The outlook
Glassy inclusions in metasedimentary enclaves, as well as their crystallized counterpart - nanogranites - in migmatites and granulites, contain the melt phase that was being produced during incongruent crustal melting and that was enclosed by the peritectic minerals growing simultaneously. Trapped during a prograde event of magma formation, these MI differ from the granitic inclusions found in minerals from plutonic and volcanic rocks of granitoid composition (e.g., Yang & Bodnar, 1994; Thomas et al., 2002; Anderson, 2003; Webster, 2006), which formed during magma crystallization upon cooling.
There is still a lot of work to be done in order to understand the best ways to observe and remelt nanogranites, to determine how representative are MI of the bulk melt composition in the system, and how and to which extent the retrograde history affects MI by interaction with the host. Nonetheless, our finding of MI in migmatites and granulites has two important consequences.
From a microstructural point of view, recognition of MI (in particular of nanogranites) is a proof that a rock was partially melted at some time in its history. While in many migmatites this can be inferred by several other methods (reviewed by Vernon, 2011) there are cases where MI may be the only textural evidence left of anatexis. One example are polymetamorphic basements, such as some internal crystalline massifs in the Alps - Gran Paradiso (Biino and Pognante, 1989) and Dora-Maira (Compagnoni & Rolfo, 2000) - where Variscan anatexis was followed by an Alpine evolution involving HP-LT metamorphism and subsequent greenschist-facies reequilibration. In this case, especially in the zones of intense deformation, it is not uncommon that textural and mineralogical evidence of anatexis has been totally erased, except for the persistence of garnet relicts that may contain nanogranite inclusions. Because of this, we believe that the occurrence of nanogranites should be included among the most reliable microstructural criteria for the former presence of felsic melt (e.g., Vernon, 2011).
From a chemical point of view, these nm- to µm-scale objects, so far studied in igneous rocks, allow for the direct analysis of natural anatectic melt compositions, overcoming the problems and uncertainties that are involved in assuming leucosomes as representative of anatectic melts in regional-scale migmatites (Brown, 2010). Melt inclusion compositions will thus provide much more reliable chemical constraints to the petrological and geochemical models of crustal melting processes. For example, trace element analyses of MI can open new developments for geochemistry and thermobarometry, as shown by Acosta-Vigil et al. (2010).
With the fast development of microanalytical tools, MI studies in migmatites and granulites may become a routine object of study in crustal petrology. At present, they also represent a promising subject for the successful application of cutting-edge techniques (for petrological purposes) such as nanoSIMS, ToF-SIMS (Rost et al., 2009), FIB TEM and synchrotron-based µ-XRD, µ-XRF and µ-CT.
We believe that many occurrences of MI have been overlooked because they simply were not searched for, and that they will be uncovered by careful re-investigation of migmatite and granulite samples worldwide. The preservation of inclusions depends on the chemical and mechanical behavior of the host-inclusion system during the post-entrapment history of the rock. Since microfracturing would allow the infiltration of fluids and modification of the primary melt composition or nanogranite assemblage, MI should be targeted in the most chemically inert and mechanically strong mineral hosts (e.g., garnet, ilmenite, rutile, spinel, zircon) from the least deformed rock domains.