ABSTRACT MAGNETIC MINERALS AND PROPERTIES OF THE MELROSE STOCK BY Dewey Dennis Sanderson An integrated petrologic-magnetic rock property study was conducted on the Melrose Stock, an early Creta- ceous Basin and Range intrusive in eastern Nevada. The stock was delineated into a monzonite and a quartz mon- zonite pluton with evidence that the former was emplaced first. The magnetite content and magnetic properties of remanence and susceptibility are about twice as great in the monzonite as in the quartz monzonite. Both rock units contain a major soft remanent component, placing the reliability of the remanent measurements and their paleomagnetic significance in doubt, even with alternat- ing field demagnetization. However, the remanence indi- cates that the horst of which the intrusive is a part has undergone structural rotation of approximately 153 The difference between the paleo-pole positions of the two plutons suggests an interval of time between their emplacement and crystallization. Only one-third of the NRM remained after magnetic cleaning to 100 oersteds. Storage tests revealed that in 100 days up to 50 percent of the NRM was a soft VRM. This, however, was easily removed with approximately Dewey Dennis Sanderson 20 oersteds of demagnetization. The opaque mineral assemblage of the two rock types is essentially identical, approximately 90 percent mag- netite and 10 percent ilmenite. The compositions are near the stoichiometric values. Within a one mile wide zone adjacent to the intrusive margin, the titanium is tied up in sphene rather than ilmenite because of the in- fluence of the carbonate host rock assimilated during crys- tallization. The ilmenite which formed deep within the magma chamber exhibits abundant exsolution of hematite. A technique (the association coefficient) developed to quantitatively determine the microscopic distribution of the magnetite with relation to the constituent min- erals in the rock suggests that the magnetite formed throughout much of the crystallization by three processes: direct precipitation, oxidation and alteration. Oxida- tion refers to magnetite formed as one ferromagnesian mineral is converted to another under oxidizing condi- tions during the normal sequence of crystallization. Mag- netite by alteration (deuteric and hydrothermal) is form- ed by late stage fluids reacting with the ferromagnesian minerals as in chloritization and serpentinization. The relative amounts of magnetite formed by direct precipitation, oxidation, and alteration in this order are 65, 25, and 10 percent in the quartz monzonite and 45, 30, and 25 percent in the monzonite. The association study is used as a Dewey Dennis Sanderson qualitative index to changes in the oxidation state of the crystallizing magma with respect to rock type, time and space. The magnetite grain size distribution of the quartz monzonite is coarser than the monzonite. Furthermore, the magnetite grains associated with the ferromagnesian minerals are notably coarser than the grains associated with the nonferromagnesian mineral fraction. The factors of time and availibility of iron by oxidation have influ- enced the variation of the magnetite grain size. The difference in the grain size distribution of the two rock types is expressed as a more stable remanence of the mon- zonite than the quartz monzonite. The magnetic susceptibility of the two rock types was effectively delineated by in situ and core specimen measurements. The in situ coil yields better quality data as well as the capacity to better resolve magnetic units than laboratory measurements on core specimens. To obtain a representative value of magnetic suscepti- bility, a site density of approximately one site per square mile was found to be sufficient for an intrusive the size of the Melrose Stock (12 square miles). The number of sites is more critical than the number of measurements per site when establishing representative susceptibilities. MAGNETIC MINERALS AND PROPERTIES OF THE MELROSE STOCK BY Dewey Dennis Sanderson A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Geology 1972 ACKNOWLEDGMENTS I wish to express my sincerest gratitude to Dr. William J. Hinze, my thesis advisor, for his guidance, advice, suggestions and time during the course of this study. In addition, his contribution to my learning while at East Lansing is most gratefully acknowledged. Dr. Thomas A. Vogel gave generously of his time throughout the course of the study and reviewing the manuscript; his help is greatly appreciated. Special thanks are extended to Mr. John D. Corbett for making this study possible while he was employed by the Anaconda Company. The author wishes to thank Drs. Harold Stonehouse, James Trow and Hugh Bennett for their comment and eval- uation of the manuscript. My wife, Sharon, deserves special recognition and thanks for her patience and understanding and for her time spent in typing the drafts of this thesis. ii Chapter TABLE OF CONTENTS I. INTRODUCTION........... 1.1 1. 2 1. 3 OijCtives o o e e e 0 Geology of the Area . . . Rock Sampling . . . . . . II. PETROLOGIC PROPERTIES . . . . . . . . 2.1 2.2 2.3 2.5 2.6 PetrOIOQYo e e e o o o e e 0 2.1.1 Introduction. . . . . . 2.1. 2 Modal Analysis 2.1. 3 Microscopic Description . . Apparent Grain Size Distribution Of Magnetite o o e e e e e e Opaque Petrology . . . . . . . 2.3.1 Introduction. . . 2.3.2 Magnetite - Ilmenite 2.3.3 sphene. C O O 0 2.3.4 Hematite . . . . Association Coefficient . . . . . 2.4.1 Introduction. . . . . . 2 4 2 Procedure. . . . . . . 2.4.3 Association Model . . . . 2 4 4 Geologic Influences on the Magnetite Association Co- foiCient o e e e e 0 2.4.5 Results and Interpretation . The Interstitial-Inclusion Index . . Magnetite and the Ferromagnesian Min- erals . . . . . . . . . . iii Page H mun- 11 11 11 12 17 22 32 32 33 37 42 SO 50 51 53 58 62 75 87 Chapter III. IV. V. 3.1 3.2 3.3 3.4 3.5 4.1 4.2 4.5 4.6 MAGNETIC SUSCEPTIBILITY . . . . . . . I ntrOdUCtion e e e e e e e 0 3.1.1 Purpose . . . . . . . 3.1.2 Sources of Variation. . . 3.1.3 Measurement Variations . . Discussion of Results . . . . . 3.2.1 Normal versus Logrithmic DiSttibUtiOno e e 0 3.2.2 Intersite Variations. . 3.2.3 In Situ . . . . . . 3.2.4 Cores. . . . . . . Sampling. . . . . . . . . . Susceptibility and the Geometry of the Intrusive e o e e e e e Susceptibility and Magnetite . . . REMANENT MAGNETIZATION . . . . . . . Introduction . . . . . . . NR“ 119.111.1280 0 o e o e e 0 4.2.1 NRM Data Evaluation . . 4.2.2 NRM Directions. . . . Alternating Field Demagnetization . 4.3.1 Preliminary Demagnetiza- tion 0 O O O O 0 4.3.2 Demagnetization at Optimim Level 3 e e e e e o 0 4.3.3 Remanent Intensities. . . Storage Tests . . . . . . . . 4.4.1 Storage Procedure. . . . 4.4.2 Results . . . . . . . 4.4.3 Demagnetization of Store Samples . . . . . . Q-Ra 1:108 e e e e e e e e e o Paleomagnetism. . . . . . . . SUMMARY 0 O O O O O O O O O O 0 iv Page 95 95 95 96 97 99 99 100 104 107 114 119 125 132 132 133 133 141 145 145 158 163 167 167 168 175 185 191 197 Chapter Page REFERENCES CITED . . . . . . . . . . . . 204 APPENDIX A . . . . . e . . . . . . . . 208 General Remarks on Magnetic Properties . . . 208 APPENDIX B . . . . . . . . . . . . . . 215 Opaque Mineral Species and Their Relative Abundance . . . . . . . . . . . . 215 APPENDIX C . . . . . . . . . . . . . . 217 NRM Results by Site . . . . . . . . . 217 Table 2-1. 2-2. 2-3. 2-4. 2-5. 3-1. 3-2 . 4-1. 4-5. LIST OF TABLES Modal analysis minimums, maximums, and means of the Melrose Stock . . . . . . . . . Percentage of opaque grains greater than 50 microns in diameter a o e e e e e e 0 Probability model for association coeffi- dents O O O O O O O O O O O O O O Source of excess magnetite associated with the ferromagnesian minerals in percent of total magnetite content . . . . . . . . Percent magnetite to total magnetite content formed by alteration, oxidation reaction, and direct precipitation . . . . . . . . . Magnetic susceptibility sampling matrix in percent deviation from the grand mean of the quartz monzonite pluton . . . . . . . . Magnetic susceptibility sampling matrix in percent deviation from the grand mean of the mnzoni ta pluton o e e e e e o e e e Comparison of NRM data of Sites 1-17 with reduction to field and laboratory orienta- tions . . . . . . . . . . . . . . . NRM results of 29 sites grouped in various combinations. . . . . . . . . . . . 50 oersted demagnetization results of 29 sites grouped in various combinations . . . 100 oersted demagnetization results of 28 sites grouped in various combinations . . . Remanent magnetization intensities ( x10-4emU/CC) e o e e e e e e e e 0 vi Page 21 30 55 69 74 115 116 140 142 160 161 164 Table Page 4-6. Remanent intensity ratios . . . . . . . 166 4-7. Results of Storage test. . . . . . . . 174 4-8. Paleomagnetic poles determined from the 1970 NRM results. . . . . . . . . . . . 194 A-le Opaque Data. 0 e e e e o o e e e o 215 A-ze NRM Data. 0 e e e e o e o e e e e 217 vii Figure 1-1. 2-2 0 2-3 . 2-4a. 2-4b. 2-5a. Z-Sb. LIST OF FIGURES Index map of east-central Nevada showing location of the Melrose Stock. Modified after GOG. snow 0 O O O O O O O O O Generalized geologic map and site loca- tions of the Melrose Stock (after 6.6. snow) 0 O O O O O C O O O O O 0 Partial modal analyses of rocks of the Melrose Stock. Rock composition fields of the U.S. Geological Survey are shown . . Modal analyses of rocks from the Melrose S tOCk O O O O O O O O O O O O 0 Rock types of the Melrose Stock based upon modal analyses. . . . . . . . . Apparent grain size distribution within the monzonite pluton. l) opaques asso- ciated with the quartz-feldspar fraction. 2) opaques associated with the ferromag- nesian fraction. . . . . . . . . . Apparent opaque grain size distribution within the monzonite pluton. 1) opaques associated with the quartz-feldspar frac- tion. 2) opaques associated with the ferro- magnesian fraction . . . . . . . . . Apparent opaque grain size distribution within the quartz monzonite pluton. 1) opaques associated with the quartz-feld- spar fraction. 2) opaques associated with the ferromagnesian fraction . . . . . . Apparent opaque grain size distribution within the quartz monzonite pluton. l) opaques associated with the quartz-feld- spar fraction. 2) opaques associated with the ferromagnesian fraction . . . . . . viii Page 13 15 16 25 26 27 28 :Efiiggtire 2-6a ' b. 2-10. 2‘11. 2‘12. 2‘13. 2‘14. 2~15. Electron microprobe microphotograph of an ilmenite grain with hematite exsolu- tion and a sphene reaction rim, a, sample current; b, titanium fluorescence . . . . Electron microprobe microphotograph of an ilmenite grain with hematite exsolu- tion and a sphene reaction rim. c, cal- cium fluorescence; d, iron fluorescence . . Distribution Of sphene. e e e e e e 0 Distribution of hematite exsolution within 1 1men1 te O O O O O O O O O O O 0 Distribution of exsolution bearing ilmenite with respect to site elevation . . . . . Opaque mineral zones within the stock re- flecting the geometry of the magma chamber. Line A-A' is the location of profile of P1911339 2’11 e e e e e e e e e e o Cross-section of the Melrose Stock showb ing the observed opaque mineral zoning. See Figure 2-9 for location of profile . . Schematic diagram of data collection pro- cedure for the association coefficients of W with respect to minerals X,Y, and Z. Graticule of ocular shown on each W grain. Counts are made at the points indicated by arrows. The resulting point counts, asso- ciations, and association coefficients are shown in the accompanying table. . . . . Frequency of occurrence of the association coefficients of the constituent minerals by rock type for magnetite . . . . . . Proposed crystallization sequence of the Melrose Stock and the mode of magnetite formation based on association coefficients and petrographic observations . . . . . Distribution of the interstitial-inclu- sion index for the quartz monzonite and monzonite rock types . . . . . . . . ix Page 35 36 39 44 46 48 49 52 63 73 76 ‘Ewiggtmze: 2-16. JZ-JLT7. :Z-JLIB. 1!-JL53. :!——:zq3. 2-21. 2.2 3. 3‘2 Page Interstitial-inclusion index spatial dis- tribution O O O O O O O O O O O O 7 7 Variation of the interstitial-inclusion in- dex with elevation within the quartz mon- zonite pluton. . . . . . . . . . . 79 Variation of quartz content with eleva- tion within the quartz monzonite pluton . . 81 Variation of the interstitial-inclusion index with elevation within the monzonite pluton . . . . . . . . . . . . . 83 Variation of quartz content with elevation within the monzonite pluton . . . . . . 85 Diagramatic sketch of the high and low interstitial-inclusion index zones with- in the monzonite pluton and the interpret- ed cross section of the pluton . . . . . 86 Relationship of magnetite content to ferro- magnesian content according to rock type. . 88 Variation of the hornblende association coefficient with ferromagnesian content within the quartz monzonite pluton. . . . 90 The relationship of the hornblende asso- ciation coefficient with elevation with- in the quartz monzonite pluton. Excludes data of sites showing appreciable hydro- thermal alteration of the ferromagnesian mineral 8 e e e e e e e e e e e e 9 3- Magnetite content versus percent hornblende to percent ferromagnesian content with respect to rock type . . . . . . . . 92 Distribution of individual in situ magnetic susceptibility measurements in the monzon- ite pluton on linear and logrithmic scales . . . . . . . . . . . . . 101 Distribution of individual in situ mag- netic susceptibility measurements in the quartz monzonite pluton on linear and logrithmic scales . . . . . . . . . 102 .Ftlgjuure 3-3. 3—4. 33“JL() Page Site susceptibility extremes versus site means of in situ measurements. Unity slope line separates maximum and minimum values. Included are the intercept, slope, standard error of the estimate, and cor- relation coefficient for each of the data groups 0 O O O O O O O O O O O O 1 O 3 Site susceptibility extremes versus site means of in situ measurements after re- jection of anomalous data. Unity slope line separates maximum and minimum values. Included are the intercept, slope, stand- ard error of the estimate, and correla- tion coefficient for each of the data group‘ e e e e e e e e e e e e e 1 0 6 Site susceptibility extremes versus site means of core specimen measurements. Unity slope line separates maximum and minimum values. Included are the inter- cept, slope, standard error of the esti- mate, and correlation coefficient for each of the data groups . . . . . . . 108 Site susceptibility extremes versus site means of core specimen measurements after rejection of anomalous data. Unity slope line separates maximum and minimum values. Included are the intercept, slope, standard error of the estimate, and correlation coefficient for each of the data groups . . 109 Magnetic susceptibility distributions of in situ and core specimen measure- ments by rock type . . . . . . . . . 111 Spatial distribution of in situ magnetic susceptibility and aeromagnetic lows . . . 120 In situ magnetic susceptibility versus distance from the south contact of the Melrose Stock. . . . . . . . . . . 121 Variation of in situ magnetic suscepti- bility with elevation within the mon- zonite pluton. . . . . . . . . . . 123 l?jugnuure 3-11e £3-JL:2. EB-JL23, £3-—JL43, 415a. 4" Sb. ““Eic. “ 5d. Variation of in situ magnetic suscepti- bility with elevation within the quartz monzonite pluton . . . . . . . . . Volume of magnetite as a function of ele- vation within the quartz monzonite pluton. Distribution of magnetite content by site with respect to rock type . . . . . . Magnetic susceptibility versus magnetite content. The expected susceptibility- magnetite relationship (Mooney and Bleifuss, 1953) is indicated by the solid line. The best fit line through the data is indicated by the dashed line . . . . . NRM site circles of confidence accord- ing to time of collection. a) 1968 specimens, 18 months of storage; b) 1970 specimens, 1 month of storage . . . NRM of Site 10, directions and intensi- ties. Equal area projection . . . . . NRM of Site 21, directions and inten- sities. Equal area projection . . . . NRM site circles of confidence before (a) and after (b) rejection of anoma- lous da ta e e e e e e e e e e e NRM directions of 29 sites grouped in various combinations. Circles of con- fidence shown for the 1970 monzonite sites and the 1970 quartz monzonite sites. 0 O O O O O I O O O O O A.f. demagnetization results of Site VIN-4 e e e e e e e e e e e e e A.f. demagnetization results of Site VN-IO. e e e e e e e e e e e e A.f. demagnetization results of Site VN-lae e e e e e e e e e e e e A.f. demagnetization results of Site W'lg e e e e e e e e e e e e e Page 124 126 127 128 134 136 137 138 143 147 148 149 150 ‘Ffilggture 4-59. 4-Sie 4—Sg. 4i-—£5]3, 4l—-:3;1, A.f. demagnetization results of Site VN-zo e e e e e e e e e e e e A.f. demagnetization results of Site W-22 e e e e e e e e e e e e A.f. demagnetization results of Site VN-25 e e e e e e e e e e e e A.f. demagnetization results of Site VN'30 e e e e e e e e e e e e A.f. demagnetization results of Site VN-31 e e e e e e e e e e e e Mean site directions for NRM, 50, 100, 150, 200, and 300 oersteds demagneti- zation of Sites 4, 10, and 13, those collected in 1968. Equal area pro- jection . . . . . . . . . . . Mean site directions for NRM, 50, 100, 150, 200, and 300 oersteds demagneti- zation of Sites 19, 20, 22, 25, 30, and 31, those collected in 1970. Equal area projection. . . . . . . . . NRM, 50, and 100 oersted demagneti- zation directions of the monzonite and quartz monzonite collected in 1968 and 1970. Numbers refer to Tables 4-2, 4-3, and 4-4 and are located at the NRM po- sitions. Equal area projection . . . Migration of remanent magnetization upon storage in the laboratory. . . . Intensity of VRM acquired during storage in the laboratory . . . . . . . . Intensity of VRM acquired for two speci- mens from Site 32 during storage in the laboratory . . . . . . . . . . Plot of VRM acquisition with respect to time o e e e e e e e e e e Storage test and demagnetization of BPOCimen VN’3OA-A . o o o o o o o xiii Page 151 152 153 154 155 156 157 162 169 170 171 173 176 Figure Page 4—13b. Storage test and demagnetization of BPBCj-men VN-32A-2 e e e e e e e e e 177 4—13c. Storage test and demagnetization of speCimen W’BZBe e e e e e e e e e 178 4—136. Storage test and demagnetization of speCj-men VN-4lBe e e e e e e e e e 179 4—136. Storage test and demagnetization of smelmen W-41C. O O O O O O O C O 180 4-1315. Storage test and demagnetization of 813001111611 W-41De e e e e e e e e e 181 “139. Storage test and demagnetization of BPBCIMGH VN'47Ae e e e e e e e e e 182 4‘1 311. Storage test and demagnetization of speCimOn VN-47De e e e e e e e e e 183 4~14. Distribution of 0 ratios by rock type . . 187 4‘1 5. Q ratios versus distance from south con- tact in the monzonite pluton . . . . . 190 ‘;“51-€5. Paleomagnetic pole positions of the 1970 NRM along with rotations of the poles corresponding to rotations of the Melrose Stock horst . . . . . . . . . . . 193 A‘Zl. Compositional relationships of the major magnetic minerals . . . . . . . . . 211 xiv CHAPTER I INTRODUCTION #1 Objectives The magnetic properties of rocks are a function of their compositions, origin, and geological history. Con- 8e(II-1.13ntly, magnetic rock properties can be extremely use- ful geological tools. The primary use of rock magnetism studies is in delineating paleomagnetic field directions. A3 a result studies have concentrated on rocks which are known to exhibit relatively stable remanent magnetism. Pat“'iidcular emphasis has been placed upon investigations of Volcanic, basic intrusive, and sedimentary rocks. In contrast, acidic to intermediate composition intrusive igneous rocks have received limited attention despite their wide spread occurrence and geological importance because of the general instability of their remanent magnetism. Pale(magnetic data obtained from granite intrusives have ken reported by Currie and others (1963), Opdyke and “Quaink (1966) Gromme (1967) May (1968) and Hanna 0 n , ( 19 69-1970). A quartz monzonite intrusive, the Melrose Stock, in the Dolly Varden Mountains of eastern Nevada is the 2 subject of this integrated magnetic property—petrologic investigation. The general objective of this study is to determine the mineralogic, petrologic, and geometric con- trols on the magnetic rock properties and their spatial variations in this Basin and Range type pluton. This 9'1LJLJL be investigated through an intensive study of the Opaque mineralogy, remanent magnetism and magnetic sus- cePtibility. The specific objectives of this investiga- tion are: 1) To determine the variations of magnetic suscepti- bility and remanent magnetization between and within the lithologies of the intrusive; 2) To relate the paragenesis, occurrence, distribu- tion, association, composition, and texture of the magnetic minerals to the susceptibility and remanence; 3) To determine the effect of the intrusive's geo- metry on the magnetic susceptibility and reman- ence; 4) To use the magnetic minerals and their proper- ties to decipher the intrusive's cooling and em- placement history: 5) To assess the effects of weathering on the magne- tic properties of surface samples; 6) To use paleomagnetism to detect structural rota- tion in the Melrose Stock. Achievement of these objectives will provide a more 3 complete knowledge regarding the origin, stability, and variations in the magnetic properties of the Melrose Stock in particular and of the Basin and Range type intrusive in general. In addition, the study will demonstrate how mag- netic rock properties can be used to solve geologic prob- lens , will provide information on sampling requirements for a magnetic rock properties study and finally will pro- Vide the magnetic interpreter with magnetic rock property data, their variation and significance for an intrusive 011' this type. Lt -.2 Geolog of the Area The Dolly Varden Mountains have been studied by only a few investigators. A rather comprehensive study of the nonntains was conducted by Geoffrey G. Snow as a Ph.D. d1 Baertation at the University of Utah in 1963. Previously Ire2E><>rted studies by J.M. Hill (1916) on the general geology and ore deposits of the range represents the only other p‘-‘12>.’l.ished work. The Dolly Varden Mountains are mentioned by Zirkel (1876) and Emmons (1877) as part of reconnai- a‘hcm geologic surveys. There has been periodic work done in the area by mining and petroleum companies, but the Ireaults of their projects are not available. The petrology of the Melrose Stock is the result of this investigation and the general geology reported herein is based on the obaervations of Snow. The Dolly Varden Range is a horst in the Basin and 4 Range Province in east-central Nevada (Figure l-l). The nearly north-south trending range rises over 2,000 feet above the surrounding alluvial filled valleys. Even though the area is semi-arid with less than 10 inches of rain a Year, vegetation and wildlife are plentiful. The moun- tains provide a wide spectrum of geologic features to ob- Serve and study. Rock types include a sedimentary sequence, an intrusive complex, both flow and pyroclastic volcanics and locally metamorphosed sediments. Tectonic activity has resulted in a number of structural relationships. The oldest exposed rocks in the range are of Mississip- pian age, but these comprise only 5 percent of the sedimen- tary column. Pennsylvanian rocks are not present. Nearly 4 - 000 feet of Permian limestones, dolomites, sandstones and siltstones are exposed throughout the range and only 100 feet of Triassic limestone attest to a Mesozoic deposi- tional record. A hiatus from Triassic to Oligicene time places Tertiary volcanics unconformably on the sediments and the intrusive. The volcanics are composed of flows, ignimbrites and pyroclastics of varying composition. Mio- Qene or Pliocene age deposits include minor and local vol- canics and fresh water sediments. Youngest of the deposits i"Te the Quaternary lake deposits, presumably related to toruler Lake Bonneville. The Melrose Stock, which occupies the core of the range, has an exposure of approximately 12 square miles. A potassium-argon age determination on biotite from the monzonite places the age of the intrusive 114‘ , 30' Wendover '0 Ic: I°¢ I 2 I I i i Antelo Vallege 40’ i 30' I .- l Steptoe | Valley ,r—wk i Melrose Dolly StOCk Varden i D011 Flat “ Var en ,U|£ Mtns. g. g 0 D 6 2| “ I s s l .2 g I .3; w I t) m i >1 5 Steptoe . h 2 Valley I' 0 '5 Elko ° ! Carson I 40. ' Cit ' 4o" 00' 0 Y Ely 00" d (LO , .9. 2‘ - a . lg 114° 4,; m i h. <1 , H l 0-! Miles 0 5 2H) 15 20 F1Qure 1-1. Index map of east-central Nevada showing location of the Melrose Stock. Modified after 6.6. Snow. 6 at 125 m.y. or early Cretaceous (Armstrong, 1963). The relationship of the stock to the surrounding host rocks is quite varied as seen in Figure 1.2. The southern margin of the stock is in contact both concordantly and discordantly with Permian limestones that have been tipped on end. The sediment-intrusive contact is sharp and the effects of heating from the intrusive do not extend more than a few feet into the sediments. Recrystallization of the limestone along the contact is not significant and there is only minor silicification of the host rock. The west side of the intrusive is truncated by a basin and rahge fault which runs the length of the range though it 18 covered by alluvium along much of its extent. The north and east sides of the stock are overlain unconformably by Tertiary ignimbrite. The ignimbrite has apparently covered uInch of the stock which had been layed bare by post-Laramide uplift and erosion. Core, recovered from drill holes approxi- mately a mile to the east of the intrusive and at a depth of nearly 1,000 feet below the surface, reveals Tertiary flow breccia lying unconformably on the intrusive. Out- liers of quartz-monzonite found to the east of the drill h(ales are believed to be apophyses of the intrusive. Unpublished aeromagnetic maps indicate the intrusive to be much larger than its outcrop exposure. Snow's con- eSlusion that the intrusive's roof is quite irregular appears to be correct from observations of the magnetic map. The lbasin fault downdrops a portion of the stock west of the T , 114' 35' ’ R65E R66E Tdi a £7 ~\:fm‘ Tdi ‘ --.__ Tdi Tdi 9 _ e 3.5 Tdi qu 8 31 3° 3'9 ‘ 0A ‘ 36 . 29 e Pg!“ '37 7 46 33. . Tdi \ .24 4o. 33 e 27° qu .41 Qal 25 ~47 4? 4 32 '28 3 20 260 4.3 e g '23 2 . ’ 44 e P m 1. 