Category: Geobiology, palaeontology, and evolution

There is a Gaelic proverb, which loosely translated goes: “There are more ways of killing a cat than drowning it in butter”.  That seems apt for mass extinctions, particularly the most severe, at the end of the Palaeozoic.  A new hypothesis points the finger towards breathing problems, but not those likely from massive, ground-hugging emissions of sulphur dioxide from the Siberian flood basalts that coincide with the P-Tr extinction: “everyone knows” that they resulted in the universal coughing reflex in all surviving land vertebrates…..  Raymond Huey and Peter Ward of the University of Washington reckon a major contributing factor for terrestrial extinctions was a fall in atmospheric oxygen (Huey, R.B. & Ward, P.D. 2005.  Hypoxia, global warming and terrestrial Late Permian extinctions.  Science, v. 308, p. 398-401).

For most of the Carboniferous and Early Permian Earth flipped in and out of glacial conditions that dominated the southern supercontinent of Gondwana.  Tropical latitudes were cloaked in dense vegetation for the first time.  Rapid sedimentation buried vast amounts of carbon in the form now taken by the world’s largest and most extensive coal deposits.  Net carbon burial for 90 to 100 Ma resulted in extraordinary oxygen concentrations in the atmosphere. One line of evidence for that is the huge size of Carboniferous and Early Permian insect fossils, such as those of dragonflies.  Insects do not breathe, but take in oxygen by a diffusive process through spiracles on the underside of their bodies.  The more oxygen the larger they can grow.  Carbon burial also links in with the global cooling that made the Carbonierous and Early Permian susceptible to astronomic forcing of glacial-interglacial cyclicity: CO₂ fell.

The present-day oxygen concentration in the air is about 22%, whereas estimates for the Carboniferous Permian peak are around 30%.  Most land animals today, including ourselves, have an altitude limit to permanent life of around 4 to 5 km, though the vast majority live much lower.  In the Early to Middle Permian, the availability of oxygen for respiration corresponding to that at sea level today would have been around 6 km altitude, and at the top of a mountain the height of Everest breathing would be easy.  The limit to altitude range of animals would have been temperature rather than oxygen availability.  So, given sufficient warmth, the area available for animal life would have been very high.  Estimates of the oxygen level at the end of the Permian are as low as about 16%.  Even living at sea level would have demanded an ability to survive at about 2.7 km today, and at 6 km during the oxygen-rich Early and Middle Permian.  Evolution of land animals during the 100 Ma long “global winter” would have adjusted to elevated oxygen availability, which Huey and Ward believe would have led to at least a limited altitude stratification of available ecosystems, governed by temperature.  Their hypothesis is that declining oxygen forced extinctions by reducing the habitable range severely, and increased competition among those taxa able to live in the reduced, low-altitude land area: probably patches of “refugia”.

The decline in oxygen was accompanied by global warming.  Permian and Triassic sedimentary records show a dramatic increase in red terrestrial sediments, coloured by iron oxide.  Iron had been released and oxidised to insoluble iron(III), possibly by increased continental weathering, which would have sequestered oxygen by the formation of iron oxide coatings to sedimentary grains.  Increased oxidation would also have encouraged biodegradation by aerobic bacteria, which may have run-away to help boost atmospheric CO₂ levels.  One testable outcome of such events is the rate of extinction during the Late Permian, which should have risen slowly, rather than plummeting at the P-Tr event.  Another is that survivors might show signs of adaptation to low oxygen levels, and indeed some Triassic reptiles do.  All in all, those times were stressful on land.  Yet the extinctions were just as severe in marine ecosystems, where the fossil record is more complete.  Less oxygen and warmer seas would have resulted in similar hypoxia for aquatic animals.

The late Jack Sepkoski did a lasting service for those who study life’s record by combing the literature to compile the first and last appearance of each marine fossil genus.  It is from this archive that we have been able to visualise mass extinctions and those less in magnitude numerically.  As well as the “Big Five” there are other die-offs, particularly through the Mesozoic and Cenozoic record.  To some extent the extinction patterns also appear among terrestrial taxa that have been less well documented, partly because few have had Sepkoski’s determination and partly because land organisms leave fewer traces.  It quickly became apparent to him and other palaeontologists that extinction occurred sharply, which is why the biologically-determined division of Phanerozoic time since 542 Ma is so well defined world-wide.  What also emerged from inspection of the time series of genus and family numbers was a pulse in the timing of significant extinctions, which appears to have been between 25 and 30 Ma.  That struck a chord with specialists in volcanic activity, and there is a good correlation between the occurrence of flood-basalt outpourings and extinctions.  But at least one of the five largest extinctions, at the K-T boundary, coincides with abundant evidence for a major impact by an extraterrestrial body.  Planetary scientists then began looking for a pulsed variation in the intensity of bombardment of the Inner Solar System.  There is no tangible evidence of that, although there are theoretical arguments that suggest that the Sun in its ~250 Ma orbit around the galactic centre wobbles through dust arranged in bands close to the galactic plane every 30 Ma.

