ON THE INTERPRETATION OF CRYSTAL SIZE DISTRIBUTIONS IN MAGMATIC SYSTEMS

The observation that basic igneous rocks most commonly are holocrystal line under a wide spectrum of cooling regimes implies that cooling and crystallization can be uncoupled and considered separately. This is t antamount to realizing that the Avrami number is large in most igneous systems. Crystallization automatically adjusts through nucleation and growth to the cooling regime, and all aspects of the ensuing crystal population reflect the relative roles of nucleation and growth, which reflect the cooling regime. The characteristic scales of crystal size, crystal number, and crystallization time are intimately tied to the c haracteristic rates of nucleation and growth, but it is the crystal si ze distributions (CSDs) that provide fundamental insight on the time v ariations of nucleation and growth and also on the dynamics of magmati c systems. Crystal size distributions for batch systems are calculated by employing the Johnson-Mehl-Avrami equations for crystallinity rela ted to exponential variations in time of both nucleation and growth. T he slope of the CSD is set by the difference a-b, where a and b are ex ponential constants describing, respectively, nucleation and growth. T he batch CSD has constant slope and systematically migrates to larger crystal size (L) with increasing crystallinity. The diminution in nucl eation with loss of melt is reflected in the CSD at late times by a st rong decrease in population density at small crystal sizes, which is r arely seen in igneous rocks themselves. Observed CSDs suggest that a-b similar-to 6-10 and that b similar-to 0. That is, growth rate is appr oximately constant and nucleation rate apparently increases exponentia lly with time. Correlations among CSD slope, intercept, and maximum cr ystal size for both batch and open systems suggest that certain diagno stic relations may be useful in interpreting the CSD of comagmatic seq uences. These systematics are explored heuristically and through the d etailed examination of comagmatic CSDs in a number of igneous and indu strial systems including, amongst others, Makaopuhi lava lake, Atka vo lcanic center. Peneplain sill, Dome Mountain lavas, Shonkin Sag laccol ith, and Kilaureu Iki lava lake. None of these systems show CSDs typic al of purely batch or purely open system, even when the system itself is known on independent grounds to be a batch system. Instead, the CSD s of each system reflect a combination of kinetic and dynamic influenc es on crystallization. Heterogeneous nucleation and annexation of smal l crystals by larger ones, entrainment of earlier grown and ripened cr ystals, rate of solidification front advance, and protracted transit o f a well-established much column are some of the effects revealed in t he observed CSDs. There may be an overall CSD evolution, reflecting th e maturity of the magmatic system, from single straight nonkinked CSDs in monogenetic systems to multiply kinked, piecewise continuous CSDs in well-established systems such as Hawaii and Mount Etna. This is not unlike the evolution of CSDs in some industrial systems. Finally, the fact that comagmatic CSDs are not often captured evolving systematica lly through large changes in nucleation rates, even in low crystallini ty systems, may suggest that magma is always laced with high populatio n densities of nuclei, supernuclei, and crystallites or clusters that together set the initial CSD at high characteristic population densiti es. Further evolution of the CSD occurs through sustained heterogeneou s nucleation and rapid annealing at all crystallines beginning at the liquidus itself and operating under more or less steady (not exponenti ally increasing) rates of nucleation.