19 q ' I; e 1.1 1.2 13 14 1 ° ° '18 Qal \ T2916 29x 3653 Pm T29N 40' 20 R65E 'rzsn Mn“ 40‘ 20““. __ r 1 I —1 | . o 1 I, 1 ‘ 114‘ 35 \ Qal 'm Qal - Quaternary alluvium Tdi - Tertiary volcanics qu - Intrusive igneous rocks Pm - Permian sediments . - Sampling sites -—"- Faults lPigure 1-2. Generalized geologic map and site locations of the Melrose Stock(after G.G. Snow). 8 range at least 780 feet for a well to that depth in the a]. luvium and west of the fault does not penetrate the quartz monzonite. This fact is also substantiated by the aeromagnetics. On a regional scale the Melrose Stock is on a magnetic high trend which can be traced from the Uinta Mountains, through Bingham Canyon and Gold Hill to the Dolly Varden Mountains (Zietz and others, 1968). The rocks in the stock are of intermediate composi- tion, porphyritic (0-15 percent phenocrysts) granitic rocks. Most commonly the rocks are from medium to coarse grained. Xenoliths from a few centimeters to a few tens of centi- uHaters are rather common throughout much of the intrusive. In hand specimen, variations exist in color and composition which permit the rocks to be easily divided into two types, a quartz-monzonite and a monzonite. The contact between the two units is transitional over a distance of about 10 feet. The monzonite constitutes about 25 percent of the exposed area of the stock. is; Rock same A total of 48 sites was sampled during the summers of 1968 and 1970. The site locations are shown in Figure 1~2. Hand samples were collected from all sites, oriented Qtires from 35 sites, and 5 additional sites have cores taken f.t-om unoriented hand specimens. A fairly uniform, site density was established over the stock. The location of alites was based on geophysical rather than geologic grounds. Drilling of the oriented cores was carried out with a portable drill unit similar to that described by Doell and Cox (1967). Drilling of a three inch core generally took about five minutes. The time required to obtain 6—8 cores ranged from one-half to one hour. Breakage during drilling resulting from incipient fractures and weathering amounted to at least 2 to 3 cores per site. The size of a collection site was in part dictated by the extent of the outcrop which generally ranged from 20 to 70 feet in the maxinum dimension. The cores were uniformally selected °Ver the entire area of the outcrop. The orientation of the cores was accomplished with a BJ:~unton compass and an orienter made of copper. The con- struction of the orienter is a slotted barrel mounted ortho- gonally to a square plate upon which the compass is placed. The precision of the orienter is estimated at 2 degrees. From one to three specimens were cut from each core arid then all specimens were stored with the same orienta- tion in the ambient field of the laboratory from the time Of preparation to measurement. Thin sections were made from the trimmings of the cores. Polishing of the thin 8actions permitted them to be used for reflected and trans- I“-‘ltted microscope and electron microprobe investigations. In addition, opaque section mounts were made for all of the 31 tea from the core trimmings. The following convention was adopted for specimen indentification. Example: VN-l4D-2. The VN is the code 10 for the area, Victoria, Nevada. The 14 is the site number in that area. The letter D refers to the fourth core drilled at Site 14 and 2 indicates that it is the second specimen from the end of the core nearest the outcrop surface. CHAPTER II PETROLOGIC PROPERTIES 2 . 1 Petrology .2 . 1 . 1 Introduction A significant aspect of this investigation is to es— tablish the relationship of the magnetic mineral suite to the constituent minerals in the rocks and to determine this relationship as expressed in the magnetic properties. The relation of the magnetic minerals to the host minerals in the rock has not been previously investigated in the manner presented in this chapter. Previous observations of other rock suites have given the stimulus to carefully Observe the magnetic minerals in their environment for what they can reveal about the history of the intrusive. The data reported in this chapter will be used to draw Conclusion, regarding the magnetic properties and history of the Melrose Stock. The petrologic investigation of this research project involved the measurement of a number of physical quantities including modal volumes, size distributions and intergranular relationships. The latter gave rise to two parameters developed by this investigator in an attempt to study in 11 12 detail the magnetic mineral suite. 2.1.2 Modal Analysis Modal analyses were carried out at a magnification of 140x with a total of 1,000 point counts per thin sec- tion, permitting constituent minerals to be determined relatively accurately. This relatively high point count was made to accurately determine the modal volumes of the ferromagnesian minerals because of their special interest in this study. The modal volume of opaque minerals was determined in conjunction with the association study of approximately 9,000 counts per thin section. Knowing the amounts of the ferromagnesians and mag- netite will permit the evaluation of their association with each other, how each is reflected in the magnetic properties, and to what extent the geometry of the magma chamber influenced the formation of these minerals. The results of the modal analyses are shown in Figure 2-1. The results of the modal analyses fall into two distinct groups, a quartz-rich and a quartz-poor, with a subgrouping in the quartz-rich group. The quartz-poor group falls into the monzonite field with the amount of K-feldspar slightly greater than the plagioclase. The data which do not fall into the general grouping may be explained in the following manner. The specimen in the syenite field may well represent the influence of the proximity of the limestone-intrusive contact which is only 13 Quartz // Gr QM Grd 0 , fi‘ . / [$..‘ 0. O ’1 .0 e 0 I {.f' .. O 3. ; ”:... ‘\. . . ..I’ e ' ~‘o‘--.” D. .1 A ' 9. ‘L 0'. 0: w\ 0 ll. 0 e .‘I M \ SYd \\ ‘1nL3-—5‘—4L -ae - K-feldspar Gr - Granite Plggioclase Volume, % Syd - Syenodiorite QM - Quartz Monzonite S - Syenite Grd - Granodiorite M - Monzonite Figure 2-1. Partial modal analyses of rocks of the Melrose Stock. Rock composition fields of the U.S. Geological Survey are shown. 14 100 feet away. The two data points in the syenodiorite field are rather coarse grained and therefore may not be representative of the rock. The other major group is in the quartz-monzonite field with the minor subgrouping in the granodiorite field. The data point well in the granite field is from Site 14 and may represent a specimen from a granitic dike, one of the accessory rock types in the intrusive. A plot of the constituent minerals on a ternary dia— gram is shown in Figure 2-2. In this representation two fields exist with no evidence of further subdivisions as in Figure 2-1. The upper group of data in Figure 2-2 corresponds to the quartz-monzonite sites of Figure 2-1, the lower to monzonite sites. The monzonite group has about 5 percent more ferromagnesian minerals and 10 percent more feldspars than the quartz-monzonite, but the signifi- cant difference is in the quartz content. The monzonite is evidently a more basic phase of the stock which is supported by the fact that it has augite as one of the ferromagnesian minerals. Furthermore, a Michel-Levy deter- mination of plagioclase composition shows values of An36 andAn4O for the quartz-monzonite and monzonite plagioclases respectively. There is a 93 percent probability that this difference is real. The lithologies delineated in Figure 2-1 are shown in Figure 2-3. A dashed line separates the general areas of these three rock types. The monzonite is confined to the 15 Quartz Monzonite Monzonite o _~u °Au° —21 ° 44e_, - s—Mam 10 20 30 40 Feldspar Ferromagnesian Modal Volume, % Figure 2-2. Modal analyses of rocks from the Melrose Stock. 16 T 114' 35' R6513 R66E ” ‘\:f\‘ \~‘ ._4. + ‘ + + “~~‘ 1”’-;\\\ + + s- \\\ “ X \\ + N : ,\ II. X \\\ + ’x x x‘\ __ fl \\ I ’’’’’ ,‘- " ‘- \\ I r o \\ x 1’, ”1’. o ‘ ‘ ‘ +,/ <9 0 0 o + ,. ' , e e . ,2; <3 G G O 4‘l' \\____ ‘ O) \ _ v + ‘~.' . £1“ T29N R6613 \T29N “ 40' 20' R65E T28N Miles 40‘ 20' - I t 1 I 1 \ l 114‘ 35' ° 4 3’ 1 Figure 2-3. + Quartz Monzonite e Monzonite x Granodiorite Rocks types of the Melrose Stock based upon modal analyses. 17 southeast portion of the exposed intrusive. The one site of monzonite in the southwest corner of the intrusive may be connected to the main monzonite mass in the subsurface. The spatial grouping of the granodiorite sites lends support to the distinct subgroup as shown on the ternary plot of Figure 2-1. For the purposes of this study only the quartz mon- zonite and monzonite rock types will be delineated. The sites having compositions in the granodiorite field are grouped with the quartz monzonite. This is done because both are quartz rich and have no augite. The magnetic properties appear to reflect differences in the quartz and augite content and not in feldspar ratios. No field con- tact is apparent between the granodiorite and quartz mon- zonite. 2.1.3 Microscopic Description A description of the various minerals will provide a picture of the events which took place during solidifi— cation of the magma. The opaque minerals appear to have formed throughout much of the cooling history, and conse- quently the character of the constituent minerals will shed light on the genesis of the opaque suite. The follow- ing description applies to both rock types. Plagioclase exists as both matrix material and large crystals, but the crystals are not as large as the microcline phenocrysts. The tabular plagioclase crystals, ranging to 18 about 5 millimeters in length show albite, pericline and carlsbad twinning in order of decreasing abundance. In the ground mass of many specimens it is difficult to differ- entiate the feldspars from each other because of the general absence of twinning in the plagioclase. Alteration of the plagioclase to calcite, epidote and albite is very common. Many plagioclase crystals diaplay a cataclastic fracturing as if late stage emplacement or movement fractured them. K-feldspar occurs as orthoclase and microcline, much of which has excellently developed perthite. Myremekitic texture is common. Most of the K-feldspar appears in the matrix and is highly poikilitic to most of the other min- erals. Carlsbad twinning is observed in the phenocrysts. Microcline gridding is common, but it is poorly preserved due to the pervasive argillic alteration. The effects of fracturing in the plagioclase seems to have its equivalent in the K-feldspar as kinking of the cleavage planes indi- cating stresses were imposed on the grains late in the cooling history of the intrusive. In a few thin sections the K-feldspar is secondary, apparently related to hydro- thermal activity. The quartz is generally quite free of inclusions as contrasted to the K-feldspar. It appears to have formed primarily before the orthoclase because only approximately one-fourth of it can be considered interstitial. The grains are generally composite, but not to the extent of being mosaic. In a few thin sections, however, a portion of the 19 quartz is obviously mosaic and related to hydrothermal activity. A few specimens reveal the quartz grains to be highly fractured. The anhydrous, ferromagnesian mineral augite is nearly completely restricted to the monzonite. A few of the quartz monzonite and monzonite specimens show traces of augite. Due to the point counter used in the study having a capacity for tallying only six minerals, augite was counted with the hornblende. Its modal volume varies from 20 to 90 percent of the hornblende volume in the sites where it occurs in appreciable amounts. Augite averages approxi- mately one-half of the hornblende in the monzonite sites as determined by counts of the hornblende-augite ratio in a few thin sections. Much of the augite is intimately associated with the hornblende apparently as the result of the incongruent conversion of augite before equilibrium was reached. Hornblende grains show a wide degree of alteration; a fine mixture of chlorite, biotite and epidote fill some hornblende pseudomorphs, while others are fresh and show excellent twinning. This variation can be seen in a single thin section and indicates hydrothermal alteration along minute avenues in the rock. The altered amphibole is generally poikilitic, especially with magnetite. There are instances of the hornblende showing peritectic con- version to biotite. Biotite is characteristically poikilitic and in the 20 monzonite occurs in distinctively larger crystals. It does not show as much alteration as the hornblende, but where it is altered, the product is chlorite. Biotite does not appear to be as closely associated with hornblende as the augite is to hornblende. The plagioclase content is nearly the same in each rock type. The quartz-monzonite has approximately 15 percent less K-feldspar than the monzonite. This effect is more than offset by the great difference in the quartz content. The monzonite has twice as much hornblende as the quartz monzonite while the amount of biotite in each remains nearly the same. Because roughly one-half of the hornblende content in the monzonite is augite, the hydrous ferromagnesian content in both rocks is approximately the same. The magnetite is twice as abundant in the monzonite. The accessory minerals in decreasing order of abun- dance are magnetite, ilmenite, sphene, apatite, zircon and corundum. The minerals appearing as alteration pro— ducts are sericite, calcite, epidote, biotite, chlorite, quartz, albite, leucoxene, and hematite. The accessory minerals relating to the opaque mineralogy will be dis- cussed individually in later sections. A compilation of the modal analyses is given in Table 2-1. Listed for each of the constituent minerals and mag- netite (including other opaque accessories) is the minimum and maximum values measured with their respective arithmetic means according to rock type. 21 m m H H N o .uos m H~ H 6 OH H .on «H AH o m m m .cuom v m o «N am mH .uuo Hv mm m mm mm mH Hummus mm mm «H on me aw .omHm new: .xmz .cwz coax .Nmz .CHZ OHHCONGOZ On. HEONGOE NHHmHHO sooum onouaoz 0:» no names cum .mssedxme .mEdEHcHE mammamsm Have: .HIN magma 22 2.2 Apparent Grain Size Distribution of Magnetite The size of the magnetic minerals was measured in an attempt to understand more fully the behavior of the mag- netic properties. Knowledge of the grain size distribution will help predict the stability of the remanent magnetism. If the bulk of the magnetite grains is in the size fraction greater than roughly 50 microns, the remanent magnetiza- tion will have a low coercivity and hence a soft remanence. The measurement of grain size cannot be taken as ab- solute since a thin section gives values of grain sizes which only approach the true size. However, if a compari- son is made on a relative basis, then valid conclusions can be drawn from the data. The size distribution count was divided into two cate- gories, those opaque grains associated with quartz and feldspar (nonferromagnesian) and those allied with the ferromagnesian minerals. The two categories represent zones where iron is present and deficient. In this manner an insight into the growth history of the magnetite grains will become clearer. The size of the magnetite grains is expected to reflect the "supply" of iron available where their growth was taking place. Zones in which the iron 'supply"was limited, as well as the time for growing, will show a smaller size of magnetite grains. The ferromagne- sian minerals are a likely source of iron if it is avail- able through some reactions. 23 The following criterion was used to define the associa- tion of a magnetite grain to a group. If a grain was in- cluded or had more than half of its area embayed by a given category (ferromagnesian vs nonferromagnesian), it was counted for that group. There was also a problem of deal- ing with composite magnetite grains. Composite grains were treated as single grains when 15 percent of their contacts were in common. The rational for this was that the magnetic properties are a function of grain size and hence a cluster of small grains together would tend to express the magnetic properties of a large grain. In order to express the properties of large grains, the cluster has to be crystallographically continuous across the boundar- ies of the individual grains. In other words, it is assumed that when more than 15 percent of the perimeters are in contact, one grain nucleated on another as an overgrowth and are structurally continuous, though not necessarily in the same crystallographic orientation. The value of 15 percent was arbitrarily chosen. Magnetite grains altered along fractures can be divided into many magnetically independent grains. Exsolution of titanomagnetite or ilmenite in a magnetite grain will also split up an appar- ently homogeneous grain. This must be taken into account as reported by Larson and others (1969), but this did not have to be applied to the samples of the Melrose Stock, because the magnetite was free of exsolution. The above distribution of magnetite represents a 24 frequency of occurrence based on the number of grains en- countered and not volume for that percentage. For example, if 96 percent and 4 percent of the magnetite grains were in the 0-5 and 5-10 micron groups respectively, the percent volume of magnetite in each fraction is approximately the same. A larger size fraction, though having a small por- tion of the magnetite in numbers of grains, can be the major contributor to the magnetic properties. A total of ten thin sections were measured for size distributions, five from each rock type. From 500 to 800 grains were counted on each thin section. A magnification of 280x was used with a graticule having 5 micron divisions as a standard of reference. The number of grains in the field of view falling into each of seven size groups was recorded. The seven size groups are 0-5, 5-10, 10-25, 25-50, 50-100, 100-200, and greater than 200 microns. Grains of 1.0-1.5 microns were at the limit of resolution. Going to a higher magnification would have increased the resolution to some extent, but the large grains would have more than covered the field of view making their determina- tion difficult. The results of the measurements are shown in Figures 2-4 and 2-5. The histograms represent the percentage of the grains encountered which fell into each of the seven size groups. All of the histograms have a maximum in the smallest size fraction and then drop off sharply in succeeding 25 mucouusouo .1 on x mm . 6H m on . mm uom cm“ .3 00¢ oom OOH oov com OCH CH 00¢ CON OOH mo hucosoeum o 2 0 Apparent Grain Size, microns within the l) opaques associated with the quartz-feldspar fraction. 2) opaques associated with the ferromagnesian fraction. Apparent grain size distribution monzonite pluton. Figure 2-4a. 100 26 100 09. oou 1 SH W cm a mu 2 V.\ (1) (2) VN-41D 00¢ 0 a 6 4 2 1 cow 6 4 D m 00H m om . . mm m \x. S \ m mucouusooo mo accosooum w oov .. oom OOH om mN OH cow com 00H memo Hum Apparent Grain Size, microns tion. 2) opaques associated with the fer- associated with the quartz-feldspar frac- romagnesian fraction. Apparent opaque grain size distribution within the monzonite pluton. 1) opaques Figure 2-4b. 27 ousouuaooo Mo mucososhm 1) opaques associated with the quartz- feld- spar fraction. 2) opaques associated with Apparent opaque grain size distribution the ferromagnesian fraction. within the quartz monzonite pluton. Figure 2-5a. 28 100 100. VN-47A " VN-34E 8 6 (1) 4. 4 2 5 s I \§§§\§ L §\ 2 \ 00¢ 00N 00H “om mm 0H 00¢ 00m 00H 0m mm 0H 60 . m m Apparent Grain Size, microns quartz monzonite pluton. 1) opaques associated with the quartz-feld- spar fraction. 2) opaques associated with Apparent opaque grain size distribution the ferromagnesian fraction. within the Figure 2-5b. 29 intervals. The size distribution is expected to display a bell shaped curve, consequently the results indicate the right hand portion of such a curve. Without greater resolu- tion, it cannot be determined whether the smallest size fraction measured actually represents the most abundant grains size group. Similar one-tailed distributions have been recorded by Larson and others (1969). In general, for both the quartz monzonite and mon- zonite the grain size of the opaques associated with the nonferromagnesian silicates is smaller than those related to the ferromagnesian minerals. Within the nonferromag- nesian group, 99 percent or more of the opaque grains fall in the size fraction of less than 50 microns (See Table 2-2). Grain sizes falling in the largest size group make up less than 1 percent of the grains. The ferromagnesian group on the other hand has much more magnetite in the coarser fraction (a-SQ/a), more than ten times the number of grains as the nonferromagnesian group. The difference which was noted between the two cate- gories suggests an influence of the ferromagnesian miner- als on the size of the magnetite grains. Because the large grains are ten times as abundant about the ferromagnesian minerals as contrasted to the quartz and feldspar, the iron silicates favored or enhanced the formation of larger magnetite grains. Perhaps these minerals served as the source of iron for the opaque oxides. This point is dis- cussed in greater depth in the association study. 30 Table 2-2. Percentage of opaque grains greater than 50 microns in diameter. *Opaque Association Rock Type Nonferromagnesian Ferromagnesian Quartz Monzonite 0.4 8.8 Monzonite 0.4 4.4 *Based on 5 thin sections from each rock type 31 A visual comparison of the histograms (Figures 2-4, 2-5) of each rock type shows that the magnetite in the quartz-feldspar category has approximately the same frac- tion distribution, but the magnetite associated with the ferromagnesians in the quartz monzonite is noticably coarser than the magnetite in the monzonite. The quartz monzonite and monzonite respectively have 8.8 and 4.4 percent of the opaque grains in the size fraction greater than 50 microns. The larger grain size of the magnetite in the quartz monzonite indicates that the magnetite had a longer time to grow and/or that more iron was available for rapid growth of the crystals. The viscosity of the monzonite would have been expected to be less than that of the quartz monzonite so that the growth potential of the magnetite grains would be enhanced in the monzonite. However, the results do not support this idea. The forthcoming section on mineral associations shows the monzonite to have had a shorter crystallization history than the quartz monzonite which suggests a shorter duration for the magnetite to grow in the monzonite than in the quartz monzonite. It is believed that a longer time span of crystallization per- mitted growth of larger magnetite crystals in the quartz monzonite, perhaps in part due to the deeper burial. The hydrous phase content should help to promote the growth of larger magnetite grains, but it is not known which magma had the greater hydrous phase content. 32 The results of the magnetite grain size distribution study suggest that the stability of the remanent magnetiza- tion will be inversely proportional to the ferromagnesian mineral content, because the large magnetite grains (:>50 ‘/~) show a preference to the ferromagnesian minerals. Fur- thermore, the higher percentage of large magnetite grains in the quartz monzonite indicates that the monzonite should have a more stable remanence than the quartz monzonite. Nagata (1961) has shown the coercivity, hence the remanent stability, of magnetite grains increases appreciably for grains less than 50 microns in size. 2.3 Opaque Petrolggy 2.3.1 Introduction An integral part of this study was the investigation of the opaque mineral suite, its connection to the genesis of the constituent minerals and their relationship to each other. To these ends, a total of 42 opaque mounts from as many sites, 25 mounts from the quartz monzonite and 17 mounts from the monzonite, were examined at 450x. The opaque species, their relative abundance by visual esti- mates, their textures and relationships to each other were recorded. The study primarily focused on the iron-titanium oxides, but the occurrence of sphene was also noted for this mineral may aid in the evaluation of the magnetic minerals. The origin of sphene and the exsolution within the ilmenite are considered as evidence for zoning within the intrusive. 33 The following opaque species were identified: mag- netite, ilmenite, hematite, goethite, maghemite, pyrite and chalcopyrite. Distinguishing between sphene and goe- thite occasionally posed a problem. 2.3.2 Magnetite and Ilmenite The two primary opaque oxides of the Melrose Stock are magnetite and ilmenite with hematite, goethite and maghemite as alteration products. The two major opaques do not show a significant difference in their relative abundance with regards to rock type. The monzonite has 12 percent ilmenite, 85 percent magnetite and the remaind- er as alteration products and the quartz monzonite has 10 percent ilmenite, 85 percent magnetite and the remainder as alteration products. Tabulations of the sites with sphene coronas and composite ilmenite-magnetite grains revealed no preference to rock type. Magnetite is present in all opaque sections and is always the dominant opaque. Microprobe results indicate the magnetite to be pure magnetite with no apparent exsolu- tion. The polish of the magnetite grains varies consider- ably, but in general it shows moderate to heavy pitting, as a result of surface alteration to goethite which was removed during polishing. The grains are anhedral to euhedral with a tendency for the euhedral grains to be more closely associated with the nonferromagnesian min- erals. The habit of magnetite varies from single grains 34 to clusters of several grains. No evidence of exsolution was noted in the magnetite, but hematite was commonly seen in the magnetite along structural planes. This occurrence of hematite is interpreted as a result of hydrothermal al- teration which is also associated with the alteration of other minerals in the specimen. In several instances, ilmenite and magnetite occur together as abutting grains. The shape of the combined unit, which is often quite regu- lar, suggests that the grains existed previously as a grain of a single species. This abutting relationship is evidence of an advanced stage of exsolution of an original titano- magnetite according to Buddington and Lindsley (1963). Two sample mounts displayed local alteration of magnetite to maghemite. Sphene coronas occur around both magnetite and ilmenite, however, they were more frequent around ilmenite. The few grains of ilmenite analyzed with the micro- probe show them to be pure ilmenite. Exsolution was ob- served in some of the ilmenite grains and is shown in Figure 2-6a. The remaining photos of the sequence are referred to in section 2.3.3. Ilmenite generally displayed its prismatic shape in cross section as contrasted to the magnetite which was more equidimensional. The polished surface of the ilmenite grains was more homogeneous than the pitted magnetite. Commonly the ilmenite was altered to goethite, but this alteration attacked the grains from the perimeter inwards. fiavzuo Figure 2-6a,b. 35 Electron microprobe microphotograph of an ilmenite grain with hematite exsolu- tion and a sphene reaction rim. a) sample current, b) titanium fluorescence. Figure 2-6c,d. 36 \OJe/diu‘ Electron microprobe microphotograph of an ilmenite grain with hematite exsolu- tion and a sphene reaction rim. c) cal- cium fluorescence, d) iron fluorescence. 