Extinctions are not, of course, the only features of the fossil record.  Primarily it charts variations in diversity, of which suddenly lowered numbers are one aspect in broader fluctuations.  Each extinction eventually precedes an increase in diversity as adaptive radiation from surviving taxa fills ecological niches left vacant or under-populated.  That part of the record has its fascinations, as complexity seems to have emerged in three great pulses, through the Palaeozoic, Mesozoic and Cenozoic Eras, each producing more diverse forms than its predecessor.  There are also slackenings in the pace and periods of apparent stasis.  Getting to numerical grips with the full record requires analysis that uses similar mathematical techniques to that which unlocked proof of Milankovich’s theory of astronomical pacing of climate from finely calibrated oceanic-sediment records.  It is possible to analyse time series in terms of discrete frequencies from which the curves can be reconstructed.  Physicists Robert Rohde and Richard Muller of the University of California have used this Fourier analysis on the 36 thousand strong catalogue published after Sepkoski’s death, with some recalibration of the time scale and some pruning of data – they removed genera with only a single record or whose age is poorly known (Rohde, R.A. & Muller, R.A. 2005.  Cycles in fossil diversity.  Nature, v. 434, p. 208-210).  There are definitely distinct frequencies that dominate the record, and they cannot be present by chance, although that is a purely statistical view.  But to their surprise, and everyone else’s, they are completely unexpected ones at 62 and 140 Ma.  It is proving exceedingly difficult to come up with plausible Earthly or extra-terrestrial explanations.  There are two interesting features: the 62 Ma periodicity dominates the record of relatively short-lived genera; and the “Big Five” seem to fit neatly into the patterns of diversity, albeit at unequally spaced intervals, when the effects of background fluctuations have been removed.  That filtering may allow for increasing preservation towards recent times.  One major control over diversity is, logically, a mixture of the number of potential niches and their geographic isolation, and both are probably related to plate tectonic activity.  Unfortunately, fluctuations in 2 and even 3 geographic dimensions have only the broadest calibration to time.  Added to that is the complex way in which global sea level has changed with time.  So we can expect a great deal of head scratching, and it may come as a relief that the crowing of some volcanologists and impact theorists may have been silenced at a single stroke!

See also:  Kirchner, J.W. & Weil, A. 2005.  Fossils make waves.  Nature, v. 434, p. 147-8.

The Doushantuo Formation of southern China dates from just before the Cambrian Explosion, and has become a source of astonishing information about animals that preceded the appearance of those with hard parts.  It contains fossil embryos, algae, achritarchs, and small bilaterians that are purportedly the Earth’s earliest animals.  Moreover the formation rests on the cap carbonates of a diamictite reckoned to represent a late Neoproterozoic glacial epoch, and provides a variable trend of carbon-isotope variation that extends up to the base of the Cambrian in southern China.  Because the sequence contains a number of volcanic ash beds it is potentially dateable.  Using a single-zircon U-Pb method, Daniel Condon of MIT and colleagues from the Chinese Academy of Science have established the ages of both top and base of the Doushantuo Formation with considerable precision (Condon, D. et al. 2005. U-Pb Ages from the Neoproterozoic Doushantuo Formation, China.  Science Express, 24 February 2005).  Sedimentation is bracketed between 635 and 550 Ma, the oldest age coinciding with that for the Ghaub tillite in Namibia.  Time-calibration of the carbon-isotope record allows it to be matched with others in Namibia, Oman and Newoundland.  There is one snag; within the sequence is a formation boundary that signifies non-deposition, which the authors correlate with a glacial epoch recognised in Newfoundland (the Gaskiers diamictite), citing sea-level withdrawal as the cause of non-deposition in China.  The well-constrained correlation suggests a major, global increase in the burial of ¹²C that produced a marked negative excursion in d¹³C that spans around 90% of the Ediacaran Period that saw the rise of large soft-bodied animals shortly before the emergence of shelly faunas.  The interpretation placed by the authors on this signature of burial of dead organic matter, which relates to no sign of glaciation, is that it would have elevated oxygen levels in the Late Neoproterozoic oceans.  That might have increased productivity by primitive eukaryotes, and possibly opportunities for predation.  The uppermost part of the Doushantuo Formation broadly coincides with the first appearance of complex trace fossils and mollusk-like bilaterians, and elsewhere there are signs of the first reef formation by weakly calcified metazoans at around that time.  Clearly, it is well-dated sections such as these that may hold the key to what exactly prompted the general secretion of skeletal material; the hallmark of the 10 Ma later explosion in fossil animals.

No graphite in Akilia apatites, no sign of life?

In the first EPN of 2005 evidence was reported that weighed against a sedimentary origin for the ~3.8 Ga ironstones of West Greenland from which isotopically light carbon had been claimed to indicate the earliest signs of life (see Iron isotopes enter the Archaean life debate January 2005 EPN).  The original work that claimed a biological signature in carbon from the oldest known metasedimentary rocks focussed on carbon-isotope analyses of apatites in them, in the belief that they would have withstood intense metamorphic alteration because of the resistance of that mineral to chemical reactions.  Following close on the heels of that revelation comes one a great deal more worrying for aficionados of biogeochemistry.  Geoscientists from Estonia, France, the US and Sweden have systematically made petrographic observations on apatite grains from the rocks of the Akilia Association, including those originally reported as carrying geochemical signs of life existing at that time (Lepland, A. et al. 2005.  Questioning the evidence for Earth’s earliest life – Akilia revisited.  Geology, v. 33, p. 77-79).  Of the 190 individual apatite grains examined in 17 rocks, not one showed the slightest trace of carbonaceous material.  It seems that apatite is unlikely to have been the host for the low d¹³C that caused such a stir in palaeobiological circles when it was first announced, and may well not be a good place to look for biomarkers.  It also throws into question what did produce the signal.  If it was the bulk rock, then the depletion in ¹³C could have resulted from temperature induced isotopic fractionation.  Another possibility is that the samples were contaminated with modern biological materials, despite the precautions taken to avoid that.

