Correlation of marine and continental Quaternary pollen records from the Northeast Pacific and western Washington

Late-glacial and postglacial pollen stratigraphy and radiocarbon chronology of a marine core from the continental slope and a core from the western Olympic Peninsula, ca. 110 km apart, are compared. Divisible into four pollen assemblage zones (L, P-1, P-2, and P-3), the cores exhibit a succession of correlative zonal prominences: grass-sedge (L), pine (P-1), alder (P-1-P-2 boundary), and hemlock (P-3). Volcanic ash of Mt. Mazama provenance is also correlative in zone P-2. Quantitative relationships of the pollen in the cores (relative and absolute numbers and pollen influx) are dissimilar, however, and are attributed to the influence of the Columbia River pollen load reaching the locale of the continental slope core compared with the local pollen rain influencing the Olympic Peninsula core site.     

Extreme fractionation in felsic magma chambers: a product of liquid-state diffusion or fractional crystallization?

Extreme fractionation of minor and trace elements commonly accompanies very modest changes in major element concentrations in highly felsic igneous sequences. In such sequences, Si increases by only a few percent while, for example, Sr, Ba, Mg, and light rare earth elements decrease drastically, commonly by a factor of 10 or more. It has been argued, most notably by Hildreth (e.g. [1]), that such trends observed in tuffs were not induced by fractional crystallization (FC), but rather are a manifestation of compositional gradients in parental magma chambers which form via liquid-state thermogravitational diffusion (LSTD). The strongest arguments against FC are that (1) crystal settling is not a viable mechanism for crystal-liquid separation, and (2) extensive recrystallization is required to produce the observed trends, yet the tuffs are relatively crystal-poor. Many workers have noted trends in plutonic as well as volcanic rocks which are strikingly similar to those for which LSTD has been proposed, and some have concluded that LSTD was the fractionating mechanism.Several lines of evidence lead us to the conclusion that FC is the dominant differentiating process in high-silica magmas: (1) elemental trends are strikingly consistent with those predicted for FC; it would be a remarkable coincidence if diffusion-induced trends mimicked FC so closely; (2) large phenocryst assemblages in high-silica tuffs indicate low-variance liquid compositions that would be improbable if crystal-liquid equilibria were not controlling differentiation; (3) highly evolved plutonic rocks in many cases do not form the caps expected for LSTD, but rather occur in dikes and pods where they apparently segregated as late liquids; (4) recent experimental studies suggest that trends induced by diffusion differ drastically from observed felsic igneous trends.We do not believe that the principal arguments against FC in high-silica systems (unlikelihood of crystal settling; crystal-poor nature of tuffs) refute the reality of the chemical process, but rather emphasize the need for a better understanding of the physical mechanisms of crystal-liquid fractionation and eruption.