37 This is in contrast to the magnetite. Pseudomorphs of goethite after ilmenite were occasionally seen. In sever- al instances the nearly complete alteration of ilmenite to sphene was seen with only a skeletal ilmenite remnent. Ilmenite in many sites has what appears to be well formed exsolution lamellae of hematite. These blebs, approximately 0.5 microns in width, and 2.0 microns in length parallel the long axis of the ilmenite grains. Minor amounts of hematite occur as an alteration product of ilmenite. The lack of apparent difference in the opaque assem- blage of the rock types suggests that the chemical and physical condition of the two phases was similar, at least in the later stages of crystallization. The two litholo- gies cannot be differentiated on the basis of their opaque assemblage. There are, however, differences in the opaque assemblage, but these reflect influences due to the geo- metry of the magma. A tabulation of the opaque data is given in the Appen- dix B. 2.3.3 Sphene Sphene was selected for investigation because it appears to be related to the ilmenite and to the geometry of the intrusive. The occurrence of sphene was noted in a qualitative way. The sphene content was classified into three categories: trace, minor and common. In terms of 38 actual modal volume, the common grouping approximates one percent. There is a lack of correlation of sphene content with lithology. However, the plotting of the areal distribution of sphene may reveal patterns related to the geometry of the intrusive. The sites in which sphene was noted as common are plotted in Figure 2-7. ‘In addition to the data obtained from the opaque mounts, results of the thin section mea- surements are also included. This was done as a check of the opaque mount data because of the possible errors of identifying sphene as goethite. In thin sections there is no problem in distinquishing between the two minerals. Figure 2-7 shows that sphene, where common, is limited to the southern part of the intrusive, an observation support- ed by both the opaque and thin section data. The fact that data from both samples at each site do not substan- tiate each other is not surprising because the two types of sections are not always from the same core. Hence, they may be sampling some of the heterogeneity existing at a given site. It is believed that the margin of the intrusive might have been an influence on the formation of sphene. A mar- ginal zone of approximately 5,000 feet width reflects the influence of the host rock on the opaque mineralogy of the intrusive. If the vertical margin did have a control on sphene formation, the interior and northern portions of 39 Figure 2-7. +' Sphene common in opaque sections C) Sphene common in thin sections Minor or traces of sphene Distribution of sphene T 114' 35' R65E R66E ” . \§:~‘~ \‘ __.._ N e G . + 1-‘+_\“ . 1‘3““ +__- °--+-- @ 0 CD - 1: g g 6 D a e . ¢, \ . ' o . £22.“ r292: R66E )T29N *- 40' 20' :5 R6SE 'rzsu Mn“ 40‘ 20. _ I I I I 1 \ l 114' 35' ° 4 ’5 1 \ 40 the stock were apparently greater than 5,000 feet from a margin or contact. It is then inferred that at least 5,000 feet of rock have been removed from the stock and additionally that the stock is much larger than its pre- sent exposure. The occurrence of the abundant sphene is believed to be the result of limestone assimilation in the crystall- izing melt. The effects of assimilation are expected to decrease towards the interior of the magma chamber result- ing in a halo or rim confined to the periphery of the intrusive body. The formation of sphene requires CaO, 8102 and TiO2 under the proper pressure-temperature con- ditions. The reaction 3 FeTiO3 + 3CaSi03 + koz-e.Fe304 + 3CaTiSiO5 as suggested by Verhoogen (1962) can well explain the observations in the Melrose Stock. Wollastonite (CaSiO3) probably was not present as a reactant but more likely as Cao from the dissociated limestone and SiO2 from the melt. The oxidizing conditions necessary to drive the reaction to the right would be enhanced by the limestone dissocia- tion, H20 derived from the sediment, and H20 already pre- sent in the melt. .The TiO2 has a large affinity for CaO as shown by the very large negative free energy (-77.1 kcal at 298'K) of the reaction. The common sphene coronas about both ilmenite and magnetite plus the sphene pseudomorphs after ilmenite give strong support to the proposed reaction. The 41 microprobe photos of Figure 2-6a through d show, in add- ition to the exsolution, a portion of an ilmenite grain with a sphene corona. Photos ”b", ”c" and "d" are of the same area as "a", however, they are made of the x-ray fluorescence of the elements, titanium, calcium and iron. Titanium covers the complete photograph (b) with nearly the same intensity, but a faint outline of the ilmenite grain can be distin- quished where the ilmenite concentration is slightly great- er. The calcium and iron photos (c and d respectively) are reverse images of each other, where the iron concen- tration is high, the calcium content is low and vice versa. The detail of the exsolution cannot be resolved in these x-ray fluorescence photos. This series of photos substan- tiates the reaction in which sphene is formed at the ex- pense of ilmenite. Further evidence of the proposed reac- tion would be obtained from the ratio of ilmenite to mag- netite which would be expected to decrease near the margins of the intrusive. The reaction produces magnetite at the expense of ilmenite. A plot of relative ilmenite content in all sites from the south contact shows no trend. This is not particularly disturbing because near the contact not all sites have abundant sphene. This may reflect in- homogeneities resulting from the stoping of limestone blocks. In other words, the effects due to the limestone assimilation are probably not pervasive but more abundant as subzones near the contact. 42 The average ilmenite content of the sites possessing common or abundant sphene in thin sections and opaque sec- tions relative to all of the opaques is 8 percent and the content in those sections having minor or a trace of sphene is 14 percent. This inverse relationship, even though there is overlapping of values of the two groups, does support the idea of the sphene forming at the expense of the ilmenite. The conditions necessary for formation of ilmenite from direct precipitation or by removal of titanium from a pyroxene structure requires a reducing atmosphere early in the crystallization history. The formation of sphene necessitates a change to an oxidizing or high partial pressure of oxygen later in the cooling history. The change in the "atmosphere" of the magma reflects the normal concentration of the hydrous phase as crystallization procedes and the dissociation of the sedimentary host rock releasing C02 and H20. It appears that the geometry of the magma has influ- enced the assemblage of magnetic minerals, but the effect is not great enough that variations of magnetic suscepti- bility due to the proposed reaction can be meaningfully delineated. However, the influence of the host rock does provide an insight into the geometry of the magma chamber. 2.3.4 Hematite The occurrence of hematite can provide useful 43 information regarding the paragensis of the opaque miner- als as well as of the intrusive. Normally the presence of hematite indicates an oxidizing condition, perhaps as a condition resulting from the abundance of intergranular fluids, primarily water. However, hematite can also form as a product of the unmixing of a hemo-ilmenite through a slow cooling history. The habit of hematite to the other opaque mineral may indicate something of the past events of the Melrose Stock. As mentioned earlier, the hematite occurs in both the magnetite and ilmenite. The hematite associated with the magnetite is undoubtedly due to late stage alteration. The hematite associated with the ilmenite is open to question regarding interpretation based on petrographic observation. The two alternative origins, of the hematite oxidation and exsolution, shall be considered. The sites having the apparent hematite exsolution within the ilmenite are distributed along the western bor- der of the range as shown in Figure 2-8. The proximity to the basin and range fault on the west side of the range may indicate a cause and effect relationship. The fault may have served as a major avenue along which fluids mi- grated, and altered the rocks of the intrusive, thus suggest- ing that the blebs of hematite are a result of alteration or oxidation. A number of abandoned prospects are found on the west side of the range indicating areas of alteration and minor mineralization. The distribution of sites having 44 Figure 2-8. 29m R66E R65E T28N Miles I ' I r I \ l 114' 35' ° 4‘ 5 1 \ O Abundant exsolution ' No exsolution(or trace) Distribution of hematite exsolution within ilmenite. T 114' 35. ’ R65E RGGE ’ A \z“ \‘ __..__ 3 e 3 O . N O . e 9 . e . . G e O 45 hematite after magnetite also fall along the western margin of the range. The evidence supports the contention that the hematite is derived by alteration. There is also evidence supporting the view that the hematite blebs are in fact exsolution lamellae. The blebs appear to be discrete entities within the ilmenite grains with no apparent avenues or partings to the perimeter of the grains as evidenced in the alteration of magnetite. If the lamellae are true exsolution, the hematite would be expected to have formed deep in the intrusive where temperatures and slow cooling rates permitted exsolution. Support to the latter hypothesis can be obtained by studying the distribution of hematite within the intrusive. A measure of depth within the magma chamber is obtained from the distance to a known contact (as on the south) and the site's elevation. In general (Figure 2-8), sites displaying the hematite lamellae are in excess of one mile from the known margin of the intrusive. The exception is Site 26 which is in the monzonite. The site elevation effect is tested by plotting vertically the site elevation positions and denoting the presence of hematite lamellae. The results are shown in Figure 2-9. There is a remark- able correlation between the presence of hematite lamellae and sites of low elevation. The occurrence of the apparent exsolution is found deep within the magma chamber. Not all sites with the apparent exsolution show mag- netite alteration to hematite. This would tend to support 46 mean on somehow suds muesmEHH madness soausaomxe mo coausnauuman .coaum>0Hm oustEHH canvas coausaomxc euaumsos suds spam 6 coaus>0Hm spam . poem .couum>mHm 0005 00mh 00Hh 0060 .mnm museum 47 the view of the exsolution for the alteration, had it occurred, should have affected both the magnetite and ilmenite. Furthermore the alteration of ilmenite would be expected to yield one or more titanium bearing minerals. The evidence seems to more strongly suggest the blebs of hematite to be true exsolution rather than due to oxi- dation. Neither alternative can be rejected based on existing evidence. The possibility cannot be ruled out that the western portion of the intrusive may have been subjected to both the effects of slow cooling and altera- tion. If the exsolution hypothesis is accepted, a broad zoning of the intrusive can be postulated based on the distribution of sphene and the exsolution within the i1- menite (Figure 2-10). An outer rim exists which is nearly a mile in thickness; this zone reflects the influence of the assimilation of the calcareous host rocks by the magma. An inner, deep seated zone is present and is in excess of a mile from a margin or roof. The interior zone would provide a favorable environment for the unmixing of an heme-ilmenite. The other zone may be intermediate to the above zones, or it may possibly be a result of the roof of the intrusive in that portion of the stock being low in calcium. Figure 2-11 illustrates the general nature of the zoning on a northwest-southeast profile drawn through the Dolly Varden Mountains. 48 T 114' 35' ’ R65E R66E ’ ‘ ‘ . ‘\“n‘ \ \‘ -é.-— \ o \ \ l 3 ‘ . ‘ O A I O . . I N exsolution ’ , _ / do exsolution /’ 'sphene’poor O / . O . I / O sphene rich. . ‘ T291“ T293 R66E )T29N *- 40' 20' - R653 T28N M1163 A. 40. 20. _ I a I ' I \ 114'35' 5‘ ’5 \ Figure 2-10. Opaque mineral zones within the stock reflecting the geometry of the magma chamber. Line A-A' is the location of profile of Figure 2-11. 49 A \ , A' , sphene rich \ exsolution exsolution, by, sphene 3:31 Alluvium poor \ I / \ Melrose Stock gégi Sedimentary rocks Figure 2-11. Cross-section of the Melrose Stock showing the observed opaque mineral zoning. See Figure 2-9 for location of profile. 50 2.4 Association Coefficients 2.4.1 Introduction In observations of rocks in thin section, relation- ships are commonly noted that are more intimate than would be expected if the distribution of the minerals were purely random. Observations on the suite of thin sections from the Melrose Stock appear to show a preferential distribution or association of magnetite to the ferromagnesian silicates. A method was developed to quantify this observation. The method developed will allow a quantitative determination of the spatial distribution of the minerals about magnetite grains. If a preferred association is suggested, or even if it is not, then a better insight will be gained into the crystallization history of the magnetite in particular and the rock in general. Though the method developed has been used in this study for the association of an accessory mineral with respect to the constituent minerals of the host rock, it could also be used to study other mineral grain assemblages in various types of rocks. When determining the association of a constituent mineral, it is advisable to modify the technique slightly to include associations of a mineral with itself. Quantitative investigations on intergranular relationships have been studied by Flinn (1969) and Kretz (1969). In the rocks of the Melrose Stock it is suspected that a portion of the magnetite associated with the 51 ferromagnesian minerals may be genetically related to the iron silicates. If the results of the association study suggest this to be statistically valid, then appropriate reactions can be postulated for the observed relationships. This will permit suggestions as to the relative changes of the oxygen fugacity in the magma chamber with respect to time and space. 2.4.2 Procedure To establish the "neighborhood" about a particular mineral, a point count is carried out in the manner as depicted in Figure 2-12. The figure shows a schematic diagram of a photomicrograph of a rock with four minerals, W,X,Y, and Z. If the association of W with respect to X,Y and Z is desired and a W grain falls under the cross- hair of the ocular, then a tally is given to each of the minerals which are at the points where the graticule line exits the W grain. For the uppermost W grain this yields two tallies for the Y. For the lowermost W grain the combination of two points is an X and a Y grain, so con- sequently, one tally is given to each of their entries. The subtotals of the respective minerals and their total for the hypothetical example are also shown in Figure 2-12. In practice the number of magnetite grains which fall under the crosshairs during a normal point count is rather small. To increase the number of opaque grains encountered a graticule with 20 points in a line was used. In this 52 W Y I W Mineral x y z Count(ST) 2 3 1 Total(T) 6 Association(A) .33 1.5 0.5 Assoc. Coef.(AC) 1.0 1.5 0.5 Figure 2-12. Schematic diagram of data collection pro- cedure for the association coefficients of W with reSpect to minerals X, Y, and Z. Graticule of ocular shown on each W grain. Counts are made at the points indicated by arrows. The resulting point counts, associ- ations, and association coefficients are shown in the accompanying table. 53 manner the probability of having the crosshairs falling on an opaque grain is increased twenty fold and in addi- tion it gives weight to grains proportional to their vol- ume. Thus, if a large grain falls under three adjacent crosshairs, three counts are given to each of the miner- als bordering the opaque grain. Approximately 30-40 minutes were spent on each thin section. Counting was expedited by traversing with the analyzer out so that it was easy to note when an opaque fell under a crosshair. The analyzer was used when a grain was encountered. From 150 to 400 counts were tall- ied in the 30 to 40 minutes. Experimentation with random spacings and directions of traverse lines reveals no significant difference with the results obtained with a uniform grid system. Where no preferred orientation of the grains exists it is safe to use the uniform grid pattern and, furthermore, it is more expedient than the random grid. If there was a pre- ferred orientation of the minerals, then a random grid system would be necessary. 2.4.3 Association Model A probability model has been chosen and parameters derived from the expected distribution of mineral grain assemblages assuming a random distribution. Values of the parameters deviating from the expected norms are interpreted as being an indication of a non-random distribution. 54 Two assumptions are made in the application of this model to a geologic situation; first, the probability (P) of finding a given mineral at a point on a thin section is equal to that mineral's mode (M) and second, if a thin section has a great number of grains, the probability of observing any combination of two grains in contact is an independent event (equal to the product of the two respec- tive probabilities). Table 2-3 shows a simple hypothetical model with four minerals (W,X,Y, and Z). A similar model could be set up for any number of minerals. The object is to deter- mine the association of W to X, Y, and Z. The first column in the table shows the possible combinations of the min- erals bounding w at the two points where the graticule line exits W. The corresponding probabilities and point counts for these combinations are found in the remaining columns. To illustrate the model, suppose that X has a modal volume of 50 percent (i.e., Px is 0.5). The pro- bability of observing an (X,X) combination, an X grain at each point where the graticule line exits W, should be 0.25. In 100 observations, 25 (X,X) combinations should occur and this would yield 50 tallies for mineral X. Similarly, if the mode of Y is 40 percent (X is still 50 percent), 40 of 100 grains recorded should be (X,Y) com- binations. This brings 40 counts to each X and Y. In this example the subtotals of X, Y and Z are respectively 100, 80 and 20 with a total of 200. The simplification of the 55 oo~ iesHmuoe ooHiacm oOHimva ooHimsa o.H iamcHuuounsm ooHimmmmsu «sum 1N.N. 82133.3: ooHAJJK mm»: 15.2 .30: 82.1133 >3. .3 :3: 82.3.3: ooHlmmxmK Jam :15 .EJC 82.3.3: 82.3 .fiu 1,3,5 :3: .23: co: xmxfia gm .3 3163 N M x thHwnmnoum msowumsansou mussoo usfiom .uucouuammeou cowpmHuoemm Mom Hoooa suHHHnmnous .mIN manna 56 sums shown in the table is a result of the following re- lationship: 1) Px+P +Pz=1000 Y The association of W with respect to X is defined as the ratio of the subtotal of X (STX) to the total T. STx 2) Ax = T where Ax is the fraction of W associated with X. In the case of a random distribution, the association of W to X is the mode of X. Thus, STx ZPx 3) Ax = “‘fir“'= 2-——'= Px' Normalization of the association parameter with the min- eral's mode (M as a fraction of 1.0) Gives the association coefficient (AC), 4) AC = x = 1.0. X With a random arrangement of grains the AC has a value of 1.0. The AC's for Y and Z are calculated in an identical manner. As an example, consider the association coefficients calculated from the data of Figure 2-11 assuming the three minerals are present in equal amounts. The results of these calculations are also given in Figure 2-11. Departures of AC values from 1.0 indicate a departure of the mineral grain assemblage from a random distribution. Departure of AC values below 1.0 indicate a deficiency asso 1.0 mine it: a1 Vi 57 association of two minerals whereas values greater than 1.0 indicate a preferential association. When certain minerals display a preferential association, this necessar- ily means that a deficiency association will be present in one or more of the other minerals. The deviation of the AC values from 1.0 has the limits given by the values 0.0 and 1.0/M, where M is the mineral's mode. The range of possible values for a particular AC is not symmetrical about 1.0. The smaller the mode of a mineral, the greater its AC upper limit. The deviation of an AC from 1.0 is a probability measure. The greater the departure from 1.0 the chances are less that the event will happen, or at least it being random. A Chi-square test can be used to test the signifi- cance of a measured set of AC values from their expected values of 1.0. Using the AC values for the test yields a more conservative estimate of the significance than using the expected and observed point count data. The difference in the test probably reflects the normalization of the data to small numbers as contrasted to the original data. From the association parameter (A) it is a simple matter to estimate the amount of a mineral associated with another mineral which would have to be explained by some association other than a random one. The excess association, the amount greater than the expected, is given by the expression 5) (A-M) x 100%. 58 This expression will be negative if the mode (M) exceeds the association and could be termed a negative excess association. The summation of the excess association will be zero. Disregarding the limited data, for the purposes of illustration, the following remarks can be made regard- ing Figure 2-12. Mineral X has no preferential associa- tion with W. Mineral Z has a deficiency preferential asso- ciation with W and as a result shows less association with W. Mineral Y is the only mineral to show an excess asso- ciation with W, by an amount of 17 percent ((0.50-0.33) x 100%). Thus, 17 percent of W's modal volume is prefer- entially associated with Y. Mineral Z has a 17 percent negative excess association. The sum of the three ex- cess associations equals zero. From the association parameter it is a simple matter to determine the amount of a mineral associated with an- other which would have to be explained by some other asso- ciation than a random one. From the results of the ex- ample given in Figure 2-12, the excess association of W related to Y is 17 percent and this amount would have to be explained by means other than a random association. 2,4,4 Geologic Influences on the Magnetite Association Coefficient The association coefficient can give valuable infor- mation concerning interrelationships of minerals. The purpose of this section is to explore some of the 59 possibilities which can explain excess associations of magnetite to the constituent minerals. Of course, many of the possibilities will be applicable to other associa- tion studies. The association parameter and coefficient were developed on a theoretical basis without reference to geologic processes that may affect their values. The geologic environment has notable differences from the mixing of balls together and analyzing them for associa- tions, although in a sedimentary environment this may not be far from true. The following discussion will primarily be concerned with associations that might arise in an intrusive igneous rock. The description will be concerned with associations divided into two categories, genetic and nongenetic. The former includes associations of the opaque minerals with other minerals which have a common chemical component. Nongenetic associations arise out of circumstances which prevail in the magmatic environment where there is no chem- ical correspondence. These shall be considered first. The formation of an igneous rock is not in most cases the mixing of preformed crystals. Rather, the formation is a long termed event in which various minerals form in response to the physical-chemical conditions of the system. As a result, minerals of different species will be pre- cipitated at different times and in some cases subsequent- ly resorbed. Some minerals may have the bulk of their modes crystallized before the arrival of another mineral. 