‘1..." m-uv. ...........-‘. MINIMUM"!WWWWIMINI!!!“ 23 10669 5244 L331: 313: :3 T; 3:: 5325713535: @1331. Univem‘ity ,1 7* This is to certify that the thesis entitled The Austurhorn and Vesturhorn Acidic and Basic Complexes in Southeastern Iceland: Examples of Magma Mixing presented by Steven R. Mattson has been accepted towards fulfillment of the requirements for M.S. degree in Geologx V //”/7/ /:::/1/uf/r/ / Mfi/ {ofessor 8 9 81 Date I / 0-7 639 OVERDUE FINES: 25¢ per «y W m- .3 341/“... "“3 ‘ W= -. q Place in book return to move “"5" a charge from ctr-caution records \ The Austurhorn and Vesturhorn Acidic and Basic Complexes in Southeastern Iceland: Examples of Magma Mixing by Steven R. Mattson AN ABSTRACT OF A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Master of Science Department of Geology 1981 ./_) ABSTRACT The Austurhorn and Vesturhorn acidic intrusions located in Southeastern Iceland have associated with them acidic and basic complexes which consist of pillow-like masses of basic rock surrounded by acidic rock and represent coexisting acidic and basic magmas. Two simplistic models for the origin of these rocks include: 1) silicate-liquid immiscibility and, 2) commingling and magma mixing. The field data, trace elements, and major element data are most consistent with commingling and limited (i.e. complex) magma mixing, in situ. Due to the wide range of compositions, found in the pillows complex models must be considered and they include: 1) the involvement of more than two compositionally different magmas, 2) a zoned magma chamber, or 3) a Soret Effect. The acidic rocks found at these intrusions are similar to acidic rocks found elsewhere in Iceland. Only simple crystal fractionation from a primitive basic magma can be eliminated as a contending hypothesis for their origin. ACKNOWLEDGEMENTS This research was supported by a National Science Foundation Grant (NSF EAR 7904125). Field study was made possible by G. E. Sigvaldason and the Nordic Volcanological Institute at the University of Iceland. Gratefully acknowledged are the many ideas, technical support, and criticisms of T. A. Vogel, J. T. Wilband, and F. W. Cambray throughout the course of this work. I would also like to express gratitude to the many denizens of the Geology Department who were often willing to deliberate geological foment with audacious alacrity (Tom Taylor, Bob Gudramovics, Rudi Meyer) and those who were more difficile and unpalpable, although laudatory in their intent (Bill Petro, Phil DeGruyter, Mark Breithart). And my greatest appreciation to Lil who supplied encouragement, libations and a retreat in times of need. Table of Contents Page List of Tables .............................................. iv List of Figures ............................................. v Introduction ...... .......................... ..... ........... 1 Field Description ...... ...... .......... . ........ ....... 3 Petrography ............................................ 27 Sampling and Analytical Methods ........................ 36 Petrochemistry .............................................. 41 Major Elements ......................................... 41 Trace Elements ................ ........ ..... ....... ..... 48 Evaluation of Mbdels of Origin . ............. ................ 56 Immiscibility ....... ...... ............... ...... . ....... 56 Magma Mixing ......... . .......... ....................... 58 Multiple Linear Regression Analysis and Trace Element Modeling ........................... 58 Complex Models ......................................... 72 P205 and Ti02 .......................................... 76 Extent of Mixing ........... ..... ....................... 77 Superheating of the Acidic Rocks ....................... 78 Origin of the Basic Rock Types ......... ............... . 78 Origin of the Acidic Rocks .................. ........... 78 ii Table of Contents (continued) Page smary 0...0...0....OOOOOOOOOOOOOOO...COOOOOOOOOOOOOOOOOOOO 81 References .00...0..0.0.0....O....0...OOOOOOOOOOOOOOOOOOOOOO 83 iii List of Tables Brief Field Description and Locality of Each Sample ....... ... ...... .... ..... ....... Analyses of Standards and Replications ...... Major Elements, C.I.P.w. Norms, and Trace Elements ............................. Major Elements From Blake (1966) and the Average Icelandic Pitchstone ..... .......... Pyroxene Compositions ................. ...... Multiple Linear Regression Analysis of btajor Elements .0...OOOOOOOOOOOOOOOOOOOOOOOO iv Page 37 40 43 46 47 List of Figures Figure Page 1 Location Map ............................. 5 2a Basaltic Xenoliths Surrounded By Acidic Rock ................................... 8 2b Basaltic Xenoliths with Cuspate Boundaries ............................. 8 2c Basaltic Xenoliths with Diffuse Margins .. 10 3 Generalized Pillow Types 1, 2, and 3 ..... 12 4a Type 3 Pillow with Cuspate and Diffuse Margin .................................. 16 4b Outcrop with all Three Types of Pillows .. 16 4c Closeup of Figure 4b Showing the Relationship Between a Type 1 and Type 3 Pillow ........................... 19 Sa Type 1 Pillow Surrounded by Material Similar to That of a Type 3 Pillow ...... 21 5b Highly Silicious Pillows ................. 21 5c Coarse Grained Acidic Rock Which is Found on the Bottoms of Many of the Pillows . ......... ....................... 25 6a Photomicrograph of the Contact of a Type 1 Pillow ........................... 29 V List of Figures (continued) Figure 6b 6c 7a 7b 9a 9b lOa,b,c 11 12 13a 13b Quench Texture Found at the Margin of a Type 1 Pillow ...................... Round Plagioclase Phenocrysts Found in the Centers of Many Type 1 Pillows ...... Type 3 Pillow Margin ..................... Texture of the Acidic Rock ............... Major Elements vs. Mg Ratio ....... ....... AFM Diagram of From the Austurhorn and Vesturhorn Acidic and Basic Complexes ... AFM Diagram From the Lavas Found in Eastern Iceland of Tertiary Age ... ...... REE Chondrite Normalized Plots ........... La vs. Sm ................................ (Ce/Lu)N vs (La/Lu)N ..................... Log-Log Plot of Sr vs. Ba for the Acidic and Basic Complexes .............. Log-Log Plot of Sr vs. Ba for the Tertiary Lavas Found in Eastern Iceland . vi Page 29 31 35 35 50 52 52 54 69 71 75 75 The Austurhorn and Vesturhorn Acidic and Basic Complexes in Southeastern Iceland: Examples of Magma Mixing INTRODUCTION The Austurhorn and Vesturhorn tertiary complexes of Southeastern Iceland contain classic examples of coexisting acidic and basic magmas which crystallized in a hypabyssal environment (Blake, 1966; Roobol, 1974). These complexes consist predominantly of acidic intrusions, but have located within them intimately associated adicic and basic rocks which have been referred to by many workers as "adicic and basic complexes" or "net-veined complexes." The net-veined complexes consist of an intimate association of acidic and basic rocks, in which the most conspicuous features are basaltic, pillow-like xenoliths with cuspate boundaries surrounded by acidic rock. A close association of acidic and basic rocks has been reported from a number of tectonic settings, on a variety of scales, and in both plutonic and volcanic regimes (For example: Holmes, 1931; Wilcox, 1944; Wager and Bailey, 1953; Chapman, 1962; King, 1964, 1965a, 1965b; Walker and Skelhorn, 1966; Philpotts, 1971, 1972, 1979; Wiebe, 1973, 1974, 1979, 1980; Yoder, 1973; Vogel and Walker, 1975; Vogel and Wilband, 1978; Holgate, 1978; Eby, 1980; McSween et al., 1979; Gamble, 1979; Taylor et al., 1980; Pallister, 1981; Pallister and Hopson, 1981). Specifically, in the Icelandic oceanic island rifting environment acidic and basic associations have been reported by Bunsen (1851), Blake et a1. (1965), Blake (1966), Walker (1966), Gunn and Watkins (1969, Roobol (1971, 1974), Sigurdasson (1971), 1 2 Sigvaldasson (1979), Jorgensen (1980), Prestvik (1980), and Sigurdasson and Sparks (1981). Review papers on acidic and basic associations have been presented by Walker (1966), Yoder (1973), and King in a series of articles (1962, 19633, 1963b, 1964, 1965a, 1965b). In many of these associations acidic and basic rocks are interpreted as representing the coexistence of two distinct liquids. Two fundamental models can be envisioned to account for the coexistence of acidic and basic magmas. The first has been termed commingling and is the failure of two miscible magmas to mix due to viscosity and temperature differences (i.e. rapid crystallization of the basic magma) upon coming into contact with each other. The second model is silicate-liquid immiscibility, whereby a homogeneous magma unmixes to form two thermodynamically stable liquids of a highly contrasting nature, one liquid essentially basic and the other acidic. The main objectives of this paper are to evaluate the two fundamental models pr0posed above and to suggest mechanisms of formation of the net-veined complexes consistent with the petrographic, major element and trace element data obtained from these complexes. FIELD DESCRIPTION The net-veined complexes occur within the Austurhorn and Vesturhorn Tertiary intrusions. These intrusions occur along the coast of Southeastern Iceland with an exposed outcrop area of 11 km2 and 19 km2 respectively (Blake, 1966; Roobol, 1974)(Fig. 1). They occur within 30 kms of each other and both have intruded basaltic lavas which gently dip to the west. These intrusions are composite stock-like bodies and were probably emplaced at depths less than 2 km, based upon the zeolite metamorphic facies of the intruded volcanics (Blake, 1966; Roobol, 1974). The intrusions are predominantly granitic rocks with subordinate basaltic and intermediate rocks. Tertiary acidic rocks in Iceland are almost exclusively associated with major volcanic centers (Walker, 1966). The net-veined complexes occur within the Austurhorn and Vesturhorn granitic intrusions and consist of differing amounts of granitic, basaltic, and intermediate rocks. The general field relationships and distribution of rock types within the Austurhorn and Vesturhorn intrusions have been reported previously by Blake (1966) and Roobol (1974). The net-veined complexes comprise only part of the total area of the intrusions, approximately 30-40% of the Austurhorn intrusion (Blake, 1966) and less than 10% of the Vesturhorn intrusion. The net-veined complexes occur in three main areas in the Austurhorn intrusion and two main areas in the Vesturhorn intrusion. The Austurhorn net-veined complexes have the best exposure. (See maps of Blake, 1966, Fig. 1;and Roobol, 1974, Fig. 2). The net-veined complexes appear to form at the margins of the granitic intrusions, although the Austurhorn could extend out Figure 1. Location of the Austurhorn (A) and Vesturhorn (V) intrusions. Rocks outside the dashed lines are predominantly tertiary in age. NEOVOLCANIC ' ZONE -----~ I I i I "Q at” I ~ ~--_ --------~ \ r 6 under the sea. All net-veined complexes consist of pillow-like xenoliths of basaltic rock with cuspate boundaries which are surrounded by acidic rock (Fig. 2a). In general, three different types of xenolithic pillows (Fig. 3) occur although gradations between pillow types do occasionally occur: Type 