 

In December 2004 EPN commented on what appears to be a serious challenge to claims of geochemical evidence that would support a major impact associated with the largest of all mass extinctions in the Phanerozoic, that at the close of the Permian Period and the Palaeozoic Era, around 251 Ma ago.  Newly published analyses from two other well-constrained P-Tr boundary sites found no signs of the elements that would be expected from a major collision with a metal or silicate-rich asteroid (Koeberl, C. et al. 2004.  Geochemistry of the end-Permian extinction event in Austria and Italy: No evidence for an extraterrestrial component.  Geology, v. 32, p. 1053-1056).  Koeberl of the University of Vienna and colleagues from the US and UK focussed on platinum-group elements (PGEs), and osmium and helium isotopes. Both sites are stratigraphically similar and dominated by carbonate sediments, with evidence from one site for deepening water that laid down organic-rich marls.  Sure enough, there is a “spike” in iridium at the level of these marls, which had been documented at the Austrian site in 1989, and there is another 50 m higher in the sequence.  The new work confirmed both, and also found the marl-related “spike” in Italy. But the reason why iridium has been used to suggest extraterrestrial impacts is because, of all the PGEs, it is the easiest to analyse at very low concentrations. That can give rise to “false positives”, for there are purely terrestrial processes that can concentrate PGEs.  An unambiguous arbiter between these processes and impacts lies in the isotopic composition of the metal osmium.  Rocks of the Earth’s crust have high rhenium (Re) and low osmium (Os) contents, whereas in meteorites the Re/Os ratio is very much smaller.  The unstable isotope ¹⁸⁷Re decays to produce a daughter ¹⁸⁷Os that adds to the common ¹⁸⁸Os isotope. Consequently, terrestrial rocks acquire high ¹⁸⁷Os/^!88Os rapidly after they crystallise from magmas and that “signature” is imparted to the entire surface environment through weathering and solution. On the other hand, meteorites have low ¹⁸⁷Os/^!88Os ratios, so the two influences on the geochemical record can be distinguished – if you have good enough analytical facilities.  The two iridium spikes fail that test, as regards an impact origin.  It seems likely that they originated through precipitation of PGEs from sea water under reducing conditions on the deep sea floor.  The helium isotope data carry the same negative message; they are typically terrestrial.

Impact-induced extinctions, particularly ones that wipe out a sizeable proportion of all organisms, are likely to be unremittingly sudden – direct effects being felt within hours over the whole planet, and secondary effects such as “nuclear winter” and acid rainfall over a matter of a few years or decades. Radiometric dating is incapable of resolving such short periods, and at the age of the P-Tr boundary probably not even several hundred millennia. Faunal sequences can give a better indication of abruptness. To most intents the marine record at the time does look as if extinction was very sharp, but it does not indicate anything by way of clear evidence for an impact, such as glass spherules, shocked quart grains and other tell-tale signs.  The continental record is pretty sparse, so has not figured much in the debate.  However, the Karoo basin of South Africa contains thick continental sediments that span the boundary, and is famous for its primitive reptile fauna, some of which became extinct around the time of the P-Tr event.  Incidentally, this die-off created the genetic conditions for the adaptive radiation in the Mesozoic that led not only to the dinosaurs but also the mammals and birds.  Charting the timing of the Karoo extinctions has proved difficult, although it appears not to have been sudden in a stratigraphic sense.  New age data has emerged from studies of palaeomagnetic field reversals in the sediments, together with variations in carbon isotopes, that allow timing to be better assessed through comparison with magnetic and carbon records from other sections (Ward, P.D. et al. 2005.  Abrupt and gradual extinction among Late Permian land vertebrates in the Karoo Basin, South Africa.  Science [soon to be published, currently available on Sciencexpress at www.sciencemag.org/sciencexpress/recent.shtml]).  The signs are that the proto-reptiles died off over tens to hundreds of thousand years due to some protracted crisis, probably connected with the giant continental flood basalt eruptions that formed the Siberian Traps. Those lavas overlap the timing of the P-Tr boundary, and would certainly have added sufficient CO₂ to give substantial global warming and also massive emissions of SO₂ that would have created chemically hazardous conditions on a global scale.

New predators on the Mesozoic block

Most people have been led to believe that, although the earliest mammals appeared in the Triassic fossil record, throughout the Mesozoic they were tiny and meekly scurried and skulked while the dinosaurs reigned supreme over land, sea and air.  They had to wait for the K-T extinction to develop their full ecological potential.  That is now a myth, for Chinese strata (yet again) have revealed much larger mammals than ever thought possible, and some of them ate dinosaurs (Hu, Y. et al. 2005.  Large Mesozoic mammals fed on young dinosaurs.  Nature, v. 433, p. 149-152).  One indisputable mammal skeleton contained the bones of young dinosaurs in its body cavity.  In fact so many that one wonders if it met its end through greed.