60 For instance, the early crystallization of mineral A and the late arrival of B precludes the possibility of having B included in A. This would limit the association of B with respect to A. On the other hand, if A and B are the last minerals to crystallize, then there is a high probabil- ity for a strong association of one to the other. An association might arise which can be termed an jJu£><>sed association. This association results from the Viscosity of the melt being sufficiently high to cause the inclusion of one mineral in another because of their in- abi lity to repel each other as one or both grow. Further- more, various minerals may differ in their power of crys- ta:l—lLization to push away adjacent magnetite grains. In addition, the strength of crystallization is likely to be IOJ-ated to crystallographic direction. The two factors, viscosity and power of crystallization, should be kept in mind when evaluating the association coefficient. If magnetite is introduced into a fractured rock, there may be certain minerals having a greater capacity of fractures per unit volume and hence, these would have a larger capacity to host the magnetite. An inherited association may arise from two minerals, A and B being originally associated, and a later mineral, C, forming at the expense of B. This would leave A in association with C. Certain minerals may act as hosts for the nuclea- tion of other minerals and hence would display an associa- tion. In some cases this may be argued as a genetic 61 association, in others as a nongenetic. Genetic associations can be broken down into forma- tive and destructive associations. A formative associa- tion in this study refers to the association of magnetite to another mineral as a result of an incongruent reaction. The candidate minerals for this relationship are the ferro- magnesian silicates. Under suitable oxidizing and P-T conditions, the incongruent melting of a ferromagnesian can provide Fe+2 ska and Fe+3 necessary for magnetite (Czaman- Magnetite formed in this manner will be re- ferred to as an oxidation reaction. , 1970). Another formative association results from exsolu- tion , a phenomenon that occasionally produces magnetite in ferromagnesian silicates. In such a case the associa- tion would be readily apparent and the association study would not provide new information. Destructive associations indicate minerals formed subsequent to the normal crystallization sequence, name- 13! deuteric and hydrothermal alterations. Magnetite re- leased during chloritization and serpentinization are examples of destructive association. Mechanism tending to destroy associations are differ- entifil movement in crystallizing magma and chemical reac- tiona which remove one of two minerals originally associated. rheag processes lead to incorrect interpretation of associa- tion coefficients. The foregoing discussion suggests several ways that I: o} [No "U :2- 9'1 62 minerals may become associated and undoubtedly other pos- sibi lities exist. The processes responsible for the asso- ciation relations can only be determined through careful observations and evaluation of the textural relationships. 2 - 4 - 5 Results and Interpretation The AC determinations of 34 thin sections from the quartz monzonite and 20 from the monzonite are shown in Figure 2-13. The ten histograms show a breakdown of the AC: values for each of the five constituent minerals for each rock type. The common abscissa is used to aid in com- Parison between rock types. The dashed line on each his- togram represents the arithmetic mean for the respective AC groups. The first striking feature of the histograms is the not-able variation of the AC values for the constituent mineral within a rock unit. A Chi-square test on the deviation of the AC's from 1.0 shows that the quartz mon- zoni te distribution is significantly different from an expected random distribution at the 99 percent level, whereas the monzonite has about a 10 percent chance of be- ing randomly distributed. Focusing attention on the results of the quartz mon- zonite, the feldspars and quartz have less magnetite asso- ciation as contrasted to the ferromagnesian minerals. Plagioclase has only one-third of the expected magnetite agsOCiation. The quartz and K-feldspar have very nearly Association Coefficient C) .p fl 4) O --( m #11 J_l O O O N H N ”II C: «Is 0+) E-H l f: 00 4) #8 U (US ml 3.9. S ’U N SN I: HJJ O H 2 mm 0:3 I GO 6'3. '0 (2'0 22 nm as O I: 5". O 01 N .p 34 I'd :5 O I“: O 0" H m 0.. U) I M W] O m 30 O '? aouaxznooo go Aouanbaig \- / O O O 20 Frequency of occurrence of the association coefficients of the con- stituent minerals by rock type for magnetite. Figure 2-13. 64 the expected associations of magnetite. The distribution of AC values for hornblende shows no overlap with the iron- free silicates. The spread of hornblende AC values is the greatest of the constituent minerals. The biotite displays a reasonably compact distribution which has twice the ex- Pee ted proportion of magnetite. There are three histograms in the monzonite group Whi ch show noteworthy contrast with the quartz monzonite. The monzonite plagioclase has, on the average, twice the all3l<>l1nt of magnetite associated with it as in the quartz monzonite. A change is noted in the association coeffi- Ci-ent of K-feldspar from a slight excess in the quartz mon- 20111. te to a slight deficiency in the (monzonite. The average association coefficient of hornblende in the quartz monzon- ite is roughly twice that observed in the monzonite and the: association coefficient of quartz and biotite are approximately equal in the two rock types. The proximity of the magnetite and ferromagnesian silicates to each other as indicated by the association coefficient shows that iron is concentrated into micro- zones, It is a problem at times trying to decide which came first, the ferromagnesian silicate or the magnetite. The inclusion of one mineral in another is not diagnostic that the inclusion came first as supported by the pheno- menon of exsolution. Exsolution in the normal sense has been recorded for the ferromagnesian-magnetite assemblage, but it is not common and the textural relationships 65 observed during this study do not indicate exsolution. The questions then posed are, did the magnetite grains, whicl'i in many cases are included, serve as centers for ferro-silicate growth by being sources of Fe, did the Inagrie‘lzite and ferromagnesians form together, or did the mag‘l'Aetite form from the ferromagnesians? Ferromagnesian mineral grains hosting several magnetite grains (a common occurrence) are not likely to have formed by the disso- ciation of magnetite grains. The idea that the magnetite and ferromagnesian min- erals formed contemporaneously will receive special atten- tion- The relationship of ferromagnesian silicates and 11°11--‘l::l.tanium oxides has been studied by Carmichael (1963, 1967 ) and Carmichael and Nicholls (1967). Their main in- t . gr eat is the effect of ferromagnesian compositions on '31an of precipitation of the magnetite and vice versa. cza‘t‘ienske (1970) focuses attention on the possibility of :§%eute formation as a product of ferromagnesian mineral P::‘lcdown to iron free silicates under oxidizing conditions. role of oxygen in the formation of iron-titanium oxides '11 b‘ctly precipitating from the melt is well known (Budding- he ‘1 and lindsley, 1964). The AC results and petrographic observations suggest “1 ‘t a portion of the magnetite formed during the incon- 9:.- “Qnt reactions of the ferromagnesian silicates. Instead Q 3 having the end product an iron free silicate which V Q‘lld imply highly oxidizing conditions, it is suggested 66 that the reactions may be somewhat less extreme: I]. ) Orthopyroxene + 02 —-> Clinopyroxene + Magnetite 2 ) Clinopyroxene + 02 --> Hornblende + Magnetite 3 ) Hornblende + 02 9 Biotite + Magnetite. A3. 1 three reactions require an oxidizing atmosphere. The Fe would be available from the solid solution of ( Mg+2, Fe+2) and/or (Mg+2, Fe+2, Al+3) in the pyroxenes and from (Mg+2, Fe+2 A1+3, Fe+3) in the hornblende. A Portion of the Fe+2 from the silicates has to be oxidized to Fe+3 in order to create magnetite. It is to be noted that HZO is needed for both the reaction of Clinopyroxene to hornblende and to produce the oxidizing atmosphere. CZamanske did note a decrease in Fe/Fe + Mg ratio in am- pI'lj—Iboles and biotites in his work, though he did not attri- 15>th this to a ferromagnesian to ferromagnesian mineral reection. Unfortunately no chemical analyses are avail- ‘bJ-Q from this investigation to determine the change in ‘6 Fe/ (Fe + Mg) ratios from the augite to the hornblende t e ~‘zihe bioti te. The suggestion that magnetite formed by oxidation of mg orthopyroxenes is speculative since there is no evi- dgnce of orthopyroxenes. The reaction of augite to horn- 131‘Qnde is seen in the monzonite as well as the change of I‘°t‘hblende to biotite, though this latter reaction appears tQ be subordinate. These reactions could well explain the 67 high association of the magnetite to the ferromagnesians. The greater amount of magnetite with the hornblende than with the biotite suggests that more of the magnetite formed when the hornblende crystallized than when the biotite formed. Magnetite once created would be stable and remain in its relative position in the rock, provided F there was little movement in the magma. If the greatest portion of the biotite formed by incongruent reaction of hornblende under oxidizing conditions, a higher associa- [ tion of magnetite to biotite would be expected as a result of "inheriting" magnetite when the hornblende was consti- tu ted. Because the AC of hornblende is three times that of biotite, it is reasonable to conclude that much of the biotite formed by direct precipitation as contrasted to 03:1 dation reaction. A number of thin sections show significant hydrother- Ina; alteration of the hornblende resulting in the release of magnetite. It appears that sites with AC values greater mar: 6.0 (the quartz monzonite) represent areas of late at‘ge alteration of the hornblende. In thin section b10- ti ta does not appear to have been as susceptible to altera- tigl‘x as the hornblende and its AC histogram reflects this o.bagrvation. The observations of the thin sections and the AC re- gults suggest that the magnetite has formed by three pro- Q‘fises. The amount of magnetite associated with the non- £§ll‘romagnesian mineral is that which has apparently formed 68 by direct precipitation. A portion of the magnetite asso- ci ated with the ferromagnesian mineral is also a result of direct precipitation. However, the volume of magnetite associated with this group which exceeds their modes is a ttributed to oxidation and hydrothermal alteration. For the mineral biotite, the average AC value of each rock type is about 2.0. Since biotite does not show appreciable hydrothermal alteration, the excess association of magne- tite to biotite is interpreted as due to an oxidation re- aCtion. Hornblende shows a bimodal frequency distribution Peak which is believed to represent the association arising from alteration of the hornblende late in the intrusive's history. Average values of 5.0 and 7.0 for the quartz mOnzonite and 3.0 and 5.0 for the monzonite can be esti- Ina ted from the histograms of Figure 2-13. The lower of the twO values were used to calculate the volume of magnetite C: e ated by the oxidation reaction and the higher value to 6‘9 termine the magnetite formed by alteration. Results of thQ excess magnetite associated with the ferromagnesian minerals are shown in Table 2-4. The results reveal that in both rock types about one- fourth of the magnetite can be attributed to oxidation reactions during the course of crystallization. In areas whfire there has been hydrothermal alteration, about one- t§hth of the volume of magnetite in the quartz-monzonite and one-quarter in the monzonite is thought to have formed by alteration. In the unaltered rock the largest portion 69 1 . rt?" III!!!“ I suspends: mnooxm scanned cmwmocmmaouuow one and: coumHoonmm ouauocome umooxo mo meadow vm on em vN m oudcowcoz mm mm Cd ON 0 ouwcoucoz wukmso cowumuouad oozedncuom macadncuom ovduofim cowuwofixo coaumvwxo cowumuoua¢ cowuocwxo Hmuoe Hmuoa .ucoucou ouwuocmms Hausa mo ucoouom ca namuocua .¢IN OHAmB 70 or approximately three-fourths of the magnetite is inter- preted as being from direct precipitation. In rocks that have been altered, approximately one-third to one-half of the magnetite in the quartz monzonite and monzonite re- spectively has a genetic association. F“- In a broad sense the histograms represent an encoding of the intrusive's cooling history. Comparison of the histograms of different rock types, or even within a unit, can help identify dissimilar cooling histories which might be too subtle to qualitatively detect. With the aid of observations made during the petrographic study, it is believed that a reasonably clear picture can be drawn of the crystallization sequence of the two primary rock types wi thin the Melrose Stock. The quartz monzonite will be considered first. Figure 2‘1 3 illustrates that the magnetite has only minor asso- c'fi-a-‘lzion with the plagioclase. This arises when the plagio- claae crystallizes as one of the earliest formed minerals. IiO‘Wever, there are some inclusions of magnetite within me plagioclase indicating that a portion of the plagio- claae formed after precipitation of the magnetite. The I<‘:E§ldspar, excluding the phenocrysts, was constituted late in the rock's history for it is primarily interstitial and QQIIaequently is poikilitic to much magnetite. The numerical results support this point, because the association coeffi- Q1Grits are greater than 1.0. Surprisingly, the quartz aDiaears to have formed before much of the K-feldspar for '71 :i. 1:: is usually equidimensional and not relegated to an in- terstitial position. It is suggested, in view of the AC values for quartz and textural observations of the thin sections, that the quartz and magnetite were crystallizing concurrently, but the magnetite exhausted itself before flue quartz. From the evaluation of the excess magnetite associa- 1t:;i.<>n of hornblende, it appears that a portion of the mag- netite formation can be correlated with the commencement and cessation of the period of hornblende formation. Textural relationships indicate that the hornblende formed ‘53ialacrly'in the crystallization sequence. Biotite for the most part formed subsequent to the hornblende. The association coefficients and textural relation- 8111 ps suggest a slightly different paragenesis in the mon- zonite. The plagioclase in the monzonite shows a closer asscaciation with the magnetite. This feature is considered to arise from the magnetite crystallizing contemparaneously “'dl”‘=]ba the plagioclase to a greater degree than in the quartz no“ Zonite. A slight deficiency in the association of mag- l‘13‘321te with K-feldspar could have resulted from the K-feld- Spar precipitating at a higher temperature. The relative auuh‘>\ant of magnetite associated with the hornblende is less ‘it‘Eili in the quartz monzonite, although in absolute volume “‘9 opposite is the case. The precipitation of biotite Elt‘fi quartz with magnetite is believed to be nearly the 3%e as in the quartz monzonite. 72 The two lithologic units do not have great dissimi- larities in their cooling histories since they do not show a sharp field contact between them and both display similar opaque assemblages. A Chi-square test of the AC assembl- ages of the two rock types shows a 25 percent chance of a real difference which is, of course, not significant. I 1:: is not believed that this invalidates suggesting the above minor differences. It is postulated that the main difference of the plutons was in their p02 which was higher in the quartz monzonite as evidenced by the greater asso- <21 ation of magnetite to hornblende. This can very reason- ably be explained by a higher concentration of H20 in the quartz monzonite. In addition the smaller range of AC a"ezlrages in the monzonite would indicate that it had a shorter crystallization history than the quartz monzonite. The foregoing interpretation of the association coeffi- cietats of magnetite with respect to the constituent miner- 318 and the petrographic relationship of the constituent minerals suggests a sequence of crystallization as illus- trated in Figure 2-14 for each rock type. The graphical rePresentation of the magnetite crystallization includes tine relative amounts of magnetite formed by direct pre- cipitation, oxidation reaction and alteration. The separa- tiQh in the K-feldspar bar indicates early formation of the p1”‘enocrysts. Table 2-5 gives the numerical values of the Inagl'letite formed by the three processes as depicted in Figure 2-14. 73 0 .n T— 3: ' I H 0 U l O o ' s: F" 3 v! I | ” vi U ’— H g T m '0 o H _- G 0 N 00 H 4" a 5‘ m “ = +3 :3 3 L '2 t N II! D. H m . 0 .p H m (U . t; H 3. '5 S '4) III e '3 0' ' o 3 o m .c 9 >2 — I 9 l... , 2 L—J k [—- n «9 § ' ~34 2: ' ”-a .9. F" H g 1 2' z . U :4 Q N F. +’ I 3 .n D-o a a I: I} I 'H o :3 m “ o o g g _J o 5 o -H '9 g g u... o “.31 an Ms 5 3. t__. ' a. 8 'g o h— ' g B E§§§§ U Direct Precipitation ‘\\\‘ Alteration 7/; Oxidation Reaction [3 One unit square equals 5% of magnetite volume Figure 2-14. Proposed crystallization sequence of the Mel- rose Stock and the mode of magnetite formation based on association coefficients and petro- graphic observations. 74 HGGOHOQ m #mOHm¢Z 08.1 me on mm evanescox mo mN oat opacoucoz uuumso codpmufimfiUGHm cowuommm uumhwn :Ofipmcfixo coavmuouad .coapmpwmfiooum voouwp can .GOauomoH cOHumpwxo .coaumuovam an poeuom scoucou ouwuocmme kuou on oufiuocmms ucouumm .mlm canoe 75 2 - 5 The Interstitial-Inclusion Index The interstitial-inclusion index (III), which is an outgrowth of the association study, measures the average number of mineral species about a magnetite grain. It does not indicate the number of grains bordering a magne- ti te grain. Higher values of the index indicate an inter- 9 31:1 tial tendency of the magnetite whereas lower values Suggest an included nature of the magnetite. The lowest ‘1 Value of the III (all the magnetite is included) is 1.0 and the maximum value is 5.0 (each magnetite is coordin- ated by the five constituent minerals). The III obviously Cannot give the true number of species neighbors since a thin section is limited to two dimensions, but on a rela- tive basis the parameter can be of value. Figure 2-15 is a compilation of the frequency of occurrence of the 111's for the monzonite and quartz mon- zonite. The histograms are quite similar in range of values and an apparent bimodal distribution. To gain an insight into the possible cause of the apparent bimodal distribu— tiOna, those sites with III) 2.2 and III< 2.0 for the quartz monzonite and III< 1.9 and III> 2.1 for the monzon- 1te are plotted in Figure 2-16. The reason for eliminat- ing the intermediate values at this point is to help delin- efite possible patterns. There does not appear to be a strong spatial pattern, though there may be a slight cluster- ing of low values on the west side of the stock and high v‘Jmes on the east. The results of Figure 2-16 prompted 76 12 _ Quartz MonZOnite :1 ‘ : III F' . . . . . . . lgilxre 2-15. Distribution of the interstitial-inclusion index for the quartz monzonite and monzonite rock types. 77 I; 0 Ban R 7 40' 20' \ ’3 l 114' 35' R658 R663 " A \:~‘ \‘ __§_ + 3 + . ~ , G . '+ ’3 N o . c 9 + + 0 + 0 e .+ ' + . . + o T29N R66E miss new ' ‘Mi'les j \ 114'35‘ ° 5‘ g’ 1 \ Monzonite sites with III<1.9 o Quartz monzonite sites with III<2.0 4-{ Monzonite sites with III’2.1 Quartz monzonite sites with III>2.2 F:lgure 2-16. Interstitial-inclusion index spatial distribution. I my 'h Tu equiv 78 plotting of the 111's as a function of site elevation for each rock type (Figures 2-17 and 2-19). The III increases with elevation in the quartz mon- zonite, though there is much scatter. There are three possible origins for the direct relationship of III to elevation in the quartz monzonite. A change in grain size of the magnetite with respect to elevation within the magma chamber can give rise to a different coordination capacity With the constituent minerals. If the magnetite increased 1) in size with elevation then there would be a greater pro- bability of having more minerals coordinated with it. This aSSumes that the constituent minerals remain roughly con- 3tant in size with elevation, an assumption which appears to be true on the basis of observation. The percentage of grains whose size is greater than 50 microns (data from Figure 2-5) indicates the size to be increasing with de- creasing elevation. This contradicts the consideration of grain size as being the influence on the variability °f the III. A second origin for the direct relationship is that the trend may suggest the degree to which the magnetite formed by oxidation reaction and direct precipitation. Maghetite forming by an oxidation reaction would have a greater likelihood of being included and hence a lower III. conversely, magnetite forming by direct precipitation is "‘Ore likely to be pushed to an interstitial position and °°hsequently display a higher III. However, there does 79 .cossam ewcoucoe uuumsw on» canvas cowum>oao spas xopcu cowmsaocwlamwuaunuouca osu mo cowumwum> .thN shaman poem .cofium>oam comm oomh OOVh OOOB 0000 Done 1 a d . 1 1 q _ a L m...” 0 \ \\ e u\\ \ x x x .. m; . \ O O O \.\ \\e\ o o \\ I. . xx x xx .. o.~ \ \\ C \O\ \ L \\ \ \\ e \ \ l NoN O \ 0 $00 \ O \ \ \ I. \ \ \ \o \ \ \\ I. *ON \ \ O III 80 appear to be a correlation of the hornblende associa- tion coefficients with elevation (Figure 2-24). This figure indicates that the III should take on lower val- ues due to greater magnetite-hornblende associations. Either the assumption for this argument is incorrect or there is another factor predominating over the associa- tion affect on the III. As a third possible origin, the trend of Figure 2-17 may reflect a viscosity variation within the cooling magma. In the early stages of crystallization (high in the chamber) the magma may be less viscous due to a slightly lower 81 lica content and the resulting more mobile melt would permit the magnetite grains to be pushed to interstitial positions. In the later cooling history, deep within the magma chamber, the viscosity would be greater and the mag- netite grains would be less easily moved aside and more likely to be included. This idea is supported by the gen- eral increase in quartz content with lower elevations as shown in Figure 2-18. It is assumed that an increase in 81lica content is reflected in the quartz content, an index of the relative viscosity of the magma. 0f the three explanations considered for the change in the amount cf included magnetite, the viscosity factor fits the observ- ed data best and is thus favored. However, this may be a sllxnplistic solution to a rather complex problem. The monzonite histogram (Figure 2-15a) is shifted 811~9htly to the left of the quartz monzonite histogram 81 .cousum ouwcowcoa nukesv on» cusps? cowam>ouo saw: ucoucoo nunssv mo coaumanm> comb poem .cowum>oam .masm museum cosh coho comm q 4, . 4 q, 4 . . . . q ms . .. . . . .om . . . .. H. “mN . . ion L :5: % ‘quenuoa zizenb 82 of the same figure. This minor difference may reflect the smaller grain size distribution of the magnetite within the monzonite (see Figures 2-4 and 2-5 and Table 2-2). Furthermore, the results presented in Table 2-4 indicate that a significant amount of magnetite was formed by alter- ation and peritectic reactions, processes that will tend to decrease the III. Therefore, both the mode of forma- tion and the grain size distribution may be the cause of the slight downward displacement of the peaks of the mon- zonite histogram. However, it is thought that the (quartz- poor) monzonite pluton, would have a lower viscosity than the quartz monzonite, thus suggesting that the III should be higher in the monzonite than the quartz monzonite. If this is true, then the above two effects more than offset the viscosity factor. Figure 2-19 shows the relationship of the III values "it): elevation within the monzonite pluton. There is no direct relationship of the parameters as evidenced in the quartz monzonite. It does appear, however, that there is a concave upward trend of the data. If viscosity was a controlling factor in the quartz monzonite pluton, then it should also apply to the monzonite pluton. The viscos- ity as shown above reflects the depth and in turn, something °=E the geometry of the magma chamber. The high elevations in the quartz monzonite pluton indicate an ascent towards the roof or a margin of the intrusive. Similarly this can be postulated for the monzonite pluton. The lower portion 83 .cousam muwcoucoe was canvas cowum>oao nuwz xopcw cownSAUCHIHmwqumHouCfl on» we cowumaum> .mHIN ohsoflm poem .cowum>0am comm comb ocwb coon comm -) 1 n 1‘ d d u q We.” \\ I O \ O O, L \ . I ma \ I \ I \\ I O \ I l . I s l \ \ . I C \ ' / o \\ (I AON \s I I 1 NON I XSPUI “DISTIIDUI-I‘BI‘I} 1381940 I 84 of the concave trend of Figure 2-19 is believed to repre- sent, by analogy, an area of the monzonite where the quartz content is higher (the magnetite is preferentially inclu- eeci). The highest and lowest elevations in the monzonite show the magnetite to be interstitial reflecting a lower quartz content. It is speculated that the monzonite pluton is a len- 11ke mass upon the quartz monzonite. The margins of the monzonite pluton are thought to be those sites which form the low grouping of the data in Figure 2-20. These are the sites having a low quartz content and are thought to have crystallized first. The upper group of data are lbealieved to represent the interior of the monzonite pluton where the quartz content is higher. Figure 2-21 is the Proposed model of the monzonite pluton. It is conjectured that the monzonite pluton was em- Placed before the quartz monzonite pluton. The margins Of the monzonite pluton may have a lower quartz content than the center portion due to the normal enrichment of quartz as crystallization proceeds and perhaps due to de- 31 licification of the magma by the host rock. The loss of cImmartz due to desilicification is probably not an impor- tQuit factor because there is only minor calc-silicate cOntact metamorphism and silicification of the host rock. The monzonite pluton was not completely crystallized at nae time of the quartz monzonite emplaced considering the gradational contact of the two plutons. The quartz 85 .cousam onwcowcoa 0:» cans“: coaus>oao nua3 acoucoo «sumac mo coauewum> .o~:~ shaman poem .cowus>oam oocm ooom cows come comm - .4 e A d a A a O . L N O n E . I 1 w nu . 3 o u 1. a o u :1, n m . % o O l m 86 I” . A . I” ,I'\ ‘ ( ' \\ . , I ’L ' \ L ” p’ (e I \ e o ” 0 \" g . u I, 0 H \\ . v . . 0 ‘ T291: T29N R66E \T29N 40' 20' . R658 T288 Mil“ 40- 20' _ I r I T W \ . o 1 114 35 o A‘ g’ \ Monzonite A I":‘Lgure 2- 21. Quartz Monzonite AI H - High III values L - Low III values Diagramatic sketch of the high and low interstitial-inclusion index zones within the monzonite pluton and the interpreted cross section of the pluton. 