1 pillows are aphantic, often porphyritic and have fine grained margins and occasional phaneritic centers; type 2 pillows are aphantic and often porphyritic pillows which may or may not have a visibly fine grained margin. The gradation boundary is typically coarser grained than the basic pillow and may have a diffuse boundary relationship or a cuspate boundary in contact with the acidic rock. These basic pillows often display a gradational boundary with the acidic rocks. Type 3 are medium grained pillows with typical grain size approximately 0.25 cm. which have no fine grained margin and display boundaries which can be sharp to gradational with the acidic rock. Pillow types are defined for discussion purposes only and are not defined to carry genetic inferences. The pillows are generally rounded, although a few have high aspect ratios, and range in size from several millimeters to 3-4 meters across with an average of approximately one meter. Many pillows have cuspate boundaries with smaller cusps develOped on the larger cusps. Third and fourth order cuspate boundaries are not uncommon (fig. 2b), but are less common on Type 3 pillows. Pillow density within a given outcrop can be highly variable, but typically occurs as depicted in figure 2a. Two outcrop patterns are observed: those which contain only Type 1 pillows (fig 2a) and others which contain all three generalized types of pillows (Fig. 4b). Figure 23. Typical outcrop which occurs within the acidic and basic complexes which contains basaltic xenoliths, with cuspate margins, surrounded by acidic rock. Figure 2b. A Type 1 pillow with chilled margins, cuspate boundaries, and a gradational mafic mineral content in the acidic rock as one approaches the pillow. Figure Za Figure 2b Figure 2c. Type 2 pillows with a diffuse zone surrounding the interior pillow all which is surrounded by the acidic rock (pillow within a pillow structure. 10 Figure 2c 11 Figure 3. The three generalized pillow types (See text for descriptions). 12 I.‘ o . o...‘ wmm>h 30.3; 13 Type 1 pillows are aphanitic at their margins with a visibly chilled boundary up to several centimeters in thickness. Chilled margins are highly visible in the field because they are usually a darker color than the rest of the pillow (Fig. 2b). These pillows gradually increase in grain size toward their centers. Some pillows have a well developed visible chilled margin, several centimeters in thickness, on one side of the pillow and a homogenous and fine-grained, margin the other. No systematic orientation of chilled margins was noted. The distribution of phenocrysts within Type 1 pillows is erratic and variable and usually no more than a few percent are present. Phenocrysts consist of pyroxene and plagioclase at the Austurhorn and plagioclase only at the Vesturhorn. In the acidic rock adjacent to many of the pillows there is an increase in the abundance of mafic minerals present as one approaches the pillow (see Fig. 2b). This feature gives many of the outcrOps a mottled texture within the acidic rock. The second type of pillows often have a diffuse zone surrounding them which is lighter grey in color and coarser grained with cuspate boundaries. The darker pillow interior also has cuspate boundaries and increases in grain size from its margin to its interior. The darker pillow interior may or may not have an obvious chilled margin. The diffuse zone, surrounding the darker pillow, does not always occur around the entire pillow. The diffuse zones often vary in width. This relationship of a more mafic pillow interior surrounded by a diffuse or hybridized zone, all of which is surrounded by the acidic rock (Fig 2c) has been termed a skialith (Chapman, 1962) or a "Pillow within a pillow" structure (Taylor 9t 31': 1980)‘ Type 2 14 pillows are often porphyritic with plagioclase and any individual pillow may contain a highly variable phenocryst content (usually less than 5%). Figure 2c also shows the highly mottled texture of the acidic rock as well as many smaller xenoliths which can have cuspate, angular, or diffuse margins and which probably represent "broken" fragments of pillows. The Type 3 pillows are coarser grained than the Type 1 or Type 2 pillows, and when present the diffuse zones surrounding them are of similar appearance and grain size to the zones surrounding the Type 2 pillows. Type 3 pillows have cuspate and angular boundaries often on the same pillow. This feature can be seen on other pillow types, but is more commonly seen with the Type 3 pillows. Gradational boundaries or diffuse zones also occur around the Type 3 pillows (Fig. 4a). Type 3 pillows can be porphyritic with respect to plagioclase, but typically are hypidiomorphic-granular (about 0.5 cm) and non-porphyritic. Higher concentrations of phenocrysts are occasionally observed in the Type 3 pillows than in other pillow types (Fig. 4a) especially in proximity to the acidic rock. In figure 4a (lower half) a pillow of Type 3 can be seen to be separated from a pillow of Type 1 by a thin lense of the acidic rock. The bottom boundary of the Type 3 pillow has a sharp cuspate margin. The upper boundary of the Type 3 pillow grades into a more leucocratic rock which in turn grades to the left and right (out of the photograph) into the typical acidic rock. At the top of the photograph is another Type 1 pillow demonstrating the intimate nature and occurrence of all pillow types in close proximity in many of the outcrops. It is emphasized that these pillow types are gradational and are 15 Figure 4a. Type 3 pillow with a gradational boundary at the t0p and a cuspate boundary at the bottom in contact with the acidic rock. Figure 4b. Large Type 3 pillows with angular and cuspate boundaries. If the pillows were fitted together in a puzzle-like manner, they would form a single Type 3 pillow with cuspate coundaries. Type 1 and Type 2 pillows are also present. 16 .;. ,,n . M‘ L" V ‘wg £332.31: 343:3: ‘, Wu Figure 4a and 4b 17 classified only to facilitate their description. Thus it is not suggested that these pillow names be carrried on to other localities. Figure 4b depicts a not unusual case in that at any given outcrop all three pillow types can be found. In general, two types of outcrops occur most often within the net-veined complexes; those which contain only Type 1 pillows and those which contain all three types. Figure 4c is a close-up of figure 4b (labelled A) and shows a Type 2 pillow in close proximity to a Type 3 pillow. The Type 1 pillow has a chilled cuspate boundary and is porphyritic with respect to plagioclase. The Type 3 pillow is coarser grained and has cuspate boundaries; although Type 3 pillows typically do not exhibit as high order cusps as Type 1 or 2 pillows. This maybe due to viscosity and temperature differences in the contrasting magmas. Many of the Type 3 pillows have cuspate boundaries on one side and angular boundaries on another. Other pillow types display this same feature, but not as commonly as the Type 3 pillows. The large cluster of Type 3 pillows in figure 4b appear to have broken up into smaller pieces by "brittle fracture and the acidic rock intruded along these fractures. The angular boundaries on the pieces of the clustered pillows could fit back together in a puzzle-like manner and form a single large pillow which had cuspate boundaries on all sides. This would seem to indicate that the type 3 pillow crystallized, broke apart and then the acidic liquid was intruded along these fractures. The complexity of events in the magma chambers can be elucidated by the presence of a Type 1 pillow with chilled boundaries 18 Figure 4c. Close-up of Figure 4b which displays the intimate association of a Type 3 pillow with a Type 1 pillow (only a thin lens of acidic rock separates the two). 19 Figure 4c 20 "sandwiched" within the fractures of the Type 3 pillow described above (Fig. 4b). With these field relationships the Type 3 pillow would have crystallized, fractured, and been intruded by a still liquid-plastic Type 1 pillow and the acidic rock. Other field relationships also indicate a complex sequence of events occurred in the magma chamber. For instance, the presence of Type 1 pillows with zones of material similar to the type 3 pillows surrounding them, all of which is surrounded by the acidic rock. In figure 5a are Type 2 and Type 3 pillows as well as a Type 1 pillow with a chilled boundary which is apparently surrounded by a material indistinguishable from a Type 3 pillow. The Type 1 pillow has chilled boundaries in contact with the intermediate rock on two sides and the acidic rock on another side. Whatever the origin of these rocks'are, the field relationships indicate the presence of several liquids coexisting at the same time, apparently all at different temperatures and viscosities. Other complexities are also found. Figure 5b shows pillows surrounded by the acidic rock. Both the pillows and the acidic rock are highly silicious with 68.9% 8102 (Ic-38b) and 73.0% SiOz (Ic-38a) respectively. These pillows were only observed at one occurrence in the Austurhorn intrusion. No apparent chilled margin is present on these rocks. The rocks have similar mineralogy, but differ texturally. The dark pillows lack the abundant granophyric texture of the acidic rock. The presence of acidic pillows with cuspate boundaries surrounded by acidic rock would seem to indicate a viscosity difference between these two compostions. The three dimensional nature of these pillows cannot be observed. 21 Figure 5a. Type 1, Type 2, and Type 3 pillows all in intimate associations and surrounded by the acidic rock. Note the Type 1 pillow surrounded by material similar to a Type 3 pillow all of which is surrounded by the acidic rock. Figure 5b. highly silicious pillows found at the Austurhorn intrusion. The three-dimensional nature of these pillows is not exposed. 22 Figure 5a Figure 5b 23 In addition, on the underside of many of the pillows, zones of coarser grained acidic rock are found (Fig. 5c). The zones are porphyritic with phenocrysts of plagioclase up to 0.75 cm. The macroscopic texture of the acidic rock is variable, but usually displays plagioclase phenocrysts which are often surrounded by graphic textures or, less commonly, hypidiomorphic textures. Within the net-veined complexes the acidic rock ofter displays a mottled texture due to the variable mafic mineral content (see fig. 2b, 2c). The acidic rocks locally contain many vugs within the net-veined complexes, and in the main body of the intrusion. The main body of the Austurhorn intrusion contains a variable plagioclase phenocryst content (from a few percent up to 43%, Blake, 1966) similar to the net-veined complex. Textural differences may represent variable mafic components or changes in the physical parameters of crystallization. There are angular xenoliths which are texturally different from those previously described found only at the Austurhorn intrusion. They are coarse grained (up to 1 cm) and have sharp or angular boundaries with the acidic rock. These xenoliths occur only in proximity to the margin of the acidic intrusion. In summary, both the Austurhorn and Vesturhorn intrusions are predominantly composed of acidic rocks. Within these intrusions there are net-veined complexes which consist of an intimate association of granitic and basaltic rocks. Xenoliths which are, generally, pillow-shaped bodies with cuspate boundaries of basic composition are surrounded by acidic rocks. The pillows are variable with respect to size, composition, and shape. Many pillows have 24 Figure 5c. Coarse acidic rock which underlies many of the basic pillows. Also, note presence of many small mafic xenoliths within the acidic rock. 25 Figure 5c 26 chilled margins. These field relationships indicate a complex history of intrusion. The acidic rock is relatively homogeneous outside the net-veined complexes and heterogeneous with respect to their mafic mineral content within the net-veined complexes. The field data is interpreted to indicate the presence of several liquids of differing compositions and crystallization temperatures which are contemporaneous in the same environment. 