3 NE 9: one nor am Roe oooe Roe .28 moo. moo ouo 38 mm on ”E mo, 92 mos Noe oo— o.o, 3.: mos 00. 8.5 8.4 R5 owe to R5 to 8.4 082 ooo.o ooo.o ooo.o Bro memo omuo mono ooto coo ooo.o ooo.o ooo.o ooo.o ooo.o ooo.o ooo.o Fooo 092 Fooo ooo.o 5o ooo.o ooo.o ooo.o ooo.o Koo 05“. o? For o? o? om. so. 98 o? 83 Eoo ooo.o ooo.o Boo ooo.o ooo.o ooo.o «So No: 98 one woo 3o 98 3o 9% QB ~06 as; oom oom oow oou o9 83 on 3 Ase 5.. 9 85% mm: mm: mm... mm... mm: mm: mm... mm... .506 outpémpémm naétnmplmmm Etwhnwtmmm outwémwémm valwémwtwmm mlébmwtomm Natwbmptmmm watwtnmptwmm n: mc_._mm cmmomo... .msflm 03.00603 new 059:8 Co EmEocooom 20.5.0 mom: can 8.6.2 .N 5029?? 111 4.2 98 :9 So 2: of mos A: 9. our N: 9: N: x: 38 9; ode .m .5 one so so :8 28 on «9 no as ”E o: mos 5.2 o: 5.9 4.2 so. we 09. o; o? 8.4 24 3.4 one 2.4 8.4 082 33 2.3 83 83 23 83 ooo.o ooo.o coo ooo.o ooo.o ooo.o ooo.o ooo.o ooo.o ooo.o ooo.o cos. ooo.o ooo.o woo woo «So on; 83 23 con. 2: 5.8 8.8 2: 2: 2: 5.9 3: 8a: ooo.o ooo.o ooo.o ooo.o ooo.o ooo.o Bo So 8E woo see 98 4.8 So So 98 8.8 ~08 so; one 8» Ba oou one 2: oo 3 as .E s 85% mm: mm: mm: mm: mm: mm: mm: mm: “59.6 on... 73.0mm ha- wt>._ 8-737354 9.4-3942: #9743725... 8-737354 «9743735.. wm-w.>._ D. macaw! :mmomoh 60:52.50 .N X52mmm< 113 3. .8 3. 30 3. 3. 3o .3 am ... 8. 8. .N. 3. 8. v.8 8. .m 8. m: .8 8. B. 8. 8. 30 mm a ”E .... 3. 3. 3. 3. 3. 3. e... 09. 83 t... 83 m3 8.... «.8 :3 m3 082 $3 83 83 83 83 83 $3 83 08 33 83 .83 83 8.3 3.3 83 888.8 092 23 88.8 8.8 83 £3 83 83 o. 3 0m". 3. 3. 3. 3. 3. 3. 3. 0% 83 83 .3 .3 .3 .3 .3 N3 .3 8: «.8 3o .8 38 3o 38 m8 3.. 8% $5 com 08 08 on. 8. on . . Ea e... 2 85...... mm: mm... mm: mm: mm... mm... mm: mm: .596 5907(9on 0&6 75.0mm mane..>._ X..n.uo7v>>._ main Twmm D. macaw camomo. ._._ .Umacacoo .N X.Ozmam< 118 0.3 E: FmF 0mm 03 am «.8 ado FNx. mom F.0m 5 new mom m5 03 wow mm ”Egg and F.FF mdF 63 9: wdF 09. 36 mod and mF.v mod «Fe ONmz vomd m Fwd Chad mmwd mm F .o m Fwd 000 08.0 N 3.0 08.0 ooo.o F Fod Bod 092 08.0 FBd omFd wood 03.0 806 Own. «.3 m.mF v.9 m.mF m.mF QoF mON_< cod cod ood cod mod mod NOE. Qmm Ymo 9mm 58 who ado N05 $5 F 000 F com com omF Ea E: 2 8:8»... mm: mm... mm: mm: mm: mm: macaw E: 753m wmcggmm S.¢.m-_._. .Umaczcou .N X52mmm< 120 Q F8 F8 Q8 Q8 Q9. Q3. Q8 Q8 FE Q8 Q8 Q8 Q8 F8 Q2 Q8 8.88 8 88 8 F8 28 F8 88 88 88 F8 8 8 “E 38 NE 28 88 88 :8 .88 88 09. 88 88 28 88 8E 88. 8QF oQF 082 F888 8 F88 F888 888 F88 F88 83 83 08 888 888 888 888 288 888 888.8 888 08.2 8.3 888 83 83 8on 88.8 888 83 08“. 88F F8F 88F 88F 88F 88F 88F F8F 8% 288 888 8 58 388 888 858 888 888 88 Q8 Q8 v.8 v.8 Q8 Q8 Q8 Q8 «08 8.; 8m 88 8F 2: 8 F 8m 88 ea. 5.285% mm: mm: mm: mm: mm: mm: 88: mm: “.888 on. WERE. mn— TEmO... Va Vino... an. Vino». Nn. .1500... Fa YEmU... mm Wan”: ma m-_mm0._. O. 32ch «5:. .Umacacoo .N x_ozwan_< 12] 88F F8~ Q8 Q8 Q8 Q3 Q3 Q8 88. SF BF 8: E 2: Q8 Q8 QF8 8 8v~ F8~ 8~ 8~ 88 88 -8 ~8~ 88 8 ”E 88.8 8.8 88.8 8.8 88 88 88 8E 0~x 8E 88 8s 8Q~ 8s 88 88 88 0~mz F88 F888 :88 ~88 F88 88.8 888 8~8 080 888 888 888 88.8 88.8 88.8 8888 F888 08.2 83 8~8 88.8 88.8 ~8~8 23 88.8 8 F~.o 08”. Z: «.8 2: v.8 v.8F Q8 8.2 Q8F 80~_< F 68 888 88.8 88.8 88.8 $88 «88 858 ~08 F8 Q8 Q8 Q88 «.8 Q8 Q8 F88 ~08 Q8; 88 88 8~ 8~ 88F 8F 8 F 2.3 at 2 8.5% mm: mm: mm: mm... mm: mm: mm: .8... 8080 on. N._mm0._. kn. Wino... an. N-.mm0._. an. Wino... tn. Winch on Winch Na «-5005. F. N.._mm0._. O. coxcmo m>_._. .Umaczcoo .N X52maa< 122 Q F8 Q8 Q8 F88 Q3 Q8 8. F8 3.8 gm 8: ~FF ~FF ~FF 8F QF8 Q8 F8 8 38 38 88 8 F8 88 88 8~ 8~ 8 8 ”E 88 8.8 88 8.8 F88 88 8F~ F88 0~x ~Q8 88 88 ~F.~ F88 88 F88 88 082 388 ~38 838 838 388 888 88 888 080 888 888 888 888 .88 ~88 888 ~88 08.2 3~8 88 38 83 8v~8 F8~8 8~8 88 08“. ~8F 88F ~8F 88F ~8F ~8F 2: 2: 80~_< $88 888 888 :88 $88 888 888 888 ~08 888 Q8 Q8 Q8 Q8 Q8 Q8 Q8 ~08 8s, 88 88 8~ 8~ 88F 8F 8 F is 2F. 2 8:38 8.8... «8: m8: 8: 88: mm: 88: 8: 8088 an. ”300... kn. Nufimmoh. on. ”800... mn— N._.mm0._. fin. N300... ma ~33: Na. Noan0h F. ”809—. D. Ic.0|>.