We are pleased that our paper has generated interest from Retallack and relish the opportunity to comment further on the intriguing timing of the end-Permian extinction. Perhaps not surprisingly, we do not agree with any of Retallack’s assertions and deal with them here in the order he presented them.
Firstly, Retallack misconstrues the nature of the ‘ Permian”foraminifera that survive into the Triassic in Tibet. These taxa are not “disaster taxa” in the understood sense, although foraminiferal disaster taxa, such as Earlandia are well known from post-extinction strata in many sections (see Hallam & Wignall 1997, which includes a definition of disaster taxa). Neither do they have attributes of dysaerobic taxa, but rather they are forms commonly encountered in shallow marine platform carbonates of the Late Permian equatorial Tethyan realm. They turn up in Tibet, in the earliest Triassic, after they had gone extinct in their normal equatorial haunts. It is the same story for the inozoan sponge taxon that Retallack does not mention. However, these foraminifera did not survive for long as they went extinct within the Griesbachian Stage, so it is perplexing that Retallack considers this a “selective survival”. It is not a survival because all the foraminifera (probably 10 species in total) and the sponge went extinct! The implication that these extinctions are unimportant because they were non- fusuline taxa is simply spurious. The end- Guadalupian mass extinction eliminated the majority of fusulines and they had not recovered greatly by the time of the end-Permian mass extinction. The fusulines were not the “primary foraminiferal casualties”of this extinction event and Stanley and Yang (1996) did not claim otherwise. Rather foraminifera with calcareous granular tests, which included the surviving fusulines and also the diverse and abundant endothyrines, are the primary end Permian casualties (except in Tibet where they survive a little longer as we showed).
Secondly, Retallack refers to the excellent, detailed analysis of the brachiopod record of Selong by Shen and colleagues (2000, 2001). We do not doubt that many brachiopod species went extinct before the Permo-Triassic (P- Tr) at Selong, as they show, and never claimed otherwise in our paper. This is part of a diachronous mass extinction story within this section.
Thirdly, Retallack challenges the ocean stagnation cause for the marine anoxia and instead invokes dysoxia caused by methane oxidation. However, his alternative does not hold water. Firstly, Berner (2002) did not claim that atmospheric oxygen levels could have plummeted from 35% to 12% in the space in the space of 20 000 years across the P- Tr boundary. This would involve the loss of around 5 x 10 19 moles of oxygen from the atmosphere. Berner’s modelling showed that the negative C isotope shift across the boundary could have been caused by the release up to 3 x 10 17 moles of methane. Clearly, oxidation of this methane would have little impact on atmospheric oxygen levels. On the other hand, rapid release of this methane into the water column could have caused transient methanogenic anoxia. The Palaeoeocene-Eocene thermal maximum (PETM) provides a possible example of this phenomenon (Dickens 2000), but this is most unlike the P- Tr superanoxic event in both duration and intensity. The PETM event was brief (a few tens of thousands of years), only associated with transient, deep-water dysoxia, and linked with the emission of around half the amount of methane released during the end-Permian C isotope event. The P- Tr superanoxic event lasted around 20 million years (it began at the start of the Late Permian and finished at the start of the Middle Triassic, Isozaki (1997)) and, at its height, true anoxia was intensely developed even within shallow water locations ( Wignall & Twitchett 2002). Furthermore, the onset of this anoxic intensification predates the d 13C excursion ( Twitchett et al. 2001; Wignall & Newton 2003). It is impossible to envisage how the required vast volumes of methane could have been continuously released throughout this time interval, caused such intense oxygen restriction, and yet only left an isotopic signal at the P- Tr boundary.
Retallack dismisses our alternative warming-induced stagnation model because of a lack of organic-rich strata. However, numerous factors control organic carbon levels including sediment accumulation rates, dilution by (for example) carbonate content and surface water productivity and only modest primary productivity is perfectly compatible with anoxic conditions ( Wignall 1994). For this reason most workers use other criteria to assess oxygen levels, such as trace metal enrichment, fabric indicators such as the presence of undisturbed lamination, pyrite framboid size distributions etc. (cf. Wignall and Twitchett 2002; Wignall and Newton 2003). Retallack falsely claims that Lower Triassic marine shales generally have low total organic carbon (TOC) contents not exceeding 3 wt % and that they are less organic C-rich than Upper Permian and Middle Triassic shales. These assertions are not supported by the literature he refers to. Thus, Suzuki et al. (1993, p.711) note that the P- Tr boundary black shales in their Japanese sections have “2.2% to 5.3%” TOC and this is after they have been heated to over 100ºC, implying an original TOC content of “4 to 8%”(p.715). The sulfur/carbon plot of Kajiwara et al.’s (1994, fig. 1) study clearly demonstrates that the P- Tr boundary black shale is more organic-rich than the bounding sediments. Our data from Spitsbergen ( Wignallet al. 1998, table 1, fig. 8) shows that lowest Triassic shales have comparable or higher TOC values than the underlying uppermost Permian shales. The Morante (1996) study was of non-marine to paralic strata and so is irrelevant in this context. Krull et al.’s (2000, fig. 4) section was developed in a volcanogenic sandstone succession which, not surprisingly, has low TOC values although the highest values are in the Lower Triassic.