At the end of the Triassic Period, around 200 Ma ago, life underwent a major crisis that so far has not been believably connected to either extraterrestrial or geological causes.  Previous studies have shown that the mass extinction was accompanied by an decrease in ¹³C in sediments that suggests a short-lived global warming of  between 2-4 °C at the Tr-J boundary.  That CO₂ levels rose is suggested by a decrease in the density of pores (stomata) on fossil leaves.  It has been suspected for some time that the largest known continental igneous event, which accompanied early rifting of the modern Atlantic Ocean basin may have been responsible, but so far the dating of this Central Atlantic magmatic province (CAMP) has not been tied to the boundary conclusively.  A large consortium of Italian, French, US, Moroccan and Swiss has addressed the sedimentary and igneous record around Tr-J times in the High Atlas of Morocco (Marzoli, A and 14 others 2004.  Synchrony of the Central Atlantic magmatic province and the Triassic-Jurassic boundary climatic and biotic crisis.  Geology, v. 32, p. 973-976).  There, one of the few uneroded continental flood basalt sequences of CAMP (most preserved CAMP magmas are in the form of sills and dykes in offshore basins) occurs among Triassic and Jurassic sediments.  Their base deforms the underlying sediments, suggesting that eruption was onto unlithified sediments, shortly after their deposition.  Fossils from the sediments are of little help in tying down the age of eruption, however, Ar-Ar ages of the lavas are all within error of 200 Ma, and tally with magnetic stratigraphy from the Tr-J boundary elsewhere.  Both age and geochemistry of the flows are remarkably similar to those of flood basalts from the other side of the Atlantic.  Magmatic duration, like that in other large igneous provinces was of short duration, no more than a couple of million years.  So it now seems that three of the “big five” mass extinctions (the others are end-Permian, connected with the Siberian Traps, and the K-T boundary and associated Deccan Traps) have at least a partial cause from CO₂ release by massive volcanism.

Iron isotopes enter the Archaean life debate

Some years ago geochemists obtained carbon-isotope data from 3.8 Ga rocks in Greenland that seemed at the time to be persuasive evidence for the emergence of life during or shortly after Earth’s most traumatic period.  Up to 3.8 Ga the Moon was bombarded by huge projectiles, and its companion Earth would have received at least 13 times the flux of destruction.  The carbon was within sturdy apatite grains from supposed iron-rich metasediments, and may have been preserved from later high-grade metamorphism.  Doubt has been cast on that hypothesis, either because of the unlikelihood of any carbon remaining unfractionated by heating, or because some aspects of the rocks’ geochemistry suggested that they we of igneous origin rather than sediments.  Readers will have seen in previous years’ EPN that a controversy rages over even tangible signs that suggest cellular material from rocks half a billion years younger.  Geochemists from France and the US have taken a different tack with the ancient Greenlandic rocks that ought to at least resolve the igneous versus sedimentary origin of the banded iron-rich rocks (Dauphas, N. et al. 2004.  Clues from Fe isotope variations on the origin of Early Archean BIFs from Greenland.  Science, v. 306, p. 2077-2080).  They found that the heavy iron isotope ⁵⁷Fe is more enriched in the ironstones than in any igneous rocks, with little chance that the difference was induced by thermal fractionation.  They are metasediments.  But therein lies a surprise.  The heavy-iron signatures are greater than in less aged banded ironstones.  One way in which that could have arisen is from biogenic precipitation of soluble reduced Fe-2, perhaps involving anoxygenic photosynthesisers – because of the strong capacity of photosynthesis for setting electrons in motion, all such organic reactions create local oxidising conditions, whether or not oxygen itself is produced.

Studies of the organic chemicals in meteorites and in “space snow” that falls continually on the Earth, show that amino acids and nucleotides (the CGAT building blocks of nucleic acids), together with other moderately complex compounds, were widespread in the solar nebula as it formed.  They can form in the absence of life.  Life’s dependence on DNA and RNA for its necessary self-replication marks a chemically complex step that assembled such building blocks by a process of polymerisation.  That presupposes an awful lot of chance reactions, none more so than the formation of the peptide bond that dominates genetic material and proteins.  Lots of mechanisms have been tested, but none work sufficiently well in a test tube to be plausible candidates for processes on the early Earth.  Perhaps the simplest, first proposed more than 30 years ago is the operation of a simple gas called carbonyl sulphide (COS).  Experiments that expose amino acids to carbonyl sulphide in water at “room temperature” yield lots of peptides in a matter of a few minutes to hours (Leman, L. et al. 2004.  Carbonyl sulphide – mediated prebiotic formation of peptides.  Science, v. 306, p. 283-286).  The more metal ions, such as those of iron, lead and cadmium, that are in the solution, the more efficient the reactions.  The likeliest place for such processes to go on would be near submarine hydrothermal vents, as COH quickly breaks down once emerged from a volcanic source.  Its role could have been crucial in the complex molecular evolution that many biochemists believe to have been intimately associated with the structures of clays and sulphide minerals that hydrothermal activity produces in abundance.