27 PETROGRAPHY A petrographic description of all exposed rocks at the Austurhorn and Vesturhorn intrusions has been reported previously, by Blake (1966) and by Roobol (1974) respectively. This report will, to avoid redundancy, deal almost exclusively with the net-veined complexes. Similar pillow types have similar mineralogy and textures. In both of these intrusions the acidic rocks are texturally similar but vary most importantly in their mafic mineral content and the amount of plagioclase phenocrysts. Type 1 pillows (Fig. 6a) near and at their margins consist of plagioclase, poikilitic biotite, opaques, poikilitic amphibole, augitic pyroxene and some biotites; augitic pyroxenes may be up to 1.5 mm across. In many chilled margins an occasional swallow-tailed plagioclase with a high aspect ratio is present. In addition, many Type 1 pillows at their margins have what is interpreted to be another quench texture (Fig. 6b), that of radiating acicular opaques presumably of a continuous crystallographic nature. This texture occurs frequently with the swallow-tailed plagioclase. These quench textures occur only at pillow margins. Type 1 pillow interiors consist of plagioclase, pyroxene and opaque grains and often have a ophitic texture. Some of the larger pillows have a equalgranular textures with grains up to 2 mm. Plagioclase as determined from microprobe analysis, have cores which range up to An85 and rims down to An32. The large plagioclases found at the Vesturhorn intrusion are often corroded and rounded by resorption (Fig. 6c). both complexes have Type 1 pillows which are porphyritic 28 Figure 6a. Photomicrograph of the contact of a Type 1 pillow with the acidic rock (7mm across photo). Figure 6b. Quench texture, consisting of radiating acicular opaques, found at the margins of many of theType l pillows commonly along with swallow-tailed plagioclase (3mm across photo). 29 Figure 6a Figure 6b 30 Figure 6c. Large plagioclase phenocrysts found in many of Type 1 pillows. Many phenocrysts are rounded and embayed (12mm across slide). 31 Figure 6c 32 with respect to plagioclase which are up to 3 mm across (Austurhorn) and 7 mm across (Vesturhorn). At the Austurhorn pyroxene phenocrysts are up to 1.5 mm across in the Type 1 pillows. Phenocrysts are not abundant in either suite; less than 2% at the Austurhorn and less than 5% at the Vesturhorn intrusion. Grain size gradually increases from the margins to the centers of the Type 1 pillows (Compare Figs. 6a,6c). The effects of hydration are also less visible in the interior of the pillow compared to the core, with the amphiboles and biotites that occur at the pillow margins and pyroxene occurring in pillow interiors. Contacts with the acidic rock are typically sharp (Fig. 6a). The Type 2 pillows are similar to the Type I pillows in their textures and mineralogy. Chilled margins may be present or absent. In general, grain size increases from the margin to the interior of the pillow. The hybridized zone surrounding these pillows are typically coarser grained, equalgranular (approximately 0.5 mm), and consist of varying amounts of plagioclase, pyroxene, amphibole, biotite, Opaque phases, and graphic intergrowths of quartz and feldspar. The contacts of these diffuse zones range from being sharp to gradational over 10 to 15 cm. Plagioclase compositions are variable and range from Ango down to andesine rims (An37). The Type 3 pillows exhibit a range of mineral constituents and contain plagioclase, opaques, widely varying amounts of graphic intergrowths of feldspar and quartz, amphibole, biotite or both, and pyroxene. Pyroxenes often have rims of amphibole or biotite. Contacts of these pillows may be sharp or gradational (Fig. 7a). Figure 7a shows a contact of a Type 3 pillow where mafic phases seem 33 to have crystallized and grown larger at the pillow boundary more readily than other phases. These pillows are predominantly hypidiomorphic-granular, with a few plagioclase or amphibole being up to 7 mm in size at the contact, with the typical grain size being 2-3 mm throughout the pillows. The acidic rock exhibits a range in textures and mineralogy. The most typical rock type consists of plagioclase, biotite, quartz, minor opaque phases, and graphic intergrowths of feldspar and quartz. The mottled texture of the acidic rock in the net-veined complexes is due to a variable mafic mineral content of amphibole, biotite, opaque phases, and variable amounts of plagioclase phenocrysts, graphic intergrowths, and quartz. Blake (1966) reported that the acidic rock near the pillows often has hypidiomorphic granular texture. However, although this texture is present locally, no systematic distribution of this texture was observed. Figures 6a and 7b show the typical textures of the acidic rock. The plagioclase phenocrysts are zoned, often rounded (Fig. 7b) and/or rimmed by turbid feldspar, and often surrounded by graphic intergrowths of turbid feldspar and quartz. Plagioclase cores ranged up to An47 Ab450r7.5. The turbid feldspars and the turbid feldspars in the graphic intergrowths are cryptoperthic. Most of the plagioclase phenocrysts are less than 0.5 cm, but a few are up to 1.0 cm. in length. Opaques exhibit an exsolution texture and probably consist of ilmenite and magnetite. The coarser grained zones of acidic rock located on the undersides of many of the pillows contain plagioclase phenocrysts up to 0.8 cm in length and euhedral amphiboles, opaques, zircon, and 34 Figure 7a. Type 3 pillow contact with the acidic rock. Note the abundant mafic minerals at the margin and the rounded plagioclase crystals in the acidic rock (7mm across photo). Figure 7b. Typical texture found in the acidic rock which consists of plagioclase phenocrysts (often rounded and rimmed) surrounded by graphic textures. The acidic rock contains variable mafic inerals (biotite, amphibole, and Opaques). 35 Figure 7a Figure 7b 36 sphene. Some of the zircons are up to 0.2 mm in size. Alteration in the net-veined complexes can be locally abundant typically along joints and fractures, but is usually of minor extent. The acidic rock alteration locally has abundant epidote group minerals, chlorite and rarely calcite. Alteration in the basic rocks is typically minor, with alteration phases of chlorite and/or epidote group minerals. Areas which displayed abundant alteration were avoided during sampling. SAMPLING AND ANALYTICAL METHODS Table 1 gives a brief description of each sample which was collected and analyzed, as well as the locality and any characteristic mineralogy or textural features. Samples were collected in the field to reflect the maximum visual variation present and thus do not represent volumes. Detailed sampling of individual pillows or pillow sequences (eg. pillow interiors, pillow margins, and the acidic rock in close proximity) has only been characterized by samples Ic-3 through Ic-9 (Austurhorn, See Table 1). Sample numbers less than forty are from the Auturhorn intrusion and those greater than forty are from the Vesturhorn Intrusion. Samples were collected from the Vesturhorn Intrusion at both the eastern (Ic-53a,b,c) and western (Ic-59 through Ic-77) occurrences of net-veined rocks as depicted on the geologic map of Roobol (1974, Fig. l). The Austurhorn Intrusion was sampled at all of the ocurrences of net‘veined complexes except a small 0.3 km2 complex located at the far northwestern end of the intrusion (see Blake, 1966, Fig. 2). All samples are from the net-veined complexes of the two intrusions. 37 Table 1 Sample Field Description Major Mineralogy or Texture Number IC-3 A Chilled margin of a Type 1 Plagioclase and pyroxene pillow from lower pillow phenocrysts in Figure 4A Ic-9 Type 3 pillow with diffuse Mineralogy of plagioclase, Ic-4 zone surrounding it. Type biotite, graphic patches, Ic-S A 3 pillow in Figure 4a from quartz, : amphibole, :_pyroxene Ic-6 t0p (Ic-9) to Bottom (Io-7) plagioclase phenocrysts Ic-7 Ic-8 A Type 1 pillow margin from Similar to Ic-3 top of Figure 4a. Ic-lOa Type 3 pillow (Ic-lOb) and Acidic rock has biotite, minor Ic-lOb A acidic rock surrounding it opaques, and phenocrysts of (Ic-lOa) plagioclase Ic-12 Type 1 pillow margin Pyroxene and plagioclase Ic-l3 A (Tc-13) and interior phenocrysts (Ic-12) Ic-l4 A Angular xenolith (see text for probable origin) similar in texture and mineralogy to a type 3 pillow Ic-ZO Taken from large outcrop- Both contain biotite, minor Ic-21 A pings of acidic rock opaques and plagioclase pheno-(3m2) within the phenocrysts surrounded by acidic and basic complexes graphic textures Ic-22a Type 1 pillow margin Acidic rock is similar to Ic-ZO Ic-22b Ic-2(Ic-22a), interior Ic-22c A (Tc-22b), and acidic rock surrounding it (1c-22c) Ic-29 Type 2 pillow diffuse zone Acid rock has hypidomorphic Ic-30 A (Io-29) and acidic rock texture with patches of graphic surrounding (Tc-30) texture Ic-31 A Type 2 diffuse zone Similar to Ic-20 with more abundant biotite Ic-28a type 2 pillow at margin Ic-28b A of gradational diffuse zone (Io-28b) and 25cm into pillow (Ic-28a) 38 Table 1 (continued) Sample Field Description Major Mineralogy or Texture Number Ic-37a Acidic rock from same Similar to Ic-20 except Ic-37b Ic-37b A pillowed outcrop with all is more leucocratic than pillow types present Ic-37a Ic-38a Highly silicious pillow Ic-38a is similar to Ic-20 and Ic-38b A Ic-38b surrounded by acidic Ic-38b, Ic-39 has hypidiomorphic Ic-39 rock Ic-38a. Ic-39 is from texture with minor patches of pillow next to Ic-38b (see graphic texture Figure 5b Ic-4l A Acidic rock collected from Similar to Ic-20 margin of acidic and basic complex Ic-4O A Angular, coarse grained, Contains coarse plagioclase, gabbroic xenolith from pyroxene, Opaques, and minor near wall rock of the apatite intrusion Ic-53a Ic-53b V Type 1 pillow margin 3cm Ic-53a and Ic-53c have large Ic-53c (Tc-53a) and 89cm (Ic-53c) plagioclase phenocrysts and from contact with acidic acidic rock Ic-53b is similar rock which surrounds the to Ic-ZO pillow Ic-59 V Acidic rock from 5m2 Similar to Ic-20 area, free of pillows within the acidic and basic complex Ic-6l V Type 1 pillow interior Pyroxene, plagioclase, and Opaques Ic-66 V Type 2 pillow from 10cm Subophitic texture Ic-67 V (Tc-66) and 30 cm (Ic-67) of contact with acidic rock Ic-68b V Acidic rock near Type 2 Ic-68b is