The Origin of a Most Contentious Rock

+ See all authors and affiliations

Science  01 Nov 2002:
Vol. 298, Issue 5595, pp. 961c-962c
DOI: 10.1126/science.298.5595.961c

In “Metasomatic origin of quartz-pyroxene rock, Akilia, Greenland and implications for Earth's earliest life” (Reports, 24 May, p. 1448), C. M. Fedo and M. J. Whitehouse conclude that rocks previously interpreted as metamorphosed sedimentary banded iron formation (BIF) and thought to contain the oldest evidence of life are instead highly deformed and metasomatized ultramafic igneous rocks. This conclusion is based, in part, on a comparison of trace element characteristics in the banded quartz-pyroxene rocks and various mafic and ultramafic igneous rocks. Unfortunately, data for only one of the ten banded rocks analyzed are shown in all of their chemical discrimination diagrams (Fig. 3). Although this one sample (AK 38) is consistent with the hypothesis, data for the nine remaining samples are very different and reveal it to be a misleading “proxy for the entire quartz-pyroxene lithology” as stated. Indeed, ratios of Th/Sc, Cr/Th, Cr/Y, and TiO2/P2O5 in the nine samples overlap those reported for the Isua BIF and are distinct from those in ultramafic rocks. It is unlikely that a combination of metasomatic gains and losses could have produced such a match. Given that the origin of banding in these rocks is equivocal and that some elemental data (rare earth element) are incompatible with either an ultramafic or BIF precursor, ratios of the least mobile trace elements would seem to be an appropriate means to constrain protolith identity. This being the case, the original BIF interpretation should be considered viable until more definitive evidence (perhaps oxygen isotopes) becomes available.


Our study integrated (i) detailed structural observations, (ii) mineralogy, (iii) geochemistry, and (iv) regional geologic events for interpreting rocks that have been repeatedly deformed and metasomatized over approximately 2 billion years. Palin questions our proposed metasomatized ultramafic igneous origin for quartz-pyroxene rocks on Akilia solely on the basis of assessment of a few geochemical ratios that overlap the field of Isua banded iron formation (BIF) (see Fig. 3A), suggesting that they support previous claims (1, 2) for a BIF protolith.

We are not surprised by the broad compositional range in Cr/Th versus Th/Sc or TiO2 versus P2O5 for the quartz-pyroxene rocks and pyroxenite boudins, considering that metasomatic quartz (and likely carbonate) has diluted their original ultramafic composition by in excess of 90% in some samples. In addition to overlapping the field of Isua BIF, which spans a factor of 20 in Cr/Th and a factor of 60 in Th/Sc, these ratios overlap average Early and Late Archean basalt, Early and Late Archean andesite, Early Archean graywacke, and Early Archean cratonic shale (3) and therefore cannot be used to constrain a unique BIF protolith. In TiO2 versus P2O5 space (Fig. 3C), our samples do not lie in the field of Isua BIFs, as Palin claims. Furthermore, Y/Ho and Zr/Hf ratios for all samples of the quartz-pyroxene rock show no affinity for sediments crystallized in equilibrium with seawater (4) but instead have values similar to ultramafic samples from Akilia. Palin rejects structural geologic, mineralogic, and geochemical data indicating that sample AK 38, a thick pyroxenite band not invaded by metasomatic quartz, is integral to the lithology, preferring instead to call it an “exception” relative to other quartz-pyroxene rocks. We categorically reject Palin's claim that the thin banding has an “equivocal” origin, for we clearly demonstrated that it is mineralogically similar to and associated with the boudinage of thicker pyroxenite bands (Fig. 2, B to D).

A BIF origin for the quartz-pyroxene rock was hypothesized on the basis of magnetite layering and “comparison” with other units, such as BIF at Isua (2). We reiterate, there is little magnetite in the quartz-pyroxene rock, some enclosing ultramafic rocks contain magnetite, and all pertinent rocks on Akilia possess a penetrative tectonic layering (foliation). Banding in the quartz-pyroxene rocks does not result from quartz-magnetite alternations, as is common at Isua. Consequently, the initial BIF hypothesis has little intrinsic merit and cannot be vindicated by a few, non-source-specific geochemical ratios.


  1. 1.
  2. 2.
  3. 3.
  4. 4.

Related Content

Navigate This Article