||||cmo m>_._. .Uozczcoo .N X_Dzmmn_< 123 F8 F8 Q8 Q8 Q8 F8 Q8 Q 8 am 3. F8 Q8 Q8 Q8 Q: 8.88 88 F8 8 88 ~ F8 8~ 8 ~38 8~ 8~ -8 8 8 ”E 8.8 8.8 -8 8Q~ 88 88 28 38 0~x 88 F 8 :8 8.8 88 88 38 F88 082 838 88 888 888 888 838 888 F 38 080 888.8 8888 888.8 888 888 888 888 888 08.2 3-8 8~8 8~8 -~8 8~8 8~8 8~8 8-8 08”. 88F F8F 88F 38F ~8F 88F 88F 88F 80~_< 888 888 888 858 888 888 888 8 :8 ~08. Q8 Q8 Q8 Q8 Q8 Q8 Q8 Q8 ~08 ”8.; 88F 8F 8 F 88F 8F 8 F 2.3 2.. 2 85% .8: 88... 8: 88: 8: mm: 8: mm: 8088 v1 m._._.. .umzcucoo .N Xazmmnz 124 00¢ 00v F.N0 0.3.. 0.0V 0.9. 0.00 0.00 am va v.00 FOF 00F 0.05 0.00 0.00 8.00 00 00m F F0 000 0Fm 00m 00m KN F00 00 ”Eng 05.0 V0.5 0N.» 00.x. «0.0 F00 00.0 00.0 09. «F.» 00.0 00.0 ~00 FN.0 00.0 00.» F00 Ommz 5 F00 3N0 00N.0 N800 F0v.0 30.0 «00.0 000.0 000 0000 F000 800.0 000.0 0000 F000 0000 «00.0 00.2 0v~0 03.0 F000 03.0 08.0 0 FN.0 F30 03.0 00". 0.2 «.mF N.0F 00F 0.0F ~.0F N.0F 0.0F MON? F000 500.0 000.0 30.0 v3.0 0000 000.0 000.0 NOrr 0.80 080 v.80 0.00 0.00 0.80 0.00 v.00 NOE 8x85 00 F 00 F 00 F 000 000 00m 00m 050 E: 2 888.88 mm: mm: mm: mm: mm... mm: mm... «0: 000mm 3 ~._._. 00:00:00 .N X_Ozmn_a< 125 0.v~ 0.0~ 0.~0 0.00 0.~.. :0 0.00 0.00 00 0~F 0~F 0. .0 0+0 0.~0 F.0~ 0.2 0.00 .0 000 30 80 3~ 0~0 000 000 00~ 00 3 ”F. 8.0 00.0 00.0 ~00 ~00 00.0 00.0 2.0 09. ~§ 03 ~00. 0: F: 2.0 00.0 ~0~ o~mz 023 000.0 09.0 23 F000 00~.o 000.0 000.0 08 30.0 80.0 F80 80.0 80.0 08.0 80.0 80.0 00.2 $3 -~.o 53 :3 00~.o 0 F~.o $03 08.0 00“. ~0F 0.2 0.2 2: F0. 2: 0.0. 0.2 «.82 8.0 6.0 5.0 5.0 8.0 ~00 3.0 ~00 8: ~00 «.00 ~00 0.00 ~00 0.00 0.00 F00 ~90 $3 00 F 00~ 8~ 00F 00F 00 F Ea 2.. 285.0,. mm: mm... 00... mm: 00: 00: mm: 00: 029.0 Na Tom~0h Fn. TOmNo0._. on. 0-2000... 0m n-2m00h Qu— nimDOP an. 0.2000... Na. n.2m00P Fm 0-2900... 0. l>|co 500 '02... 60:00:00 .N X52maa< 126 0.00 ~0~ 0.00 0.00 0.00 0.~v ~00 0.00 00 03 00F 00F 02 0: FFF 0.00 ~00 .0 F00 0E 00. v2. F0~ 000 ~0~ 0.00 00 00 ”E ~00 000 000 00.0 0.0 -0 9.0 00.0 09. E ~00 ~00. :0 00.0 00.0 F~.~ F00 0~02 00.0 000.0 0000 F000 20.0 000.0 03.0 000.0 000 000.0 800 000.0 000.0 000.0 000.0 000.0 0.00.0 00.2 00~.0 00~.0 00~.0 00~.0 00~.0 0F~0 0-.0 Fv~.0 00“. 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0o~_< 20.0 03.0 08.0 0~0.0 -0.0 0~0.0 30.0 0000 St ~00 000 0.00 0.00 000 0.00 0.00 0.00 ~90 30.; 00F 00F 00 F 00F 00F 00 F Ea a: 2 820% mm: mm... mm... mm: mm: mm... mm: 00: 00000 '0. 0.50000. 00. 0.2000... Nn. 92000.0 00. 0.2000... tn. F-<0N00. an. F.. _.._.| 00200000 .N 50200.0( 128 0.00 00 ~00 0.00 0.. F0 ~00 0. Fe 0.0~ 00 00F 0.00 0. F0 ~00 F0F 0F F 0.00 00F .0 0~ F~0 000 000 F00 00~ 00~ 000 00 00 ”E 00.0 ~00 00.0 F00 0~.0 00.0 00.0 ~F.0 09. 00.0 0.0 0.0 .00 00.0 ~0.0 9.0 00.0 0~02 000.0 03.0 03.0 000.0 0~0.0 F000 :00 000.0 900 00~.0 ~00.0 000.0 000.0 00.0 000.0 000.0 ~F0.0 o0: 00~.0 ~0F.0 ~F~.0 0-.0 0F~.0 00~.0 0-.0 00~.0 000 ~.0F 0.0F F.0F 0.0F 0.0F ~.0F 0.0F 0.0F 0o~_< 000.0 000.0 000.0 00.0 000.0 10.0 000.0 ~F0.0 ~90 F00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 ~90 0.00; F 00F 00F 00 F F 000. 00F E0 2.. 2 802...... m0: 00: 00: m0: m0: 00: 00: 00: .5000 Fn. «-2000H 0000 0-<0N0... nu 0-<0N0... mn. 0.<0N0... E0 0.<0N0... E... 0.0me... 0000 0.00N0... 00. 0.2000... 0. 000.000. .02.... 00000000 .N 50200.0( 129 0.3 ~00 0.00 0.3 N. FN 00~ v.00 0.~0 gm 0. F0 SF 0.00 :3 F0F «NF «NF 0: .0 000 03 F0~ 00~ F00 00~ 00~ 00~ 00 a ”E F00 $0 8.0 00.0 00.0 $0 00.0 00.0 09. 00.0 00.0 20 00.0 0:. Es 00s 03 802 0~0.0 0~0.0 000.0 000.0 «3 000.0 80.0 000.0 000 000.0 08.0 08.0 000.0 80.0 000.0 08.0 30.0 00.2 00~.0 05.0 00~.0 00~.0 05.0 :3 003 08.0 00“. F.0F F.0F 0.0F 0.0F 0.8 0.2 0.0F 0.2 082 30.0 F80 0~0.0 30.0 000.0 000.0 03.0 08.0 8: 30 0.0 $0 :0 0.00 0.00 «.00 0.00 ~90 ”$0; 00F 00F 00 F 00~ 00F 00F 00 Ea 502.23% mm... m0: m0: mm: mm: mm: mm: mm: $9.0 Va. ”dwNOF an. némmoh Na mémNOP Pm néMNOh an. «inch Va ~-EmmOP an. Nimmoh. Na Nimmoh. D. :0. 