Abstract

One of the most persistent questions regarding the phase equilibria of mid‐ocean ridge basalts (MORB) pertains to the petrogenesis of the anorthitic plagioclase phenocrysts (>An90) that are characteristic of the more primitive members of such suites. Anorthitic phenocrysts are present in many if not most MORB suites in spite of the fact that no naturally occurring MORB glasses have ever been discovered to be in equilibrium with plagioclase more calcic than An85. We have addressed this paradox by attempting to saturate natural basalts with anorthite in a series of 1 atm experiments using three different natural basaltic starting compositions: an N‐MORB, an E‐MORB, and a continental high‐alumina basalt. To ensure duplication of the olivine and anorthite saturation observed in natural anorthite‐bearing basalt, the experiments were run in An93‐6 capsules with Fo92 olivine added to the starting glass. The compositions of experimental liquids are generally colinear with the trends observed in the lava suites used as the source material for the starting glasses. Significantly, aluminous spinel (Al2O3 contents of 61–68 wt%) was produced at 1290°C in all compositions and chromites (Al2O3 contents of 33–42 wt%) at lower temperatures in N‐MORB‐derived liquids despite no spinel having been added to the starting mixture. In addition, the experiments produced basaltic liquid in equilibrium with both >Fo89 olivine and >An85 feldspar at temperatures of 1230° and 1210°. These liquids have compositions with Mg# (at% Mg/Mg + FeT*100) that range from 63 to >85. The TiO2‐MgO correlation indicates large (∼16–23%) amounts of crystallization for each percent decrease in MgO. These results suggest the possibility that dry, anorthite‐bearing basaltic magmas are the product of the interaction between primary melt and Al‐spinel‐bearing upper mantle. In addition, the results indicate that MORB magmas can undergo a large amount (>50%) of crystallization prior to reaching 8% MgO. Further, although anorthite‐bearing magmas have characteristics consistent with their being a significant volumetric component of MORB “parent” magmas, the reaction mechanism suggested for their petrogenesis indicates that they are not necessarily primary magmas.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call