Mineralogical Association of Canada Short Course, 36 MELT INCLUSIONS IN PLUTONIC ROCKS

The greatest scientific value of melt inclusion research lies in its ability to constrain the compositions of parental magmatic liquids, and reveal the trends of liquid evolution in specific suites of natural igneous rocks. The cumulative number of melt inclusion papers has been growing exponentially during the last three decades (Bodnar & De Vivo 2003), and many have dealt with the most voluminous basaltic liquids, which play an exceptionally important role in magmatic differentiation of the Earth and other terrestrial planets. Although it is fair to conclude that melt inclusions have "come of age" in volcanic systems (Lowenstern 2003), the research on melt inclusions in mafic plutonic rocks is still in its infancy. The number of papers on melt inclusions in gabbros is at least 10 times smaller than the number of published studies of melt inclusions in the basaltic volcanic equivalents. Furthermore, many researchers express serious doubts about the very existence of gabbroic melt inclusions. The principles of identification and interpretation of crystallized melt inclusions in plutonic products of basaltic magma remain unclear. Much has been written already about the difficulty of recognizing fully crystallized melt inclusions in plutonic rocks (e.g., Roedder 1984a, Bodnar & Student 2006). The difficulty is probably greater in gabbroic rocks than in diorite, syenite, or granite, because of the broader temperature interval, in which basaltic magma crystallizes. Chemical compositions of basaltic liquids change more extensively in the course of magma crystallization than the composition of semi-eutectic syenitic or granitic melts, and prolonged magma evolution is likely to result in a diverse sequence of daughter mineral assemblages in crystallized melt inclusions. The identification and interpretation of mafic melt inclusions are further complicated by host-liquid reactions and other post-entrapment phenomena, which are examined in this chapter in detail. Apart from the difficulties imposed by nature, there are also psychological biases, the burden of conventional truths, and local scientific traditions. Very often people see only what they are prepared to see. The discovery of silicate liquid immiscibility, at first in melt inclusions from lunar samples brought by Apollo missions, and afterwards in ordinary terrestrial basalts (Roedder 1984a, b), is a vivid historic example of roundabout