2:00 m>.|_._. .Uwzczcoo .N X_Dzman_< 130 0.00 F.0~ 0.F0 0.00 0.00 0.00 «.00 0.~0 5. SF «NF 00F RF 8F ~0F 00F SF .0 ~0~ 00~ 00~ 00~ 00~ 00~ ~0~ 0F0 00 3 “E 00.0 00.0 00.0 00.0 00.0 3.0 ~00 00~ 09. 00.F 0: F00. 0: BF 0: BF F0 802 F000 :00 80.0 00.0 000.0 003 F000 000.0 08 F50 000.0 80.0 08.0 000.0 08.0 08.0 000.0 00: ~03 00~.0 0:0 83 00~.0 «0~.0 00~.0 00~.0 00“. 0.0F 0.0F 0.2 0.2 F.0F 0.0F 0.0F 0.0F 802 80.0 30.0 000.0 08.0 0~0.0 000.0 FF0.0 08.0 3: 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 ~90 ”$0; 80 00~ 00~ 00F 00F 00 F 00~ 2.3 g. 2 85% mm: mm: m0: m0: mm: mm: mm: mm: “596 ha. szmOP on. 729.6% mm ..-meOF Va .3me0... mm 0-20.0.0... an. ..-2mm0._. E 72000». ma nuowNOh O. coxcmo m>_._. .Umzczcoo .N X_ozmn_a< 131 0.~0 :0 0.3 0. FF. 0. F0 F00 0.~0 0.8 am 0.00 0.00 F00 0.00 0.00 0.00 2: 0F F .0 ~00 000 F~0 «a 00~ F00 ~0~ 0mm 00 a ”E 0~.0 Fog 00.0 00.0 3.0 00.0 00.0 00.0 09. 00.0 00.0 0: 0: 0: 00.0 ~0~ 00.2 802 000.0 000.0 000.0 83 :00 000.0 F000 0~0.0 000 000.0 08.0 80.0 000.0 03.0 F 80 F000 0000 00.2 «F~0 00~.0 $3 «0~.0 ~0F.0 00~.0 00~.0 03.0 000 ~.0F 0.0F 0.2 0.2 ~.0F ~.0F v.0F 0.0F 00~: 05.0 20.0 03.0 30.0 0~0.0 000.0 05.0 30.0 «o: 0.00 ~00 0.00 0.00 0.00 500 0.00 0.00 8.0 .22; 00~ 00~ 00F 00F 00 F 000 000 .22 2.. 2 8.22.. 00: m0: m0: m0: m0: m0: m0: m0: .580 on. w-_mm0._. ma s-_mm0h Va ”amok ma h-_mm0._. Na Wino... ..n. ~-_mm0._. ma szmOP on. szmOP Q. co>c00 m>_._. .banEOU .N X_02mn_n_< 132 m.mm 300 0. we mdh who Now who am 99‘ 0.3. Q: 0.: NNF QNF m.m—. hw Yov can mfim 0N0 Non m. ..m v. 5 mm ”Eng mm.» on.» n F. .w Sta 0‘0 Now 5s 3d ONX and «08 mmé 3.0 09m 5.»... mad or .v ONmZ de wad sud mud hwd and 36 mmd Om0 mood ooo.o wood wood hood ooo.o 000.0 000.0 002 9.0 the 3.0 NNd 3.0 9.6 and 9 .0 Own— odp 02 NB 0.9. 0.2 0d? 0.9 m6? MON? ooo.o @nod Fmod omod mmod NNod ooo.o mood NOE. v.00 mhm 0.00 ado v.00 Q8 0.8 v.50 N05 HAXLE, F com omN CON on F 00 P on w A83 E: 2 8:206 mm: mm: mm... mm... mm: mm... mm... mm... 95.5 Fa- Tmumh< hm- PomN.m.r< on- TmNAhh< on... wumwéhk Va. TmN.m._.< ma- YQNAEK Na. meéhk Pm- Tawébx D— 0xcm... mEoEE< 6025.30 .N x_ozman_< 133 ode Now 0. E. ...mo ooh 5.00 m. ..o v.3 am odw NmF oi. F.0F «.9 0.9. «.9 03 5 com v.3 0.3 Nam 98 0.8 Now 0% mm ”Eon ovd mud mod 9 .o and on.» 2% 3.0 09. ovh «v.0 ooé 2.0 cod 5.0 San mod ONmz Foo woo and mod omo moo omo «no 000 Boo ooo.o ooo.o ooo.o ooo.o mood ooo.o ooo.o Gas. to Ed ofio So to 3d Fmo 9o Om“. For Now v.3 o9. 0.9 0.3 For 0.9 mON_< «rod ooo.o ooo.o ooo.o m Foo Nooo m Foo :oo «0:. 0.00 «do oso moo v.8 v.00 «do «.8 N05 ”023 ooF om F omw ooN on? 80 on E3 E: 2 8030.0 mm: mm... mm... mm: mm: mm... mm: mm... 96.0 8.04.0.0: «2.04.0.0: E04005 02.70.05 02. 70.0.2 «2.70.05 8. 70.0: «2.70.05. 0. 0xcmh mEoEE< 60:50:00 .N X_Dzwan_< 134 Q: _..oo o. o o. 5 «.vo woo «.oo odo om o.«« tom odd «do od« o. 5 v.«« n.o« .m Eon mi 09 RF 3: wow o«w v.3 am ”End oo« «...o o«.o omd ood o«.o «vd vod o«x nod rod 3.0 «od odd to mod ond o«mz nvd ovd :Vd omd omd So «no omd 000 0 rod ooo.o vood mood ooo.o ooo.o ooo.o mood 092 o«.o o_.d o«.o n F .o o F .o o«.o o«.o \. _. .o 00“. odr 93 dd? dd? 9.9 dd? «.3 Fd_. mO«_< ooo.o ooo.o vood o«od n«od ooo.o ooo.o ooo.o «O; odo moo «do «do woo vdo «.oo moo «05 ”$5, F oov oov oom oom oo« oo« 9.: 3.3 E: 2 80320 mm: mm: mm: mm: mm: mm: mm: mm: 30.0 70-00-05 22.04.00? 02.0-5.0: 0.0.0-805 5.04.0.0: 02.0-5.0? 02.0-5.0: 3.0-50.2 0. 0x5... mEoEE< .Umaczcoo .N Xazwma< 135 0.00 0.00 F00 :0 F00 0.00 0.00 0.00 am F.0~ 0.: 0.2 0.0F EF 0.0F v.0F 0.0F .0 00F 02 0: 0.00 0.00 0.00 0.00 3L 00 ”End F0.» 000 ~F.0 F0.“ 00.0 ~F.0 00.0 00.0 09. 00.0 2.0 00.0 0~.0 00.0 00.0 00... 00... 0002 00.0 00.0 00.0 00.0 00.0 00.0 3.0 00F 08 80.0 000.0 80.0 08.0 000.0 000.0 000.0 000.0 00.2 ~F.0 00.0 ~F.0 02. 