87 monzonite pluton was emplaced below and to the marginal side of the monzonite pluton. _2_.fi 6 Magnetite and the Ferromagnesian Minerals The study of the association of magnetite with the other minerals has shown preferential association to the ferromagnesian minerals and furthermore, this association is more common in the quartz monzonite than the monzonite. A closer examination of the relationship of the ferromag- nesian and magnetite contents may shed further light on the cooling history. In Figure 2-22 the relationship of the total ferro- magnesian content to the magnetite content is shown with respect to rock type. No particular pattern seems appar- ent for the monzonite, but the quartz monzonite, on the other hand, has a general inverse relationship of the two cI'--‘l<‘=alntities. A decrease in magnetite content is accompan- ied by an increase in the ferromagnesians. This suggests conditions existed which favored greater amounts of iron in the oxide structure in some cases and greater amounts in the silicate structure for other cases. The atmosphere Of the magma chamber, as determined primarily by the par- tial pressure of oxygen (p02), controls the structure in which iron will ultimately reside. An oxidizing atmosphere in the magma chamber would create magnetite and therefore 1°33 iron would be available for the ferromagnesians. In a reClucing environment the iron would find its way into 88 .omhu #00» cu mcapuooum ucoucou anamocmssouuom on acoucoo ouwuocmos mo Qanncoausaom .-a~ shaman & .ucoucou csanocmmsouuom 0* on ON cH mil! . u 1) 4 a I.) 1i o l .. W s0")! 2 . sII .. Inna e . e I cuflCONcoz a o . so o Is a? .I H a 3235: + x . .. .2. . n. I. O 0 II .17 I s s I l a +1. 2 3 O O a I J. II . "Cu .r lib e\\ n 3. +s. + - L a + + N u .v + + 1 3. + l + 1 d» l .9 + a m 89 the silicates. If the initial p02 was high, then the magnetite should have little association with the ferromagnesians, but if the p02 increased later in the cooling history, then we might expect a high magnetite-iron silicate association. A graph of the association values of the hornblende within the quartz monzonite with the ferromagnesian content can help to clarify this speculation (Figure 2-23). Disre- garding the data that are high due to hydrothermal altera— tion, the remaining data exhibit an inverse relationship. As a result, it is believed that the magnetite-ferromagne- sian inverse association arises from a loss of iron in the breakdown of the ferromagnesian minerals. The oxida- tion reaction which yields magnetite gives a high associa- tion of magnetite to hornblende at the expense of the iron B111.cate content. The variation of the hornblende association coeffi- cient with elevation suggests that more magnetite formed by an oxidation reaction at higher elevations (Figure 2“-24). Thus, it is concluded that the upper part of the ll“Roma chamber represents a more oxidizing condition than the more deep-seated interior throughout much of the crys- tallization history. The concentration and migration of the volatile phase upwards during crystallization and assimi- lation of host rock could create the proposed oxidizing e1'lvironment. Figure 2-25 shows a breakdown of the ferromagnesians 90 I” ' \ \ 1° F’ Due to hydro-\ ' \ ‘/ thermal altera-‘\ ‘\ p tion . \ . \ \ \ \ \ 8 + \ \ \ \ 4) __ \ ' ‘ 5 P I’v \‘\\ \\ .: H ‘. e ‘o "' .3 6 \ e. \\ D \ \ 31 \ . \\ \ 8 * .\ . O U \\ ' \ c ~. ’ ‘ \ -3 “ ' \\ ° ' ' \ u \\ . . 6 so g D \\\-"” 3 2 C u- 0 1 4 4 L a 1 j 1 1 n 4 8 12 16 20 Ferromagnesian Content, % Ffiigure 2-23. variation of the hornblende association coef- ficient with ferromagnesian content within the quartz monzonite pluton. 91 7.0 - ” I I e , .l I- . I ’ I 6.0 I. ’ I u I ’ 5 , -' ‘06 I- I I '6'. ’ ’ m I . ' / 8 5.0 ’ I o. I U ’ /. 8 I , b 0 ‘g / / on ’ C 8 4.0 P ‘ O I a ‘ ‘I a: ’ , I p . I / 3.0 - I’ / 1 1 1 J 6000 7000 8000 Figure 2-24. Elevation, feet The relationship of the hornblende associa- tion coefficient with elevation within the quartz monzonite pluton. Excludes data of sites showing appreciable hydrothermal al- teration of the ferromagnesian minerals. 92 .omxu soon on avenues suaz scoucoo cmanocmmeouuom scouuoa ou moccancuos acouuem msmHo> scoucoo ouwuocmmz .mmlm shaman mcmwnocmmeonuom x\0pcoancuom x c.H m.c moo v.0 Noc q a d d u m e a» J n O . w e Q 0 ‘ b o u L. o e. o e j H M“ e e 0 on T. O . q o o o. 8 + o 0 My .7 u a. o... 6 1. + + + 4 N m + + +1. :1. + + up ouwcoucoz .o ... ++ ‘ ouacowcoz + o m 93 as a normalized ratio of hornblende to total iron sili- cate content plotted against magnetite content. The ratio of the hornblende to the total ferromagnesian content is greater in the monzonite than in the quartz monzonite. This variation probably represents a higher calcium con- tent in the monzonite due to either an initially more calcic magma or a greater amount of assimilation of carbonate sediments. Hornblende and augite are able to accommodate more Ca than biotite in their structures and are probably more likely to form as a result of the dissociations of limestone. This is substantiated by the observations made of a thin section of a xenolith (originally limestone?) adjacent the limestone contact (Site 11). The hornblende content in the xenolith is more than three times that of the surrounding host rock. The magnetic susceptibility of the xenoliths is from 50-100 percent greater than the host rock. In addition, a thin section from Site 18, previously not mentioned, which is a diorite-dike material intruded into the limestone near the contact with the monzonite, shows 60 percent plagioclase, 30 percent horn- blende, 8 percent K-feldspar and the remaining as accessor- ies. Clearly this has been modified by the surrounding limestone. A check of the sites at which xenoliths were noted reveals that 14 of 16 were in the quartz monzonite. Even taking into account the greater number of quartz monzonite sites than monzonite sites, a disproportionately large 94 number of xenoliths occur in the former lithology. It is suggested that the more basic monzonite pluton was able to more completely assimilate the xenoliths than the more siliceous pluton. The evidence presented in this section indicates that the magmas were modified to some extent by the surround- ing host rock. This may account for some of the higher magnetite content in the monzonite in contrast to the quartz monzonite. Chemical analyses are necessary to determine the extent to which assimilation may have modi- fied the magma and provide information on the iron content and its distribution between the oxide and silicate phases. CHAPTER III MAGNETIC SUSCEPTIBILITY 3.1 Introduction 3.1.1 Purpose Variations in magnetic susceptibility among geologic Ibodies permits this physical property to be the basis for the magnetic geOphysical method. The expression of the :magnetic susceptibility through the magnetic mineral is a .sensitive indicator of past events. Then it follows that tflae examination of the magnetism of a geologic body, the determination of the association coefficients and inter- stitial-inclusion indices, and the investigation of the INetrologic properties of the magnetic minerals can reveal idiformation leading to a more complete knowledge ofgeolo- gic bodies. The magnetic susceptibility of the Melrose Shock has been investigated in order to determine its de- Pendence on lithology, and the intrusive's geometry, its el'tpected and actual variations, and the manner in which it can be analyzed to obtain a representative susceptibility. ’Phe magnetic susceptibility was investigated by measurements in situ and on cores. 95 96 3.1.2 Sources of Variation The measurements of magnetic susceptibility are sub- ject to both random and systematic variations. A basic assumption is that the magnetic susceptibility of a rock unit is deterministic. Variations in susceptibility are due to the primary distribution of the magnetic minerals in the rock. The susceptibility will be the summation of the discrete sus- ceptibilities of each magnetic constituent (assuming negli- gible magnetic interaction for magnetite contents of a few percent or less). The measured variation or dispersion in the susceptibility is a function of the size and number of samples and their homogeneity. If a sufficient number of data are available, then it should be possible to deter- mine whether the susceptibility displays a normal or Gaus- sian distribution or a logrithmic distribution in addi- tion to a mean value. It has been found that a logrithmic distribution best fits the scattered data common to rock magnetism (Runcorn, 1967). The primary distribution of magnetic minerals can be subsequently changed by secondary chemical and physical processes. Mechanical weathering, such as the breakdown of rock by frost action, probably does not directly affect the magnetic properties, but it does help to promote more rapid chemical weathering. Weathering reduces the sus- ceptibility as a result of the relatively nonmagnetic oxides forming at the expense of the magnetite. Deuteric 97 and hydrothermal alteration can also change the primary susceptibility. Hematization of pre-existing magnetite lowers the susceptibility, whereas breakdown of the ferro- magnesian minerals can increase the susceptibility with formation of additional magnetite. 3.1.3 Measurement Variations Scatter in data can result from instrumentation, the number and size of the samples, and their manner of sel- ‘ection. The dispersion introduced into the data from the repeatibility of the susceptibility bridges is less than one percent standard deviation. The least weathered por- tions of the outcrop surface were selected for in situ measurements and coring. This would obviously eliminate some dispersion. The collection of several measurements at each site makes recognition of anomalous data easier than when only few data are available. In addition, unusu— ally low in situ readings were rejected in the field. Six to eight in situ measurements and four to fifteen core specimen susceptibilities were obtained for each site. Systematic susceptibility due to calibration errors in the instruments used for the in situ and core measure- ments was eliminated by measuring approximately 100 uncut field cores (3 inches length) with the internal coil sys- tem of each bridge. The mean of the differences for each specimen was used to determine a correction factor. The core specimens measurement were reduced by 5 percent. 98 The dispersion of magnetic minerals in a rock poses the problem of determining a representative value of mag- netic susceptibility. It would be expected that one mea- surement of susceptibility on a rock unit would not suffice, unless the measured volume was very large. Since a sample volume is limited, the number of measurements must necessar- ily increase. The two methods employed in this study will help to establish necessary guide lines for determination of magnetic susceptibility. The measurement of susceptibility was performed on two different size volumes, 0.70 cubic inches, the core specimens and 57 cubic inches, the effective hemispheri- cal volume measured by the in situ coil. The ratio of these volumes to each other is approximately 80 to l. The large difference in the volumes measured by each technique allows the evaluation of the heterogeneity of the rock units within the scale of the specimens. If homogeneity of the rock unit exists at the 0.70 cubic inch volume, then the dispersion in the data between the two techniques should be approximately the same. If the in situ volume is nearer the level of magnetic homogeneity, these data will show less dispersion than the core specimen data. And lastly, if both methods give wide dispersion, it can be inferred that the rock unit is heterogeneous on a scale greater than the in situ volume. The effective depth of measurement for the core and in situ methods is 3 inches. The in situ technique does 99 penetrate deeper than 3 inches, but the contribution is not appreciable. The specimens permit observations of the susceptibility with depth from the outcrop surface, however, no systematic increase of susceptibility with depth was noted within three inches of the outcrop sur- face. This observation does not lead to an insight into the amount of dispersion attributable to weathering. 3.2 Discussion of Results 3.2.1 Normal versus Loqrithmic Distribution The individual in situ data were qualitatively anal- yzed to determine the type of distribution of the measure- ments. In the following sections, the analyses of the data assume a normal distribution of the data and that assumption is verified. The wide scatter of values of magnetic data in many cases best fits a lognormal distribution rather than a normal distribution, this scatter ranging as much as sev- eral magnitudes (Irving and others, 1966: Runcorn, 1967). However, the data of this study does not show a wide scatter and hence suggests that the distribution may be normal or Gaussian. The plotting of frequency distribution of lognormal data on a linear scale gives a graph which is skewed towards increasing values of the abscissa. The same data plotted as the logrithim of the data will display a bell-shaped curve. On the other hand, plotting data possessing a 100 normal distribution on a logrithmic scale yields a curve which is skewed towards decreasing values of the data and, once again, plotting it on a linear scale yields the bell- shaped curve. All of the individual in situ measurements collected are plotted, according to rock type, on both linear and logrithmic abscissa histograms as shown in Figures 3-1 and 3-2. The data from the monzonite pluton (Figure 3-1) appears to be skewed slightly to the left on the lognormal plot and symmetrical on the linear or normal plot. The same respective shapes are noted for the data from the quartz monzonite pluton, although it is less marked than in Figure 3-1. The results shown in these histograms supports the contention that the distribution of the susceptibilities is normal. This places greater confidence in the mean- ingfulness of the arithmetic mean and subsequent analysis of the magnetic susceptibility. 3.2.2 Inter-site Variations This section is concerned with the analysis of the data, determining which are reliable, which should be discarded and how many data are necessary to establish a representative susceptibility. To this end, the six to eight magnetic susceptibility means at each site were averaged and the extreme values, the lowest and highest, are plotted against the site mean as in Figure 3-3. Thus, 101 50 , 40 b 30 - 20 b 10 b 102 ////////A /////z \\ Figure 3-2. \\ F 1 2 3 4 Susceptibility, 10-3cgs units Distribution of individual in situ mag- netic susceptibility measurements in the quartz monzonite pluton on linear and logrithmic scales. s- 4 ‘P m 4.) --i c a m .3? u m I 0 ,4 >1 +: 2 - vi H -a .0 H .p o. m 8 1 a U) 0 103 Intercept a 0.18 Slope = 1.06 Std. Err. ESt. :3 0.3.8 0 Cor. Coef. a ' Intercept =-0.27 .e . Slope = 0.95 0 Std. Err. Est. a 0.29 Cor. Coef. = 0.93 00 \ Anomalous and Rejected Data I I l I Figure 3-3. 1 2 3 4 s Susceptibility, 10‘3cgs units Site susceptibility extremes versus site means of in situ measurements. Unity slope line separates maximum amd minimum values. Included are the intercept, slope, standard error of the estimate, and correlation coef- ficient for each of the two data groups. 104 each site is represented by two points, one each above and below the line of unity slope at the value of the site mean. If the data are evenly distributed about the mean, the distance of the two extreme points off the unity line will be the same. In this manner the dispersion can be generalized and the sites which obviously have anomalous data are easily recognized. To delineate data which fall noticably outside of the two linear groupings paralleling the unity slope line, linear regression lines were computed for each of the two data groups. Data falling outside of two standard error of the estimates (S.E.E.) from the upper data group were rejected. The pertinent parameters of zero intercept, slope, S.E.E. and correlation coefficient are given adja- cent to each data grouping as in Figure 3-3. The S.E.E., which is analogous to the standard deviation, allows a comparison of the scatter of the two data groups. 3.2.3 In Situ The S.E.E. of the upper in situ data group (Figure 3-3) was chosen as a reference for eliminating poor qual- ity or erroneous data because of its smaller dispersion than the lower group. With a 95 percent certainty that the data points rejected are not representative of the sampled population, points falling greater than two S.E.E. (0.36 3 x 10- cgs units) from the least square lines were re- jected (See Figure 3-3). Seven data points representing 105 6 sites fall outside of the acceptance limits. The site means were recalculated after discarding the anomalous data and the results are shown in Figure 3-4. The S.E.E.'s have been greatly reduced, especially for the lower group, and the two values do not differ appreciably from each other. It is concluded that the data at this stage rea- sonably represents a normal distribution with a mean being representative of the surface rock at the sites. A calculation based on the least squares lines from Figure 3-4 indicates that the percent difference between the extreme points near 4,000 x 10'6 cgs units is only one-half (20 percent) of the value calculated at 1,000 x 10"6 cgs units (40 percent). This indicates that the rocks of lower susceptibility show greater heterogeneity either due to primary origin or secondary alteration. Interestingly, there is a distinct change in the dis- tribution of points on either side of a site mean of 2400 x 10"6 cgs units. Below this value the data cluster near the unity slope line as contrasted to above where they are regularly farther away from the line. This value roughly coincides with the values separating the magnetic suscepti- bility of the two major rock units. The majority of the monzonite sites have a mean susceptibility of greater than 2400 x 10"6 cgs units while the quartz monzonite means are generally less than this value. The origin of the varia- tions in the extreme values may be either original homo- geneity or subsequent alteration. 106 5 F Intercept - 0.18 Std. Err. Est. s 0.12 .' Cor. Coef. - 4 P m 4..) -s c s m 8‘ 3 *- m I o H ‘ >2 .3.) I} 2 - .a h 3 Intercept --0.13 8a : : Slope = 0.93 3 Cor. Coef. c 0.98 m 1 1 1 l l _l O 1 2 3 4 5 Figure 3-4. Susceptibility, 10-3cgs units Site susceptibility extremes versus site means of in situ measurements after rejec- tion of anomalous data. Unity slope line separates maximum and minimum values. Included are the intercept, slope, standard error of the estimate, and correlation coefficient for each of the two data groups. 107 A number of the quartz monzonite sites show little» alteration, hence less dispersion, and would represent the data falling near the unity line. Some of the quartz monzonite sites are finer grained and therefore may show greater homogeneity and less dispersion. 3.2.4 Egggg The susceptibility measurements of the core speci- mens were subjected to a similar analysis. The origin- al data, Figure 3-5, show a much greater scatter than the data in Figure 3-3 as shown by a comparison of the S.E.E.. The same criterion was used to eliminate anoma- 1ous as in the in situ analysis. The recalculated data are presented in Figure 3-6. Even after rejection of more data than in the in situ method, the S.E.E. values of the core specimens are larger than those of the in situ method. This difference is a result of the two different volumes sampled by the in situ coil and core measurements. Twenty-nine percent of the core sites required removal of poor data as con- trasted to only 14 percent with the in situ sites. It is believed that greater inhomogeneities exist on the level of the core volume than on the in situ volume, which has led to greater dispersion in the data of the former. The in situ method apparently is able to "average out" many of the inhomogeneities which are observed by the smaller volume measurements. The small sample volume 108 5 r Intercept a 0.43 0 Std. Err. Est. - 0.34 Cor. Coef. - 0.94 ‘ S r . 6° . f} 4 ~ ' .4 s . 3 o o '8» - . 0 '. . O m 0 lo 3 _ O o 0 o 0 Pi . , >2 6 o . . 0 a , -° «4 . o ' Fl . o .3 . C .0 .0 '0'. 2 .- o . ° G) g. '. e 8 . ° ' - 0 g 0 Intercept --0.29 m ,. ‘ Slope - 0.83 1 '- Std. Err. ESt. B 0.48 0 Cor. Coef. - 0.84 0 0 O G h\Anomalous and Rejected Data 0 0 _1 l l J I 1 2 3 4 5 Figure 3-5. Susceptibility, 10-3cgs units Site susceptibility extremes versus site means of core specimen measurements. Unity slope line separates maximum and minimum values. Included are the intercept, slope, standard error of the estimate, and correlation coef- ficient for each of the two data groups. 109 6. Intercept = 0.36 Slope = 1.04 Std. Err. Est. a 0.21 ' 5 Cor. Coef. s 0.98 m ‘4' 4) "5 " g e... . . o , . , '1’ 3 r O ' 0 fl :. .e e. . >2 .. - ,' '0', a 0 fl : a e .. . ‘0 .o 2 ' ' .. "', w-l 4" 0 e . 8' . . ' Intercept =-0.ll g '. Slope - 0.87 a .. Std. Err. ESt. a 0.17 m 1 _ : Cor. Coef. - 0.98 O l l L l l 2 3 4 S Susceptibility, 10-3cgs units Figure 3-6. Site susceptibility extremes versus site means of core specimen measurements after rejection of anomalous data. Unity slope line separates maximum and minimum values. Included are the intercept, slope, standard error of the esti- mate, and correlation coefficient for each of the two data groups. 110 shows about 50 percent more dispersion than the large volume. This is based upon the separation of the linear regression lines of Figures 3-4 and 3-6 in the middle of the mean susceptibility range. The analysis clearly demonstrates that the core data is subject to notably more dispersion than the in situ data, which necessitates a closer inspection of the data, preferably analytically. The results of the foregoing analysis indicate the "poor“ data were almost always low values which are thought to be the effects of weathering. Weathering destroys the magnetite and consequently lowers the susceptibility of the rock. The in situ data do not appear to need as careful scrutiny as the core data. If volumes greater than the 57 cubic inches were measured, the amount of dispersion and anomalous data would be expected to decrease. The chances of collecting anomalous susceptibility data are inversely proportional to the sample volume. A test of the effectiveness of removing anomalous data comes from a comparison of the results of the two methods. The susceptibility values should display a nor- mal distribution and the means of the two methods should be equal. Figure 3-7 shows the resulting means of Figures 3-4 and 3-6 separated according to rock type. The range and overlap of the rock types for each method of measurement are nearly identical with less than 10 percent separating 11 H 15 P ¥\\ In Situ Means 10 ' \\\\ _. ~ m\\ ///- / m u c m ‘5 3°‘ 0 l 3 4 5 #4 0 >10. 6’ ‘\ :3 / 3' Core Specimen Means 5i- o. \L.. 1 2 3 4 5 Susceptibility, 10-3cgs units Z2 Quartz Monzoni te Monzonite Figure 3-7. Magnetic susceptibility distributions of in situ and core specimen measurements by rock type. 112 the respective means. The in situ histograms (Figure 3-7 ) show an approximate, normal distribution with a distinct separation between rock types. The core samples show less distinction between rock types, although the mean values agree reasonably well with the in situ method. This lack of a strong separation may, in part, reflect the smaller sample population of the core measurements as opposed to the in situ measurements. The in situ means and standard deviations are (1880 i.4°°) x 10'6 and (3130,: 640) x 10'6 cgs units for the quartz monzonite and monzonite respectively. The core specimen means and standard deviations are (2030 i 460) x 10"6 and (3400': 700) x 10"6 cgs units. A student t test shows the difference between the susceptibility means of the rock types to be highly significant for both methods of measurement. There is concern as to the amount of weathering the outcrop surfaces have undergone. The field generated by the in situ coil does penetrate the outcrop surface more than the effective hemispherical radius of the coil (3 in- ches). If the field does couple with fresh rock, then the in situ means would be expected to be slightly higher than the core sample mean. This difference is not observed. Also, specimens from a given field core do not show a consistent increase of susceptibility with depth. These observations suggest that no weathering has occurred in the top few inches of the outcrop surface or that there is uniform weathering of the top several inches. 113 In situ measurements were made on a few boulders which had apparently broken by rolling downhill. The values obtained at a distance of about one foot from the surface of the boulders were approximately 30 percent greater than the surface in situ values of the boulder. This suggests that there may be a uniformly weathered zone, perhaps a foot deep on the outcrop surface. How- ever, lack of detailed evidence on the boulder measure- ments precludes any definitive statement. Cores of sev- eral feet in length would provide the necessary informa- tion to resolve the problem. The following conclusions can be stated from the foregoing discussion. 1) Larger sample volumes lead to appreciably less dis- persion in the original data. 2) Larger sample volumes are less subject to anomalous data. 3) The delineation of magnetic rock units is noticably better with larger sample volumes than small volumes. 4) The approach of plotting site means and extremes pro- vides a way of recognizing poor quality data. 5) Two magnetically distinct units were delineated in the Melrose Stock. These are correlative to the lith- ologic units. The quartz monzonite has a mean magnetic susceptibility of 2000 x 10'6 cgs units, the monzonite, 3200 x 10"6 cgs units. 114 3,3 Sampling A fundamental problem in determining the suscepti- bility of a rock unit is knowing the number of samples and sites that are necessary for a representative value of the unit (e.g., Case, 1966). Sufficient data are avail- able from this study to suggest guidelines for sampling of intrusives similar to the Melrose Stock. To aid in the evaluation of the sampling procedure, matrices were constructed with the number of sites sampled as rows and the number of measurements per site as columns. The in situ data of 29 quartz monzonite sites were used to generate one matrix and 17 monzonite sites were used for another. The entries in a sampling matrix are suscepti- bility means based upon the number of sites and samples per site. All of the means in a matrix will approach the grand mean, the average of all the sites and samples. Tables 3-1 and 3-2 are sampling matrices which, instead of giving the various means, give the percentage deviation of the various means from the grand mean. The grand mean is considered a representative value of the rock unit. In this manner, the number of samples and sites can be determined which bring a particular matrix entry or mean to within a certain percentage of the grand mean. The entries of Tables 3-1 and 3-2 were derived in the following manner. Numbers for each quartz monzonite site were put into a box, shaken, and then drawn out randomly to obtain the necessary data for Table 3-1. This gave a 115 o T m+ m+ T H+ m+ m+ 3+ m H: T m+ v+ a: ~+ 3+ 3+ 3+ m 3 I'll.