similar to Ic-20 and Ic-68c pillow interior (Ic-68c) Ic-68c is similar to Ic-6l Ic-74a Type 2 pillow, Ic-75 is Ic-74b from the interior of the Ic-74c V pillow and Ic-74a is from Ic-75 the contact with the acidic rock. Ic-74b if 3cm and Ic-74c is 8 cm from the margin Ic-77 V Type 1 pillow interior Similar to Ic-6l 39 Before chemical analysis each sample was examined petrographically and samples which had abundant secondary mineralogy or miariolitic cavities were not used. Samples Ic-38a,b are exceptions in that alteration along two fractures spaced approximately 2.5 cm apart were present. Due to the numerous mafic "inclusions" in the acidic rock, each piece was sliced to a thickness of 1.5 cm. and any mafic inclusions were ground or cut from the sample. Samples were then crushed and ground to a fine rock flour between porcelain plates. Forty-three samples were analysed for major elements; twenty-nine samples were from the Austurhorn Intrusion and fourteen samples from the Vesturhorn Intrusion. The analyses were performed by Barringer Magenta by inductively coupled plasma emission spectrometry (I.C.P.E.S.). Eight trace elements were also determined by the same methods and they include Ni, Cu, Zn, Cr, Co, Sr, V, and Ba. Table 2 lists the standards (mrg-l and sy-2) as unknowns and lists the averages for these standards as published by Abbey (1978). Replicate analysis of Ic-20 and Ic-6l were also run and these are listed in table 2 for comparison. Methodologies and error analysis using the I.C.P.E.S. methods has been evaluated in geological samples by Dahlquist and Knoll (1978). Ferrous iron was determined by the authors using a wet chemical technique described and tested by Weis (1974). Thirty four samples were analysed for Na, La, Ce, Sm, Eu, Tb, and Lu by the authors using instrumental neutron activation analysis (INAA) methods and the techniques of Gordon et al. (1968). Th was analysed by INAA using the techniques of Korotev (1976). Twenty-five 40 MRG-l SY-Z 1c-20 Ic-él 1 2 3 4 5 6 7 8 5102 40.8 39.24 60.7 60.09 73.9 74.3 48.9 50.7 T102 3.89 3.75 0.13 0.15 0.35 0.36 3.51 4.34 A1203 8.56 8.56 12.6 12.15 13.8 14.2 12.3 13.2 13:203 17.4 17.80 6.18 6.29 3.11 3.13 13.4 13.6 MnO 0.15 0.17 0.29 0.32 0.04 0.04 0.18 0.18 M30 12.1 13.51 2.62 2.69 0.35 0.35 4.79 4.76 C80 12.3 14.72 7.39 8.00 1.10 1.10 6.79 7.80 Nazo 0.82 0.71 4.17 4.35 4.57“ 4.579 3.07” 3.07’ K20 0.71 0.18 4.47 4.51 3.56 3.61 1.04 1.13 P20, 0.04 0.07 0.36 0.44 0.03 0.02 0.40 0.42 L.O.1. 1.1 - 0.9 - 0.9 0.9 0.60 0.70 C02 - 1.01 - 0.46 - - - - 1'120 - 0.99 - 0.46 -— - - - TOTAL 97.87 100.71 99.81 99.91 101.71 102.58 94.98 99.90 N1 179 200 18 10 5 7 30 38 Cu 115 135 2.5 4.0 4.7 '4.8 45.1 52 Zn 173 185 232 250 43 50 65 63 Cr 315 420 22 110 9.3 8.5 7.4 21.1 C0 104 87 23 10(7) 3 3 53 68 Sr 244 260 250 270 67.4 80.8 311 361 V 491 520 49 50 17 .1 16.8 343 365 Ba 1 55(?) 410 460 510 520 90 90 l Analysis-Barringer Magenta (I.C.P.E.S.) 5,6,7,8-Bannger Magenta (ICPES) 2 Analysis-Average of S. Abbey (1978) 3 Analysis-Barringer Magenta (I.C.P.E.S.) a Analysis-Average of 5. Abbey (1978) * Average Na Values from I.C.P.E..S. and Neutron Activation (this study) TABLE 2 41 samples were run from the Austurhorn Intrusion and nine samples from the Vesturhorn Intrusion. Where Na was determined by both INAA and I.C.P.E.S., the values from INAA. were used. Major elements, trace elements, and C.I.P.W. norms are reported in table 3. Table 4 lists the reported major element analyses for acidic rocks and basic pillows from the Austurhorn Intrusion as reported by Blake (1966) and the average Icelandic pitchstone as reported by Walker (1962). Pyroxene compositions (Table 5) and plagioclase compositions were determined using an ARL EMX microprobe. Operating conditions of 15Kv and a sample current of 20 mA were used for most elements except Ti, Mg, Fe, and Mn for which 20Kv was used. A defocused beam of 0.4 u. was used for most analyses. Feldspars in the acidic rocks were cryptoperthitic so a wider beam was occasionally used. PETROCHEMISTRY Major Elements: Due to the complex chemical variation in these rocks, for ease of discussion, they have been separated into three main groups: acidic rocks (greater than 65% $102), intermediate rocks (55% to 65% 8102), and basic rocks (less than 55% $102). C.I.P.W. norms (Table 3) show two interesting features. Within the basic rocks none of the rocks are olivine normative and within the acidic rocks many of the rocks are corundum normative. Plots of the major oxides versus Mg ratio indicate that the variation of the major elements is similar in both of the rock suites except in T102 and P205 (fig. 8). Ti02 is generally higher in the most basic rocks found at the Vesturhorn net-veined complexes (up to 4%) than at the Austurhorn net-veined complexes (up to 2.89 g0). P205 is 42 Table 3. Major elements, trace elements and C.I.P.W. norms. 43 can «0. a.~ ~8- man 0.: o.a n._ ..a n.o «a an. ex - O I I ago—anon ...‘z sasazasazas=sz=== - .0.- 0A.. -°.~ 8.0 3.0 ”fl 3.. :6 0n.— ”.0. a... a... a: no. can u..— a: «.3 5 .3 0.- ..~ 0.0 0.: ... as;ags=q:==ase::= 2’:.~'“OG 'g'n'c'ég' o... 3.. an; 2.: ~90 a~.~ a... :8 uni an... s... a. a: a: ... ..8 .... ... .8 2. .o 3 a. .... ... ... 3. a 2 a... ... ... 2 O. 2 0.. ... ... ... ... ... ... ..~ , ... ... ~.. 0.. . .nn 2 2 3. 3. 2. a. 3 a «Ugandaanum ...... 3.8 «.... 8.8 3... ...: .... a... .... .... .... 8.. 8.. .... 3.. ... ... ... 2d .... .... a... .... 3.. a... 3.. a... .... 8.. .... .... 3.. 8.. ...e a... .... 3... ...... a... .0... 3.9. 3... 3... 8... ...: 3.. .... .... 3... 2.8 3.8 34% 3 ad a.” .... .... .... .... 3.. 3.. .... 3.. 3.. .... n... .... 8.. o..~ 36 00.. a... :6 a... O... 3.. 3.. ...~ ...~ 0... .6 3.. 3... on... o. ... .... 3.. .... a... .... u... «M. mi... 22”.... ... .2 .... .... ... ... a .. ... .... ... ... ... .... . 2 ... .. ... .- ... .. ... I z I ... 3 .. .. ...: ...: 8... ...: .... .... .... 3.. .... .... a. on. ... .... .... .... .... .... 3. 3a 3. 3. 8.. .... .... .... an... 3.3 00.: «a... .... .... 2... 2... .3... a..." .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... 2.. 3.. .... ...: 2... .... .... .... a... ”a. ...»... .2 8 .... . x. a .. cu .... .O .. ... ... ..O : ... ... ... ... ... ... .. ... ..u A. ...n ... no a 3 so. = as ...8 93.. .... ...... 0.. 3. .... ..O .. . .... .... ... .... .3 .... ... .... z. .... Ob .... a: 8... 2.. .. ... ... a... ..O .... u a... O 3."! “fig .... ...O... .... . «O... .... O... 2.. of... a .. O6 .... O... .... Oca .... O... .... .O... ...: .On... .... ~O... ... z .2. a 9.. «.0 a... 3 o.- a. «.0 ... a.- Q.— a.- 0. ca. ~a~°- h.- 3.3.3.3833?! - 3 3.. ...n =.« a... an .0 3.0 8‘ a... Co... 00.. i... a: 0.: u: 0.: 0.. ... s.- n.~ 0.: a: 3 389948384 aeaa~---- 33 a... .06 8.0 a... 3.. a... as.- on... 00.0. no .0 a... :0 u... 0.. an a... :- q.~ a. ... o.— o.~ ..n u. u: 2 .3 ..a. ... ~.. in ... o. .2 ... ... ... .... .... ... .... .... ... . . .. ... . .. ... a .. ... ... ... . .. o . . «.3 n .0. 1 . .. ... .. ... ... .. ... ... .. ...... .... .. ... ... -- a 3 -. mafia ...: ...: .... .... ...: a... .... .... .... .... .... .... .... .... .... ... 3. ... ... ... .... .... .... 3.. a. .... .... ... ... ... .... .... .... .... .... .... .... .... .... ...: 3... .. ... S... ...! ...: .... ... .... .. ... a... a. .. flag 3 33 a 3.. .... .... .... .... .... .... .... .... 00.0 3.0 00.0 .... .. .. .... .... 2 .. .... .... .... .. .. .... .... .... .... .... .... I... ...: ...... .... z .. z .. a... ...: a .2 .ma .ma .nm «gag... ... ... .2 3 . ... ... .... . ... .... .... a .. a a co ... ... ... .O .. S 2. ... ... ... ... ..O . . o .z .. ... . . .. ... .. ... ... ... .. ... ... .. I ... ... a. .. 2 s .... .. ... ... oO .. .. ... 3 1.... .... .... 2.3.3.3.. 8.3.... .313. 8.5:... 38......) . ...... . ...: ... .. .... .O... ...: .... .... O3. .... .... .... ...... .... .... 2.. ...: 8.. 8.. so. 8 ... ... ... .¢ .... .... 3.. d .... 2.. a a .... .... ... .. .. ... :0 :0 - o z. .... .... a .. Op .... .... .... .2 .... .... .... 2.. ...: ...: a... ... ...: a... 2 ... ..o .... .... 8.. O ...: ... .. .. ... O «a ”in a JqHOH .... .... .... ...OJ .... .... .... .O... .... .. .. .... O... a... .... .... OP: .... .... .... O6 .... .. .. .... O... .... .... .... O5. 2.. o. .. .... O... .... .... .... no... ...: ...: ...: .O..< .... .... .... .O: ...: a... ...: .Oa “Mal“ a «an! E93F§asaii=800 5%: P‘VS’QPE’ZngPI'C‘F 76.9 7.30 I0.60 7.06 6.06 0." 0.06 0) I0.0 0.0 L7 0., 0.] 'O’ l).’ 44 W 70.0 70.0 37.7 7.00 7.07 7. 3| ".60 I0.” ".60 1.” 3.63 1.28 6.“ 6.33 7.82 0. [7 0.70 0. |0 7.00 7.03 3.00 5.00 7.30 7.00 L00 0.13 3.00 I.“ L70 1.09 0." 0.06 0.6) 0.76 0.07. L“ 192.11 m 21.21 W I0.0I 6.76 U.“ 0.“ 0.“ 0.“ [0.70 I0.” 0.00 ”.05 M) ”.11 I3.” [7.0. 0.“ m m w 0.” I.” 6.00 7.67 0.70 0. l0 6.60 6.93 0.66 ”I 0.9' 3.1) L” 0.” L68 I” L50 m 0.“ 7.61 7.03 0.81 7.30 I0.” ”.21 a.“ 20.77 17.1) 0|.l0 03.35 03.6) W - ,1 - — I) - — 0.0 - - .0, - - ‘0. ‘- 1.’ 0.) 7.. I7 70 I, no.1 32.. IL: 7| I“ ” ”.0 “.6 7., )3 67 )1 MO 1). 30 I50 I93 I“ 200 060 2'0 ”J 1.7, I0.” I. I, 6.67 0.“ ). II 6.06 ).90 L67 6.66 L00 7) 0.) 7.7 0.7 7.8 l6.7 I) “7 38 I0 0.6 7.! 0.0 1.0 l7 19.7 liézyn 45 9.. ss- nna .. o.~n A.- ~-- 0. ..~ n.. ... o.~ n.- on. ~r~ ..nu . a. n..~ ... ~.— ..~ n.o ... ~.~ 0.. o- n. on. 9.. no. nn~ no~ ... ... ... an a. no o.~ a..n o... .. ... .. .... ... a .. .. .. ... ... ... ... ... ... o.- o.- o.- .. ... ... . z . no on no a. mu 0. «Hammad—dug 8... 3... I... 2... x... ...: ...: ...: .... ...: ...: ...: .... .... .... .... ... ... .... .... .... .... .... .... .... .... .... ...: ...: ...: .... .... .... .... .... .... ... .... 3... ... ...: ...: .. .... .... .. .... .... .... .... .... 3 and an an a... .... . Q..- .o.. on.. an.. a... u... a... ~0.n no.n -.a .... .... .... .... .... .... a... a... o... .... .... .... I... a... ...— ...n- .-.a- cu.n- .... .... .... .... .. .. .. .. ”an” “a "ME ..u .an him no ..s~ oo- .... ..n s.. n.- o.n .... a. ~- .~.. On.— -.~ 0... -.o as.- a... 0:... -.~ a... any 3.33.3:326 any 0a ...-(lug ...o... o .. a x 6.... Avon .0... .o..< 0.- 51¢ S!§§!%83%§%!§§!€4 ~.~-..“~‘.~ .~.-~ a- - - an {5:83 --.- .... .... .... .... .... 3... .... .... ”NM. 283388S8333232282 I a O o I o O o I I O I O a I can ... ... _.. ..a ';-:: :gsaxaazzax=s=aaa a... no. o~. a. ..o n.. n.- ~.n ... zaaaxxsszaazaaacfi ... 8 ... 3. ... ... ... ... ... ... ... ... ... ... ... .. .. .. .. .. .... .... .... .... .... ... ... ... ... ... .... ... .... .... .... .. .. .. .. .. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... .... .... .. a .. .. .. .. .. a .. 2 g ...-1...... ...-......2 .... .o 1...... z . ...-...... so]... I z... a... «.... 3... ..... z... ...: .... ...: 2... ...: ...: ...: .... ...: ...: ...: ...: ...: ...: .... .... .... .... .... ... ... ... .... ... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... ...: ...: ...: .... ...: .... .... .... .... .... .... .... .... .... .... ...: ...: ...: ...: ...: 3... ...... ...: ...: .... .... .... .... ...: .... .... .... .... .... .... .... .... .... .... .... 3... ...... ... ... z... ...... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... s... ...n .o.— a... a... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... ...: ...: ...: ...: 2... .... .... .... .... .... .... .... .... .... 