02. 8.0 2.0 2.0 000 F.0F ~.0F 0.0F 0.0F F.0F F.0F 0.0F F.0~ 0002 :00 30.0 08.0 05.0 000.0 03.0 000.0 08.0 00: 0.00 0.00 0.00 F00 0.00 0.00 0.00 0.00 00.0 ”-0.; 80 000 80 000 80 00F 00F 00 2.2 .2. 2 8.2% m0: m0: m0: m0... m0: m0: m0: m0: 08.0 mad-00.0.2 oa-o-nm-Ek E-o-nn-obk 8.0-5.0.; 0.0-0-00-0.2 {-04-0.0.5 na-o-nnooK «ad-5-0.? n: chmo. mEoEEd 60:50:00 .« X52mam< 136 0.0F 8.0 0.00 0.00 0. F» 0.00 0. F0 0.00 am 0.00 0.00 0.00 0.00 0.00 0.00 ~0F 0.0F .0 0.00 0.00 0.00 0.00 «.00 0.5 0.00 FNF 00 ”End Re «0.0 00.0 00.0 F00 0.0 00.0 02. 09. 00.0 F00 ~F.0 F~0 00.0 3.0 00.0 2.0 0002 0F.F FNF 0F.F 00.0 00.0 00.F 2: F00 000 08.0 F000 08.0 08.0 000.0 80.0 000.0 80.0 00.2 ~F.0 02. 00.0 2.0 02. 02. 00.0 00.0 00.. F.0~ 0.00 0.00 0.0F 0.0F 0.0F 0.0F 0.0F 0002 F80 000.0 000.0 000.0 08.0 80.0 08.0 08.0 00: 0.00 0.00 F.00 0.00 0.00 0.00 F00 F00 00.0 “2...; 000 000 000 00F 00F 00 F 000 Ea 2.. 2 8.22.. m0... m0: m0... m0... m0: m0... m0: m0: 08.0 Roi-0.0.; duo-30.2 duo-5.0.2 «no-50.2 ma-o-nn-obx «0.0-no.8." Edi-0.0.: own-0-8.0.2 n: 0x00? mEoEE< 00:50:00 .« X_Ozmn_n_< 137 You «...-o o.vo ...oV 0d? oNN oéN oéo am 93 N. ..v «do 0.2 now now u. PN mom ..w «3 who won odm now r. 5 V. ..m NNV 0m ”Eon ohd mod and mod 55¢ «F6 o F .0 mod ON! and mod mod ow.» mog- de no.0 ..Nd mez o«.o o«.o o«.o moo woo Fm... :.w or... 000 ooo.o mood ooo.o. Noo.o ooo.o ooo.o ooo.o hood 092 to _.—..o owo owo 3o Fwd o«.o o«.o 00n— o.o_. adv «.9. 0.9. oo« oo« o.o—. vow mON_< o..o.o FNoo n ..o.o ovoo ooo.o ooo.o ooo.o ooo.o NO:- moo moo oNo 0.8 «do odo moo ado N05 He\o§> oo w on _. om _. oo« on r own :03 E: 2 8:806 mm: mm... mm... mm... mm: mm... mm... mm... 0:90 nQ-Ynn-oh< «Q-V-nm-mkx‘ Fa+hmfih< vfl-n.o«-o...< nm-m-oN-mbx «mo-o«.o»? «mun-o«.obx om-o-hn-Ek O. 0x00... m_00EE< 00300000 .« X_Ozmn_m< 138 0.00 0.00 0.00 0.00 F. E 0.00 0. F0 0.00 gm 0.0F 0.00 0.00 000 0.3- 0.00 0.00 F00 .0 0.0 0.00 000 02 02 3.. 000 000 00 ”0.00 00.0 03 00.0 00.0 00.0 2.0 00.0 00.0 09. 00.0 000 00.0 00.0 00.0 00.0 00.0 00.0 0002 00F 02 3.0 00.0 00.0 00.0 00.0 0~.0 000 80.0 000.0 30.0 000.0 000.0 000.0 F80 08.0 00.2 :00 0:. 8.0 00.0 9.0 2.0 2.0 00.0 00”. 0.00 F.0~ ~.0F ~.0F ~.0F F.0F F.0F 0.0F 0002 06.0 000.0 000.0 000.0 08.0 30.0 80.0 80.0 00: 0.00 0.00 0.00 F00 F00 F00 F00 0.00 000 ”2...; 00 F 80 000 000 000 000 00F Ea e... 2 8.2% m0... m0: m0: m0... m0: m0... m0: m0: 08.0 «0. 9-5-0.: E 94.0.0.2 0.0-0150.2 003-50.; 0.0-$50.5 8.0.5.0.; 0.04-50.2 wt%-5.0.2 0. 0x00... 0.00.002 00:00:00 .« X.Dzmn_n_< 139 0.0F 0.00 0.. F0 0.0.0 000 0. F0 0. E 0.00 am 0. F0 F.0~ 0.00 0.00 0. F0 F.0~ 0.0F ~.0F .0 0.00 02 FNF 00F 0F F 0.00 0.: 0.00 00 ”0.00 00.0 00.0 8.0 3.0 00.0 00.0 00.0 3.0 09. 00.0 F00 00.0 00.0 00.0 00.0 00.0 00.0 0002 0.0. F 000.0 000.0 0000 F000 .000 000.0 000.0 000 000.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0 00.2 02.0 000.0 0 F .0 00 F .0 00 02.0 00F.0 02.0 00“. 0.0F 0.0F 0.0F F.0F ~.0F 0.0F 0.0F 0.0F 0002 0~0.0 000.0 000.0 000.0 000.0 0000 F000 0000 00: 0.00 0.00 000 0.00 0.00 0.00 0.00 0.00 00.0 ”-0.; 000 00F- 000 000 000 000 00F 02 2.3 2.. 2 8.2% m0: m0... m0: m0: m0... m0... m0: m0... 08.0 2000-00.0.2 00. 9.5.0.2 on. 9.5.0.; E 3.5.0.? on. 9-5-obt- on. 9-5-0.: 3. 9-5-Bk on. 9.3.0.; 0. 0x00... mEoEE< 00300000 .N 02029.5( 140 0.00 0.: 0.00 0.F0 0.00 0.F0 0E 0.F0 00 0.0F 0.00 0.0F F.0F 0.0F 0.00 0.0F 0.0F .0 0.00 0.00 0.00 0.00 0.00 F00 0.00 0.0 00 ”0.00 F00 F00 00.0 00.0 00.0 00.0 00.0 00.0 002 00.0 00.0 00.0 F00 00.0 00.0 F00 00.0 0002 000.0 000.0 000.0 000.0 000.0 0F0.0 20.0 000.0 000 F000 0000 000.0 000.0 0000 F F00 0000 F000 00.2 08.0 03.0 09.0 000 02.0 000 02.0 000 000 0.2 F.0F F.0F 0.0F 0.0F 0.0F 0.0F 0.0F 000_< 000.0 08.0 000.0 000.0 00.0 000.0 00.0 000.0 00.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 00.0 00.; 000 000 000 000 00F 00F 00 F .22 2.. 