The volcanism versus impact debate about the K-T boundary runs and runs, as newshounds tend to say.  Things are not so evenly balanced for the biggest of all mass extinctions at the end of the Permian.  Although signs have been reported, a link with an impacting extraterrestrial body has not convinced a decisive majority.  On the other hand, there is a 1-2 Ma mismatch between the well-determined age (around 253 Ma) of the Siberian Traps and previous dates for the end of Permian stratigraphy in sections that have no depositional break with the Triassic.  The extinction has all the hallmarks of a catastrophe, by definition a sudden event, so tying down its age and that of a plausible cause is essential.  Not being able to do that for the K-T event and the Deccan Traps, and with uncertainties about the relationship of impact rocks to signs of extinction at the Chicxulub site, add fuel to that long-running debate.  The accepted “golden spike” or GSSP for the Permian-Triassic boundary is at Meishan in eastern China, and there are other sites in China that run it close.  The sections contain several volcanic ash layers, so zeroing in on a date for the extinction would seem straightforward, using U/Pb zircon dating.  There is a problem.  Some of the zircons in the ashes are xenocrysts rather than having formed during the various magmatic episodes, and they are microscopically indistinguishable from those that should give precise dates.  All the zircons also show signs of having lost radiogenic lead during later alteration of the beds.  The last could explain the mismatch with the Ar-Ar age of the Siberian Traps, the generally favoured culprits for the extinction.  US and Australian geochemists have taken a new tack in dealing with these problems (Mundil, R et al. 2004.  Age and timing of the Permian mass extinction: U/Pb dating of closed system zircons.  Science, v. 305, p. 1760-1763).  They have “aggressively” treated zircon grains to remove outer parts from which radiogenic lead has been lost, so leaving isotopically undisturbed cores of the grains.  Their U/Pb data are mainly from a boundary section in central China (Shangsi), dating 8 separate ash layers, plus one from the boundary clay itself at the Meishan GSSP.  The dates agree well with the stratigraphic sequence of the ashes, and hare high precision.  Judging the actual age of the boundary at Shangsi relies on statistical analysis of the sequence of ages from the different ashes, and gives a date of 252.6±0.2  Ma.  That is within error of the accepted Ar-Ar age of the Siberian Traps.  As usual, this is not cut and dried, because there are other ages for the Siberian Traps, including one using the same U/Pb zircon method that suggests a 251.4 Ma age.  Clearly the mismatches for the end-Permian events will be a meaty bone of contention, when all respected geochronologists turn up for a meeting early in 2005 to thrash out the conflicts that continually inflame their passions.

As with several major extinction events, the Permian-Triassic boundary is characterised by a major excursion in carbon isotopes of marine towards negative d¹³C.  This is often taken to indicate a reduction in the burial of dead organic matter, perhaps because of low global biomass.  US, Chinese and Canadian geoscientists have added great detail to the P-Tr carbon-isotope record  from analysis of three continuous sections through carbonate-dominated sequences in an Early Triassic reef system in southern China (Payne, J.L. et al. 2004.  Large perturbations of the carbon cycle during recovery from the end-Permian extinction.  Science, v. 305, p. 506-509).  This is no ordinary reef, for it was built by carbonate secretions by micro-organisms, either algae or bacteria.  The tabulate coral reef builders of the Palaeozoic became extinct at the end of the Permian (251 Ma), and their successors, scleractinian corals, do not appear until about 10 Ma later.  The Early Triassic was undoubtedly characterised by low animal diversity, before adaptive radiation could “re-stock” a devastated biosphere.  The authors found a remarkable series of ups and downs in d¹³C within the reef carbonates, some of the negative excursions being even more severe than that just after the mass extinction.  Some of the positive peaks go far beyond the d¹³C levels in preceding and following times, and could be due to periods of extremely high burial of organic matter.  But the fossil record shows that such burial probably involved a restricted number of taxa, so perhaps there were huge “blooms” among a few groups that filled vacant ecological niches only to collapse.  As suddenly as this see-sawing of the carbon cycle had begun, at about 246 Ma it settled to a more or less constant level, just after the start of the Middle Triassic.  There are two reasonable explanations for the fluctuations.  One is that biotic recovery from the mass extinction was set back three of four times by further environmental upheavals, thereby dashing diversification.  The other is that the fluctuations reflect instability in the simple ecosystems of the Early Triassic and their control on carbon burial.