“ T N: v+ m+ T H+ m+ 0+ 2+ a m. Ill mu m: m: T. H+ H+ ~+ 3+ 5+ 2+ m m Trllll S Nu m: ¢+ m+ ~+ ~+ m+ m+ 3+ N T m: m+ ...+ ~+ N+ s+ m+ 3+ H mm _ mm _ ow _ ma _ 3 _ s h m _ m _ m nmmfimmo nonesz ‘ w .couSHm ouwcoucoa nuumsv «nu mo some Ucmum may Scum cowumw>ov psmoumm cw xwuume mcHHmEmm aufidwnwumoomsm usumcomz .Hlm magma 116 Table 3-2. Magnetic susceptibility sampling matrix in percent deviation from the grand mean of monzonite pluton. Number of Sites 2 I 3 I 5 l 7 l 10 [ 15 1 17 1 -1o -5 +6 -3 +1 +1 +3 u. 2 -lO -6 +3 -5 +2 0 +4 g 3 -8 -4 +4 -4 +1 -1 +2 '3 4 -8 ~6 +2 -6 -1 -2 +1 3.""“ to 5 -6 -5 +2 -6 -2 -3 ,_____., 6 -13 -5 +3 -5 -2 -3 117 random selection order of sites in the quartz monzonite and the sequence of data at each site were taken as recorded in the field notes. Originally the sites were selected to give a relatively uniform sample pattern over the intrusive. To arrive at the upper left entry in the Table 3-1, the first in situ readings at the first two randomly cho- sen sites were averaged, subtracted from the grand mean, and then calculated in terms of percent from the grand mean. The algebraic sign indicates whether the array entry was greater or less than the grand mean. As a further example, the entry for 7 sites and S in situ readings per site is determined by taking the average of the first 5 readings from the first 7 randomly selected sites. This mean is 6 percent less than the grand mean (Table 3-2). The results of the quartz monzonite measurements in Table 3-1 show a decrease in the deviation with increasing number of sites, but no trend is apparent from the number of samples per site. Similar results are shown for the monzonite (Table 3-2). These data indicate little ad- vantage is gained by increasing the number of measure- ments taken at a site. Thus, greater variation exists among the sites than within the sites. In the quartz mon- zonite the deviation is reduced to less than 10 percent of the grand mean with 7 sites in the rock unit, but in the monzonite only 3 sites are necessary to bring the mean within 10 percent. Part of the difference in the indi- cated number of sites can be attributed to the size 118 difference of the two plutons. The monzonite has an ex- posure of approximately 3 square miles and with 3 sites necessary to bring the mean within 10 percent of the grand mean, at least one site for each square mile of surface exposure is needed. The quartz monzonite has an areal extent of nearly 9 square miles, so the necessity of 7 sites suggest a collection density of just slightly less than one site per square mile. The two values agree rather closely with each other. The results suggest that the number of sites to be sampled in order to obtain representative magnetic sus- ceptibility values is in proportion to the size of the pluton, roughly one site to each square mile. The stated site density would also permit the distinction of major magnetic units within an intrusive which might be over- looked with just a few sites within the intrusive. Ob- viously, the greater the number of sites, the greater the probability of approaching the representative suscepti- bility of the unit. This study indicates little adVantage is gained by several measurements per site, although it should be remembered that in the preceding section several measurements per site permitted easier removal of anoma- lous data. There is little extra time required to take a few additional data at a site. More time is consumed establishing additional sites. The quartz monzonite showed less dispersion than the monzonite (See 3.2.3) which suggests the need for a greater 119 sample density within the monzonite pluton. However, on a relative basis the dispersion (reflected by the stan- dard deviation) of the two rock units is nearly identical. Hence, the guidelines as outlined above are consistent with the difference in dispersion. 3,4 Susceptibility and the Geometry of the Intrusive The presentation of the magnetic susceptibility data with respect to the geometry of intrusives has been limit- ed to profiles (Hay, 1967: Pothacamury, 1970). Sufficient data are available from this investigation to study the areal distribution of the magnetic susceptibility. The magnetic susceptibility of the Melrose Stock is contoured in Figure 3-8 irrespective of rock type. The dashed lines delineate the areas of relative aeromagnetic highs and lows. The area of highest susceptibility, of course, falls over the monzonite pluton. Within the mon- zonite zone, no real distinction or trends exist. In the 'quartz monzonite there is a general increase of suscepti- bility to the east. The lack of apparent correlation or response of the contours to the known contacts suggest little relation between the two quantities. This observa- tion is corroborated by Figure 3-9 which indicates no systematic trend in the susceptibility with respect to the contact distance for either rock type. Either there is no magnetic zoning related to the geometry of the intru- sive or the observed contact is not a good representation 120 T a. c; N 0. Figure 3-8. T 114' 35' RGSE R66E v 29H R663 \TZSN R673 T288 "11.8 40'20'._ r W t W \ 1 114' 35' ° " 3’ 1 \ ~.\ -__. .Aeromagnetic lows “‘~— Magnetic susceptibility, lO-Gcgs units ‘”“fi-~ Monzonite-Quartz monzonite contact Spatial distribution of in situ magnetic susceptibility and aeromagnetic lows. 121 4 + + + m + + b + -H 3 fi . s + + . a ”8‘ + . e e . '0 . 0 e. ' H 2 ‘ -o- 0 a . . . a , . - . .d . H .H o .0 :3 1 4}- a. o 0 m 0 a + Monzonite . Quartz Monzonite 0 1 l l l 1 2 3 4 Distance, miles In situ magnetic susceptibility versus distance from the south contact of the Melrose Stock. 122 of the body's geometry. An interesting feature on the map is the transcurrent nature of the large magnetic lows to the susceptibility contours. The southern of the two magnetic lows may be due to topographic effects of a large ridge. The other magnetic low seems to be independent of topography and consequently, it must be attributed to mag- netization variations within the intrusive. The surface susceptibility pattern bears no correspondence with the aeromagnetic pattern. This infers that the susceptibility of the subsurface is not the same as that on the surface. In the areas of aeromagnetic lows the susceptibility in the subsurface is less than what is expressed at the surface, assuming the effects of remanent magnetization to be uni- form throughout the intrusive. Pigures 3-10 and 3-11 were constructed in order to determine the relationship of the susceptibility to ver- tical position in the intrusive. The monzonite (Figure 3-10) shows no apparent relation of susceptibility with elevation. 0n the other hand, the quartz monzonite (Fig- ure 3-11) displays an increase in susceptibility with ele- vation at a rate of about 450 x 10'6 cgs units per 1,000 feet. This change can be attributed to either primary or secondary causes. The decrease in susceptibility at lower elevations could be due to alteration of the magnetite increasing downward and, hence, cause a general lowering of the susceptibility. The evidence for this suggestion is not strong for most sites at lower elevations do not 123 4 e f . 3 e .3 . . . D . s e e 3 3 - MO I o O s H ' . >: s .t: 2 P e v-l .d .0 '0'! 4.! D4 0 8 3 l s U) 0 j 1 L l l l l l 6600 7000 7400 7800 8200 Elevation, feet Figuree3-10. Variation of in situ magnetic susceptibility with elevation within the monzonite pluton. 124 3 3 - ed . £3 :3 . o m s U‘ ‘ . e 0 . f? 0 <3 2 ' . ' . H . Q C. . Q . . >‘ O 4’ . '1'" .0 H 2 L - ,, 1 .D D. 0 U V) :3 0 1 J 1 l s 6200 6600 7000 7400 7800 8200 Elevation, feet Figure 3-11. Variation of in situ magnetic susceptibility with elevation within the quartz monzonite pluton. 125 show significant amounts of hematite, an indication of alteration. A primary increase of magnetite upwards, is another possibility. The magnetite content does in- crease with elevation, but only in a very general way as shown in Figure 3-12. The change of susceptibility with elevation is too great to be attributable to a var- iation in grain size. It is likely that the susceptibility increase with elevation is not simply a function of one variable, but rather due to several factors. 3.5 Susceptibility and Magnetite Content The magnetic susceptibility of the monzonite and quartz monzonite has been shown to average 3200 x 10-6 and 2000 x 10'"6 cgs units respectively. Therefore, as expected the magnetite content which was determined dur- ing the modal analysis study also shows a difference be- tween the monzonite and quartz monzonite. A normal shaped distribution is shown in Figure 3-13 for the frequency of occurrence of the opaque content in the two rock types. Averages for the monzonite and quartz monzonite are re- spectively 2.1 and 1.2 percent and a correction for the 10 percent of the opaques which is ilmenite reduces these values to 1.9 and 1.1 percent for the magnetite content. A plot of the average in situ susceptibility as a function of opaque content is shown in Figure 3-14. The Opaque content data are in most cases limited to a value from one thin section whereas the susceptibility is the 126 .sousam opacoNsos suumsv on» ceases coaum>mao no soapucsw a ma ouuuocmme mo oEsHo> .manm shaman poem .cowus>mam comm ooom comp ooos comm case 7 . . . _ m6 0. M e s o J o H m.— . . . m . . . as O 0 D O % . . L m; L o.~ 127 \\\\\\\V 2 Quartz Monzonite aouasxnooo JO Kouenbexa Distribution of magnetite content by site with respect to rock type. Figure 3-13. Susceptibility, 10-3cgs units Figure 3.140 128 e 2.9x10-3cg. s unit; 1% Magnetite l l l i l 1 2 3 Volume, % Magnetic susceptibility versus magnetite content. The expected susceptibility— mag- netite relationship (Mooney and Bleifuss, 1953) is indicated by the solid line. The best fit line through the data is indicated by the dashed line. 129 site mean based on several readings. Greater scatter is shown on a graph of the specimen susceptibility and opaque content of a given field core than is displayed in Figure 3-14. The use of means tends to eliminate scatter in the data by means averaging out the scatter shown in individual data. A direct relationship of the two is readily appar- ent. However, a least squares line through the points does not intersect the origin. A number of linear approximations of susceptibility versus magnetite content have been derived for small modal volumes of magnetite (»lOO). 4.2 NRM Results 4.2.1 NRM Dataigyaluation The results of the NRM measurements including number of specimens, mean site direction and intensity, and cir- cles of confidence are found in Appendix C. The circles of confidence of the 34 sites span a large range, 11.9° to 127.5: In most published results the values are typi- cally from a few degrees to approximately 15: Since changes in magnetization could be detected during measurement, it is suspected that the specimens have a component of secondary magnetization which is soft. A compilation of the site circles of confidence according to when the specimens were collected is displayed in Figure 4-1. Circles of confidence from the 1968 sites range to over lZd’as contrasted to the 1970 sites which have circles less than 80: Just three 1968 sites have circles of Frequency of Occurrence Figure 4-1. 134 a l///////// 8//// ’//////// fl Ti‘ \\\\\\\\§ 1 120 160 ////// T b Circle of Confidence, degrees NRM according to time of collection. site circles of confidence a) 1968 specimens, 18 months of storage; b) 1970 specimens, 1 month of storage. 135 confidence under 30: This comparison strongly suggests that the 1968 specimens have acquired significant second- ary components while in storage. The intensity and direction of magnetization in coarse grained igneous rock is rather sensitive to magnetic modi- fications from the time of formation to measurement. There- fore, anomalous NRM data are anticipated and must be elim- inated. Anomalous data are illustrated from two typical sites in Figures 4-2a and b. Anomalous results have a remanent intensity several times greater than the grouping of the other specimens of the sites and in many cases their direction will also be scattered from the others. In Site 10, specimen B and in Site 21, specimens F-l, F-2, and F-3 were eliminated from further analysis. All site stereo- grams and intensities were inspected in a similar manner. Revised Fisher statistics were calculated after rejection of anomalous data. A histogram, Figure 4-3, shows a comparison of the circles of confidence of the sites before and after removal of the anomalous data. Not surprisingly, there is a not- icable shift of the circles of confidence to lower values. However, there are still a large number of sites with re- sults that could not be expected to yield reliable remanent directions. Suitable remanent directions might be obtained upon demagnetization of sites having circles of confidence less than 30: Only three of the sixteen sites having circles of confidence of less than 3d’are from 1968. 136 Specimen Intensity VN-lOA 1.09 xlO-4emu/cc VN-lOB 26.60 " VN-lOC 5.13 " VN-lOF 1.74 " VN-lOG 5.50 " VN-lOH 0.90 " Figure 4-2a. NRM of Site 10, directions and intensi- ties. Equal area projection. 137 Specimen VN-ZlA VN-ZlB-l VN-ZlC-l VN-ZlD-l VN-ZlB-l VN-ZlF-l Intensity 8.12 x 10'3 4.43 4.83 3.86 7.07 23.50 emu/cc Figure 4-2b. NRM of Site 21,directions and intensities. Equal area projection. 138 O 7.” 0 l oucouunooo mo hocusvon Circles of Confidence, degrees te circles of confidence before (a) and after (h) rejection of anomalous data. NRM si Figure 4-3. 139 Chi-square test shows there is a less than one percent probability of this being due to chance. This observa- tion supports the view of dispersion being a direct func- tion of storage time. A check of the number of monzonite and quartz-monzonite sites in the group of 16 sites shows no preference of dispersion to rock type. To further check the effects of storage, a test was made on the samples of Sites 32 which were drilled from a rock pile adjacent to an old prospect pit. The seven specimens from this site were each given the same arbi- trary orientation which resulted in a site circle of con- fidence of 46: a value less than many sites given their correct field orientation. This prompted consideration of the other sites using the laboratory orientation in calculating their mean directions and dispersion. All of the specimens from Sites 1-17 (1968) were grouped together and their mean direction and dispersion calculated suing both their field and laboratory orien- tations. The laboratory orientation is the same for each specimen while the field orientation varies from sample to sample. The results are shown in Table 4-1. The directions resulting from the two sets of orien- tation seemingly differ greatly in declination, but in reality only lé'of great circle distance separate the two directions. If there was little modification during storage then the laboratory orientation should have given wide dispersion because it represents a reduction of each 140 o.~ NH mm Hma asapmuonma ~.~ ma on as eaoaa .Hmm .oum — .soo .Hau _ .ucH +_ .umn cowumusoawo .mcoausscowuo auoumuonmd pas cause 0» coauuseou spa: saua mouam mo mums zmz mo consumasoo .H-v magma 141 specimen to an orientation different from that of the collection. The circles of confidence and precision para- meter are very much the same, indicating with little doubt that storage in the laboratory has significantly modified the remanent magnetism of the 1968 samples in particular. 4.2.2 NRM Directions The mean NRM directions and corresponding disper- sion statistics of specimens from 29 sites are compiled in Table 4-2. Five sites with high scatter of remanent directions and intensities were eliminated from the study due to the poor quality of the data. The data were group- ed according to collection date and lithology and analyzed in various ways. The only appreciable difference in the 1968 and 1970 samples is in declination. The declination is shifted towards the north for the sites sampled in 1970. In all comparisons the monzonite shows less dispersion than the quartz monzonite. The location of 7 of the directions given in Table 4-2 are graphically presented in Figure 4-4. Point 1 in the figure is the mean direction of all 29 sites and it is midway between the monzonite and quartz monzonite directions, points 2 and 3, respectively. Although not plotted, the circles of confidence of these two groups overlap with only one degree of closure and represent near- ly significantly different directions. An interesting 142 .vav musmwm ca pom: soaumcmamen s m.m m.ma 6.5» H.Hm~ as «no: .0 on. s m.vm m.oa H.55 m.mmm m .ucoz oh. m H.oa m.ma o.om4 H.mom a «so: .0 mm. m s.~¢ «.ma m.Ha m.mm m .ucoz me. e p.m v.aa «.ms e.mm~ as .ucoz .o m m.~e m.s m.es m.aa oa .ucoz m m.oa m.HH «.ms o.mom as ohms m.~a A.NH m.ms m.mmm NH moms m.HH ~.m m.om m.m~m ma Has a .Hmm .0um— .coo .ku .OsH .Omn Henssz umuwm a .mcoaumcaneou macaum> ca pomsouo mouam mN mo mudsmmu zmz .va wanna 143 x Present field 4- Axial dipole field Figure 4-4. NRM directions of 29 sites grouped in various combinations. Circles of confi- dence shown for the 1970 monzonite sites and the 1970 quartz monzonite sites. Equal area projection. 144 feature of the sterogram is the positions of the monzon- ite (points 4 and 6) and quartz monzonite (points 5 and 7) site means of 1968 and 1970. In both cases the points of 1968 are in approximately the same relative direction away from the 1970 points and with about the same angular dis- tance (6°for quartz monzonite, lO’for monzonite). The coincidence of the relative difference in direc- tion of the two 1968 means from the 1970 means suggests that length of storage or magnetic field of the storage facility influenced the shift. In view of this, the best direction for the NRM should be shown by the 1970 samples. Drawn about the mean directions of the two rock types collected in 1970 are the respective circles of confidence. The circles show that there could be a difference in the mean directions, though it is not a significant difference at the 95 percent confidence limit. If the two plutons have the same mean direction, then the monzonite pluton has had its NRM changed through geolo- gic time in the direction of the earth's field. The re- sults illustrated in Figure 4-4 suggest that the monzon- ite unit has a larger and more unstable VRM component. Demagnetization and a storage test may provide further information on whether the NRM directions are the same as the TRM direction, the directions assumed to be representa- tive of the paleomagnetic direction. If the two directions are different upon demagnetization, then there may be support to the idea that the two directions are different 145 paleo-pole positions. 4.3 Alternating Field Demagnetization 4.3.1 Preliminary Demagnetization Twenty-nine of the 34 sites with oriented specimens were chosen for demagnetization. Samples from all sites established in 1970 were demagnetized. The measurements from sites sampled in 1968 with the largest circles of confidence were all eliminated except one, the highest. This extreme was included as a check of the dispersion reduction upon demagnetization. One specimen from each core was selected for the demagnetization study. Af demagnetization was used to isolate the TRM acquir- ed at the time of the intrusive's consolidation. Char- acteristically, coarse grained igneous rocks contain com- ponents of VRM and other stray components. The level at which the demagnetization removes unwanted components varies, but in general the best level of demagnetizing for coarse grained intrusive rocks falls below an alter- nating field of 300 oersted peak strength (As and Zijder- veld, 1958). Initially a group of nine sites from the 29 sites were randomly chosen for demagnetization at five levels, 50, 100, 150, 200 and 300 oersteds. The technique employed by Gromme and others (1967) for determining the level of demagnetization which gives minimum dispersion stabilization of the remanent magnetization was followed in this study. 146 The major criterion for choosing the optimum level of demagnetization is the circle of confidence which should be a minimum. At minimum dispersion the direction of magnetization should be stablized. Inclination, declina- tion, intensity and circle of confidence are shown as functions of the level of magnetic cleaning in Figures 4-Sa to 1. Seven of the nine sites yield a minimum dispersion at 50 oersteds while the remaining two sites show least dispersion for the NRM. It could be argued that Sites 4, 10, 19 and 25 have the best level of demagnetization at 100 oersteds, in view of the leveling of the intensity and only minor increase in the circles of confidence be- tween 50 and 100 oersteds demagnetization. In comparison to the results of Gromme and others (1967), the data presented here is much more variable and of poorer quality. There is greater change in the in- clination and declination values with demagnetization. The amount of reduction in the remanent intensity by de- magnetizing fields as low as 100 oersteds is rarely ex- ceeded by other published results. These soft and rela- tively large moments of magnetization are characteristic of secondary VRM. Figures 4-6 and 4-7 are equal area projections show- ing the paths of the mean site directions with increas- ing demagnetization intensity for specimens collected in 1968 and 1970 respectively. It is apparent that the 147 1 V V T T I I 80 160 3 .. a 120 D g1 - 40 '0 80 6‘ '1 ,, 40 8 - o 0 320 I --40 +120 2 _ - 8 Int. ; __ ‘80 B 0 1' d s 1 ~ c , -40 .3 c: + H o | 1 L I i 1 1 o 100 200 300 Demagnetization, oersteds Figure 4-5a. A.f. demagnetization results of Site VN’4. Cir. Con., degrees Inc., degrees 148 220 Y r V I T, T 1 a: I D o e . u 0‘ .8 180 - m : I 8° 3’» o ' H a 3‘ 140 1 "1 6O . o s H 40 U) 0 o u 100 3. o o \ 'U 2 80 . 0 . v c 2: 5° 8 H 0 : 40 .t: u u c H 20 O 100 200 300 Demagnetization, oersteds Figure 4-5b. A.f. demagnetization results of Site VN-J-o. 149 60 I t i ’,t t v v 90 I A 30 C \ fi 70 m ~ 0 \v a '0 ‘50 . :330' 6 3 ‘ '5 Q 300 F a 30 270 n 90 C0 .. (D 0 ‘8 2 .. 0 U '0 p . 5 q 60 5 8 8 1- .J 3 1 .3: ‘ .. O .5 5 1 30 o l L I l l l J 100 200 300 Demagnetization, oersteds Figure 4-5c. A.f. demagnetization results of Site VN-l3. 10 3 o 0 ‘5 m 350 'O .‘340 u _ 0 Q 10 U U E 8 E V.0) I 6 O H . 4 43 E. 2 0 Figure 4-5d. 150 100 200 300 Demagnetization, oersteds A.f. demagnetization results of 80 U] 0 70 3 D 8 60 so 2 H 120 90 U) 0 0 § '0 a‘ 60 0 U :3 .d U 30 151 I '490 m 4 3 Q) Q) 2 u 3’ - so 8‘ o o a? ‘ 2 G H 4 70 -«40 C -1 3 :3 (D U -130 3. 3 a) g 'o 0 2 q : V' {.1 lo O H o . ‘ 20 . ' 1 .3: .p ‘3 Int. 0 H .- l 1 l l L l 1 o 10 100 200 300 Demagnetization, oersteds Figure 4-5e. A.f. demagnetization results of Site VN-ZO. 152 360 I v v I r 1 T 70 U) m (D ,. 3.. .. C" (a m m 1’ 340 b a : q 60 5* U - 'U 3 . -J o . D 320 g 4 F H ‘ 50 3 8 8 * 20 3 \ U1 2 2 8 o C _ 3' : o c H O . + 10 o J 1 Int. J a H H a o O l l l l l i; P 0 100 200 300 Demagnetization, oersteds Figure 4-5f. A.f. demagnetization results of Site VN-ZZ. 153 140 v v I r ”T 1 v " 80 m 3 120 w 3 0 q H a a.» q, 100 r D ,0 '0 - 60 . .‘ 30 " 6 U .4 C: 0 H Q 60 I— I ‘ 40 to 8 8 E q 80 ‘6‘ C 5 _ 8 q . .0 o H -1 60 g ‘ U +3 _ . 5 Int. ,2 U -*40 l L 1 1 A 5 1 100 200 300 Demagnetization, oersteds :Pigure 4-Sg. A.f. demagnetization results of Site VN-ZS. . 100 270 - ‘ 80 3 a) u: u 0 - o m a» a 'o g 210 -1 so 6‘ ~ r: 0 - H 8 I G D « 40 150 b 1 so 4 m 0 m J u 40 8‘ o 'o 0 \ . . :3 o S 8 1' ~ 30 0 O o H d a : o .p '5 20 0 1 4 L n J j A 100 200 300 Demagnetization, oersteds Figure 4-Sh. A.f. demagnetization results of Site VN-BO. 155 "80 m U) 2: ~ é? §’210 8: 'O ‘ 60 'U . 200 : 6 “ 8 ‘g 190 H - 40 ~ 80 m as 0 - u 8‘ C o 3 2 - 60 ~ 2 a a - o 7" ° 3 - 4o 3 . 1 o 4.: - E, .. v f —:Int. " 20 0 1 L l _L L l L 100 200 300 Demagnetization, oersteds Figure 4-51. A.f. demagnetization results of Site VN-31. 156 Present field Axial dipole field 1Figure 4-6. Mean site directions for NRM, 50, 100, 150, 200, and 300 oersteds demagnetization of Sites 4, 10, and 13, those collected in 1968. Equal area projection. 157 X Present field ‘* Axial dipole field :Figure 4-7. Mean site directions for NRM, 50, 100, 150, 200, and 300 oersteds demagnetization of Sites 19, 20, 22, 25, 30, and 31, those col- lected in 1970. Equal area projection. 158 direction of the remanent magnetization of the three sites collected in 1968 show considerable movement upon demag- netization. The 1968 and 1970 sites show respectively l7d°and Sd’average great circle movement from the NRM to 300 oersteds demagnetization levels. Obviously, stor- age time has had considerable effect on the earlier collect- ed samples. The acquisition of VRM will in general skew a dis- tribution of points from the primary magnetization dir- ection towards the direction of the ambient field along the great circle including the two directions. Inspec- tion of the paths in Figure 4-6 and 4-7 does not indicate consistent movement along great circles from the NRM to the 300 oersted positions. A reasonable explanation of the complex pattern of paths is that the samples have a soft TRM of significant proportions coupled with a more stable VRM acquired through "long storage" in the field. The TRM also is probably soft and is being removed at low demagnetizing fields. If there were no VRM acquired dur- ing laboratory storage then the demagnetization paths should have moved along great circles away from the dir- ection of the earth's magnetic field to the TRM direction. 4.3.2 Demagnetization at Optimum Levels The 20 remaining sites were demagnetized at both 50 and 100 oersteds in order to be reasonably certain that the best level was being attained, yet of course, certainty 159 of this does not exist unless the sites are demagnetized at many levels. The remanent magnetization of the specimens after demagnetization is tabulated in Tables 4-3 and 4-4 in the same format as Table 4-2. In all but one of the list- ings, the NRM shows higher quality data than either the 50 or 100 oersted demagnetization results. The added dis- persion was not expected and this undoubtedly reflects the “soft" nature of the remanence to the demagnetization apparatus. To aid in the evaluation of the magnetic stability and the storage, the directions listed in Tables 4—2, 4-3 and 4-4 for both lithologies and years are plotted in Figure 4-8‘. The remanent directions of the 1968 specimens show greater migration than the 1970 samples upon demag- netization, averaging 1" and 7’ of great circle movement respectively. Only a few degrees of angular distance separate the NRM and 100 oersted directions of both the 1970 monzonite and quartz monzonite specimens. This in- dicates that the remanence removed during demagnetization lies nearly parallel to the HRH direction because the directions remained very close while the intensity decreased from the NRM value by over 50 percent. The path of the 1968 quartz monzonite specimens upon magnetic cleaning is on a great circle towards the direc- tion of the earth's magnetic field in the field area. This is unusual behavior for the residual remanent vector, 160 .mte madman a“ poms coaumcmamon t m.m v.mH o.mh m.mom NH use: .0 as. h v.m~ m.