3... mg n.na an n.— ... u.- o.o a... as .... .... .... .... .... .... I... .... 8... 35 I5 is ID 15 «I a» I: 8- 8d An in :63 1.:a0.i_uauannv .. saga-.0 ...-gnu . cg "Xxds 43.1...“ ...o... ...? o x o .z 0.0 o... .0... .o..< o: an 46 wam Coma E mafia .6523: >_NON _N.N 3% EN 3.3 SN 3% N... mmNOu N... NL NL N.Nc N.m N.N T— mmO 7m _.v w._ PNN Fm Nb Te 2:0 93 9o... 95 N... PB 9; o.— ZmO 0.; o.Nw on To: Pm Pm ob OmO TNm TN. N.au u.NN N.N N.N Te ZmNO uL 9N 9N FE. u.N a; Fm KNO 9m u.N N.N N.NN To T: N.« mNOu oLN Pom ohm 0.; ohm 9.5 0.3 INO .mu ... .5 .3 7w Tu Fm Hon: So... 59o Bob 3.3 Eek 50L 593 flaw—h. ¢ 47 rub-— _n-uuN _n..m.._ _n..m-N Tmuu EL uuN .9.— wnu >33.me cm @3983... .33 .33 _ ....Eoémz 3:35.703 03$ 48 also generally higher at the Vesturhorn net-veined complexes (up to 0.8% as compared to 0.47%). Both suites, however, have enriched values of T102 and P205 in some of the intermediate rocks. This is most pronounced with respect to P205. The angular xenolith analysed (Tc-40), which differs mineralogically and texturally from the pillows described, is enriched with respect to total FeO and P205 compared to all other rocks found at either intrusion and is not included on the variation diagrams. Both suites of rocks exhibit a "gap" in MgO. Figure 9a is an AFM plot for all of the rocks from the acidic and basic complexes found at the two intrusions which illustrates the presence of this "gap". If these rocks are compared with typical variation of the lavas in the Tertiary of Eastern Iceland which also exhibit a gap (Wood, 1978), the size of the "gap" is greatly reduced (See figure 9b for comparison). Recent examination of the rocks present from the 1875 Askja eruption indicate that no "gap" is present in MgO (Sigurdsson and Sparks, 1981). Thus the end members of the rocks found at the Austurhorn and Vesturhorn acidic and basic complexes are similar to the volcanic variations found throughout Iceland. However, the intermediate rocks found are apparently less common, but do occur (Walker, 1966; Sigurdsson, 1977). The AFM plot (fig. 9a) also indicates that primitive high MgO rich rocks found in many volcanic sequences in Iceland are not present at the intrusions under discussion. Trace Elements The Chondrite normalized REE distribution patterns indicate that all of the rocks under study are LREE enriched (Fig. lOa,b,c). Acidic 49 Figure 8. Major element variation of the rocks from the Vesturhorn and Austurhorn acidic and Basic complexes vs. Mg. 50 .00 HEW o o 8. ... 8... 8... IS. an ocu m ow .m m ‘ m m #2 o n... W .n.. 8.10 9... 22...! [oodu 02: w 1L N N R R m... ... W. . W W D@ rm... fa rl Mu Mu mum mm 0 AH V QmW m D O D 31.. mg $3 3 w 8.... 8.... 8...... 8.20 Noam M60 M00 M00 8.00 mm. mu mm... mmwo.m m mu .w mom a mom a mu mu mm gm m. 8... 8...! 8... 8&u omcz mm mm.m .m m. . m m .m 8... 8... 8.... 1...... momma 4.00 2.00 1.00 M00 M00 M00 I ' o .00 V 6.00 2.00 % m. no G mu d 8... 8... Rocha». 00.. on. own $1... H00 MGD H00 Figure 8 51 Figure 9a. AFM plot from the Austurhorn and Vesturhorn acidic and basic complexes. Figure 9b. AFM plot of the lava variation found from the Tertiary of Eastern Iceland (Data from Wood, 1978). 52 (FE203 + FEB] 90 O BUSY. I. VEST. COHPLEXES \l \l \l \l \l \l \l \l [N920 + K20) [FE203 + FED) 0 (“000.1978) 90 [N920 + K20) M00 —-——-— H H, 7 7 , 7 Figure 9a and 9b 53 Figure lOa,b,c. REE Chondrite normalized plots for basic, intermediate, and acidic rocks found at the acidic and basic complexes. 54 ; D turcnncomrc kocn rvrcs I x CRRlezc Rocu rvpgs ‘ , 3 x 2 l x X = . 1 X x ! E , X g ' a .0 '9, g m " 3 C O ‘3 X g E 3 § g n B C I 3 g S E] E] a : 31 3 g i. 3b 39,: 1.1 e t , 5 Ln c5 Sn EU ra Y8 LU g , 1 Ln CE sn EU re Y8 LU 91 Y I Y Y I Y Y Fr 4 3! fir T Y T Y I V Y I' T Y T ‘1' 1' T 55 57 $3 59 so an 52 63 s: 55 as 57 56 59 70 i: 72 55 57 53 59 so SI 52 63 so 55 66 57 so 59 70 71 72 RTOnIC NUHBER RTOflIC NUHBER “O 4 1 o ansxc aocx "His 1 EL 6 2 2 3 E - 3 § 3 a E g E 2 , 5 5 5 1 ._, -_, 5 ,1; : If 0 E E, w E 9 3 3'9! ”1 . i LR ca sn EU ra va LU 0. —7 Y Y T V U T I Y 7 U I Y vi r V 56 S7 58 59 60 5} 52 63 54 65 66 S7 68 69 70 71 72 RTDHIC NUHBER ' Figure lOa,b,c 55 rocks are the strongest LREE enriched and have a negative europium anomaly. The intermediate rocks plot in between acidic and basic compositions with some overlap with both categories. The intermediate rocks range from having slight positive to slight negative europium anomalies. The basic rocks have generally flat europium profiles. All of the rocks have high (La/Lu)N ratios (which are comparable ratios to the (Ce/Yb)N ratios used by most workers) than are exhibited in most Icelandic volcanics (for example, see Wood, 1978) except for those rocks which are associated with volcanic centers of rhyolite to basalt variation (O'Nions and Gronvold, 1973). 56 EVALUATION OF MODELS OF ORIGIN Two simple models can be proposed to account for the observed field relationships, major element variation, and trace element variation found at the Austurhorn and Vesturhorn net-veined complexes. The first is commingling and is the failure of two miscible magmas to mix due do viscosity and temperature differences upon coming into contact with each other. Limited mixing may occur if temperature and viscosity differences are low enough such that extremely rapid crystallization of the basic phase does not result. The second model is silicate-liquid immiscibility, whereby, a homogeneous magma unmixes to form two thermodynamically stable liquids of a highly contrasting nature, one liquid essentially basic and the other acidic. Immiscibility Much interest has been renewed in immiscibility as a potential explanation of the variation of rock types in many environments. Experimental work and petrographic observations (Roedder, 1951; Ferguson and Currie, 1971, 1972; Massion and Koster Van Groos, 1973; Rutherford et al., 1974; McBirney and Nakamura, 1974; Irvine, 1975, 1976; Visser and Koster Van Groos, 1976, 1977, 1979; Watson, 1975, 1976; Naslund, 1976; Watson and Naslund, 1977; Roeder, 1978; Freestone, 1978; Visser, 1979; Ryerson and Hess, 1978, 1980) has provided evidence that immiscibility can potentially occur in natural silicate systems. Experimental investigations in immiscible systems indicate that the REES, phosphorous, and titanium are partitioned toward the basic 57 liquid (Watson, 1976; Ryerson and Hess, 1978, 1980; Visser and Roster van Groos, 1979). Watson reported a partitioning of a five fold increase in the basic liquid and Ryerson and Hess a ten fold increase with respect to the REES. The differences may be accounted for by the differing Si/O ratios (Eby, 1980), which could affect melt structure and thus partitioning, used in the experimental charges. Trace elements can thus be used to evaluate an immiscibility model. The REE'S in the Austurhorn and Vesturhorn net-veined complexes are enriched in the acidic rock in comparison to the basic rock types. Some overlap occurs between the REES in the basic rocks and the intermediate rocks. The REES in the intermediate rocks extensively overlap with the REEs in the acidic rocks, although the average REE'S in the acidic rock is higher than the average REE in the intermediate rocks (see Fig. 10,a,b,c). In a given sequence of rocks (e.g. a basic pillow interior, a diffuse zone surrounding the pillow and the acidic rock) all in close association, the acidic rock is more enriched in the REES than the intermediate rock and the basic rock. Thus, the REE abundances observed are not consistent with immiscibility, because the REES are partitioned in the acidic rocks. Differences in Si/O ratios, observed in some alkaline rocks interpreted as immiscible systems (Eby, 1980) cannot explain the observed REE distributions with regard to immiscibility. Phosphorous and titanium are depleted in the acidic rocks relative to the intermediate and basic compositions as might be expected in an immiscible system, but this same relationship could be generated by almost any fractionation process and thus is not a critical test of immiscibility. 58 The fact that basic pillows are chilled indicates a temperature difference between the magmas. This is inconsistent with immiscibility in place because, by definition, immiscibility occurring in a single homogeneous liquid is a thermal equilibrium process (i.e both liquids would be at the same temperature). Other field data, such as diffuse zones which surround many of the pillows, is also not supportive of immiscibility in place. Common phases (e.g. plagioclase) in immiscible systems should be of the same composition, which is not found at these net-veined complexes. In addition, more complex models involving immiscibility, such as the separation of the liquid phases at depth and then later coming into contact at a higher structural level (i.e. lower pressure) is inconsistent with the observed REE distributions as well as requiring an extremely complex "plumbing system". Magma Mixing The field relationships at the Austurhorn and Vesturhorn net-veined complexes strongly support commingling and partial mixing model between basaltic and granitic magmas. The field observations which are interpreted to support this model includes the presence of diffuse zones which surround many of the basic pillows and the highly mottled textures of the acidic rock due to an increase in mafic mineral content as the pillows are approached. Multiple Linear Regression Analysis and Trace Element Modeling Evaluation of mixing models or crystal fractionation models by multiple linear regression analysis (Wright and Doherty, 1970) has become commonplace in igneous petrology. Several errors (eg. choice 59 of parents, limited testing of alternative hypotheses) can be introduced by the investigator. In some cases the evaluation serves only to provide broad constraints on the data. Thus, the analysis only can lend support to a given model. In most cases, a simple multiple linear regression analysis of the major element data using a mixing of acidic and basic end members to produce intermediate rocks yields unsatisfactory residuals. The major influences in these regression analyses are SiOZ, A1203, total FeO, P205, K20, and T102. More complex models of magma mixing were thus considered. Two complex models of magma mixing are 1) that mixing proceeds in a stepwise fashion and 2) that magma mixing and crystal fractionation are occurring Simultaneously. In the first model magma A and B mix to form magma C. This product in turn mixes with either magma A or B again. Although this Situation should still produce a linear major element variation the process can become more complex or non-linear if one of both of the mixing magmas have crystals present which are fractionating (e.g. crystal settling or filter pressing) or if thermodynamic properties such as rates of diffusion become the rate limiting step (i.e. disequilibrium). If magma mixing is the dominant process which has occurred at the Austurhorn and Vesturhorn acidic and basic complexes then, by definition on the scale of the individual complexes, the rocks represent a case of disequilibrium because a Single homogeneous product of mixing is not present. Locally the rocks may represent a quasi-stable equilibrium. Multiple linear regression analysis of rocks from the most thoroughly sampled "pillow" outcrop, Ic-3 through Ic-9, indicate that a stepwise magma 60 Haw—m m zoom: >_NOw _w.wo true New Irma —¢.u¢ PC: “00 N.mo :.Vw Pow :.Nw :.wu Poe 3:0 For o.