Calcium in the ocean and the Cambrian Explosion

If ever there was a geoscientific topic that would “run and run”, it would be explaining why creatures with hard parts just popped into being 542 Ma ago.  Physiologically, members at the phylum level of the Cambrian fauna have little in common apart from hard parts made from calcium compounds, either carbonate or phosphate.  Calcium carbonate was secreted as stromatolites by blue-green bacteria as far back as the Archaean, but not in an organised form linked to their function.  In the very latest Precambrian, the Ediacaran, there are tiny shell-like bits and pieces in its very uppermost strata (the “small shelly fauna”) but they suggest no obvious function and no association with any of the various soft-bodied metazoans that define that Period.  The Cambrian Explosion has no rudimentary precursor.  Because calcium is an element with a very narrow tolerance in cells, from the level needed for viable function (it has a “messenger” function) to that at which it is fatally toxic, and it is a common element in all environments, adoption of calciferous hard parts seems very likely to have a risen as a means of avoiding toxicity, without any other role.  Once established in large animals, hard parts provide a means of and a defence against predation, so losing the ability to secrete hard parts would be an evolutionary risky strategy; once established it cannot be lost except when substituted by other effective defences or mealtime tackle.  There were times in the Precambrian record when calcium compounds exceeded their solubility, and they are marked by inorganically precipitated crystalline forms in sediments.  The early Archaean was one such period, but if levels of Fe-2 are high in water those solubilities are enhanced.  Therein lies a link between Archaean and Palaeoproterozoic stromatolites, banded iron formations and the oxidation potential of seawater.  In fact precipitation of BIFs seems to link nicely with the abundance of stromatolites, because the production of oxygen by blue-green bacteria would locally have consumed electrons to oxidise soluble Fe-2 to Fe-3 that has insoluble oxides and hydroxides.  This connection returned several times in the Neoproterozoic, oddly at the times of so-called “Snowball Earth” episodes, first noted by Preston Cloud.  Could the last of these have triggered adoption of calcium secretion by the early metazoan animals?  That is hard to judge, because it preceded the Cambrian by several tens of million years.  Geochemists from the US Geological Survey, the State University of New York and the US Oak Ridge National Laboratory have taken a cunning route to shedding some light on the biggest of all palaeontological mysteries (Brennan, S.T. et al. 2004.  Seawater chemistry and the advent of biocalcification.  Geology, v. 32, p. 473-476).  They sought crystals of evaporitic halite that spanned the Precambrian-Cambrian boundary, and which usually contain fluid inclusion containing samples of the brine from which they formed, hopefully seawater.  So far, they have two sets of suitable halites that can be assigned to a marine environment, from Siberia and the Oman, and their measurements of calcium concentrations are very precise.  The first is dated around 515 Ma the other set from 544 Ma.  Two sample points are not enough to prove a role for elevated calcium levels in the ocean, but the results are encouraging.  Calcium concentrations (with suitable corrections for changes during evaporation of restricted seas) jumped by a factor of 3 from the very end of Precambrian to Cambrian times.  Over the same period, it is thought that global sea-floor spreading rates were much higher than at present, and there is also strontium-isotope evidence for an increase in ocean-floor hydrothermal activity that adds elements derived from oceanic basalts to seawater.  That, however post-dates the start of the Cambrian by about 15 Ma.  With a CO₂-rich atmosphere and elevated continental weathering calcium is likely to have been supplied from the continents.  Whatever, the results fit with models based on variation of continental and oceanic additions to seawater with changing spreading rates (Hardie, L.A. 2003.  Secular variations in Precambrian seawater chemistry and the timing of Precambrian aragonite seas and calcite seas.  Geology, v. 31, p. 785-788).  Hardie suggested that calcium in seawater fell to very low levels during the Neoproterozoic from an unprecedented high at its outset at 1000 Ma.  That is a time when metazoans were probably not around, while the period when they appear in the later Neoproterozoic record was one of calcium-poor conditions.  Large animals may have evolved when there was little danger of calcium shock, only to face it once they were well established.  Then would have had to rid their cells of it very efficiently. Studies of fluid inclusions from marine precipitates seem likely to grow following Brennan et al.’s important discovery, though suitable samples are likely to be few and far between.  One important role they need to play is verifying Hardie’s model for secular variation in seawater chemistry, which depends on difficult interpretations of rates of sea-floor spreading and continental erosion.

Ancestral animal?

The significant feature of the first appearance of widespread, large fossils during the Cambrian Explosion about 542 Ma ago was really the adoption of hard parts by most of the existing (and some now extinct) phylla of animals.  The preceding Neoproterozoic Ediacaran Period witnessed lots of large life forms, but preserved them only as imprints; they were soft  bodied.  Superficially, the outset of the Cambrian appears to marked the simultaneous emergences of the rough blueprints of all subsequent animals.  In reality, this was probably not a faunal explosion, but one of biochemical processes, wherein many phylla turned the fundamental cell process of excreting excess calcium as carbonate and phosphate to generating functional parts of their bodies.  Why that happened explosively is still a mystery.  Looking for the origin of animals requires going further back in geological time, and an element of luck as regards exceptional preservation of soft tissue.  The other way is using a molecular clock approach to the genetic differences among modern phylla, but that is fraught with uncertainties and gives a very large time range (possibly 1500 to 600 Ma) in which to find tangible evidence.  The maximum limit is around 2200 Ma, when oxygen became significant in the atmosphere and the upper ocean – the prime condition for eukaryote life.  A rather dull carbonaceous fossil, with a spiral form and thought to be the first known multicelled eukaryote (Grypania) appears in the record about 1500 Ma ago, but what it was is unclear.  The best place to look for ancestral animals is in known repositories of well preserved organisms.  One such lagerstät is the Doushanto Formation in SW China.  This goes back to the last “Snowball Earth” event at 600 Ma, and has been heavily mined for primitive life forms.  Chinese palaeontologists, teamed up with others from the USA have indeed found something intriguing (Chan, J-Y. et al. 2004.  Small Bilaterian Fossils from 40 to 55 Million Years Before the Cambrian.  Science, v. 305, p. 218-222).  Only about 0.2 mm across, 10 specimens seems to show microscopic signs of all the basic elements of many members of the Animal Kingdom: bilateral symmetry, a mouth and gut, skin tissue and possible sensory organs.  The layers from which they were extracted are between 580 to 600 Ma, well before the Cambrian Explosion.  However, micropalaeontologists in general subscribe to the “once bitten, twice shy” outlook, especially following controversies over even earlier evidence for small organisms and those purported to occur in Martian meteorites, which are as likely to be results of inorganic mineralisation as fossils.  Various mineral crusts and films, formed inorganically, can mimic organic structures.  The one feature that persuades Chen and colleagues is that the same features show up in all the specimens, and they are all the same size.  That is highly unlikely from some inorganic process.