¢H o.h> m.mmm m .usoz on. m 0.5H >.¢H v.5b m.m¢m u use: .0 mm. m >.N n.5m m.mw ~.mo m .Ncoz mm. v >.m H.NH o.mh m.m~m mH .Nsoz .0 m ~.m m.m~ m.hh v.mm 0H .nsoz N w.m m.~H H.m> H.m~m ma osmH m.m N.o~ ~.mh ~.m~ mH momH H.> N.OH ~.om o.m¢m mm Had H .Hmm .mum _ .sou .uwo _ .usH .omn Hunfisz mmuwm a .msowumswnsoo msoaum> cw condone woven mm mo mvHsmou coaumuauocmmEop pmumuoo om .mlv oHnma 161 .mlv eunufim :H can: soaumcmamma .4. mo? HoNN bohb mommm NH NGOZ .0 Oh. h MomH VomH m.mh oommm m oNGOZ Ch. 0 HoNH coma m.Hh b.m h NGOE .0 mm. m Coma moVN $.00 Noflm v oNGOE mm. fi 000 oomH nomb N.bNm mH oNGOZ .0 m ¢ovH moMH Vomr momN m oNGOS N How homH m.mb OoNHm 5H ObmH moNH V.MH Hlo hovN HH mme Mob moOH nomh mommm mm add H .umm .mum— .coo .Hao — .OGH _ .omn — uwnesz mwuwm a .mcoaumsdneoo msowhm> cw Ummaoum mopwm mm mo mpHsmou soaumuwuocmmamv vmumuco ooH .vlv mHnma Figure 4-8. 162 x Present field + Axial dipole field NRM, 50, and 100 oersted demagnetization directions of the monzonite and quartz mon- zonite collected in 1968 and 1970. Numbers refer to Tables 4-2, 4-3, and 4-3 and are located at the NRM positions. Equal area projection. 163 as it is expected to move in the opposite sense upon de- magnetization. Since it is the only group to show this movement, it is considered anomalous. This can be explain- ed by a TRM coercivity spectrum which is lower than the VRM spectrum, however, this is unusual. The monzonite shows less movement upon demagnetiza- tion than the quartz monzonite. The demagnetization to 300 oersteds of three sites collected from each of the rock types shown in Figure 4-7 gives an average movement of the remanence of 38° for the monzonite and 71° for the quartz monzonite. The 1970 specimen results shown in Figure 4-8 reveal that the remanence of the monzonite sites moved 4'upon demagnetization as contrasted to 11? for the quartz monzonite. The greater stability of the monzonite is believed to be influenced by the difference in the magnetite size distribution as shown in Section 2.2. The results of the demagnetization are not strongly convincing as would have been anticipated. The remanent .results from the 1970 specimens are believed to be more .representative of the remanent directions. No distinction can.be made between the REM and demagnetized directions of remanence because both are nearly parallel. ‘4.3.3 Remanent Intensities ———— A tabulation of the intensities of magnetization before and after demagnetization are presented in Table 4-5. ‘1‘. \~ . ' l! t i. 164 OJ» _ v5 v6 A 5.0 m.m _ mfn mmamnyxm W ~.H m.~ s.m cam: m m.m _ m.o o.ma _ m.H m.ma _ s.~ mmsmuuxm w H.m o.m «.5 saw: w OHGO OOH OHQO cm a .AOO\580 IOHxv mofiuwmcmucw :owumnaumcvme ucmcmamm .mlv mHQmB v 165 Individual specimens have magnetizations which range from l x 10"2 emu/cc to S x 10-5 emu/cc. Most samples, however, are of the order of magnitude 10'"4 emu/cc. The extremes for each entry in the table show a wide separation. The difference within a rock type (NRM, 50 oer., 100 oer.), and between rock types are not statistically significant. However, the means of the different entries appear to be. consistent. The calculations are based upon specimens from the 29 demagnetized sites. The range for each of the entries into Table 4-5 is about one magnitude, even after demagnetization. This suggests that the variation in the NRM is not from a range of intensities of superimposed secondary components on a uniform TRM. If this were true, the range or extremes for the two rock types should have decreased with each successive level of demagnetization. The variation is likely due to the normal distribution inherent to the lithologies. There is nearly a two to one relationship of the monzonite to quartz monzonite intensities at each of the three levels, which directly reflects the magnetite content. Included in Table 4-6 are various ratios of remanent intensities. Each step of demagnetization removed approx- imately 40 percent of the remanence and at 100 oersteds, slightly over one-third of the remanence remains in each rock type. In contrast, basalts generally show only a few percent decrease in magnetization at 100 oersteds. The 166 Table 4-6. Remanent intensity ratios. R R .59.. R100 _.l£9_ NRM NRM R50 Monzonite 0.59 0.39 0.60 O. Monzonite 0.62 0.36 0.60 167 he removal of the remanence is typical (Strangway, 970, p. 79) of that for granitic rocks in which the agnetite is coarse grained. L. 4 S toragg Tests 4 . 4 . 1 Storage Procedure Nine specimens from four sites were selected for storage and subsequent demagnetization tests. The object of these tests was to determine the significance of the VRM contribution and how easily it could be removed. Movements along great circles of as much as 50° due to VRM acquisition have been noted by Akimoto and Kushiro (1959) for a suite of dolerite specimens. Experimentally, ‘Rimbert (1958) has shown that VRM is more stable to af demagnetization the longer the period over which it has been acquired. Extrapolating the results of Rimbert, in conjunction with the results of this study, should suggest whether the amount of VRM acquired during storage, both for the 1968 and 1970 specimens, has been removed by the af demagnetization. The specimens used for this experiment were given arbitrary orientations because they were cored from unorient- ed hand samples. After the cores had been prepared and stored in the same position (ambient field of the labora- tory) for several months, they were rotated 180° about an axis nearly at right angles (8. off) to the magnetic merid- ian. The proximity of the magnetic meridian to the plane 168 of rotation permits, as a reasonable approximation, one- half of the difference in the change of the three mutually orthogonal components as the means to calculate the amount of VRM acquired during storage. The experiment commenced with the rotation of the specimens and the subsequent determination of the resulting remanence from zero time B. to over 100 days later. 4. 4. 2 Results ' The migration of the NRM direction for each of the 9 specimens is shown in Figure 4-9. The best fitting great circle was visually drawn through each set of points. These intersect about the direction of the ambient field in the laboratory. The NRM directions have been moved through 15°to Bd’of great circle arc in three months. The magnitude of this movement indicates a large ratio of VRM to the magnetically stable components and shows that the original remanence can be significantly altered in a short period of time. The magnitude of van attained as a function of the storage time was calculated and the results are shown in Figures 4-10 and 4-11. The VRM intensity shows an expo- nential growth with time and in just a few days an appre- ciable quantity of VRM is acquired. The acquisition of VRM is a time dependent phenomenon thought to be due to thermal agitations allowing the mag- netization to be "trapped" in the direction of the ambient 169 32A-2 Figure 4-9. Migration of remanent magnetization upon storage in the laboratory. 170 .huoumuonmd esp cw mwmuosu ocfiusp pmuwsvum zm> mo wuwmcmucH hump .mewa 00H om oo 0* 0N - d n u u q a q u q «hv T Ill!!! ) ‘ 95¢ UH¢ QHv wt. QH¢ - n - ‘l‘ .osne magmas \\\ r-‘I ao/nmav_ot 'WHA 171 .auoumuonma 0:» ca comuoum Unease mm ouam Eoum nauseoman 03» How pmuasvum zm> mo huwmcmucH .HHue shaman wasp .oewe 00H ONH om ov a J 1 q a q u a n O mmm m.o Do/mu3v_0I ‘waA 172 field when it crosses over energy barriers. Mathemati- cally the acquisition of VRM is expressed by the formula: 1) I = constant + s logt where I is the acquired remanence during time t and s is an index of magnetic viscosity. The data of Figures 4-10 and 4-11 are plotted on a semilog plot as shown in Figure 4-12. The trends of the semilog plot can be projected in order to estimate the in- tensity of VRM acquired during long periods of storage. Equation 1 predicts that the growth of the VRM component is a straight line when plotted on a similog graph. Figure 4-12 shows data from Sites 30A and 32 are nearly linear whereas the data from Sites. 41 and 47 are not linear. A simple calculation shows that in less than 1 million years, the samples from Site 41 would attain a VRM equal to their VRM. It is apparent that the equation 1 does not hold its semilog linearity for a great number of magnitudes. How- ever, it does lend insight to the fact that the remanence of the rock at Site 41 must be largely VRM. Projecting the trends of the specimens from Sites 30A and 32 reveal that approximately three-fourths of the NRM can be attri- buted to VRM for a storage time of 125 m.y. from the time the rocks were formed (125 m.y.). The ratio of the VRM acquired during storage to the mm at the commencement of the test is compiled in Table 4-7 (data in the last column is referred to in the follow- ing section). No distinction can be made as to which rock 173 .meu on pummuwu nuwa coauauwsvum Em> mo scam OdesuauooHfiEmm .NHIe musmwm wasp .meaa OOH 0m om 0.“. ON OH m 0 fi N a . . 4 q j 1 j fiwq - T 1 A - Chv tIANOm “ N'fim .1 I. h «Idem ‘11 new 14 \\\\\\\ L UHV GHV mac .I:t(\\\\q1 - b . p - p p! p p p b p p — P - 03H m.H Da/nm8,_0I 'waA Table 4-7. 174 Results of storage test. F Specimen Storage * VRM Added "(111M Demag. (Days) % Field VNBBOA-A 98 40 25 VN-BOA-B 69 30 - VN-32A-2 13S 15 20 VN-32B 117 25 20 VN-4lB 87 35 20 VN-41C 100 40 10 VN-41D 99 25 5 VN-47A 100 35 5 VN-47D 100 50 20 *( van at 100 days/ NRM at 0 days ) x 100 ** af demagnetization in oersteds necessary to remove VRM acquired during storage 175 type has larger amounts of VRM. It is clear, however, that the NRM can be significantly changed in a matter of a few months and has certainly been appreciably modified since the time of rock formation by VRM. 4.433 Demagnetization of Stored Samples Eight of the samples were progressively af demagne- tized in order to determine the level of demagnetization necessary to remove the VRM acquired during storage. The demagnetization results will indicate whether the VRM's of the 1970 specimens acquired during a month of storage were removed with the magnetic cleaning. With the aid of Rimbert's demagnetization results, a reasonable extra- polation can be made to determine whether the VRM was also removed from the 1968 specimens. Figures 4-l3a through h show the results of demag- netization of the stored sampled for five progressive steps of af demagnetization (20, 40, 60, 80 and 100 oer- steds). The positions of the net component removed are connected by great circle segments. The direction of the component of the magnetization removed in each interval is shown by the isolated points. The first point of the net component and of the first interval are, of course, coinci- dent. The zero time position, terminal storage direction and the direction of the total VRM acquired are also illus- trated in the stereograms. Figure 4-13a shows the first component of 176 VRM Direction of acquired VRM component "0” NRM at start of storage test T NRM at termination of storage test Connected segments are net components removed at successive steps of demagnetization Isolated points are individual components removed during demagnetization Figure 4-13a. Storage test and demagnetization of specimen VN-3OA-A. Equal area projection. 177 VRM Direction of acquired VRM component ”0" NRM at start of storage test T NRM at termination of storage test Connected segments are net components removed at successive steps of demagnetization Individual points are individual components removed during demagnetization Figure 4-13b. Storage test and demagnetization of specimen VN-32A-2. Equal area projec- tion. 178 T o 0 VII“ VRM Direction of acquired VRM component "0" NRM at start of storage test T NRM at termination of storage test Connected segments are net components removed at successive steps of demagnetization Isolated points are individual components removed during demagnetization Figure 4-13c. Storage test and demagnetization of specimen VN-BZB. Equal area projection. 179 VRM Direction of acquired VRM component ”0" NRM at start of storage test T NRM at termination of storage test Connected segments are net components removed at successive steps of demagnetization Isolated points are individual components removed during demagnetization Figure 4-13d. Storage test and demagnetization of specimen VN-4lB. Equal area projection. 180 VRM Direction of acquired VRM component "0" NRM at start of storage test T NRM at termination of storage test Connected segments are net components removed at successive steps of demagnetization Isolated points are individual components removed during demagnetization Figure 4-13e. Storage test and demagnetization of specimen VN-41C. Equal area projection. 181 VRM Direction of acquired VRM component ”0" NRM at start of storage test T NRM at termination of storage test Connected segments are net components removed at successive steps of demagnetization Isolated points are individual components removed during demagnetization Figure 4-13f. Storage test and demagnetization of specimen VN-41D. Equal area projection. 182 3, 5 9 D I l O I 0 ‘A | I ? 2"“ .4 VRM Direction of acquired VRM component "0" NRM at start of storage test T NRM at termination of storage test Connected segments are net components removed at successive steps of demagnetization Isolated points are individual components removed during demagnetization Figure 4-13g. Storage test and demagnetization of specimen VN-47A. Equal area projection. 183 VRM Direction of acquired VRM component ”0" NRM at start of storage test T NRM at termination of storage test Connected segments are net components removed at successive steps of demagnetization Isolated points are individual components removed during demagnetization Figure 4-l3h. Storage test and demagnetization of specimen VN-47D. Equal area projection. 184 demagnetization to be nearly in the same direction as the VRM direction. This indicates that at 20 oersteds the demagnetizing field had removed a component which was near- 1y parallel to the storage acquired VRM. The remaining four interval demagnetizations show reversely polarized components and, hence, would not be due to remanence attain- ed during storage. The migration of the net component away from the ambient field direction shows that the effect of the recently acquired VRM is quickly and easily mask- ed by remanence removed in the direction of the NRM. In similar manner the remaining figures in the group can be inspected. All but two of the samples behave in a fash- ion as specimen VN-BOA-A. Specimens VN-41D and VN-47A show little indication of having recently acquired a VRM, because the directions of the first components removed are near their respective NRM directions. Clearly the VRM of these two specimens was very soft. The demagnetization fields necessary to remove the storage VRM were estimated from orthogonal projection curves and the results are included in Table 4-7. For a low inducing field, approximately 30 oersteds of demagnetization are necessary to remove the VRM acquired for each magnitude increase of storage time according to Rimbert's results. Applying this fact to the data of this study indicate that, if 20 oersteds of demagnetization remove the VRM of three months, then 55 oersteds should remove the VRM acquired in 30 months. The VRM picked up 185 by the 1968 specimens during two years of storage should have been removed by the demagnetization as carried out in this study. As previously mentioned, it is generally thought that soft components of remanent magnetization, such as VRM, are removed in fields of 300 oersteds or less. This is particularly true with the coarse grained magnetite bear- ing rocks of this study. However, the amount of long term VRM acquired during ”storage" in the field which was removed by the demagnetizing process is unknown. Gener- ally, this can be determined by demagnetizing until a level is reached at which the remanence is stabilized. However, the demagnetization process leads to erratic results indicating that the remanence is very soft and easily distributed in spurious ways by the demagnetization apparatus. This is substantiated by the erratic results obtained from demagnetization of a few specimens to 800 oersteds. 4.5 _~Q3Ratios Koenigsberger (1938) recognized the value of defin- ing a ratio, Q, as the remanent to induced magnetization. The magnetic expression of a body is a composite of the remanent and induced components. Thus, the Q ratio express- es the relative importance of the two components to each other. Books (1962) has shown that the magnetic anomaly over volcanic buttes can be primarily due to remanent 186 magnetization. Characteristically, volcanic rocks have Q's greater than 1.0. On the other hand, in most intru- sive igneous rock Q is less than 0.5. Thus, the remanent component is generally neglected in magnetic modeling of anomalies derived from plutonic rocks. It will be shown that there is good reason to consider remanence, even though it may be soft and unstable. The Q ratios were calculated using an inducing field of 0.56 oersteds, the mean NRM, and susceptibility for each of the 34 in situ cored sites. The specimens with anomalously high remanence were eliminated. The results of the calculations are illustrated in Figure 4-14. The mean values of Q for the monzonite and quartz monzonite are 0.42 and 0.38 respectively. The remanence has been shown to be unstable and hence it would be very sensitive to mechanical, thermal, and magnetic disturbances. Light- ning strikes, slight heating during drilling of the speci- mens, magnetic fields of the transporting vehicles and fields encountered during preparation will disturb the magnetism and may increase it. The Q ratio would increase because of the relatively constant value of magnetic sus- ceptibility. The effects of weathering would decrease the susceptibility and remanence proportionately and hence have no appreciable affect on O. This assumes the product of weathering to be the relatively nonmagnetic goethite. The median value was chosen as the most representa- tive Q value for each rock type. This was done because 187 .mmhu xuou he mowumu 0 mo coausnauumfin m.H o.H m.o \\\\\\\\\\\\\\\\ Rh \\ h. evacoucoz npumso LIAIIIJ m.H ooH m.o J l l wuwcoucoz l L OH aouaxxnooo go Aouanbexa .eaue ounces 188 righthanded skewness of the histograms in Figure 4-14 is considered to reflect magnetic disturbances in the reman- ence. The median values are essentially the same, 0.28 and 0.29 with the monzonite being slightly larger. Thus, the remanence is approximately one-fourth of the induced magnetization. Since the remanence is in the general direction of the earth's magnetic field, the to- tal magnetization of the intrusive will be low by 25 per- cent if remanence is disregarded. For the quartz monzon- 3 ite the induced magnetization is about 1.1 x 10' emu/cc and with the effects of remanence included this value rises to nearly 1.4 x 10"3 emu/cc. Similarly, correspond- ing values for the monzonite are 1.7 x 10""3 emu/cc and 2.1 x 10"3 emu/cc. This difference may be quite impor- tant in magnetic interpretation. How much of this argument for including remanence in the total magnetization can be extended to other in- trusive bodies is difficult to assess. However, it should be remembered that, even though the remanence may be soft and possess large VRM components, this magnetization will align itself with the direction of the earth's magnetic field and the ommision of this remanence will effectively lower the magnetization of the body. A study of the effect of the intrusive's geometry on Q values indicates that there is little relation of the Q values of the quartz monzonite to the distance from the limestone-intrusive contact on the southern margin. 189 However, the monzonite (Figure 4-15) shows a decrease in Q away from the contact. As mentioned previously, the magnetic susceptibility shows no discernible relationship to the border of the intrusive. As a result, the trend shown in Figure 4-15 indicates a decrease in the remanence away from the contact. The data in the figure suggest that the Q values start leveling off at approximately one- half mile from the contact. It is postulated that the mag- netite fraction near the margin of the intrusive has a higher coercivity spectrum and hence, a more stable reman- ence of the TRM. In fact, those sites nearest the con- tact do show a more stable remanence than sites farther away from the contact, thereby suggesting higher block- ing temperatures. This supports the view of the magne- tite having an appreciably greater number of grains in the monodomain region near the contact. There are insuffi- cient data from the magnetite grain size study to help verify this point. Thermal demagnetization would be help- ful in revealing the blocking temperature spectrum. If the monzonite were the first pluton to be intruded, it would have a significant thermal gradient with the country rock. This temperature differential could well explain a somewhat faster crystallizing margin than the interior of the pluton. As the cooling proceeds the country rock would be heated. Then, if the quartz monzon- ite was intruded into an already heated host rock, the temperature gradients would be less than within the 190 1.3 .| | ‘h 0.6 t‘ ‘0 \ \ ’ \ ' \ \ \ 0.4 b \ \O \ \ O \ (I F \ \ \ O o . 0.2 ’ , \‘ \ ~ . \ 0 I J L l l P l L 1 J l I 0.4 0.8 1.2 Distance to Contact, miles Figure 4-15. Q ratios versus distance from south contact in the monzonite pluton. 191 monzonite pluton at the time of crystallization. As a result the grain size of the magnetite would be more homo- geneous, leading to homogeneity in the stability of the remanence, and more uniform remanence within the quartz monzonite as observed. 4.6 Paleomaggetism The determination of paleomagnetic poles has allowed geophysicists, as well as geologists, to reconstruct past events. Paleo-poles can be used to date magnetic rock units if certain structural relationships are known, or they can be used to decipher the structural history, if the age of the magnetic rock unit is known. The age of the Melrose Stock is known to be early Cretaceous, or perhaps late Jurassic. The position of the Cretaceous paleo-magnetic pole with respect to North America is fairly well defined at 1ed’w, 64°N (Strangway, 1970). The limited scattered paleomagnetic data from known Jurassic rocks suggests that there may be little differ- ence in the pole positions of the two geologic periods. With the age of the intrusive known, the tectonic movement (rotation) can be determined. The Melrose Stock is part of a horst which has a strike of Nl6°E. Snow (1963) has indicated that the fault block has been rotated approximately 15: upthrown on the west side and downthrown on the east, which is based on the orientation of sedi- ments overlying the intrusive. This evidence may not be 192 conclusive due to distortion of the beds when the intru- sive was emplaced as readily seen by the upturned beds along the southern margin of intrusive. However, the fault, which nearly parallels the range on the east side (Figure 1-2) and repeats the volcanic-intrusive sequence in outcrop, indicates the east side to be downthrown. The paleo-pole position of the Cretaceous period and the poles obtained from the 1970 data with structural rotation are shown in Figure 4-16 and the corresponding numerical values are listed in Table 4-8. These are cal- culated from the NRM values of specimens collected in 1970 (the demagnetized positions differ little from the HRH). The positions of the NRM poles after 10: 15'and 20. of rotation about an axis of N16°E, the strike of the horst which includes the Melrose Stock, are shown in sequence. To obtain the pole positions shown in Figure 4-16, the NRM vector was rotated to the west or the horst was up- lifted on the west in order for the pole position to approach the Cretaceous pole. Therefore, the horst was originally differentially uplifted on the east during the Basin and Range faulting. Both rock unit poles prior to application of the rotation are approximately the same distance from the Cretaceous pole. A structural rotation of 15°brings the monzonite pole to within 4’of great circle arc of the Cretaceous pole position while the quartz monzonite and the average of 1970 site poles do not closely approach the 193 .unuos xuoum enouHmz 0:» mo mcowumuou ou mcwpcomnmhuou noHom on» we ncowumuou sue: oGOHm zmz cemH on» no chHanom oHom uwumcmmaomHmm .mHIv madman coaumHOH.OH now UnmomH "0H mo spas oHom uneconcoa .v zmz osmH v cowumuou sues mHom .ucoa zmz oemH m coaumuou oHom nsomomumuo H OH sues maven oemH Ho maom zmz.cmoz m mHom uconoum + e .o 194 Table 4-8. Paleomagnetic poles determined from the 1970 NRM results. Cretaceous 68N, 180W 1970 sites 48N, 139W lO'rotation 43N, 103W 1970 monzonite 65N, 115W 10° rotation 64N, 155w 15° " 61N, 171w 20" " 5m, 179E 1970 Q. monzonite 4lN, 148W 10°rotation 38N, 168W 195 Cretaceous pole. The position of the monzonite pole is nearly on a great circle with the quartz monzonite and present pole positions. This suggests migration of the poles due to VRM, although the storage tests do not indicate one rock type to be more susceptible to VRM than the other. Also, the monzonite pole is nearly in between the Cretaceous and the present magnetic pole positions. Unfortunately, no conclusive statement can be made regarding the posi- tion of the monzonite pole. It may have migrated from the Cretaceous pole position or the quartz monzonite posi- tion or it may represent its primary remanent magnetiza- tion position. The different pole positions of the two major rock types of the stock may represent the magnetic field at two different times. The monzonite pole position with a 15°rotation nearly coincides with the established Cre- taceous pole of North America. The quartz monzonite, which is thought to have been emplaced later, may have crys- tallized at a sufficiently different time to have