Thesis (Ph. D.)--University of Wisconsin--Madison, 1949. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 78-81).

The iron apatite mineralizations of northern Sweden are mainly situated in supracrustal rocks and are thought to have formed during approximately the same time as the major plutonic suites in northern Sweden. In this thesis the Malmberget iron apatite ore deposit have been compared to the Kiruna iron apatite ore deposit to see whether the role of hydrothermal processes are different between the two ore districts. Also, since it has proven problematic to distinguish between the different mafic and ultramafic members of the major plutonic suites in northern Sweden (i.e. the Haparanda suite (1.94-1.85 Ga), the Perthite-monzonite suite (PMS) (1.87 Ga) and the Edefors suite (1.80-1.79 Ga)), a second goal is to trace the distribution and distinguish between these suites. Major elements, REE and trace elements have been studied along with polished sections from the ores and oxygen isotopes of magnetites from the ore districts to trace the ore formation process and separate between suites. All ore districts display a close similarity in geochemistry between host rocks and magnetites, especially regarding the REE pattern. Oxygen isotopes show that samples from the Kiruna district plot exclusively at positive δ18O-values with the majority of them at values < 1‰. Samples from the Malmberget district display δ18O-values either close to the igenous range or at slightly negative values. However, samples that display very low or negative δ18O-values show signs of either high temperature hydrothermal alteration, oxidation or crustal contamination. The δ18O composition along with textural observations from the polished sections and geochemistry, points towards a magmatic origin for these deposits. No apparent evidence has been found to support a primary ore formation by hydrothermal process in any of the districts studied. Regarding the suites, normalized spider plot patterns of REEs and trace elements served as a basis for subdivision into groups by similarities in key identification parameters. Several patterns are characterized by a pronounced Eu-trough while the middle- to HREEs display a rather flat trend, criteria that are related to rocks from the Haparanda suite, while samples with positive Eu-anomaly (Eu/Eu* > 1) and a nearly flat trend from La to Pr with a marked positive peak at Sr and generally less enriched in LILE, are consistent with the Edefors suite. Also, the Edefors suite generally display low concentrations of both Th and U but are somewhat enriched in Sr with relation to Pr. No distinct geochemical features could help to separate between the Haparanda and the PMS suites.