2 8.22.. m0: m0... m0... m0: m0: m0: m0: 002 000.0 00.0-0.0.2 00.0-0.0.2 8.0.0.0.: 00.0-0.050 00.0-90.2 00.0-0.002 «0.0-0.0.? {-0.0-0.: 0. 0200.0 0_coEE< 00200000 .N X52wmm< 141 Eon vdo ndo o.oo «do no» o.on F. K vdo or. do? 0.: Q? nor do? o.or 0.00 0.5 o.F« .0 Foo o.oo o.oo o. 00 0: o.ov adv o.«n Fdn 0m ”E00 «0d ovd ovd «vd «od and «vd mod and 09. So ond «No {.0 mud ood ood ond nod o«mz mnmd «ood ooo.o o«od Budd oo«.o «omd ovmd ooo.o 000 oFod ooo.o ooo.o ooo.o ooo.o rood ooo.o ooo.o ooo.o 092 word did a F«.o movd 09d «0. F .o «opd oowd om F d 00“. 0.00 0.00 «.00 dd? 0.3 0.0.. odr 0.00 «.00 m0«_< Sod ooo.o ooo.o mood o Pod «ood ««od «ood wood «0:. Noo «.oo o.oo o.oo o.oo «do o.oo o.oo «do «Qm ”0005 com oo« oo« oor oop do 0 oov oow :50 c... 2 85.0.0 mm: mm: mm: mm: mm: mm... mm: mm: mm: 020.0 5-0.00.0: 00-0-00.02 00-0-00.0: 3-0.00.0: 00-0-00.0.2 00-0-00.0.2 F000-00.0.2 05.0.0.0? 0.0-0.0.05 o. 00.00... 0_000.E< 00300000 .N X_Dzmn_n_< 142 00F 00F 0FF 0FF 00F 00F 00F 00F 00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 .0 F00 0. F0 000 0.00 0.00 0.00 0.00 0.00 00 00 ”E 0. FF 0.2 0.2 0.0F 0.0F 0.0F 0.2 20 002 00.0 00.0 00.0 00.0 00.0 00.0 F2. 00.0 0002 02.0 80 03.0 80 09.0 80 03.0 000.0 000 0 00.0 0 00.0 0 0 0 0 00.2 000.0 000.0 000.0 000.0 02.0 000.0 000.0 000.0 000 F.0F 0.0F 0.0F 0.0F 0.0F 0.0F 0.0F F.0F 0002 00.0 00.0 0 F00 F00 F00 0 0 00: 0.00 0.00 0.00 00.00 0.00 0.00 0.00 0.00 00.0 .2. E 000 000 000 000 00F 00F 00 F 0.2 2.. 2 02.2.... F002 F002 F00: F00: F00: F002 F002 F002 00000 00.0-33.00 0.0-0.32.50 00.0-32.50 00.0-33.00 {-0.0-5:00 00.0-02.50 «0.0-32.50 3.0-12.50 0. 000.2 .0_0_0m 00300000 .« X_Dzmn_n_< 143 0.0F 00F «FF FFF 00F 00F 0.00 00F e... 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 .0 F.0F F.0~. 0.00 F: 0.00 0.00 0.00 0.00 00 2.000 00.F 00.: 0.2 00.2 00.2 29 3F 2: 002 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0002 00.0 0:0 000 03.0 02.0 02.0 03.0 03.0 000 0 0 0 0 08.0 0 0 0 00 00.2 02.0 02.0 000.0 000.0 000.0 000.0 000.0 000.0 00“. 0. F0 0.0F 0.0F 0.0F 0.0F 0.0F 0.0F 0.0F 0002 0 00.0 00.0 0 0 00.0 00.0 0 00: 0.00 0.00 0.00 F00 0.00 0.00 0.00 F.00 00.0 H0\o «)3 F 000 00F 00F 00 F 000 000 222 2.. 2 02.22.. F200: F00: F00: F00: F002 F002 F002 . F00: 00000 E 0.0.0 :20 00.0-0.0020 3.0-0.0221 00.0-00:20 «...-0.0.3.20 {-0.0-32¢ own-0.32.50 0.0-0.32.50 D. 000.2 .2501 00300000 .« x_ozman.< 144 00F 00F 00F 00F SF 0.00 0.0F F.FF 00. 0.00 0.00 0.00 0.00 00.00 0. F0 0. F0 0.00 .0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0F 00 00 ”E 00.0 0.0F 0.0F 0.0F 0.0F 0. FF 3: 0....F 002 00.0 0F... 00.0 0F... 00.0 00.0 00.0 0F.0 0002 000 000 000.0 000.0 00.0 000 00.0 00.0 000 0 0 0 00.0 0 0 0 0 00.2 000.0 000.0 83 000.0 0000 S F0 020 03.0 00.. 0.0F F.0F 0.0F 0.0F 0.0F 0.0F 0. F0 0. F0 000.< 000.0 0 0 000.0 0 00.0 0 00.0 000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 F00 00.0 ”2.. S F 000 00F 00F 00 F 02 00 “.2... 2.. 2 02.2.... F00: F002 F00: F00... F00... F002 F00: F00: 0000.0 {oo«-32¢ 00-0-0.9.2”. .00-No.92”. 00.0-9005.”. «0.4-0.00.2”. «0.4-0.002”. on 0.0-022m. «0. 0.0.9.5.. 0. wwms. .253. 00300000 .« X.ozman.< 145 00F 00F F0F 00F 0: 0: ..3 0: 00. 0.00 0.00 0.0F 0.0F 0.00 0.00 0.00 F00 .0 0.00 0.: 0.00 0.00 0.00 0.00 0.00 0.00 00 2000 0.: 0.: 0.: 0.: 0.: 0.: 0.0F 0.9 002 R0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0002 000.0 020 2:0 30 0:0 0:0 02.0 02.0 000 00.0 0 00.0 0 0 0 00.0 06.0 00.2 0000 20 000.0 000.0 000.0 000.0 000.0 000.0 000. 0.0F 0.0F 0.0F 0.0F 0.0F 0.0F 0.0F 0.0F 0002. 00.0 00.0 00.0 00.0 000.0 0 0 000.0 00... 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 00.0 .00.; 00F 00F 00 F 000 00F 00F 00 H2.... 0... 2 02.20... 0-0.0: 0.0.0.. 0.0.02 0-0.02 F00: F00: F00: F002 0500.0 3.4-«0E 00.40.30 «0.02-«.03. {+0.90 00.0-8.0020 va-o.o«-o«s.m 00.0-3.0320 «0.0-8.8.20 0. 000.2 0.0.01 00300000 .