taken on a different remanent direction. The age date of the Melrose Stock was determined on the monzonite pluton. In a composite batholith such as the Boulder Batholith, emplacement of individual plutons can take place over a time span of 10 m.y. (Tilling and others., 1968). It is not suggested that emplacement of the two exposed plutons in the Melrose Stock required that time interval. It 196 does, however, lend support to the suggestion that the relative position of the magnetic north pole may have changed. Secular variations of the field may be respon- sible for the difference between the two pole positions, however, the period of the secular variation is in the order of thousands of years and hence should be averaged out on the time scale for the emplacement and cooling of the Melrose Stock. The structural rotation of the Melrose Stock as de- termined from the remanent magnetism is in the opposite sense to the geological interpretation by Snow. Both the paleomagnetic and Snow's method are subject to error and with the present data, the conflict cannot be resolved. CHAPTER V SUMMARY As a result of this study a better understanding has been obtained of the many factors which play a role in determining the ultimate magnetic expression of an igneous intrusive. The Melrose Stock provided the evi- dence that permitted the answering of many questions and at the same time the evidence brought an awareness that there is much more work needed to be done on igneous intrusives. Hopefully, the Melrose Stock will serve as a representative model for future academic and commercial investigations of intrusives in the Basin and Range Pro- Vince. The Melrose Stock was emplaced into Permian sediments. The injection of magma into the host rock was forceful as evidenced by upturned sediments and cataclastic textures revealed microscopically. The magma must have been large- ly a crystal mush at the time of emplacement because of the minor reaction with the host and common occurrence of xenoliths. However, the assimilation of the carbonate host rock did modify the magma locally which is reflected in the iron oxides. The exposed portion of the Melrose 197 198 Stock shows it to be comprised of at least two plutons of monzonite and quartz monzonite compositions. A radio- genic age date of the monzonite pluton places it as very early Cretaceous and it is considered to be the old- er of the two plutons. From the evidence of the association coefficient, it is surmised that the monzonite pluton had a shorter crystallization history than the quartz monzonite pluton. The magnetic minerals appear to have had a long history of formation. The growth of the magnetite was genetically related to the formation of the ferromagnesian minerals. Magnetite was found to have a much greater association to hornblende than any other constituent mineral in the rock. Magnetite formed by three processes as indicated by the association coefficients. These are direct pre- ciptation, oxidation reactions and hydrothermal or deu- teric alteration. The percentage of magnetite formed in these three ways can be estimated from the association parameter. The relative crystallization sequence of each pluton was portrayed from the evidence of the asso- ciation coefficient and petrographic observations. The important magnetic minerals found in the Melrose Stock are ilmenite, which in some cases has exsolved hematite, and a pure magnetite. The opaque content in the monzonite is 2.1 percent and 1.2 percent in the quartz monzonite. The ilmenite averages 10 percent of the opaque assemblage and is not preferentially distributed with 199 respect to rock type. In both rock types, the opaques associated with the ferromagnesian fraction have a larg- er grain size than the opaques associated with the quartz- feldspar fraction. In addition, the quartz monzonite has a larger grain size distribution than the monzonite. The larger size distributions are thought to reflect both the availability of iron and the time available for growth. The variation of the interstitial-inclusion in- dex revealed that the relative intergranular nature of magnetite most likely changed in response to the vis- cosity of the consolidating magma. The magnetic properties of the Melrose Stock were delineated by rock type. The magnetic susceptibility of the quartz monzonite is 2,000 x 10'6 cgs units and the mon- zonite is 3,200 x 10'"6 cgs units. These values are an average of values determined by in situ and specimen core methods. Both methods agree within 10 percent of each other. The susceptibility means are suspected to be low- er than the true value. Weathering has reduced the sus- ceptibility by an estimated 30 to 40 percent. The mon- zonite appears to be more susceptible to the effects of weathering than the quartz monzonite. The areal suscepti- bility distribution does not correlate with the aeromag- netic pattern. There are two notable areas within the stock where the susceptibility distribution in the subsur- face is appreciably less than on the outcrop surface. This is not thought to be influenced by the remanent 200 magnetization. The results of the remanent magnetization study were not totally satisfying. The remanence of these igneous rocks, not unexpectedly, was rather unstable and soft and possessed large components of VRM. Signifi- cant dispersion was added to the remanence by storage in the laboratory. However, a combination storage-demag- netization test revealed that a 50 oersted demagnetiz- ing field to be sufficient for removing VRM acquired in the specimens stored for 18 months. Progressive demag- netization of nine sites showed 50 oersteds gave mini- mum dispersion of the site's remanent direction. When the site directions were statistically analyzed in var- ious lithologic and time of collection combinations, it was found that the NRM results were as good or better than the demagnetized results. This reflects the un- stable nature of the remanence that is commonly found in coarse grained granitic rocks. Hence, the reliability of the remanent magnetization for paleomagnetic inter- pretation is open to question. Two remanent directions were obtained, one for each pluton of the stock. The remanent directions remained nearly stationary with demagnetization at 50 and 100 oersteds. The storage test showed the NRM to change by as much as 50 percent in three months and demagnetization to 100 oersteds removed approximately two-thirds of the NRM. This may indicate that a higher demagnetization 201 level is necessary to remove the VRM acquired through "geologic storage" in the field. Unfortunately, the specimens were very susceptible to picking up spurious moments of magnetization during demagnetization, espec- ially at fields greater than 50 oersteds. If the two remanent directions are different, they may represent the position of the earth's magnetic field at different times. The 1970 remanent data were used to determine the structural rotation of the horst of which the Melrose Stock is a part. The data indicate the Dolly Varden Range to be upthrown on the east and downthrown on the west side. This conflicts with the geological interpre- tation by Snow. Both interpretations are subject to error and cannot be resolved without further investiga- tion. . Thegeometry of the intrusive has had a number of effects on the magnetic minerals of the Melrose Stock. It was found that near the southern margin of the stock sphene was much more abundant than elsewhere in the stock. This is a result of the influence of the carbonate host providing calcium and the breakdown of ilmenite provid- ing the titanium to form sphene. In the lowest portion of the stock and away from known contacts, ilmenite showed exsolution of hematite, presumably in response to slow cooling deep within the magma chamber. These two obser- vations give the stock a zoned character and allow 202 conjecture to the size and amount of the stock removed by erosion. Within the quartz monzonite pluton the interstit- ial-inclusion index increases with elevation, an obser- vation presumably indicative of lower viscosity higher in the magma chamber. This is supported by a decreas- ing quartz content at higher elevations. Both the sus- ceptibility and hornblende association also increase upwards in the quartz monzonite pluton. These are re- lated to each other and are thought to represent the effects of a more oxidizing atmosphere higher in the magma chamber. The variation of the p02 with elevation could either reflect concentration of the volatile phase to the upper portions of the magma chamber or the assimi- lation of the sedimentary host rock. The only expression of geometric effects within the monzonite pluton was on the Q ratios. Apparently the monzonite had a larger temperature differential with the host rock than did the quartz monzonite, for the former shows an exponential decrease of Q away from the southern margin contact. There were sufficient data so as to permit suggest- ions regarding sampling guidelines for magnetic suscepti- bility. Susceptibility measurements were made both in situ and on specimen cores with a volume difference of eighty times. Twice as many data were rejected as anomalous for 203 the specimen cores in comparison to the in situ method. The capacity to delineate magnetic units is in favor of the in situ method. This points out that the larger the sample volume, the greater the chances of having better quality data and of separating magnetic units. It was found that the number of sites sampled was more impor- tant than the number of samples per sites. Two or three in situ measurements per site was sufficient, but it in- volves little expenditure of time to take a few addition- al measurements. A site density of one site per square mile was determined to be sufficient in order to obtain a representative susceptibility for an intrusive such as the Melrose Stock. Also, this density gives the investi- gator the opportunity of detecting multiple pluton intru- sives. From the integrated approach of this thesis, it is believed that a significant amount of information was obtained which has led to a clearer picture of events in the history of the Melrose Stock than would have been possible by independent approaches of geophysics and petrology. Obviously, many points were not fully answer- ed due to insufficient data and knowledge. Further in- sight would be gained by pursuing the geochemical aspects of this investigation. However, the scope of this investi- gation did reveal a much better understanding of the mag- netic properties of an igneous intrusive. REFERENCES CITED REFERENCES CITED Armstrong, R.L., 1963, Geochronology and Geology of the Eastern Great Basin (Ph.D. thesis): New Haven, Yale University. As, J.A. and Zijderveld, J.D.A., 1958, Magnetic Cleaning of Rocks in Paleomagnetic Research: Geophys. Jour., v. 1, p. 308-319. Books, K.G., 1962, Remanent Magnetism as a Contributor to Some Aeromagnetic Anomalies: Geophys., v. 27, pe 359-3750 Buddington, A.F., and Lindsley, D.H., 1964, Iron-Titan- ium Oxide Minerals and Synthetic Equivalents: JOUI. PetIOIQ' V. 5' p. 310-357. Carmichael, I.S.E., 1963, The Occurrence of Magnesian Pyroxenes and Magnetite in Porphyritic Acid Glasses: Mineral. Mag., v. 33, p. 394-403. Carmichael, I.S.E., 1967, The Iron-Titanium Oxides of Salic Volcanic Rocks and their Associated Ferro- magnesian Silicates: Contr. Mineral. and Petrol., V. 14. p. 36-64. Carmichael, I.S.E. and Nicholls, J., 1967, Iron-Titan- ium Oxides and Oxygen Fugacities in Volcanic Rocks: Jour. Geophys. Research, v. 72, p. 4665-4687. Case, J.E., 1966, Geophysical Anomalies over-Precambrian Rocks, Northwestern Uncompahgre Plateau, Utah and Colorada: Bull. Amer. Assoc. Petroleum Geologists, v. 50, p. 1423-1443. Chikazumi, S., 1964, Physics of Magnetism: New York, John Wiley and Sons, Inc., 554p. I Currie, R.G., Gromme, C.S., and Verhoogen, J., 1963, Remanent Magnetization of Some Upper Cretaceous Granitic Plutons in the Sierra Nevada, California: Jour. Geophys. Research, v. 68, p. 2263-2279. 204 205 Czamanske, G.K. and Wones, D.R., 1970, Amphiboles as Indicators of Oxidation during Magmatic Differ- entiation: Geol. Soc. Am. Abs. with Programs, v. 2, p. 531. Doell, R.D., and Cox, A., 1967, Paleomagnetic Sampling with a Portable Coring Drill, in Collinson, D.W., Creer, K.M., and Runcorn, S.K., eds., Methods in Paleomagnetism: New York, American Elsevier Publ. Co., p. 21-25. Emmons, S.F., 1877, Descriptive Geology: Geological Exploration of the Fortieth Parallel, v. 2, p. 476-483. Fisher, R.A., 1953, Dispersion on a Sphere: Proc. Roy. Soc. London, v. 217, p. 295-305. Flinn, D., 1969, Grain Contacts in Crystalline Rocks: Lithos' V. 3' p. 361-370. Gromme, C.S., and Merrill, R.T., 1965, Paleomagnetism of Late Cretaceous Granitic Plutons in the Sierra Nevada, California: Further Results: Jour. Geophys. Research, v. 70, p. 3407-3420. Gromme, C. 5., Merrill, R. T., and Verhoogen, J., 1967, Paleomagnetism of Jurassic and Cretaceous Plutonic Rocks in the Sierra Nevada, California, and Its Significance for Polar Wandering and Continen- tal Drift: Jour. Geophys. Research, v. 72, p. 5661- 5684. Hanna, W. F. 1969, Negative Aeromagnetic Anomalies over Mineralized Areas of the Boulder Batholith, Montana: U.S. Geol. Survey Prof. Paper 650-D p. DlS9-Dl67. Hanna, W.F., 1970, Personal Communication. Hill, S.M., 1916, Notes on Some Mining Districts in Eastern Nevada: U.S. Geol. Survey Bull. 648, p. 78-88. Irving, E., 1964, Paleomagnetism and Its Application to Geological and Geophysical Problems: New York, John Wiley and Sons, Inc., p. 399. Irving, E., Molyneux, L., and Runcorn, S.K., 1966, The Analysis of Remanent Magnetization and Sus- ceptibilities of Rocks: Geophys. Jour., v. 10, p. 451-464. 206 Koenigsberger, J.G., 1938, Natural Residual Magnetism of Eruptive Rocks, parts I and II: Terr. Mag. Atmos. Elec., v. 43, p. 119-127 and 299-320. Kretz, R., 1969, On the Spatial Distribution of Crystals in Rocks: Lithos, v. 2, p. 33-66. Larson, E., Ozima, M., Ozima, M., Nagata, T., and Strangway, D., 1969, Stability of Remanent Mag- netization of Igneous Rocks: Geophys. Jour. Royal Astr. Soc., v. 17, p. 263-292. May, B.T., 1968, Magnetic Properties of Rocks Associated with the New Cornelia Porphyry Copper Deposit, Pima County, Arizona (Ph.D. thesis): Tucson, University of Arizona, 159 p. Mooney, H.M. and Bleifuss, R., 1953, Magnetic Sus- ceptibility Measurements in Minnesota Part II: Analysis of Field Results: Geophysics, v. 18, p. 383-393. Nagata, T., 1961, Rock Magnetism: Tokyo, Maruzen Co. Ltd., 350 p. Opdyke, N.D. and Wensink, H., 1966, Paleomagnetism of Rock from the White Mountain Plutonic-Volcanic Series in New Hampshire and Vermont: Jour. Geophys. Research, v. 71, p. 3045-3051. Pothacamury, 1., 1970, Magnetic Properties of the Boulder Batholith near Helena, Montana and their Use in Magnetic Interpretation (M.S. thesis): East Lansing, Michigan State University, 64 p. Rimbert, E., 1958, Thesis, University of Paris. Runcorn, S.K., 1967, Statistical Discussion of Mag- netization of Rock Samples, in Collinson, D.W., Creer, K.M., and Runcorn, S.K., eds., Methods in Paleomagnetism: New York, American Elsevier Publ. Co., p. 329-339. Shandley, P.D. and Bacon, L.O., 1966, Analysis for Magnetite Utilizing Magnetic Susceptibility: Geophysics, v. 31 p. 398-409. Slichter, L.B., 1929, Certain Aspects of Magnetic Surveying: A.I.M.E. Trans, v. 81, p. 238. Snow, 6.6., 1963, Mineralogy and Geology of the Dolly Varden Mountains, Elko County, Nevada (Ph.D. thesis): Salt Lake City, University of Utah, 153 p. 207 Strangway, D.W., 1967, Magnetic Characteristics of Rocks, in Mining Geophysics, Volume II (theory): Tulsa, Soc. Explor. Geophysicists, p. 454-473. Strangway, D.W., 1970, History of the Earth's Magnetic Field: New York, McGraw-Hill Book Co., 168 p. Thellier, E., 1951, Propietes magnetiques des terres cuites et des roches: Jour. Physique et Radium, v. 12, p. 205-218. Tilling, R.T., Klepper, M.R., and Obradovich, J.D., 1968, K-Ar Ages and Time Span of Emplacement of the Boulder Batholith, Montana: Am. Jour. Sci., v. 266, p. 671- 689. Verhoogen, J., 1962, Distribution of Titanium between silicates and Oxides in Igneous Rocks: Am. Jour. SCiQ' V. 260' p. 211-220. Zietz, 1., Bateman, P.C., Case, J.E., Crittenden, M.D. Jr., Griscom, A., King, E.R., Roberts, R.J., and Lerentzen, G.R., 1968, Aeromagnetic Investigation of Crustal Structure for a strip across the Western United States: Geol. Soc. Am. Bull., v. 80, p. 1703- 1714. Zirkel, F., 1876, Microscopic Petrography: Geological Exploration of the Fourtieth Parallel, v. 6, p. 49, 166-167. APPENDICES APPENDIX A GENERAL REMARKS ON MAGNETIC PROPERTIES APPENDIX A GENERAL REMARKS ON MAGNETIC PROPERTIES Magnetic Minerals Most igneous rocks contain ferromagnetic miner- als which lead to both induced and remanent magnetism. Ideally, the latter can be used to determine certain aspects of the rock's history providing this magnetism is sufficiently large and stable. The magnetic memory stored in a rock's remanent magnetization can be modi- fied subsequent to acquisition; these modifications may or may not be desirable depending on the object- ives of the particular investigation. Information is retrieved from the magnetic memory of the rock held in the iron-titanium oxides which are generally a minor constituent of the rock's total volume. The most important magnetic minerals fall into two crys- tallographic systems, the cubic and rhombohedral. According to their atomic and magnetic structures, these minerals are either ferromagnetic or antiferro- magnetic. The most common and important magnetic mineral is magnetite. Though it has the capacity to acquire 208 209 large amounts of remanent magnetism, it is not necessar- ily able to retain it, especially when coarse grained. If titanium is available under proper physical-chemical conditions, it can reside in the inverse spinel struc- ture which may form a solid solution with magnetite. This gives rise to the titanomagnetite series with ulvo- spinel and magnetite as end members. The magnetic pro- perties are diminished with increasing titanium content. Also forming a series with magnetite is maghemite (X Fe203), a low temperature variety of hematite which has an in- verse spinel structure. It possesses magnetic proper- ties similar to that of magnetite, but it is thermally unstable and reverts to the more stable but less magnetic hematite, the change involving a structural transition from the cubic to rhombohedral system. Ilmenite and hematite occur in the rhombohedral crystal system and are and members of a continuous ser- ies. Hematite has no titanium in its structure, and it is much less magnetic than most titanomagnetites. However, it has a very strong coercive force and hence has a stable remanence. Ilmenite is an accessory to the magnetic properties of a rock for it is essentially non- magnetic, however, the physical relationship of ilmenite to magnetic minerals can affect their overall magnetic properties. The hydrous iron oxide, goethite, generally a pro- duct of surface alteration of magnetite or ilmenite, can 210 carry a weak but stable remanence when cooled from mod- erate temperatures in the presence of a magnetic field. The magnetic susceptibility of goethite is negligible in contrast to that of magnetite. The compositional relationship of the magnetic min- erals to each other is shown in Figure A-l, the FeO- FeZOB-Tio ternary system. Remanent Magnetization Natural remanent magnetism (NRM) is the remanence possessed by a specimen in its natural state in the field while the primary remanent magnetization is the magnetism that was acquired at the time of the rock's formation. The NRM and primary magnetization are not necessarily equivalent because of changes in the remanence following acquisition of the primary magnetization. Furthermore, the NRM can be altered subsequent to collection of the specimens. A brief outline of the various types of rema- nent magnetization that NRM can include follows. The acquisition of remanent magnetism by cooling a mineral through its Curie point is referred to as ther- mo-remanent magnetization (TRM). The majority of the TRM is acquired near the Curie point, but due to a spectrum of blocking tempertures, the TRM is actually acquired over a wide range of tempertures below the Curie point. Thellier (1951) found that the TRM acquired in any temp- erature interval is independent of the TRM obtained in 211 TiO FeTiO3 Ilmenite Ilmeno- FezTio4 hematite Ulvospinel magnetite Magnetite Hematitehx) Maghemite(6) Figure A-l. Compositional relationships of the major magnetic minerals. 212 other temperature intervals. This led to the property of partial thermoremanent magnetization (PTRM) being simply an additive phenomenom. The stability of TRM appears to vary inversely with the size of the magnetic minerals in the rock. Remanence obtained at a constant temperature gives rise to isothermal remanent magnetization (IRM). One variety of IRM is ARM, anhysteritic remanent magnetiza- tion, which is acquired by a magnetic material in the presence of an ambient field with a superimposed decay- ing alternating field. The acquisition of a component of remanent magnetization as a result of a specimen re- maining in a constant magnetic field for a period of time is known as viscous remanent magnetization (VRM), the time dependent variety of IRM. During a chemical reaction involving a magnetic min- eral, a remanence can be acquired which is termed chemi- cal remanent magnetization (CRM). A remanence also can be obtained from the deposition (DRM) of magnetic minerals in the genesis of a sedimentary rock. In order to obtain the primary remanent magnetism of igneous rocks, which is generally TRM, it is necessary to magnetically "clean" the rocks. VRM is present in various proportions and can actually mask the original TRM direction so that the NRM direction is almost dia- metrically opposed to the TRM direction (Opdyke and Wensink, 1966). Two basic methods are available for the 213 recovery of the primary magnetization, alternating field (af) and thermal demagnetization. The af demagnetization technique was employed in this study for it is generally faster and easier and does not lead to mineralogic changes which may occur during heating of specimens. However, the thermal method can give an insight into the magnetic minerals present in the rock. Maggetic Susceptibility Induced magnetization is developed in a substance in response to an applied magnetic field and is lost upon removal of the inducing field. The intensity of magnetization (M) is related to inducing field (H) by the constant (k) which is the magnetic susceptibility; MakH For inducing fields of less than a few oersteds, the magnetic susceptibility remains relatively constant. In magnetically isotropic material the direction of the induced magnetization is parallel to the inducing field and in anisotropic material the two vector quantities may be in slightly different directions. Magnetic susceptibility is a function of a miner- al's composition. Specifically, the presence of titan- ium in both iron-titanium series decreases the suscepti- bility. The rhombohedral and hydrous iron oxides con- tribute little to the overall susceptibility of a rock. Even with a fixed composition and volume of magnetic 214 minerals within a rock the magnetic susceptibility can vary in a complex way with such factors as percent re- manent saturation, grain size and exsolution. An in- crease in remanence in a rock is accompanied by a de- crease in susceptibility (Shandley and Bacon, 1963). It is well known that as the grain size of the magnetite decreases so does the susceptibility. Shandley and Bacon (1963) determined a critical size limit dependence below 40 microns. In a closely related manner, exsolu- tion tends to divide or partition a magnetic mineral into smaller subgrains which in turn expresses the be- havior of many small grains. The description of magnetic susceptibility and re- manent magnetism presented here is quite brief, serving as only an introduction to this subject; for further in- formation the reader is referred to several texts on the subject, (Nagata, 1961; Chikizumi, 1964; Strangway, 1967 and 1970). APPENDIX B OPAQUE MINERAL SPECIES AND THEIR RELATIVE ABUNDANCE APPENDIX B OPAQUE MINERAL SPECIES AND THEIR RELATIVE ABUNDANCE Opaque Data Table A-1. seesaw seasonoez assesses opaooaaa ouaooaaa openness: Guam 15 15 40 70 60 75 80 15 10 20 15 85 85 25 28 13 70 70 85 25 40 75 60 85 10 11 12 12 15 10 100 75 13 14 15 16 17 18 80 x15 78 75 NICO x20 15 80 19 20 21 9O 75 10 15 215 216 £3 3 3 :3 . s .4 .a o 55 .l 14 .3 :n g 22 80 10 1o - - c 23 85 x12 - 3 - - 24 83 x10 5 2 - - 25 92 x 3 5 - - c 26 80 x15 5 - - 27 85 x10 5 - - 28 83 x 2 3 10 - - 29 82 x13 2 3 - m 30 75 x 7 4 7 7 m 30A 80 x10 5 5 m m 31 75 x 7 3 15 - m 32 70 S 15 10 - - 33 99 - - 1 - C 34 93 - 2 _ c 40 90 8 2 - - m 41 85 15 - - - - 42 78 20 2 - - 43 85 5 10 - - 44 95 2 2 1 - c 45 80 x20 - - - - 46 89 x 8 - 3 - 47 90 10 - - _ (BOX ilmenite with exsolution common or rich, approximately 1% minor, approximately 3% trace or absent APPENDIX C NRM RESULTS BY SITE APPENDIX C NRM RESULTS BY SITE Table A-2. NRM Data Rejected to Cir. con. After removal of Site total beforelafter anomalous data specimens rejection Dec.lInc.lInt. 1 2/6 99.1 63.9 9.4 55.3 4.95 2 1/8 52.8 49.0 84.8 55.9 3.44 3 0/6 127.5 127.5 355.4 65.7 4.52 4 0/7 41.6 41.6 327.3 47.7 2.73 5 1/6 91.3 151.5 304.9 30.2 2.39 6 1/7 48.1 49.4 146.9 - 0.9 5.17 7 0/6 14.0 14.0 325.9 77.1 3.15 8 2/5 112.8 70.7 160.7 71.1 3.93 9 1/6 80.3 118.2 92.7 77.8 4.67 10 1/6 45.5 37.4 219.7 52.1 2.88 11 0/6 19.3 19.3 350.7 71.5 1.54 12 1/5 33.3 39.0 62.5 74.8 5.07 13 1/6 53.9 40.8 067.4 60.3 2.32 14 0/4 40.3 40.3 921.1 55.1 3.49 15 1/6 27.0 15.1 59.5 61.3 18.60 16 1/7 69.5 95.7 69.4 35.5 2.14 17 1/6 83.8 89.6 51.6 63.2 3.31 19 5/13 47.3 40.0 1.5 55.0 9.25 217 218 Rejected to Cir. Con. After removal of Site total before! after anomalous data specimens rejection Dec.LInc. I Int. 20 2/14 12.1 13.8 333.2 74. 6 11. 007 21 3/13 23.4 11.6 3.4 79.1 5.65 22 3/13 18.6 10.8 354.5 61.7 3.64 23 3/11 42.2 46.2 44.7 75.1 6.30 24 3/11 35.1 23.3 13.2 85.9 1.92 25 0/13 19.5 19.5 163.0 80.0 1.46 26 0/15 14.8 14.8 327.1 78.7 4.95 27 1/12 26.6 27.7 276.9 65.4 9.30 28 0/10 25.1 25.1 355.0 77.9 1.91 29 5/10 50.3 45.4 256.0 43.0 7.04 30 0/11 25.1 25.1 88.3 84.0 1.86 31 3/10 67.2 18.6 160.0 84.9 1.47 33 0/6 24.8 24.8 85.2 71.5 2.19 34 0/9 11.9 11.9 314.8 52.0 2.45 45 3/7 47.5 24.9 282.7 34.9 9.03 46 1/8 62.9 46.0 352.5 55.3 2.84 Intensity, x 10’4emu/cc Circle of confidence, declination, and inclination in degrees