~o 90¢ OLw o.~o nob— Zmo 98 3.8 3.8 3.8 3.8 98 AUG _.—0 wk: 3&— V.ma yew no.om ZwNO fuw Puw Paw wk: Pam no.Nu .ANO w.um _.~—. o.o —.wu Two Pom QNOu o.ow 0.: o.o oba ob— o.ou 2:8“ N.mu* mw.c* 6.me .8 u 988 8,8 ... 988 64 ...mEm m ZOO—u. < >_~Ou 3.3 :73 3.8 3.3 Pee 3.3 3.3 9.3 Two Pmo v.3 who 2.3 «.3 2.3 2.3 -93 2:0 99. F; 92.: 93 Pom o.~o o.~c Poo 250 ohm Yum ~.~u 27% 3.3 ~73 Fan Po— an 2.3 ab» 8.3 v.3 3.2 «...: wbm 93 meO Fun F3. ...uu wk: 95 why PS 93 xNo Uoua No—o —0VU —OWU °o° —0N— —0NV icoca v~0u 98 93 92 obm ob 9.2 92 Poo .N u 988 2% u 98... 65 mixing model along with minor fractionation of pyroxene can produce the observed sequence of rocks (Table 6). In model I acidic rock (Ic-ZO) can be mixed with pillow" margin (Ic-6) type 3 pillow, to produce an intermediate (Ic-S) a hybrid of the same "pillow". See figure 2a,b for a photograph of the positions of the samples. In model II a hybrid from the same 'pillow' (Ic-4c) can be mixed with 'pillow' margin (Ic-6) to produce another hybrid (Ic-7). In model III hybrid (Ic-4c) and acidic-granophyric rock (Ic-20) can be mixed to produce hybrid (Tc-9). In model IVa acidic rock (Ic-ZO) is mixed with a Type 1 chilled "pillow" margin (Ic-3) to produce the Type 3 "Pillow" margin (Ic-6) (not shown). A slight variation on model IVa is to add pyroxene into the model as a fractionating phase from chilled "pillow" margin (Ic-3) which contains pyroxene phenocrysts (Model IVb). It is realized that the B coefficient generated for pyroxene in this equation is not significant and changes the residuals only slightly, but has been included since its involvement is geologically reasonable. Pyroxene may or may not be actually involved. Note its negative coefficient as would be expected (Table ) if it was removed by fractional crystallization. Thus in a stepwise mixing model these rocks can be accounted for whereas unsatisfactory results are obtained in endmember mixing between acidic rock (Ic-ZO) and a chilled "pillow" margin (Ic-3) to produce intermediate (Ic-6). Model V collectively models the stepwise model by mixing acidic rock (Ic-ZO) with hybrids (Ic-S, Ic-Ac) and 'pillow' margin (Ic-6) with minor pyroxene fractionation to produce Type 1 "pillow" margin (Ic-3). Within all of these equations the amount of acidic rock to mix in is similar. 66 Models which involve crystal fraction from the basic rock types to yield the intermediate rocks or the acidic rocks do not produce satisfactory residuals using the phenocrysts phases present in these rocks; nor phenocryst phase common to Icelandic lavas. The phenocryst phase compositions used are from Icelandic lavas as reported by Carmichael (1967). Many of the compositionally intermediate pillows contain anomalous P205 and TiOz values higher than would be expected from a simple mixing model. The Austurhorn net-veined complexes contains intermediate pillows which are high in both P205 and T102 (for example, Ic-28a,b) relative to the most basic pillows with chilled margins (Ic-12,Ic-3). The Vesturhorn net-veined complexes also contains intermediate pillows which have anomalously high P205 contents (Ic-74a,b,c for example, Ic-74b is a chilled margin). T102 in the Vesturhorn net-veined complexes is not as enriched in comparison to all of the chilled basic pillows. For instance, many of the intermediates are higher in T102 than chilled basic pillows Ic-53a,c or Ic-75, but lower than chilled basic pillows Ic-6l or Ic68c. A complex situation is presented in which all of these compositions (i.e. acidic , low P205-T102 intermediates, high P205 intermediates and basic rocks) are present in a liquid state at the same time. The variation in P203 and T102 cannot be explained by simple mixing, mixing and crystal fractionation, or silicate liquid immiscibility. A more complex, at present unknown, model will have to account for this variation. Mixing and Trace Elements Trace element mixing models based upon the B coefficients 67 reported in Table 6 produce only marginaly satisfactory results and this may be due to disequilibrium. For example, for model V in Table 6 the following are the calculated REE abundances from the model and the actual measured REE contents in Ic-3 respectively: La (18-29), Ce (51-59), Sm (8.7-9.6), Eu (2.82-2.80), Tb (1.41-1.58), and Lu (0.78-0.48). However, the trace element trends when viewed collectively for all rock types are most consistent with mixing. Plots of la vs. Sm or Ce vs Sm (Fig. 11) in general yield linear plots as would be expected from endmember mixing. A stronger test of mixing is a plot of a LREE/HREE vs another incompatible ratio (Langmuir, 1978). Theoretically, rocks which have mixed between two end members produce a linear trend on a ratio-ratio plot of incompatible elements when both ratios have a common denominator (Langmuir, 1978). Figure 12 is a plot of (Ce/Lu)N vs. (La/Lu)N and in general a linear trend is produced although a small degree of scatter is present. The scatter of the plotted points is more apparent from the Vesturhorn than from the Austurhorn mixed rocks. These types of plots are particularly good in differentiating between crystal fractionation and magma mixing since crystal fractionation produces little variation on such a plot. The observed variation of the mixed rocks is greater by a factor of two when compared to the maximum theoretical variation produced by crystal fraction. Some of the chilled margins exhibit a significant departure from the linear trend which may indicate localized conditions of disequilibrium. This conclusion is supported by the fact that Ic-53a and Ic-53c, two samples from the same pillow, both show departures from a linear trend. Other pillows exhibit internal consistency; 68 Figure 11. Plot La vs. SM which in general generates a linear trend. 69 Figure 11 oo.mm 2m oo.m~ oo.n~. oo.m~ oo.ma . oo.m~ oouo co“. oo.pu flu w 9%9 o e a a 9% % .m e as m 8% mm a a .m a w rt 8 zmozmimfi e m 7:55.325 3 m .m 0 .m r2 mw mu 0 £6 t 0 .m 6.1 70 Figure 12. Plot of (Ce/Lu)N vs. (La/Lu)N 71 s m m a. m N a o 2 .............. h pppppp 2 ....... 2 »»»»»»» 2 ....... L ...... 2 ..... 2 »»»»»»» 0 2 zmozmimg a ”1.. 555552. e m 9 H13] 2 3 n 3 O / e 0 ee 131: as H H manages? 946. H [ 6 n e as a -vN Q .. Q . vs 72 for example, Ic-12 (interior) and Ic-13 (margin). Thus, disequilibrium affects may be apparent in some of the samples, but the trend produced on this plot is most consistent with magma mixing. Complex Models Other trace elements are not consistent with crystal fractionation or immiscibility. Log-Log plots of Sr vs. Ba, Sr vs. La, Sr vs. Ce, or V vs. Ba have similar trends. These plots have an initial positive SIOpe for basic rocks and then a sharp negative slope for acidic rocks. Rocks which are intermediate compositionally plot at the change in slope (see Fig. 14a). This variation observed with respect to Sr is similar to the lava variation observed all over Iceland (Gerasimovsky et al., 1976). Theoretical plots of Sr vs. La indicate that the observed trends in these intrusions cannot be produced by crystal fractionation alone. With the fractionation of olivine, orthopyroxene, augite, clinOpyroxene, or a plagioclase having a high or low distribution coefficient cannot produce the observed trends (Fig. lha,b) by simple Rayleigh crystal fractionation. The observed trends indeed cannot be produced by any combination of these minerals fractionating. Crystal fractionation of minor mineral phases, such as apatite or an opaque phase, may be able to strongly affect the trace element trends. The distribution coefficients for these phases with regard to V and Sr are not known. However, using reasonable estimates of fractionating mineral proportions of five percent apatite and ten percent of an opaque phase, and an initial concentration in a basic or intermediate parent parent of 300 pm Sr to produce an acidic rock 73 with 80 pppm Sr, the calculated distribution coefficients using a Rayleigh law model are approximately Kd-6O (apatite) and KD = 60 (opaque phase). The distribution coefficients needed to produce the Sr values observed are unreasonably high. Estimates of Sr distribution coefficients can be made by comparisons of the distribution coefficients for Eu, which should behave similarly. A maximum value, KB of 50 (apatite) for Eu (Nagasawa, 1970), for a rhyolithic composition would estimate the maximum value expected for Sr since lower Si02 magmas would likely have lower KD values. Reported values for Sr in magnetites from andesite compositions range from KDs of 0.23 to 0.42 (Luhr and Carmichael, 1980). Additional consideration of plagioclase as a fractionating phase would only slightly change these estimated coefficients and have only a minor effect upon the theoretical distribution of these elements in comparison to the trends observed. In addition, apatite and opaque phases are not observed as phenocryst phases. Thus crystal fractionation alone cannot produce the observed variation in these rocks. Simple magma mixing between acidic and basic compositions would not explain the observed trends of Sr vs. La because a linear trend would be expected between endmembers. A model involving magma mixing and crystal fractionation is also difficult to reconcile with observed trends because of the continuous but non-linear variation between Sr vs La. Fractionation of apatite or plagioclase produces a similar trend as mixing between end members and cannot explain the horizontal trend between basic and intermediate compositions. It should be remembered that both basic and intermediate pillows have 74 Figure 13a. Log-Log plot of Sr vs. Ba. The solid dark line represents the trend found in Figure 13b. Figure 13b. Log-Log plot of Sr vs. Ba. For the Tertiary lavas found in Eastern Iceland (Data from Wood, 1978). 