Source:  Stokstad, E.  2004.  Controversial fossil could shed light on early animals’ blueprint.  Science, v.  304, p. 1425.

The case of the stranded, tiny mammoths

It does seem likely that our ancestors ate all the mammoths (Mammuthus primigenius), a species that had wandered over the northern tundras bordering the Northern Hemisphere ice sheets through several glacial-interglacial periods.  But some of them did escape to survive into the Holocene.  They were stranded on high-latitude islands off NE Siberia and Alaska as sea levels rose.  The last of them died on Wrangel island about 4 thousand years ago.  A common tendency in small populations of large mammals that are restricted to islands is that they become smaller and smaller with each generation.  This happened to the stranded mammoths of the Bering Straits islands, remains of which are often dwarfs.  (Guthrie, R.D. 2004.  Radiocarbon evidence of mid-Holocene mammoths stranded on an Alaskan Bering Sea island.  Nature, v. 429, p. 746-749).  St Paul Island is now only 91 km² in area, too small to support even tiny, woolly elephants, but it was probably much larger when sea-level rise first isolated it from the vast Bering steppe across which mammoths roamed.  It was that isolation about 13 thousand years ago that probably helped the stranded mammoth population avoid the hunters who colonised the Americas, until 7 900 when the last mammoth there died.   The even later population on much larger Wrangel Island fell to human colonisation, but there are no signs of human intervention on St Paul.  The earlier extinction there was probably a result of shrinking browse as sea level steadily rose., when St Paul would have been 5 to 10 times larger than it is now.

The four largest extinction events of the Phanerozoic (late Devonian, 370 Ma; end-Permian, 251 Ma; end-Triassic 201 Ma; end-Cretaceous , 65 Ma) each coincide with periods of rapid and voluminous continental flood-basalt volcanism.  There is also evidence from the extinction horizons that each coincided with a major impact event as well, most widely accepted for the end-Cretaceous event.  Geological time is so long that pure chance cannot be ruled out entirely to explain coeval impacts and CFB events, but is unlikely (a 1 in 8 chance for one coincidence, but 1 in 3500 for four).  So there has been a long-running controversy over a volcanic or an extraterrestrial cause for extinctions, together with speculation that large impacts can somehow trigger CFB events.  The last does not work for the end-Cretaceous extinction, because the Deccan volcanism began somewhat before the formation of the “smoking-gun” Chicxulub crater, and a linking mechanism is not clear.  Taking into account lesser extinctions and CFB events, there is a rough periodicity of 30 Ma and similar ages for both.  Geoscientists at the Geomar Institute of the University of Kiel in Germany have stoked up the controversy by taking a very different view of events (Phipps Morgan, J. et al. 2004.  Contemporaneous mass extinctions, continental flood basalts, and ‘impact signals’: are mantle plume-induced lithospheric gas explosions the causal link?  Earth and Planetary Science Letters, v. 217, p. 263-284) albeit not a completely new one.  They consider the processes at depth that presage CFB events, where rising mantle material impacts at the base of thick continental lithosphere.  Each of the CFB provinces linked in time to the four large extinctions lies on an ancient craton, devoid of tectonic activity for over a billion years, and greatly depleted in heat-producing elements.  Lithosphere beneath them is over 300 km thick and might have acted in the manner of the lid on a pressure cooker, building up gas pressure during the delay in breaking through overlying rock.  Eventually pressure would be sufficient to breach the lithosphere, and gases (CO₂ and SO₂) would be explosively vented, perhaps creating globally toxic conditions.  Release of the pressure would lead to collapse above the plume head that would propagate upwards, at hypersonic speeds according to the authors.  Maybe that would fling enormous amounts of rock into the stratosphere.  Some chunks might be large enough to cause big impact structures at the surface when they fell back, so explaining the coincidence.  They account for the pre-extinction start of CFB outpourings, as in the case of the Deccan traps, by lateral and upwards migration of part of the plume to locally thinned lithosphere.  The power involved in such an event extending through the entire lithosphere could account for the shocked grains, microspherules and fullerenes in known extinction horizons.  Being sourced in mantle rock that may once have resided near the core-mantle boundary, such a process could also eject high iridium concentrations that were the signs that first led to the Alvarez’ hypothesis of impact-induced extinctions, but without an extraterrestrial culprit.  Despite the attractions of the impact theory, no sign of meteoritic debris has been found in any of the ejecta horizons or the craters themselves.  On Phipps Morgan and colleagues’ account that is not surprising, because the impacting objects would have been common Earth rock.  The authors decided to dub these hypothetical events “Verneshots” after Jules Verne’s book From the Earth to the Moon, which involved a giant gun firing the space craft moonwards.  If there is anything in the idea, then surely there would be spectacular evidence of the source of the blasts, but perhaps they are conveniently buried by later CFBs.  Geophysical studies do show signs of circular features beneath both the Deccan and Siberian Traps.  However, the associated seismic shock waves would pervade large volumes of crust outside the blast vent, and signs of that, such as shatter cones, are perhaps an easier target.  As with all departures from “accepted wisdom”, the Geomar group’s ideas will come in for a lot of stick, quite possibly from the fans of giant impacts, who not so long ago were themselves dismissed as “whizz-bang kids” by many geoscientists.