« x.02wn.n.< 146 N: 02 0.5 v0. 5 5 v2 09 5. 0.3 0.05 30 09:0 0.8 0..& 0.~0 ..2 .0 Eu vaw new «.00 ..E 0.: «.00 0.00 mm ”Ema 0.: 0d 09 0.: 0.: 0.: 0.: 0.: 09. 0.0 00.0 00~ 6.0 2.0 0~.0 :0 0.0 Oumz 80... 8:. 8m... 83 83 83 83 83 08 o o o o o o o Sod 00.2 83. 8:. 83 8°... 83 08... 82. ooo.o Om“. 0.2 0.2 2: 0.0. 0.8 0.2 0.2 0.9 8% 3... o 8... 6.0 o So... So... Sod «o: 0.00 0.00 0.00 0.00 0.00 0.00 0.00 000 «90 xx. .3 on P Sc Sc 80 80 o0~ 8w ”.2... 2.. 2 85% 3.0: «-00: 3.0: 3.0: «-00: 3.0: «.00... 3.0... $9.0 ~a¢~épm Fa¢~épm oE.v.~.0 3... ma+~é E oaéfié E 5.34.0 E Ewe”. ma+~éwm o. 000.2 .2501 6253.50 .N X_Dzman_< 147 0: 0.: 0: on? 02 02 F: 0: am 0.00 6.8 3.. 90.. 0.3 0.0 «.8 0.00 .0 S: 03.0.. 0.2 0.05 wt «.00 0% E 00 a ”E ...: 0.: w: 0.: 0.: v: 0.: 0.: 09. 00.0 00.0 00.0 2.0 2.0 ~00 00.0 00.0 0002 20.0 No 83 83 83 SN... 23 83 000 o o o 5.0 o o o 00.2 83 3.0 8o... 22. 80.0 83 83 83 00“. F.0F 0.3 0.8 0.2 0.8 o0. 0.2 0.2 0002 o Sod So 3.0 o 8... o 30 «0: ~00 «.00 0.00 «.00 0.00 «.00 0.00 «.00 «0.0 ..x. .3 one 8w 8.0 80 00~ SN 02 02 “.2... 2.. s 8%.»... 3.0: 3.0: 0-00... «-00... «-00... «.00... 0-00... «-00... 02000 05.0.0.3”. 3.0.0.05... 00.00.05. 3.0.0.03. 00.00.05". 00.00.05 3.90.05 00.0.NOE D. 000.2 ..0_c_0m dos—5:00 .N x_ozman_< 148 E 00. 0: 00. 3: 000 00. 00. 00 0.00 0.00 0.00 0.00 0.00 00. 0.0.. 0.00 .0 0.00 0.00 0.00 0.00 0.0 000 0.0.. 0.00 00 00 ”E 0.: 0.: ...: 0.: ...: 0:. 0.2 0.: 000. 00.0 00.0 .00 00.0 00.0 .00 00.0 00.0 0002 000.0 20.0 000.0 000.0 000.0 000.0 0.0.0 0.0.0 000 0 0 0 0 0.0.0 000.0 0 800 005. 000.0 000.0 000.0 2.0.0 000.0 000.0 0. .0 000.0 00.. 0.0. 0.0. 0.0. 0.2 0.0. .00 0.0. 0.0. 000_< 08.0 0 0 0 06.0 0 05.0 06.0 00.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 00.0 ”.0 E 000 80 00. 8. 00 F 000 80 H...... 5.. 28.5% 0-00: 0-00: 0-00.. 0-00.. 0.00.. 0-00: 0-00.. 0.00: 00000 00.0 0-0.0 E 00.0 0.0.0:. 3.0 0-0.0 E 00.0 70.0 E 00.00.0000. {.0 0-0.0 E 0 5.0.0.0 E 30.0.0001 Q 0002 .0_c_00_ 60:50:00 .0 X_Dzmn.n_< 149 0.. .0. 0.00 00. .0. 00. 0.00 00. 00 ... .0 0.00 0.00 0.00 0.: 0.0. 0.00 0.00 .0 00. 00. 000 .3 000 000 000 0.00 00 ”£00 0... 0... 0... 0... 0... 0... 00.0 0... 00.. 00.0 00.0 00.0 00.0 00.0 00.0 00... 00.0 0002 00.0 00.0 00.0 00.0 0.0.0 0000 00.0 0.0.0 000 0 .00 0 0 0 0.0.0 0.0.0 00.... 00.0 ...0 00.0 0. ..0 0. ..0 00.0 000.0 000.0 00.. 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. ..0. 0.0. 0002 .00 0 0 0.0.0 0 0.0.0 0 0 00.. 0.00 0.00 00.00 0.00 0.00 0.00 0.00 0.00 00.0 a... E 000 000 000 00. 00. 00 . 000 H...... 2.. 28.3.... 0.00.. 0-00.. 0.00.. 0-00.. 0.00.. 0-00.. 0-00: 0.00.. .5000 .00-00-0.0.0 800-00.". 00.00-00 .m {.00-0.0.x mum-00-0.0.0 0.0-00-0.0.0. {.00-0.0 E 0.0-0.0.0.1 a. $02 .9500. 60:50:00 .N 5029.30. 150 00. 00. 00. 00. 00. 00. 0.00 0.. 00 0.00 ..00 0.00 0.00 0.00 0.00 0.0 0.00 .0 0.3 0.00 ..00 0.00 0.00 0.. 0.. ... 00 ”0.00 0... 0... 0... 0... 0... 0... 0... 0... 00V. .00 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0002 0.0 00.0 .00 .00 00.0 00.0 00.0 00.0 000 0 .00 .00 0 .00 0 00.2 00.0 ..0 00.0 00.0 00.0 00.0 00.0 00.0 00.. 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. 0002 00.0 00.0 .00 00.0 .00 0 0 .00 00.. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 00.0 ”.0 ..s 000 000 000 000 000 000 000 000 0..... .... 2 .25.. 0-00.. 0-00.. 0.00: 0.00: 0-00.. 0-00: 0-00.. 0-00.. .5000 00.0-0.0.0 «0.0.00.0 .00-0.00.0 00000-0.0 0.3.00.0 0.0-00.0.0.0 0000-000 0000.000 9 000.2 .2000 60300000 .0 x_ozw00< 151 0.00 0.00 00. 0.. .... .0. 00. ..0 .00 000 0.00 .00 .00 0.00 0.00 .0 .03 000.. 0.0. 0. .0 0... .0. 0.0. 00 ”0.00 0.0. 00.0 0... 0... ...... 0... 0... 0... 00v. 0.} 00... 00.0 00.0 00.0 3.0 00.0 00.0 0002 00.0 00.0 00.0 0.0 00.0 00.0 0.0 00.0 000 0 0 0 0 0 .00 00s. ...0 0.0 .00 00.0 0.0 .00 ..0 00.0 00.. 0.0. 0.0. ..0. ..0. 0.0. 0.0. 0.0. 0.0. 