Attribution: You must attribute the work in the manner specified by the author or licensor ( but no in any way that suggests that they endorse you or your use of the work).Noncommercial ‒ you may not use this work for commercial purpose.No Derivative works ‒ You may not alter, transform, or build upon this work.Sharing ‒ Individual scientists are hereby granted permission, without fees or further requests to GSA, to use a single figure, a single table, and/or a brief paragraph of text in other subsequent works and to make unlimited photo copies of items in this journal for noncommercial use in classrooms to further education and science.

We thank Huggett and Hesselbo for their interest in our work and substantial agreement with our conclusions when they state, “We do not dispute the possibility of a postapocalyptic greenhouse, nor that this could explain the existence at high latitudes of a mineral normally associated with low-latitude soils.” What they question are the details of mineral identification and genesis. We'll address each of their three issues individually.

1. Berthierine from modern settings is not associated with greenhouse conditions. We cited two examples of berthierine occurrences in soils/paleosols in our original article (Sheldon and Retallack, 2002). Our point, in contrast to the quotation above, is that berthierine is exceptionally rare in soils forming at any latitude, yet we found large (20–30 cm) nodules of it in the paleosols. Further, berthierine is uniquely found in the first few paleosols above the isotopically defined Permian-Triassic boundary (Sheldon and Retallack, 2002); other iron minerals including pyrite are found throughout the sections. In the Sydney Basin the total section examined in detail is 42 m (Retallack, 1999), at Mount Crean it is 140 m, in Allan Hills it is 290 m, and at Graphite Peak it is 560 m (Retallack and Krull, 1999). We think large nodules of berthierine unique to paleosols immediately above the Permian-Triassic boundary in four separate and extensive localities is compelling evidence of something unusual.

2. If iron in the 7 Å green clay is ferric, it may have been odinite. Using an electron microprobe, it is not possible to distinguish between the two different redox states of iron, but titration combined with bulk chemical analyses (Retallack and Krull, 1999) show very low amounts of ferric iron in these green nodules. It is also possible to discount odinite as the 7 Å green clay in this study from published data for odinite (Bailey, 1988), which give molar Mg/Al ratios of 1.4–5.7, while our clays have molar Mg/Al ratios of 0.2–1.0 and other published berthierine data (Brindley, 1982) give molar Mg/Al ratios of 0.1–1.3. Odinite is significantly more Al-poor than berthierine as indicated by molar Si/Al ratios of 5.0–10.2 for odinite (Bailey, 1988) versus molar Si/Al ratios of 2.1–3.6 for our clays (Sheldon and Retallack, 2002) and molar Si/Al ratios of 1.2–3.6 for other published berthierine data (Brindley, 1982). Our analyses overlap with published berthierine analyses, but not with published odinite analyses.

3. Berthierine replacement of odinite during burial diagenesis. We disagree with the suggestion that berthierine could be replacing odinite for a variety of reasons. First, as we illustrated in our original paper, the scanning electron microscope (SEM) textures of the berthierines are inconsistent with replacement based on comparisons with published SEM images of secondary berthierines (Retallack and Krinsley, 1993). Second, the assertion that berthierine associated with laterites indicates formation in oxidizing conditions and subsequent diagenetic reduction is misleading. The china clays of laterites are chemically reduced and very strongly weathered, while nearby horizons are oxidized and plinthic. Third, no laterites or ooids were seen in the Antarctic paleosols, so comparison with lateritic or marine ooids is not appropriate.

Finally, in preparing this reply, we came across two errors in our original paper that we wish to correct. First, while the text described the calculation of pCO2 values correctly, equation 4 was incorrect. Equation 4 should have read:  

Second, when we calculated the pCO2 values, the activity of silica was inadvertently placed in the denominator of equation 1. Recalculating a “worst case” pCO2 with our original parameters (25 °C, quartz saturation, and ~160 ppm CH4) using this corrected equation gives a soil pCO2 of 2.29 bars, or ~6000 times present atmospheric levels (PAL) (Fig. 1) as opposed to our published limit of ~60 PAL (0.021 bars pCO2) discussed in our original paper (Sheldon and Retallack, 2002). Is this value still realistic? Tropical and subtropical soils typically have soil pCO2 values in excess of 100–200 PAL (Brook et al., 1983; Volk, 1987), or about an order of magnitude lower than our theoretical limit. As Retallack (1999) originally reported, earliest Triassic paleosols indicate exceptional degrees of weathering at very high paleolatitudes (65–85°S) with no modern analogue except for tropical ecosystems forming at low to mid-latitudes (<45°N or S). Given that the observed 7‰–10‰ δ13C shift at the Permian-Triassic boundary can only be adequately explained by a release of ~4200 Gt C from marine or permafrost clathrates (Berner, 2002), earliest Triassic soil pCO2 levels could have exceeded values in modern tropical soils.

We stand by our original identification of the green phyllosilicate in the four high paleolatitude sites as berthierine. Berthierine is abundant only at this stratigraphic level and is consistent with dysoxic and hypercapnic formation conditions (Sheldon and Retallack, 2002) as a result of elevated atmospheric and soil pCH4 levels, which are indicated by carbon isotopic data (Krull and Retallack, 2000) and mass balance modeling of the carbon cycle (Berner, 2002).