That gas build-up might lead to catastrophic crustal collapse gets some support from a modelling study on the processes involved in volcanic collapse (Reid, M.E. 2004.  Massive collapse of volcano edifices triggered by hydrothermal pressurization.  Geology, v. 32, p. 373-376), albeit in miniature.  Mark Reid of the USGS focuses on those volcano collapses that occur without any warning signs from eruptions and seismicity.  His study examines the effects of deep intrusion of magma on the groundwater systems within stratovolcanoes.  This could promote increases in gas pressures deep within the edifice.  Their upward propagation would destabilise the entire volcanic structure, leading to its collapse in extreme situations.  The modelling indicates increased likelihood of over-pressuring where permeability is low; a crude analogy to Phipps Morgan and colleagues’ pressure lid of inert cratonic lithosphere.  Gas-rich magmas can emerge explosively in continental flood basalt provinces, normally regarded as forming by episodic, quiet outpourings from fissure systems.  That is well demonstrated by the Ethiopian-Yemeni CFB province.  The main basaltic trap sequence is followed by very widespread felsic ignimbrites on both sides of the Red Sea that formed by lateral blasts of incandescent debris and felsic lava shards.  Only one example of an ignimbrite centre is known from the province.  Lying about 60 km south of Sa’ana, near the small town of Mabar, it is a circular structure about 18 km across with clear concentric zoning.  Interestingly the zones dip steeply towards the centre of the structure, in an inverted cone, that is possibly due to collapse even more dramatic than in the calderas that sourced the more familiar ignimbrites of the Andes.

See also:   Ravilious, K. 2004.  Four days that shook the world.  New Scientist * may 2004, p. 32-35.

How, when and under what circumstances vertebrates got limbs to take them charging across the forested land of the late Palaeozoic form a central issue in our own evolution, as well as that of the other four-footed land animals.  By negative analogy with the functional though rather rudimentary enlarged fins of various modern fish that flop from pond to pond during dry seasons, many vertebrate palaeontologists have considered limbs as evolutionary adaptations in air-breathing fish once they made this a habit.  As so often, the fossil record has not given up enough evidence for that to be certain.  Well, an upper foreleg bone (humerus) has turned up in Late Devonian rocks from Pennsylvania at a time and in a context that strongly suggests it was carried by a fish (Shubin, N.H. et al. 2004.  The early evolution of the tetrapod humerus.  Science, v. 304, p. 90-93).  While not able to ride a bicycle, the advanced fish probably used what became limbs to hold itself motionless while lying in ambush for its prey.  That would provide a plausible point of departure from which walking might develop.

Early biomarkers in South African pillow lavas

It is now established that various kinds of bacteria infest rocks down to depths of 2 km or more, one particularly favourable habitat being in sea-floor basalts though which hydrothermal fluids travel.  Although the majority probably inhabits cracks and joints, some seem to work actively to corrode rock, especially volcanic glass, thereby obtaining mineral nutrients.  Signs of this microbial corrosion in modern volcanic glasses are radiating tubes on a scale of a few micrometres, that show up in micrographs, and many may have been overlooked by petrographers in all kinds of rock.  That they are definitely formed by organic activity is demonstrated by the presence of nucleic acids, carbon and nitrogen in the tubules.  Carbon isotopes from them show the strong depletion in ¹³C that is the hallmark of organic fractionation of natural carbon.  A team of geoscientists, from Norway, Canada and the USA, who have steadily accumulated evidence for biological rotting in modern oceanic basalts, turned their focus to the oldest, well- preserved pillow lavas in the 3.5 billion-year old Barberton greenstone belt of north-eastern South Africa (Furnes, H. et al. 2004.  Early life recorded in Archean pillow lavas.  Science, v. 304, p. 578-581).  Virtually identical microtubules seem common in them too, particularly in hydrated glasses that are now tinged with the low-grade metamorphic mineral chlorite.  Indeed, chlorite seems to have grown preferentially from clusters of the holes, which suggests that they formed before metamorphism of the basalts.  Micro-geochemical studies confirm the presence of hydrocarbons with low d¹³C.  The bulk of the tubules occur in the inter-pillow debris, that probably formed as glassy rinds as magma protruded on the Archaean sea floor.  As well as adding to evidence for ancient terrestrial life, the find has inevitably opened up the search for such signs in meteorites reckoned to have come from Mars.  In two, olivine grains show similar structures, although why the olivine hadn’t broken down in the presence of water that is essential for life makes such observations worth taking with a pinch of salt. A number of studies have stymied claims for early bacterial fossils (see Artificial Archaean “fossils” and Doubt cast on earliest bacterial fossils, April 2002 and December 2003 issues of EPN) and inorganic processes conceivably might create structures that can be mistaken for ones formed by biological action.  The Fischer- Tropsch  process is capable of producing hydrocarbons, and produces depletion in ¹³C abiogenically.  In the on-line April edition of Science Express (www.sciencexpress.org) experiments are reported that highlight the possible influence of chromium-bearing mineral catalysts in hydrothermal generation of hydrocarbons from inorganic carbon dioxide(Foustoukos, D.I. & Seyfried, W.E. 2004.  Hydrocarbons in hydrothermal vent fluids: the role of chrome-bearing catalysts.  Science Express, April 2004).  The Barberton greenstone belt is well known for ultramafic lavas rich in chromium, as are most early volcanic sequences.

See also:  Kerr, R.A. 2004.  New biomarker proposed for earliest life on Earth.  Science, v. 304, p. 503.