75 [I] RUSTURHORN O VESTURHORN 1.1.1.03 [llU SR 1 lllLll 1 U 1 I lllIllll l IIIIIWI 0 1 2 10 10 10 89 T IIIITHI 103 a ”Q ‘ 2.3 m TERTIRRY LRVRS (WOOD.1978) \ j l I z . an N ‘ HEM mm Q CE] 2.3 g c: 3 m 03—4 2 C) H I Illltrrl’ I lllllllli I FIlHll] U 1 2 3 10 10 10 10 BR _—— Figure 13a and 13b 76 apparently mixed with the acidic rock, yet the differences occur between these compositions (i.e., there is not a linear trend from basic to acidic compositions). The observed trends involving Sr and V remain problemmatical. £205 and T192 Field observations strongly support magma mixing as being the dominant process which has occurred in these net-veined complexes. Hypotheses which could account for the variation of rock types, especially the high P205 and TiOz intermediate compositions and the behavior of Sr and V observed, concurrent with mixing, include 1) that more than two end members are involved in the mixing process, 2) a zoned basic magma chamber is mixing with the acidic magma, similar to the zoned magma chambers envisioned by Hildreth (1979) and/or Rice (1981), and 3) that the compositions present represent disequilibrium products, such that, some elements such as phosphorous and titanium have diffused preferentially under a Soret effect or quasi-stable immiscibility. These processes would have had to occurred at depth and not at the exposed level of outcrop. If more than two unrelated magmas are involved as endmembers then it is surprising that simple mixing or complex mixing between three endmembers cannot produce the variation observed. Multiple linear regression models testing this hypothesis also yield unsatisfactory residuals. Various intermediate samples were chosen to be used in multiple linear regression models involving a third magma type. Involvement of a third magma type from a separate and unrelated magma chamber would not seem to be justified, although the d. data does not clearly discriminate between the possibilities suggeste 77 It is of interest that the Austurhorn and Vesturhorn net-veined complexes exhibit such a similar range of compositions, especially with respect to the high P205 and T102 intermediate compositions. Other examples, of a similar nature to the range of compositions present at these net-veined complexes occur elsewhere in Iceland. The 1875 Askja eruption has a similar range of compositions and has been interpreted as representing the intimate coexistence of acidic, intermediate, and basic magmas (Sigurdsson and Sparks, 1981). High P205 and T102 intermediate lavas are also present. These occurrences over widely separated areas suggest more than a casual relationship. Neither the data presented in this report, nor the data presented elsewhere, allow one to distinguish between or reject mixing models involving zoned magma chambers, more than two separate magmas, or the preferential diffusion of ions (Soret effect) such as P or Ti. Extent of Mixing The field evidence at both of these net-veined complexes indicates that magma mixing, in situ, was widespread, but spatially limited to the localized areas surrounding many of the pillows. Extensive mixing between basaltic and granitic magmas is unlikely unless the acidic magma is quite superheated. If the volume of the acidic magma is significantly larger than the basic magma, then homogeneous mixing even with superheating is probably prohibitive due to viscosity and temperature differences. It is likely that small amounts of granitic magma may be mixed with a large amount of basaltic magma, especially if the granitic magma is superheated. 78 Superheating of the Acidic Rocks The acidic rocks found in the net-veined complexes were likely somewhat superheated. Evidence for this includes: 1) The lack of chilled boundaries on many of the intermediate rocks (Type 3 pillows); and 2) the lack of chilled boundaries on some edges of the more basic pillows (Type 2 pillows). Superheating of acidic compositions, if common in Iceland, may have important ramifications in producing the variation of rock types observed as well as the eruption of acidic compostions. Magma mixing as a triggering mechanism for eruption of acidic magmas (Sparks et a1. 1977) may be heavily dependent upon the volumes present and degree of superheating of the acidic magma. Evidence from the 1875 Askja eruption indicates that the acidic magma was superheated. Temperatures of 990°C to lOlO°C were reported (Sigurdsson and Sparks, 1981) on the basis of plagioclase geothermometry. Origin of the Basic Rock Types The most primitive basic pillow compositions closely resemble the lava compositions of ferrobasalt and basaltic andesite found in the Tertiary lavas of Eastern Iceland (Wood, 1978). These compositions are not generally thought to represent primary mantle melts, but more evolved melts which have undergone fractionation. Thus, the basic rocks found in the acidic and basic complexes probably represent evolved compositions and not primary mantle melts. Origin of the Acidic Rocks The abundances of rock types in Iceland have a bimodal distribution with approximately 85 percent basalt and a minimum of 79 10-12 percent rhyolite (Walker, 1966; Sigurdsson, 1977). This bimodality is apparently unique and quite different from normal oceanic crust. The acidic rocks found at the Austurhorn and Vesturhorn intrusions are similar to the acidic compositions of lavas found in Iceland. Models which have been proposed to account for the origin of the acidic rocks include: 1) simple crystal fractionation of primary basic compositions (Wood, 1978); 2) the remobilization of deep-seated acidic rocks towards the surface by a complex history of refusion, fractional crystallization, and hybridization (Sigurdsson and Sparks, 1981) and similarly the melting of plagiogranites (Sigurdsson, 1977); 3) the partial melting of amphibolitized basalt or gabbro at the base of the oceanic crust. The partial melting is due to the sinking of a hydrated basaltic crust which is subjected to an increase in the geothermal gradient as the crust moves away from Icelandic ridge system (Oskarsson et al., 1979). Experimental investigations on the fusion of the Picture Gorge Thoeliite at Skbar (H20) indicate that acidic compositions (similar to many of those found in Iceland) can be produced (Heltz, 1976); and 4) the partial melting of oceanic layer 3 to produce acidic rocks (small degrees of melting) to intermediate (more extensive melting) compositions (O'Nions and Gronvold, 1973) Of these models, simple crystal fractionation from primary basic compositions can be eliminated as the only process involved in the origin of acidic rocks in Iceland. Simple crystal fraction cannot explain the wide occurence of Na>K in the acidic rocks (especially those acidic rocks associated with tholeiitic basalts), nor the high 80 REE contents of many of the acidic rocks. The high (Ce/Yb)N or (La/Lu)N ratios found in most Icelandic acidic rocks can not be produced by simple crystal fractionation even when using a wide range of distribution coefficients and either a simple Rayleigh crystal fractionation or a more complex model of batch crystal fractionation (McCarthy and Hasty, 1976). Mineral pr0portions and primary liquids used in these calculations are from the multiple linear regression analysis of Wood (1978). In addition, the trace element trends of Sr vs Ba or Sr vs La, for example, can not be explained by simple crystal fractionation as previously discussed. The more complex models which have been proposed have not been, as yet, rigorously tested. The following observations constrain hypotheses for the petrogeneiss of acidic rocks in Iceland: (1) The greater abundances of acidic rocks in Iceland (Walker, 1966; Sigurdsson, 1977) than can be explained by simple crystal fraction (Sigurdsson and Sparks, 1981). (2) The close association of acidic rocks with central volcanic complexes (Walker, 1966; O'Nion and Gronvold, 1973). (3) The LREE enrichment, negative Eu anomalies, and higher total REE abundances which is reflected by the high (LREE/HREE)N ratios found in acidic compositions (O'Nions and Gronvold, 1973; Oskarsson et al., 1979; and this study). (4) The association of alkaline rhyolites with alkali basalts and peraluminous rhyolites with tholeiitic basalts (Sigurdsson, 1971)(5) The high degree of variability in the 87Sr/86Sr ratios 0f 81 acidic rocks from different localities (O'Nions and Gronvold, 1973). (6) The often close association in time of rhyolitic and basic lavas in no predictable order (O'Nion and Gronvold, 1973) and their existence in the same environment often comtemporaneously (for example, Sigvaldason, 1979; Jorgensen, 1980; and Sigurdasson and Sparks, 1981; This study). Factors which may play an important role in the origin of acidic rocks in Iceland include: The greater thickness of Icelandic crust than typical oceanic crust, the degree and depth of hydration of the oceanic crust, the regional or local geothermal gradient in association with volcanic centers, and the rates of plate movement. SUMMARY The rocks at both the Austurhorn and Vesturhorn net-veined complexes are most consistent with commingling magma mixing between acidic and basic magmas with regard to major element, trace element and field evidence. Silicate liquid immiscibility is not consistent with the data. Complex models of mixing are necessary to explain the observed features at these net-veined complexes. A complex history of intrusion and interaction between coexisting acidic, intermediate, and basic magmas is indicated by the field data as well as the chemical variations. Of particular interest is the presence of high P205 and T102 intermediate compositions which cannot be related by fractional crystallization or simple mixing. Trace element variation of V and Sr are also problemmatical. Models which could account for the chemical variation observed included: 1) the involvement of more than two compositionally different magmas in the same magmatic chamber, 2) the 82 involvement of a zoned basic magma with an acidic magma or, 3) a preferential partitioning of elements, such as P and Ti under quasi-stable equilibrium conditions (Soret Effect), which is likely compositionally and temperature dependent. Acidic and basic magmas are not likely to mix extensively, forming a homogeneous product, unless the volume of the basic magma is much larger than the acidic magma and unless the acidic magma is superheated. The acidic magmas at the Austurhorn and Vesturhorn net-veined complexes are interpreted as having been superheated. Extensive mixing between acidic and basic compositions did not occur indicating that mixing between small volumes of basic magma and larger volumes of acidic magma is likely prohibitive due to temperature and viscosity differences. Mixing between large volumes of basic magma and small volumes of acidic magma may be possible. The major element compositions and trace element trends in the basic pillows are most consistent with their origin resulting from a fraction and/or contamination process. The basic pillow types probably do not represent a primitive melt from the mantle. The acidic rocks in these intrusions are similar to acidic lavas found all over Iceland of tholeiitic affinities. The acidic rocks are not the result of simple crystal fractionation from a primitive basic parent based upon the trace element variations. The origin of acidic rocks in Iceland could be the result of many processes or interacting processes which have not, as yet, been rigorously tested. REFERENCES REFERENCES Abbey, S, 1978, IV Studies in Standard Samples: for use in the General Analysis of Silicate Rocks and Minerals Part 5: X-ray Spectrometry, 7, 99-122. Blake, D. H., Elwell, R. W. D., Gibson, I. L., Skelhorn, R. R., and Walker, G. P. L., 1965, Some Relationships Resulting from the Intimate Association of Acid and Basic Magmas: Quart. Jour. Geol. Soc. Lond., V.121, 31-49. Blake, D. H., 1966, The Net-veined Complex of the Austurhorn Intrusion, Southeastern Iceland: Jour. Geol., v.74, 891-907. Bunsen, R. W., 1851, Uber die Processe der Vulkanischen Gesteinsbildungen Islands: Annalen der Physik und Chemie, 83, Carmichael, I. S. E., 1967, The Mineralogy of Thingmuli, A Tertiary Volcano in Eastern Iceland: Am. Minera1., 52, 1815-1840. Chapman, C. A., 1962, Diabase-granite composite dikes, with pillow-like structure Mount Desert Island, Maine: Jour. Geol., 70, 539-564. Dahlquist, R. L. and Knoll, J. W., 1978, Inductively Coupled Plasma -Atomic Emission Spectrometry: Analysis of Biological Materials and Soils for Major, Trace, and Ultra-trace Elements: Applied SpectroscOpy, 32, 1-29. Eby, G. N., 1980, Minor and Trace Element Partitioning Between Immiscible Ocelli-Matrix Pairs from LamprOphyre Dikes and Sills, Monteregian Hills Petrographic Province, Quebec, Contrib. Mineral. Petro., 75, 269-278. Ferguson, J. and Currie, K. L., 1971, Liquid Immiscibility in Alkaline Ultrabasic Dikes at Callender Bay, Ontario: J. Petrology, 12, 561-585. Ferguson, J. and Currie, K. L., 1972, Silicate Immiscibility in the Ancient "basalts" of the Barberton Mountain Land, Transvaal: Nature (Phys. Sci), 235, 86-89. Freestone, I.C., 1978, Liquid Immiscibility in Alkali-Rich Magmas: Chem. Geol. 23, 115-123. Gamble, J. A., 1979, Some Relations Between Coexisting Granitic and Basaltic Magmas and the Genesis of Hybrid Rocks in th Tertiary Central Complex of Slieve Gullion, Northeast Ireland: Jour. Vol. Geotherm. Res., V.5, 297-316. 