0002 00.0 00.0 0 .00 .00 .00 .00 0 00.. ..00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 00.0 .... E 00 . 0... 00. 00. 000 000 000 ...... s... 2 8.2.... 0.00.. 0-00: 0-00: 0-00.. 0-00.. 0-00.. 0-00.. 0-00: .5000 00.70-00.20 .00-.0020 30.00.00 .0 0.000.000 0040.00 .0 0000.000 00.0-0.0.0 0000.000 0. .00'05. 0.0.00 00300000 .0 X_n_zm00< 152 5.0 2.0 03 $0 8.0 8.». Ed 0.8 gm 2: 02 02 a? 2: a? 05 2: .0 0.8 30 EN 20 QR o.ou £0 88 mm a ”E a? a? :1 8.. «3 N? a? 8.0 09. 03. 0~.0 an.» 00.0 NS 8.0 8.0 a? 0%2 03‘ 3.0 $0 :e 8.0 8.0 Be :00 08 o o 8.0 5.0 o 8.0 o o 092 No No :3 8.0 8.0 2.0 o«.o 8.0 08 9% RN ~0~ RN o.«« cam 00~ <2 00~: 8.0 8.0 8.0 3.0 60 No.0 3.0 o No: 50 memo 3.0 2.0 0.8 v8 0.~0 30 ~00 0;, SN 80 02 02 on r 8. 2: $3 2.. 2 85% ~00: ~00: «00: «00... «.00: ~00: «-00: «.00: “5000 00.F0~.0~_20 m0é0~0~20 00.780320 00.009020 «0.730020 0.73.820 0.0440030 00.70.0520 0. 000.2 52:00 00:50:00 .N X_Dzm00< 153 000 0: 000 v: 0: 0: 0: $2 00 0.00 0.00 0.00 :0 0.00 0.2. 0.: 0.00 .0 0: 000 02 02 000 a: 000 0.000 00 ”E00 0.: 0.: 0.: 0.: 0.: 0.: 0.: 0.: 09. 2.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0002 000.0 000.0 000.0 000.0 00~.0 20.0 000.0 000.0 000 0 0 000.0 0 0 000.0 0 5.0 0 00.2 000.0 000.0 . 000.0 0000 t 0.0 000.0 000.0 00.0 00". 0.0. 0.2 0.00 0.00 0.00 0.0. 0.00 0.00 0002 30.0 000.0 05.0 0 03.0 000.0 0 :00 ~0F F00 0.00 0.00 0.00 0.00 0.00 0.00 30 000 ”0. E 000 000 000 000 02 000 00 F U2.0 2.: 2 .2200 0.00: 0.00: 0-00: 0-00: 0-00: 0-00: 0-00: 0-00... .5000 00 «0.0520 00 «000220 00 «000220 00 «000.20 v0 «000220 00 «000220 «0 «000.20 .0 $000020 0. 000.2 .2500 60:50:00 .N X52m00< 154 J 0.00 0.00 0.00 0.00 «.3 0.3 «.0. 0.3 00 0.0. .00 0.00 0.00 0.0. 0.00 0.00 «+0 .0 0.0. 0.3 0.: 03 .00 30 000 ... 00 ”E00 0... .... .... .... .... .... .... N... 09. 00.0 00.0 0.0 00.0 00.0 £0 E... E0 0002 000.0 00.0 .000 000.0 000.0 000.0 000.0 00.0 000 000.0 0 0 30.0 000.0 .000 .000 000.0 00.2 .000 R00 000 000.0 000.0 .20 .000 00.0 00“. 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. 0002 000.0 .000 0 000.0 000.0 000.0 0 30.0 00.. 0.00 0.00 0.00 .00 0.00 .00 0.00 0.00 ~00 0x. E 000 00~ 000 00. 00. 00 . 00¢ u2.3 2.. 2 85% 0-00: 0-00: 0.00: 0-00: 0.00: 0.00: 0.00: «.00: 00000 0.0 «70.0.5.0 00 90.0.20 00 «70.0.20 00 3.0.0.20 n0 90.0.20 «0 9.0.0.2} .0 90.0.5.0?0 «0.0.20 0. 000.2 02500 60300000 .N X_Dzw00< 155 .00. v.00 n. .0 3. 0.00 0.00 0.00 0.0 00 ..00 0.00 0.00 0.0.» 0.00 0.00 0.00 v.3 0m v.00 0.00 0.00 0.2. 0.0.0 ~.~0. 0.00. N00 00 ”E00 0... .... 0... .... o... N... .... N... 09.. 00.0 00.0 00.0 00.0 00.0 .00 «0.0 2.0 Oumz 00~.0 mvmd 0~.0 «mud .mwd 00~.0 0&0 00 . .o 000 000.0 30.0 wood «00.0 .000 o 0 wood 092 30.0 000.0 «00.0 V0.0 0N. .0 000.0 03.0 35.0 000 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. 0.0. 00~? o o 00.0 .00 o o 0 mod N00. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 N05 ”00 .3 000 com 00~ com 00. oo. 00 . ”:03 E: 2 8030.0 «-001 «-00: «.00... N00... «.001 «-00: «-00... «-00... 03000 00 ..0.0.20 .0 30.0.20 00 20.0.5050 20.0.20 30 30.0.2050 20.0.20 «0 3.0.0.255 2.0.0.205. 9022 002051 00:00:00 .N 50200.0( 156 «.8 0.3 0.8 «,3 0.0V 0.3, 0.2 «.2 gm 0: .NN cum 98 E08 03 08 a: .0 «no 80. 88 8... .08 .2 3. m. .m 8 a ”E to. No. 3. No. no. no. N... .... 09. a? mm... 3.. «N... ...v a... 8.” B... 00.2 93 33 «~00 .3 m .3 8m... 2.... a3 08 mood 390 0 ~80 o o 80.0 ~80 022 8.0 83 0~.0 m. ..o v. ..o 3.0 08.0 83. 0...“. 3. ..m. ..m. 3. ..m. 2.. ..m. 3. 8% 03.0 .3 o .2. mad Rod 8.0 3.0 «0: $0 30 $0 0.3 $0 ..8 0.8 30 ~05 ”x. .3 com com on. 8. on . 8.. 8. He... at 2 8:3. 3.0... 9mm: 9mm: 9mm... 9mm: 0mm... 3.01 3.0... “59.0 on. «-mvée‘m ma ”Ni—2m #1 ”~32". nu ”~32". «Q ”~ng .& n-~v.m§¢ o.n. ...m..m.2¢ on. ...m..o.2¢ D. 0005. .9501 $35.80 .N 5023.? 157 888 888 8.8 8.8 «8.. ...8 88v 888 81 8. v8 8. 8. 8. 8. 8.. 8. .m 8.8 8.... 88.. 88. ..N. .8. 88 8.. mm 8 ”E 88. N8. ..8. 88 ..8. 88. .8. 88. 