In finishing, we note that our claim of a diachronous P- Tr mass extinction was also based on evidence from British Columbia, which Retallack does not challenge. However, the BC data also provides crucial evidence for the diachroneity with a diverse radiolarian fauna disappearing over a meter below the loss of the benthic fauna (composed of abundant hexactinellid sponge spicules of diverse types). In our original paper we noted that the diverse radiolaria were replaced for a time by apparently simple radiolaria, which we referred to as “ sphaeroids”. Subsequently, Peter Cejchan (Prague) has succeeded in extracting the “ sphaeroids” from their matrix to reveal that they are simple, thin-walled siliceous spheres that lack diagnostic pores or ornament of radiolarians. This discovery serves to emphasise the abrupt demise of the radiolarian populations. Thus the evidence from BC and Tibet indicates a distinct, phased nature of the P- Tr extinctions in the marine realm:-
Phase 1: Abrupt elimination of radiolarian communities, probably several tens of thousands of years prior to phase 2, assuming typical pelagic sedimentation rates for the deep-water biogenic cherts of the BC section.
Phase 2: Mass extinction of marine benthos, including deep-water sponges in BC and many shallow-water taxa including brachiopods in Tibet.
Phase 3: Onset of a negative C isotope excursion that may record methane release from hydrates in a warming ocean. Foraminifera and sponge refugees from the tropics turn up in high latitude Tibet.
-- P- Tr boundary --
Phase 4. Extinction of refugees in Tibet due to the delayed onset of dysoxia in this region.
In detail, the negative d 13C excursion, and by implication the release of gas hydrates, occurs after the extinction phases indicating that it is a consequence of the end-Permian environmental crisis and not a factor in the extinction story. A similar sequence of timings is seen in the East Greenland which has an expanded record of the end-Permian mass extinction ( Twitchett et al. 2001).
Berner, R.A. 2002, Examination of hypotheses for the Permo-Triassic boundary extinction by carbon cycling modelling: Proceedings of the National Academy of Sciences, v. 99, p. 4172-4177.
Dickens, G.R. 2000, Methane oxidation during the Late Palaeocene Thermal Maximum: Bulletin de la Societe Geologique France, v.171, p. 37-49.
Hallam, A. and W ignall, P.B. 1997. Mass extinctions and their aftermath. Oxford University Press, Oxford, 320pp.
Isozaki, Y. 1997, Permian-Triassic superanoxia and stratified superocean: records from the lost deep sea: Science, v. 276, p. 235-238.
Krull, E.S., Retallack, G.J., Campbell, H.J. and Lyon, G.L. 2000. d 13C org chemostratigraphy of the Permian-Triassic boundary in the Maitai group, New Zealand: evidence for high latitude methane release: New Zealand Journal of Geology and Geophysics, v.43, p. 21-32.
Morante, R. 1996, Permian and Early Triassic isotopic records of crbon and strontum in Australia and a scenario of events about the Permian-Triassic boundary: Historical Biology, v. 11, p. 289-310.
S hen S-Z., A rchbold, N.W., S HI, G.-R. & C hen Z-Q. 2000, Permian brachiopods across the Selong Xishan section, Xizang (Tibet), China. Part 1: Straigraphy, Strophomenida, Productida and Rhynchonellida: Geobios, v.33, p. 725-752.
Shen S-Z., A rchbold, N.W., S HI, G.-R. & C hen Z-Q. 2001, Permian brachiopods across the Selong Xishan section, Xizang (Tibet), China. Part 2: Palaeobiogeographical and palaeoecological implications: Spirifierida, Athyridida and Terebratulida: Geobios, v.34, p. 157-182.
Stanley, S .M and Y ang, X. 1996, A double mass extinction at the end of the Paleozoic era: Science, v. 266, p. 1340-1344.
Suzuki, N., I shida, K. and I shiga, H. 1993, Organic geochemical implications of black shales relative to the Permian-Triassic boundary, Tamba Belt, southwest Japan: Japanese Geological Survey Bulletin, v. 44, p. 707-720.
Twitchett, R.J., L ooy, C.V., M orante, R., V isscher, H. and W ignall, P.B. 2001, Rapid and synchronous collapse of marine and terrestrial ecosystems during the end-Permian biotic crisis: Geology, v. 29, p. 351-354.
Wignall, P.B. and N ewton, R. 2003, Contrasting deep-water records from the Upper Permian and Lower Triassic of south Tibet and British Columbia: Evidence for a diachronous mass extinction: PALAIOS, v.18, p. 153-167.
Wignall, P.B. 1994. Black Shales. Oxford University Press, Oxford, 130 pp.
Wignall, P.B., M orante, R. and N ewton, R. 1998, The Permo-Triassic transition in Spitsbergen: d 13C org chemostratigraphy, Fe and S geochemistry, facies, fauna and trace fossils: Geological Magazine, v. 135, p.47-62.
Wignall, P.B. and T witchett, R.J. 2002, Extent, duration and nature of the Permian-Triassic superanoxic event: in Koeberl, C. and MacLeod, K.C., eds., Catastrophic events and mass extinctions: Impacts and beyond: Geological Soociety of America Special Paper 356, p. 395-413.