83 84 Gerasimovskiy, V. 1., Laktionova, N.V., Kovalenker, V. G., 1976, Strontium and Barium in Icelandic Lavas, Geochim. Inter., 13, 143-149. Gordon, G. E., Randle, K., Goles, G. G., Corliss, J. B., Beeson, M. H., and Oxley, S. S., 1968, Instrumental activation analysis of standard rocks with high-resolution gamma-ray defectors: Geochim. Cosmochim. Acta, V.32, 369-396. Gunn, B. M., and Watkins, N.D., 1969, The Petrochemical effect of simultaneous cooling of adjoining basaltic and rhyolitic magmas: Geochim. Cosmochim. Acta, V.33, 341-356. Helz, R. T., 1976, Phase Relations of Basalts in Their Melting Ranges at P320 = Skb. Part II. Melt Compositions: Jour. Petrology, v. 17, Part 2, 139-193. Hildreth, W., 1979, The Bishop Tuff: Evidence for the Origin of Compositional Zonation in Silicic Magma Chambers: in Ash Flow Tuffs, G.S.A. Spec. Paper 180, Chapin, C. E. and Wolfgang, E. E. (eds.), 43-72. Holgate, N., 1978, A composite Tholeiite dyke at Impachar, Isle of Arran: its petrogenesis and associated pyrometamorphism: Mineral. Mag. 42, 141-142. Holmes, A., 1931, The problem of the association of acid and basic rocks in central complexes: Geol. Mag., v.68, p. 241-255. Irvine, T. N., 1975, The silica immiscibility effect in magmas: Carnegie inst. Wash. yearbook 74, 484-492. Irvine, T. N., 1976, Metastable liquid immiscibility and MgO,FeO, Si02 fractionation patterns in the system Mgz 8104 = Fez 3104 - Ca A12 81208 - K A1 Si30©x8 - 8102: Carnegie Inst. Wash. yearbook 75, 597-611. Jorgensen, K. A., 1980, The Thorsmork Ignimbrite: an unusual comenditic pyroclastic flow in Southern Iceland: Jour. Volc. Geotherm. Res., 8, 7-22. King, B. C., 1962, The nature of basic igneous rocks and their relations with associated acid rocks: Part I: Sci. progr., V.L, 616-629. King, B. C., 1963a, The nature of basic igneous rocks and their relations with associated acid rocks: part II; Sci, Progr., V.LI, 259-2750 King, B. G., 1963b, The nature of basic igneous rocks and their relations with associated acid rocks: Part III: Sci. Progr. V.LI, 610-628. 85 King, B. G., 1964, The nature of basic igneous rocks and their relations with associated acid rocks: part IV: Sci. progr., v. 52, 282-292. King, B. C., 1965a, The nature of basic igneous rocks and their relations with associated acid rocks; Part V; Sci. progr., v.53, 437-446. King, B. G., 1965b, The nature of basic igneous rocks and their relations with associated acid rocks: Part VI: Sci. Progr., V. 53, 117-125. Korotev, R. L., 1976, Geochemical Modeling of the Distribution of Rare-earth and other elements in a Basalt and Grain Size fraction of soils from the Apollo 17 valley floor and a well tested procedure for accurate instrumental neutron activation analysis of geological materials. Ph.D. Thesis, Univ. of Wisconsin, Madison, 264.p. Langmuir, C. H., Vocke, R. D., Jr., Hanson,G. M., and Hart, S. R., 1978. A General Mixing Equation with Applications to the Icelandic Basalts: Earth Planet Sci. Lett., V.37, 380-392. Luhr, J. F. and Carmichael, I. S. E., 1980, The Colima Volcanic Complex, Mexico I. Post-Caldera Andesites from Volcano Colima: Contrib. Mineral. Petrol., v.71, 343-372. Mission, P. J., and Koster Van Gross, A. F., 1973, Liquid immiscibility in silicates: Nature (Phys. Sci.), V.245, 60—63 0 McBirney, A.R., and Nakamura, Y., 1974, Immiscibility in late-stage magmas of the Skaergaard intrusion: Carnegie Inst. Wash. Yearbook 73, 348-352. McCarthy, T. S., and Hasty, R. A., 1976, Trace element distribution patterns and their relationship to crystallization of granitic melts: Geochim. Cosmochim. Acta, 40, 1351-1358. McSween, H. Y. Jr., Coish, R. A., Norman,M. D., 1979, Coexisting acidic and basic melts: Geochemistry of a Composite Dike - A Discussion: Jour. Geol., 87, 209-214. Nagaswa, H., and Schnetzler, C. C., 1971, Partitioning of rare earth, alkali, and alkaline earth elements between phenocrysts and acidic igneous magma: Geochim. Acta, V. 35, 953-968. Naslund, H. R., 1976, Liquid immiscibility in the system KAl 81308 - Na Al 81308 - FeO - Fe203 - $10 and its application to natural systems: Carnegie Inst. Wash. Yearbook 75, 592-597. 86 Oskarsson, N., Sigvaldason, G. E., Steinthorsson, S., 1979, A Dynamic Model of Rift Zone Petrogenesis and the Regional Petrology of Iceland: Nordic Vol. Inst. 79 05, p. 104. O'Nions, R. K., and Gronvold, K., 1973, Petrogenetic relationships of Acid and Basic Rocks in Iceland: Sr-Isotopes and Rare-earth Elements in Late and Postglacial Volcanics, Earth Planet. Sci. Lett., V.19, 397-409. Pallister, J.S., and H0pson, C. A., 1981, Samail Ophiolite plutonic suite: Field relations, phase variations cryptic variation and layering, and a model of a spreading ridge magma chamber: Jour.Ge0phys. Res., 86, 2593-2644. Pallister, J.S., 1981, Structure of the sheeted dike complex of the Samail Ophiolite near Ibra, Oman: Jour. GeOphys. Res., 86, 2610-2672. Philpotts, A.R., 1971, Immiscibility between feldspathic and gabbroic magmas: Nature (Physical Science), V.229, 107-109. Philpotts, A. R., 1972, Density surface tension and viscosity of the immiscible phase in a basic, alkaline magma: Lithos, V.5, 1-18. Philpotts,A. R., 1979,Silicate immiscibility in tholeiitic basalts: Jour. Petrology, V.20, 99-0118. Prestvik, T., 1980, Petrology of hybrid intermediate and Silicic rocks from Oraefajokull, Southeast Iceland: Geologisica Foreningens i Sockholm Forhandlingar, 101, 299-307. Rice, A., 1981, Corrective fractionation: A mechanism to provide cryptic zoning (Macrosegregation, layering, crescumulates, bande tuffs and explosive volcanism in igneous processes: Jour. GeOphys. Res. 86, 405-417. Roedder, E., 1951, Low temperature liquid immiscibility in the system KZO-FeO-A1203-SiOZ: Amer. Minera1., V.36, 282-286. Roedder, E., 1978, Silicate immiscibility in magmas and in the system KZO-FeO - A1203 - $102: an example of serendipity: Geochim. Cosmochim. Acta., V.42, 1597-1617. Roobol, M. J., 1971, Some relations between common acid-basic associations: Geol. Mag., V.108, 525-531. Roobol, M. J., 1974, The geology of the Vesturhorn intrusion, SE Rutherford, M. J., Hess,P.C., and Garland, D. H., 1974, Experimental liquid line of descent and liquid immiscibility for basalt 70017: Proceedings of the fifth Lunar Science conf., Suppl. 5, Geochim. Cosmochim., Acta, v.1, 569-583. 87 Ryerson, F. J. and Hess, P. C., 1978, Implications of liquid-liquid distribution coefficients to mineral-liquid partitioning: Geochim. Cosmochim. Acta, 42, 921-932. Ryerson, F. J., and Hess, P. C., 1980, The role of P203 in silicate melts: Geochim. Cosmochim. Acta, 44, 611-624. Sigurdsson, H., 1971, Feldspar relations in a composite magma. Lithos, V.4, 231-238. Sigurdsson, H., 1977, Generation of Icelandic rhyolites by melting of plagiogranites in the oceanic layer: Nature, 269, 25-28. Sigurdsson, H., and Sparks, R. S. J., 1981, Petrology of rhyolitic and mixed magma ejecta from the 1975 eruption of Askja, Iceland: Jour. Petro, 22, 484. Sigvaldason, G. E., 1979, Rifting, magmatic activity and interaction between acid an basic liquids: The 1875 Askja eruption in Iceland, Nordic Vol. Inst. 79 03, p. 54. Sparks, S. R. J., Sigurdsson, H., and Wilson, L., 1977, Magma mixing: A mechanism for triggering acid explosive eruptions: Nature, V.267, 315-318. Taylor, T. R., Vogel, T. A., and Wilband, J. T., 1980, The composite dikes at Mount Desert Island, Maine: an example of coexisting acidic and basic magmas: Jour. Geol., 88, 433-444. Visser, W., 1979, Experimental investigation of silicate liquid immiscibility int he ssytem K20 - FeO - A1203 - SiOz - T102 - P205: Unpub. Ph.D. Thesis, Uniersity of Illinois at Chicago Circle, p. 106. Visser, W., and Koster Van Groos, A. F., 1976, Liquid immiscibility in K20 - FeO - A1203 - 102: Nature, V.264, 426-427. Visser, W., and Koster Van Groos,A.F., 1977, Liquid immiscibility in K20 - FeO - A1203 - $102, a reply: Nature, V.267, p.560. Visser, W., and Koster Van Groos, A. F., 1979, Effects of P205 and TiOz on liquid-liquid equilibria in the system K20 - FeO - A1203 - 3102: Am. Jour. Sci., 279, 970-988. Wager, L.R., and Bailey, E. E. 1953, Basic magma chilled against acid magma: Nature, V.172, 68-69. Walker, G. P. L., 1962, Tertiary welded tuffs in eastern Iceland: Quart. Jour.Geol.Sco. Lond., 119, 29-63. Walker, G. P. L., 1966, Acid volcanic rocks in Iceland: Bull. Volcanol. 29, 375-402. 88 Walker, G. P. L., and Skelhorn, R. R., 1966, Some associations of acid and basic igneous rocks, Earth Sci. Rev., V.2, 93-109. Watson, E. B., 1975, Partitioning of trace and minor elements between coexisting silicate liquids: Carnegie Inst. Wash. Yearbook 74, 500-504. Watson, E. B., 1976, Two liquid partition coefficients: experimental data and geochemical implications: Contrib. Mineral. Petrol. V.56, 119-134. Watson, E.B., and Naslund, h. R., 1977, The effect of pressure on liquid immiscibility in the system K20 - FeO - A1203 - Si02 - C02: Carnegie Inst. Wash. Yearbook 76, 410-414. Weis, L. A., 1974, A petrochemical investigation of the Winona quadrangle basalts, northern Michigan: M.S. Thesis, Univ. of Toledo, 75 p. Wiebe, R. A., 1973, Relations between coexisting basaltic and granitic magmas in a composite dike: Am. Jour. Sci., V.273, 130-151. Wiebe, R. A., 1974, Coexisting intermediate and basic magmas, Ingonish, Cape Breton Island: Jour Geol., v.82, 74-87. Wiebe, R. A., 1979, Fractionation and liquid immiscibility in an anorthositic pluton of the Nain Complex, Labrabor.: J. Petrology, V.20, 239-269. Wiebe, R. A., 1980, Commingling of contrasted magmas in the plutonic environment: Examples from the Nain anorthosite complex: Jour. Geol., 88, 197-209. Wilcox, R. E., 1944, Rhyolite-basalt complex on Gardiner River, Yellowstone Park, Wyoming: Geol. Soc. America Bull., v.55, 1047-1081. Wright, T. L., and Doherty, p. C., 1970, A linear-programming and least squares computer method for solving petrologic mixing problems: Geol. Soc. America, Bull., V.81, 1995-2008. Wood, D.A., 1978, Major and trace element variations in the tertiary lavas of eastern Iceland and their significance with respect to the Iceland geochemical anomaly: Jour. Petrology, V.19, 393-436. Yoder, H. 8., Jr., 1973, Contemporaneous basaltic and rhyolitic magmas: Am. Mineral, v.58, 152-171. MICHIGAN STATE UNIV. LIBRnRIEs VIlWWI"!"HINWIHW“WWI 31293106695244