08 8... 8+ 8+ 88... 88 83. 8.8 88V 082 888 ....8 .88 ~88 .88 8.8 888 83.8 080 .88 .88 8 888 8 .88 .88 .88 082 88.8 838 3.8 8.8 818 2.8 8.8 8.8 08“. 88. 88. 88. .8. .8. 88. 88. 88. 808. 888 8 .88 ...88 888 .88 .88 v.88 «0: 888 888 ..8 888 888 888 888 88 ~08 8.. E 88 88 8. 88. 8 . 88 88 85. e. 2 8.2% 9mm: 8-me 3.0: 8.881 $me 8-me 8&0: 8-me .5080 on. .-NYwfim mn— .-~v.o§m VA 78185... 9.... .-NYQEK Na ..NYmSE .n. .-NYQZK on. Wag R. 99.33 D. 9'05. .950”; 0035.80 .N X_Dzmn_n_< 158 ..8 8.8 88.. 888 8.8 88.. 08.. .88 88 88. 8. 8.. .8. .8. .8. .8. 8. .m 88. 88. 88. 888. .8. 88. 88.. .88. 88 8 NE 88. 8.8. 88. N8. 88. 88. N8. 88. 08. 88 a... 88V 88 88 88 88... «88 082 $88 888 8888 888 888 888 8.88 .888 080 888 .88 8 8 8 8 888 8.88 08.2 8.8 8.8 88.8 88.8 8.8 3.8 ....8 838 08”. 88. .8. 88. 88. 88. 88. .8. N8. 8082 888 888 888 888 8 8 .88 «88 «Q. 888 888 .88 888 .88 8.8 888 88 ~08 H0\0 a; 88 88 88 88 8v 8.. 88 88 Ea s. 2 8:28.. 8-88: 8-88: 8-88: 8-88: 8-88: 8-88: 8-88: 8-88: 8080 v.1 .-Nv.021 0.1 .-«121 «.1 .-NYosE ..1 54.321 0.1 79.921 01 73.021 on. 7810.21 E ..lem D. 000.2 .2501 00:03:00 .N X_Dzwn_n_< 159 888 .88 88. 8.8. 8.8 .8. 8.8 8.8 88 88. 08. 8.. 8.. 8.. ... v8. 88. .8 8.. 88 88¢ 88.. 88¢ 88.. 888 .8. 88 8 ”E 88. 88. 8.8. .8. 88. ..8. ..8. 88. 09. 888 88... 8.... 8.8 ...... 888 8.8 83 082 8888 «88 8.88 88 8888 3.8 8.88 8.88 08 8 8.88 8 8 8 ..88 8 8.88 08.2 8.8 .8.8 888.8 888 8.8 888 8.8 3.8 08“. 8.8. 8.8. ..8. 88. ..8. ..8. 8.8. 88. 8082 8 8.88 8.8.8 8 888 888 8 888 80: 8.88 8.88 8.88 888 8.88 .88 8.88 888 80.8 8. E 88 88 88 8. 88. 8 . 88. 222 2.. 2 8.22. 8-88: 8-88: 8-88: 8-88: 8-88: 8-80: 8-88: 8-88: .5080 5 .73-021 01 ...3.021 01 2-3.021 01 ..-3-021 01 ..--3.021 «1 ..-3-021 .1 .73-021 23.-3.021 0_ 0002 32:01 .0030ch0 .N x_ozm11< 160 8.88 8.8 .. .8 8.8 ..88 888 8.88 888 80 88 8 .8 .88 828 8.. 8. 8.. .8. .8 8888. 8888 . .88 .88. 8. 28¢ 88.. .8.. 88 8 ”F. ..8. ..8. ... .. ...8. v8. 88. 8.8. v8. 09. .8... 8.... 83 888 888 .8... .88 88... 082 888 888 8.88 8888 8888 .88 .888 888 080 .88 8 8 8 888.8 .88 8 888.8 08.2 8.8 8.8 8.8 88 888 88.8 88.8 888 08“. ..8. ..8. v8. ..8. ..8. 8.8. 8.8. 8.8. 8082 888 8 8 888.8 8.8.8 388 8 8 80_. 8.88 ..88 8.8 8.88 8.88 8.88 8.88 8.88 808 a... E 88. 88. 88 . 888 88.. 8.. 888 222 2.. 2 8.22.. 82 88 88 8.. 8-082 8-88... 8-88: 8.082 88080 va-m-nTnN—Zm wt%-8.0321 NQ-n-n ..nmfim .Qé-n ..nNEm. Soon-N321 0.1 3-98-@2131 ..-NYQEK ma ..-NY-wzm n: 0802 .0_0_01 00303000 .N X_szn.n_< 161 8.88 8.88 8.8 88 8.8 8.8 .3. .8 .88 .8.. 88.. 88 ”E00 88.8 88. 88.8 09. 83. 8.8 8.8 082 8.88 8.88 888 080 8 .88 8 082 8.8 8.8 8.8 08. 8.8. 8.8. 8.8. 8082 .888 8 888 80: 38 8.8 8.88 808 28.; 888 888 88 .22 2.. 2 8.22.. 82 88 88 88088 .8.-8.0.8.821 0188.8«21 01.0.0.8N21 0. $02 .2201 00300000 .N X_Ozm11< 162 APPENDIX 3 163 164 V8. 888 888 888 .8. 888 888 888 ..8. 2822.88. 88 8.88 8. .8 8.8 8.8 8.88 8.8 8.8 8.8 8.88 .8228. 8._ 8... 8.88 8.8 8.8 8.8. 8.8 8.8 8.8 8... .8258. ... 8.88 88. 88. 8.. 8.8 8.. ....8 8.. 88. 88 88.. 8.8. 8.8 8... 8.} ..8. 8.8. 8.8. .88 82 8.8. .8.. v... 8... 88.8 8.88 8.8. 8... .88 8._ 3. 8.8. 8.8. 8.8. .88 8... .... ..8. ...8. ...- “0.00 88... 8: 88.8 88... 8.8 8.8 88.8 88.8 8.8 09. 88.8 8.8 8.... ...8 8.8 .8.. 8... 88.. .88 082 8888 888 8.88 888 8888 888 .888 8888 888 080 0.2 88.8 .88 8.8.8 888.8 888 .888 888.8 .888 888 08.2 8888 .....8 .8.8 8.88 .888 .888 .88 ....8 888.8 088 8.8. 8.8. 8.8. 8.8. 8.8. ..8. 8.8. 8.8. 8.8. 8082 8888 8.8.8 8.8 .888 888.8 888 88.8 88.8 8888 No: 8.8. v8. 8... 8.8. 8.8. 8.8. 8.8. ..8. 8.8. 808 .88 88: 88: 88: 882 88: 88: 88... 88: 88... 8588... .89. 70-007000 70-08.10 70.00000 70700-000 70-07010 70-80000 0-.-<0000 .-.-(0000 7700.000 02020:. 05.00 000000.. 00.8: 0.0 08000 .800 000 020.0 0582.05 :00. 05 .0802 .5501 .0... .EoEmg 020.20 .800 05 .0 055809000 01 000 _0._ 0.. 00. 0.0 008: 082.200 .0802 .2501 000 88.00.. 0_00EE