Origin of Magnetite in the Zoned Ultramafic Complexes of Southeastern Alaska1
Abstract
Appreciable concentrations of titaniferous magnetite occur in several of the ultramafic complexes that are exposed along a linear belt 350 miles in length in southeastern Alaska. Four of the larger bodies, at Klukwan, Snettisham, Union Bay, and Duke Island, are of economic interest. Large magnetite concentrations (15–20 percent of the rock by volume) of very uniform grade occur in the hornblende pyroxenite units of these complexes.
In several of the complexes (including Union Bay and Duke Island) a consistent zoning of ultramafic rock types occurs, with dunite, peridotite, olivine pyroxenite, and hornblende pyroxenite occurring in sequence from the center outward to the margin. Only the hornblende pyroxenite zone is exposed at Klukwan and Snettisham, but the mineralogy of this zone is identical in all the complexes. The minerals are diopsidic augite (Di85Hed15 to Di70Hed30), hornblende (abnormally high in A12O3 and low in SiO2 relative to most igneous hornblendes), magnetite, and accessory ilmenite and hercynitic spinel. Accessory amounts of olivine may occur in the interior parts of the magnetite pyroxenite zones of the complexes that contain olivine pyroxenite and/or dunite, but the magnetite concentrations drop abruptly as the amount of olivine increases; magnetite is largely absent from the olivine pyroxenite zones.
The many common features of these ultramafic bodies indicate that they formed by a single mechanism, namely by crystallization of ultramafic magmas. Evidence for this includes: (1) graded layering in rhythmically-bedded olivine pyroxenite and magnetite pyroxenite; (2) significant contact metasomatism and metamorphism at the margins of the bodies; and (3) good correspondence between observed rock types and experimentally studied liquids in the system diopside-olivine-iron oxide. It is suggested that the complexes formed by a combination of fractional crystallization and multiple intrusion of ultramafic magmas, and that the magnetite-hornblende pyroxenite zones in particular crystallized from a liquid roughly corresponding to their present chemical composition. This liquid must have had a total Fe content of about 15–18 percent by weight.
Figures & Tables
Contents
Magmatic Ore Deposits

This monograph on Magmatic Ore Deposits has resulted from a Symposium held at Stanford University on November 12 and 13, 1966. All except three of the papers that were presented are published in this volume as well as some of the discussion and the summation of the symposium. Unfortunately much of the discussion cannot be included because the volume is already so large. The best introduction to this volume is, perhaps, the introduction as it was presented at the symposium:
This symposium was conceived in 1962 when the Program Policy Committee recommended that the . Society of Economic Geologists should sponsor a symposium on magmatic ore deposits. The Committee under the chairmanship of John K. Gustafson believed this to be an effective method of advancing geologic thought. It is fitting that the symposium should finally be held during Gustafson’s presidential year. The proposal of the Program Policy Committee was approved by Council at its meeting in November, 1962. A special committee consisting of G. Kullerud, J. A. Noble, C. H. Smith, T. P. Thayer, with H. D. B. Wilson as chairman, was appointed by the President, Olaf N. Rove, in February 1963 to make arrangements for the symposium. E. N. Cameron, Secretary of the Society, was ex officio member of the special committee and remained as an active member when he resigned the secretaryship. C. H. Park, Jr. joined the committee shortly after its formation.
The Program Policy Committee was prompted to recommend the symposium by the realization that the underlying theory of the formation of magmatic ore deposits was formulated many decades ago., In the intervening years, much new data have been acquired from systematic research. It seemed to the Program Policy Committee that it was time for those with an abiding interest in the magmatic deposits to meet to assess this new data and to point out the unresolved problems.
The symposium was entitled “Symposium on Magmatic Ore Deposits.” The special committee accepted the terminology in the “Glossary of Geology and Related Sciences,” Edition 2, page 175.
Magmatic Deposits Certain kinds of mineral deposits form integral parts of igneous rock masses and permit the inference that they have originated, in their present form, by processes of differentiation and cooling in molten magmas. (Lindgren p. 863, 1929).
The symposium committee has added the term “ore” to attempt to keep the discussions centered on ore, or near ore material, or with