Two suites of olivine-rich ultramafic and feldspathic rocks appear to be present in the Mid-Atlantic Ridge: one which seems to have alkalic affinities, and one similar to the chromitite- bearing alpine peridotite-gabbro complexes. The similarities of rocks in the two environments—continental and oceanic—imply that much about the petrology of mid-oceanic ridges may be learned from studies of continental complexes, and that silicic rocks have been formed in the mantle. Although gabbros in St. Paul Rocks and similar rocks at Tinaquillo, Venezuela, and Lizard, England, have been interpreted as not comagmatic with intimately associated peridotite by some petrologists, evidence to the contrary at Lizard is discussed. Association of fresh gneissic gabbro, some containing quartz, with talcose serpentinite, amphibole schist, quartz diorite and epidotic but unsheared basalts along the Mid-Atlantic Ridge is believed to indicate presence of alpine-type rocks that occur normally in eugeosynclinal belts.
Gabbro, described as partly interlayered with peridotite by gravitational differentiation, forms major parts of three widely separated ultramafic complexes which have been interpreted as slices of oceanic crust and upper mantle: the Troodos massif in Cyprus, the Bowutu Mountains in Papua, and the Camagiiey complex in central Cuba. If, as Dietz has suggested, peridotite and related rocks in eugeosynclines represent fragments of ocean rind formed along mid-oceanic ridges and moved laterally by ocean-floor spreading, gabbro must be an essential constituent of the upper mantle. This could account for many geophysical anomalies, but would complicate some postulated mechanisms involved in ocean-floor spreading.